Haematologica, Volume 108, Issue 12 by Haematologica

haematologica Journal of the Ferrata Storti Foundation

haematologica.org

VOL.

108 DECEMBER 2023

ISSN 0390 - 6078

haematologica Editor-in-Chief Jacob M. Rowe (Jerusalem)

Deputy Editors Carlo Balduini (Pavia), Jerry Radich (Seattle)

Associate Editors Shai Izraeli (Tel Aviv), Steve Lane (Brisbane), Pier Mannuccio Mannucci (Milan), Pavan Reddy (Houston), David C. Rees (London), Paul G. Richardson (Boston), Francesco Rodeghiero (Vicenza), Gilles Salles (New York), Kerry Savage (Vancouver), Aaron Schimmer (Toronto), Richard F. Schlenk (Heidelberg), Sonali Smith (Chicago)

Statistical Consultant Catherine Klersy (Pavia)

Editorial Board Walter Ageno (Varese), Sarit Assouline (Montreal), Andrea Bacigalupo (Roma), Taman Bakchoul (Tübingen), Pablo Bartolucci (Créteil), Katherine Borden (Montreal), Marco Cattaneo (Milan), Corey Cutler (Boston), Kate Cwynarski (London), Ahmet Dogan (New York), Mary Eapen (Milwaukee), Francesca Gay (Torino), Ajay Gopal (Seattle), Alex Herrera (Duarte), Martin Kaiser (London), Marina Konopleva (Houston), Johanna A. Kremer Hovinga (Bern), Nicolaus Kröger (Hamburg), Austin Kulasekararaj (London), Shaji Kumar (Rochester), Ann LaCasce (Boston), Anthony R. Mato (New York), Matthew J. Mauer (Rochester) Neha Mehta-Shah (St. Louis), Moshe Mittelman (Tel Aviv), Alison Moskowitz (New York), Yishai Ofran (Haifa), Farhad Ravandi (Houston), John W. Semple (Lund), Liran Shlush (Toronto), Sarah K. Tasian (Philadelphia), Pieter van Vlieberghe (Ghent), Ofir Wolach (Haifa), Loic Ysebaert (Toulouse)

Managing Director Antonio Majocchi (Pavia)

Editorial Office Lorella Ripari (Office & Peer Review Manager), Simona Giri (Production & Marketing Manager), Paola Cariati (Graphic Designer), Giulia Carlini (Graphic Designer), Debora Moscatelli (Graphic Designer), Igor Poletti (Graphic Designer), Marta Fossati (Peer Review), Diana Serena Ravera (Peer Review), Laura Sterza (Account Administrator)

Assistant Editors Britta Dost (English Editor), Rachel Stenner (English Editor), Anne Freckleton (English Editor), Rosangela Invernizzi (Scientific Consultant), Marianna Rossi (Scientific Consultant), Massimo Senna (Information Technology), Luk Cox (Graphic Artist)

Haematologica | 108 - December 2023

Brief information on Haematologica Haematologica (print edition, pISSN 0390-6078, eISSN 1592-8721) publishes peer-reviewed papers on all areas of experimental and clinical hematology. The journal is owned by a non-profit organization, the Ferrata Storti Foundation, and serves the scientific community following the recommendations of the World Association of Medical Editors (www.wame.org) and the International Committee of Medical Journal Editors (www.icmje.org). Haematologica publishes Editorials, Original articles, Review articles, Perspective articles, Editorials, Guideline articles, Letters to the Editor, Case reports & Case series and Comments. Manuscripts should be prepared according to our guidelines (www.haematologica.org/information-for-authors), and the Uniform Requirements for Manuscripts Submitted to Biomedical Journals, prepared by the International Committee of Medical Journal Editors (www.icmje.org). Manuscripts should be submitted online at http://www.haematologica.org/. Conflict of interests. According to the International Committee of Medical Journal Editors (http://www.icmje.org/#conflicts), “Public trust in the peer review process and the credibility of published articles depend in part on how well conflict of interest is handled during writing, peer review, and editorial decision making”. The ad hoc journal’s policy is reported in detail at www.haematologica.org/content/policies. Transfer of Copyright and Permission to Reproduce Parts of Published Papers. Authors will grant copyright of their articles to the Ferrata Storti Foundation. No formal permission will be required to reproduce parts (tables or illustrations) of published papers, provided the source is quoted appropriately and reproduction has no commercial intent. Reproductions with commercial intent will require written permission and payment of royalties. Subscription. Detailed information about subscriptions is available at www.haematologica.org. Haematologica is an open access journal and access to the online journal is free. For subscriptions to the printed issue of the journal, please contact: Haematologica Office, via Giuseppe Belli 4, 27100 Pavia, Italy (phone +39.0382.27129, fax +39.0382.394705, E-mail: info@haematologica.org). Rates of the printed edition for the year 2022 are as following: Institutional: Euro 700 Personal: Euro 170 Advertisements. Contact the Advertising Manager, Haematologica Office, via Giuseppe Belli 4, 27100 Pavia, Italy (phone +39.0382.27129, fax +39.0382.394705, e-mail: marketing@haematologica.org). Disclaimer. Whilst every effort is made by the publishers and the editorial board to see that no inaccurate or misleading data, opinion or statement appears in this journal, they wish to make it clear that the data and opinions appearing in the articles or advertisements herein are the responsibility of the contributor or advisor concerned. Accordingly, the publisher, the editorial board and their respective employees, officers and agents accept no liability whatsoever for the consequences of any inaccurate or misleading data, opinion or statement. Whilst all due care is taken to ensure that drug doses and other quantities are presented accurately, readers are advised that new methods and techniques involving drug usage, and described within this journal, should only be followed in conjunction with the drug manufacturer’s own published literature.

Direttore responsabile: Prof. Carlo Balduini; Autorizzazione del Tribunale di Pavia n. 63 del 5 marzo 1955. Printing: Press Up, zona Via Cassia Km 36, 300 Zona Ind.le Settevene - 01036 Nepi (VT)

Associated with USPI, Unione Stampa Periodica Italiana. Premiato per l’alto valore culturale dal Ministero dei Beni Culturali ed Ambientali Haematologica | 108 - December 2023

Table of Contents Volume 108, Issue 12: December 2023

About the Cover Image taken from the review by Robert Stuver et al. in this issue.

Landmark Papers in Hematology 3191

Clonal evolution of leukemia from G6PD studies Tal Bacharach and Liran I. Shlush https://doi.org/10.3324/haematol.2023.284215

3193

Central nervous system therapy in acute lymphoblastic leukemia: no more, no less Susana Rives https://doi.org/10.3324/haematol.2023.283275

3195

Expanding the bacterial origins of nodular lymphocyte-predominant Hodgkin lymphoma Maher K. Gandhi and Colm Keane https://doi.org/10.3324/haematol.2023.283392

3197

Early splenectomy in sickle cell disease: another piece of the puzzle Raffaella Colombatti and Maddalena Casale https://doi.org/10.3324/haematol.2023.283481

3199

Long-term survivors in severe aplastic anemia: standard mortality rates now comparable to controls? Andrea Bacigalupo https://doi.org/10.3324/haematol.2023.283523

3201

Worms on stage Aleksandra A. Pandyra and Arndt Borkhardt https://doi.org/10.3324/haematol.2023.283709

Editorials

Review Series on Peripheral T-Cell Lymphomas 3204

Introduction to the peripheral T-cell lymphoma review series: advances in molecular characterization, classification refinement and treatment optimization Kerry J. Savage and Laurence de Leval https://doi.org/10.3324/haematol.2023.282719

3211

Past, present and future therapeutic approaches in nodal peripheral T-cell lymphomas Henry S. Ngu and Kerry J. Savage https://doi.org/10.3324/haematol.2021.280275

3227

Pathobiology of nodal peripheral T-cell lymphomas: current understanding and future directions Bettina Bisig, Kerry J. Savage and Laurence de Leval https://doi.org/10.3324/haematol.2023.282716

Haematologica | 108 - December 2023

3244

Few and far between: clinical management of rare extranodal subtypes of mature T-cell and NK-cell lymphomas Robert Stuver, Zachary D. Epstein-Peterson and Steven M. Horwitz https://doi.org/10.3324/haematol.2023.282717

3261

Biology and genetics of extranodal mature T-cell and NK-cell lymphomas and lymphoproliferative disorders Natasha E. Lewis et al. https://doi.org/10.3324/haematol.2023.282718

Articles 3278

3287

3298

3308

3321

3333

3347

3359

3372

Acute Lymphoblastic Leukemia Effect of two additional doses of intrathecal methotrexate during induction therapy on serious infectious toxicity in pediatric patients with acute lymphoblastic leukemia Janina Heilmann et al. https://doi.org/10.3324/haematol.2022.281788 Acute Lymphoblastic Leukemia Impact of central nervous system involvement in adult patients with Philadelphia-negative acute lymphoblastic leukemia: a GRAALL-2005 Corentin Orvain et al. https://doi.org/10.3324/haematol.2022.282332 Bone Marrow Failure Conditional survival and standardized mortality ratios of patients with severe aplastic anemia surviving at least one year after hematopoietic cell transplantation or immunosuppressive therapy Ryotaro Nakamura et al. https://doi.org/10.3324/haematol.2023.282781 Hematopoiesis Clonal hematopoiesis with DNMT3A and PPM1D mutations impairs regeneration in autologous stem cell transplant recipients Patrick Stelmach et al. https://doi.org/10.3324/haematol.2023.282992 Myeloproliferative Disorders Haploidentical donor hematopoietic cell transplantation for myelodysplastic/myeloproliferative overlap neoplasms: results from a North American collaboration Tania Jain et al. https://doi.org/10.3324/haematol.2023.283426 Non-Hodgkin Lymphoma Long non-coding RNA mitophagy and ALK-negative anaplastic lymphoma-associated transcript: a novel regulator of mitophagy in T-cell lymphoma Valentina Mularoni et al. https://doi.org/10.3324/haematol.2022.282552 Hodgkin Lymphoma B-cell receptor reactivity against Rothia mucilaginosa in nodular lymphocyte-predominant Hodgkin lymphoma Lorenz Thurner et al. https://doi.org/10.3324/haematol.2023.282698 Plasma Cell Disorders Amyloidogenic light chains impair plasma cell survival Marjorie Pick et al. https://doi.org/10.3324/haematol.2022.282484 Plasma Cell Disorders IL6Myc mouse is an immunocompetent model for the development of aggressive multiple myeloma Michael D. Pisano et al. https://doi.org/10.3324/haematol.2022.282538 Haematologica | 108 - December 2023

3384

3392

3399

3409

3418

Plasma Cell Disorders Cancer-specific mortality in multiple myeloma: a population-based retrospective cohort study Arleigh McCurdy et al. https://doi.org/10.3324/haematol.2023.282905 Plasma Cell Disorders Disease associations with monoclonal gammopathy of undetermined significance can only be evaluated using screened cohorts: results from the population-based iStopMM study Aðalbjörg Ýr Sigurbergsdóttir et al. https://doi.org/10.3324/haematol.2023.283191 Plasma Cell Disorders Changing trends in the risk factors for second primary malignancies after autologous stem cell transplantation for multiple myeloma before and after the introduction of proteasome inhibitors and immunomodulatory drugs Hiroyuki Takamatsu et al. https://doi.org/10.3324/haematol.2023.283202 Red Cell Biology & its Disorders Early splenectomy in a large cohort of children with sickle cell anemia: risks and consequences Aimen Mechraoui et al. https://doi.org/10.3324/haematol.2022.282556 Red Cell Biology & its Disorders Metabolic signatures of cardiorenal dysfunction in plasma from sickle cell patients as a function of therapeutic transfusion and hydroxyurea treatment Angelo D’Alessandro et al. https://doi.org/10.3324/haematol.2023.283288

Letters 3433

Prognostic value of early positron emission tomography in patients with large B-cell lymphoma treated with anti-CD19 chimeric antigen receptor T-cell therapy Jennifer L. Crombie et al. https://doi.org/10.3324/haematol.2022.282345

3438

Hematological abnormalities in Jacobsen syndrome: cytopenia of varying severities and morphological abnormalities in peripheral blood and bone marrow Daiki Yamashita et al. https://doi.org/10.3324/haematol.2022.282513

3444

Specific blood monocyte distribution in histiocytoses correlates with vascular involvement and disease activity Jerome Razanamahery et al. https://doi.org/10.3324/haematol.2023.282739

3449

Mechanisms of endothelial injury and transplant-associated thrombotic microangiopathy in tandem autologous hematopoietic stem cell transplant for neuroblastoma Anthony Sabulski et al. https://doi.org/10.3324/haematol.2023.283351

3454

Cross-intolerance with bosutinib after prior tyrosine kinase inhibitors for Philadelphia chromosome-positive leukemia: long-term analysis of a phase I/II study Jorge E. Cortes et al. https://doi.org/10.3324/haematol.2022.281944

3460

Efficacy and toxicity of midostaurin with idarubicin and cytarabine induction in FLT3-mutated acute myeloid leukemia Julia S. Lee et al. https://doi.org/10.3324/haematol.2022.281967

Haematologica | 108 - December 2023

III

3464

Latexin deletion protects against radiation-induced hematopoietic damages via selective activation of Bcl-2 prosurvival pathway Cuiping Zhang et al. https://doi.org/10.3324/haematol.2022.282028

3471

ETV6 fusions from insertions of exons 3-5 in pediatric hematologic malignancies Sarah B. Mueller et al. https://doi.org/10.3324/haematol.2022.282498

3477

Modified carfilzomib dosing is associated with improved treatment responses and longer time on treatment in patients with multiple myeloma Abdullah M. Khan et al. https://doi.org/10.3324/haematol.2022.282521

3480

Impact of pinworm infection on the development of murine B-cell leukemia/lymphoma in the presence and absence of ETV6::RUNX1 Briana A. Fitch et al. https://doi.org/10.3324/haematol.2022.282591

3485

CAR T-cell therapy for central nervous system lymphomas: blood and cerebrospinal fluid biology, and outcomes Claire Lacan et al. https://doi.org/10.3324/haematol.2023.282875

3491

CD47 expression in acute myeloid leukemia varies according to genotype Andrea Marra et al. https://doi.org/10.3324/haematol.2023.283154

3496

Co-stimulatory and immune checkpoint molecules are important in the tumor microenvironment of Hodgkin-like adult T-cell leukemia/lymphoma Mai Takeuchi, et al. https://doi.org/10.3324/haematol.2023.283163

3502

Higher cyclophosphamide dose grants optimal stem cell collection after daratumumab-based induction in multiple myeloma Carmine Liberatore et al. https://doi.org/10.3324/haematol.2023.283452

Case Reports & Case Series 3506

Myeloid/lymphoid neoplasms associated with eosinophilia and rearrangements of PCM1::JAK2 with erythroblastic sarcoma: a case report and literature review Lina Zhang et al. https://doi.org/10.3324/haematol.2022.282228

3511

Myeloid lineage switch following CD7-targeted chimeric antigen receptor T-cell therapy in relapsed/refractory T-cell acute lymphoblastic leukemia Ibrahim Aldoss et al. https://doi.org/10.3324/haematol.2023.283566

Haematologica | 108 - December 2023

LANDMARK PAPER IN HEMATOLOGY

T. Bacharach and L.I. Shlush

Clonal evolution of leukemia from G6PD studies Tal Bacharach1 and Liran I. Shlush1,2 1

Department of Molecular Cell Biology, Weizmann Institute of Science, Rehovot and 2Molecular Hematology Clinic Maccabi Healthcare Services, Tel Aviv, Israel E-mail: liranshlush3@gmail.com

TITLE

Clonal origin of chronic myelocytic leukemia in man.

AUTHORS Fialkow PJ, Gartler SM and Yoshida A. JOURNAL Proceedings of the National Academy of Siences 1967;58(4):1468-1471. PMID: 5237880.

It is now well-established that acute myeloid leukemia (AML) emerges following a long evolution involving a preleukemic phase, in which mutations occur in normal hematopoietic stem cells, which clonally expand while maintaining differentiation. The basic evidence for this concept came from a series of studies by Philip J. Fialkow, in which simple and elegant experiments led to groundbreaking conclusions regarding the clonality, cell of origin, and evolution of AML. These works rely on the inference of the active X chromosome in women as a clonal marker. In women, due to X-inactivation, each cell expresses either the paternal or the maternal X chromosome. Thus, a healthy adult tissue consists of a mixture of cells expressing either the paternal or the maternal X, while a clonal leukemia arises from one cell and, therefore, consists entirely of cells expressing either the paternal or the maternal X. In 1967, Fialkow studied 3 chronic myeloid leukemia (CML) patients who were heterozygous for X-linked gene G6PD.1 He measured the G6PD gene product as a clonal marker in patient-derived erythrocytes and granulocytes, as well as skin-derived fibroblasts. While the fibroblasts expressed both G6PD types, the erythrocytes and granulocytes expressed only one type, suggesting they were derived from one cell. From this he concluded that CML is a clonal disease, and suggested that it is derived from a stem/progenitor cell common to both erythrocytes and granulocytes. The same methodology was applied to studies of AML, in which Fialkow demonstrated that AML too is a clonal disease.2 Furthermore, he performed lineage tracing in AML patients by studying the G6PD gene product in various hematopoietic cell types, and established that AML can

originate in two distinct cell types: either a multipotent stem cell or a progenitor of granulocytes and macrophages.2,3 By studying the G6PD gene product in samples from AML remission, he showed two distinct remission types, either non-clonal (suggested to originate in normal stem cells that re-populated the bone marrow after treatment) or clonal remission.2,4 He further linked the cell of origin with remission type in AML. He observed that one form of AML is typical of younger patients, in which the leukemia originates in a granulocyte-macrophage progenitor, and non-clonal remission is obtained from normal stem cells; the second type, typical to older patients, originates in a multipotent stem cell, in which remission is clonal.5 The observation that in 100% of patients with clonal remission, the clone in remission shows the same marker as the clone in diagnosis, proved that clonal remission starts from a pre-leukemic stem cell.6 This led to the conclusion that AML occurs in a multi-step fashion, starting with a preleukemic phase which involves the clonal proliferation of hematopoietic stem cells, followed by a late stage involving the acquisition of 'late' chromosomal changes which give a selective advantage of subclones to evolve to leukemia.6 The important concepts introduced by these experiments are the consequence of scrupulous observation, but even more so, of bold and brilliant interpretation that has revolutionized the understanding of the evolution of AML. Disclosure No conflicts of interest to disclose. Contributions Both authors contributed equally.

Haematologica | 108 December 2023

3191

T. Bacharach and L.I. Shlush

LANDMARK PAPER IN HEMATOLOGY

Figure 1. Main contributions of the works by Fialkow to the study of acute myeloid leukemia (AML) evolution over the years. CML: chronic myeloid leukemia.

References 1. Fialkow PJ, Gartler SM, Yoshida A. Clonal origin of chronic myelocytic leukemia in man. Proc Natl Acad Sci. 1967;58(4):1468-1471. 2. Fialkow PJ, Singer JW, Adamson JW, et al. Acute nonlymphocytic leukemia. N Engl J Med. 1979;301(1):1-5. 3. Fialkow P, Singer JW, Adamson JW, et al. Acute nonlymphocytic leukemia: heterogeneity of stem cell origin. Blood. 1981;57(6):1068-1073. 4. Jacobson RJ, Temple MJ, Singer JW, Raskind W, Powell J,

Fialkow PJ. A clonal complete remission in a patient with acute nonlymphocytic leukemia originating in a multipotent stem cell. N Engl J Med. 1984;310(23):1513-1517. 5. Fialkow PJ, Singer JW, Raskind WH, et al. Clonal development, stem-cell differentiation, and clinical remissions in acute nonlymphocytic leukemia. N Engl J Med. 1987;317(8):468-473. 6. Fialkow P, Janssen J, Bartram C. Clonal remissions in acute nonlymphocytic leukemia: evidence for a multistep pathogenesis of the malignancy. Blood. 1991;77(7):1415-1417.

Haematologica | 108 December 2023

3192

EDITORIAL

S. Rives

Central nervous system therapy in acute lymphoblastic leukemia: no more, no less Susana Rives Hospital Sant Joan de Déu de Barcelona and Institut de Recerca Sant Joan de Déu, Barcelona, Spain

Correspondence: S. Rives susana.rives@sjd.es Received: Accepted: Early view:

May 31, 2023. June 7, 2023. June 15, 2023.

Central nervous system (CNS) infiltration at diagnosis remains a poor prognostic factor in acute lymphoblastic leukemia (ALL) with over 30% of relapses involving CNS. Treatment with antileukemic agents against CNS leukemia is an essential component for cure in ALL. With the intensification of CNS-directed therapy, cranial radiotherapy (CRT) has been completely omitted in current pediatric ALL protocols, or limited to a minority of high-risk patients, without an increase in relapse rates.1 Avoiding CRT is an important step forward due to its long-term adverse effects, such as endocrinopathies, neurocognitive deficits, and second cancers. Nevertheless, contemporary intensive systemic and intrathecal (IT) chemotherapy is also associated to short- and long-term neurotoxicity and it seems to be a threshold dose effect for the association between IT chemotherapy exposure and cognitive impairment, particularly in younger children.2 The intensity of conventional chemotherapy has reached its limit in current therapeutic protocols, with the rate of toxic deaths approaching that of deaths from relapse in pediatric patients with ALL.1 Concerns about acute and delayed toxicity in children treated with intensive treatment has changed the focus in new protocols of therapy aiming to avoid undertreatment, but also overtreatment. Possible overtreatment of patients with low levels of leukemic CNS involvement (CNS2) is analyzed by Heilmann and colleagues in this issue of Haematologica.3 They assessed the effect of three versus five doses of IT methotrexate (MTX) in induction therapy on systemic toxicity in children and adolescents with ALL aged 1-17 years. In a retrospective analysis of 6,136 patients enrolled in the AIEOP-BFM 2009 clinical trial, the authors found that the addition of two extra doses of IT MTX in patients with initial CNS involvement (CNS2 and CNS3) was associated with a significant increase in life-threatening and fatal infections. Patients with CNS2 or CNS3 involvement who received five IT MTX showed an incidence of life-threatening and fatal infections of 4.4% and 1.6%, respectively, versus 1.6% and 0.3%, respectively, in patients receiving three

doses. The authors attributed this increase in infections to the systemic effect of IT MTX. Although the group of patients with CNS involvement also had other factors associated with an increased risk for infection (such as age > 10 years or dexamethasone therapy in induction), having received these two additional doses of IT MTX was the strongest risk factor for severe infections in the multivariate analysis (Odds Ratio [HR]: 2.85; 95% Confidence Interval [CI]: 1.96-4.14; P<0.001). CNS3 status was associated with greater risk for relapse in B-cell precursor (BCP) ALL (HR: 1.59; 95% CI: 0.85-3.0; P=0.15) and in T-ALL (HR: 2.65; 95% CI: 1.56-4.51; P<0.001), but CNS2 status was not. Based on the severe adverse events associated with the intensification of IT therapy in patients with CNS involvement at diagnosis, and the unclear relevance of CNS2 for relapse risk, the number of intrathecal doses has been reduced in patients with CNS2 involvement in the current AIEOP-BFM ALL 2017 treatment protocol. Frank CNS infiltration (CNS3) is an adverse prognostic factor in pediatric ALL patients. However, the impact on outcome of CNS2 is controversial and varies between different treatment groups and protocols, and according to ALL immunophenotype. The uncertain value of CNS2 may be due to several reasons. First, there are differences in systemic and CNS-directed treatment, which includes cranial radiotherapy in high-risk patients in some protocols that can modify, and even abrogate, the prognostic impact of low levels of CNS leukemia. Thus, in some clinical trials in which CNS-directed therapy was not modified in CNS2 patients, CNS2 status had an adverse prognostic impact.4 In this regard, the Children's Oncology Group (COG) published their results from three clinical trials in which patients with BCP-ALL with CNS2 involvement had higher CNS relapse rate and lower 5-year event-free survival (85% vs. 76%; P<0.001).5 In these treatment protocols, patients with CNS2 did not receive extra doses of IT therapy. However, in other clinical trials, in which patients with CNS2 received additional doses of IT chemotherapy during induction, CNS2 lost its prognostic significance6,7 or only

Haematologica | 108 December 2023

3193

EDITORIAL

S. Rives

retained its significance in T-ALL.1 Of note, CNS2 status did not have any adverse prognostic impact in patients treated with the AIEOP-BFM ALL 2009 protocol, as reported by Heilmann et al. in this issue of Haematologica.3 However, as these patients did receive two additional intrathecal doses, we cannot rule out the possibility that their omission might increase the risk of relapse. Second, there might be differences in the prognostic value of CNS involvement across different biologic subgroups of ALL, in which the modification in therapy may have a different impact.1 Indeed, unlike BCP ALL, CNS2 status did not have any impact on outcome in T-ALL patients treated according to augmented BFM therapy in the COG AALL0434 and AALL1231 clinical trials.8 Finally, the incidence of CNS2 status varies widely between groups (from 2.4% to 42%), probably due to pre-analytical and analytical factors rather than to true clinical differences.9 The way of stratifying CNS, based on leukocyte and red blood cell counts in the chamber together with a cytologic study of the cyto-centrifuged CSF sample has low

sensitivity, specificity, and reproducibility. This heterogeneity makes it difficult to compare the prognostic value of CNS2 status across trials. Clearly, better methods are required to assess the degree of CNS infiltration at diagnosis, as well as biomarkers to measure response to treatment, as with MRD assessment in systemic disease. In this regard, CSF flow cytometry may achieve greater sensitivity and predictive prognostic value, as recently demonstrated by the Nordic Society of Paediatric Haematology and Oncology (NOPHO) group.10 The study by Heilmann and colleagues identifies important issues in the management of ALL patients, such as the need to reduce toxicity and improve stratification of CNS leukemia. Better biomarkers of CNS disease that allow administration of the right dose to prevent relapse are required to avoid overtreatment and toxicity. No more, no less. Disclosures No conflicts of interest to disclose.

References 1. Jeha S, Pei D, Choi J, et al. Improved CNS control of childhood acute lymphoblastic leukemia without cranial irradiation: St Jude Total Therapy Study 16. J Clin Oncol. 2019;37(35):3377-3391. 2. Jacola LM, Conklin HM, Krull KR, et al. The impact of intensified CNS-directed therapy on neurocognitive outcomes in survivors of childhood acute lymphoblastic leukemia treated without cranial irradiation. J Clin Oncol. 2022;40(36):4218-4227. 3. Heilmann J, Vieth S, Möricke A, et al. Effect of two additional doses of intrathecal methotrexate during induction therapy on serious infectious toxicity in pediatric patients with acute lymphoblastic leukemia. Haematologica. 2023;108(12):3278-3286. 4. Mahmoud HH, Rivera GK, Hancock ML, et al. Low leukocyte counts with blast cells in cerebrospinal fluid of children with newly diagnosed acute lymphoblastic leukemia. N Engl J Med. 1993;329(5):314-319. 5. Winick N, Devidas M, Chen S, et al. Impact of initial CSF findings on outcome among patients with National Cancer Institute standard- and high-risk B-cell acute lymphoblastic leukemia: a report from the Children's Oncology Group. J Clin Oncol. 2017;35(22):2527-2534. 6. Möricke A, Reiter A, Zimmermann M, et al. Risk-adjusted therapy of acute lymphoblastic leukemia can decrease treatment burden

and improve survival: treatment results of 2169 unselected pediatric and adolescent patients enrolled in the trial ALL-BFM 95. Blood. 2008;111(9):4477-4489. 7. Vrooman LM, Stevenson KE, Supko JG, et al. Postinduction dexamethasone and individualized dosing of Escherichia Coli L-asparaginase each improve outcome of children and adolescents with newly diagnosed acute lymphoblastic leukemia: results from a randomized study--Dana-Farber Cancer Institute ALL Consortium Protocol 00-01. J Clin Oncol. 2013;31(9):1202-1210. 8. Gossai NP, Devidas M, Chen Z, et al. Central nervous system status is prognostic in T-cell acute lymphoblastic leukemia: a Children's Oncology Group report. Blood. 2023;141(15):1802-1811. 9. Thastrup M, Duguid A, Mirian C, Schmiegelow K, Halsey C. Central nervous system involvement in childhood acute lymphoblastic leukemia: challenges and solutions. Leukemia. 2022;36(12):2751-2768. 10. Thastrup M, Marquart HV, Levinsen M, et al. Flow cytometric detection of leukemic blasts in cerebrospinal fluid predicts risk of relapse in childhood acute lymphoblastic leukemia: a Nordic Society of Pediatric Hematology and Oncology study. Leukemia. 2020;34(2):336-346.

Haematologica | 108 December 2023

3194

EDITORIAL

M.K. Gandhi and C. Keane

Expanding the bacterial origins of nodular lymphocytepredominant Hodgkin lymphoma Maher K. Gandhi1,2 and Colm Keane2,3 1

Mater Research, University of Queensland; 2Princess Alexandra Hospital and 3Frazer Institute, University of Queensland, Brisbane, Queensland, Australia

Correspondence: C. Keane c.keane@uq.edu.au Received: Accepted: Early view:

May 31, 2023. June 20, 2023. June 29, 2023.

In the current issue of Haematologica, Thurner et al. provide further insights into the growing knowledge base on specific tumor cell reactivity in the unique disease of nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL).1 This HL differs from classical variants by the fact that the malignant cell clearly retains its B-cell identity. It expresses functional B-cell receptors (BCR), is surrounded by rosettes of T-follicular helper cells, retains CD20 expression, and has no CD30 expression. It is not believed to have any association with Epstein-Barr virus and is genetically distinct from classical HL.2 Clinically, it has distinguishing features from classical HL with a very strong male preponderance and, in most cases, a limited stage presentation. In recognition of the distinct pathologic, biologic, and clinical differences, in the 2022 International Consensus Classification of Mature Lymphoid Neoplasms (2022 ICC) it has now been renamed nodular lymphocyte predominant B-cell lymphoma (NLPBL).3 (It should be noted that it retains its NLPHL nomenclature within the 2022 WHO classification.4) It is the most indolent of HL, with population-based studies indicating only a slightly higher lymphoma-related death rate compared to death from other causes.5 Its relative rarity has hindered our understanding of the disease. The optimal treatment strategies are still unclear, but in advanced disease rituximab is generally added to systemic chemotherapy given the expression of CD20 on the malignant cells. Radiation therapy with or without chemotherapy is very effective for the more common limited-stage presentation.6 Its distinct presentation and phenotype had led to some suggesting that perhaps this lymphoma could be antigendriven. In addition, the very strong familial risk may indicate that specific environmental factors could be responsible.7 Previous work has demonstrated antigen specificity for Moraxella catarrhalis with specific IGVH genotypes and associated HLA class II haplotypes.8,9 These HLA associations showed specific T-follicular helper cell reactivity and direct immune synapses between these

supporting T cells and the malignant B cells, further enhancing the hypothesis that antigenic stimulation supports malignant B-cell propagation. The current study expands on this work to identify additional antigens that are associated with NLPHL.1 Due to frequent presentation of NLPHL with cervical lymphadenopathy, the authors focused on several bacteria commonly found in the oral cavity, upper respiratory tract, and gut, such as Rothia spp, Enterococcus spp, and Lactobacillus spp. With this expanded bacterial screening of NLPHL cases, two antigens of R. mucilaginosa (a common oral commensal that commonly resides in cervical lymph node drainage areas) were identified as targets of NLPHL BCR, and, like M. Catarrhalis RpoC, light chain-restricted antibodies to R. mucilaginosa Gltf and 2,3-BDH were found in the sera of patients with the disease, although at low concentration. Screening of patients with classical HL and primary mediastinal B-cell lymphoma (PMBCL), and T-cell/histiocyte-rich large B-cell lymphoma found no similar antibodies in the sera of any of these patients. The authors also confirmed that R. mucilaginosa Gltf and 2,3-BDH induced growth by activating the BCR pathway. It should be noted that, unlike M. catarrhalis cases, no HLA restriction was found in cases of R. mucilaginosa, and only one of the 5 cases showed IgD positivity. In newly presenting patients with NLPHL, the authors also identified 3 further cases showing reactivity to Moraxella spp. Thus, 10 out of 22 of analyzed NLPHL to date have reactivity to this same bacterium, which confirms the findings of a previous work.9 As is the case for most publications on NLPHL, the study by Thurner et al. is characterized by a relatively low number of cases but, intriguingly, 15 of 22 cases showed reactivity to bacterial species.1 It is possible that a wider screen of bacteria, and even viruses, may lead to the identification of additional cases that might be related to chronic antigenic stimulation. As stated by Thurner et al., screening of other lymphomas such as marginal zone lymphoma or duodenal follicular

Haematologica | 108 December 2023

3195

EDITORIAL

M.K. Gandhi and C. Keane

lymphoma could lead to the identification of other causative or supportive organisms. The description of antigenic stimulation and the indolent and sometimes waxing/waning course of NLPHL does, indeed, warrant investigation of new theraputic strategies that could include vaccination or treatment with antibiotics. The rarity of the disease makes trials of such approaches difficult, but perhaps the indolent nature of the disease could allow novel interventions to be tested. A priority for this disease is the development of better pre-clinical models and cell lines that more accurately mimic the disease, given the current dependence for functional work on the DEV cell line when it has lost expression of MHC II, which contrasts with the almost universal expression found in clinical cases. Better understanding of the microbiome indicates that we may only be touching the surface of understanding how bacteria influence, not only tumor development, but also response to therapy. Future work on the microbiome and gut microflora may provide further insights into rare lymphoma subtypes and their response to therapy. It is likely

that an international collaboration or consortium will be required to investigate these key questions in sufficient numbers of patients. In summary, Thurner et al. should be commended for their work on this rare lymphoma and for the fascinating description of a second bacterium that is likely a key factor in the disease pathogenesis. Almost 70% of tested cases demonstrated bacterial reactivity,1 which provides a launching pad not only for exploration of further potential causative organisms, but that also encourages the development of novel agents and approaches for this unique disease. Disclosures CK has sat on advisory boards, and received consulting/speaker fees for Roche, Janssen, Astra Zeneca, MSD Takeda and Beigene. MKG has received supplies of study drugs for investigator-led clinical trials from Beigene and Janssen. Contributions MKG and CK both wrote the manuscript.

References 1. Thurner L, Fadle N, Regitz E, et al. B-cell receptor reactivity against Rothia mucilaginosa in nodular lymphocytepredominant Hodgkin lymphoma. Haematologica. 2023;108(12):3347-3358. 2. Hartmann S, Schuhmacher B, Rausch T, et al. Highly recurrent mutations of SGK1, DUSP2 and JUNB in nodular lymphocyte predominant Hodgkin lymphoma. Leukemia. 2016;30(4):844-853. 3. Tousseyn TA, King RL, Fend F, Feldman AL, Brousset P, Jaffe ES. Evolution in the definition and diagnosis of the Hodgkin lymphomas and related entities. Virchows Arch. 2023;482(1):207-226. 4. Alaggio R, Amador C, Anagnostopoulos I, et al. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: Lymphoid Neoplasms. Leukemia. 2022;36(7):1720-1748. 5. Posthuma HLA, Zijlstra JM, Visser O, Lugtenburg PJ, Kersten MJ, Dinmohamed AG. Primary therapy and survival among patients with nodular lymphocyte-predominant Hodgkin lymphoma: a

population-based analysis in the Netherlands, 1993-2016. Br J Haematol. 2020;189(1):117-121. 6. Eichenauer DA, Engert A. How I treat nodular lymphocytepredominant Hodgkin lymphoma. Blood. 2020;136(26):2987-2993. 7. Saarinen S, Pukkala E, Vahteristo P, Makinen MJ, Franssila K, Aaltonen LA. High familial risk in nodular lymphocytepredominant Hodgkin lymphoma. J Clin Oncol. 2013;31(7):938-943. 8. Bein J, Thurner L, Hansmann ML, Hartmann S. Lymphocyte predominant cells of nodular lymphocyte predominant Hodgkin lymphoma interact with rosetting T cells in an immunological synapse. Am J Hematol. 2020;95(12):1495-1502. 9. Thurner L, Hartmann S, Fadle N, et al. Lymphocyte predominant cells detect Moraxella catarrhalis-derived antigens in nodular lymphocyte-predominant Hodgkin lymphoma. Nat Commun. 2020;11(1):2465.

Haematologica | 108 December 2023

3196

EDITORIAL

R. Colombatti and M. Casale

Early splenectomy in sickle cell disease: another piece of the puzzle Raffaella Colombatti1 and Maddalena Casale2 1

Pediatric Hematology-Oncology Unit, Department of Women’s and Child’s Health, University of Padova, Padova and 2Università della Campania, Luigi Vanvitelli, Napoli, Italy

Correspondence: R. Colombatti raffaella.colombatti@unipd.it Received: Accepted: Early view:

June 20, 2023. July 3, 2023. July 13, 2023.

Splenic involvement occurs very early in life in sickle cell anemia (SCA), which includes the SS and Sβ° genotypes, and later on in the SC and Sβ+ forms of sickle cell disease (SCD). Slow blood flow and open microcirculation determine deoxygenation and sickling in the splenic vascular bed starting at six months of age.1,2 Multiple mechanisms have been hypothesized to contribute to the damage in all the three histological areas (red pulp, white pulp, marginal zone) and to functional asplenia. Clinically, the main splenic complications are acute splenic sequestration (ASS) and splenomegaly, with or without hypersplenism. ASS is a frequent life-threatening cause of morbidity in SCA with a need for acute admissions and emergency red blood cell (RBC) transfusions; it also remains a significant cause of death under five years of age in recent newborn cohorts3,4 in spite of early diagnosis of SCD and educational initiatives for parents. In France, among 4,682 children with SCD diagnosed through newborn screening and followed for 15 years in reference centers, 9 deaths occurred due to ASS (7 in SS/ Sβ° and 2 in Sβ+); in Spain, data from the national registry show that three deaths occurred due to ASS, all in very young children (1 year of age). While the prevalence of ASS is between 10-25%, the recurrence rate after one episode of ASS can be as high as 67%, especially in younger children experiencing the first episode before one year of age,5 with higher mortality rates after recurrence. Hypersplenism is the other main splenic complication in SCA and a possible indication for splenectomy due to high transfusion burden.6 The risk of pneumococcal invasive infection, especially in the short term after splenectomy, is one of the frequent and severe causes of death in high-income countries, and this has hampered the utilization of this procedure before the age of five years as a therapeutic measure for ASS.6 The treatment of ASS is mainly supportive, while prevention strategies after a first episode utilizing RBC transfusion or hydroxyurea (HU) have limited efficacy. Hence the specific need to identify early splenectomy in SCD as a safe possible treatment and preventive strategy in order

to avoid the high risk of recurrence and death in very young children.5-8 In this issue of Haematologica, Mechraoui et al.9 present the results of an observational retrospective study investigating risks and benefits of early splenectomy in a large cohort of 1,167 SCA children followed from birth and diagnosed through neonatal screening in France. Among the 188 splenectomized patients, 123 (65.4%) patients underwent splenectomy after three years of age and 65 (34.6%) were under three years of age at splenectomy. The control group included the non-splenectomized patients of the newborn cohort. The authors provide an answer to one of the challenging questions in the management of children with SCA: at what age is it safe to perform splenectomy? Does the benefit of splenectomy in managing life-threatening complications in very young children overcome the potential risks? They demonstrate that, in the case of clinical indication, splenectomy can be safely performed as early as three years of age if the immunization schedule is complete and the antibiotic prophylaxis is continued with compliance. In fact, in their large cohort, with splenectomy performed at a median age of 4.1 years (range 2.5-7.2), with a high rate of full immunization and adherence to antibiotic prophylaxis continued until the age of ten, none of the children experienced life-threatening infections by capsulated bacteria during the follow-up period (mean 5.9 years; range 2.7-9.2), and the overall incidence of invasive bacterial infections and thromboembolic events was low (0.005/patient years [PY] and 0.003/PY, respectively). Data from other high-income settings had shown worse outcomes, with death from severe infections and sepsis in the first year after splenectomy, even in children who had been splenectomized at an older age. Neither occurred in this cohort. The high rate of full immunization with the non-conjugate and conjugate vaccines, as well as the high compliance with penicillin administered until ten years of age, can surely have eliminated the infection

Haematologica | 108 December 2023

3197

EDITORIAL

R. Colombatti and M. Casale

risk due to capsulated bacteria in this cohort, where only six cases of invasive infection occurred, but none from encapsulated bacteria. Continuation of prophylaxis for those children over five years of age, and even for their entire lifetime, is more common in European settings than in the United States, and can certainly be protective. The observation time after splenectomy in Mechraoui et al.’s cohort is still in the medium term (5.9 years of follow-up) with a median age at the time of study of 14.4 years; therefore, a longer follow-up is needed to confirm the low risk of infection throughout the patient's lifespan, even if splenectomy is performed as early as three years of age. Furthermore, the authors take into consideration the rate of occurrence of post-splenectomy sickle cell-related complications in relationship to HU treatment. It is noteworthy that whereas vaso-occlusive events (VOE) were significantly lower in splenectomized patients who were not treated with HU, they were significantly higher in splenectomized patients who underwent HU treatment, in contrast to patients who were not splenectomized. Furthermore, there has been a growing trend in patients initiating HU due to clinical complications after splenectomy. Cerebrovascular events were more prevalent in splenectomized patients and, specifically, in those splenectomized before three years of age. These findings raise a further question: does splenectomy increase the risk of sickle-related complications? Is the need for splenectomy in SCD simply an epiphenomenon of a more severe hemolytic disease? Or does splenectomy itself trigger or worsen the sickle crisis and vascular damage? This question is especially relevant for infants, who only have limited treatment options10 and, therefore, the evaluation of the risk-benefit ratio of splenectomy requires not only the assessment of splenectomy-related complications, but also of the potential risk of disease progression and onset of sickle-related complications. The

prevalence of VOE after splenectomy varied across studies, but the link between the surgical procedure and VOE remains elusive.5-8 Mechraoui et al. report a worrying higher rate of cerebrovascular events (abnormal transcranial Doppler, stenosis on magnetic resonance angiography or overt stroke) post splenectomy in HU-treated patients and in children splenectomized before the age of three years. The authors' suggested hypothesis for this finding is that those children could be the ones with the most severe clinical phenotype and that spleen complications might serve as a marker of disease severity. However, the higher rate of cerebrovascular events post splenectomy in children who were symptomatic enough to warrant initiation of HU therapy post splenectomy also points towards the intriguing consideration of a link between the brain and the spleen with the involvement of the brain-spleen axis in SCA, as in inflammatory and cerebral disorders.11 Further research on the anatomical neural circuits, and soluble inflammatory and immune mediators of the brain-spleen axis in SCA could enhance the understanding of severe phenotypes early in life and the risk of SCA-related complications post splenectomy. In conclusion, Machroui et al. provide us with a new, important piece of the puzzle: if indicated, splenectomy should not be delayed until five years of age as long as there is good immunization and prophylactic coverage. Deeper understanding of the correlation between splenectomy and sickle-related complications will help clinicians plan comprehensive management of the disease and its evolution. Disclosures No conflicts of interests to disclose. Contributions RC and MC wrote the manuscript and reviewed it.

References 1. Brousse V, Buffet P, Rees D. The spleen and sickle cell disease: the sick(led) spleen. Br J Haematol. 2014;166(2):165-176. 2. Pizzi M, Fuligni F, Santoro L, et al. Spleen histology in children with sickle cell disease and hereditary spherocytosis: hints on the disease pathophysiology. Hum Pathol. 2017;60:95-103. 3. Desselas E, Thuret I, Kaguelidou F, et al. Mortality in children with sickle cell disease in mainland France from 2000 to 2015. Haematologica. 2020;105(9):e440-443. 4. Cela E, Bellón JM, de la Cruz M, et al; SEHOPHemoglobinopathies Study Group (Sociedad Española de Hematología y Oncología Pediátricas). National registry of hemoglobinopathies in Spain (REPHem). Pediatr Blood Cancer. 2017;64(7). 5. Brousse V, Elie C, Benkerrou M, et al. Acute splenic sequestration crisis in sickle cell disease: cohort study of 190 paediatric patients. Br J Haematol. 2012;156(5):643-648. 6. Iolascon A, Andolfo I, Barcellini W, et al; Working Study Group on

Red Cells and Iron of the EHA. Recommendations regarding splenectomy in hereditary hemolytic anemias. Haematologica. 2017;102(8):1304-1313. 7. Hall BJ, Reiter AJ, Englum BR, et al. Long-term hematologic and clinical outcomes of splenectomy in children with hereditary spherocytosis and sickle cell diseae. Pediatr Blood Cancer. 2020;67(8):e28290. 8. Machado NO, Grant CS, Alkindi S, et al. Splenectomy for haematological disorders: a single center study in 150 patients from Oman. Int J Surg. 2009;7(5):476-481. 9. Mechraoui A, Ithier G, Pages J, et al. Early splenectomy in a large cohort of children with sickle cell anaemia: risks and consequences. Haematologica. 2023;108(12):3409-3417. 10. Inusa BPD, Casale M, Campbell A, Archer N. Will the changing therapeutic landscape meet the needs of patients with sickle cell disease? Lancet Haematol. 2021;8(5):e306-e307.

Haematologica | 108 December 2023

3198

EDITORIAL

A. Bacigalupo

Long-term survivors in severe aplastic anemia: standard mortality rates now comparable to controls? Andrea Bacigalupo

Correspondence: A. Bacigalupo andrea.bacigalupo@unicatt.it

Fondazione Policlinico Universitario “A. Gemelli” IRCCS, Roma, Italy

Received: Accepted: Early view:

June 28, 2023. July 4, 2023. July 13, 2023.

In the current issue of Haematologica, Nakamura and coworkers report standard mortality rates (SMR) for patients with severe aplastic anemia (SAA) alive one year after receiving either immunosuppressive therapy (IST) or a hematopoietic stem cell transplant (HSCT).1 SMR compare the mortality risk of SAA patients with the general population, adjusted for age, gender, and race/ethnicity. The bottom line is that 1-year survivors after IST or HSCT can expect to live beyond five or ten years, becoming comparable to that of the general US population. Severe aplastic anemia is a rare hematologic disorder, which was, in the 1970s, lethal in most cases.2 The outcome has improved considerably over the past decades with the widespread use of IST and HSCT. What remains to be ascertained is the long-term outcome of these therapeutic procedures, particularly since both have long-term complications: 20% cumulative incidence of tumors has been reported 12 years after IST and 5% after HSCT.3 Chronic graft-versus-host disease (cGvHD) after HSCT and persistent transfusion requirement/relapse after IST are also well-known complications, together with non-lethal issues such as osteonecrosis, fertility problems, cataracts, hypothyroidism. For these reasons, the long-term outcome of SAA patients has been regarded as never reaching the contemporaneous control curves. There are several findings which I find surprising in the study of Nakamura and co-workers:1 first, the fact that SMR normalized after a given time post treatment; second, that this is true both for IST as well as for HSCT; and third, that alternative donor transplants seem to have similar SMR to matched sibling transplants. As to the gradual reduction of the SMR in 5- and 10-year survivors, this tells us that patients are still at a higher risk of lethal complications between one and five years, including failed response for IST and cGvHD for HSCT;

however, it also suggests that patients who survive five years, especially over the last decade, can be considered to be cured of their disease. So, a combination of bad and good news. Treatment with IST has never been regarded as curative, and the term “postponing the inevitable” was defined many years ago,4 being the inevitable clonal evolution into myelodysplasia or overt leukemia. This study now suggests that patients surviving five years following IST can expect a similar survival to the control population, although a proportion (20%) of IST patients did receive an allograft for incomplete response. Finally, we learn that alternative donor transplants have similar long-term outcome, in terms of SMR at 5-10 years, to that of matched sibling grafts. This is an important confirmation of comparable survival,5 but also confirms that for patients who lack an HLA matched sibling, an unrelated transplant should be considered early in the course of the disease. A very recent analysis of the SAA Working Party of the European Group for Blood and Marrow Transplantation (EBMT) has shown rather convincingly that patients allografted upfront from an HLA matched related donor have a significantly better outcome than patients allografted after having failed IST;6 this is true for patients under the age of 40, and also for subgroups of patients over the age of 40 years. The Authors admit that more patients need to be studied, and with longer follow-up, especially for more recent patients, and that causes of death in the first five years after treatment need to be carefully evaluated. Nevertheless, a gradual reduction of SMR to approximately that of the general population is, for a rare, once lethal disease, a great achievement of the scientific community. Disclosures No conflicts of interest to disclose.

Haematologica | 108 December 2023

3199

EDITORIAL

A. Bacigalupo

References 1. Nakamura R, Patel BA, Kim B, et al. Conditional survival and standardized mortality ratios of patients with severe aplastic anemia surviving at least one year after hematopoietic cell transplantation or immunosuppressive therapy. Haematologica. 2023;108(12):3298-3307. 2. Young NS. Aplastic anemia. N Engl J Med. 2018;379(17):1643-1656. 3. Socié G, Henry-Amar M, Bacigalupo A, et al. Malignant tumors occurring after treatment of aplastic anemia. European Bone Marrow Transplantation-Severe Aplastic Anaemia Working Party.

N Engl J Med. 1993;329(16):1152-1157. 4. Moore MA, Castro-Malaspina H. Immunosuppression in aplastic anemia--postponing the inevitable? N Engl J Med. 1991;324(19):1358-1360. 5. Bacigalupo A, Socié G, Hamladji RM, et al. Aplastic Anemia Working Party of the European Group for Blood Marrow Transplantation. Current outcome of HLA identical sibling versus unrelated donor transplants in severe aplastic anemia: an EBMT analysis. Haematologica. 2015;100(5):696-702.

Haematologica | 108 December 2023

3200

EDITORIAL

A.A. Pandyra and A. Borkhardt

Worms on stage Aleksandra A. Pandyra and Arndt Borkhardt Department of Pediatric Oncology, Hematology and Clinical Immunology, Medical Faculty, Center of Child and Adolescent Health, Heinrich-Heine-University, Düsseldorf, Germany

Correspondence: A. Borkhardt Arndt.Borkhardt@med.uni-duesseldorf.de Received: Accepted: Early view:

June 29, 2023. July 5, 2023. July 13, 2023.

The eloquent quote by pencillin-discoverer Alexander Fleming stating that 'the unprepapred mind cannot see the outstretched hand of opportunity’ resonates with a paper published in the current issue of Haematologica in which the authors clearly saw the opportunity to report on their serendipitous discovery stemming from an initially frustrating lack of reproducibility.1 In 2013, Scott Kogan’s group assessed the interplay between ETV6::Runx1 (formerly TEL-AML1) inducible transgene expressed in hematopoietic cells (E6R1+) and Cdkn2a loss (whole body knockout, Cdkn2a-/-) in the context of B-cell lymphoblastic lymphoma/leukemia development. A clear oncogenic cooperation was observed between Cdkn2a null mice that expressed ETV6::Runx1 (E6R1+ Cdkn2a-/-) compared to Cdkn2a-/- controls, an effect that was further exacerbated by irradiation.2 However, following the 2013 publication, attempts to recapitulate the main phenotype of the E6R1+ Cdkn2a-/- mice (development of B-cell neoplasms faster and with a higher incidence compared to Cdkn2a-/- controls) failed. A review of the Specific-Pathogen-Free (SPF) facility records revealed that there was an outbreak of the pinworm Aspicularis during the time of the 2013 study. While such an infection is certainly not uncommon in animal facilities, what is commendable is that the authors then sought to systematically unroot the cause of the discrepancy in the reproducibility of their research and subsequently report on the new discovery that the pinworm infection was likely responsible for the initially observed differences from the 2013 publication. Surprisingly, after medicinal eradication of the pinworm outbreak, the differences in leukemia latency and incidence between the double hit E6R1+Cdkn2a-/- model and the single hit Cdkn2a-/- model disappeared. The authors concluded that pinworm infection was protective in the Cdkn2a-/- model but not in the E6R1+Cdkn2a-/- one. Indeed, as described in the current issue of Haematologica,1 when E6R1+Cdkn2a-/- and control Cdkn2a-/- mice were prospectively transferred at 4 weeks of age from the SPF facility to a conventional facility (CF) infected with Aspicularis, the disease phenotype was restored between E6R1+Cdkn2a-/and control Cdkn2a-/- mice (Cdkn2a-/--pinworm vs. E6R1+Cdkn2a-/--pinworm). As before in the 2013 publication,

pinworm infection significantly delayed leukemia/lymphoma development in the single leukemogenic hit Cdkn2a-/- model compared to the two-hit E6R1+Cdkn2a-/model (Figure 1A). Interestingly, when the authors combined their cohorts to assess genotype-environmental (Cdkn2a-/--SPF vs. Cdkn2a-/--pinworm, E6R1+Cdkn2a-/--SPF vs. E6R1+Cdkn2a-/--pinworm) comparisons, there were no lymphoma/leukemia-associated survival advantages in the pinworm-CF. Rather, the median latency in pinworm-infected E6R1+Cdkn2a-/- mice was slightly shorter (231 vs. 253 days) than in SPF-housed controls. It might be expected that the normally silent ETV6::Runx1 predisposition becomes detrimental when combined with another cancerinducing genetic alternation and that the leukemogenic effects of this ‘double hit’ model were too potent to overcome even when combined with the “protective pinworm intervention”. However, the lack of pinworm protective effects in the Cdkn2a-/--SPF vs. Cdkn2a-/--pinworm comparison is surprising. However, this cohort was aggregated from two different experimental settings. In one cohort (20092011), the mice were likely continuously infected with pinworm from the developmental, in utero stage to adulthood, while in the second cohort, mice were transferred 4 weeks after birth (when immune competence has already been largely established) into a pinworm-prevalent CF. These intriguing observations raise several important questions about delayed infection in an untrained immune system and its contribution to the development of B-cell precursor acute lymphoblastic leukemia.3,4 Infection of mice with other helminths has been shown to affect anti-viral and bacterial immunity and to protect mice from inflammation.5 Therefore, this kind of bystander infection has not only experimental consequences but potential disease-specific translational relevance. Indeed, in adults, helminth infections have been retrospectively demonstrated to protect against several inflammatory disease states and to be prospectively effective as vaccines in inflammatory bowel diseases.6 Immunologically, this has been linked to their induction of Th2like regulatory responses. In children, helminth exposure has been proposed to have protective effects in unique isolated settings7 and intestinal nematodes led to decreases in

Haematologica | 108 December 2023

3201

EDITORIAL

A.A. Pandyra and A. Borkhardt

Figure 1. Summary of significant findings and future directions regarding pinworm infection and leukemia models. (A) Pinworm infection is protective against leukemia/lymphoma in the single hit Cdkn2a-/- model when compared to the double hit E6R1+Cdkn2a-/- model. (B) Potential future avenues of investigation to further delineate the mechanisms governing protective effects of pinworm infection in leukemia models. ns: difference not statistically significant.

markers of intestinal inflammation in some studies.8 However, due to the risk of heavier parasite burdens and other associated comorbidities, disease-specific preventative strategies in children will likely favor probiotic interventions rather than helminth infection by intention. Given the strong disease penetrance of the above models, bulk differences could be directly relevant to the observed phenotype. Serial sample stool and blood collection in potential future investigations would enable a thorough characterization of multiplexed cytokine, microbiome and immunological profiles including immunophenotyping of key immune populations (T cells, monocytes, B cells, NK cells) for exhaustion/activation/immunosuppressive markers (Figure 1B). Another important question is whether Th2-like responses could contribute to early immune surveillance and eradication of the pre-leukemic clone.

Helminth infections induce central trained immunity responses,9 which are known to delay tumor progression.4,10 Therefore, the mechanisms governing trained immunity responses using the above leukemia predisposition models following infection with pinworms and other trained immunity inducers should be further explored. Taken together, the authors’ serendipitous discovery provides support for further exploring the hypothesis that immune training influences the development of B-cell acute lymphoblastic leukemia. Disclosures No conflicts of interest to disclose. Contributions AB and AP contributed equally to this work.

Haematologica | 108 December 2023

3202

EDITORIAL

A.A. Pandyra and A. Borkhardt

References 1. Fitch BA, Situ J, Wiemels JL, Kogan SC, Zhou M. Impact of pinworm infection on the development of murine B-cell leukemia/lymphoma in the presence and absence of ETV6::RUNX1. Haematologica. 2023;108(12):3480-3484. 2. Li M, Jones L, Gaillard C, et al. Initially disadvantaged, TELAML1 cells expand and initiate leukemia in response to irradiation and cooperating mutations. Leukemia. 2013;27(7):1570-1573. 3. Greaves MF. Speculations on the cause of childhood acute lymphoblastic leukemia. Leukemia. 1988;2(2):120-125. 4. Hauer J, Fischer U, Borkhardt A. Toward prevention of childhood ALL by early-life immune training. Blood. 2021; 138(16):1412-1428. 5. Tao L, Reese TA. Making mouse models that reflect human immune responses. Trends Immunol. 2017;38(3):181-193. 6. Summers RW, Elliott DE, Urban JF Jr, Thompson R, Weinstock

JV. Trichuris suis therapy in Crohn's disease. Gut. 2005;54(1):87-90. 7. Soloski MJ, Poulain M, Pes GM. Does the trained immune system play an important role in the extreme longevity that is seen in the Sardinian blue zone? Front Aging. 2022;3:1069415. 8. Gildner TE, Schrock J, et al. Soil-transmitted helminth infection and intestinal inflammation among the Shuar of Amazonian Ecuador. Am J Phys Anthropol. 2019;170(1):65-74. 9. Cunningham KT, Finlay CM, Mills KHG. Helminth imprinting of hematopoietic stem cells sustains anti-inflammatory trained innate immunity that attenuates autoimmune disease. J Immunol. 2021;206(7):1618-1630. 10. Kalafati L, Kourtzelis I, Schulte-Schrepping J, et al. Innate immune training of granulopoiesis promotes anti-tumor activity. Cell. 2020;183(3):771-785.e12.

Haematologica | 108 December 2023

3203

INTRODUCTION TO REVIEW SERIES

Introduction to the peripheral T-cell lymphoma review series: advances in molecular characterization, classification refinement and treatment optimization Kerry J. Savage1 and Laurence de Leval2

Correspondence: K.J. Savage

Center for Lymphoid Cancer, Division of Medical Oncology, BC Cancer and the University of British Columbia, British Columbia, Canada and 2Institute of Pathology, Department of Laboratory Medicine and Pathology, Lausanne University Hospital and Lausanne University, Lausanne, Switzerland 1

ksavage@bccancer.bc.ca Received: Accepted:

June 16, 2023. August 29, 2023.

Peripheral T-cell lymphomas (PTCL), also known as mature T- and NK-cell lymphomas, are a diverse group of diseases with variable clinical presentations, including predominantly nodal or extranodal involvement, as well as rare leukemic forms. In Western populations PTCL represent 10% of all non-Hodgkin lymphomas; however, 20% of all non-Hodgkin lymphomas in Asia have a T/NK-cell phenotype with a higher proportion of Epstein-Barr virus (EBV)- or human T-lymphoma virus 1 (HTV1)-associated entities also seen. In the present review series, four papers cover our current understanding of PTCL disease biology, as well as optimal therapeutics in 2023 based on best available evidence, with a focus on nodal and selected extranodal entities. Eleven distinct or provisional PTCL entities were described in the Revised European American Lymphoma classification in 1994, and the subsequent World Health Organization (WHO) updates evolved to describe over 30 different PTCL subtypes today. In 2022, two separate classifications were developed. The 5th edition of the WHO classification (WHOHAEM5), led by experts designated by the International Agency for Research on Cancer (IARC), was published as a review1 and a beta version of the blue book is available online pending publication in a definitive form. A separate International Consensus Classification (ICC) was proposed by the Clinical Advisory Committee, a group which had directed the WHO classifications in the past, and is led by international lymphoma experts encompassing clinical, pathology and basic science expertise.2 Both classifications are referred to here for completeness, and differences from, as well as comparisons to, the 2017 4th revised WHO classification (WHO-HAEM4R) are highlighted in Table 1. While the ICC maintains the distinction between provisional and definitive disease entities, all diseases are established along the same rank in the WHO-HAEM5. Several entities introduced in the WHO-HAEM4R as provisional have been confirmed in both classifications.

From the first identification of a T-cell receptor rearrangement to establish monoclonality in T-cell lymphomas,3 genetics has always been integral to the diagnosis of PTCL. However, apart from the fusion of NPM::ALK as a result of t(2;5)(p23;q35) in anaplastic lymphoma kinase (ALK)-positive (ALK+) anaplastic large cell lymphoma (ALCL),4 and the more recently discovered DUSP22 and TP63 rearrangements in ALK-negative (ALK–) ALCL,5,6 very few structural diseasedefining aberrations have been described. Comprehensive studies evaluating the molecular profiles and mutational landscapes of PTCL have markedly enriched the characterization and defining criteria of several disease entities and enhanced our understanding of their pathobiology and pathogenesis, translating into significant classification changes. In some cases, genomic analysis has also provided prognostic information and helped to identify key oncogenic drivers which have informed therapeutics. Importantly, the ICC published a companion paper describing the current role as well as the promise of genomics and molecular testing across all lymphomas, including PTCL.7 Although there are a daunting number of PTCL entities and the projection is further expansion, being disease ‘splitters’ (vs. ‘lumpers’) has the advantage of highlighting the peculiar biology and natural history of each entity with parallel optimization of treatment approaches. The so-called ‘nodal’ PTCL include PTCL not otherwise specified (NOS), ALCL and follicular helper T-cell lymphoma (TFHL), this last being referred to as nodal T-follicular helper (TFH) cell lymphomas in the WHO-HAEM5. A new entity has also been added to nodal PTCL: ‘(primary) nodal EBV-positive T/NK-cell lymphoma’ and is deemed a provisional entity in the ICC (Table 1). In addition to EBV positivity, tumor cells are positive for cytotoxic markers and limited studies have suggested a very poor outcome, not unlike that of advanced stage extranodal NK/T-cell lymphoma.8 Although the term ‘nodal’ is something of a misnomer as extranodal sites of involvement are

Haematologica | 108 December 2023

3204

K.J. Savage and L. de Leval

INTRODUCTION TO REVIEW SERIES

Table 1. Mature T- and NK-cell neoplasms in the International Consensus Classification (ICC) and World Health Organization (WHO)-HAEM5 classification (2022) in reference to the WHO-HAEM4R classification (2017) (adapted from Alaggio et al.1 and Campo et al.2). WHO-HAEM4R 2017

ICC 2022

WHO-HAEM5 2022

T-prolymphocytic leukemia

T-cell prolymphocytic leukemia

T-prolymphocytic leukemia

T-cell large granular lymphocytic leukemia

T-large granular lymphocytic leukemia

Chronic lymphoproliferative disorder of NK cells

NK-large granular lymphocytic leukemia

Adult T-cell leukemia/lymphoma

EBV-positive T-cell/NK-cell lymphoproliferative disorders of childhood

EBV-positive T- and NK-cell lymphoid proliferations and lymphomas of childhood

Hydroa vacciniforme-like lymphoproliferative disorder

Hydroa vacciniforme lymphoproliferative disorder, classic type and systemic type

Hydroa vacciniforme lymphoproliferative disorder

Severe mosquito bite allergy

Chronic active EBV infection of T- and NKcell type, systemic form

Chronic active EBV disease, systemic (Tcell and NK-cell phenotype)

Systemic chronic active EBV disease

Systemic EBV-positive T-cell lymphoma of childhood

Extranodal NK/T-cell lymphoma, nasal type

Extranodal NK/T-cell lymphoma

Aggressive NK-cell leukemia

Not listed as an entity, subtype of peripheral T-cell lymphoma, not otherwise specified (PTCL-NOS)

Primary nodal EBV+ T-cell/NK-cell lymphoma

EBV+ nodal T- and NK-cell lymphoma

Enteropathy-associated T-cell lymphoma

Not listed as an entity

Type II refractory celiac disease

Not listed as an entity

Monomorphic epitheliotropic intestinal T-cell lymphoma

Intestinal T-cell lymphoma, NOS

Indolent T-cell lymphoproliferative disorder of Indolent clonal T-cell lymphoproliferative the gastrointestinal tract disorder of the gastrointestinal tract

Indolent T-cell lymphoma of the gastrointestinal tract

Not listed

Indolent NK-cell lymphoproliferative disorder of the gastrointestinal tract

Hepatosplenic T-cell lymphoma

Mycosis fungoides

Sézary syndrome

Primary cutaneous CD30+ T-cell lymphoproliferative disorders

Primary cutaneous CD30+ T-cell lymphoproliferative disorder: Lymphomatoid papulosis

Lymphomatoid papulosis

Primary cutaneous CD30+ T-cell lymphoproliferative disorder

Primary cutaneous anaplastic large cell lymphoma

Primary cutaneous CD4+ small/medium T-cell lymphoproliferative disorder

Primary cutaneous small/medium CD4+ T-cell lymphoproliferative disorder

Subcutaneous panniculitis-like T-cell lymphoma

Subcutaneous panniculitis-like T-cell lymphoma Continued on following page.

Haematologica | 108 December 2023

3205

K.J. Savage and L. de Leval

INTRODUCTION TO REVIEW SERIES

WHO-HAEM4R 2017

ICC 2022

WHO-HAEM5 2022

Primary cutaneous gamma-delta T-cell lymphoma

Primary cutaneous acral CD8+ T-cell lymphoma

Primary cutaneous acral CD8+ T-cell lymphoproliferative disorder

Primary cutaneous CD8+ aggressive epidermotropic cytotoxic T-cell lymphoma

Not listed

Primary cutaneous peripheral T-cell lymphoma, NOS

Peripheral T-cell lymphoma, NOS

Nodal lymphomas of T follicular helper origin

Follicular helper T-cell lymphoma

Nodal T-follicular helper (TFH) cell lymphoma

Angioimmunoblastic T-cell lymphoma

Follicular helper T-cell lymphoma, angioimmunoblastic type (angioimmunoblastic Tcell lymphoma)

Follicular T-cell lymphoma

Follicular helper T-cell lymphoma, follicular type

Nodal peripheral T-cell lymphoma with T follicular helper phenotype

Follicular helper T-cell lymphoma, NOS

Anaplastic large cell lymphoma, ALK-positive Anaplastic large cell lymphoma, ALK-positive ALK-positive anaplastic large cell lymphoma Anaplastic large cell lymphoma, ALK-negative Anaplastic large cell lymphoma, ALK-negative ALK-negative anaplastic large cell lymphoma Breast implant-associated anaplastic large cell lymphoma

Breast implant-associated anaplastic large cell lymphoma

The entities are listed according to the order in which they appear in the ICC 2022. Shading/no shading denotes groups of entities. Provisional entities in WHO-HAEM4R and the ICC are shown in italics. EBV: Epstein-Barr virus; NOS: not otherwise specified; ALCL: anaplastic large cell lymphoma.

not uncommon across all subtypes, this terminology is useful to distinguish these subtypes of lymphoma from those arising primarily in extranodal sites or skin as well as leukemic PTCL subtypes. PTCL-NOS has long been recognized as the ‘wastebasket’ diagnosis since it captures entities that do not meet the criteria for a defined PTCL subtype. As such, PTCL-NOS is a heterogenous group that still has not been fully delineated. Bisig and de Leval9 describe the proposed subtypes, GATA3 and TBX21, which were first identified through Affymetrix molecular profiling.10 This distinction is reproduced using an immunohistochemistry algorithm11 and it is hoped that in the future, a molecular Nanostring assay12 utilizing RNA derived from formalin-fixed, paraffin-embedded tissue can be applied. The GATA3 subtype with a signature of TH2 cells, including GATA3 and targets as well as components of the PI3K pathway, has an inferior prognosis compared to the TB2X1 subtype which has higher expression of TBX21 (T-bet) and EOMES and their targets (e.g., CCL3, IFNγ).10 However, the story has emerged as being more complicated, because a subset of TBX21 associated with the presence of DNMT3A mutations and cytotoxic gene expression signature has also an inferior prognosis.13 Furthermore, several cases remain ‘unclassified’. Although not yet incorporated in current classifications, correlative studies integrating molecular clas-

sification of PTCL-NOS in clinical trials may elucidate whether the GATA3 and TBX21 subtypes have a differential response to novel therapies. In the WHO-HAEM4R 2017 classification, a new category for TFHL was created to expand beyond the prototype, angioimmunoblastic T-cell lymphoma (AITL), to encompass cases previously classified as PTCL-NOS that have a TFH phenotype, currently defined as the presence of two or more of BCL6, CD10, CXCL13, PD1 and ICOS by immunohistochemistry, but lacking the morphological criteria for AITL. In addition, the TFHL category includes the less common morphological variant, follicular T-cell lymphoma. Minor nomenclature differences can be noted when comparing the ICC and WHO-HAEM5 (Table 1). The genomic landscape of TFHL has been well characterized with recurrent mutations in TET2, DNMT3A, IDH2 and RHOA. Although not mandatory, mutational profiling is now routinely performed at many centers and is helpful for the diagnosis of TFHL,7,14 especially for borderline cases, and can guide therapeutics.15 ALCL is broadly divided into ALK+ ALCL, identified by the presence of the ALK protein detected by immunohistochemistry, and the morphologically similar ALK– ALCL. ALK– ALCL was once proposed to be combined with PTCL-NOS because of its poor outcome,16 but later studies confirmed a more favorable prognosis.17 Almost 10 years ago, further

Haematologica | 108 December 2023

3206

INTRODUCTION TO REVIEW SERIES

K.J. Savage and L. de Leval

genetic heterogeneity was discovered with the identification of a DUSP22 rearrangement in 20-30% cases.18 As described by Bisig and de Leval, DUSP22-rearranged cases have distinct morphological, immunohistochemical and molecular features, establishing them as a genetic subtype in the ICC, but not in the WHO-HAEM5 classification, due to disparate prognoses in studies which collectively suggest that they are likely more intermediate between ALK+ ALCL and ‘triple negative’ ALCL which lack any rearrangement of ALK, DUSP22 or TP63.19-21 Approximately 2-8% of cases of ALK– ALCL harbor a TP63 rearrangement which is associated with a dismal outcome.18 These cases are not yet a distinct genetic subtype as further studies are needed to fully char-

acterize them. Extranodal entities, as detailed by Lewis, Zhou and Dogan,22 with a clinical lens provided by Stuver, Epstein-Peterson and Horwitz,23 are identified by their primary location. They constitute a diverse and rare group of diseases with a limited number of dedicated clinical trials, making therapeutic advancement very challenging. Subcutaneous panniculitis-like T-cell lymphoma was first recognized in the 2001 WHO classification which was later refined to exclude cases with a γδ phenotype that are lumped into primary cutaneous γδ T-cell lymphoma because of a similar poor prognosis.24 Approximately 20% are associated with hemophagocytic lymphohistiocytosis, which can be particularly severe in those

Figure 1. Systemic agents approved, previously studied, or under investigation in peripheral T-cell lymphomas/mature T/NKlymphomas. *Approved in the USA; brentuximab vedotin (BV) is globally approved in relapsed/refractory (R/R) anaplastic large cell lymphoma (ALCL) and for front-line treatment in CD30+ PTCL; BV-CHP (BV, cyclophosphamide, doxorubicin, prednisolone) in Europe approved in ALCL only. #Romidepsin was previously approved in the USA/Canada but withdrawn following report of the negative findings of the phase III, RO-CHOP (romidepsin, cyclophosphamide, doxorubicin, vincristine, predisone) versus CHOP study; pralatrexate is not approved in Europe. ^Approved outside the USA, mogamulizumab is approved in Japan only for CCR4+ R/R PTCL (and adult T-cell leukemia/lymphoma [positive for human T-lymphotropic virus 1]); chidamide is approved in China only for R/R PTCL. Please also see Table 2 and the reviews by Ngu and Savage15 and Stuver, Epstein and Horwitz25 for drug approvals and disease rationale. ALK: anaplastic lymphoma kinase; CAR: chimeric antigen receptor; cIAP1: cytoplasmic inhibitor of apoptosis 1; CRBN: cereblon; DHFR: dihydrofolate reductase; EBV: Epstein-Barr virus; EZH: enhancer of zest Haematologica | 108 December 2023

3207

INTRODUCTION TO REVIEW SERIES

K.J. Savage and L. de Leval

Table 2. Selected drugs, their targets and disease rationale under study in peripheral T-cell lymphoma/mature NK/T-cell neoplasms. Drug class, type and/or target

Regulatory approval/indication as of July 2023

Disease rationale

Anti-CD30 targeting agents Antibody-drug conjugate: brentuximab vedotin BV-CHP

R/R ALCL (Global) Front-line CD30+ PTCL (FDA) Front-line ALCL and CD30+ PTCL-NOS/AITL (Health Canada) Front-line ALCL (EMA)

ALCL > EATL > other CD30+ PTCL

Anti-CD30 CAR T-cell therapy

None

ALCL > other CD30+ PTCL

Bi-specific antibody e.g., AFM 16 (CD30/CD16 [NK cell])

None

ALCL > other CD30+ PTCL

CAR T-cell and allogeneic T-cell therapy targets beyond CD30 CD70

None

CD70+ PTCL

CD5

None

CD5+ PTCL

TRBC1 positive

None

TRBC1+ PTCL

PD1 inhibitor: pembrolizumab, nivolumab PDL inhibitor: sugemalimab,a avelumab

Nonea

Extranodal NK/TCL, ? primary nodal EBV+ T/NK-cell lymphoma

Anti-CD47 antibody and CD47 decoy

None

CTCL>PTCL(nodal)

EBV-specific cytotoxic T cells

None

Extranodal NK/TCL, ? primary nodal EBV+ T/NK-cell lymphoma

Nanostinostat and valganiclovir

None

NK/TCL > EBV+ nodal PTCL

Romidepsinb (FDA, Health Canada) Belinostat (FDA) Chidamide (NMDA China)

TFHL > other PTCL (nodal)

Hypomethylating agents: 5-azacitidine, decitabine

None

TFHL > other PTCL (nodal)

EZH2 inhibitors: valemetostat

None

TFHL > other PTCL (nodal)

Farnesyltransferase inhibitor: tipifarnib

None

AITL and CXCL12 wild type?

PI3γ kinase inhibitor: duvelisib

None

TFHL > other PTCL (nodal)

JAK/STAT inhibitor: ruxolitinib

None

JAK2 mutation, pSTAT3+ > unselected Hemophagocytic lymphohistiocytosis

R/R ALK+ ALCL 1-21 years (FDA)

R/R ALK+ ALCL

Aurora kinase inhibitor: alisertib

None

Unknown

Inhibitor of apoptosis: tolinapant

None

Unknown

Lenalidomide

None

?TFHL

Cyclosporine

None

SCPTCL (some TFHL)

Pralatrexate (USA, Canada)

PTCL-NOS/ALCL > AITL

None

Extranodal NKTCL, speculation primary nodal EBV+ T/NKCL

EBV-directed therapies

Epigenetic therapies HDAC inhibitors

Oncogenic pathway inhibition

ALK inhibitor: crizotinib

Immunomodulators

PTCL-specific chemotherapy Dihydrofolate reductase inhibitor Asparagine depletion: pegylated asparaginase

Sugemalimab: Food and Drug Administration (FDA) breakthrough therapy and orphan drug designation and breakthrough designation by the National Medical Products Administration (NMPA) in China. bDe-listed by the FDA and Health Canada after the negative CHOP versus RoCHOP study. BV-CHP: brentuximab vedotin-cyclophosphamide, doxorubicin and prednisone; R/R: relapsed/refractory; ALCL: anaplastic large cell lymphoma; PTCL: peripheral T-cell lymphoma; NOS: not otherwise specified; AITL: angioimmunoblastic T-cell lymphoma; EMA: European Medicines Agency; EATL: enteropathy-associated T-cell lymphoma; CAR: chimeric antigen receptor; PD1: programmed cell death 1; PDL: programmed cell death ligand; TCL: T-cell lymphoma; EBV: Epstein-Barr virus; HDAC: histone deacetylase; TFHL: follicular helper T-cell lymphomas; SCPTCL: subcutaneous panniculitis like T-cell lymphoma; T/NKCL: T/NK-cell lymphoma.

Haematologica | 108 December 2023

3208

INTRODUCTION TO REVIEW SERIES

K.J. Savage and L. de Leval

harboring a germline mutation in HAVCR2.25 Hepatosplenic Tcell lymphoma often occurs in the setting of chronic immunosuppression, especially in patients receiving combination therapy with tumor necrosis-α inhibitor therapy and thiopurines (e.g., azathioprine). Intestinal T/NK cell lymphoma includes the familiar enteropathy-associated T-cell lymphoma (previously referred to as type 1 EATL) linked to celiac disease, which was importantly distinguished from monomorphic epitheliotropic intestinal T-cell lymphoma (MEITL) (previously referred to as ‘type 2’ EATL) in the WHO-HAEM4R classification, each having a distinct genetic landscape, with highly recurrent SETD2 mutations observed in the latter. Intestinal T-cell lymphoma NOS was introduced in the WHOHAEM4R classification to capture cases that would have previously been included with PTCL-NOS and is retained in the 2022 proposals. Indolent T- and NK-cell lymphoproliferative disorders of the gastrointestinal tract (with the term lymphoma used in the WHO-HAEM5 for T-cell cases) are critical to recognize in order to avoid use of chemotherapy which does not appear to impact the natural history. Historically, treatment paradigms in mature T- and NK-cell neoplasms have been largely modelled on those for aggressive B-cell lymphomas, leading to default use of CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) or CHOPlike chemotherapy,26 which remains the standard comparison arm in clinical trials today. With a still high risk of relapse, etoposide is often added to CHOP chemotherapy (CHOEP) coupled with consolidative autologous stem cell transplant; however, in the absence of a randomized trial, practices vary. With improved understanding of the molecular basis of disease and exciting expansion of novel therapies under investigation in relapsed/refractory PTCL (Figure 1, Table 2), as outlined by Ngu and Savage,15 more recent studies have integrated these agents in the front-line setting, including chemotherapy-free approaches, with a particular interest in TFHL. CHOP (and CHOEP) has the best track record in ALCL, especially ALK+ ALCL. The practice-changing ECHELON-2 study in CD30+ PTCL evaluated the addition of the anti-CD30 antibody drug conjugate brentuximab vedotin (BV) to CHP (CHOP with the omission of vincristine because of the overlapping peripheral neuropathy), and demonstrated improved progression-free and overall survival compared to CHOP chemotherapy, leading to a new treatment paradigm with particular relevance in ALCL.27 Uncertainties remaining around the efficacy of CHP-BV in non-ALCL CD30+ PTCL subtypes have led to differences in regulatory approval, with use of this regimen in Europe restricted to ALCL alone15 (Table 2). With a growing realization that a ‘one size fits all’ approach is suboptimal in PTCL, entity-specific studies have evolved, leading to the complete abandonment of CHOP in some entities, even in the absence of a randomized controlled trial. This is best exemplified in the extranodal subtypes.23 Extranodal NK/T-cell lymphoma is typified by overexpression of

P-glycoprotein which imparts a multi-drug resistance phenotype, including resistance to anthracyclines. As outlined by Stuver and colleagues,23 this led to the exploration of nonCHOP chemotherapy, with clear improvement in outcomes with asparaginase-based regimens as well as cisplatin-based combinations in early-stage disease along with early integration of radiotherapy for disease that could be encompassed in a radiation field. Since extranodal NK/T-cell lymphoma is an EBV-driven disease, programmed death ligand (PDL) and PD1 inhibitors show clear therapeutic promise and are under investigation in a variety of settings (Figure 1, Table 2).23 NonCHOP regimens (e.g., ICE - ifosfamide, carboplatin, etoposide) are also preferred in the primary therapy of hepatosplenic Tcell lymphoma followed typically by an allogeneic stem cell transplant. At the other end of the spectrum, immunomodulator therapy, such as cyclosporine, has evolved to be the treatment of choice in subcutaneous panniculitis like T-cell lymphoma with a recent small series also indicating that the JAK1/2 inhibitor ruxolitinib is highly effective in those with refractory hemophagocytic lymphohistiocytosis.23,28 With an increased understanding of the genetic and molecular basis of PTCL, more rational, biologically driven and subtype-specific research is underway. As highlighted by Ngu and Savage,15 TFHL represent the ‘poster child’ for more personalized therapies as studies emerge showing sensitivity to epigenetic modifiers, with some long-term remissions being achieved29 (Figure 1, Table 2). This has led to studies of combinations of novel agents, most often integrating epigenetic therapies and even including cohorts of treatment-naïve patients; however, the curative potential of these combinations is still unknown. Overall, although progress in mature NK/T-cell neoplasms has lagged behind that in B-cell lymphomas, a renewed focus has led to important advances in understanding the molecular basis of these diseases as well as success in shifting the therapeutic bar in many subtypes; despite this, there is still work to be done. Modern clinical trials have evolved in some cases to focus on specific entities or pathways, often integrating rich correlative studies to elucidate markers of response and resistance. Ultimately, for our community of disease splitters, this ‘divide and conquer’ approach will lead to more customized treatment approaches and improved outcomes across the whole spectrum of PTCL. Disclosures KJS has received honoraria from and provided consulting for BMS, Merck, Seagen, and Janssen; has sat on a steering committee for Beigene; has received research funding from BMS; has received institutional research funding from Roche; and sat on a Data Safety and Monitoring Committee for DSMC. LdeL has no conflicts of interest to disclose. Contributions KJS and LdL wrote the manuscript.

Haematologica | 108 December 2023

3209

INTRODUCTION TO REVIEW SERIES

K.J. Savage and L. de Leval

References 1. Alaggio R, Amador C, Anagnostopoulos I, et al. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: lymphoid neoplasms. Leukemia. 2022;36(7):1720-1748. 2. Campo E, Jaffe ES, Cook JR, et al. The International Consensus Classification of mature lymphoid neoplasms: a report from the Clinical Advisory Committee. Blood. 2022;140(11):1229-1253. 3. Diss TC, Watts M, Pan LX, Burke M, Linch D, Isaacson PG. The polymerase chain reaction in the demonstration of monoclonality in T cell lymphomas. J Clin Pathol. 1995;48(11):1045-1050. 4. Rimokh R, Magaud JP, Berger F, et al. A translocation involving a specific breakpoint (q35) on chromosome 5 is characteristic of anaplastic large cell lymphoma ('Ki-1 lymphoma'). Br J Haematol. 1989;71(1):31-36. 5. Feldman AL, Dogan A, Smith DI, et al. Discovery of recurrent t(6;7)(p25.3;q32.3) translocations in ALK-negative anaplastic large cell lymphomas by massively parallel genomic sequencing. Blood. 2011;117(3):915-919. 6. Vasmatzis G, Johnson SH, Knudson RA, et al. Genome-wide analysis reveals recurrent structural abnormalities of TP63 and other p53-related genes in peripheral T-cell lymphomas. Blood. 2012;120(11):2280-2289. 7. de Leval L, Alizadeh AA, Bergsagel PL, et al. Genomic profiling for clinical decision making in lymphoid neoplasms. Blood. 2022;140(21):2193-2227. 8. Wai CMM, Chen S, Phyu T, et al. Immune pathway upregulation and lower genomic instability distinguish EBV-positive nodal T/NK-cell lymphoma from ENKTL and PTCL-NOS. Haematologica. 2022;107(8):1864-1879. 9. Bisig B, Savage KJ, de Leval L. Pathobiology of nodal peripheral T-cell lymphomas: current understanding and future directions. Haematologica 2023;108(12):3227-3243. 10. Iqbal J, Wright G, Wang C, et al. Gene expression signatures delineate biological and prognostic subgroups in peripheral Tcell lymphoma. Blood. 2014;123(19):2915-2923. 11. Amador C, Greiner TC, Heavican TB, et al. Reproducing the molecular subclassification of peripheral T-cell lymphoma-NOS by immunohistochemistry. Blood. 2019;134(24):2159-2170. 12. Amador C, Bouska A, Wright G, et al. Gene expression signatures for the accurate diagnosis of peripheral T-cell lymphoma entities in the routine clinical practice. J Clin Oncol. 2022;40(36):4261-4275. 13. Herek TA, Bouska A, Lone W, et al. DNMT3A mutations define a unique biological and prognostic subgroup associated with cytotoxic T cells in PTCL-NOS. Blood. 2022;140(11):1278-1290. 14. Ondrejka SL, Amador C, Climent F, et al. Follicular helper T-cell lymphomas: disease spectrum, relationship with clonal hematopoiesis, and mimics - a report of the 2022 EA4HP/SH lymphoma workshop. Virchows Arch. 2023;483(3):349-365. 15. Ngu HS, Savage KJ. Past, present and future therapeutic approaches in nodal peripheral T-cell lymphomas. Haematologica 2023;108(12):3211-3226. 16. ten Berge RL, de Bruin PC, Oudejans JJ, Ossenkoppele GJ, van

der Valk P, Meijer CJ. ALK-negative anaplastic large-cell lymphoma demonstrates similar poor prognosis to peripheral Tcell lymphoma, unspecified. Histopathology. 2003;43(5):462-469. 17. Savage KJ, Harris NL, Vose JM, et al. ALK- anaplastic large-cell lymphoma is clinically and immunophenotypically different from both ALK+ ALCL and peripheral T-cell lymphoma, not otherwise specified: report from the International Peripheral TCell Lymphoma Project. Blood. 2008;111(12):5496-5504. 18. Parrilla Castellar ER, Jaffe ES, Said JW, et al. ALK-negative anaplastic large cell lymphoma is a genetically heterogeneous disease with widely disparate clinical outcomes. Blood. 2014;124(9):1473-1480. 19. Savage KJ, Slack GW. DUSP22 rearranged ALK-negative ALCL is a pathogenetically distinct disease but can have variable clinical outcome. Haematologica. 2023;108(6):1463-1467. 20. Sibon D, Bisig B, Bonnet C, et al. ALK-negative anaplastic large cell lymphoma with DUSP22 rearrangement has distinctive disease characteristics with better progression-free survival: a LYSA study. Haematologica. 2023;108(6):1590-1603. 21. Qiu L, Tang G, Li S, et al. DUSP22 rearrangement is associated with distinctive immunophenotype but not outcome in patients with systemic ALK-negative anaplastic large cell lymphoma. Haematologica. 2023;108(6):1604-1615. 22. Lewis NE, Zhou T, Dogan A. Biology and genetics of rare extranodal mature T-cell and NK-cell lymphomas. Haematologica 2023;108(12)3261-3277. 23. Stuver R, Epstein-Peterson ZD, Horwitz SM. Few and far between: clinical management of rare extranodal subtypes of mature T-cell and NK-cell lymphomas. Haematologica 2023;108(12)3244-3260. 24. Swerdlow SH, Campo E, Harris NL, et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissue. IARC 2008; 4th edition. 25. Gayden T, Sepulveda FE, Khuong-Quang DA, et al. Germline HAVCR2 mutations altering TIM-3 characterize subcutaneous panniculitis-like T cell lymphomas with hemophagocytic lymphohistiocytic syndrome. Nat Genet. 2018;50(12):1650-1657. 26. Fisher RI, Gaynor ER, Dahlberg S, et al. Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin's lymphoma. N Engl J Med. 1993;328(14):1002-1006. 27. Horwitz S, O'Connor OA, Pro B, et al. Brentuximab vedotin with chemotherapy for CD30-positive peripheral T-cell lymphoma (ECHELON-2): a global, double-blind, randomised, phase 3 trial. Lancet. 2019;393(10168):229-240. 28. Zhang Q, Zhou CJ, Li DH, et al. Efficacy of ruxolitinib for HAVCR2 mutation-associated hemophagocytic lymphohistiocytosis and panniculitis manifestations in children. Br J Haematol. 2023;202(1):135-146. 29. Pro B, Horwitz SM, Prince HM, et al. Romidepsin induces durable responses in patients with relapsed or refractory angioimmunoblastic T-cell lymphoma. Hematol Oncol. 2017;35(4):914-917.

Haematologica | 108 December 2023

3210

REVIEW SERIES

Past, present and future therapeutic approaches in nodal peripheral T-cell lymphomas Henry S. Ngu° and Kerry J. Savage

Correspondence: K.J. Savage

Center for Lymphoid Cancer, Division of Medical Oncology BC Cancer and the University of British Columbia, British Columbia, Vancouver, Canada °

Current address: Department of Hematology, Auckland City Hospital, Auckland, New Zealand

ksavage@bccancer.bc.ca Received: Accepted:

May 18, 2023. August 14, 2023.

Abstract Peripheral T-cell lymphomas (PTCL) encompass over 30 different entities and although they share post-thymic T- or NK-cell derivation, the disease biology and genomic landscape are very diverse across subtypes. In Western populations, nodal PTCL are the most frequently encountered entities in clinical practice and although important achievements have been made in deciphering the underlying biology and in therapeutic advances, there are still large gaps in disease understanding and clinical scenarios in which controversy over best practice continues. CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone)based chemotherapy continues to be the ‘standard’ treatment, with the addition of brentuximab vedotin (BV) in the combination CHP (cyclosphosphamide, doxorubicin, prednisone)-BV representing a new treatment paradigm in CD30+ PTCL although its benefit is less certain in the non-anaplastic large cell lymphoma subtypes. Given the high risk of relapse, consolidative autologous stem cell transplant is considered in nodal PTCL, outside of ALK-positive anaplastic large cell lymphoma; however, in the absence of a randomized controlled trials, practices vary. Beyond CHP-BV, most study activity has focused on adding a novel agent to CHOP (i.e., CHOP + drug X). However, with high complete remission rates observed with some novel therapy combinations, these regimens are being tested in the front-line setting, with a particular rationale in follicular helper T-cell lymphomas which have a clear sensitivity to epigenetic modifying therapies. This is well exemplified in the relapsed/refractory setting in which rational combination therapies are being developed for specific subtypes or guided by underlying biology. Taken together, we have finally moved into an era of a more personalized approach to the management of nodal PTCL.

Introduction Peripheral T-cell lymphomas (PTCL) represent approximately 10% of all non-Hodgkin lymphomas although geographic variation is notable.1,2 In 2022, both the World Health Organization (WHO) 5th edition lymphoma classification update (WHO-HAEM5)1 and the new International Consensus Classification (ICC)2 were published. These two publications contained classification refinements, including for mature T/NK-cell neoplasms. Most of these refinements are concordant in the two classification systems, although there a few exceptions (see Table 1 in the Introduction to the Review Series on Lymphoma published in this issue of Haematologica3). The so-called ‘nodal’ PTCL subtypes, grouped to separate them from predominantly leukemic, extranodal and cutaneous subtypes of PTCL, represent about 60% of all PTCL in Western populations, and include PTCL not otherwise specified (NOS), anaplastic large cell lymphoma (ALCL) and follicular helper T-cell lymphomas. The last group are referred to as nodal T-follicular helper lympho-

mas in the WHO-HAEM5 and are considered a family of three lymphomas – angioimmunoblastic type (angioimmunoblastic T-cell lymphoma [AITL]), a NOS type and a follicular type – whereas the ICC describes one entity, follicular helper T-cell lymphoma, with three subtypes (angioimmunoblastic type, follicular type and NOS).3 For the purpose of this review, they are collectively referred to as TFHL with specific subtypes referenced where appropriate. In addition, a new rare nodal PTCL subtype is now recognized, primary nodal Epstein-Barr virus (EBV)+ T/NK-cell lymphoma in the ICC (provisional entity) or EBV+ nodal T/NK-cell lymphoma in WHO-HAEM5 (distinct entity) (Table 1 in the Introduction to the Review Series3) which, in addition to being EBV+, has an activated cytotoxic phenotype (TIA+ ± granzyme B/perforin) and a poor outcome that is more similar to that of extranodal NK/T-cell lymphoma.4 While progress in PTCL has lagged behind that in B-cell lymphomas, the past decade has been a period of great advancement, both in elucidating disease pathogenesis and in the development of new therapies. Recognizing that drug

Haematologica | 108 December 2023

3211

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

Table 1. Largest retrospective series including nodal peripheral T-cell lymphomas treated with primarily anthracycline-based chemotherapy.

First author Study

Period of PTCL diagnosis (age for enrollment)

Vose7a International Peripheral T-cell Lymphoma Project

1990-2002 (≥19 yrs)

Ellin8 Swedish Registry

2000-2009 (≥18 yrs)

Brink9 Netherlands Cancer Registry

1989-2018 (18-65 yrs)

Nodal PTCL subtypes

PTCL-NOS AITL ALK– ALCL ALK+ ALCL PTCL-NOS AITL ALK– ALCL ALK+ ALCL PTCL-NOS AITL ALK– ALCL ALK+ ALCL

340 243 72 87 256 104 115 68 692 294 89 139

Received CHOP/CHOPlike therapy, % 80 82 95 88 84 (All)b

NRc

5-year PFS %

5-year OS %

20 18 36 60 21 20 38 63

32 32 49 70 28 32 31 79 32 44 52 72

Median follow-up reported in subsequent subtype-specific publications. bTreatment information was available for 708/757 cases of peripheral T-cell lymphoma: 84% received CHOP/CHOP-like chemotherapy but this estimate includes other non-cutaneous, non-leukemic subtypes. cIn the whole cohort 1,369/1,427 (96%) patients received chemotherapy but type was not specified with the overall survival estimates. PTCL: peripheral T-cell lymphoma; CHOP: cyclophosphamide, doxorubicin, vincristine, prednisone; PFS: progression-free survival; OS: overall survival; yrs; years; NOS: not otherwise specified; AITL: angioimmunoblastic T-cell lymphoma; ALK: anaplastic lymphoma kinase; ALCL: anaplastic large cell lymphoma; NR: not reported.

regimen as the standard combination chemotherapy for all aggressive lymphomas. The study was conducted when diagnoses were based on the Working Formulation, and prior to the integration of routine immunophenotyping, but CHOP treatment was similarly adopted for PTCL.6 Several large, retrospective series have documented the poor outcome in most patients with PTCL treated with primarily anthracycline-based chemotherapy, including subtype-specific results (Table 1).7-9 Notably, there is limited information on the new entity, primary nodal EBV+ T/NK-cell lymphoma, but it appears to have an outcome that is inferior to that of PTCL-NOS.4 In these series, given their only recent recognition, primary nodal EBV+ T/NK-cell lymphoma and TFHL other than AITL, are combined in the PTCL-NOS subgroup. Of note, the International Prognostic Index (IPI) effectively stratifies patients with PTCL-NOS and ALCL into risk groups7,10-12 but its usefulness in AITL has been more limited given that most patients fall into a high-risk category.7,13,14 Overall, patients with ALK+ ALCL have a more favorable outcome, although this is in part driven by a younger age at diagnosis and, importantly, those with a high IPI score have a poor outcome, not unlike that of patients with other PTCL subtypes (IPI score ≥3; 5-year progression-free survival [PFS] 23-54%; 5-year overall survival [OS] 23-62%).10,15 Outcomes are also better in ALK− ALCL,7-11 a finding that is more evident in larger series and with central pathology review, given the potential for mis-diagnosis as CD30+ PTCL-NOS. The outcome of ALK− ALCL may also be impacted by the The origin of CHOP and impact of proportion of cases with DUSP22 rearrangement and/or P63 prognostic factors in PTCL rearrangement, as outlined by Bisig, Savage and de Leval.5 Almost 30 years ago, the landmark Southwest Oncology A recent study suggested that the presence of TP53 mutaGroup (SWOG) phase III study established the CHOP tions correlates with an inferior PFS in patients treated with sensitivities may be different, clinical trials have evolved to focus specifically on PTCL and more recently, on specific subtypes. With the exception of ALCL, in which brentuximab vedotin (BV)-CHP (cyclosphosphamide, doxorubicin, prednisone) is considered the new standard treatment, uncertainty remains regarding the optimal front-line therapeutic regimen and the role of consolidative high-dose chemotherapy and autologous stem cell transplant (auto-SCT). This is particularly apparent in TFHL in which sensitivity to epigenetic therapies has led to numerous studies evaluating chemotherapy-free, novel therapy combination treatment approaches, also in treatment-naïve patients. Furthermore, as the genomic landscape is uncovered, evolving studies are targeting specific pathway vulnerabilities (e.g., Janus kinase [JAK]/signal transducer and activator of transcription [STAT], phosphatidyl inositol 3 kinase [PI3K]) as well as integrating biological correlates in an effort to understand biomarkers of response and resistance. As a follow-up to the nodal PTCL pathobiology paper by Bisig, Savage and de Leval in this issue, we review the history of CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) and transplant for nodal PTCL, highlight new treatment advances in the front-line and relapsed/refractory (R/R) settings as well as the promise of a more 'personalized' therapy approach.

Haematologica | 108 December 2023

3212

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

CHOP or CHOP-like chemotherapy and is found in approximately one-third of cases of ALK− ALCL and PTCL-NOS.16 Overall, longer term follow-up is important in PTCL since, among patients alive and event-free at 24 months, the 5year risk of subsequent relapse is still 23% for nodal PTCL, compared to only 7% in a prior study of diffuse large B-cell lymphoma.17,18 There has been some debate on whether anthracyclines are essential for cure, particularly in PTCL-NOS and AITL,7,19 leading to studies exploring alternative combination therapy regimens. The UK National Cancer Research Institute (NCRI) Lymphoma Clinical study group compared CHOP to GEM-P (gemcitabine, solumedrol, cisplatin) in a randomized phase II study (CHEMO-T), but results were similar with a 2-year PFS of 37% versus 38%, respectively (P=0.82) and a 2-year OS of 51% versus 64%, respectively (P=0.31).20 Although exploratory, patients with PTCL-NOS (as well as enteropathy-associated T-cell lymphoma) had a significantly better outcome with CHOP (odds ratio [OR]=0.036; P=0.012). In contrast, outcomes by treatment arm were similar in AITL (OR=0.69; P=0.578). The SWOG PEGS (cisplatin, etoposide, gemcitabine, solumedrol) regimen, built on the premise of using drugs not effluxed by multidrug resistance-1/permeability glycoprotein (MDR-1/P-gp), gave disappointing results with a 2-year PFS of only 12%, which may also reflect the absence of cyclophosphamide in the regimen. Although data are very limited, CHOP appears to be suboptimal in primary nodal EBV+ T/NK-cell lymphoma and it may be that regimens used in extranodal NK/T-cell lymphoma are more appropriate in this entity but this requires further study. Taken together, CHOP appears to be most important in ALCL, variably effective in PTCL-NOS especially given disease heterogeneity, and there is less certainty in AITL. Notably, there are limited data on outcomes with CHOP in the other TFHL subtypes.21 More intensive regimens, including the integration of etoposide, have also been explored in the front-line treatment of PTCL. This was first evaluated in a retrospective analysis of all patients with a PTCL diagnosis enrolled on prospective German High-Grade Non-Hodgkin Lymphoma Study Group (DSHNHL) studies (now called the German Lymphoma Alliance [GLA]).18 An improved event-free survival was observed in young (≤60 years), good-risk (normal lactate dehydrogenase level) patients who received etoposide with a front-line CHOP/CHOP-like regimen; however, the benefit was most evident in ALK+ ALCL (3-year event-free survival: 91% vs. 57%; P=0.01) with a trend observed in other nodal PTCL subtypes (3-year event-free survival: 61% vs. 48.3%; P=0.057). The findings did not translate into an OS benefit, and it was an unadjusted analysis.22 Other studies have supported improved outcomes with CHOP plus etoposide (CHOEP) over CHOP alone specifically in ALK+ ALCL,9,12 but this is also the group that derives the greatest benefit from upfront BV-CHP as

outlined below. Results with CHOEP are more mixed in the other nodal PTCL.8,9 In the Netherlands Cancer Registry (NCR), there was no OS benefit of CHOEP over CHOP in a multivariate analysis when adjusted for the IPI score and use of auto-SCT.9 Taken together, CHOEP may still be considered a chemotherapy alternative in select patients but should be avoided in older patients given additional toxicity and lack of firm evidence of benefit.

Randomized front-line studies in PTCL The earliest phase III randomized controlled study was conducted by the Groupe Ouest Est d’Etude des Leucemies et Autres Maladies due Sang (GOELAMS) comparing a doseintensive regimen, VIP-reinforced-ABVD (etoposide, ifosfamide, cisplatin; doxorubicin, bleomycin, vinblastine, decarbazine), to CHOP in newly diagnosed PTCL, including patients with ALCL (ALK status not specified), who represented almost half of enrolled patients (42/86). There was no difference in 2-year event-free survival which was 45% for all patients.23 The most significant advancement to date in the treatment of nodal PTCL has been the impact of BV in ALCL. BV is an anti-CD30 antibody-drug conjugate linked to the anti-tubulin agent monomethyl auristatin E and was initially developed and approved for classical Hodgkin lymphoma and systemic ALCL in the R/R setting (see below). A phase I study established the safety and promise of CHP-BV in CD30+ PTCL (n=26 [ALCL n=19]: 5-year PFS 52%, 5-year OS 80%), with the omission of vincristine due to overlapping peripheral neuropathy with BV.24 The subsequent practicechanging, double-blind, double-dummy, randomized phase III ECHELON-2 trial demonstrated improved PFS (3-year: 57.1% vs. 44.4%; P<0.011) and OS (3 year: 76.8% vs. 69.1%; P=0.024) with BV-CHP compared to CHOP, in 552 newly diagnosed patients with CD30+ PTCL (CD30 ≥10%; ALK+ ALCL (IPI score ≥2 only).25 Results were maintained in the 5-year update (5-year PFS 51.4% vs. 43%; P=0.0077 and 5-year OS 70.1% vs. 61%; P=0.042).25,26 The benefit was most striking in ALCL (5-year PFS 60.6% vs. 48.4%; P=0.0009 and 5-year OS 75.8% vs. 68.7%; P=0.053). As a regulatory requirement, the study was powered to evaluate ALCL, and the nonALCL subtypes were under-represented (AITL n=54, PTCLNOS n=72). As a result, the statistical comparisons for AITL and PTCL-NOS were unplanned and underpowered but since confidence intervals all crossed 1, the benefit of BVCHP remains uncertain.26 This also led to differences in regulatory approval of BV-CHP, with the USA Food and Drug Administration including the intent-to-treat population of CD30+ PTCL eligible for the study, Health Canada approving it for ALCL and CD30+ PTCL-NOS or AITL only, whereas in Europe and the UK, BV-CHP is only approved for ALCL. Several additional comments can be made about ECHE-

Haematologica | 108 December 2023

3213

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

LON-2. The study was restricted to cases with CD30 expression of ≥10% but regulatory approval did not specify a CD30 cutoff. A separate phase II study in CD30+ (<10%) PTCL is ongoing (NCT04569032) (Online Supplementary Table S1). Previous studies in R/R PTCL have not shown a good correlation between CD30 expression and response to BV, which may reflect the insensitivity of immunohistochemical detection or a broader mechanism of action, targeting the microenvironment.27,28 In addition, information on the presence of DUSP22 and P63 rearrangements in ALK– ALCL is not available, which could have an impact on prognosis. Some have also argued whether CHOP is the appropriate comparator. Although CHOEP is an option in PTCL, in the absence of randomized controlled studies, it has not replaced CHOP as a preferred standard therapy. The strongest retrospective data for the use of CHOEP are those for ALK+ ALCL, but this is also the group with excellent outcomes with CHP-BV in ECHELON-2 with a 5-year PFS of 87% versus 67% (P=0.0372), despite restriction to patients with an IPI score ≥2. Outcomes were also more favorable in ALK– ALCL, although there is still room for improvement (5-year PFS 49% vs. 39%; P=0.0054).26 A separate phase II study explored CHEP-BV (cyclophosphamide, doxorubicin, etoposide, prednisone plus BV) in CD30+ PTCL, including ALCL, with encouraging results (objective response rate [ORR] 91%, complete response [CR 80%]), but the rate of febrile neutropenia was 21% with routine granulocyte colony-stimulating factor support.29 This regimen may be taken into consideration in very high-risk, younger ALCL patients especially those with an elevated central nervous system risk (e.g., high lactate dehydrogenase, involvement of multiple extranodal sites), given the additional penetration of etoposide across the blood-brain barrier, which does not appear to occur with BV.29 Romidepsin and alemtuzumab have also been evaluated in combination with CHOP/CHOP-like therapy in phase III trials. Romidepsin is a selective histone deacetylase (HDAC) inhibitor and demonstrated modest efficacy in a phase II study in R/R PTCL (see below), ultimately leading to a phase III Lymphoma Study Association (LYSA) study evaluating romidepsin plus CHOP (Ro-CHOP) versus CHOP in previously untreated PTCL, excluding ALK+ ALCL.30 The addition of romidepsin did not translate into an improved PFS in the intention-to-treat population (P=0.096) and was associated with increased toxicity.30 The negative results have unfortunately led to de-listing of romidepsin in the R/R setting in the USA and Canada (see below). Alemtuzumab is an anti-CD52 humanized monoclonal antibody. CD52 antigen is a glycosylphosphatidylinositol-linked glycoprotein that is expressed on lymphocytes and monocytes and is variably expressed in PTCL.31 Alemtuzumab (A) and CHOP-14 (or CHOEP-14) versus CHOP-14 (or CHOEP-14) were evaluated in a collaboration between the Nordic and DSHNHL/GLA groups in parallel, phase III studies in younger (CHOEP-14,

18-65 years; ACT-1 trial)32 and older (CHOP-14, >65 years; ACT-2 trial)33 patients with newly diagnosed PTCL, the former incorporating auto-SCT into both treatment arms. In ACT-2, the 3-year event-free survival was 27% versus 24% in the A-CHOP and CHOP arms, respectively (P=0.248) and the experimental arm was associated with significant toxicity.33 Similarly, there was no benefit of A-CHOP observed in the ACT-1 trial (3-year event-free survival: A-CHOP 35% vs. 26%).32 Role of consolidative autologous transplant in front-line treatment of nodal PTCL: how strong is the evidence? With the high relapse rate in PTCL, auto-SCT is often considered in first remission. However, in the absence of randomized controlled studies, there is a lack of consensus and, as a result, guidelines, as well as clinical practice, vary. This is highlighted in the randomized studies above in which use of consolidative auto-SCT was at the investigators’ discretion in the ECHELON-2 (blinded) trial, not allowed in the Ro-CHOP study and was integrated into both treatment arms in the younger patients in the ACT-1 study.26,30,34 Overall, it is challenging to compare studies with the inclusion of diverse subtypes, variable responses leading into the transplant and analyses either in an ‘intent to transplant’ population or from the point of view of auto-SCT with the comparison group for the latter often including non-responders (Table 2).35-37 The largest prospective study evaluating consolidative auto-SCT was conducted by the Nordic Lymphoma Group (NLG-T-01),38 which enrolled 115 patients with newly diagnosed PTCL, excluding ALK+ ALCL, in 24 centers. Patients received CHOEP or CHOP-14 (>60 years) and 70% proceeded to auto-SCT. With a median follow-up of 5 years, the 5-year PFS and OS were 44% and 51%, respectively. With uncertainty around auto-SCT, the GLA and LYSA study groups conducted a phase III study comparing consolidative auto-SCT to allogeneic (allo)-SCT in newly diagnosed, poor-risk PTCL following four courses of CHOEP and one course of DHAP (dexamethasone, high-dose cytarabine, cisplatin); however, this study was stopped early due to futility.39 Ultimately, with an overall transplant rate of 65%, the 3-year event-free survival was 43% in the alloSCT group and 38% in the auto-SCT group. Notably, the 3year cumulative incidence of relapse was 40% in the auto-SCT group and 0% in the allo-SCT group, supporting a graft-versus-lymphoma effect; however, non-relapse mortality was significantly higher in the allo-SCT group (3year 23% vs. 0%), offsetting any overall benefit.39 Retrospective studies have given conflicting results regarding the benefit of consolidative auto-SCT (Table 2).8,9,40 The LYSA group performed a propensity score matched analysis of auto-SCT in an intention-to-treat population of nodal PTCL patients (n=269) and did not find a PFS or OS benefit

Haematologica | 108 December 2023

3214

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

of auto-SCT in multivariate analysis. In contrast, the NCR noted improved OS in patients with nodal PTCL (excluding ALK+ ALCL) treated in the more recent era when auto-SCT was more routinely applied. Furthermore, in a multivariate

analysis of patients diagnosed between 2014-2018 who received CHOP or CHOEP, omission of auto-SCT with primary therapy was associated with a higher risk of death in patients with non-ALK+ ALCL subtypes.9 As this analysis was

Table 2. Selected large studies evaluating consolidative autologous stem cell transplant in nodal peripheral T-cell lymphomas.

First author Study typea

Response Benefit of prior to auto-SCT auto-SCT CR/PR, %

Reimer108/ Wilhelm109 Phase II

Maybe

D’Amore38 Phase II NLG-01

Maybe

Yes (UVA)f Abramson41,b USA multicenter No (CR) Ellin8,c Swedish Registry Cederleuf42 Swedish/Danish Fossard40 LYSA Park110 COMPLETE Janikova111 CLSG GarciaSancho112,d GELTAMO/FIL Brink9 Netherlands Registry

Savage43 Phase III ECHELON-2 subgroup

Outcome of intent to Outcome after auto-SCT auto-SCT (vs. no (vs. no auto-SCT) auto-SCT) PFS, % OS, % PFS, % OS, %

Comment auto-SCT

3-yr, 71 (vs. 11)* 5-yr, 57 (vs. 23)*

5-yr analysis: 28 additional patients analyzed (on protocol)

5-yr, 51

5-yr, 61 (vs. 28)*

No auto-SCT group includes those with no response/PD

3-yr, 74 (vs. 53)

62/20e

3-yr, 36 5-yr, 39

3-yr, 48 5-yr, 44

53/31

5-yr, 44

No PFS/OS benefit of auto-SCT in MVA All CR: no PFS/OS benefit of auto-SCT PFS*/OS* benefit in ‘intent to auto-SCT’ group <70 yr (not adjusted for response)

61/12

3-yr, 58 (vs. 30)*

Yes

Auto-SCT MVA*

CR (All by design)

2-yr, 66 (vs. 67)

2-yr, 76 (vs. 80)

CR/PR 57

5-yr, 46 (vs. 40.5)

59 (vs. 60)

CR (All by design)

P=0.23

2-yr, 88 (vs. 70)

5-yr, 46 (vs. 41)

5-yr, 59.5 (vs. 49)

Yes

CR (All by design)

5-yr, 63 (vs. 49)*

5-yr, 74 (vs. 62)

All CR: PFS*/OS* benefit in MVA

Landmark 5-yr, 78 (vs. 45)* CR: 5-yr, 82 (vs. 47)*

MVA: improved OS*

All CR: PFS* benefit in BV-CHP arm (adjusted for age/region)

3-yr, 80 (vs. 55)* 5-yr, 65 (vs. 46)* 3-yr, 67 (vs. 54)g 5-yr, 49 (vs. 51)g

Yes

BV-CHP Yes CR (All by design)

CHOP No

MVA: no OS benefit No PFS/OS benefit using propensity score matching MVA: no PFS/OS benefit Improved OS with auto-SCT in IPI score 2-4 Adjusted by IPI score: no PFS/OS benefit with auto-SCT

All CR: no PFS benefit in CHOP arm (adjusted for age/region)

All included nodal PTCL and excluded ALK+ anaplastic large cell lymphoma (ALCL) with the exception of Abramson et al. (7% ALK+) and Cederleuf et al. (19% ALK+). bAlso included extranodal subtypes: total PTCL, not otherwise specified/angioimmunoblastic T-cell lymphoma/ALK– ALCL=67% (in addition to 6% ALK unknown). cAutologous stem cell transplantation (auto-SCT)-evaluated patients <70 years who received CHOP or CHOEP also included those with enteropathy associated T-cell lymphoma (EATL) and T-cell lymphoma ‘unspecified’. dAlso included patients with EATL, hepatosplenic T-cell lymphoma, NK/T-cell lymphoma, and primary cutaneous γδ lymphoma. eResponse after induction CHOP in 91/111 (82%) of patients, ultimately 75 (68%) transplanted. fUnivariate analysis of subset of patients treated with CHOP/CHOP-like versus CHOP/CHOP-like + autoSCT, not adjusted by response or prognostic factors. gP values not reported for this exploratory analysis, confidence intervals cross 1.0. BV-CHP: brentuximab vedotin - cyclophosphamide, doxorubicin, prednisone; CHOP: cyclophosphamide, doxorubicin, vincristine, prednisone; NLG: Nordic Lymphoma Group; LYSA: Lymphoma Study Group; CLSG: Czech Lymphoma Study Group; GELTAMO: Spanish Lymphoma Group; FIL: Fondazione Italiana Linfomi; auto-SCT: autologous stem cell transplantation; PFS: progression-free survival; OS: overall survival; yr: year; UVA: univariate analysis; MVA: multivariate analysis. Estimates are rounded; NR: not reported; IPI: International Prognostic Index. *Statistically significant. Haematologica | 108 December 2023

3215

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

not adjusted for response, a separate 9-month landmark analysis and a sensitivity analysis were performed exclusively in CR patients, which also demonstrated improved OS in the auto-SCT group.9 Given the variability in interpretation and definition of a partial response in retrospective studies, some studies have evaluated the role of auto-SCT only in patients in CR at the end of treatment, again with mixed results (Table 2).41,42 In a similar fashion, a subgroup analysis of patients in CR following BV-CHP enrolled on the ECHELON-2 study was performed and documented an improved PFS with consolidative SCT (auto-SCT n=36; alloSCT n=2) (5-year PFS: 65.3% vs. 46.4%, hazard ratio [HR]=0.36, 95% confidence interval [95% CI]: 0.17-0.77) but a similar benefit of SCT was not observed in the CHOP arm (Table 2).43 The recently activated TRANSCRIPT trial will address the role of auto-SCT in patients with nodal PTCL (excluding ALK+ ALCL) in CR following induction therapy (NCT05444712) (Online Supplementary Table S1). Despite data limitations, consolidative auto-SCT should still be a strategy to consider with upfront treatment but guidelines differ about whether it should be performed exclusively in patients in CR.34,44 Further studies are needed, ideally by subtype, to identify lower-risk patients in whom auto-SCT may be omitted and, conversely, determine whether there are molecular markers, such as P53 or DNMT3A mutations, that identify cases in which auto-SCT is futile.

removal if textured implants are used. Complete surgical excision in patients with stage 1 disease yields a 5-year disease-specific survival of 95%. For those presenting with stage 2 disease, there are limited data to guide recommendations. Surgery should include removal of the mass and sampling/removal of suspicious lymph nodes.45 With incomplete resection, radiation may be considered and in rare cases adjuvant BV has been administered although data supporting this approach are lacking. Although patients with BIA-ALCL were not included in ECHELON-2, BVCHP would be reasonable in those with disease outside of the breast and lymph nodes, or with lymph node involvement.45

Breast-implant-associated anaplastic large cell lymphoma After its original description in 1997, breast implant-associated (BIA) was defined as a provisional entity in the revised 4th edition of the WHO classification (WHO-HAEM4R) and was upgraded to a distinct entity by both the ICC and the WHO-HAEM51,2 (Table 1 in the Introduction to the Review Series3). Although not a ‘nodal’ PTCL, given its primary extranodal location, it is described here to distinguish it from systemic ALCL as the work-up, management and prognosis differ. The risk of BIA-ALCL is exclusively associated with textured implants and the time from implant to development of ALCL is 7-11 years. The overall risk varies in series but is likely between 1:1,000 to 1:10,000.45 Peri-implant effusion is the most common presentation with 85% of patients having stage 1 disease limited to the seroma ± capsule. The effusion, preferably as a large volume, should be sent for cytology with cell block preparation and flow cytometry including CD30 in the panel. A positron electron tomography scan should be done before surgery as post-surgical inflammatory changes can complicate interpretation.45 As recently reviewed,45 the mainstay of treatment is implant removal and en-bloc complete capsulectomy, with bilateral

Treatment options in relapsed/ refractory PTCL and the promise of personalized therapy Unfortunately, despite advances in the front-line setting a large proportion of PTCL patients have lymphoma relapse or have primary refractory disease. The only established curative treatment is SCT, although rare long-term remissions have been observed following systemic therapy alone, which, in some cases, may reflect more indolent disease biology.46-48 With the emergence of genomic techniques, there is a greater understanding of underlying disease biology which has also helped to inform therapeutics. This is best shown in TFHL, which are typified by recurrent mutations in epigenetic modifiers,5 with growing evidence of sensitivity to a broad range of agents of this class (Tables 3-5). In a proportion of ALCL and other rarer PTCL subtypes, there is evidence of JAK/STAT pathway activation, leading to recent trials with JAK inhibitors.49 Although studies of PTCL-NOS have elucidated the GATA3 and TBX21 molecular subtypes, how this informs treatment decisions remains unknown. Here, we review the overall management of R/R nodal PTCL, highlighting situations in which biology can guide treatment options. Transplant or no transplant? Outcomes are historically poor in patients with R/R PTCL, with a median OS from first relapse/progression typically <6 months in those who are not transplanted.46 Thus, the first therapeutic decision is whether or not a patient is a transplant candidate. SCT is limited to fit, often younger patients with chemosensitive disease, a term that should be redefined as ‘systemic therapy-sensitive’ with the expanding compendium of modern, novel therapies that also serve as an effective bridge to SCT.50-53 The prospective International T-Cell Project collected data on 633 patients with relapsed (n=197) or refractory (n=436) PTCL, including those managed with intent to transplant;

Haematologica | 108 December 2023

3216

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

the median OS for all patients was still only about 6 months, and the 3-year OS was 23%. Overall, only 99 patients (16%) underwent SCT (type not specified) as part of salvage therapy and, not surprisingly, this group had a superior 3-year OS of 48% compared to 30% in patients in a partial remission or CR and were not transplanted (for any reason).54 Data are more limited in patients managed with ‘intent to transplant’, especially with auto-SCT. Two retrospective, single-institution studies evaluating outcomes from the point of relapse/progression with intention to incorporate SCT suggested a cure rate of 20-35% with autoSCT, with dismal outcomes in patients with refractory disease.55,56 Outside of ALCL, most evidence supports the use of alloSCT in relapsed and especially refractory PTCL. The Center for International Bone Marrow Transplant Research (CIBMTR)57 evaluated outcomes in 241 patients with PTCL undergoing SCT between 1996 to 2006 and 2018. Confining the analysis to those beyond first CR, a superior outcome was observed with auto-SCT over allo-SCT (3-year OS 62% vs. 33%, respectively; P=0.0088) with a lower transplantrelated mortality (5% vs. 32%; P=0.0088), but PFS and relapse/progression rates were similar. This study was largely driven by a high proportion of ALCL patients who may derive the greatest benefit from auto-SCT in the relapse setting (3-year PFS 53%, 3-year OS 65%) compared to PTCL-NOS (3-year PFS 29%, 3-year OS 42%), although specific outcomes by ALK status were not reported. Only six patients with AITL underwent auto-SCT, limiting the evaluation of this group. The proportion of patients with refractory disease was not specified by subtype, a factor that also strongly influences the information on the utility of auto-SCT. Studies have highlighted favorable outcomes with allo-SCT particularly in R/R AITL, including a recent combined retrospective registry study from the European Society for Blood and Marrow Transplantation (EBMT) and the CIBMTR, which evaluated 1,942 PTCL patients (AITL, PTCL-NOS, ALCL) undergoing allo-SCT between 2008 and 2018 primarily with R/R disease (70%).52 Overall, the 3-year PFS was 50% and the 3-year OS was 60%, highlighting better outcomes in the more modern treatment era. Furthermore, using PTCL-NOS as the reference group, a reduced risk of lymphoma relapse (P<0.001) was observed for AITL, highlighting a unique sensitivity to the graft-versus-lymphoma effect, which is consistent with other studies.58-60 In contrast, an increased risk of relapse was observed for ALCL (HR=1.3, 95% CI: 1.1-1.6; P=0.01). Not surprisingly, patients in CR had better outcomes than those with a partial response or resistant disease (3-year PFS 57% vs. 47% vs. 36%, respectively; P<0.0001) with response remaining significant in a multivariate analysis. Although efforts should be made to achieve a deeper response, given that one-third can still be cured, patients should not be denied an allo-SCT if criteria for a

partial response are not met. Interestingly, outcomes after haplo-SCT and matched donor transplants were comparable, and use of intensified conditioning did not appear to be advantageous compared with non-myeloablative approaches.52 Collectively, studies suggest that auto-SCT may be considered in relapsed ALCL if not used with front-line therapy, but allo-SCT should be favored in AITL and patients with PTCL-NOS should most likely receive an alloSCT. Apart from some cases of ALK+ ALCL, for those with refractory disease, allo-SCT would be preferable. As outlined by Bisig, Savage and de Leval,5 the characteristic landscape of TFHL supports a multi-step process derived from underlying clonal hematopoiesis (see Figure 4 in the review published in this issue of Haematologica5). TET2 and DNMT3A mutations, which typify TFHL, are also common in clonal hematopoiesis. A recent study using bone marrow samples as well as flow-sorted bone marrow and peripheral blood samples supported the finding that clonal hematopoiesis is prevalent in AITL and showed that progenitor cells harboring identical TET2 and DNMT3A mutations can divergently evolve to AITL and myeloid neoplasms.61 With still limited data, it remains unclear whether the presence of clonal hematopoiesis should affect treatment choices. However, if available, testing for clonal hematopoiesis should ideally be performed as it may inform on the use of auto-SCT in older patients and whether an allo-SCT may be preferred in some settings, as well as follow-up surveillance. Is there a preferred salvage chemotherapy in transplanteligible relapsed/refractory PTCL? As for diffuse large B-cell lymphoma, there is no standard first-line salvage therapy for R/R PTCL patients, and very few studies have detailed outcomes specifically in PTCL. A subgroup analysis of the Canadian Cancer Trials Group phase III LY.12 study comparing GDP (gemcitabine, dexamethasone, cisplatin) and DHAP salvage chemotherapy before auto-SCT in R/R aggressive lymphomas, including PTCL (n=59), demonstrated an ORR of 36% with GDP (compared to 46% in aggressive B-cell lymphomas; P=0.12) which was similar to that achieved by DHAP.62,63 The population was high risk with most patients either having refractory disease (41%) or had relapsed within 1 year (37%), which may have contributed to overall lower response rates, regardless of the study arm. ICE (ifosfamide, carboplatin, etopsoside) chemotherapy is also frequently used,55 however, there are no comparative studies in PTCL. Given the high frequency of chemorefractory disease, novel agents have been increasingly used as a bridge to transplant as outlined below. Novel agents in the management of relapsed/refractory PTCL Over the past decade, there has been a pivot to perform novel therapy studies specifically in R/R PTCL and, more

Haematologica | 108 December 2023

3217

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

recently, even in specific subtypes. The majority are singlearm phase II studies and, apart from BV, drug approval may be country specific (Table 2 and Figure 1 from the Introduction to review Series3). The scope of agents under investigation is wide but very few are approved (Table 3).3 Pralatrexate, a folate analog metabolic inhibitor which competitively inhibits dihydrofolate reductase, was the first novel agent to be studied in PTCL after early studies showed preferential sensitivity in PTCL (including cutaneous T-cell lymphoma) compared to B-cell lymphomas.64 The PROPEL study51 evaluated 115 patients with R/R PTCL and the ORR for all patients was 29% (11% CR), the median PFS was 3.5 months, and the median duration of response was 10.1 months (Table 3). The response rate was notably lower in AITL (8%). Subsequent studies explored a different ramped up dosing schedule and use of leucovorin to mitigate mucositis (‘Columbia regimen’) which improved tolerance and appears to maintain efficacy.65 BV was developed for use in Hodgkin lymphoma and ALCL due to the disappointing efficacy of the nascent anti-CD30 antibody in these lymphomas. In the phase II registration study for systemic ALCL, patients were eligible following failure of front-line anthracycline-based therapy. The efficacy was striking with an ORR of 86% and CR rate of 57%. The median PFS and median duration of response were 13.3 months and 12.6 months, respectively47 (Table 3). In the 5year follow-up, the PFS was 39% and, overall, 14% of patients remained in CR following single-agent BV in the

absence of transplant, suggesting that cure was possible in a minority of patients.52 This led to global approval of BV in R/R ALCL (Table 3). In a separate study, BV was evaluated in R/R non-ALCL CD30+ PTCL and, although less impressive, efficacy was demonstrated with an ORR of 41% across the entire cohort (n=34) and 54% in AITL, but duration of response was short (all patients 7.6 months). Although not approved, it remains an option in R/R CD30+ PTCL especially as a bridge to SCT if funding is available.66 More recently, the ALK inhibitor crizotinib was approved by the Food and Drug Administration for patients up to 21 years old with ALK+ ALCL, based on a robust CR rate (81%)67 (Table 3). Alisertib is an aurora A kinase inhibitor that produced an ORR of 50% in eight PTCL patients enrolled in a phase I study of hematologic malignancies.68 This study was followed by the Lumiere study, which was the first randomized phase III study in R/R PTCL, and compared alisertib to investigators’ choice of therapy (pralatrexate, romidepsin, gemcitabine) (Table 4). It did not show superiority of alisertib (ORR 33% [alisertib] vs. 45% [comparators]; median PFS 3.8 months vs. 3.5 months, respectively). This study demonstrated that phase III trials were possible in this setting but also highlighted the challenges with disease heterogeneity.69 CCR4 is expressed in 30-40% of cases of PTCL and is associated with the GATA3 subtype of PTCL-NOS. Mogamulizumab is a CCR4 monoclonal antibody and was explored in R/R CCR4+ nodal PTCL in a phase II study in Japan. Among 29 patients, the ORR was 34% with CR in 17%; the median

Table 3. Approved novel agents for the treatment of relapsed/refractory peripheral T-cell lymphomas: global perspective.

Type of agent DHFR inhibitor

Study phase

Country approval

PTCL subtype(s)

USA/Canada

PTCL/tMF

29/11

10.1

3.5

14.5

ADC CD30

Global

ALCL

86/57

12.6a

13.3

Not reached*

Romidepsin50,71, b*

HDAC inhibitor

USA/Canada (de-listed)

PTCL AITL

25/15 27/19

17b -

4 -

11.3 -

Belinostat72

HDAC inhibitor

USA

PTCL AITL

26/11 45.5

13.6 -

1.6 -

7.9 -

Chidamide73

HDAC inhibitor

China

PTCL AITL

28/14 50/40

9.9

2.1

21.4

Forodesine113

PNP inhibitor

Japan

PTCL

25/10

10.4

1.9

15.6

CCR4 antibody

Japan

CCR4+ PTCLc (2014)

34/17

2.0

14.2

ALK inhibitor

USA 1-21 yr

ALK+ ALCL

88/81

Agent Pralatrexate51 Brentuximab vedotin66

Mogamulizumab114,c Crizotinib67

ORR/CR Median DoR Median PFS Median OS % in months in months in months

Updated analysis of brentuximab vedotin with a median follow-up of 22.3 months: the median duration of response (DoR) was 28 months. Updated analysis of romidepsin with a median follow-up of 71 months: the median DoR was 25.6 months. cPeripheral T-cell lymphoma (PTCL) subtypes: PTCL, not otherwise specified; angioimmunoblastic T-cell lymphoma; ALK– anaplastic large cell lymphoma. *Approval of the PTCL indication for romidepsin was withdrawn by the Food and Drug Administration in the USA (2021) and Canada (2023) following the negative results of the Ro-CHOP phase III study. ORR: objective response rate; CR: complete response; PFS: progression-free survival; OS: overall survival; DHFR: dihydrofolate reductase; tMF: transformed mycosis fungoides; ADC: antibody-drug conjugate; ALCL: anaplastic large cell lymphoma; AITL: gioimmunoblastic T-cell lymphomas; HDAC: histone deacetylase; PNP: purine nucleoside phosphorylase; NR: not reported. b

Haematologica | 108 December 2023

3218

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

PFS was 3 months (Table 3) and led to the approval of mogamulizumab in 2014 in Japan in CCR4+ nodal PTCL.114

Recurrent mutations in epigenetic modifying genes, including TET2, DNMT3A and IDH2R17 as well as the diseasespecific RhoAG17V mutation, characterize TFHL. As a group these lymphomas respond better to epigenetic therapies (Figure 1 in the Introduction to the Review Series3), showing much higher response rates than those achieved when considering all PTCL (Tables 3 and 4). Histone modifier gene mutations have been reported in 36% of PTCL-NOS and are associated with inferior PFS but may also define a group with an increased response to the HDAC inhibitor chidamide, suggesting that there may be a larger scope of patients who could benefit from a more personalized approach to therapy.70 Romidepsin was the first HDAC inhibitor approved for use in R/R PTCL and was associated with an ORR of 25% (CR

Epigenetic therapies and a personalized treatment approach in T-follicular helper cell lymphomas Epigenetics reflect changes in gene expression in the absence of DNA sequence changes and include histone modification, DNA methylation, noncoding RNA effects and chromatin reorganization. Abnormal activity of HDAC can affect gene expression with epigenetic silencing of tumor suppressor genes and oncogene activation. Aberrant epigenetic alterations play a role in the pathogenesis of TFHL.

Table 4. Selected novel agent/combination therapy in relapsed/refractory peripheral T-cell lymphomas.

Agents Study

Target

Phase

PTCL subtype (N)

ORR/CR %

Median DoR in months

Median PFS in months

Aurora kinase

III

PTCL (271 total)c

33/18 45/27

7.4 5.6

3.8 3.5

Immunomodulatory Anti-angiogenic

Duvelisib89

PI3K γδ

Cerdulatinib91

Dual JAK/SYK

PTCL (54) AITL (26) PTCL (78) AITL (21) PTCL (65) TFHL (29)

22/11 31/15 50/32 67/48 35 52

3.6 3.5 7.8 NR NR 12.9

2.5 4.6 3.6 NR NR 4.6

PTCL (53) Cohort 1 JAK/STAT+ Cohort 2 pSTAT3+ Cohort 3 unselected

25 33 29 12

8.4 7.5 14.7 Not reached

2.8 NR NR NR

PTCL (51) AITL (20) CXCL12 3’UTR (12) AITL (11)d

43/22 60 42/25 45/27

Not reached

NR NR

Alisertib69 vs. Investigators’ choicea Lumiere Lenalidomide82

Ruxolitinib49

JAK1/2

Golidocitinib JAKPOT892

JAK1

I/II

Tipifarnib90

Farnesyltransferase

Azacitidine76 vs. Investigators’ choiceb ORACLE

DNMT1

III

TFHL (86 total)

33/12 43/23

NR NR

5.6* 2.8

Valemetostat78

EZH2

PTCL (45) AITL

56/24 70.6

NR NR

PTCL (25) TFHL (15) PTCL (55) TFHL (19)

61/43 80/60 58/42 68/58

20.3 NR 8.1 NR

8.0 8.9 6.9 NR

PTCL (14)

71/29

Combination therapies Romidepsin + HDAC + DNMT1 azacitidine95 Romidepsin + HDAC + PI3K γδ duvelisib53 Romidepsin + HDAC + DHFR pralatrexate115

II TN/RR I I

Comparator ‘investigators’ choice’: gemcitabine (N=30), romidepsin (N=23), pralatrexate (N=80). bComparator ‘investigators’ choice’: gemcitabine (N=24), romidepsin (N=4), bendamustine (N=16). *P=0.042, however, the pre-specified P value was P=0.025. cAngioimmunoblastic T-cell lymphoma (N=61; 23% of all patients); objective response rate to alisertib 28% vs. 46% for comparators. dAngioimmunoblastic T-cell lymphoma 11/23 evaluable. Estimates are rounded. PTCL: peripheral T-cell lymphoma; ORR: objective response rate; CR: complete response; DoR: duration of response; PFS: progression-free survival; AITL: angioimmunoblastic T-cell lymphoma; PI3K: phosphoinositide 3 kinase; NR: not reported; JAK: Janus kinase; SYK: spleen tyrosine kinase; UTR: untranslated region; DNMT: DNA methyltransferase; TFHL: T-follicular helper cell lymphoma; EZH2: enhancer of zeste homolog; HDAC: histone deactylase; TN: treatment-naïve; R/R: relapsed/refractory; DHFR: dihydrofolate reductase. Haematologica | 108 December 2023

3219

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

15%) (Table 3).50 Although responses were infrequent and overall median PFS was only 3 months, some responses were notably durable with a median duration of 28 months with longer follow-up.50,71 Furthermore, 4/27 (15%) of patients enrolled with AITL still remained in CR over 3 years after entering the study.48 Unfortunately, in 2021, the PTCL indication for romidepsin was withdrawn from the USA market and more recently Canada has followed, due to the negative results of a phase III study evaluating RoCHOP in the first-line setting. Belinostat, a hydroxamic acid-derived pan-HDAC inhibitor, and chidamide, the only oral class I/II HDAC inhibitor, are approved in the USA and China, respectively, and have similar efficacy to romidepsin and also a higher ORR in AITL (ORR 46% with belinostat; 50% with chidamide) (Table 3).72,73 A retrospective multicenter study that compared the efficacy of HDAC inhibitors in TFHL versus PTCL-NOS confirmed a higher response rate in the former (ORR and CR 56.5% and 28.9%, respectively, in TFHL vs. 9.4% and 19.6%, respectively, in PTCL-NOS; P=0.0035)74 and in those PTCL cases with ‘typical’ AITL mutations as described above. Beyond HDAC inhibitors, other epigenetic therapies produce high response rates in TFHL (Table 4; Figure 1 in the Introduction to the Review Series3). Oral 5-azacitidine (CC486) is a hypomethylating agent that inhibits DNA methyltransferase and was first evaluated in a retrospective study of 12 patients with R/R AITL, five of whom had a concurrent myeloid neoplasm; the ORR was 75%, the CR rate was 50% and the median PFS was 15 months.75 This led to the recently reported ORACLE phase III study (NCT03593018) comparing oral azacitidine (n=42) to investigators’ choice of therapy (n=44) (romidepsin n=4, gemcitabine n=24, bendamustine n=16) in TFHL.61 The median PFS favored 5-azacitidine (5.6 months vs. 2.8 months; P=0.042) but did not reach the pre-specified significance level of P<0.025, suggesting that the study may have been underpowered. Interestingly, despite the favorable PFS, lower ORR and CR rate were observed with 5-azacitidine (ORR and CR 33% and 12%, respectively, with 5-azacitidine vs. 43% and 23%, respectively, with investigators’ choice of therapy) supporting that a greater proportion of patients may have had stable disease as best response, which is also reflected in a more favorable OS (median OS 18.4 months vs. 10.3 months) (HR=0.56, 95% CI: 0.323-0.961).76 A separate, ongoing, phase III study in Japan is comparing oral azacitidine to investigators’ choice of therapy (romidepsin or gemcitabine) in R/R AITL, but the results have not yet been reported (NCT03703375) (Online Supplementary Table S1). Valemetostat is a potent, selective dual inhibitor of enhancer of zeste homologs (EZH2 and EZH1) and stimulates the expression of pro-apoptotic and tumor suppressor genes (Table 4; Figure 1 in the Introduction to the Review Series3).77 A phase I dose-escalation trial reported an ORR

of 55.6% (24% CR) and 70.6% in AITL (Table 4).78 The phase II VALENTINE-PTCL01 trial (NCT04703192) has completed accrual, but results have not yet been reported. There is also emerging evidence to support the integration of epigenetic therapies into the front-line therapy of TFHL. An unplanned subgroup analysis of patients with TFHL enrolled on the LYSA Ro-CHOP study demonstrated a PFS benefit of Ro-CHOP versus CHOP (P=0.046)30 (Table 5). Belinostat-CHOP was evaluated in a phase I study of 23 patients, with AITL as the predominant subtype (n=10, 43%); the ORR was 86% and the CR rate was 57%.81 In a phase I study of vorinostat-CHOP, all evaluable patients (n=12) achieved a CR, providing additional support for this approach.79 Several additional, ongoing studies are integrating chidamide in the front-line setting (Online Supplementary Table S1). Similarly, a phase II study evaluated CHOP with oral 5-azacitidine (CC-486) in 20 newly diagnosed patients with PTCL, of whom 81% had TFHL.80 The azacitidine was given as ‘priming’ to enhance chemosensitization and deepen the response. The ORR and CR rate was 75%, rising to 88% for both in TFHL. Overall, half of all patients underwent consolidative auto-SCT and, with a median follow-up of 21 months, the 2-year PFS was 65.8% for all patients and 69.2% in patients with TFHL.80 The presence of TET2 mutations was associated with a more favorable PFS (P=0.014), whereas a trend towards an inferior PFS was observed in cases with a DNMT3A mutation. Building on this regimen, the Alliance group launched a randomized phase II study comparing CHOP (or CHOEP for patients <60 years old) with either the same backbone with oral azacitidine or duvelisib in a 1:1:1 design (Online Supplementary Table S1). Not surprisingly, adding a novel therapy to the CHOP backbone is associated with additional toxicity. Febrile neutropenia occurred in 21% and 10% of patients treated with Ro-CHOP and CHOP, respectively (Table 5).81

Beyond epigenetic therapies: promise in T-follicular helper lymphomas and beyond Lenalidomide is an immunomodulatory drug with a complex mechanism of action including direct effects on the tumor cell mediated by inhibition of the protein CEREBLON as well as anti-angiogenic and immunogenic effects through its impact on the tumor microenvironment (Figure 1 in the Introduction to the Review Series3). Previous studies have shown modest single-agent activity in R/R PTCL (ORR 22-30%, CR 0-11%) and a median PFS of approximately 3 months across studies82-85 including patients with AITL (ORR 31%, CR 15%).82 A phase II study of CHOP and lenalidomide in patients 60-80 years old with

Haematologica | 108 December 2023

3220

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

Table 5. Studies evaluating CHOP + epigenetic therapies versus epigenetic therapy combinations in treatment naïve PTCL: focus on T-follicular helper lymphomas.

Experimental regimen RomidepsinCHOP30 Phase III AzacitidineCHOP80 Phase II

PTCL Median FU subtype (N) in months PTCL (421)# TFHL (101) PTCL (20)# TFHL (17)

75/75 88/88

86/57-71b 89

PTCL (14) AITL (5)

PTCL (11 TN) PTCL (29)# AITL (11)

Belinostat-CHOP81 PTCL/tMF (23) Phase I AITL (10) VorinostatCHOP79,c Phase I Romidepsin + 5-azacitidine95 Phase II Romidepsin + lenalidomide97 Phase II

27.5

Grade 3/4 Grade 3/4 ORR/CR 2-year PFS 2-year OS ↓ neutrophils, % ↓ platelets, % % % % 64 49 (FN 21) 50 63/41 (Ro-CHOP) 43 63 33 (FN 10) 10 60/37 (CHOP) 36 + GCSF All ~50a 66 69

68 76

71 (FN 14) + GCSF All

30 (FN 17) + GCSF as per ASCO guidelines

93/93c

57 (FN 13) + GCSF schedule B

13.5 (All)

70/50 (TN)

75/30 85/38.5

54 (1-year)

76 (1-year)

40 (FN 12) (All) GCSF as per investigator discretion 45 (FN NR) (All) GCSF guidelines NR

Subgroup analysis of T-follicular helper lymphoma: point estimates are estimated from Kaplan-Meier curves. bComplete response 57% in cohort 3; 71% at maximum tolerated dose (1,000 mg/m2 belinostat x 5 days). cVorinostat days 5-14 schedule A 200 mg bid, schedule B 300 mg tid; two patients considered not evaluable because of early treatment discontinuation. #Excludes ALK+ anaplastic large cell lymphoma. Estimates are rounded. FU: follow-up; ORR: objective response rate; CR: complete response; PFD: progression-free survival; OS: overall survival; CHOP: cyclophosphamide, doxorubicin, vincristine, prednisone; Ro: romidepsin; TFH: T-follicular helper; FN: febrile neutropenia; GCSF: granulocyte colony-stimulating factor; AITL: angioimmunoblastic T-cell lymphoma; NR: not reported; ASCO: American Society of Clinical Oncology; TN: treatment-naïve; PTCL: peripheral T-cell lymphomas; tMF: transformed mycosis fungoides.

AITL demonstrated a CR rate of 41%, which was below the pre-specified target of 55%, although the 2-year PFS and OS rates of 42% and 59% were better than expected based on results from historical series.86 Duvelisib is a dual inhibitor of PI3K-δ and PI3K-γ and showed encouraging activity in a phase I study in which the ORR was 50%.87 The findings of the phase II PRIMO registration trial of duvelisib in R/R PTCL were reported at the 2021 American Society of Hematology (ASH) meeting. The ORR was 50% (32% CR); results were not detailed by histological subtype (Table 4).88,89 The GATA3 molecular subtype of PTCL-NOS is enriched for PI3K-induced signatures providing a potential rationale for duvelisib (Table 4; Figure 1 in the Introduction to the Review Series3). PI3K inhibitors can cause immune-mediated toxicities reminiscent of those caused by checkpoint inhibitors, and these toxicities have led to discontinuation of drug development in many settings in B-cell lymphomas/leukemias. In the PRIMO study, duvelisib was associated with grade 3/4 transaminitis in 27% of patients and pneumonitis in two patients, both of whom died.89 However, immune toxicities may be favorably modified by certain concurrent therapies (see below). Tipifarnib is an oral inhibitor of farnesyltransferase which reduces CXCL12, a chemokine that is highly expressed in AITL and some PTCL-NOS. In a phase II study of tipifarnib for the treatment of PTCL the ORR was 39.7% but was

56.3% in AITL (CR 28.5%) and wildtype CXCL12 genotype in PTCL-NOS was predictive of response (ORR 40%).90 With activation of the JAK/STAT pathway being a feature in many PTCL, including ALK+ ALCL and a subset of ALK– ALCL, a JAK inhibitor (ruxolitinib) and a dual JAK/SYK inhibitor (cerdulatinib) have also been explored (Table 4; Figure 1 in the Introduction to the Review Series3).49,91 Cerdulatinib produced an ORR of 35% across R/R PTCL, increasing to 52% for those with a TFH phenotype.91 Ruxolitinib was associated with an ORR of 33% in cases with activating mutations in JAK/STAT (cohort 1) and 29% in those pSTAT3-positive by immunohistochemistry (cohort 2).49 A preliminary report of a phase I/II study of golidocitinib (JAKPOT8), a selective JAK1 inhibitor, in R/R PTCL described an ORR of 43% (65% in AITL), a CR of 22% and a median duration of response that had not been reached at that time (Table 4).92 EBV is present in most AITL (70-80%) in the surrounding B-cell immunoblasts, as well as in a proportion of PTCLNOS, although some of the latter cases would be re-classified today as primary nodal EBV+ T-cell/NK-cell lymphoma by the WHO-HAEM5 if also positive for cytotoxic markers. Regardless, nanatinostat, which induces EBV kinase genes, and valganciclovir which is subsequently activated (i.e., ‘kick and kill’), are being evaluated in a phase II basket study (NAVAL-1) (Figure 1 in the Introduction to the Review Series3) including a cohort with

Haematologica | 108 December 2023

3221

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

nodal (non-ALCL) PTCL (Online Supplementary Table S1) following an initial report of an ORR of 40% (CR 19%). In the six patients with PTCL-NOS/AITL from the phase I study, the ORR was 67% (CR 50%).93

68%/58%) but interestingly, also less hepatic toxicity than lead-in treatment with duvelisib alone (40% vs. 8%), suggesting that romidepsin may offset the immunotoxicity of duvelisib. Long-term follow-up is needed to determine the curative potential of these chemotherapy-free approaches and it is important to note that the toxicities are not negligible (Table 5). There are ongoing studies evaluating both combination therapies as well as CHOP + novel therapy approaches in treatment-naïve nodal PTCL.

Are combination novel therapies ready for prime time in treatment-naïve patients? In an effort to deepen CR, activity in R/R PTCL trials has centered around combination therapies chosen to capitalize on complementary, additive or synergistic activities. In some cases, these chemotherapy-free combination therapies are also being evaluated in treatment-naïve cohorts, challenging the paradigm of upfront CHOP-based approaches (Table 5). A phase I study of romidepsin and 5-azacitidine noted sensitivity in T-cell lymphomas with an ORR of 73% in five patients in an expansion cohort, four of whom had a CR (AITL n=3).94 A phase II study followed in PTCL including both treatment-naïve and R/R PTCL cohorts, with the analysis including the five patients from the earlier study.95 Considering the evaluable patients, the ORR was 61% (CR 43%) and was 70% (CR 50%) in treatment-naïve patients (n=10) versus 54% (CR 38%) in R/R PTCL patients. With a median followup of 13.5 months, the median PFS for all patients was 8 months.95 For the TFHL subgroup, the ORR was 80% and the CR was 67%, with a median PFS of 8.9 months versus 2.3 months in other PTCL subtypes. Considering all patients, grade 3/4 thrombocytopenia occurred in 48%, grade 3/4 neutropenia in 40%, and febrile neutropenia in 12% (Table 5). Targeted mutation information was available for 15 patients: numerically higher ORR and CR rate were demonstrated in those with TET2 mutations (69% and 53%, respectively) compared to those with a wildtype genotype (40% and 20%, respectively); however, the differences did not reach statistical significance. A retrospective series evaluating azacitidine (oral or subcutaneous) and romidepsin demonstrated similar favorable efficacy in 26 patients with R/R PTCL, 23 of whom had TFHL, of which one was a composite with diffuse large B-cell lymphoma. The ORR was 76.9% (CR 53%) in the TFHL subgroup.96 The combination of azacitidine and romidepsin is being compared to investigators’ choice of therapy (belinostat, pralatrexate, gemcitabine) in an ongoing phase III trial (Online Supplementary Table S1). Romidepsin and lenalidomide were evaluated in a phase I study in treatment-naïve PTCL patients ≥60 years or those <60 years and not considered chemotherapy candidates by the treating physician. Of 20 evaluable patients, 13 (65%) had AITL. Overall the ORR was 75% (CR 30%), rising to 85% (CR 38.5%) in the patients with AITL97 (Table 5). Although not evaluated in a treatment-naïve population, a phase I study of duvelisib with romidepsin in R/R patients demonstrated encouraging efficacy (ORR 58%/42%; TFHL

What is the promise of immunotherapy in PTCL? The efficacy of therapy targeting the programmed death pathway, either through programmed cell death protein 1 (PD1) or its ligand (PDL1), has been well described in extranodal NK/T-cell lymphomas85 and with upregulation of PDL1 in primary nodal EBV+ T/NK-cell lymphoma, it may also be a potential therapeutic approach in this rare entity.4 There are limited data for the remaining nodal PTCL, but in all PTCL, there is a potential concern about hyperprogression given that PD1 on T cells may function as a tumor suppressor.98 In a phase I study of 12 patients with R/R PTCL, half of whom had AITL, the ORR was only 33% and median PFS <3 months with hyperprogression reported to occur in four patients.99 A combination study of romidepsin and pembrolizumab showed more encouraging efficacy (ORR 47.3%, CR 37%), but hyperprogression was reported in two patients, so further studies are needed.100 Cellular therapy is still in the development phase in PTCL and a full review is beyond the scope of this article, but some of the ongoing studies are highlighted in Online Supplementary Table S1. The majority of studies in PTCL have focused on CD30 as a target and some have reported CR in ALCL but the numbers of patients and the follow-up time are limited and thus the curative potential is still unknown.101,102 Third-generation products utilizing the CD28 (CD28z) co-stimulatory domain may persist long term and appear to have a more potent anti-tumor effect.102 Apart from CD30, chimeric antigen receptor (CAR) T-cell targets in PTCL have been challenging due to three main barriers: (i) T-cell aplasia; (ii) fratricide; and (iii) the potential for contamination of CAR T-cell products with malignant cells.103 Strategies to circumvent fratricide include capitalizing on the selection of either TRBC1 or TRBC2 for the βchain constant region to spare normal T cells104 as detailed in the phase I/II AUTO4 trial (NCT03590574) evaluating TRBC1+ PTCL as well as use of NK-cell CAR products which do not express T-cell antigens.105 Early results from AUTO4 demonstrated that five of nine patients with PTCL achieved a CR, but lack of CAR T-cell expansion may limit the durability of responses.106 Use of ‘off the shelf’ products, such

Haematologica | 108 December 2023

3222

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

as allogeneic CD70 CRISPR-Cas9-engineered T cells from healthy donors which incorporate editing of T-cell receptor α and β2-microglobulin genes (NCT04502446 COBALT-LYM) avoids normal T-cell killing and does not rely on personalized manufacturing. AFM13, an innate bispecific CD16/CD30 antibody, is also under investigation in CD30+ PTCL (REDIRECT NCT04101331), including ALCL, demonstrating ORR of 40% in a phase Ib/II study (NCT03192202).107

ment, but definitive recommendations are difficult to make in the absence of randomized data. Recent studies highlight the sensitivity of TFHL to epigenetic therapies and, in the future, genomic information may also inform therapy. Future studies should focus on the evaluation of new treatments in specific PTCL subtypes or molecularly defined subgroups, to further refine personalized therapeutic options across a broader range of PTCL.

Disclosures HSN has no conflicts of interest to disclose. KJS has received Conclusions honoraria from and provided consulting for BMS, Merck, Advances in the understanding of the biology and molecu- Seagen, and Janssen; has sat on a steering committee for lar underpinnings of PTCL have refined both classifications Beigene; has received research funding from BMS; has reand therapeutic approaches. BV-CHP has changed the ceived institutional research funding from Roche; and sat on treatment landscape in ALCL and may be considered in a Data Safety and Monitoring Committee for DSMC. selected cases of CD30+ PTCL-NOS/AITL but it is recognized that the data are not definitive in the latter subtypes. Contributions Auto-SCT still remains a possible choice in upfront treat- KJS and HSN co-wrote the paper.

lymphoma is clinically and immunophenotypically different from both ALK+ ALCL and peripheral T-cell lymphoma, not otherwise specified: report from the International Peripheral T-Cell Lymphoma Project. Blood. 2008;111(12):5496-5504. 11. Shustov A, Cabrera ME, Civallero M, et al. ALK-negative anaplastic large cell lymphoma: features and outcomes of 235 patients from the International T-Cell Project. Blood Adv. 2021;5(3):640-648. 12. Sibon D, Nguyen DP, Schmitz N, et al. ALK-positive anaplastic large-cell lymphoma in adults: an individual patient data pooled analysis of 263 patients. Haematologica. 2019;104(12):e562-e565. 13. Federico M, Bellei M, Marcheselli L, et al. Peripheral T cell lymphoma, not otherwise specified (PTCL-NOS). A new prognostic model developed by the International T cell Project Network. Br J Haematol. 2018;181(6):760-769. 14. Advani R, Skrypets T, Civallero M, et al. Outcomes and prognostic factors in angioimmunoblastic T-cell lymphoma: final report from the International T-Cell Project. Blood. 2021;138(3):213-220. 15. Sibon D, Fournier M, Brière J, et al. Long-term outcome of adults with systemic anaplastic large-cell lymphoma treated within the Groupe d’Étude des Lymphomes de l’Adulte trials. J Clin Oncol. 2012;30(32):3939-3946. 16. Johnson WT, Ganesan N, Epstein-Peterson ZD, et al. TP53 mutations identify high-risk events for peripheral T-cell lymphoma treated with CHOP-based chemotherapy. Blood Adv. 2023;7(17):5172-5186. 17. Maurer MJ, Ghesquières H, Jais JP, et al. Event-free survival at 24 months is a robust end point for disease-related outcome in diffuse large B-cell lymphoma treated with immunochemotherapy. J Clin Oncol. 2014;32(10):1066-1073. 18. Maurer MJ, Ellin F, Jerkeman M, et al. International assessment of event-free survival at 24 months and subsequent survival in peripheral T-cell lymphoma. J Clin Oncol. 2017;35(36):4019-4026. 19. Briski R, Feldman AL, Bailey NG, et al. The role of front-line anthracycline-containing chemotherapy regimens in peripheral

Haematologica | 108 December 2023

3223

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

T-cell lymphomas. Blood Cancer J. 2014;4(5):e214. 20. Gleeson M, Peckitt C, To YM, et al. CHOP versus GEM-P in previously untreated patients with peripheral T-cell lymphoma (CHEMO-T): a phase 2, multicentre, randomised, open-label trial. Lancet Haematol. 2018;5(5):e190-e200. 21. Martín García‐Sancho A, Rodríguez‐Pinilla SM, Domingo‐Domenech E, et al. Peripheral T-cell lymphoma with a T follicular-helper phenotype: a different entity? Results of the Spanish Real-T study. Br J Haematol. 2023;203(2):182-193. 22. Schmitz N, Trümper L, Ziepert M, et al. Treatment and prognosis of mature T-cell and NK-cell lymphoma: an analysis of patients with T-cell lymphoma treated in studies of the German HighGrade Non-Hodgkin Lymphoma Study Group. Blood. 2010;116(18):3418-3425. 23. Simon A, Peoch M, Casassus P, et al. Upfront VIP-reinforcedABVD (VIP-rABVD) is not superior to CHOP/21 in newly diagnosed peripheral T cell lymphoma. Results of the randomized phase III trial GOELAMS-LTP95. Br J Haematol. 2010;151(2):159-166. 24. Fanale MA, Horwitz SM, Forero-Torres A, et al. Brentuximab vedotin in the front-line treatment of patients with CD30+ peripheral T-cell lymphomas: results of a phase I study. J Clin Oncol. 2014;32(28):3137-3143. 25. Horwitz S, O’Connor OA, Pro B, et al. Brentuximab vedotin with chemotherapy for CD30-positive peripheral T-cell lymphoma (ECHELON-2): a global, double-blind, randomised, phase 3 trial. Lancet. 2019;393(10168):229-240. 26. Horwitz S, O’Connor OA, Pro B, et al. The ECHELON-2 trial: 5year results of a randomized, phase III study of brentuximab vedotin with chemotherapy for CD30-positive peripheral T-cell lymphoma. Ann Oncol. 2022;33(3):288-298. 27. Horwitz SM, Advani RH, Bartlett NL, et al. Objective responses in relapsed T-cell lymphomas with single-agent brentuximab vedotin. Blood. 2014;123(20):3095-3100. 28. Gardai SJ, Epp A, Law C-L. Brentuximab vedotin-mediated immunogenic cell death. Cancer Res. 2015;75(15_Suppl):2469. 29. Herrera AF, Zain J, Savage KJ, et al. Brentuximab vedotin plus cyclophosphamide, doxorubicin, etoposide, and prednisone (CHEP-BV) followed by BV consolidation in patients with CD30expressing peripheral T-cell lymphomas. Blood. 2021;138(Suppl 1):133. 30. Bachy E, Camus V, Thieblemont C, et al. Romidepsin plus CHOP versus CHOP in patients with previously untreated peripheral Tcell lymphoma: results of the Ro-CHOP phase III study (conducted by LYSA). J Clin Oncol. 2021;40242-40251. 31. Rodig SJ, Abramson JS, Pinkus GS, et al. Heterogeneous CD52 expression among hematologic neoplasms: implications for the use of alemtuzumab (CAMPATH-1H). Clin Cancer Res. 2006;12(23):7174-7179. 32. d’Amore F, Leppä S, Silva MG da, et al. Final analysis of the front-line phase III randomized ACT-1 trial in younger patients with systemic peripheral T-cell lymphoma treated with CHOP chemotherapy with or without alemtuzumab and consolidated by autologous hematopoietic stem cell transplant. Blood. 2018;132(Suppl 1):998. 33. Wulf GG, Altmann B, Ziepert M, et al. Alemtuzumab plus CHOP versus CHOP in elderly patients with peripheral T-cell lymphoma: the DSHNHL2006-1B/ACT-2 trial. Leukemia. 2021;35(1):143-155. 34. d’Amore F, Gaulard P, Trümper L, et al. Peripheral T-cell lymphomas: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2015;26(Suppl 5):v108–v115. 35. Blystad AK, Enblad G, Kvaløy S, et al. High-dose therapy with

autologous stem cell transplantation in patients with peripheral T cell lymphomas. Bone Marrow Transplant. 2001;27(7):711-716. 36. Rodríguez J, Caballero MD, Gutiérrez A, et al. High-dose chemotherapy and autologous stem cell transplantation in peripheral T-cell lymphoma: the GEL-TAMO experience. Ann Oncol. 2003;14(12):1768-1775. 37. Jantunen E, Wiklund T, Juvonen E, et al. Autologous stem cell transplantation in adult patients with peripheral T-cell lymphoma: a nation-wide survey. Bone Marrow Transplant. 2004;33(4):405-410. 38. D’Amore F, Relander T, Lauritzsen GF, et al. Up-front autologous stem-cell transplantation in peripheral T-cell lymphoma: NLG-T01. J Clin. Oncol. 2012;30(25):3093-3099. 39. Schmitz N, Truemper L, Bouabdallah K, et al. A randomized phase 3 trial of autologous vs allogeneic transplantation as part of first-line therapy in poor-risk peripheral T-NHL. Blood. 2021;137(19):2646-2656. 40. Fossard G, Broussais F, Coelho I, et al. Role of up-front autologous stem-cell transplantation in peripheral T-cell lymphoma for patients in response after induction: an analysis of patients from LYSA centers. Ann Oncol. 2018;29(3):715-723. 41. Abramson JS, Feldman T, Kroll-Desrosiers AR, et al. Peripheral T-cell lymphomas in a large US multicenter cohort: prognostication in the modern era including impact of frontline therapy. Ann Oncol. 2014;25(11):2211-2217. 42. Cederleuf H, Hjort Jakobsen L, Ellin F, et al. Outcome of peripheral T-cell lymphoma in first complete remission: a Danish-Swedish population-based study. Leuk Lymphoma. 2017;58(12):2815-2823. 43. Savage KJ, Horwitz SM, Advani R, et al. Role of stem cell transplant in CD30+ PTCL following frontline brentuximab vedotin plus CHP or CHOP in ECHELON-2. Blood Adv. 2022;6(19):5550-5555. 44. Horwitz SM, Ansell S, Ai WZ, et al. T-cell lymphomas, version 2.2022, NCCN clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2022;20(3):285-308. 45. Mehta-Shah N, Ghione P. An updated approach and understanding of breast implant-associated anaplastic large cell lymphoma. J Natl Compr Canc Netw. 2022;20(3):309-315. 46. Mak V, Hamm J, Chhanabhai M, et al. Survival of patients with peripheral T-cell lymphoma after first relapse or progression: spectrum of disease and rare long-term survivors. J Clin Oncol. 2013;31(16):1970-1976. 47. Pro B, Advani R, Brice P, et al. Brentuximab vedotin (SGN-35) in patients with relapsed or refractory systemic anaplastic largecell lymphoma: results of a phase II study. J Clin Oncol. 2012;30(18):2190-2196. 48. Pro B, Horwitz SM, Prince HM, et al. Romidepsin induces durable responses in patients with relapsed or refractory angioimmunoblastic T‐cell lymphoma. Hematol Oncol. 2017;35(4):914-917. 49. Moskowitz AJ, Ghione P, Jacobsen E, et al. A phase 2 biomarkerdriven study of ruxolitinib demonstrates effectiveness of JAK/STAT targeting in T-cell lymphomas. Blood. 2021;138(26):2828-2837. 50. Coiffier B, Pro B, Prince HM, et al. Results from a pivotal, openlabel, phase II study of romidepsin in relapsed or refractory peripheral T-cell lymphoma after prior systemic therapy. J Clin Oncol. 2012;30(6):631-636. 51. O’Connor OA, Pro B, Pinter-Brown L, et al. Pralatrexate in patients with relapsed or refractory peripheral T-cell lymphoma: results from the pivotal PROPEL study. J Clin Oncol. 2011;29(9):1182-1189. 52. Pro B, Advani R, Brice P, et al. Five-year results of brentuximab

Haematologica | 108 December 2023

3224

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

vedotin in patients with relapsed or refractory systemic anaplastic large cell lymphoma. Blood. 2017;130(25):2709-2717. 53. Horwitz SM, Moskowitz AJ, Mehta-Shah N, et al. The combination of duvelisib and romidepsin (DR) is highly active against relapsed/refractory peripheral T-cell lymphoma with low rates of transaminitis: final results. Hematol Oncol. 2021;39(S2):100-101. 54. Bellei M, Foss FM, Shustov AR, et al. The outcome of peripheral T-cell lymphoma patients failing first-line therapy: a report from the prospective, International T-Cell Project. Haematologica. 2018;103(7):1191-1197. 55. Horwitz S, Moskowitz C, Kewalramani T, et al. Second-line therapy with ICE followed by high dose therapy and autologous stem cell transplantation for relapsed/refractory peripheral Tcell lymphomas: minimal benefit when analyzed by intent to treat. Blood. 2005;106(11):2679. 56. Ngu HS, Parkin S, Scott DW, et al. Outcome of relapsed and refractory peripheral T-cell lymphoma (PTCL) with intention for curative therapy incorporating high dose chemotherapy and hematopoietic stem cell transplant (HDC/SCT). Blood. 2021;138(Suppl 1):624. 57. Smith SM, Burns LJ, Inwards DJ, et al. Hematopoietic cell transplantation for systemic mature T-cell non-Hodgkin lymphoma. J Clin Oncol. 2013;31(25):3100-3109. 58. Le Gouill S, Milpied N, Buzyn A, et al. Graft-versus-lymphoma effect for aggressive T-cell lymphomas in adults: a study by the Société Francaise de Greffe de Moëlle et de Thérapie Cellulaire. J Clin Oncol. 2008;26(14):2264-2271. 59. Epperla N, Ahn KW, Litovich C, et al. Allogeneic hematopoietic cell transplantation provides effective salvage despite refractory disease or failed prior autologous transplant in angioimmunoblastic T-cell lymphoma: a CIBMTR analysis. J Hematol Oncol. 2019;12(1):6. 60. Hamadani M, Ngoya M, Sureda A, et al. Outcome of allogeneic transplantation for mature T-cell lymphomas: impact of donor source and disease characteristics. Blood Adv. 2022;6(3):920-930. 61. Lewis NE, Petrova-Drus K, Huet S, et al. Clonal hematopoiesis in angioimmunoblastic T-cell lymphoma with divergent evolution to myeloid neoplasms. Blood Adv. 2020;4(10):2261-2271. 62. Crump M, Kuruvilla J, Couban S, et al. Randomized comparison of gemcitabine, dexamethasone, and cisplatin versus dexamethasone, cytarabine, and cisplatin chemotherapy before autologous stem-cell transplantation for relapsed and refractory aggressive lymphomas: NCIC-CTG LY.12. J Clin Oncol. 2014;32(31):3490-3496. 63. Skamene T, Crump M, Savage KJ, et al. Salvage chemotherapy and autologous stem cell transplantation for peripheral T-cell lymphoma: a subset analysis of the Canadian Cancer Trials Group LY.12 randomized phase 3 study. Leuk Lymphoma. 2017;58(10):2319-2327. 64. O’Connor OA, Hamlin PA, Portlock C, et al. Pralatrexate, a novel class of antifol with high affinity for the reduced folate carriertype 1, produces marked complete and durable remissions in a diversity of chemotherapy refractory cases of T-cell lymphoma. Br J Haematol. 2007;139(3):425-428. 65. O’Connor OA, Amengual J, Colbourn D, Deng C, Sawas A. Pralatrexate: a comprehensive update on pharmacology, clinical activity and strategies to optimize use. Leuk Lymphoma. 2017;58(11):2548-2557. 66. Horwitz SM, Advani RH, Bartlett NL, et al. Objective responses in relapsed T-cell lymphomas with single-agent brentuximab vedotin. Blood. 2014;123(20):3095-3100. 67. Merino M, Kasamon Y, Li H, et al. FDA approval summary:

crizotinib for pediatric and young adult patients with relapsed or refractory systemic anaplastic large cell lymphoma. Pediatr Blood Cancer. 2022;69(8):e29602. 68. Friedberg JW, Mahadevan D, Cebula E, et al. Phase II study of alisertib, a selective aurora A kinase inhibitor, in relapsed and refractory aggressive B- and T-cell non-Hodgkin lymphomas. J Clin Oncol. 2014;32(1):44-50. 69. O’Connor OA, Ozcan M, Jacobsen ED, et al. Randomized phase III study of alisertib or investigator’s choice (selected single agent) in patients with relapsed or refractory peripheral T-cell lymphoma. J Clin Oncol. 2019;37(8):613-623. 70. Ji MM, Huang YH, Huang JY, et al. Histone modifier gene mutations in peripheral T-cell lymphoma not otherwise specified. Haematologica. 2018;103(4):679-687. 71. Coiffier B, Pro B, Prince HM, et al. Romidepsin for the treatment of relapsed/refractory peripheral T-cell lymphoma: pivotal study update demonstrates durable responses. J Hematol Oncol. 2014;7(1):11. 72. O’Connor OA, Horwitz S, Masszi T, et al. Belinostat in patients with relapsed or refractory peripheral T-cell lymphoma: results of the pivotal phase II BELIEF (CLN-19) study. J Clin Oncol. 2015;33(23):2492-2499. 73. Shi YK, Dong M, Hong X, et al. Results from a multicenter, openlabel, pivotal phase II study of chidamide in relapsed or refractory peripheral T-cell lymphoma. Ann Oncol. 2015;26(8):1766-1771. 74. Ghione P, Faruque P, Mehta-Shah N, et al. T follicular helper phenotype predicts response to histone deacetylase inhibitors in relapsed/refractory peripheral T-cell lymphoma. Blood Adv. 2020;4(19):4640-4647. 75. Lemonnier F, Dupuis J, Sujobert P, et al. Treatment with 5azacytidine induces a sustained response in patients with angioimmunoblastic T-cell lymphoma. Blood. 2018;132(21):2305-2309. 76. Dupuis J, Tsukasaki K, Bachy E, et al. Oral azacytidine in patients with relapsed/refractory angioimmunoblastic T-cell lymphoma: final analysis of the ORACLE phase III study. Blood. 2022;140(Suppl 1):2310-2312. 77. Honma D, Kanno O, Watanabe J, et al. Novel orally bioavailable EZH1/2 dual inhibitors with greater antitumor efficacy than an EZH2 selective inhibitor. Cancer Sci. 2017;108(10):2069-2078. 78. Ishitsuka K, Izutsu K, Maruyama D, et al. First-in-human study of the EZH1 and EZH2 dual inhibitor (DS-3201b) in patients with relapsed or refractory non-Hodgkin lymphomas. Hematol Oncol. 2021;39(s2):38-40. 79. Oki Y, Younes A, Copeland A, et al. Phase I study of vorinostat in combination with standard CHOP in patients with newly diagnosed peripheral T-cell lymphoma. Br J Haematol. 2013;162(1):138-141. 80. Ruan J, Moskowitz AJ, Mehta-Shah N, et al. Multicenter phase 2 study of oral azacitidine (CC-486) plus CHOP as initial treatment for peripheral T-cell lymphoma. Blood. 2023;141(18):2194-2205. 81. Johnston PB, Cashen AF, Nikolinakos PG, et al. Belinostat in combination with standard cyclophosphamide, doxorubicin, vincristine and prednisone as first-line treatment for patients with newly diagnosed peripheral T-cell lymphoma. Exp Hematol Oncol. 2021;10(1):1-11. 82. Morschhauser F, Fitoussi O, Haioun C, et al. A phase 2, multicentre, single-arm, open-label study to evaluate the safety and efficacy of single-agent lenalidomide (Revlimid®) in subjects with relapsed or refractory peripheral T-cell nonHodgkin lymphoma: the EXPECT trial. Eur J Cancer. 2013;49(13):2869-2876. 83. Fabbri A, Cencini E, Pietrini A, et al. Impressive activity of

Haematologica | 108 December 2023

3225

REVIEW SERIES - Therapeutic approaches in nodal PTCL

H.S. Ngu and K.J. Savage

lenalidomide monotherapy in refractory angioimmunoblastic Tcell lymphoma: report of a case with long-term follow-up. Hematol Oncol. 2013;31(4):213-217. 84. Dueck G, Chua N, Prasad A, et al. Interim report of a phase 2 clinical trial of lenalidomide for T-cell non-Hodgkin lymphoma. Cancer. 2010;116(19):4541-4548. 85. Toumishey E, Prasad A, Dueck G, et al. Final report of a phase 2 clinical trial of lenalidomide monotherapy for patients with Tcell lymphoma. Cancer. 2015;121(5):716-723. 86. Lemonnier F, Safar V, Beldi-Ferchiou A, et al. Integrative analysis of a phase 2 trial combining lenalidomide with CHOP in angioimmunoblastic T-cell lymphoma. Blood Adv. 2021;5(2):539-548. 87. Horwitz SM, Koch R, Porcu P, et al. Activity of the PI3K-δ,γ inhibitor duvelisib in a phase 1 trial and preclinical models of Tcell lymphoma. Blood. 2018;131(8):888-898. 88. Pro B, Brammer JE, Casulo C, et al. Duvelisib in patients with relapsed/refractory peripheral T-cell lymphoma from the phase 2 PRIMO trial: dose optimization efficacy update and expansion phase initial results. Blood. 2020;136(Suppl 1):38-39. 89. Brammer JE, Zinzani PL, Zain J, et al. Duvelisib in patients with relapsed/refractory peripheral T-cell lymphoma from the phase 2 PRIMO trial: results of an interim analysis. Blood. 2021;138(Suppl 1):2456. 90. Witzig TE, Sokol L, Foss FM, et al. Proof of concept for tipifarnib in relapsed or refractory angioimmunoblastic T-cell lymphoma (AITL) and CXCL12+ peripheral T-cell lymphoma (PTCL): preliminary results from an open-label, phase 2 study. Blood. 2019;134(Suppl_1):468. 91. Horwitz SM, Feldman TA, Ye JC, et al. Phase 2a study of the dual SYK/JAK inhibitor cerdulatinib (ALXN2075) as monotherapy in patients with relapsed/refractory peripheral T-cell lymphoma. Blood. 2021;138(Suppl 1):622. 92. Kim WS, Yoon DH, Song Y, et al. A phase I/II study of golidocitinib, a selective JAK1 inhibitor, in refractory or relapsed peripheral T-cell lymphoma. J Clin Oncol. 2022;40(16_suppl):7563. 93. Haverkos BM, Alpdogan O, Baiocchi R, et al. Nanatinostat (Nstat) and valganciclovir (VGCV) in relapsed/refractory (R/R) EpsteinBarr virus-positive (EBV +) lymphomas: final results from the phase 1b/2 VT3996-201 study. Blood. 2021;138(Suppl 1):623. 94. O’Connor OA, Falchi L, Lue JK, et al. Oral 5-azacytidine and romidepsin exhibit marked activity in patients with PTCL: a multicenter phase 1 study. Blood. 2019;134(17):1395-1405. 95. Falchi L, Ma H, Klein S, et al. Combined oral 5-azacytidine and romidepsin are highly effective in patients with PTCL: a multicenter phase 2 study. Blood. 2021;137(16):2161-2170. 96. Kalac M, Jain S, Tam CS, et al. Real-world experience of combined treatment with azacitidine and romidepsin in patients with peripheral T-cell lymphoma. Blood Adv. 2023;7(14):3760-3763. 97. Ruan J, Zain JM, Palmer B, et al. Multicenter phase II study of romidepsin plus lenalidomide for patients with previously untreated peripheral T-cell lymphoma (PTCL). J Clin Oncol. 2021;39(15_suppl):7514. 98. Rauch DA, Conlon KC, Janakiram M, et al. Rapid progression of adult T-cell leukemia/lymphoma as tumor-infiltrating Tregs after PD-1 blockade. Blood. 2019;134(17):1406-1414. 99. Nora Bennani N, Kim HJ, Pederson LD, et al. Nivolumab in patients with relapsed or refractory peripheral T-cell lymphoma: modest activity and cases of hyperprogression. J Immunother Cancer. 2022;10(6):e004984.

100. Agbedia OO, Prakash R, Xu J, et al. Updated results of an investigator-initiated phase II study of pembrolizumab and romidepsin for patients with relapsed or refractory T-cell lymphoma (TCL) with survival analysis. Blood. 2022;140(Suppl 1):2313-2315. 101. Ramos CA, Ballard B, Zhang H, et al. Clinical and immunological responses after CD30-specific chimeric antigen receptorredirected lymphocytes. J Clin Invest. 2017;127(9):3462-3471. 102. Zhang S, Gu C, Huang L, et al. The third-generation anti-CD30 CAR T-cells specifically homing to the tumor and mediating powerful antitumor activity. Sci Rep. 2022;12(1):10488. 103. Fleischer LC, Spencer HT, Raikar SS. Targeting T cell malignancies using CAR-based immunotherapy: challenges and potential solutions. J Hematol Oncol. 2019;12(1):141. 104. Maciocia PM, Wawrzyniecka PA, Philip B, et al. Targeting the T cell receptor β-chain constant region for immunotherapy of T cell malignancies. Nat Med. 2017;23(12):1416-1423. 105. Daher M, Melo Garcia L, Li Y, Rezvani K. CAR‐NK cells: the next wave of cellular therapy for cancer. Clin Transl Immunol. 2021;10(4):e1274. 106. Cwynarski K, Iacoboni G, Tholouli E, et al. First in human study of AUTO4, a TRBC1-targeting CAR T-cell therapy in relapsed/refractory TRBC1-positive peripheral T-cell lymphoma. Blood. 2022;140(Suppl 1):10316-10317. 107. Sawas A, Elgedawe H, Vlad G, et al. Clinical and biological evaluation of the novel CD30/CD16A tetravalent bispecific antibody (AFM13) in relapsed or refractory CD30-positive lymphoma with cutaneous presentation: a biomarker phase Ib/IIa study (NCT03192202). Blood. 2018;132(Suppl 1):2908. 108. Reimer P, Rüdiger T, Geissinger E, et al. Autologous stem-cell transplantation as first-line therapy in peripheral T-cell lymphomas: results of a prospective multicenter study. J Clinl Oncol. 2009;27(1):106-113. 109. Wilhelm M, Smetak M, Reimer P, et al. First-line therapy of peripheral T-cell lymphoma: extension and long-term follow-up of a study investigating the role of autologous stem cell transplantation. Blood Cancer J. 2016;6(7):e452. 110. Park SI, Horwitz SM, Foss FM, et al. The role of autologous stem cell transplantation in patients with nodal peripheral T-cell lymphomas in first complete remission: report from COMPLETE, a prospective, multicenter cohort study. Cancer. 2019;125(9):1507-1517. 111. Janikova A, Chloupkova R, Campr V, et al. First-line therapy for T cell lymphomas: a retrospective population-based analysis of 906 T cell lymphoma patients. Ann Hematol. 2019;98(8):1961-1972. 112. García-Sancho AM, Bellei M, López-Parra M, et al. Autologous stem-cell transplantation as consolidation of first-line chemotherapy in patients with peripheral T-cell lymphoma: a multicenter GELTAMO/FIL study. Haematologica. 2022;107(11):2675-2684. 113. Maruyama D, Tsukasaki K, Uchida T, et al. Multicenter phase 1/2 study of forodesine in patients with relapsed peripheral T cell lymphoma. Ann Hematol. 2019;98(1):131-142. 114. Ogura M, Ishida T, Inagaki H, et al. Multicenter phase II study of mogamulizumab (KW-0761), a defucosylated anti-CC chemokine receptor 4 antibody, in patients with relapsed peripheral T-cell lymphoma and cutaneous T-cell lymphoma. J Clin Oncol. 2014;32(11):1157-1163. 115. Amengual JE, Lichtenstein R, Lue J, et al. A phase 1 study of romidepsin and pralatrexate reveals marked activity in relapsed and refractory T-cell lymphoma. Blood. 2018;131(4):397-407.

Haematologica | 108 December 2023

3226

REVIEW SERIES

Pathobiology of nodal peripheral T-cell lymphomas: current understanding and future directions Bettina Bisig,1 Kerry J. Savage2 and Laurence de Leval1 Correspondence: Prof. L. de Leval Institute of Pathology, Department of Laboratory Medicine and Pathology, Lausanne University Hospital and Lausanne University, Lausanne, Switzerland and 2Centre for Lymphoid Cancer, Division of Medical Oncology, BC Cancer and University of British Columbia, Vancouver, British Columbia, Canada 1

Laurence.deLeval@chuv.ch Received: Accepted:

June 12, 2023. August 23, 2023.

Abstract Predominantly nodal is the most common clinical presentation of peripheral T- (and NK-) cell lymphomas (PTCL), which comprise three main groups of diseases: (i) systemic anaplastic large cell lymphomas (ALCL), whether positive or negative for anaplastic lymphoma kinase (ALK); (ii) follicular helper T-cell lymphomas (TFHL); and (iii) PTCL, not otherwise specified (NOS). Recent advances in the genomic and molecular characterization of PTCL, with enhanced understanding of pathobiology, have translated into significant updates in the latest 2022 classifications of lymphomas. ALK-negative ALCL is now recognized to be genetically heterogeneous, with identification of DUSP22 rearrangements in approximately 20-30% of cases, correlated with distinctive pathological and biological features. The notion of cell-of-origin as an important determinant of the classification of nodal PTCL is best exemplified by TFHL, considered as one disease or a group of related entities, sharing oncogenic pathways with frequent recurrent epigenetic mutations as well as a relationship to clonal hematopoiesis. Data are emerging to support that a similar cell-of-origin concept might be relevant to characterize meaningful subgroups within PTCL, NOS, based on cytotoxic and/or Th1 versus Th2 signatures. The small group of primary nodal Epstein-Barr virus-positive lymphomas of T- or NK-cell derivation, formerly considered PTCL, NOS, is now classified separately, due to distinctive features, and notably an aggressive course. This review summarizes current knowledge of the pathology and biology of nodal-based PTCL entities, with an emphasis on recent findings and underlying oncogenic mechanisms.

Introduction Peripheral T-cell lymphomas (PTCL) collectively refer to neoplasms of mature NK or T cells. They constitute approximately 10% of all lymphomas in the Western world and up to 20% in Asia. PTCL exhibit significant clinical and biological diversity, leading to the recognition of over 30 distinct entities in the latest classification proposals.1,2 These entities can be categorized into three groups based on their typical clinical presentation: involvement of lymph nodes, skin and other extranodal organs, or dissemination/leukemic manifestations. This clinical grouping reflects, to some extent, the distinct cell of origin for each entity, characterized by specific functional, homing, or trafficking properties, as well as varying oncogenic mechanisms. The nodal-based PTCL entities represent the large majority of PTCL cases. They include anaplastic large cell lymphomas (ALCL), categorized as positive or negative for anaplastic lymphoma kinase

(ALK+ and ALK–, respectively), lymphomas originating from T-follicular helper (TFH) cells, primary nodal Epstein-Barr virus (EBV)-associated PTCL, and PTCL, not otherwise specified (PTCL, NOS) (Figure 1, Tables 1 and 2). The two proposed classifications emanated in 2022, the International Consensus Classification (ICC) of mature lymphoid neoplasms1 and the 5th edition of the World Health Organization classification of lymphoid neoplasms (WHOHAEM5),2 represent updates of the former 2017 revised 4th edition of the World Health Organization classification (WHO-HAEM4R),3 and both maintain the principle of a multiparametric definition of lymphoma entities adopted since 1994. Morphology, immunophenotype, genetic and clinical features as well as the putative normal cellular counterpart are all taken into account for the classification. Research advances elucidating the genomic landscape and molecular characterization of many PTCL, in addition to enhanced understanding of the pathobiology and pathogen-

Haematologica | 108 December 2023

3227

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology

B. Bisig et al.

Figure 1. Representation of nodal T- and NK-cell lymphoma entities according to the 2022 classifications. NOS: not otherwise specified; TFH: T follicular helper cell; ICC: International Consensus Classification; WHO: World Health Organization; ALCL: anaplastic large cell lymphoma; ALK: anaplastic lymphoma kinase; EBV: Epstein-Barr virus.

esis with information derived from both clinical and experimental models, have translated into significant updates in both classifications. The main adjustments and changes introduced in both proposals reflect similar conceptual shifts. Here, we present a review of our current knowledge of the pathology and biology of nodal-based PTCL, the understanding of oncogenic mechanisms, and discuss future directions.

Anaplastic large cell lymphomas ALCL encompass four entities having in common large pleomorphic tumor cells with strong expression of CD30, but being distinct in their pathogenesis, clinical presentation and outcome. The two systemic entities, ALK+ and ALK– ALCL, most often present as nodal-based disease and are considered nodal PTCL, but may show the involvement of a variety of extranodal sites. The other two ALCL entities are extranodal and site-specific. Although not a nodal PTCL entity, breast implant-associated (BIA)-ALCL is described here to highlight both overlapping and distinguishing features compared with ALK– ALCL, and the fact that it may disseminate to lymph nodes.4 Primary cutaneous ALCL (ALK–), classified with other primary cutaneous CD30+ T-cell lymphoproliferative disorders, is not included in this review.1,2 The classification and diagnostic criteria of ALCL entities is essentially identical in the two 2022 classifications (Table 1). Provisional in WHO-HAEM4R, BIA-ALCL is now a definitive entity. Among ALK– ALCL, DUSP22-rearranged (DUSP22-R) cases constitute a distinct genetic subtype according to the ICC 2022, but not in WHO-HAEM5, as discussed below.1,2

Anaplastic large cell lymphomas, ALK-positive In ALK+ ALCL, there is aberrant expression of the ALK protein secondary to rearrangements involving the ALK gene.3 ALK+ ALCL accounts for 7-9% of (non-cutaneous) PTCL worldwide,5,6 most commonly affects young male patients (median age, 30-35 years), but may occur at any age including in children.7,8 ALK+ ALCL usually presents at an advanced stage, and frequently involves lymph nodes and/or various extranodal sites, including bone, skin and lung.7,8 Rare cases, mostly of the small cell variant, present as leukemia.9 ALK+ ALCL has the most favorable prognosis among systemic PTCL, with a long-term overall survival reaching 70-90% in adults and up to 95% in children. This is in part related to the younger age at diagnosis.6-8 ALK+ ALCL comprises several morphological variants,10 all containing “hallmark cells”, with a large, often kidneyshaped nucleus, abundant cytoplasm and a prominent Golgi zone, although in variable proportions.3 In the classic form (common pattern) (Figure 2A-G), hallmark cells are numerous and form cohesive sheets. Nodal infiltration typically shows an intrasinusoidal growth pattern. The small cell and lymphohistiocytic patterns, which are characterized by smaller hallmark cells with a preferential perivascular distribution and a dominant infiltrate of reactive histiocytes, respectively,3 are often encountered together, and have been associated with an adverse prognosis in the pediatric population, but not in adults.11 A rare pattern mimicking nodular sclerosis classic Hodgkin lymphoma (Hodgkin-like pattern) may be diagnostically challenging.12 CD30 expression is strong and diffuse in the classic and Hodgkin-like variants, but may be more focal in the other patterns.3,10 Chimeric ALK protein is by definition expressed, but its subcellular location depends on the translocation partner. The most common t(2;5)(p23;q35) translocation

Haematologica | 108 December 2023

3228

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology

B. Bisig et al.

Table 1. List of nodal mature T and NK-cell neoplasms in the International Consensus Classification (ICC) and the World Health Organization (WHO)-HAEM5 classification (2022) in reference to the WHO-HAEM4R scheme (2017) (adapted from Campo et al.1 and Alaggio et al.2) WHO-HAEM4R

ICC 2022

WHO-HAEM5

Follicular helper T-cell lymphoma

Nodal T-follicular helper (TFH) cell lymphoma

Angioimmunoblastic T-cell lymphoma

Follicular helper T-cell lymphoma, angioimmunoblastic type (angioimmunoblastic T-cell lymphoma)

Nodal TFH cell lymphoma, angioimmunoblastic type

Follicular T-cell lymphoma

Follicular helper T-cell lymphoma, follicular type

Nodal TFH cell lymphoma, follicular type

Nodal peripheral T-cell lymphoma with T follicular helper phenotype

Follicular helper T-cell lymphoma, NOS

Nodal TFH cell lymphoma, NOS

Not listed as an entity, subtype of peripheral T-cell lymphoma, not otherwise specified (PTCL, NOS)

Primary nodal EBV+ T-cell/NK-cell lymphoma

EBV+ nodal T- and NK-cell lymphoma

Peripheral T-cell lymphoma, NOS

The entities are listed according to the order in which they appear in the International Consensus Classification 2022. Italics indicate provisional entities. WHO: World Health Organization; HAEM4R: revised 4th edition of the WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues; ICC: International Consensus Classification of lymphoid neoplasms; HAEM5: the 5th edition of the WHO Classification of Lymphoid Neoplasms; ALK: anaplastic lymphoma kinase; TFH: T-follicular helper; PTCL: peripheral T-cell lymphoma; NOS: not otherwise specified; EBV: Epstein-Barr virus.

(80% of the cases) results in a NPM1::ALK fusion that shuttles between the nucleus and the cytoplasm, due to the preserved N-terminal portion of NPM1, and is detected by ALK immunohistochemistry in both compartments.13 With alternative partners, ALK staining may be purely cytoplasmic (e.g., ATIC::ALK), membranous (MSN::ALK), or both (TPM3::ALK).14,15 Epithelial membrane antigen (EMA) is typically positive and EBV is always negative. The neoplastic cells often show loss of T-cell receptor (TCR) molecules (“null” immunophenotype) and of several T-cell antigens (typically CD3, CD5 and CD7 commonly negative), while CD2, CD4 and CD43 are more frequently preserved. CD8 is usually negative. Cytotoxic markers (granzyme B, perforin, T-cell intracellular antigen 1 [TIA-1]) are generally expressed.3 The ALK fusion proteins have constitutive tyrosine kinase activity which activates multiple signaling cascades involved in oncogenesis, including the JAK-STAT3 pathway.15,16 Accordingly, nuclear phospho-STAT3 (pSTAT3) is detected by immunohistochemistry in nearly all cases.16,17 In relation to the crucial driver role of ALK fusions, ALK inhibitors have demonstrated high response rates in pediatric and young adult patients with relapsed/refractory ALK+ ALCL.8 Nevertheless, as for ALK-rearranged lung carcinomas, mechanisms of resistance may develop, including acquired ALK mutations.18

Mutational profiling of ALK+ ALCL has revealed NOTCH1 mutations in 20% of the cases. NOTCH1 expression is also induced via STAT3 activation mediated by the ALK fusion protein. In line with these observations, inhibition of the NOTCH1 pathway by γ-secretase inhibitors is effective in vitro and a promising treatment target to be further explored.19 Other recurrently mutated genes in ALK+ ALCL include LRP1B (19%) and TP53 (11%), as well as epigenetic modifier genes such as EP300, KMT2C and KMT2D. TP53 mutations have been associated with an inferior outcome.20 Anaplastic large cell lymphomas, ALK-negative ALK– ALCL account for 6-15% of PTCL5,6,21 and occur later in life than ALK+ ALCL (median age, 54 years), with a slight male predominance.21 Most patients present with advanced stage disease, and extranodal sites of involvement (50%) include lung, liver and bone.7,22 The 5-year overall survival is, as a whole, worse (∼50% with CHOP-like chemotherapy) for ALK– ALCL patients than for ALK+ patients.7,21 The diagnosis of ALK– ALCL requires typical morphology, with hallmark cells identical to the classic form of ALK+ ALCL, uniformly strong CD30 expression, negativity for ALK expression or ALK rearrangement, and absence of EBV. Tumor cells frequently show loss of T-cell antigens,

Haematologica | 108 December 2023

3229

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology

B. Bisig et al.

Table 2. Usual features of the main nodal peripheral T-cell lymphoma entities. Immunophenotype

Cell of origin

Genes altered, cytogenetic features

TFH lymphoma Small to medium atypical neoplastic cells with clear cytoplasm and often prominent polymorphous environment

CD4+; TFH cell markers (BCL6, CD279 [PD1], ICOS, CXCL13, CD10); expanded CD21+ follicular dendritic cell meshworks in angioimmunoblastic type; CD20+ B immunoblasts, often EBV+

CD4 TFH cell

TET2, DNMT3A, IDH2, RHOA, CD28, FYN, LCK, PLCG1, VAV1

ALCL, ALK-positive Large pleomorphic cells, cohesive

CD30+, ALK+, often extensive loss of pan-T-cell antigens, frequently cytotoxic (TIA-1, granzyme B, perforin), EBV–

CD4 αβ T cell

ALK fusions NOTCH1, TP53

ALCL, ALK-negative Large pleomorphic cells, cohesive

CD30+, ALK–, variable expression of pan-T-cell antigens and of cytotoxic markers (TIA-1, granzyme B, perforin), EBV–

CD4 αβ T cell

JAK1, JAK3, STAT3, MSC Rearrangements: DUSP22, TP63, JAK2

PTCL, NOS Variable cytology of the neoplastic cells and variable microenvironment

Variable

CD4 > CD8, αβ > γδ, Th1 or Th2 subsets

TP53, PRDM1, TET1, TET3, DNMT3A, ATM, PTEN, RB1…

EBV+, CD8+ and CD56–, cytotoxic (TIA-1, granzyme B, perforin)

CD8 T cell > NK cell

TET2, PIK3CD, STAT3

PTCL entities

Primary nodal EBV+ T/NK-cell lymphoma Large cells, monomorphic, no angiocentricity

PTCL: peripheral T-cell lymphoma; TFH: T follicular helper; EBV: Epstein-Barr virus; ALCL: anaplastic large cell lymphoma; ALK: anaplastic lymphoma kinase; TIA-1: T-cell intracellular antigen 1; NOS: not otherwise specified.

EMA expression and a cytotoxic phenotype, although not as consistently as observed in ALK+ ALCL.3 Approximately half of all ALK– ALCL express nuclear pSTAT3, as a consequence of translocations or mutations that lead to constitutive JAK-STAT3 activation, including JAK1 and/or STAT3 mutations in 20-30% of the samples.20,22-24 TP53 mutations are observed in 23% of the cases, and an inferior outcome has been reported for ALK– ALCL with TP53 or STAT3 mutations.20 Beyond these shared features, the last decade has brought new data that highlight the genetic heterogeneity of ALK– ALCL, with the identification of recurrent structural aberrations, some of which may influence the clinical outcome. The largest genetic subgroup (20-30% of ALK– ALCL) is characterized by the presence of a rearrangement in the DUSP22 gene,22,25,26 which induces the downregulation of dual specificity phosphatase, thus preventing its inhibitory effect on various signaling pathways. DUSP22-R ALK– ALCL have distinctive biological features, including lack of JAK-STAT3 activation, widespread DNA hypomethylation, expression of the cancer-testis antigens and an immunogenic phenotype.24 One-third of cases have a musculin (MSC) hotspot mutation (E116K), which promotes the CD30–IRF4–MYC axis.27 DUSP22-R ALK– ALCL (Figure 2H-N) frequently demonstrate doughnut-shaped

nuclei and, compared to DUSP22 non-rearranged (NR) cases, are more frequently CD3+ but less commonly express EMA and cytotoxic molecules.22,25,26 Conversely, strong and uniform nuclear expression of lymphoid enhancer binding factor 1 (LEF1) is highly associated with DUSP22-R.28 Clinically, the DUSP22-R subgroup has been the subject of a number of recent studies with somewhat disparate results (5-year overall survival: 40%-100%; 5year progression-free survival: 40-57%), but, although study sizes still remain small, collectively it appears to have a more favorable prognosis, intermediate between ALK+ ALCL and DUSP22-NR ALK– ALCL.22,25,26,29 Consequently, DUSP22-R ALCL is now considered a genetic subtype among ALK– ALCL in the ICC 2022 – which recommends fluorescence in situ hybridization testing if available –, but not in the WHO-HAEM5 classification due to uncertainties around prognosis.1,2 The presence of a TP63-R characterizes a smaller genetic subgroup of ALK– ALCL, accounting for 2-8% of cases and associated with an adverse prognosis.25,26,30 A few cases with JAK2-R have been described, featuring Reed-Sternberg-like cells in an inflammatory background with eosinophilia, reminiscent of classic Hodgkin lymphoma.31 A Hodgkin-like morphology has also been observed in a subset of cases expressing truncated ERBB4 transcripts.32

Haematologica | 108 December 2023

3230

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology

B. Bisig et al.

Figure 2. Nodal involvement by systemic anaplastic large cell lymphoma. (A-G) Anaplastic lymphoma kinase (ALK)-positive anaplastic large cell lymphoma (ALCL), classic pattern. This case comprises cohesive sheets of large pleomorphic cells (A, hematoxylin & eosin), shows diffuse strong expression of CD30 (B) and loss of several T-cell antigens including CD3 (C). Nuclear and cytoplasmic expression of ALK protein (D) is indicative of a NPM1::ALK fusion. An ALK gene rearrangement can be confirmed by break-apart fluorescence in situ hybridization (FISH) (E, the rearranged ALK allele is represented by split red and green signals). The cells are positive for cytotoxic markers (F, perforin) and nuclear phospho-STAT3 (G). (H-N) ALK-negative ALCL with DUSP22 rearrangement. This case comprises cohesive sheets of neoplastic cells, including kidney-shaped hallmark cells (H, hematoxylin & eosin, arrows) and cells with doughnut-shaped nuclei (H, arrowheads). The cells are strongly CD30-positive (I), weakly positive for CD3 (J), and ALK-negative (K). Break-apart FISH shows a DUSP22 locus rearrangement (L, the rearranged DUSP22 allele is represented by split red and green signals). The cells are negative for cytotoxic markers (M, T-cell intracellular antigen 1 [TIA-1]) and phospho-STAT3 (N).

Other gene fusions reported in ALK– ALCL involve the tyrosine kinase domain of FRK, ROS1 or TYK2.23,33 Nevertheless, these cases represent very small and heterogeneous subsets of patients, and further data are needed before stratifying DUSP22-NR ALK– ALCL into further genetic subtypes. Breast implant-associated anaplastic large cell lymphoma BIA-ALCL is a rare complication of textured breast implants, with a median of 9 years from implantation to diagnosis, and generally an excellent prognosis.34 Recent data suggest that germline BRCA1 or BRCA2 mutations represent a risk factor for developing BIA-ALCL.35 Most often, BIA-ALCL occurs as a peri-prosthetic effusion and is diagnosed on cytological samples.36 In surgical capsulectomies, lymphoma cell deposits are found along the inner surface of the capsule. In the infiltrative forms, tumor cells extend into the capsule and rarely

may form a mass.36 Regional lymph node involvement may occur, typically with low tumor burden with sinusoidal and perifollicular growth patterns.37 Although BIA-ALCL resembles systemic ALK– ALCL morphologically and immunophenotypically,36 it develops in a peculiar confined microenvironment, in which bacterial infection, chronic inflammation and hypoxia are believed to be distinctive pathogenic determinants.38 The genetic aberrations acquired in this setting and transformation mechanisms partially overlap with those of systemic ALK– ALCL. The JAK-STAT3 pathway appears to be consistently activated, either by mutations (in 6090% of cases, most frequently STAT3 and JAK1) or by other mechanisms, as witnessed by the constant pSTAT3 expression.39 In contrast, DUSP22 and TP63 rearrangements have not been found in BIA-ALCL.34,40 Furthermore, loss of 20q13.13, detected by shallow whole-genome sequencing in 66% of cases in one series, is reportedly highly characteristic of BIA-ALCL

Haematologica | 108 December 2023

3231

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology compared to systemic ALK+ or ALK– ALCL.38 Other recurrent alterations involve epigenetic modifiers, with frequent mutations in KMT2C, CHD2, CREBBP and KMT2D (up to 75% of cases), while TET2 and DNMT3A mutations, typically observed in TFH lymphomas, are uncommonly detected.39 Altogether, the unique clinical association, peculiar pathogenic background and distinctive pathobiological features of BIA-ALCL support its classification as a separate entity.

Follicular helper T-cell lymphoma T follicular helper (TFH) cells represent a functional subset of CD4+ lymphocytes that are necessary for the formation and maintenance of germinal centers, interacting with germinal center B cells and aiding their differentiation.41 In 2007, immunophenotyping and gene expression profiling analyses identified the TFH cell as the cell of origin of angioimmunoblastic T-cell lymphoma (AITL),42 and this specific cellular derivation has become a cardinal defining feature of the disease. In practice, a TFH phenotype is evaluated by immunohistochemistry by the expression of PD1, ICOS, CXCL13, CD10 and BCL6, which are markers of normal TFH cells, with at least any two positive required for the definition.3 In 2017, the developing concept that TFH cell derivation represents a unifying feature of a larger group of nodal CD4+ T-cell lymphomas led to the creation of an umbrella term “nodal T-cell lymphoma of T follicular helper origin” to encompass AITL, follicular T-cell lymphoma, and nodal PTCL with a TFH phenotype.3

B. Bisig et al.

The grouping was supported by the fact that these three entities also share a similar genetic landscape, characterized by recurrent mutations in epigenetic modifier genes, RHOA and other TCR signaling genes.42-48 Identification of a TFH phenotype might also be relevant to treatment decisions.4,49 Therefore the ICC considers one single disease entity, namely follicular helper T-cell lymphoma, with three subtypes, angioimmunoblastic, follicular and NOS (Table 3).50 This entity by definition applies to CD4+ PTCL, implies significant expression of TFH markers, and excludes primary cutaneous CD4+ T-cell lymphoproliferative disorders with a TFH phenotype. The WHO-HAEM5 proposal considers a family of three related entities of nodal T-follicular helper cell lymphomas, angioimmunoblastic-type, follicular-type and NOS types. Collectively, TFH lymphoma(s) outnumber(s) PTCL, NOS in recent epidemiological studies.5,51,52 Follicular helper T-cell lymphoma, angioimmunoblastic type AITL is the prototypic and most frequent subtype of TFH lymphoma.42 It manifests as a systemic disease in adults, usually in the elderly. Patients present with generalized lymphadenopathy, often with extranodal involvement (e.g., skin, tonsil, liver, spleen,...), systemic symptoms, and various immune abnormalities. The median survival is <3 years, but a subset of patients experience long-term survival.53 Histologically (Figure 3A-F) (for a review, see de Leval et al.54), AITL comprises a polymorphous infiltrate including variable proportions of neoplastic T cells, typically outnumbered by reactive small lymphocytes, histiocytes, im-

Table 3. Comparison of the three subtypes of follicular helper T-cell lymphoma. Follicular helper T-cell lymphoma

Angoimmunoblastic type

Follicular type

Not otherwise specified

Frequent, often accompanied by general symptoms and biological abnormalities

Very rare

20% of PTCL, NOS

Diffuse with FDC expansion OR perifollicular without FDC expansion

Follicular lymphoma-like OR PTGC-like (with IgD+ B cells) FDC restricted to follicles

Diffuse and no FDC expansion OR T-zone pattern No/minimal FDC expansion

Microenvironment

Abundant, polymorphous

Minimal (FL-like) B cells (PTGC-like)

Minimal

Vascular proliferation

Abundant

Absent or minimal

EBV+/- B-cell blasts

Typically present

Often present

May be present

Several TFH markers

Several TFH markers, strong

At least 2 (ideally 3) TFH markers

Epidemiology

Pattern of growth and FDC distribution

TFH phenotype

PTCL: peripheral T-cell lymphoma; NOS: not otherwise specified; FDC: follicular dendritic cells; PTGC: progressive transformation of germinal centers; FL: follicular lymphoma; EBV: Epstein-Barr virus; TFH: T follicular helper cell. Haematologica | 108 December 2023

3232

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology munoblasts, eosinophils and plasma cells. The neoplastic lymphoid cells are usually small to medium-sized, with moderately abundant, clear cytoplasm, but may lack significant atypia. Some cases may contain a larger proportion of neoplastic cells or large cells with abundant, clear cytoplasm. Large B-cell immunoblasts, sometimes resembling Hodgkin and Reed-Sternberg cells, represent a typical component of AITL; they are usually scattered, but sometimes numerous. Some cases rich in histiocytes may resemble lymphoepithelioid (Lennert) lymphoma. Plasma cells may be abundant, and some patients may even present with peripheral blood plasmacytosis. A stromal component is typically present, consisting of a marked proliferation of arborizing high endothelial venules, and abnormal extrafollicular expansion of follicular dendritic cells (FDC) best demonstrated by CD21 or CD23 immunostaining. The lymphoproliferation is usually diffuse (pattern III). In less common instances AITL is associated with hyperplastic or regressive follicles (patterns I and II). Transitions between different patterns have been observed in consecutive biopsies, and a combination of patterns may be seen at one time. Patterns I and II correspond to lower tumor cell burden reflecting partial nodal involvement, but do not correlate with lower stage disease.54-56 Two studies found that a subset of AITL enriched in a B-cell signature was associated with a better outcome,57,58 and interestingly the favorable significance of a B-cell signature was also found in PTCL, NOS in another study.59 The neoplastic cells of AITL consist of mature CD3+ CD4+ CD8– TCRαβ+ cells. An aberrant T-cell immunophenotype (most commonly reduced or absent surface CD3, CD5 or CD7) is frequently observed, especially by flow cytometry. The neoplastic cells express several TFH markers, including the CXCL13 chemokine; PD1 (CD279), ICOS, CD10 and CD200 membrane receptors; and BCL6 and cMAF transcription factors.60,61 Overall, PD1 and ICOS are more sensitive for identifying the neoplastic TFH cells than CXCL13 or CD10, which are more specific.56 Aberrant co-expression of CD20 and/or partial expression of CD30 by the neoplastic cells is not unusual.62,63 Reactive CD8+ cells are variably abundant and may outnumber the CD4+ neoplastic cells.64 The large B-cell immunoblasts are positive for CD20, PAX5 and CD79a, often CD30, and may also sometimes co-express CD15. They are usually, but not always, infected by EBV (positive for EBV-encoded RNA [EBER] and latent membrane protein 1 [LMP-1]).65 The spectrum of B-cell-derived expansions in AITL also comprises EBV– large B-cell proliferations, and polytypic (or sometimes monotypic) plasma cell expansions. Up to one-third of cases, particularly those with an increased number of B cells, harbor (oligo)clonal rearrangements of the immunoglobulin genes (IG), in addition to monoclonal or oligoclonal T-cell receptor gene (TR) rearrangements. Some patients develop secondary large B-cell lymphomas.66

B. Bisig et al.

Most cases demonstrate a characteristic mutational landscape that recapitulates a multi-step oncogenic process derived from underlying clonal hematopoiesis (Figure 4).67 The profile of AITL alterations typically consists of epigenetic deregulation (TET2 +/- DNMT3A mutations, occurring at early stages in hematopoietic progenitors, present in about 80% and 30-35% of the cases, respectively),44,46 and second-hit mutations with a more restricted distribution.68 TET2 and DNMT3A mutations are inactivating, TET2 mutations are often multiple (2 or 3), and alterations in both genes tend to co-occur. Their global effect is DNA hypermethylation. DNMT3A mutations were found in one study associated with a shorter progression-free survival,69 and in an experimental mouse model of AITL generated by TET2 inactivation and RHOA G17V, addition of DNMT3A R882H accelerated the development of the disease.70 Secondhit mutations include a hotspot RHOA G17V mutation encoding a dominant negative variant of the protein in up to 70% of cases, and other gain-of-function mutations targeting the TCR pathway (PLCG1, CD28, FYN, PIK3 components, CARD11, etc.).45,47,71 RNA fusions involving CD28 with ICOS or rarely CTLA4, mutually exclusive with CD28 mutations, are detected in a small subset of patients.72 Mouse models have shown that RHOA G17V induces TFH differentiation and autoimmunity, and promotes lymphomagenesis in the presence of TET2 inactivation, indicating the synergistic effect of both mutations.73-75 AITL harboring the RHOA G17V mutation tend to have higher microvessel density, more FDC proliferation and a more pronounced TFH immunophenotype compared to wildtype cases, but no prognostic significance was observed.76,77 One study showed that sensitive RHOA G17V mutation analysis may be valuable for the early diagnosis of TFH lymphoma, as this alteration may be found prior to conclusive histopathological changes.78 IDH2 point mutations at the R172 residue, present in about onethird of cases,79,80 modify IDH2 enzymatic activity resulting in the production of an oncometabolite (2 hydroxyglutarate) ultimately altering DNA and histone methylation.80,81 IDH2-mutated AITL has a characteristic morphology with prominent medium-sized to large clear cells (Figure 3G-H), and tends to show strong CD10 and CXCL13 expression.82 In an AITL mouse model driven by IDH2 and TET2 mutations, the tumors show abundant angiogenesis and plasma cells, and the malignant TFH cells display aberrant transcriptomic and epigenetic programs that impair TCR signaling and alter cross-talk with germinal center B cells, promoting B-cell clonal expansion while decreasing the Fas-FasL interaction and reducing B-cell apoptosis.83 Gains of chromosomes 5 and 21 are frequent, especially in IDH2-mutated cases, and copy number losses in genes regulating the PI3K-AKTmTOR pathway are enriched in IDH2-wild-type cases.84

Haematologica | 108 December 2023

3233

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology

B. Bisig et al.

Figure 3. Follicular helper T-cell lymphoma. (A-H) Follicular helper T-cell (TFH) lymphoma, angioimmunoblastic type comprises a polymorphous cellular infiltrate and prominent venules (A, hematoxylin & eosin); immunostains show an irregular proliferation of follicular dendritic cell meshworks (B, CD21), aggregates of small B cells and scattered large blasts positive for CD20 (C), a diffuse infiltrate of CD4+ cells (D), which are positive for TFH markers (E, PD1 and F, CXCL13). In IDH2-mutated cases (G, H), large cells with abundant clear cytoplasm are prominent (G, hematoxylin & eosin), and expression of the IDH2 R172K variant is highlighted by immunohistochemistry (H). (I, J) TFH lymphoma, follicular type is characterized by large nodules resembling progressive transformation of germinal centers at low magnification (I, hematoxylin & eosin), and TFH markers demonstrate aggregates of pale neoplastic cells (J, ICOS).

Detailed genetic analyses of AITL have shown that these lymphomas may contain one or multiple clonal TR gene rearrangements associated with the same TET2 mutation(s), indicating parallel neoplastic evolution from a common TET2-mutant hematopoietic progenitor pool. A biased TRBV usage was also found, suggesting the role of antigenic stimulation in promoting T cells to clonal expansion and malignant transformation.85 Moreover, the study of microdissected T- and B-cell populations has shown that TET2 mutations of AITL clones are frequently present in the B cells as well, indicating that clonal hematopoiesis also generates a large population of mutated mature B cells and, while RHOA and IDH2 mutations are confined to the T-cell compartment, other mutations, notably in NOTCH1, may be identified in the B cells, providing an explanation for the frequently associated B-cell expansions in AITL (Figure 4).86,87 Clonal hematopoiesis appears to be the source of the observed association with myeloid neoplasms in certain patients88 and the increased incidence of myeloid neoplasms after TFH lymphoma-directed therapy, through divergent clonal evolution (Figure 4).4,67

Follicular helper T-cell lymphoma, follicular type This least common subtype of TFH lymphoma presents either a truly follicular pattern, mimicking follicular lymphoma (FL-like), or more commonly a pattern resembling progressive transformation of germinal centers (PTGClike). In FL-like cases, nodular aggregates of neoplastic cells are sustained by a meshwork of FDC. In PTGC-like cases (Figure 3I, J), pale aggregates of medium-sized atypical T cells are distributed within expanded mantle zones in large nodules mostly composed of small IgD+ B cells.43,89 TFH lymphoma, follicular type lacks both the extrafollicular expansion of FDC and the proliferation of high endothelial venules characteristic of AITL. The clinical presenting features overlap with those of AITL.89,90 A subset of patients has long-term survival despite sometimes multiple relapses and the prognosis might be slightly better than that of AITL.89,90 The neoplastic cells are CD3+ CD4+ and usually show extensive positivity for most TFH markers (PD1, ICOS, CXCL13, BCL6, CD10, and sometimes CD57).89,91 In one study most cases had at least partial expression of CD30 in the neoplastic cells.91 A component of large blastic EBV+ or EBV– B

Haematologica | 108 December 2023

3234

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology

B. Bisig et al. Figure 4. Oncogenic model of follicular helper T-cell lymphoma in relationship to clonal hematopoiesis. TFH: follicular helper T-cell

cells is often identified, frequently with Reed-Sternberglike morphology and immunophenotype.65,89,91-93 The t(5;9)(q33;q22) translocation, resulting in an ITK::SYK fusion, is found in about 20% of the follicular type, and has been reported thus far in only one case of typical AITL.89,94 Based on limited data, the mutational pattern of follicular PTCL appears otherwise to overlap with that of AITL.48

lymphoma, NOS overlaps with that of AITL, including frequent mutations in TET2 and DNMT3A but less frequent RHOA mutations and infrequent IDH2 mutations.44-47,58 According to limited data available, the outcome related to TFH lymphoma, NOS, is similar to that of TFH lymphoma of the angioimmunoblastic type but larger studies are needed.48,96

Follicular helper T-cell lymphoma, not otherwise specified TFH lymphoma, NOS, formerly nodal T-cell lymphoma with a T follicular helper phenotype,3 encompasses cases without specific pathological features, but showing imprints of the TFH signature and/or expression of TFH markers, and/or cases exhibiting some characteristics of AITL (e.g., increased vascularity, presence of EBV+ Bblasts).42,44 According to the WHO-HAEM4R, qualification for a TFH lymphoma required the expression of at least two or ideally three TFH markers (among the 5 recommended for routine testing: PD1, ICOS, CD10, BCL6, CXCL13) by the neoplastic cells, in addition to CD4.3 This criterion is retained in the current proposals (Table 3). Some cases show perifollicular involvement and may mimic marginal zone lymphoma.56 A subset of cases may present as what was previously called the “T-zone variant” of PTCL, NOS, in which there is preserved architecture with residual sometimes hyperplastic B-cell follicles, and interfollicular lymphomatous involvement.95 Since FDC proliferation is generally considered as a typical hallmark of AITL, cases with some FDC expansion are better qualified as tumor-cellrich AITL, but the border between PTCLTFH and AITL is not well delineated, likely reflecting a biological continuum.48,56 The genetic background of TFH

Primary nodal Epstein-Barr viruspositive T/NK-cell lymphoma Primary EBV+ nodal T-cell or NK-cell lymphoma was introduced in the WHO-HAEM4R as a variant of PTCL, NOS.3 In the light of novel data confirming its distinctive features from extranodal EBV+ NK/T-cell lymphoma, nasal type (ENKTCL), and supporting specific characteristics, this rare disease is a new entity in the 2022 classifications, with slightly different nomenclatures (Table 1).1,2 Most cases were described in reports from Asia.97-99 Primary nodal EBV+ T/NK-cell lymphoma involves lymph nodes and tends to occur in elderly adults who present with generalized lymphadenopathy, frequent dissemination to the liver or spleen but lack of nasal involvement, sometimes in association with human immunodeficiency virus infection or immunodeficient conditions.56,100 The outcome of patients with primary nodal EBV+ T/NK-cell lymphoma is dismal, being significantly worse than that of patients with ENKTCL or PTCL, NOS.101 Pathological features distinct from ENKTCL include a usually monomorphic large cell morphology, lack of prominent angiocentricity or necrosis, negativity for CD56, positivity for CD8, and more frequent derivation from T cells than

Haematologica | 108 December 2023

3235

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology from NK cells.100,102 The lymphoma cells are CD3+ CD5–/+ with an activated cytotoxic phenotype, with EBV detected in the majority of tumor cells by in situ hybridization or expression of LMP-1. Most cases are of T-cell lineage, carry clonally rearranged TR genes associated with 14q11.2 loss, and variably express the TCR.100 Compared with ENKTCL, primary nodal EBV+ T/NK-cell lymphoma is characterized by low genomic instability, upregulation of immune pathways (checkpoint protein PD-L1) that promote immune evasion, and downregulation of EBV microRNA.101 Few cases have been investigated by high-throughput sequencing; recurrent mutations have been found in TET2, DNMT3A, STAT3, PIK3CD and DDX3X.101,102

Peripheral T-cell lymphoma, not otherwise specified PTCL, NOS remains defined as a diagnosis of exclusion for those cases of PTCL lacking specific features that qualify for another “specific” PTCL entity. While the definition is unchanged from the previous WHO classifications, subsets of cases formerly included in PTCL, NOS, such as those expressing a TFH phenotype or positive for EBV, are now classified into distinct entities, and therefore the boundaries of PTCL, NOS are narrowing.1-3 Accordingly, while PTCL, NOS has for decades been reported as the most frequent type of PTCL,6 in recent years the reported prevalence of PTCL, NOS is tending to decrease, accounting for 21-27% of PTCL.5,51,52 PTCL, NOS nearly always affects adults.103,104 The presentation is usually nodal, with frequent concurrent extranodal involvement, especially of the skin. Most patients have disseminated disease, constitutional symptoms, intermediate- to high-risk International Prognostic Index score and sometimes blood eosinophilia.103 The overall outcome is 2030% survival at 5 years.103 A small minority of patients have a preceding lymphoproliferative variant of the hypereosinophilic syndrome105 or chronic lymphocytic leukemia.106 PTCL, NOS is morphologically heterogeneous (Figure 5AN). Many cases show predominantly medium-sized or large cells with irregular nuclei and prominent nucleoli. Less commonly, others have a predominance of atypical small cells with irregular nuclei.3 Morphological grading is not recommended for clinical purposes, but tumors with a predominance of large cells have been found to have a worse outcome.103 Many cases have an admixture of reactive small lymphocytes, eosinophils, histiocytes, B cells and plasma cells. Any of the microenvironmental components can be dominant, obscure the neoplastic cells, and represent a confounding factor to establishing a correct diagnosis. Cases with a prominent infiltrate of epithelioid histiocytes,

B. Bisig et al.

referred to as lymphoepithelioid lymphoma (Lennert lymphoma), represented less than 10% of PTCL, NOS in historical series and were associated with an overall better prognosis than other PTCL, NOS.103 In recent years it has turned out that many cases of “Lennert lymphoma” correspond to histiocyte-rich TFH lymphomas.95,107,108 It is unclear at the present time whether the Lennert/lymphohistiocytic lymphomas remaining categorized as PTCL, NOS, which are often derived from CD8+ cells with a nonactivated cytotoxic immunophenotype, should be considered as a distinct subgroup of PTCL, NOS.109-111 The neoplastic cells in PTCL, NOS are positive for pan-Tcell antigens (CD3, CD2, CD5, CD7), but one or several of these (most commonly CD5 or CD7) may show reduced or absent expression; they are most commonly CD4+ CD8–, less frequently CD4– CD8+, uncommonly CD4– CD8– or CD4+ CD8+.110 More than 85% of cases express the αβ TCR, and a minority of cases are either of γδ derivation, or TCR-silent.112 Loss of BCL2 expression, observed in 45-60% of the cases, can be a useful marker indicative of T-cell malignancy.113,114 A small proportion of PTCL, NOS (5% or less) express CD20 (or other B-cell markers) in a subset of the neoplastic cells.115 By definition, the neoplastic cells in CD4+ PTCL, NOS must lack a TFH immunophenotype.1,2 CD30 expression is frequent (∼30%) and variable.7,62,63 In a study of 141 cases of PTCL, NOS, over 20% of the cases had more than 50% CD30+ tumor cells,62 and staining extent and intensity were higher in cases with large cell morphology. Strong CD30 expression by a majority of the tumor cells is seen occasionally, and raises the need for a differential diagnosis from ALCL. PTCL, NOS is usually positive for LEF1,63 and negative for BCL6, FOXP3 and TCL1 transcription factors which are, respectively, critical for TFH differentiation and function, related to regulatory T cells, and overexpressed in T-cell prolymphocytic leukemia.116,117 A few cases of FOXP3+ PTCL, NOS have been described in patients negative for HTLV1 infection; these cases were composed of large cells, some had EBV reactivation in bystander cells and the clinical course was aggressive.118 The presence of EBV+ (or EBV–) B cells and plasma cell expansions typical of TFH lymphomas, has been described in PTCL, NOS as well, albeit less frequently; several of these reports, however, antedate the recognition of nodal PTCL of TFH derivation, and therefore their significance is uncertain in the light of the current definition of PTCL, NOS.119,120 Conventional cytogenetics and array-based studies have documented many aberrations and complex patterns of imbalances.121 A whole-genome sequencing study showed that CDKN2A and PTEN deletions are frequent (46% and 26% of the cases, respectively), and may co-occur; this event is specifically associated with PTCL, NOS and never observed in AITL or ALCL.122 CDKN2A deletions, which were associated with shorter survival in that study, are recurrent

Haematologica | 108 December 2023

3236

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology

B. Bisig et al.

Figure 5. Heterogeneity of peripheral T-cell lymphoma, not otherwise specified. (A-G) Cytotoxic peripheral T-cell lymphoma, not otherwise specified (PTCL, NOS) with large cell morphology and TBX21 (Th1) phenotype. This PTCL, NOS is composed of large pleomorphic cells, with numerous apoptotic bodies (A, hematoxylin & eosin), expresses pan-T-cell markers (B, CD3 and C, CD5), CD4 (D) and cytotoxic markers (E, perforin). Although TBX21 (F) is negative, diffuse expression of CXCR3 (G) subclassifies this lymphoma as PTCL-TBX21. CD30, EBV, GATA3 and CCR4 are also negative (not shown). (H-N) PTCL, NOS with small to mediumsized cells and a GATA3 (Th2) phenotype. This PTCL, NOS comprises diffuse sheets of small to medium-sized cells, with irregular nuclei and occasionally an abundant clear cytoplasm (H, hematoxylin & eosin), is positive for pan-T-cell markers including CD2 (I), and CD4+ (J), and is negative for cytotoxic markers (K, TIA-1). In the absence of TBX21 and CXCR3 expression (not shown), diffuse positivity for GATA3 (L) and CCR4 (M) subclassifies this lymphoma as PTCL-GATA3. P53 protein is diffusely overexpressed (N), reflecting a mutated TP53 gene status (confirmed by sequencing).

in the GATA3 molecular subgroup of PTCL, NOS (see below).84 Few recurrent translocations and several fusion transcripts have been characterized. Overall, their individual prevalence is low, and they are not specific to PTCL, NOS, as they also occur in other entities, in particular ALK– ALCL and TFH lymphomas.72,123-125 The t(6;14)(p25;q11.2) involving the IRF4 locus, has been reported in clinically aggressive cytotoxic PTCL.126,127 TP63 rearrangements with TBL1XR1 or other partner genes, are associated with an aggressive clinical course and bad outcome, as observed in ALK– ALCL.30 Fusions involving VAV1 (VAV1::MYOF1, VAV1::THAP4, VAV1::S100A7) result in increased activation of VAV1 effector pathways and the oncogenic properties of VAV1::MYOF1 were demonstrated in mice in vivo.128 The FYN::TRAF3IP2 fusion, found in PTCL, NOS and TFH lymphomas, activates the NF-κB pathway.125,129 Fusions involving CD28 (CD28::CTLA4 and CD28::ICOS) occur in PTCL, NOS but are more common in TFH lymphomas and adult T-cell leukemia/lymphoma.72

PTCL, NOS harbors recurrent mutations in epigenetic modifiers, most often TET2 or DNMT3A,130 less commonly SMARCA4 or KMT2D, and in genes related to the TCR signaling pathway, notably activating mutations in PLCG1, CD28 and VAV1.71,102,123,131 Mutations in RHOA and IDH2, recurrent in TFH lymphomas, are essentially absent. Alterations in TP53 (mutations and/or deletions, often biallelic) are detected in 40% of the cases, and portend an adverse prognostic significance.84,132 One study found that alterations in TP53 and/or CDKN2A delineate a group of PTCL, NOS characterized by marked genomic instability, mutations in genes related to immune surveillance and immune evasion (HLA-A, HLA-B, CIITA, CD58, CD274), mutations in transcriptional and post-transcriptional regulators, and a worse outcome.132 As outlined below, research continues into the identification of meaningful subgroups of PTCL, NOS. Cytotoxic molecule-positive PTCL, NOS. A subset of PTCL, NOS, ranging from 15% to 30-40% of the cases in various series, express one or several cytotoxic granule-associated

Haematologica | 108 December 2023

3237

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology molecules (TIA-1 and/or granzyme B and/or perforin) indicative of a resting or more commonly activated cytotoxic immunophenotype.109,110,133 Of note, most series of cytotoxic PTCL, NOS have been reported from Asia, and often contained EBV+ cases, which are now classified separately (see above).1,2 In a recently published European cohort of 45 EBV– nodal cytotoxic PTCL,102 the disease affected predominantly males at a median age of 60 years and one-fifth of the patients had a previous history of B-cell lymphoma, solid tumor or underlying immune disorder. Besides a primary nodal presentation, most patients had extranodal disease, and the median survival was only 13 months. Comparison with non-cytotoxic PTCL, NOS in another study demonstrated an inferior overall survival for cytotoxic PTCL, NOS.133 The morphology of these tumors is variable, with predominantly medium-sized to large neoplastic cells and a more or less abundant microenvironment. The neoplastic cells are most commonly CD8+ or CD4– CD8–, less commonly CD4+ or CD4+ CD8+, and in most cases TCRβF1+ or TCR-silent, rarely TCRγδ+.102,110,133 PTCL, NOS with a cytotoxic phenotype express Th1-associated markers, TBX21 or CXCR3, accounting for a subset of PTCL-TBX21 with a more aggressive course (see below); they harbor frequent mutations in epigenetic modifiers, notably in TET2 and DNMT3A, recurrent alterations affecting the TCR and JAK/STAT signaling pathways, including fusions involving VAV1 and CD28, and TP53 mutations in 18% of the cases.102 Cell-of-origin subgroups. Earlier studies suggested that subclasses of PTCL, NOS might be delineated by their immunological profile according to the expression markers associated with Th1 (CXCR3, CCR5, CD134/OX40, CD69, Tbet) or Th2 (CCR4, CXCR4, ST2[L]) differentiation.134,135 These

B. Bisig et al.

classifiers have not been widely applied due to technical difficulty in assessing the markers, often requiring freshfrozen tissue, and are lacking validation studies. Two molecular subgroups of PTCL, NOS, PTCL-TBX21 and PTCL-GATA3, were identified by transcriptome profiling (Table 4), based on signatures similar to those regulated by the transcription factors TBX21 (T-bet) and GATA3, which are master regulators of Th1 and Th2 differentiation pathways, respectively.136 The TBX21 subgroup shows high expression of TBX21 and its target genes (CCL3, CXCR3, EOMES, IFNG, ILR2B), and enrichment of the NF-κB pathway; conversely the GATA3 subgroup is characterized by high expression of GATA3 and its target genes (CCR4, CXCR7, IL18RA), high MYC and proliferation signatures.136 Histologically, PTCL-TBX21 (Figure 5A-G) tends to be polymorphic with a background of reactive inflammatory cells, including cases of lymphoepithelioid (Lennert) lymphoma, while PTCL-GATA3 (Figure 5H-N) tends to lack a prominent inflammatory microenvironment and shows sheets of medium-sized tumor cells with abundant clear cytoplasm or clusters or sheets of large tumor cells.137 Furthermore, GATA3+ PTCL, NOS and cases with a cytotoxic phenotype (clustered within the TBX21 subgroup) were found to have a worse outcome than non-cytotoxic TBX21+ tumors.136,138 In addition, subsequent studies showed distinctive genetic features. PTCL-TBX21 has fewer copy number aberrations, and a higher frequency of mutations in epigenetic modifying genes, especially those involved in DNA methylation (TET1, TET3, and DNMT3A). PTCL-GATA3 has greater genomic complexity; frequent losses of TP53, PTEN, RB1, CDKN2A/B, and PRDM1; gains of STAT3 and MYC; and recurrent mutations of TP53 and PRDM1.84 An immunohisto-

Table 4. Comparison of TBX21 and GATA3 subgroups of peripheral T-cell lymphoma, not otherwise specified. Feature

TBX21

GATA3

Frequency

50-60% of PTCL, NOS

30-40% of PTCL, NOS

High expression of TBX21 and its target genes (CCL3, CXCR3, EOMES, IFNG, ILR2B)

High expression of GATA3 and its target genes (CCR4, CXCR7, IL18RA)

Polymorphous

Medium-sized to large tumor cells Little environment

Cytotoxic subset TBX21+ and/or CXCR3+

TBX21– and CXCR3– GATA3+ and/or CCR4+

Pathways

Enrichment of the NF-κB pathway

High MYC and proliferation signatures

Genetics

Relatively lower genomic complexity, mutations in epigenetic modifiers frequent

Higher genomic complexity, recurrent TP53 alterations, CDKN2A and PTEN deletions

Outcome

Better than GATA3 except for the subset of cytotoxic cases

Worse than TBX21

Defining signature

Morphology

Immunohistochemistry

PTCL: peripheral T-cell lymphoma; NOS: not otherwise specified. Haematologica | 108 December 2023

3238

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology chemical algorithm has been developed as a surrogate to the gene expression profiling-based classification. The algorithm uses four antibodies to TBX21, CXCR3 (a TBX21 transcriptional target), GATA3, and CCR4 (a GATA3 transcriptional target), which are interpreted sequentially.137 Positivity for TBX21 or CXCR3 in 20% or more of the neoplastic cells defines the TBX21 subgroup (Figure 5A-G). Lymphomas negative or below the threshold for TBX21 markers are classified as GATA3 if 50% or more of the neoplastic cells are GATA3+ or CCR4+ (Figure 5H-N). The GATA3 group identified by immunohistochemistry was similarly associated with an inferior overall survival. Other cases remain unclassified. More recently, a simplified transcriptomic assay based on quantification of 153 selected transcripts dedicated for the molecular diagnosis of the major PTCL entities, including the two subtypes of PTCL, NOS, has been implemented on a digital gene expression profiling platform for routinely processed biopsies.139 The identification of the TBX21 and GATA3 subgroups currently does not have an impact on frontline clinical management of the patients and it remains unclear whether there is differential sensitivity to novel therapies. Thus, it is still considered a research tool and not currently required in standard diagnostic practice.

Conclusions and future directions The current schemes for nodal PTCL classification (Figure 1), in continuity with the previous model, emphasize the role of distinct genetic drivers in the definition of PTCL entities or subtypes, and reinforce the concept of cellular derivation being an important determinant of PTCL biology and of clinical relevance. Advances in the genetic characterization of nodal PTCL have added more or less specific defining features of distinct entities (Table 2). Accordingly, genetic testing is increasingly used for diagnostic purposes, and its role as an aid to clinical decision-making is likely to expand in the near future.140 Among ALCL, the ALK– entity is no longer simply “negative for ALK” but a set of genetically distinct subgroups, which require further characterization to assess their clinical and biological relevance. Gray zones remain around the demarcation between CD30+ PTCL, NOS and ALK– ALCL with, in-

B. Bisig et al.

terestingly, some overlap at the genetic level. For example, recurrent JAK2 rearrangements were recently described in PTCL, which had anaplastic features, sometimes ReedSternberg-like cells, frequent CD15 positivity (80%),31 and had been diagnosed in some cases as CD30+ PTCL, NOS and in other cases as ALK– ALCL. Features overlap with cases of PTCL, NOS co-expressing CD30 and CD15 reported earlier.141 While it seems premature to jump to definitive conclusions without collecting additional cases, the question is whether genetics will supersede classic morphological and phenotypic criteria to define ALK– ALCL boundaries. While there are multiple lines of evidence in support of the concept of TFH lymphoma(s), the consideration of one entity versus a group of related disorders differs between the two 2022 classifications, and the distinction between the three types, relying essentially on morphology and immunoarchitecture, may be difficult to apply to a significant fraction of cases showing overlapping features. It is felt that the current definition of the TFH phenotype might be insufficient to capture TFH lymphoma, NOS precisely and distinguish it from PTCL, NOS, and additional criteria, possibly including genetic features, require further research. PTCL, NOS is still viewed as a heterogeneous group of neoplasms that likely do not constitute a single entity, awaiting further identification of meaningful subgroups and substratification. While recent efforts have been made in that direction, and there is growing evidence to substantiate the rationale for molecular or functional subgrouping among PTCL, NOS, there is still a lack of large-scale studies and, at present, these are not recognized as new diagnostic subtypes. Disclosures KJS has received honoraria from and provided consulting for BMS, Merck, Seagen, and Janssen; has sat on a steering committee for Beigene; has received research funding from BMS; has received institutional research funding from Roche; and sat on a Data Safety and Monitoring Committee for DSMC. BB and LdeL have no conflicts of interest to disclose. Contributions BB and LdL conceived the content of the paper, wrote the manuscript and prepared the figures. KS edited the manuscript.

References 1. Campo E, Jaffe ES, Cook JR, et al. The International Consensus Classification of mature lymphoid neoplasms: a report from the Clinical Advisory Committee. Blood. 2022;140(11):1229-1253. 2. Alaggio R, Amador C, Anagnostopoulos I, et al. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: lymphoid neoplasms. Leukemia. 2022;36(7):1720-1748. 3. Swerdlow SH, Campo E, Harris NL, et al. WHO Classification of

Tumours of Haematopoietic and Lymphoid Tissues. Revised 4th ed. Lyon, France: IARC, 2017. 4. Ngu HS, Savage KJ. Past, present and future therapeutic approaches in nodal peripheral T-cell lymphomas. Haematologica 2023;108(12):3211-3226. 5. de Leval L, Parrens M, Le Bras F, et al. Angioimmunoblastic T-cell lymphoma is the most common T-cell lymphoma in two distinct

Haematologica | 108 December 2023

3239

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology French information data sets. Haematologica. 2015;100(9):e361-364. 6. Vose J, Armitage J, Weisenburger D; International T-cell Lymphoma Project. International peripheral T-cell and natural killer/T-cell lymphoma study: pathology findings and clinical outcomes. J Clin Oncol. 2008;26(25):4124-4130. 7. Savage KJ, Harris NL, Vose JM, et al. ALK- anaplastic large-cell lymphoma is clinically and immunophenotypically different from both ALK+ ALCL and peripheral T-cell lymphoma, not otherwise specified: report from the International Peripheral T-Cell Lymphoma Project. Blood. 2008;111(12):5496-5504. 8. Rigaud C, Knorr F, Brugieres L, Woessmann W. Diagnosis and management of ALK-positive anaplastic large cell lymphoma in children and adolescents. Best Pract Res Clin Haematol. 2023;36(1):101444. 9. Graetz D, Crews KR, Azzato EM, et al. Leukemic presentation of ALK-positive anaplastic large cell lymphoma with a novel partner, poly(A) binding protein cytoplasmic 1 (PABPC1), responding to single-agent crizotinib. Haematologica. 2019;104(5):e218-e221. 10. Falini B, Bigerna B, Fizzotti M, et al. ALK expression defines a distinct group of T/null lymphomas ("ALK lymphomas") with a wide morphological spectrum. Am J Pathol. 1998;153(3):875-886. 11. Khanlari M, Li S, Miranda RN, et al. Small cell/lymphohistiocytic morphology is associated with peripheral blood involvement, CD8 positivity and retained T-cell antigens, but not outcome in adults with ALK+ anaplastic large cell lymphoma. Mod Pathol. 2022;35(3):412-418. 12. Vassallo J, Lamant L, Brugieres L, et al. ALK-positive anaplastic large cell lymphoma mimicking nodular sclerosis Hodgkin's lymphoma: report of 10 cases. Am J Surg Pathol. 2006;30(2):223-229. 13. Cordell JL, Pulford KA, Bigerna B, et al. Detection of normal and chimeric nucleophosmin in human cells. Blood. 1999;93(2):632-642. 14. de Leval L. Approach to nodal-based T-cell lymphomas. Pathology. 2020;52(1):78-99. 15. Hallberg B, Palmer RH. Mechanistic insight into ALK receptor tyrosine kinase in human cancer biology. Nat Rev Cancer. 2013;13(10):685-700. 16. Chiarle R, Simmons WJ, Cai H, et al. Stat3 is required for ALKmediated lymphomagenesis and provides a possible therapeutic target. Nat Med. 2005;11(6):623-629. 17. Bisig B, de Reynies A, Bonnet C, et al. CD30-positive peripheral Tcell lymphomas share molecular and phenotypic features. Haematologica. 2013;98(8):1250-1258. 18. Hare L, Burke GAA, Turner SD. Resistance to targeted agents used to treat paediatric ALK-positive ALCL. Cancers (Basel). 2021;13(23):6003. 19. Larose H, Prokoph N, Matthews JD, et al. Whole exome sequencing reveals NOTCH1 mutations in anaplastic large cell lymphoma and points to Notch both as a key pathway and a potential therapeutic target. Haematologica. 2021;106(6):1693-1704. 20. Lobello C, Tichy B, Bystry V, et al. STAT3 and TP53 mutations associate with poor prognosis in anaplastic large cell lymphoma. Leukemia. 2021;35(5):1500-1505. 21. Shustov A, Cabrera ME, Civallero M, et al. ALK-negative anaplastic large cell lymphoma: features and outcomes of 235 patients from the International T-Cell Project. Blood Adv. 2021;5(3):640-648. 22. Sibon D, Bisig B, Bonnet C, et al. ALK-negative anaplastic large cell lymphoma with DUSP22 rearrangement has distinctive disease characteristics with better progression-free survival: a

B. Bisig et al.

LYSA study. Haematologica. 2023;108(6):1590-1603. 23. Crescenzo R, Abate F, Lasorsa E, et al. Convergent mutations and kinase fusions lead to oncogenic STAT3 activation in anaplastic large cell lymphoma. Cancer Cell. 2015;27(4):516-532. 24. Luchtel RA, Dasari S, Oishi N, et al. Molecular profiling reveals immunogenic cues in anaplastic large cell lymphomas with DUSP22 rearrangements. Blood. 2018;132(13):1386-1398. 25. Parrilla Castellar ER, Jaffe ES, Said JW, et al. ALK-negative anaplastic large cell lymphoma is a genetically heterogeneous disease with widely disparate clinical outcomes. Blood. 2014;124(9):1473-1480. 26. Savage KJ, Slack GW. DUSP22-rearranged ALK-negative anaplastic large cell lymphoma is a pathogenetically distinct disease but can have variable clinical outcome. Haematologica. 2023;108(6):1463-1467. 27. Luchtel RA, Zimmermann MT, Hu G, et al. Recurrent MSC (E116K) mutations in ALK-negative anaplastic large cell lymphoma. Blood. 2019;133(26):2776-2789. 28. Feldman AL, Oishi N, Ketterling RP, Ansell SM, Shi M, Dasari S. Immunohistochemical approach to genetic subtyping of anaplastic large cell lymphoma. Am J Surg Pathol. 2022;46(11):1490-1499. 29. Qiu L, Tang G, Li S, et al. DUSP22 rearrangement is associated with a distinctive immunophenotype but not outcome in patients with systemic ALK-negative anaplastic large cell lymphoma. Haematologica. 2023;108(6):1604-1615. 30. Peterson JF, Pearce KE, Meyer RG, et al. Fluorescence in-situ hybridisation for TP63 rearrangements in T cell lymphomas: single-site experience of 470 patients and implications for clinical testing. Histopathology. 2020;76(3):481-485. 31. Fitzpatrick MJ, Massoth LR, Marcus C, et al. JAK2 rearrangements are a recurrent alteration in CD30+ systemic T-cell lymphomas with anaplastic morphology. Am J Surg Pathol. 2021;45(7):895-904. 32. Scarfo I, Pellegrino E, Mereu E, et al. Identification of a new subclass of ALK-negative ALCL expressing aberrant levels of ERBB4 transcripts. Blood. 2016;127(2):221-232. 33. Hu G, Dasari S, Asmann YW, et al. Targetable fusions of the FRK tyrosine kinase in ALK-negative anaplastic large cell lymphoma. Leukemia. 2018;32(2):565-569. 34. Laurent C, Delas A, Gaulard P, et al. Breast implant-associated anaplastic large cell lymphoma: two distinct clinicopathological variants with different outcomes. Ann Oncol. 2016;27(2):306-314. 35. Ghione P, Arcila M, Joseph V, et al. BRCA1/2 mutations impact on the development of breast implant-associated lymphoma (BIAALCL) in women with breast cancer reconstructed with textured breast implants. Hematol Oncol. 2023;41(S2):193-194. 36. Jaffe ES, Ashar BS, Clemens MW, et al. Best pactices guideline for the pathologic diagnosis of breast implant-associated anaplastic large-cell lymphoma. J Clin Oncol. 2020;38(10):1102-1111. 37. Ferrufino-Schmidt MC, Medeiros LJ, Liu H, et al. Clinicopathologic features and prognostic impact of lymph node involvement in patients with breast implant-associated anaplastic large cell lymphoma. Am J Surg Pathol. 2018;42(3):293-305. 38. de Leval L. Chromosomes in breast lymphoma. Blood. 2020;136(25):2848-2849. 39. Laurent C, Nicolae A, Laurent C, et al. Gene alterations in epigenetic modifiers and JAK-STAT signaling are frequent in breast implant-associated ALCL. Blood. 2020;135(5):360-370. 40. Letourneau A, Maerevoet M, Milowich D, et al. Dual JAK1 and STAT3 mutations in a breast implant-associated anaplastic large

Haematologica | 108 December 2023

3240

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology cell lymphoma. Virchows Arch. 2018;473(4):505-511. 41. Crotty S. T follicular helper cell biology: a decade of discovery and diseases. Immunity. 2019;50(5):1132-1148. 42. de Leval L, Rickman DS, Thielen C, et al. The gene expression profile of nodal peripheral T-cell lymphoma demonstrates a molecular link between angioimmunoblastic T-cell lymphoma (AITL) and follicular helper T (TFH) cells. Blood. 2007;109(11):4952-4963. 43. de Leval L, Savilo E, Longtine J, Ferry JA, Harris NL. Peripheral Tcell lymphoma with follicular involvement and a CD4+/bcl-6+ phenotype. Am J Surg Pathol. 2001;25(3):395-400. 44. Lemonnier F, Couronne L, Parrens M, et al. Recurrent TET2 mutations in peripheral T-cell lymphomas correlate with TFH-like features and adverse clinical parameters. Blood. 2012;120(7):1466-1469. 45. Palomero T, Couronne L, Khiabanian H, et al. Recurrent mutations in epigenetic regulators, RHOA and FYN kinase in peripheral T cell lymphomas. Nat Genet. 2014;46(2):166-170. 46. Sakata-Yanagimoto M, Enami T, Yoshida K, et al. Somatic RHOA mutation in angioimmunoblastic T cell lymphoma. Nat Genet. 2014;46(2):171-175. 47. Vallois D, Dobay MP, Morin RD, et al. Activating mutations in genes related to TCR signaling in angioimmunoblastic and other follicular helper T-cell-derived lymphomas. Blood. 2016;128(11):1490-1502. 48. Dobay MP, Lemonnier F, Missiaglia E, et al. Integrative clinicopathological and molecular analyses of angioimmunoblastic T-cell lymphoma and other nodal lymphomas of follicular helper T-cell origin. Haematologica. 2017;102(4):e148-e151. 49. Bachy E, Camus V, Thieblemont C, et al. Romidepsin plus CHOP versus CHOP in patients with previously untreated peripheral Tcell lymphoma: results of the Ro-CHOP phase III study (conducted by LYSA). J Clin Oncol. 2022;40(3):242-251. 50. Feldman AL, Laurent C, Narbaitz M, et al. Classification and diagnostic evaluation of nodal T- and NK-cell lymphomas. Virchows Arch. 2023;482(1):265-279. 51. Hsi ED, Said J, Macon WR, et al. Diagnostic accuracy of a defined immunophenotypic and molecular genetic approach for peripheral T/NK-cell lymphomas. A North American PTCL study group project. Am J Surg Pathol. 2014;38(6):768-775. 52. Yoon SE, Song Y, Kim SJ, et al. Comprehensive analysis of peripheral T-cell and natural killer/T-cell lymphoma in Asian patients: a multinational, multicenter, prospective registry study in Asia. Lancet Reg Health West Pac. 2021;10:100126. 53. Federico M, Rudiger T, Bellei M, et al. Clinicopathologic characteristics of angioimmunoblastic T-cell lymphoma: analysis of the International Peripheral T-Cell Lymphoma project. J Clin Oncol. 2013;31(2):240-246. 54. de Leval L, Gisselbrecht C, Gaulard P. Advances in the understanding and management of angioimmunoblastic T-cell lymphoma. Br J Haematol. 2010;148(5):673-689. 55. Rodriguez-Justo M, Attygalle AD, Munson P, Roncador G, Marafioti T, Piris MA. Angioimmunoblastic T-cell lymphoma with hyperplastic germinal centres: a neoplasia with origin in the outer zone of the germinal centre? Clinicopathological and immunohistochemical study of 10 cases with follicular T-cell markers. Mod Pathol. 2009;22(6):753-761. 56. Attygalle AD, Cabecadas J, Gaulard P, et al. Peripheral T-cell and NK-cell lymphomas and their mimics; taking a step forward report on the lymphoma workshop of the XVIth meeting of the European Association for Haematopathology and the Society for Hematopathology. Histopathology. 2014;64(2):171-199. 57. Iqbal J, Weisenburger DD, Greiner TC, et al. Molecular signatures to improve diagnosis in peripheral T-cell lymphoma and

B. Bisig et al.

prognostication in angioimmunoblastic T-cell lymphoma. Blood. 2010;115(5):1026-1036. 58. Rodriguez M, Alonso-Alonso R, Tomas-Roca L, et al. Peripheral Tcell lymphoma: molecular profiling recognizes subclasses and identifies prognostic markers. Blood Adv. 2021;5(24):5588-5598. 59. Sugio T, Miyawaki K, Kato K, et al. Microenvironmental immune cell signatures dictate clinical outcomes for PTCL-NOS. Blood Adv. 2018;2(17):2242-2252. 60. Bisig B, Thielen C, Herens C, et al. c-Maf expression in angioimmunoblastic T-cell lymphoma reflects follicular helper Tcell derivation rather than oncogenesis. Histopathology. 2012;60(2):371-376. 61. Gaulard P, de Leval L. Follicular helper T cells: implications in neoplastic hematopathology. Semin Diagn Pathol. 2011;28(3):202-213. 62. Bossard C, Dobay MP, Parrens M, et al. Immunohistochemistry as a valuable tool to assess CD30 expression in peripheral T-cell lymphomas: high correlation with mRNA levels. Blood. 2014;124(19):2983-2986. 63. Onaindia A, Martinez N, Montes-Moreno S, et al. CD30 expression by B and T cells: a frequent finding in angioimmunoblastic T-cell lymphoma and peripheral T-cell lymphoma-not otherwise specified. Am J Surg Pathol. 2016;40(3):378-385. 64. Chen Z, Zhu Q, Deng X, et al. Angioimmunoblastic T-cell lymphoma with predominant CD8+ tumor-infiltrating T-cells is a distinct immune pattern with an immunosuppressive microenvironment. Front Immunol. 2022;13:987227. 65. Nicolae A, Pittaluga S, Venkataraman G, et al. Peripheral T-cell lymphomas of follicular T-helper cell derivation with Hodgkin/Reed-Sternberg cells of B-cell lineage: both EBVpositive and EBV-negative variants exist. Am J Surg Pathol. 2013;37(6):816-826. 66. Willenbrock K, Brauninger A, Hansmann ML. Frequent occurrence of B-cell lymphomas in angioimmunoblastic T-cell lymphoma and proliferation of Epstein-Barr virus-infected cells in early cases. Br J Haematol. 2007;138(6):733-739. 67. Lewis NE, Petrova-Drus K, Huet S, et al. Clonal hematopoiesis in angioimmunoblastic T-cell lymphoma with divergent evolution to myeloid neoplasms. Blood Adv. 2020;4(10):2261-2271. 68. Lemonnier F, Gaulard P, de Leval L. New insights in the pathogenesis of T-cell lymphomas. Curr Opin Oncol. 2018;30(5):277-284. 69. Lemonnier F, Safar V, Beldi-Ferchiou A, et al. Integrative analysis of a phase 2 trial combining lenalidomide with CHOP in angioimmunoblastic T-cell lymphoma. Blood Adv. 2021;5(2):539-548. 70. Zheng J, Wang Z, Pan X, et al. DNMT3A(R882H) accelerates angioimmunoblastic T-cell lymphoma in mice. Oncogene. 2023;42(23):1940-1950. 71. Rohr J, Guo S, Huo J, et al. Recurrent activating mutations of CD28 in peripheral T-cell lymphomas. Leukemia. 2016;30(5):1062-1070. 72. Vallois D, Dupuy A, Lemonnier F, et al. RNA fusions involving CD28 are rare in peripheral T-cell lymphomas and concentrate mainly in those derived from follicular helper T cells. Haematologica. 2018;103(8):e360-e363. 73. Zang S, Li J, Yang H, et al. Mutations in 5-methylcytosine oxidase TET2 and RhoA cooperatively disrupt T cell homeostasis. J Clin Invest. 2017;127(8):2998-3012. 74. Ng SY, Brown L, Stevenson K, et al. RhoA G17V is sufficient to induce autoimmunity and promotes T-cell lymphomagenesis in mice. Blood. 2018;132(9):935-947. 75. Cortes JR, Ambesi-Impiombato A, Couronne L, et al. RHOA G17V induces T follicular helper cell specification and promotes

Haematologica | 108 December 2023

3241

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology lymphomagenesis. Cancer Cell. 2018;33(2):259-273. 76. Ondrejka SL, Grzywacz B, Bodo J, et al. Angioimmunoblastic Tcell lymphomas with the RHOA p.Gly17Val mutation have classic clinical and pathologic features. Am J Surg Pathol. 2016;40(3):335-341. 77. Nagao R, Kikuti YY, Carreras J, et al. Clinicopathologic analysis of angioimmunoblastic T-cell lymphoma with or without RHOA G17V mutation using formalin-fixed paraffin-embedded sections. Am J Surg Pathol. 2016;40(8):1041-1050. 78. Dobson R, Du PY, Raso-Barnett L, et al. Early detection of T-cell lymphoma with T follicular helper phenotype by RHOA mutation analysis. Haematologica. 2022;107(2):489-499. 79. Cairns RA, Iqbal J, Lemonnier F, et al. IDH2 mutations are frequent in angioimmunoblastic T-cell lymphoma. Blood. 2012;119(8):1901-1903. 80. Lemonnier F, Cairns RA, Inoue S, et al. The IDH2 R172K mutation associated with angioimmunoblastic T-cell lymphoma produces 2HG in T cells and impacts lymphoid development. Proc Natl Acad Sci U S A. 2016;113(52):15084-15089. 81. Wang C, McKeithan TW, Gong Q, et al. IDH2R172 mutations define a unique subgroup of patients with angioimmunoblastic T-cell lymphoma. Blood. 2015;126(15):1741-1752. 82. Steinhilber J, Mederake M, Bonzheim I, et al. The pathological features of angioimmunoblastic T-cell lymphomas with IDH2(R172) mutations. Mod Pathol. 2019;32(8):1123-1134. 83. Leca J, Lemonnier F, Meydan C, et al. IDH2 and TET2 mutations synergize to modulate T follicular helper cell functional interaction with the AITL microenvironment. Cancer Cell. 2023;41(2):323-339. 84. Heavican TB, Bouska A, Yu J, et al. Genetic drivers of oncogenic pathways in molecular subgroups of peripheral T-cell lymphoma. Blood. 2019;133(15):1664-1676. 85. Yao WQ, Wu F, Zhang W, et al. Angioimmunoblastic T-cell lymphoma contains multiple clonal T-cell populations derived from a common TET2 mutant progenitor cell. J Pathol. 2020;250(3):346-357. 86. Nguyen TB, Sakata-Yanagimoto M, Asabe Y, et al. Identification of cell-type-specific mutations in nodal T-cell lymphomas. Blood Cancer J. 2017;7(1):e516. 87. Schwartz FH, Cai Q, Fellmann E, et al. TET2 mutations in B cells of patients affected by angioimmunoblastic T-cell lymphoma. J Pathol. 2017;242(2):129-133. 88. Lemonnier F, Dupuis J, Sujobert P, et al. Treatment with 5azacytidine induces a sustained response in patients with angioimmunoblastic T-cell lymphoma. Blood. 2018;132(21):2305-2309. 89. Huang Y, Moreau A, Dupuis J, et al. Peripheral T-cell lymphomas with a follicular growth pattern are derived from follicular helper T cells (TFH) and may show overlapping features with angioimmunoblastic T-cell lymphomas. Am J Surg Pathol. 2009;33(5):682-690. 90. Miyoshi H, Sato K, Niino D, et al. Clinicopathologic analysis of peripheral T-cell lymphoma, follicular variant, and comparison with angioimmunoblastic T-cell lymphoma: Bcl-6 expression might affect progression between these disorders. Am J Clin Pathol. 2012;137(6):879-889. 91. Hartmann S, Goncharova O, Portyanko A, et al. CD30 expression in neoplastic T cells of follicular T cell lymphoma is a helpful diagnostic tool in the differential diagnosis of Hodgkin lymphoma. Mod Pathol. 2019;32(1):37-47. 92. Moroch J, Copie-Bergman C, de Leval L, et al. Follicular peripheral T-cell lymphoma expands the spectrum of classical Hodgkin lymphoma mimics. Am J Surg Pathol. 2012;36(11):1636-1646. 93. Alikhan M, Song JY, Sohani AR, et al. Peripheral T-cell lymphomas

B. Bisig et al.

of follicular helper T-cell type frequently display an aberrant CD3(-/dim)CD4(+) population by flow cytometry: an important clue to the diagnosis of a Hodgkin lymphoma mimic. Mod Pathol. 2016;29(10):1173-1182. 94. Attygalle AD, Feldman AL, Dogan A. ITK/SYK translocation in angioimmunoblastic T-cell lymphoma. Am J Surg Pathol. 2013;37(9):1456-1457. 95. Agostinelli C, Hartmann S, Klapper W, et al. Peripheral T cell lymphomas with follicular T helper phenotype: a new basket or a distinct entity? Revising Karl Lennert's personal archive. Histopathology. 2011;59(4):679-691. 96. Yoon SE, Cho J, Kim YJ, et al. Comprehensive analysis of clinical, pathological, and genomic characteristics of follicular helper Tcell derived lymphomas. Exp Hematol Oncol. 2021;10(1):33. 97. Kato S, Takahashi E, Asano N, et al. Nodal cytotoxic molecule (CM)-positive Epstein-Barr virus (EBV)-associated peripheral T cell lymphoma (PTCL): a clinicopathological study of 26 cases. Histopathology. 2012;61(2):186-199. 98. Asano N, Kato S, Nakamura S. Epstein-Barr virus-associated natural killer/T-cell lymphomas. Best Pract Res Clin Haematol. 2013;26(1):15-21. 99. Jeon YK, Kim JH, Sung JY, Han JH, Ko YH, Hematopathology Study Group of the Korean Society of Pathologists. Epstein-Barr virus-positive nodal T/NK-cell lymphoma: an analysis of 15 cases with distinct clinicopathological features. Hum Pathol. 2015;46(7):981-990. 100. Kato S, Yamashita D, Nakamura S. Nodal EBV+ cytotoxic T-cell lymphoma: a literature review based on the 2017 WHO classification. J Clin Exp Hematop. 2020;60(2):30-36. 101. Wai CMM, Chen S, Phyu T, et al. Immune pathway upregulation and lower genomic instability distinguish EBV-positive nodal T/NK-cell lymphoma from ENKTL and PTCL-NOS. Haematologica. 2022;107(8):1864-1879. 102. Nicolae A, Bouilly J, Lara D, et al. Nodal cytotoxic peripheral Tcell lymphoma occurs frequently in the clinical setting of immunodysregulation and is associated with recurrent epigenetic alterations. Mod Pathol. 2022;35(8):1126-1136. 103. Weisenburger DD, Savage KJ, Harris NL, et al. Peripheral T-cell lymphoma, not otherwise specified: a report of 340 cases from the International Peripheral T-cell Lymphoma Project. Blood. 2011;117(12):3402-3408. 104. Federico M, Bellei M, Marcheselli L, et al. Peripheral T cell lymphoma, not otherwise specified (PTCL-NOS). A new prognostic model developed by the International T cell Project Network. Br J Haematol. 2018;181(6):760-769. 105. Lefevre G, Copin MC, Roumier C, et al. CD3-CD4+ lymphoid variant of hypereosinophilic syndrome: nodal and extranodal histopathological and immunophenotypic features of a peripheral indolent clonal T-cell lymphoproliferative disorder. Haematologica. 2015;100(8):1086-1095. 106. Trimech M, Letourneau A, Missiaglia E, et al. Angioimmunoblastic T-cell lymphoma and chronic lymphocytic leukemia/small lymphocytic lymphoma: a novel form of composite lymphoma potentially mimicking Richter syndrome. Am J Surg Pathol. 2021;45(6):773-786. 107. Hartmann S, Agostinelli C, Klapper W, et al. Revising the historical collection of epithelioid cell-rich lymphomas of the Kiel Lymph Node Registry: what is Lennert's lymphoma nowadays? Histopathology. 2011;59(6):1173-1182. 108. Kurita D, Miyoshi H, Yoshida N, et al. A clinicopathologic study of Lennert lymphoma and possible prognostic factors: the importance of follicular helper T-cell markers and the association with angioimmunoblastic T-cell lymphoma. Am J Surg Pathol. 2016;40(9):1249-1260.

Haematologica | 108 December 2023

3242

REVIEW SERIES - Nodal peripheral T-cell lymphoma pathobiology 109. Kagami Y, Suzuki R, Taji H, et al. Nodal cytotoxic lymphoma spectrum: a clinicopathologic study of 66 patients. Am J Surg Pathol. 1999;23(10):1184-1200. 110. Geissinger E, Odenwald T, Lee SS, et al. Nodal peripheral T-cell lymphomas and, in particular, their lymphoepithelioid (Lennert's) variant are often derived from CD8(+) cytotoxic T-cells. Virchows Arch. 2004;445(4):334-343. 111. Gafencu GA, Selicean SE, Petrushev B, et al. Clinicopathological analysis of a case series of peripheral T-cell lymphomas, not otherwise specified, of lymphoepithelioid variant (Lennert's lymphoma). A Central European single-center study. Hum Pathol. 2016;53(192-194. 112. Bonzheim I, Geissinger E, Roth S, et al. Anaplastic large cell lymphomas lack the expression of T-cell receptor molecules or molecules of proximal T-cell receptor signaling. Blood. 2004;104(10):3358-3360. 113. O'Malley DP, Chizhevsky V, Grimm KE, Hii A, Weiss LM. Utility of BCL2, PD1, and CD25 immunohistochemical expression in the diagnosis of T-cell lymphomas. Appl Immunohistochem Mol Morphol. 2014;22(2):99-104. 114. Siddiqui F, Perez Silos V, Karube K, et al. B-cell lymphoma-2 downregulation is a useful feature supporting a neoplastic phenotype in mature T-cell lymphomas. Hum Pathol. 2022;125:48-58. 115. Rahemtullah A, Longtine JA, Harris NL, et al. CD20+ T-cell lymphoma: clinicopathologic analysis of 9 cases and a review of the literature. Am J Surg Pathol. 2008;32(11):1593-1607. 116. Bonzheim I, Geissinger E, Tinguely M, et al. Evaluation of FoxP3 expression in peripheral T-cell lymphoma. Am J Clin Pathol. 2008;130(4):613-619. 117. Narducci MG, Pescarmona E, Lazzeri C, et al. Regulation of TCL1 expression in B- and T-cell lymphomas and reactive lymphoid tissues. Cancer Res. 2000;60(8):2095-2100. 118. Satou A, Asano N, Kato S, et al. FoxP3-positive T cell lymphoma arising in non-HTLV1 carrier: clinicopathological analysis of 11 cases of PTCL-NOS and 2 cases of mycosis fungoides. Histopathology. 2016;68(7):1099-1108. 119. Balague O, Martinez A, Colomo L, et al. Epstein-Barr virus negative clonal plasma cell proliferations and lymphomas in peripheral T-cell lymphomas: a phenomenon with distinctive clinicopathologic features. Am J Surg Pathol. 2007;31(9):1310-1322. 120. Eladl AE, Satou A, Elsayed AA, et al. Clinicopathological study of 30 cases of peripheral T-cell lymphoma with Hodgkin and ReedSternberg-like B-cells from Japan. Am J Surg Pathol. 2017;41(4):506-516. 121. de Leval L, Bisig B, Thielen C, Boniver J, Gaulard P. Molecular classification of T-cell lymphomas. Crit Rev Oncol Hematol. 2009;72(2):125-143. 122. Maura F, Dodero A, Carniti C, et al. CDKN2A deletion is a frequent event associated with poor outcome in patients with peripheral T-cell lymphoma not otherwise specified (PTCL-NOS). Haematologica. 2021;106(11):2918-2926. 123. Abate F, da Silva-Almeida AC, Zairis S, et al. Activating mutations and translocations in the guanine exchange factor VAV1 in peripheral T-cell lymphomas. Proc Natl Acad Sci U S A. 2017;114(4):764-769. 124. Drieux F, Ruminy P, Sater V, et al. Detection of gene fusion transcripts in peripheral T-cell lymphoma using a multiplexed targeted sequencing assay. J Mol Diagn. 2021;23(8):929-940.

B. Bisig et al.

125. Debackere K, Marcelis L, Demeyer S, et al. Fusion transcripts FYN-TRAF3IP2 and KHDRBS1-LCK hijack T cell receptor signaling in peripheral T-cell lymphoma, not otherwise specified. Nat Commun. 2021;12(1):3705. 126. Somja J, Bisig B, Bonnet C, Herens C, Siebert R, de Leval L. Peripheral T-cell lymphoma with t(6;14)(p25;q11.2) translocation presenting with massive splenomegaly. Virchows Arch. 2014;464(6):735-741. 127. Feldman AL, Law M, Remstein ED, et al. Recurrent translocations involving the IRF4 oncogene locus in peripheral T-cell lymphomas. Leukemia. 2009;23(3):574-580. 128. Cortes JR, Filip I, Albero R, et al. Oncogenic Vav1-Myo1f induces therapeutically targetable macrophage-rich tumor microenvironment in peripheral T cell lymphoma. Cell Rep. 2022;39(3):110695. 129. Moon CS, Reglero C, Cortes JR, et al. FYN-TRAF3IP2 induces NFkappaB signaling-driven peripheral T cell lymphoma. Nat Cancer. 2021;2(1):98-113. 130. Herek TA, Bouska A, Lone W, et al. DNMT3A mutations define a unique biological and prognostic subgroup associated with cytotoxic T cells in PTCL-NOS. Blood. 2022;140(11):1278-1290. 131. Manso R, Rodriguez-Pinilla SM, Gonzalez-Rincon J, et al. Recurrent presence of the PLCG1 S345F mutation in nodal peripheral T-cell lymphomas. Haematologica. 2015;100(1):e25-27. 132. Watatani Y, Sato Y, Miyoshi H, et al. Molecular heterogeneity in peripheral T-cell lymphoma, not otherwise specified revealed by comprehensive genetic profiling. Leukemia. 2019;33(12):2867-2883. 133. Asano N, Suzuki R, Kagami Y, et al. Clinicopathologic and prognostic significance of cytotoxic molecule expression in nodal peripheral T-cell lymphoma, unspecified. Am J Surg Pathol. 2005;29(10):1284-1293. 134. Jones D, O'Hara C, Kraus MD, et al. Expression pattern of T-cellassociated chemokine receptors and their chemokines correlates with specific subtypes of T-cell non-Hodgkin lymphoma. Blood. 2000;96(2):685-690. 135. Tsuchiya T, Ohshima K, Karube K, et al. Th1, Th2, and activated Tcell marker and clinical prognosis in peripheral T-cell lymphoma, unspecified: comparison with AILD, ALCL, lymphoblastic lymphoma, and ATLL. Blood. 2004;103(1):236-241. 136. Iqbal J, Wright G, Wang C, et al. Gene expression signatures delineate biological and prognostic subgroups in peripheral T-cell lymphoma. Blood. 2014;123(19):2915-2923. 137. Amador C, Greiner TC, Heavican TB, et al. Reproducing the molecular subclassification of peripheral T-cell lymphoma-NOS by immunohistochemistry. Blood. 2019;134(24):2159-2170. 138. Wang T, Feldman AL, Wada DA, et al. GATA-3 expression identifies a high-risk subset of PTCL, NOS with distinct molecular and clinical features. Blood. 2014;123(19):3007-3015. 139. Amador C, Bouska A, Wright G, et al. Gene expression signatures for the accurate diagnosis of peripheral T-cell lymphoma entities in the routine clinical practice. J Clin Oncol. 2022;40(36):4261-4275. 140. de Leval L, Alizadeh AA, Bergsagel PL, et al. Genomic profiling for clinical decision making in lymphoid neoplasms. Blood. 2022;140(21):2193-2227. 141. Barry TS, Jaffe ES, Sorbara L, Raffeld M, Pittaluga S. Peripheral Tcell lymphomas expressing CD30 and CD15. Am J Surg Pathol. 2003;27(12):1513-1522.

Haematologica | 108 December 2023

3243

REVIEW SERIES

Few and far between: clinical management of rare extranodal subtypes of mature T-cell and NK-cell lymphomas Correspondence:

R. Stuver

stuverr@mskcc.org Received: Accepted:

Robert Stuver,1 Zachary D. Epstein-Peterson1,2 and Steven M. Horwitz1,2,3 1

Lymphoma Service, Department of Medicine, Memorial Sloan Kettering Cancer Center; Department of Medicine, Weill Cornell Medical College and 3Cellular Therapy Service, Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA

May 2, 2023. July 3, 2023.

Abstract While all peripheral T-cell lymphomas are uncommon, certain subtypes are truly rare, with less than a few hundred cases per year in the USA. There are often no dedicated clinical trials in these rare subtypes, and data are generally limited to case reports and retrospective case series. Therefore, clinical management is often based on this limited literature and extrapolation of data from the more common, nodal T-cell lymphomas in conjunction with personal experience. Nevertheless, thanks to tremendous pre-clinical efforts to understand these rare diseases, an increasing appreciation of the biological changes that underlie these entities is forming. In this review, we attempt to summarize the relevant literature regarding the initial management of certain rare subtypes, specifically subcutaneous panniculitis-like T-cell lymphoma, hepatosplenic T-cell lymphoma, intestinal T-cell lymphomas, and extranodal NK/T-cell lymphoma. While unequivocally established approaches in these diseases do not exist, we make cautious efforts to provide our approaches to clinical management when possible.

Introduction The mature T-cell and NK-cell lymphomas are a broad class of heterogeneous clinicopathological entities. The list of unique disease subtypes encompassed in this group is seemingly ever-growing, now numbering 34 distinct subtypes in the recently updated fifth edition of the World Health Organization (WHO) Classification of Haematolymphoid Tumours.1 Similarly, in the first edition of the International Consensus Classification (ICC) of Mature Lymphoid Neoplasms, over 30 entities are named.2 Despite the breadth of pathology, T-cell and NK-cell lymphomas are exceptionally rare diseases, with an annual incidence in the USA of under 2 cases per 100,000 persons.3 Individual disease incidence rates vary from 0.4-0.5 per 100,000 for the most common subtypes, such as peripheral T-cell lymphoma, not otherwise specified (PTCLNOS), to ≤0.1 per 100,000 for rarer subtypes, such as hepatosplenic T-cell lymphoma (HSTCL) and extranodal NK/T-cell lymphoma (ENKL).3 This rarity has made dedicated study and clinical management of T-cell lymphomas particularly challenging, as clinicians are often forced to rely on case series and anecdotal reports in the face of relatively few clinical trials, especially for the rarest subtypes. Acknowledging this challenge, our goal herein is to

summarize relevant literature on the clinical management for certain rare subtypes of mature T-cell and NK-cell lymphomas (Figure 1).

Subcutaneous panniculitis-like T-cell lymphoma Subcutaneous panniculitis-like T-cell lymphoma (SPTCL) falls under the category of primary cutaneous T-cell lymphomas, including mycosis fungoides, which are reviewed elsewhere.4 SPTCL was first recognized as a distinct WHO entity in 2001, although until reliable immunohistochemistry markers became readily available in the early 2000s, SPTCL was often grouped with the related but distinct primary cutaneous γδ T-cell lymphoma (PCGDTCL), which demands a different therapeutic approach.5 However, it is now realized that SPTCL evolves from a mature cytotoxic αβ T cell, and cases expressing the γδ T-cell receptor are considered PCGDTCL. SPTCL classically presents in middle-aged females as multiple subcutaneous, non-ulcerated nodules or plaques.6,7 A personal or family history of autoimmunity, especially lupus erythematosus, may be present in up to one-third of patients, and the distinction between SPTCL

Haematologica | 108 December 2023

3244

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas

R. Stuver et al.

Figure 1. Overview of the rare subtypes of T-cell lymphoma covered in this review. Figure created at BioRender.com. GI: gastrointestinal; LPD: lymphoproliferative disorder.

and lupus panniculitis is often challenging.8 Involvement of the legs seems to be slightly more common, but involvement of the head/neck and trunk has been documented.6,7 Extracutaneous involvement, including nodal, bone marrow, and/or visceral disease, appears to be very uncommon and should be confirmed with biopsy. Many patients have constitutional symptoms, such as fevers and night sweats.6,7 In roughly one-fifth of patients, hemophagocytic lymphohistiocytosis (HLH) is present.7 HLH appears to be more common and more severe in individuals harboring germline mutations in HAVCR2.9-12 A suggested management approach for SPTCL in shown in Figure 2. The clinical evaluation is rooted in a proper pathological confirmation showing a lymphoid infiltrate involving fat lobules, with neoplastic cells having a mature αβ T-cell phenotype, usually CD8+ and negative for Epstein-Barr virus (EBV), with expression of cytotoxic

markers. We perform staging evaluation with positron emission tomography/computed tomography (PET/CT), as most lesions are avid with a median SUV near 9.0.7 In addition, we undertake routine laboratory studies, including serology for human T-cell lymphotropic virus-1 to exclude cutaneous manifestations of adult T-cell leukemia/lymphoma. Diagnostic evaluation for HLH, such as measurement of ferritin, fibrinogen, triglycerides, and soluble IL-2 receptor, can be performed (we consider such evaluation in all patients as a baseline and always evaluate patients with high clinical suspicion). We often perform a rudimentary investigation for autoimmune disorders, particularly in patients in whom lupus panniculitis is entertained, by determining whether antinuclear antibodies and other serological markers are present. Positive or ambiguous results should prompt a formal rheumatology evaluation. Finally, we consider referring patients with a suspected or

Haematologica | 108 December 2023

3245

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas

R. Stuver et al.

Figure 2. Suggested clinical management schema for subcutaneous panniculitis-like T-cell lymphoma. Subcutaneous panniculitis-like T-cell lymphoma (SPTCL) usually presents as multifocal subcutaneous, non-ulcerated nodules, commonly involving the legs. Constitutional symptoms can be present, and rarely patients present with evidence of hemophagocytic lymphohistiocytosis (HLH). HLH appears to be more common in those with germline HAVCR2 mutations. Considerations in diagnostic and staging procedures are shown. In the absence of HLH and/or clinical deterioration, SPTCL can be managed with conservative immunomodulatory strategies. Figure created at BioRender.com. SLE: systemic lupus erythematosus; PET/CT: positron electron tomography/computed tomography; CBC: complete blood count; CMP: chemistry panel; LDH: lactate dehydrogenase; HTLV-1: human T-cell lymphotropic virus-1.

confirmed diagnosis to clinical genetics for HAVCR2 singlegene testing as a risk-stratification tool, although this testing is not widely available and not required for optimal management. No specific recommendations for additional testing of family members exist for patients with detected HAVCR2 aberrations, and we rely on genetic counseling for formal advice. The management of SPTCL should be framed by the understanding that this is most often an indolent disorder

for which immunomodulatory approaches are often effective.6,7 As noted previously, prior grouping of SPTCL with PCGDTCL – a generally very aggressive primary cutaneous lymphoma – initially led to concern that SPTCL was a similarly aggressive subtype. However, in 2008, the European Organization for Research and Treatment of Cancer compared clinical outcomes of 63 cases of SPTCL (at that time, referred to as αβ SPTCL) versus 20 cases of PCGDTCL (at that time, referred to as γδ SPTCL).6 Most patients in

Haematologica | 108 December 2023

3246

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas this series were treated with anthracycline-based combination chemotherapy, such as CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone). Five-year overall survival (OS) was markedly different between the two groups, at 82% for SPTCL versus 11% for PCGDTCL (P<0.001). A more recent, multi-institutional cohort of 95 cases of SPTCL and related adipotropic lymphoproliferative disorders between 1998 and 2018 confirmed the generally indolent behavior of SPTCL.7 With a median follow-up of 56 months, 67% of patients achieved a complete response (CR) to a median of three cumulative therapies. While relapses were common, no patients died of disease or HLH. Immunomodulatory agents included systemic steroids, cyclosporine, methotrexate, and others, with an objective response rate (ORR) of 52%. In particular, most patients treated with cyclosporine had a response (94%), and methotrexate as a first-line agent in seven patients produced a response in all seven. Therefore, for most patients, our approach is conservative immunosuppression with any of the agents above (in particular, cyclosporine or methotrexate), taking into consideration current symptoms, coexisting conditions, and concurrent medications. Other, smaller series show equally high response rates to immunosuppression.13-15 We assess response with PET/CT, and in those with evidence of HLH, with continued assessment of abnormal clinical and laboratory parameters. For those with relapsed/refractory (R/R) disease, we sequence immunosuppressive agents in efforts to avoid combination chemotherapy. In clinically aggressive or multiply relapsed disease, combination chemotherapy, often with consolidative transplant (autologous and allogeneic) has efficacy, similar to paradigms for PCGDTL or nodal peripheral T-cell lymphomas.6,7,16 The optimal management of frank HLH is unclear, and treatment of the underlying SPTCL is logical, with consideration for steroids and etoposide.17 Emerging data on ruxolitinib in the treatment of HLH, predominantly in the pediatric population, are encouraging,18 and we consider this agent in persistent HLH. At least one case report describes this approach in a patient with underlying SPTCL.19 Finally, CD30 expression is rare and, if detected, usually weak in intensity.6,7,20 Therefore, while we could consider brentuximab vedotin in multiply R/R disease, we acknowledge the absence of data.

Hepatosplenic T-cell lymphoma HSTCL classically presents in males in the setting of chronic immune suppression or dysregulation, specifically inflammatory bowel disease (IBD) or after solid organ transplantation.21-23 In three representative case series, up to 20% of patients had IBD, an autoimmune disorder other than IBD, or had received a solid organ transplant.21-23 The

R. Stuver et al.

association with IBD and IBD-directed therapy, in particular, has been reviewed in detail, stemming from a 2006 report of eight cases (six fatal) in seven males and one female with IBD who were treated with the tumor necrosis factor-α inhibitor, infliximab, in combination with additional immunosuppressant therapy.24 This led to a Food and Drug Administration boxed warning for infliximab in 2006. An epidemiological survey of over three million reports to the Food and Drug Administration Adverse Event Reporting System identified 30 unique cases of incident HSTCL in patients under chronic immunosuppression.25 Available evidence seems to suggest that the greatest risk for HSTCL is in those receiving concomitant tumor necrosis factor-α blockade and thiopurines. A study from the International T-cell Lymphoma Project (ITCP) shows that HSTCL can present in older individuals as well.26 HSTCL generally presents aggressively with hepatosplenomegaly and cytopenias. Bone marrow is nearly always involved, but lymphadenopathy is uncommon and should prompt consideration of a biopsy.22-24 Bone marrow or liver biopsy is mandatory to demonstrate a predominantly intrasinusoidal infiltrate of mature γδ (rarely αβ) cytotoxic T cells, commonly with isochrome 7q and trisomy 8 chromosomal abnormalities (detectable by fluorescence in situ hybridization)27-31 and mutational signatures with enrichment in genes of the JAK/STAT pathway and chromatin modification, such as SETD2.32-35 The diagnosis must be distinguished from other γδ T-cell lymphomas, particularly γδ T-cell large granular lymphocytic leukemia, through careful clinical and pathological review, focusing on histology, cytogenetics, and molecular findings if available.36 Therapy of curative intent for HSTCL entails combination chemotherapy induction followed by allogeneic hematopoietic stem cell transplant (alloHCT) in first remission. The optimal induction regimen is unclear (Table 1), although available evidence appears to suggest that CHOP is inadequate with refractory disease being common. In an early evaluation of 21 cases, all except two of whom were treated with CHOP/CHOP-like therapy, the median OS was 16 months; all patients died except the two patients who received non-CHOP induction.21 A second report of 15 cases from the MD Anderson Cancer Center similarly found that of six patients treated with CHOP/CHOP-like induction, all died within 2 years.22 In this series, all four surviving patients were treated with nonCHOP induction. An analysis of 166 cases of HSTCL between 1990 and 2018 showed a significantly increased ORR in patients receiving cytarabine/etoposide/platinumcontaining regimens (ORR: 82%; CR: 56%) versus CHOP/CHOP-like regimens (ORR: 52%; CR: 38%).37 The median OS was significantly prolonged in the non-CHOP induction group at 36.5 versus 18 months (hazard ratio [HR]=0.33, 95% confidence interval [95% CI]: 0.19-0.58). In the absence of prospective or randomized data, this

Haematologica | 108 December 2023

3247

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas

R. Stuver et al.

Table 1. Selected series detailing outcomes in hepatosplenic T-cell lymphoma.

Series Falchook N=15

Induction

ORR, CR

Transplant

CHOP-like: 5 hyperCVAD: 4 other: 6

CHOP/CHOP-like: 40%, 40% hyperCVAD: 100%, 75%

Auto: 2 Allo: 2

Belhadj21 N=21

CHOP/CHOP-like: CHOP/CHOP-like: 19 61%, 47% platinum-cyplatinum-cytarabinetarabine-based: based: 2 100%, 0%

Voss N=15

CHOP/CHOP-like: 4 ICE/IVAC: 8 other: 2

Yabe23 N=28

Tanase N=25 Foss26 N=31

†42

PFS^ mths

Auto: 2 Allo: 1

CHOP/CHOP-like: 75%, 50% ICE/IVAC: 75%, 38%

Auto: 4 Allo: 5

CHOP/CHOP-like: 9 hyperCVAD: 15 other: 19*

Auto: 7 Allo: 5

CHOP: 8 cytarabine/etoposide/ platinum-containing: 14 other: 3

Auto: 7 Allo: 18

13.3

28.3

Notes

No patients receiving CHOP/CHOP-like induction survived.

Only 2 patients survived, both treated with platinumcytarabine (alive at 42 and 52 mths). All other patients died.

6 of 7 surviving patients received non-CHOP induction and 5 of 7 surviving patients received allo.

28.3

Transplant (auto or allo) associated with longer OS (HR=0.3, 95% CI: 0.1-1.3) and EFS (HR=0.2, 95% CI: 0.1-0.9)

3-yr: 48% 3-yr: 54%

Anthracycline: 60% 40% CR (response by 8 (auto vs. allo non-anthracycline: 40% regimen not specified) not specified)

CHOP/CHOP-like: 52%, 38% CHOP/CHOP-like: 50 Klebaner‡37 cytarabine/etoposide/ cytarabine/etoposide/ N=84 platinum-containing: 34 platinum-containing: 82%, 56%

OS^ mths

Auto: 20 Allo: 15

5 of 7 patients receiving auto relapsed and died.

3-yr PFS and OS of 40%.

CHOP/CHOP-like: 18 mths. see notes Cytarabine/etoposide/platinum-containing: 36.5 mths.

^ Median unless specified. *First-line regimens not specified. Therefore, the total number of regimens reported is greater than the total number of patients. †European Society for Bone and Marrow Transplantation (EBMT) transplant database series (all patients received transplant). ‡Meta-analysis of published cases. Allo: allogeneic hematopoietic stem cell transplant; Auto: autologous hematopoietic stem cell transplant; CHOP: cyclophosphamide, doxorubicin, vincristine, prednisone; CI: confidence interval; CR: complete response; EFS: event-free survival; HR: hazard ratio; hyperCVAD: cyclophosphamide, vincristine, doxorubicin, and dexamethasone alternating with high-dose methotrexate and cytarabine; ICE: ifosphamide, carboplatin, etoposide; IVAC: ifosphamide, etoposide, cytarabine; mths: months; NR: not reported; ORR: objective response rate; OS: overall survival; PFS: progression-free survival; yr: year.

analysis suggests that cytarabine/etoposide/platinumcontaining induction regimens are potentially superior to CHOP-based approaches and should be prioritized in curative intent strategies. Our institutional preference is ifosfamide-containing regimens, such as ICE (ifosfamide, carboplatin, etoposide), usually for two or three cycles followed by response assessment with PET/CT evaluation and bone marrow biopsy. In patients who achieve CR or near CR, we proceed quickly to alloHCT, usually with an additional cycle of therapy to minimize any break in therapy. In our published experience of 14 patients with HSTCL, most of whom were treated with ICE or IVAC (ifosfamide, etoposide, cytarabine), we observed a median OS of 59 months, with six of seven surviving patients receiving non-CHOP induction, and five of seven surviving patients receiving alloHCT.38 Most National Comprehensive Cancer

Network (NCCN) centers use ICE as their initial approach although the guidelines39 also list DHAP(X) (dexamethasone, cytarabine, and a platinum), dose-adjusted EPOCH (etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin), hyper-CVAD (hyperfractionated cyclophosphamide, vincristine, doxorubicin, cytarabine),22,23 and IVAC (ifosfamide etoposide, cytarabine) as other recommended regimens. European guidelines from the European Society of Medical Oncology propose ICE, IVAC, and CHOEP (CHOP plus etoposide) as induction regimens for HSTCL.40 A graft-versus-lymphoma effect is likely a critical factor in long-term survival, as most patients will relapse in the absence of alloHCT. In the analysis of 166 published cases, 2-year OS was 12% for those who did not receive any transplant versus 56% for those receiving alloHCT (although the non-transplant group likely included non-re-

Haematologica | 108 December 2023

3248

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas sponders to therapy).37 In the largest report of alloHCT in HSTCL, estimated 3-year relapse-free survival and OS in 54 patients was 42% and 56%, respectively.41 The European Society for Bone and Marrow Transplantation made similar findings.42 While consolidation with autologous hematopoietic stem cell transplantation (autoHCT) can be considered,38 our preference, based on limited available data, is early transplant evaluation and alloHCT in eligible patients. In patients with R/R disease, we treat with alternative regimens in the absence of a clinical trial. Donor lymphocyte infusion in those who relapse after alloHCT may have effect.38

Intestinal T/NK-cell lymphomas Enteropathy-associated T-cell lymphoma The intestinal T-cell and NK-cell lymphomas are unified by their primary involvement of the gastrointestinal tract (Figure 3). The two most recognized entities, enteropathy-associated T-cell lymphoma (EATL, formerly referred to as type 1 EATL) and monomorphic epitheliotropic T-cell lymphoma (MEITL, formerly referred to as type II EATL) are now recognized as genomically distinct diseases.43-58 EATL is the most common intestinal T-cell lymphoma in Western countries, presenting with abdominal symptoms in individuals with a preceding or concomitant diagnosis of celiac disease. Direct visualization reveals destructive ulcerating lesions, or at times a frank mass, most commonly in the small bowel, although multifocal lesions involving other intestinal or extraintestinal sites are not uncommon.43-58 Biopsies show a pleomorphic population of medium/large-sized cells in an inflammatory background, with the neoplastic cells being most often CD4– /CD8– mature T cells, although CD4/CD8 expression is seen, frequently with 9q34 gains, 16q12 deletions, and JAK/STAT mutations.50,59-61 EATL is a recognized complication of celiac disease, although the pathogenesis is complex. EATL may be preceded by a condition known as refractory celiac disease, defined as persistent gastrointestinal symptoms and abnormal histological findings despite a strict gluten-free diet for ≥6–12 months. This etiological relationship has been reviewed by Dogan and colleagues and elsewhere (see excellent reviews and guidelines,62 recent genomic analysis and commentary,63,64 and clinical practice guidelines for refractory celiac disease from the American Gastroenterological Association65). The optimal management of EATL is undefined, and interpretation of primary literature is challenging due to the rarity of the condition, evolving classifications (previously EATL could refer to EATL, MEITL or other intestinal T-cell lymphomas), and sub-analyses of larger trials that include various histologies (Table 2). Still, essentially all reports describe an aggressive natural history with generally unsatisfactory

R. Stuver et al.

long-term outcomes. For example, in the ITCP, 3-year progression-free survival (PFS) and OS for EATL was 28% and 30%, respectively, with a median OS of 11 months.26 The best outcomes have been observed with combination chemotherapy followed by consolidative transplantation, although the optimal induction is unclear. In the ITCP, in which nearly all patients with EATL received upfront anthracycline-based therapy, the CR rate to first-line therapy was 30%.26 Intensified efforts have varied results. In a small series of ten patients treated with six cycles of CHOEP, the response rate was again 30%.48 However, in the Nordic NLG-T-01 trial, one of the largest prospective trials in T-cell lymphoma, CHOEP14 for six cycles (CHOP-14 in patients >60 years) followed by autoHCT in 21 patients with EATL resulted in a 5-year PFS and OS of 38% and 48%, respectively, and updated results showed a 10-year PFS and OS of 29%, highlighting potential for long-term survival with this program.66 Other studied options include a novel regimen consisting of two cycles of IVE (ifosfamide, etoposide, epirubicin) followed by two cycles of high-dose methotrexate and autoHCT.51 In a preliminary evaluation of this regimen in six patients, four achieved a CR and remained free of disease at over 1.5 years. A larger, retrospective analysis of a modified version of this regimen (later referred to as the Newcastle Regimen) tested one cycle of CHOP followed by three courses of IVE alternating with intermediate-dose methotrexate and autoHCT in 26 patients.55 Five-year PFS and OS rates were both 68%. In comparison to a historical control of 31 patients treated with anthracycline-base therapy, there was a trend to improved CR rate with IVE/methotrexate (65% vs. 42%; P=0.06) and death was less frequent (39% vs. 81%, P=0.001). A phase II trial of this approach in T-cell lymphomas (including 11 patients with EATL) reported comparatively shorter but still encouraging survival in patients with EATL, with a 1-year PFS and OS of 45%.67 Given the frequent expression of CD30, EATL was included in the ECHELON 2 study, but only three patients were enrolled. A separate phase II study, the EATL-001 trial, evaluated BV-CHP (brentuximab vedotin, cyclophosphamide, doxorubicin and prednisolone) followed by autoHCT in 14 patients with EATL.68 A CR was observed in 64%. Three patients had primary progressive disease, but in all others, no relapses occurred, with a 2-year PFS and OS of 63% and 68%, respectively. Taken together, intensified approaches with CHOEP or BV-CHP in CD30+ tumors followed by autoHCT would be our preferred upfront approaches. Treatment of R/R disease is empiric. Second-line chemotherapy can be attempted.57 A durable response to CD30-directed chimeric antigen receptor T-cell therapy has been described.69 We assess disease response in EATL (and MEITL, see below) with PET/CT, and we consult closely with colleagues in Radiology and Gastroenterology to determine the most appropriate imaging and surveillance modality if there is difficulty in fully visualizing involved bowel regions.

Haematologica | 108 December 2023

3249

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas

R. Stuver et al.

Figure 3. Intestinal T-cell lymphomas. Five subtypes of intestinal T-cell lymphomas are recognized in current classification schemas. Indolent disorders consist of indolent T-cell lymphoma of the gastrointestinal (GI) tract and indolent NK-cell lymphoproliferative disorder of the GI tract, and aggressive disorders consist of enteropathy-associated T-cell lymphoma and monomorphic epitheliotropic T-cell lymphoma. A fifth subtype, intestinal T-cell lymphoma, is undefined and used for intestinal T-cell lymphomas not meeting diagnostic criteria for the other subtypes. *Indolent T-cell lymphoma of the GI tract is named as such in the World Health Organization schema1 but referred to as indolent T-cell lymphoproliferative disorder of the GI tract in the International Consensus Classification.2 Figure created at BioRender.com. NOS: not otherwise specified; PTCL: peripheral Tcell lymphoma; EBV: Epstein-Barr virus; mts: mutations; RCD: refractory celiac disease; autoHCT: autologous hematopoietic stem cell transplantation; CHOEP: cyclophosphamide, doxorubicin, vincristine, etoposide, prednisone; BV-CHP: brentuximab vedotin, cyclophosphamide, doxorubicin, prednisolone; alts: alterations.

Monomorphic epitheliotropic T-cell lymphoma Like EATL, MEITL is a primary intestinal T-cell lymphoma, although there is no clear association with celiac disease and there may be a predilection in those of Asian descent.7073 As MEITL was previously referred to as type II EATL (and before that, often grouped with intestinal T-cell lymphomas), dedicated literature is very sparse. Still, genomic studies increasingly demonstrate that MEITL is distinct from

EATL.59 As the name implies, MEITL is monomorphic, usually positive for CD8 and CD56, and most commonly derived from intraepithelial γδ T cells. Cytogenetic analyses and mutational profiling are increasingly defining the genomic landscape of this disease.59,74,75 MEITL is aggressive, and while sporadic patients are included in clinical trials of patients with other peripheral Tcell lymphomas, there have been no dedicated treatment

Haematologica | 108 December 2023

3250

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas

R. Stuver et al.

Table 2. Selected series detailing outcomes in enteropathy-associated T-cell lymphoma.

Series

Induction

ORR, CR

Transplant

PFS

Notes All patients with CD. Prior to induction, 24-h NPO with oral antibiotics for “gut sterilization.” All patients received enteral feeding. Retrospective analysis. Significantly improved outcomes compared to historical controls receiving anthracyclinebased therapy.

Bishton51 N=6

IVE/HD MTX + BEAM auto

100%, 83%

Auto: 6 Allo: 0

Not reported, 2 relapses at 0.2 and 1.7 yr

2 relapsed and died, all others alive (1.8-4.3 yr post-transplant)

Sieniawski55 N=26

Newcastle (CHOP x 1, IVE/MTX + BEAM auto)

69%, 65%

Auto: 14 Allo: 0

5-yr: 52%

5-yr: 60%

Foss26 N=65

Anthracycline: 97% non-anthracycl.: 3%

30% CR

10 (auto vs. allo not specified)

7 mths; 3-yr: 28%

11 mths; 3-yr: 30%

Not specified whether EATL included other intestinal T-cell lymphomas.

Phillips67 N=11

Newcastle (CHOP x 1, IVE/MTX + BEAM auto)

55%, 55%

Auto: 13 Allo: 0

1-yr: 45%

Prospective phase II trial. Included other histologies, including suspected MEITL.

2-yr: 68%

Prospective phase II trial in only EATL. All patients CD30+ (≥10%). Aside from 3 primary refractory patients, no relapses occurred.

10-yr: 29%

Prospective evaluation. Not specified whether EATL included other intestinal T-cell lymphomas.

BV+CHP/etoposide/ HD MTX + BEAM auto

Sibon68 N=14

Relander N=21

79%, 64%

CHOEP-14 82%, 51% (CHOP-14 >60 yr) (overall trial + auto population)

1-yr: 45%

Auto: 11 Allo: 0

2-yr: 63%

Not reported by histology

10-yr: 29%

Allo: allogeneic hematopoietic stem cell transplant; anthracycl.: anthracycline; Auto: autologous hematopoietic stem cell transplant; BEAM: carmustine, etoposide, cytarabine, melphalan; BV-CHP: brentuximab vedotin, cyclophosphamide, doxorubicin, prednisone; CD: celiac disease; CHOEP: cyclophosphamide, doxorubicin, vincristine, etoposide, prednisone; CHOP: cyclophosphamide, doxorubicin, vincristine, prednisone; CR: complete response; EATL: enteropathy-associated T-cell lymphoma h: hours; HD: high dose; IVE/MTX: ifosphamide, etoposide, epirubicin, methotrexate; MEITL: monomorphic epitheliotropic T-cell lymphoma; mths: months; NPO: nothing by mouth; ORR: objective response rate; OS: overall survival; PFS: progression-free survival; yr: year.

studies and the optimal approach is unclear. Patients often present with abdominal symptoms secondary to intestinal lesions, including perforation requiring upfront emergency resection.71,72 Many patients may be unable to tolerate further therapy due to debilitation. For example, in a multinational report from Asia of 38 patients, 26% did not receive any systemic therapy (all died). In this series, the median OS was 7 months, with most patients receiving anthracyclinebased therapy.70 In a slightly larger series of 42 patients there was an improved median OS of 14.8 months.71 More patients in this series received chemotherapy (88%), which was predominantly CHOP (71%), with a CR rate to first-line therapy of 38%. Ability to receive chemotherapy, response to first-line therapy, and receipt of autoHCT were all significantly associated with improved OS in univariate analyses. A non-significant increase in OS was seen in seven patients who received non-CHOP induction including CHOEP, EPOCH and non-anthracycline based regimens. In the largest series to date of 71 patients, the median OS was 7.8 months and 2-year OS was only 15%.75 High frequencies of alterations

and mutations in MYC, SETD2, STAT5B, and JAK3 were observed, as well as TP53 mutations in over one-third of patients. Eight patients survived beyond 2 years; all underwent surgery and six subsequently received chemotherapy. We would prioritize clinical trial enrollment in all eligible patients. In the absence of a clinical trial, we most often treat MEITL as nodal PTCL with CHOEP and plan consolidation of CR with autoHCT. However, due to the scarcity of data and the apparent higher rates of chemorefractory disease, we often restage early and change course with alternate therapy and planned alloHCT if the response is inadequate. Efforts to capitalize on aberrancies in the JAK/STAT pathway are worth considering in this aggressive disease and warrant further study. Intestinal T-cell lymphoma, not otherwise specified Intestinal T-cell lymphoma, NOS, introduced in the WHOHAEM4R and listed in both the WHO1 and ICC2 classification systems, is a non-descript category for T-cell lymphomas of the intestines that do not conform to diagnostic criteria

Haematologica | 108 December 2023

3251

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas for EATL or MEITL. Whether this is a unique entity or simply represents PTCL-NOS with intestinal involvement is unclear.76 Comprehensive immunophenotyping and mutational profiling, if available, are recommended. We would most likely treat intestinal T-cell lymphoma, NOS, similarly to EATL or PTCL-NOS once confidently distinguished from the indolent processes described below. Indolent T-cell and NK-cell disorders of the gastrointestinal tract Indolent T-cell lymphoma of the gastrointestinal tract (referred to as indolent clonal T-cell lymphoproliferative disorder of the gastrointestinal tract in the ICC) is a clonal process marked by a chronic natural history generally without progressive, disseminated disease. Only case reports and cases series exist.77-88 Presenting symptoms include dyspepsia, vomiting, and diarrhea.86 Disease has mostly been documented in the small intestine, but involvement of the oral cavity, esophagus, and large bowel has also been reported.86 While mesenteric adenopathy can be seen, frank peripheral adenopathy or extraintestinal involvement is uncommon.86 Macroscopically, intestinal polyps may be seen, and microscopically, a non-destructive, superficial lymphoid infiltrate of mature, αβ T cells with a low proliferation rate is observed.86 Recently, recurrent STAT3-JAK2 fusions and additional mutational events resulting in JAK-STAT activation have been identified.87,88 The optimal management of indolent T-cell lymphoma of the gastrointestinal tract is unknown and has been non-uniform in the literature, but options include observation, resection, budesonide, mesalamine, interferon, and combination chemotherapy.86,88 Three tentative conclusions can be made: (i) OS is long regardless of management, including with observation, with most patients in available reports being alive, some beyond 20 years; (ii) the natural history appears to be a locally chronic disease without high risk of aggressive transformation or spread; and (iii) no therapy appears reliably curative. Based on all available literature, if symptoms need treatment, we would favor conservative therapies at first and avoid the use of chemotherapy. The related indolent NK-cell lymphoproliferative disorder of the gastrointestinal tract is a new entry in the WHO classification and also designated in the ICC, and was previously referred to as lymphomatoid gastropathy89 or NKcell enteropathy.90 This entity was first documented in a Japanese case series in 2010, describing ten cases of a self-limited, EBV-negative NK-cell proliferation in the stomach.89 Patients were middle-aged with an equal sex distribution. Three cases were discovered upon follow-up of prior gastric malignancy, and the others were discovered via gastric cancer screening. No patients had extragastric disease, and all patients were observed and alive at the time of publication. Only three patients had

R. Stuver et al.

recurrent disease after initial resection, and in each case recurrent lesions self-resolved without intervention. Soon after this report, an additional eight cases from the USA were published, again describing an EBV-negative NK-cell proliferation confined to the gastrointestinal tract.90 In this series, most patients presented with vague abdominal complaints. Again, no extraintestinal involvement was detected. All patients were observed endoscopically and none developed progression. No patients died. Recurrent somatic JAK3 mutations were recently identified in some cases in a series led by our center.91 This series and others92-94 all reiterate that the condition has a chronic, at times spontaneously remitting natural history for which observation may be appropriate. As in indolent T-cell lymphoma of the gastrointestinal tract, we perform full staging, including endoscopic bowel evaluation, peripheral blood flow cytometry, bone marrow assessment, and mutational profiling, with a dedicated effort to rule out PTCL or ENKL, which require treatment (see below). Close observation is recommended with endoscopic follow-up in the absence of concerning clinical or pathological features.

Extranodal NK/T-cell lymphoma ENKL are predominantly extranodal diseases with a predilection for the upper aerodigestive tract, as first noted by Ng and colleagues over 30 years ago, recognizing a pattern of cases of sino-nasal lymphoma showing NK-cell markers.95 ENKL involving the upper aerodigestive tract has traditionally been called nasal type ENKL, whereas non-nasal ENKL or more disseminated disease with an unknown site of origin is sometimes called extranasal ENKL. The qualifier ‘nasal type’ has been retained by the ICC2 but not the WHO.1 The frequency of this disease is higher in Asian populations.96 A common presentation for nasal type ENKL is a rapidly growing, destructive mass involving any number of facial structures, including the nasopharynx or palate, with frequent invasion of local structures such as skin, sinuses, or orbits. Systemic symptoms and hemophagocytosis may be present at diagnosis. Key histopathological findings include a diffuse, aggressive lymphocytic infiltrate displaying NK-cell or Tcell markers along with universal EBER positivity. We evaluate patients with newly diagnosed, nasal type ENKL with direct visualization of the nasopharynx, CT or magnetic resonance imaging of nasal structures, PET/CT, laboratory studies including quantitative assessment of EBV DNA by polymerase chain reaction,97 and bone marrow biopsy to confirm stage. Given our nearly universal use of radiation therapy (RT) as an adjuvant or consolidation for patients with early-stage disease, we routinely consult Radiation Oncology colleagues during the initial workup

Haematologica | 108 December 2023

3252

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas for assessment of pretreatment disease volumes to facilitate planning of post-chemotherapy radiation. While we evaluate prognosis with the prognostic index of NK lymphoma plus EBV DNA scoring system (PINK-E; risk factors: age >60 years, stage III/IV, distant [non-regional] lymph node involvement, non-nasal subtype, detectable plasma EBV-DNA at diagnosis),98 our treatment decisions are primarily driven by stage of disease and the patient’s fitness, as described below. Our treatment approach is broadly based on the stage and extent of disease, the patient’s comorbidities, candidacy for HCT, and goals of care. A number of overarching principles shape our management, including recognition of: (i) the importance of RT for local disease control; (ii) the additive benefits of systemic therapy in addition to RT, primarily with pegaspargase-based combination chemotherapy regimens to overcome inherent chemoresistance and address micrometastatic disease; (iii) the ability to consolidate response with either autoHCT or alloHCT; and (iv) the promising activity of PD-1 and PD-L1 targeting monoclonal antibodies in R/R ENKL. For fit patients with stage I/II nasal ENKL, we typically use combined modality therapy that incorporates local RT directed at the main sites of disease plus systemic therapy. While early retrospective analyses were conflicting regarding the benefit of systemic therapy in addition to RT,99,100 later population-level data from both the International Peripheral T-Cell Lymphoma Project and the China Lymphoma Collaborative Group showed a clear survival benefit when systemic therapy is added to RT.96,101 The conflicting data from earlier reports may stem from the use of CHOP or CHOP-like chemotherapy regimens that are now recognized as suboptimal for treating ENKL. The benefits of systemic therapy in addition to RT in stage I/II ENKL were recently documented in a small, multicenter randomized control trial in China, in which 65 patients with stage I/II ENKL were randomized to RT alone versus sequential asparaginase-based chemotherapy followed by RT.102 The response rate (83.3% vs. 60.0%), 5-year PFS (82.9% vs. 56.5%), and 5-year OS (85.7% vs. 60.4%) were significantly improved with the addition of chemotherapy to RT. As expected, combined modality therapy resulted in statistically significantly increased myelosuppression, although grade III/IV events were uncommon overall. For patients ineligible for multiagent chemotherapy with stage I/II nasal ENKL, we typically use RT alone (generally at doses of 50 to 55 Gy) or combine RT with less intensive chemotherapy. There is uncertainty regarding the optimal sequencing and type of therapy in this paradigm, and supportive data exist for sequential chemoradiation,102-104 sandwich chemoradiation,105-107 and concurrent chemoradiation (typically followed by additional chemotherapy alone).108-113 As RT alone is highly effective (especially for stage I disease)114 and

R. Stuver et al.

concurrent chemotherapy generally compromises the choice and intensity of chemotherapy as well as adding toxicity, we typically favor sequential approaches instead, which can avoid the additive toxicity of concurrent chemoradiation and still enable early incorporation of effective systemic therapy. In our institutional practice, for fit patients, we favor one to two cycles of the modified SMILE regimen (dexamethasone, methotrexate, ifosphamide, pegaspargase, etoposide) (see full regimen descriptions in Table 3) for systemic therapy followed by RT.103 In patients with early-stage and localized nasal ENKL who attain a CR after combined modality therapy, we generally observe rather than consolidate with transplantation. Table 3 details key studies and results for different therapeutic strategies in early-stage ENKL. Dosing, radiation strategies, and even staging vary between published reports, and clinicians should consult the primary literature for specific treatment details of any particular treatment program. For patients with advanced-stage ENKL, we use combination chemotherapy and often consider combined modality therapy in those with a symptomatic, invasive mass in the upper aerodigestive tract given the effectiveness of RT and the morbidity associated with local treatment failure. In advanced-stage disease, systemic therapy regimens all integrate L-asparaginase or pegaspargase (Table 3) and include modified SMILE,103,115,116 P-GemOx (pegaspargase, gemcitabine, oxaliplatin),117 DDGP (cisplatin, dexamethasone, gemcitabine, pegaspargase),118 and AspaMetDex (Lasparaginase, methotrexate, dexamethasone).119 At our institution, we use modified SMILE in most fit patients due to institutional experience with the regimen and its known long-term efficacy.101 Still, we acknowledge the efficacy of other asparaginase-containing regimens listed above, in particular those containing gemcitabine, which show largely comparable results with potentially less toxicity, but for which there are noticeably fewer data and less long-term follow-up in comparison to those for the SMILE regimen. Outside of China, these regimens have not been compared directly, and the most critical principle in practice is to use asparaginase-based induction and not an anthracycline-containing regimen. Of note, a randomized trial of SMILE versus DDGP was conducted in China in patients with newly diagnosed, stage III/IV ENKL.120 Patients treated with DDGP achieved improved PFS (median not reached vs. 6.8 months, HR=0.42, 95% CI: 0.23-0.77) and OS (median not reached vs. 75.2 months, HR=0.41, 95% CI: 0.19-0.89) compared to those treated with SMILE. The ORR favored the DDGP arm (90.0% vs. 60.0%, P=0.067). However, there are important caveats regarding this trial that limit its routine use in clinical practice. First, the trial applied an alternative staging system that is not routinely used in North America (called the Chinese Southwest Oncology Group and Asia

Haematologica | 108 December 2023

3253

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas Lymphoma Study Group system).121 In this system, stage III disease is any lesion with regional lymph node involvement, which would typically be considered stage II disease in the Lugano Modification of the Ann Arbor Staging System. More patients with stage III disease in this system

R. Stuver et al.

(vs. stage IV) were in the DDGP arm, which may indicate that the DDGP arm was enriched with patients who would otherwise be considered to have early-stage disease (and likely more favorable prognoses). In addition, the SMILE regimen was significantly modified and resulted in exces-

Table 3. Selected series in untreated extranodal NK/T-cell lymphoma.

Setting

Regimen

Transplant information†

Administration

Cisplatin-RT followed Weekly cisplatin x 4 by VIPD108 with RT followed by VIPD x 3 Early stage: concurrent chemoradiation

Early stage: sequential chemoradiation

Early stage: sandwich chemoradiation

Not stated

3-yr PFS: 86% 3-yr OS: 85%

Notes Cisplatin-RT has been followed by other regimens (see reference 114). DVIP dosing differs from VIPD, (consult reference 112). 2/3 dosing of DeVIC (2/3DeVIC) was primarily used due to toxicity. 94% ORR (89% CR) in stage I-II disease. Randomized trial comparing DDGP-RT versus RT alone.

DEP-RT followed by DVIP112

DEP x 2 with RT followed by DVIP x 2

Not stated

5-yr PFS: 60% 5-yr OS: 66%

DeVIC-RT110

DeVIC x 3 with RT

Auto: 0 (0%) Allo: 0 (0%)

5-yr PFS: 67% 5-yr OS: 73%

mSMILE followed mSMILE x 1-2 cycles by RT103 followed by RT

Auto: 1 (6%) Allo: 0 (0%)

median PFS/OS: not reached

DDGP followed by RT102

Not stated

5-yr PFS: 83% 5-yr OS: 86%%

Auto: 0 (0%) Allo: 0 (0%)

5-yr PFS: 66% 5-y OS: 81%

Not stated

5-yr PFS: 74% 5-yr OS: 85%

Not stated

5-yr PFS: 64% 5-yr OS: 64%

DDGP x 3 cycles followed by RT

P-GemOx x 2-6 cyP-GemOx followed cles (not defined) folby RT104 lowed by RT GeLOx x ≥ 2 cycles followed by RT folloGeLOx-RT-GeLOx105 wed by GeLOx x 2-4 cycles (6 total cycles) LVP-RT-LVP106

Advanced stage: combination chemotherapy^

Survival

LVP x 2 cycles followed by RT followed by LVP x 2-4 cycles

mSMILE103

mSMILE x 3-4 cycles ± RT

Auto: 3 (30%) Allo: 4 (40%)

DDGP118

DDGP x 4-6 cycles

Not stated

P-GemOx

P-GemOx x 6-8 cycles (usually followed by RT)

Auto: 7 (20%)* Allo: 0 (0%)

117

AspaMetDex119

AspaMetDex x 3 cycles

Auto: 5 (26%) Allo: 0 (0%)

†

Key differences in mSMILE and median PFS: 8 mths SMILE116,117 are median OS: 11 mths summarized in reference 104. See text for DDGP 3-yr PFS: 62% versus SMILE 3-yr OS: 75% comments. Reference includes 3-yr PFS: 39% newly-diagnosed 3-yr OS: 65% advanced and R/R. See notes

Trial in R/R ENKL regardless of stage, consult reference for detailed outcomes.

This column refers to the number (and percentage) of patients in each series who underwent transplant consolidation in remission following the stated regimen listed under the Administration column. *Publication does not state whether these 7 patients had newly diagnosed or R/R disease. allo: allogeneic hematopoietic stem cell transplant; AspaMetDex: L-asparaginase, methotrexate, dexamethasone; auto: autologous stem cell transplantation (or high-dose therapy with autologous stem cell rescue); DDGP: cisplatin, dexamethasone, gemcitabine, pegaspargase; DEP: dexamethasone, etoposide, cisplatin; DeVIC, dexamethasone, etoposide, ifosphamide, carboplatin; DVIP: dexamethasone, etoposide, ifosphamide, cisplatin; ENKL: extranodal NK/T-cell lymphoma; GeLOx: gemcitabine, L-asparaginase, oxaliplatin; LVP: L-asparaginase, vincristine, prednisone; mths: months; mSMILE: dexamethasone, methotrexate, ifosphamide, pegaspargase, etoposide; P-GemOx: pegaspargase, gemcitabine, oxaliplatin; R/R: relapsed/refractory; RT: radiation therapy; SMILE: dexamethasone, methotrexate, ifosphamide, L-asparaginase, etoposide; VIPD: etoposide, ifosphamide, cisplatin, dexamethasone; yr: year. Haematologica | 108 December 2023

3254

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas sive toxicity including seven treatment-related deaths (17.5% of patients treated with SMILE) and lower efficacy than in other published studies. In our own institutional experience of using the modified SMILE regimen, we did not document any treatment-related deaths,103 and we are hesitant to extrapolate the results of this Chinese trial given the non-conventional staging system used and the unusually high treatment-related mortality in the SMILE arm. We strongly consider consolidative HCT in patients with advanced-stage disease who attain complete remission, although the decision whether to perform autoHCT or alloHCT is highly individualized based on disease presentation, chemosensitivity, treatment tolerability, patients’ fitness, and donor availability. Our practice is to consider alloHCT when possible, especially for patients with advanced-stage disease involving multiple extranodal sites, although there is no clear consensus. If the goal is alloHCT, there is no minimum number of cycles of therapy that must be performed (we proceed as fast as possible once remission has been obtained). In terms of autoHCT, a recently reported retrospective study of 20 patients who underwent upfront autoHCT who were compared to a matched, 60-patient control group who were observed suggested improved PFS and OS with autoHCT.123 There are other reports documenting the efficacy of autoHCT.123,124 In terms of alloHCT, the Center for International Blood and Marrow Transplant Research used pooled data to evaluate 82 patients with ENKL who underwent alloHCT (30 patients in first CR).125 Of note, 22% in this series were refractory to frontline therapy and only 38% received pegaspargase prior to alloHCT. With a median follow-up of 36 months, the 3-year PFS and OS was 28% and 34%, respectively. Non-relapse mortality at 3 years was 30%. The Asia Lymphoma Study Group reported outcomes for 18 patients undergoing alloHCT (9 patients in first CR), demonstrating 5-year event-free survival and OS of 51%

R. Stuver et al.

and 57%, respectively.126 For patients with prolonged followup in both series, survival plateaus suggest curative potential. Driven by EBV, PD-L1 is frequently expressed on NK/T cells as a mechanism to avert immune surveillance. For R/R ENKL, recent results on the use of anti-PD-1 and anti-PDL1 monoclonal antibodies are encouraging (Table 4).127-133 These agents are the preferred therapies in the R/R setting, if accessible.39 Table 4 details outcomes of the various agents. None is approved in the USA and Europe, and only pembrolizumab and nivolumab are compendium-listed by the NCCN guidelines (to the best of our knowledge, none of these agents is approved globally at the time of publication).39 While pseudo-progression has been reported with these agents, hyperprogression akin to that reported in adult T-cell leukemia-lymphoma134 has not been consistently observed in ENKL. We await final results from larger series with longer follow-up in the R/R setting to assess response durability as well as the use of checkpoint blockade prior to alloHCT, given concerns regarding graft-versus-host disease. Ongoing studies are evaluating the incorporation of checkpoint blockade in first-line regimens (NCT03728972). Finally, additional EBV-directed therapies for R/R ENKL are under investigation, including nanatinostat, a selective histone deacetylase inhibitor, in combination with valganciclovir,135 and autologous EBV-specific T cells.136

A word on non-cytotoxic therapies The last decade has witnessed numerous attempts to utilize non-cytotoxic therapies in T-cell lymphomas, primarily in the R/R setting. These agents include brentuximab vedotin,137,138 belinostat,139 romidepsin,140,141 pralatrexate,142 bortezomib,143,144 lenalidomide,145-147 duvelisib,148 and 149 ruxolitinib, some of which are approved for use in the

Table 4. Anti-PD-1 and PD-L1 therapies in relapsed/refractory extranodal NK/T-cell lymphoma.

Agent Nivolumab128 N=3, case series

ORR (CR) Median PFS Median OS DOR Of 3 patients with R/R disease, 1 achieved CR and remained free of disease at publication. The other 2 patients exhibited clinical response in observable lesions but died of infection within 3 mths.

Pembrolizumab127 N=7, case series

Of 7 patients with R/R disease, 4 achieved a response (ORR 57%) with 2 CR (29%). No survival data provided. Not reported, 5 of 21 pa38% (24%) 2.7 mths Not reached tients with ongoing treatment beyond 12 mths Not reached 75% (21%) Not reported 4.1 mths (2-yr OS: 79%)

Avelumab131 N=21 Sintilimab133 N=28 Tislelizumab132 N=22 Sugemalimab129 N=80

32% (18%)

2.7 mths

9 mths

Not reached

45% (36%)

Not reported

Not reached (18-mth OS: 58%)

Not reached (18-mth DOR rate: 83%)

CR: complete response; DOR: duration of response; mths: months; ORR: objective response rate; OS: overall survival; PFS: progression free survival; R/R: relapsed/refractory; yr: year. Haematologica | 108 December 2023

3255

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas USA and Europe, and others which can be variably used off-label. The referenced clinical trials of these agents are dominated by the more common nodal T-cell lymphomas (PTCL-NOS, angioimmunoblastic T-cell lymphoma, and anaplastic large cell lymphoma), and only a handful of patients with the rare subtypes discussed in this review were included in these trials. Therefore, when referencing response rates and survival times following the use of these agents, it is imperative to realize that the true efficacy of many novel therapies remains largely unclear in the rare subtypes. An increasing reliance on molecular drivers of response and resistance may allow for a more nuanced application of therapies, as histology-specific clinical trials are challenging to conduct (although not insurmountable68) given the sheer rarity of these entities.

R. Stuver et al.

management decisions. Thanks to tremendous basic science and translational reaearch, as well as multicenter, often international clinical efforts in the dedicated study of T-cell lymphomas, we are hopeful that the next decade can provide further insights into these rare entities and refine treatment paradigms based on greater degrees of evidence.

Disclosures There are no disclosures related to the present manuscript. Outside of this work, RS and ZDEP have no disclosures. SMH has received research support from ADC Therapeutics, Affimed, Aileron, Celgene, Crispr Therapeutics, Daiichi Sankyo, Forty Seven, Inc., Kyowa Hakko Kirin, Millennium /Takeda, Seattle Genetics, Trillium Therapeutics, and Verastem/SecuraBio. He has also receives consultancy fees from Acrotech Biopharma, ADC Therapeutics, Astex, Auxilus Pharma, Merck, C4 Therapeutics, Celgene, Cimieo TheraConclusions peutics, Daiichi Sankyo, Janssen, Kura Oncology, Kyowa Treating rare diseases is a challenging endeavor. In- Hakko Kirin, Myeloid Therapeutics, ONO Pharmaceuticals, Seattle Genetics, SecuraBio, Shoreline Biosciences, Inc, Taformative clinical trials are lacking, and management keda, Trillium Therapeutics, Tubulis, Verastem/SecuraBio, decisions are consequently derived from imperfect data. Vividion Therapeutics and Yingli Pharma Limited. Still, patients with rare diseases need treatment, and as clinicians we must therefore thoughtfully review prior literature, searching for themes and guiding principles that, Contributions when coupled with knowledge on underlying biology and All authors wrote the paper and approved the final version our own personal experiences, can allow for reasonable of the manuscript.

References 1. Alaggio R, Amador C, Anagnostopoulos I, et al. The 5th edition of the World Health Organization classification of haematolymphoid tumours: lymphoid neoplasms. Leukemia. 2022;36(7):1720-1748. 2. Campo E, Jaffe ES, Cook JR, et al. The International Consensus Classification of Mature Lymphoid Neoplasms: a report from the Clinical Advisory Committee. Blood. 2022;140(11):1229-1253. 3. Howlander N, Noone A, Krapcho M, Garshell J, Miller D. SEER Cancer Statistics Review, 1975-2011. Bethesda, MD. 4. Khan N, Noor SJ, Horwitz S. How we treat mycosis fungoides and Sézary syndrome. Clin Adv Hematol Oncol. 2021;19(9):573-581. 5. David KA, Pulitzer M, Guitart J, et al. Characteristics, treatment patterns, and outcomes in primary cutaneous gamma delta T cell lymphoma (PCGDTCL): a real world multi-institutional analysis of a rare malignancy. Blood. 2019;134(Suppl_1):4028. 6. Willemze R, Jansen PM, Cerroni L, et al. Subcutaneous panniculitis-like T-cell lymphoma: definition, classification, and prognostic factors: an EORTC Cutaneous Lymphoma Group study of 83 cases. Blood. 2008;111(2):838-845. 7. Guitart J, Mangold AR, Martinez-Escala ME, et al. Clinical and pathological characteristics and outcomes among patients with subcutaneous panniculitis-like T-cell lymphoma and related adipotropic lymphoproliferative disorders. JAMA Dermatol. 2022;158(10):1167-1174. 8. Rangel LK, Villa-Ruiz C, Lo K, et al. Clinical characteristics of lupus erythematosus panniculitis/profundus: a retrospective review of 61 patients. JAMA Dermatol. 2020;156(11):1264-1266. 9. Gayden T, Sepulveda FE, Khuong-Quang DA, et al. Germline

HAVCR2 mutations altering TIM-3 characterize subcutaneous panniculitis-like T cell lymphomas with hemophagocytic lymphohistiocytic syndrome. Nat Genet. 2018;50(12):1650-1657. 10. Polprasert C, Takeuchi Y, Kakiuchi N, et al. Frequent germline mutations of HAVCR2 in sporadic subcutaneous panniculitis-like T-cell lymphoma. Blood Adv. 2019;3(4):588-595. 11. Sonigo G, Battistella M, Beylot-Barry M, et al. HAVCR2 mutations are associated with severe hemophagocytic syndrome in subcutaneous panniculitis-like T-cell lymphoma. Blood. 2020;135(13):1058-1061. 12. Koh J, Jang I, Mun S, et al. Genetic profiles of subcutaneous panniculitis-like T-cell lymphoma and clinicopathological impact of HAVCR2 mutations. Blood Adv. 2021;5(20):3919-3930. 13. Guenova E, Schanz S, Hoetzenecker W, et al. Systemic corticosteroids for subcutaneous panniculitis-like T-cell lymphoma. Br J Dermatol. 2014;171(4):891-894. 14. Michonneau D, Petrella T, Ortonne N, et al. Subcutaneous panniculitis-like T-cell lymphoma: immunosuppressive drugs induce better response than polychemotherapy. Acta Derm Venereol. 2017;97(3):358-364. 15. López-Lerma I, Peñate Y, Gallardo F, et al. Subcutaneous panniculitis-like T-cell lymphoma: clinical features, therapeutic approach, and outcome in a case series of 16 patients. J Am Acad Dermatol. 2018;79(5):892-898. 16. Gibson JF, Alpdogan O, Subtil A, et al. Hematopoietic stem cell transplantation for primary cutaneous γδ T-cell lymphoma and refractory subcutaneous panniculitis-like T-cell lymphoma. J Am

Haematologica | 108 December 2023

3256

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas Acad Dermatol. 2015;72(6):1010-1015.e5. 17. Rosée P la, Horne AC, Hines M, et al. Recommendations for the management of hemophagocytic lymphohistiocytosis in adults. Blood. 2019;133(23):2465-2477. 18. Zhang Q, Zhao YZ, Ma HH, et al. A study of ruxolitinib responsebased stratified treatment for pediatric hemophagocytic lymphohistiocytosis. Blood. 2022;139(24):3493-3504. 19. Lévy R, Fusaro M, Guerin F, et al. Efficacy of ruxolitinib in subcutaneous panniculitis-like T-cell lymphoma and hemophagocytic lymphohistiocytosis. Blood Adv. 2020;4(7):1383. 20. Wehkamp U, Mitteldorf C, Stendel S, et al. Most rare subtypes of cutaneous lymphoma display variable CD30 expression: analysis of the German Cutaneous Lymphoma Network. Br J Dermatol. 2021;185(1):228-230. 21. Belhadj K, Reyes F, Farcet JP, et al. Hepatosplenic gammadelta Tcell lymphoma is a rare clinicopathologic entity with poor outcome: report on a series of 21 patients. Blood. 2003;102(13):4261-4269. 22. Falchook GS, Vega F, Dang NH, et al. Hepatosplenic gamma-delta T-cell lymphoma: clinicopathological features and treatment. Ann Oncol. 2009;20(6):1080-1085. 23. Yabe M, Medeiros LJ, Tang G, et al. Prognostic factors of hepatosplenic T-cell lymphoma: clinicopathologic study of 28 cases. Am J Surg Pathol. 2016;40(5):676-688. 24. Mackey AC, Green L, Liang LC, Dinndorf P, Avigan M. Hepatosplenic T cell lymphoma associated with infliximab use in young patients treated for inflammatory bowel disease. J Pediatr Gastroenterol Nutr. 2007;44(2):265-267. 25. Deepak P, Sifuentes H, Sherid M, et al. T-cell non-Hodgkinʼs lymphomas reported to the FDA AERS with tumor necrosis factoralpha (TNF-α) inhibitors: results of the REFURBISH study. Am J Gastroenterol. 2013;108(1):99-105. 26. Foss FM, Horwitz SM, Civallero M, et al. Incidence and outcomes of rare T cell lymphomas from the T Cell Project: hepatosplenic, enteropathy associated and peripheral gamma delta T cell lymphomas. Am J Hematol. 2020;95(2):151-155. 27. Wang CC, Tien HF, Lin MT, et al. Consistent presence of isochromosome 7q in hepatosplenic T gamma/delta lymphoma: a new cytogenetic-clinicopathologic entity. Genes Chromosomes Cancer. 1995;12(3):161-164. 28. Yao M, Tien HF, Lin MT, et al. Clinical and hematological characteristics of hepatosplenic T gamma/delta lymphoma with isochromosome for long arm of chromosome 7. Leuk Lymphoma. 1996;22(5-6):495-500. 29. Jonveaux P, Daniel MT, Martel V, Maarek O, Berger R. Isochromosome 7q and trisomy 8 are consistent primary, nonrandom chromosomal abnormalities associated with hepatosplenic T γ/δ lymphoma. Leukemia. 1996;10(9):1453-1455. 30. Alonsozana ELC, Stamberg J, Kumar D, et al. Isochromosome 7q: the primary cytogenetic abnormality in hepatosplenic gammadelta T cell lymphoma. Leukemia. 1997;11(8):1367-1372. 31. Wlodarska I, Martin-Garcia N, Achten R, et al. Fluorescence in situ hybridization study of chromosome 7 aberrations in hepatosplenic T-cell lymphoma: isochromosome 7q as a common abnormality accumulating in forms with features of cytologic progression. Genes Chromosomes Cancer. 2002;33(3):243-251. 32. Travert M, Huang Y, de Leval L, et al. Molecular features of hepatosplenic T-cell lymphoma unravels potential novel therapeutic targets. Blood. 2012;119(24):5795-5806. 33. Ferreiro JF, Rouhigharabaei L, Urbankova H, et al. Integrative genomic and transcriptomic analysis identified candidate genes implicated in the pathogenesis of hepatosplenic T-cell lymphoma. PLoS One. 2014;9(7):e102977. 34. Nicolae A, Xi L, Pittaluga S, et al. Frequent STAT5B mutations in γδ

R. Stuver et al.

hepatosplenic T-cell lymphomas. Leukemia. 2014;28(11):2244-2248. 35. McKinney M, Moffitt AB, Gaulard P, et al. The genetic basis of hepatosplenic T-cell lymphoma. Cancer Discov. 2017;7(4):369-379. 36. Yabe M, Medeiros LJ, Wang SA, et al. Distinguishing between hepatosplenic T-cell lymphoma and γδ T-cell large granular lymphocytic leukemia: a clinicopathologic, immunophenotypic, and molecular analysis. Am J Surg Pathol. 2017;41(1):82-93. 37. Klebaner D, Koura D, Tzachanis D, et al. Intensive induction therapy compared with CHOP for hepatosplenic T-cell lymphoma. Clin Lymphoma Myeloma Leuk. 2020;20(7):437. 38. Voss MH, Lunning MA, Maragulia JC, et al. Intensive induction chemotherapy followed by early high-dose therapy and hematopoietic stem cell transplantation results in improved outcome for patients with hepatosplenic T-cell lymphoma: a single institution experience. Clin Lymphoma Myeloma Leuk. 2013;13(1):8-14. 39. Horwitz SM, Ansell S, Ai WZ, et al. T-cell lymphomas, version 2.2022, NCCN clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2022;20(3):285-308. 40. d’Amore F, Gaulard P, Trümper L, et al. Peripheral T-cell lymphomas: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2015;26(Suppl 5):v108-v115. 41. Rashidi A, Cashen AF. Outcomes of allogeneic stem cell transplantation in hepatosplenic T-cell lymphoma. Blood Cancer J. 2015;5(6):e318. 42. Tanase A, Schmitz N, Stein H, et al. Allogeneic and autologous stem cell transplantation for hepatosplenic T-cell lymphoma: a retrospective study of the EBMT Lymphoma Working Party. Leukemia. 2015;29(3):686-688. 43. Tutt ANJ, Brada M, Sampson SA. Enteropathy associated T cell lymphoma presenting as an isolated CNS lymphoma three years after diagnosis of coeliac disease: T cell receptor polymerase chain reaction studies failed to show the original enteropathy to be a clonal disorder. Gut. 1997;40(6):801-803. 44. Gale J, Simmonds PD, Mead GM, Sweetenham JW, Wright DH. Enteropathy-type intestinal T-cell lymphoma: clinical features and treatment of 31 patients in a single center. J Clin Oncol. 2000;18(4):795-803. 45. Daum S, Ullrich R, Heise W, et al. Intestinal non-Hodgkin’s lymphoma: a multicenter prospective clinical study from the German Study Group on Intestinal non-Hodgkin’s Lymphoma. J Clin Oncol. 2003;21(14):2740-2746. 46. Jantunen E, Juvonen E, Wiklund T, Putkonen M, Nousiainen T. High-dose therapy supported by autologous stem cell transplantation in patients with enteropathy-associated T-cell lymphoma. Leuk Lymphoma. 2003;44(12):2163-2164. 47. Gobbi C, Buess M, Probst A, et al. Enteropathy-associated T-cell lymphoma with initial manifestation in the CNS. Neurology. 2003;60(10):1718-1719. 48. Wöhrer S, Chott A, Drach J, et al. Chemotherapy with cyclophosphamide, doxorubicin, etoposide, vincristine and prednisone (CHOEP) is not effective in patients with enteropathytype intestinal T-cell lymphoma. Ann Oncol. 2004;15(11):1680-1683. 49. Novakovic BJ, Novakovic S, Frkovic-Grazio S. A single-center report on clinical features and treatment response in patients with intestinal T cell non-Hodgkin’s lymphomas. Oncol Rep. 2006;16(1):191-195. 50. deLeeuw RJ, Zettl A, Klinker E, et al. Whole-genome analysis and HLA genotyping of enteropathy-type T-cell lymphoma reveals 2 distinct lymphoma subtypes. Gastroenterology. 2007;132(5):1902-1911. 51. Bishton MJ, Haynes AP. Combination chemotherapy followed by autologous stem cell transplant for enteropathy-associated T cell

Haematologica | 108 December 2023

3257

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas lymphoma. Br J Haematol. 2007;136(1):111-113. 52. Al-toma A, Verbeek WHM, Visser OJ, et al. Disappointing outcome of autologous stem cell transplantation for enteropathyassociated T-cell lymphoma. Dig Liver Dis. 2007;39(7):634-641. 53. Al-toma A, Verbeek WHM, Hadithi M, von Blomberg BME, Mulder CJJ. Survival in refractory coeliac disease and enteropathyassociated T-cell lymphoma: retrospective evaluation of single-centre experience. Gut. 2007;56(10):1373-1378. 54. Verbeek WHM, van de Water JMW, Al-Toma A, et al. Incidence of enteropathy--associated T-cell lymphoma: a nation-wide study of a population-based registry in The Netherlands. Scand J Gastroenterol. 2008;43(11):1322-1328. 55. Sieniawski M, Angamuthu N, Boyd K, et al. Evaluation of enteropathy-associated T-cell lymphoma comparing standard therapies with a novel regimen including autologous stem cell transplantation. Blood. 2010;115(18):3664-3670. 56. Delabie J, Holte H, Vose JM, et al. Enteropathy-associated T-cell lymphoma: clinical and histological findings from the International Peripheral T-Cell Lymphoma Project. Blood. 2011;118(1):148-155. 57. Raderer M, Troch M, Kiesewetter B, et al. Second line chemotherapy in patients with enteropathy-associated T cell lymphoma: a retrospective single center analysis. Ann Hematol. 2012;91(1):57-61. 58. Malamut G, Chandesris O, Verkarre V, et al. Enteropathy associated T cell lymphoma in celiac disease: a large retrospective study. Dig Liver Dis. 2013;45(5):377-384. 59. Tomita S, Kikuti YY, Carreras J, et al. Genomic and immunohistochemical profiles of enteropathy-associated T-cell lymphoma in Japan. Mod Pathol. 2015;28(10):1286-1296. 60. Nicolae A, Xi L, Pham TH, et al. Mutations in the JAK/STAT and RAS signaling pathways are common in intestinal T-cell lymphomas. Leukemia. 2016;30(11):2245-2247. 61. Nairismägi ML, Tan J, Lim JQ, et al. JAK-STAT and G-proteincoupled receptor signaling pathways are frequently altered in epitheliotropic intestinal T-cell lymphoma. Leukemia. 2016;30(6):1311-1319. 62. Chander U, Leeman-Neill RJ, Bhagat G. Pathogenesis of enteropathy-associated T cell lymphoma. Curr Hematol Malig Rep. 2018;13(4):308-317. 63. Cording S, Lhermitte L, Malamut G, et al. Oncogenetic landscape of lymphomagenesis in coeliac disease. Gut. 2022;71(3):497-508. 64. Murray JA. Several faces of refractory coeliac disease type 2. Gut. 2022;71(3):449-450. 65. Green PHR, Paski S, Ko CW, Rubio-Tapia A. AGA clinical practice update on management of refractory celiac disease: expert review. Gastroenterology. 2022;163(5):1461-1469. 66. Relander T, Pederson MB, Ellin F, et al. Long-term follow-up of clinical outcome determinants and correlative biological features from the Nordic NLG-T-01 trial. Blood. 2022;140(Suppl 1):1479-1480. 67. Phillips EH, Lannon MM, Lopes A, et al. High-dose chemotherapy and autologous stem cell transplantation in enteropathyassociated and other aggressive T-cell lymphomas: a UK NCRI/Cancer Research UK phase II study. Bone Marrow Transplant. 2019;54(3):465-468. 68. Sibon D, Khater S, Bruneau J, et al. The EATL-001 trial: results of a phase 2 study of brentuximab vedotin and CHP followed by consolidation with high-dose therapy - autologous stem-cell transplantation (HDT-ASCT) in the frontline treatment of patients with enteropathy-associated T-cell lymphoma. Blood. 2021;138(Suppl 1):136. 69. Voorhees TJ, Ghosh N, Grover N, et al. Long-term remission in multiply relapsed enteropathy-associated T-cell lymphoma following CD30 CAR T-cell therapy. Blood Adv.

R. Stuver et al.

2020;4(23):5925-5928. 70. Sun J, Lu Z, Yang D, Chen J. Primary intestinal T-cell and NK-cell lymphomas: a clinicopathological and molecular study from China focused on type II enteropathy-associated T-cell lymphoma and primary intestinal NK-cell lymphoma. Mod Pathol. 2011;24:983-992. 71. Yi JH, Lee GW, Do YR, et al. Multicenter retrospective analysis of the clinicopathologic features of monomorphic epitheliotropic intestinal T-cell lymphoma. Ann Hematol. 2019;98(11):2541-2550. 72. Tse E, Gill H, Loong F, et al. Type II enteropathy-associated T-cell lymphoma: a multicenter analysis from the Asia Lymphoma Study Group. Am J Hematol. 2012;87(7):663-668. 73. Huang D, Lim JQ, Cheah DMZ, et al. Whole-genome sequencing reveals potent therapeutic strategy for monomorphic epitheliotropic intestinal T-cell lymphoma. Blood Adv. 2020;4(19):4769-4774. 74. Roberti A, Dobay MP, Bisig B, et al. Type II enteropathy-associated T-cell lymphoma features a unique genomic profile with highly recurrent SETD2 alterations. Nat Commun. 2016;7:12602. 75. Veloza L, Cavalieri D, Missiaglia E, et al. Monomorphic epitheliotropic intestinal T-cell lymphoma comprises morphologic and genomic heterogeneity impacting outcome. Haematologica. 2023;108(1):181-195. 76. Attygalle AD, Cabeçadas J, Gaulard P, et al. Peripheral T-cell and NK-cell lymphomas and their mimics; taking a step forward report on the Lymphoma Workshop of the XVIth meeting of the European Association for Haematopathology and the Society for Hematopathology. Histopathology. 2014;64(2):171-199. 77. Carbonnel F, Lavergne A, Messing B, et al. Extensive small intestinal T-cell lymphoma of low-grade malignancy associated with a new chromosomal translocation. Cancer. 1994;73(4):1286-1291. 78. Egawa N, Fukayama M, Kawaguchi K, et al. Relapsing oral and colonic ulcers with monoclonal T-cell infiltration. Cancer. 1995;75(7):1728-1733. 79. Hirakawa K, Fuchigami T, Nakamura S, et al. Primary gastrointestinal T-cell lymphoma resembling multiple lymphomatous polyposis. Gastroenterology. 1996;111(3):778-782. 80. Carbonnel F, D’Almagne H, Lavergne A, et al. The clinicopathological features of extensive small intestinal CD4 T cell infiltration. Gut. 1999;45(5):662-667. 81. Ranheim EA, Jones C, Zehnder JL, Warnke R, Yuen A. Spontaneously relapsing clonal, mucosal cytotoxic T-cell lymphoproliferative disorder: case report and review of the literature. Am J Surg Pathol. 2000;24(2):296-301. 82. Zivny J, Banner BF, Agrawal S, Pihan G, Barnard GF. CD4+ T-cell lymphoproliferative disorder of the gut clinically mimicking celiac sprue. Dig Dis Sci. 2004;49(4):551-555. 83. Isomoto H, Maeda T, Akashi T, et al. Multiple lymphomatous polyposis of the colon originating from T-cells: a case report. Dig Liver Dis. 2004;36(3):218-221. 84. Svrcek M, Garderet L, Sebbagh V, et al. Small intestinal CD4+ Tcell lymphoma: a rare distinctive clinicopathological entity associated with prolonged survival. Virchows Archiv. 2007;451(6):1091-1093. 85. Margolskee E, Jobanputra V, Lewis SK, et al. Indolent small intestinal CD4+ T-cell lymphoma is a distinct entity with unique biologic and clinical features. PLoS One. 2013;8(7):e68343. 86. Perry AM, Warnke RA, Hu Q, et al. Indolent T-cell lymphoproliferative disease of the gastrointestinal tract. Blood. 2013;122(22):3599-3606. 87. Sharma A, Oishi N, Boddicker RL, et al. Recurrent STAT3-JAK2 fusions in indolent T-cell lymphoproliferative disorder of the gastrointestinal tract. Blood. 2018;131(20):2262-2266.

Haematologica | 108 December 2023

3258

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas 88. Soderquist CR, Patel N, Murty V, et al. Genetic and phenotypic characterization of indolent T-cell lymphoproliferative disorders of the gastrointestinal tract. Haematologica. 2020;105(7):1895. 89. Mansoor A, Pittaluga S, Beck PL, et al. NK-cell enteropathy: a benign NK-cell lymphoproliferative disease mimicking intestinal lymphoma: clinicopathologic features and follow-up in a unique case series. Blood. 2011;117(5):1447. 90. Takeuchi K, Yokoyama M, Ishizawa S, et al. Lymphomatoid gastropathy: a distinct clinicopathologic entity of self-limited pseudomalignant NK-cell proliferation. Blood. 2010;116(25):5631-5637. 91. Xiao W, Gupta GK, Yao J, et al. Recurrent somatic JAK3 mutations in NK-cell enteropathy. Blood. 2019;134(12):986-991. 92. McElroy MK, Read WL, Harmon GS, Weidner N. A unique case of an indolent CD56-positive T-cell lymphoproliferative disorder of the gastrointestinal tract: a lesion potentially misdiagnosed as natural killer/T-cell lymphoma. Ann Diagn Pathol. 2011;15(5):370-375. 93. Isom JA, Arroyo MR, Reddy D, et al. NK cell enteropathy: a case report with 10 years of indolent clinical behaviour. Histopathology. 2018;73(2):345-350. 94. Xia D, Morgan EA, Berger D, et al. NK-cell enteropathy and similar indolent lymphoproliferative disorders. Am J Clin Pathol. 2019;151(1):75-85. 95. Chan JKC, Lo STH. Expression of natural killer cell markers in non-Hodgkin’s lymphomas. Hum Pathol. 1987;18(12):1257-1262. 96. Au WY, Weisenburger DD, Intragumtornchai T, et al. Clinical differences between nasal and extranasal natural killer/T-cell lymphoma: a study of 136 cases from the International Peripheral T-Cell Lymphoma Project. Blood. 2009;113(17):3931-3937. 97. Au WY, Pang A, Choy C, Chim CS, Kwong YL. Quantification of circulating Epstein-Barr virus (EBV) DNA in the diagnosis and monitoring of natural killer cell and EBV-positive lymphomas in immunocompetent patients. Blood. 2004;104(1):243-249. 98. Kim SJ, Yoon DH, Jaccard A, et al. A prognostic index for natural killer cell lymphoma after non-anthracycline-based treatment: a multicentre, retrospective analysis. Lancet Oncol. 2016;17(3):389-400. 99. Cheung MMC, Chan JKC, Lau WH, Ngan RKC, Foo WWL. Early stage nasal NK/T-cell lymphoma: clinical outcome, prognostic factors, and the effect of treatment modality. Int J Radiat Oncol Biol Phys. 2002;54(1):182-190. 100. Kim K, Chie EK, Kim CW, Kim IH, Park CI. Treatment outcome of angiocentric T-cell and NK/T-cell lymphoma, nasal type: radiotherapy versus chemoradiotherapy. Jpn J Clin Oncol. 2005;35(1):1-5. 101. Deng XW, Wu JX, Wu T, et al. Radiotherapy is essential after complete response to asparaginase-containing chemotherapy in early-stage extranodal nasal-type NK/T-cell lymphoma: a multicenter study from the China Lymphoma Collaborative Group (CLCG). Radiother Oncol. 2018;129(1):3-9. 102. Zhang L, Wang Y, Li X, et al. Radiotherapy vs sequential pegaspargase, gemcitabine, cisplatin and dexamethasone and radiotherapy in newly diagnosed early natural killer/T-cell lymphoma: a randomized, controlled, open-label, multicenter study. Int J Cancer. 2021;148(6):1470-1477. 103. Ghione P, Qi S, Imber BS, et al. Modified SMILE (mSMILE) and intensity-modulated radiotherapy (IMRT) for extranodal NK-T lymphoma nasal type in a single-center population. Leuk Lymphoma. 2020;61(14):3331-3341. 104. Zhang Y, Ma S, Cai J, et al. Sequential P-GEMOX and radiotherapy for early-stage extranodal natural killer/T-cell lymphoma: a multicenter study. Am J Hematol. 2021;96(11):1481-1490. 105. Wang L, Wang ZH, Chen XQ, et al. First-line combination of GELOX

R. Stuver et al.

followed by radiation therapy for patients with stage IE/IIE ENKTL: an updated analysis with long-term follow-up. Oncol Lett. 2015;10(2):1036. 106. Zhang L, Jiang M, Xie L, et al. Five-year analysis from phase 2 trial of “sandwich” chemoradiotherapy in newly diagnosed, stage IE to IIE, nasal type, extranodal natural killer/T-cell lymphoma. Cancer Med. 2016;5(1):33-40. 107. Jing X-M, Zhang Z-H, Wu P, et al. Efficacy and tolerance of pegaspargase, gemcitabine and oxaliplatin with sandwiched radiotherapy in the treatment of newly-diagnosed extranodal nature killer (NK)/T cell lymphoma. Leuk Res. 2016;47:26-31. 108. Kim SJ, Kim K, Kim BS, et al. Phase II trial of concurrent radiation and weekly cisplatin followed by VIPD chemotherapy in newly diagnosed, stage IE to IIE, nasal, extranodal NK/T-cell lymphoma: Consortium for Improving Survival of Lymphoma study. J Clin Oncol. 2009;27(35):6027-6032. 109. Yamaguchi M, Tobinai K, Oguchi M, et al. Phase I/II study of concurrent chemoradiotherapy for localized nasal natural killer/Tcell lymphoma: Japan Clinical Oncology Group study JCOG0211. J Clin Oncol. 2009;27(33):5594-5600. 110. Yamaguchi M, Tobinai K, Oguchi M, et al. Concurrent chemoradiotherapy for localized nasal natural killer/T-cell lymphoma: an updated analysis of the Japan Clinical Oncology Group study JCOG0211. J Clin Oncol. 2012;30(32):4044-4046. 111. Yamaguchi M, Suzuki R, Oguchi M, et al. Treatments and outcomes of patients with extranodal natural killer/T-cell lymphoma diagnosed between 2000 and 2013: a cooperative study in Japan. J Clin Oncol. 2017;35(1):32-39. 112. Tsai HJ, Lin SF, Chen CC, et al. Long-term results of a phase II trial with frontline concurrent chemoradiotherapy followed by consolidation chemotherapy for localized nasal natural killer/Tcell lymphoma. Eur J Haematol. 2015;94(2):130-137. 113. Oh D, Ahn YC, Kim SJ, Kim WS, Ko YH. Concurrent chemoradiation therapy followed by consolidation chemotherapy for localized extranodal natural killer/T-cell lymphoma, nasal type. Int J Radiat Oncol Biol Phys. 2015;93(3):677-683. 114. Li YX, Wang H, Jin J, et al. Radiotherapy alone with curative intent in patients with stage I extranodal nasal-type NK/T-cell lymphoma. Int J Radiat Oncol Biol Phys. 2012;82(5):1809-1815. 115. Yamaguchi M, Kwong YL, Kim WS, et al. Phase II study of SMILE chemotherapy for newly diagnosed stage IV, relapsed, or refractory extranodal natural killer (NK)/T-cell lymphoma, nasal type: the NK-Cell Tumor Study Group study. J Clin Oncol. 2011;29(33):4410-4416. 116. Kwong YL, Kim WS, Lim ST, et al. SMILE for natural killer/T-cell lymphoma: analysis of safety and efficacy from the Asia Lymphoma Study Group. Blood. 2012;120(15):2973-2980. 117. Wang JH, Wang L, Liu CC, et al. Efficacy of combined gemcitabine, oxaliplatin and pegaspargase (P-gemox regimen) in patients with newly diagnosed advanced-stage or relapsed/refractory extranodal NK/T-cell lymphoma. Oncotarget. 2016;7(20):29092-29101. 118. Zhao Q, Fan S, Chang Y, et al. Clinical efficacy of cisplatin, dexamethasone, gemcitabine and pegaspargase (DDGP) in the initial treatment of advanced stage (stage III–IV) extranodal NK/Tcell lymphoma, and its correlation with Epstein-Barr virus. Cancer Manag Res. 2019;11:3564. 119. Jaccard A, Gachard N, Marin B, et al. Efficacy of L-asparaginase with methotrexate and dexamethasone (AspaMetDex regimen) in patients with refractory or relapsing extranodal NK/T-cell lymphoma, a phase 2 study. Blood. 2011;117(6):1834-1839. 120. Wang X, Zhang L, Liu X, et al. Efficacy and safety of a pegasparaginase-based chemotherapy regimen vs an Lasparaginase-based chemotherapy regimen for newly diagnosed

Haematologica | 108 December 2023

3259

REVIEW SERIES - Clinical management of rare mature T/NK lymphomas advanced extranodal natural killer/T-cell lymphoma: a randomized clinical trial. JAMA Oncol. 2022;8(7):1035-1041. 121. Hong H, Li Y, Lim ST, et al. A proposal for a new staging system for extranodal natural killer T-cell lymphoma: a multicenter study from China and Asia Lymphoma Study Group. Leukemia. 2020;34(8):2243-2248. 122. Wang J, Wei L, Ye J, et al. Autologous hematopoietic stem cell transplantation may improve long-term outcomes in patients with newly diagnosed extranodal natural killer/T-cell lymphoma, nasal type: a retrospective controlled study in a single center. Int J Hematol. 2018;107(1):98-104. 123. Lee J, Au WY, Park MJ, et al. Autologous hematopoietic stem cell transplantation in extranodal natural killer/T cell lymphoma: a multinational, multicenter, matched controlled study. Biol Blood Marrow Transplant. 2008;14(12):1356-1364. 124. Yhim HY, Kim JS, Mun YC, et al. Clinical outcomes and prognostic factors of up-front autologous stem cell transplantation in patients with extranodal natural killer/T cell lymphoma. Biol Blood Marrow Transplant. 2015;21(9):1597-1604. 125. Kanate AS, DiGilio A, Ahn KW, et al. Allogeneic haematopoietic cell transplantation for extranodal natural killer/T-cell lymphoma, nasal type: a CIBMTR analysis. Br J Haematol. 2018;182(6):916-920. 126. Tse E, Chan TSY, Koh LP, et al. Allogeneic haematopoietic SCT for natural killer/T-cell lymphoma: a multicentre analysis from the Asia Lymphoma Study Group. Bone Marrow Transplant. 2014;49(7):902-906. 127. Kwong YL, Chan TSY, Tan D, et al. PD1 blockade with pembrolizumab is highly effective in relapsed or refractory NK/Tcell lymphoma failing l-asparaginase. Blood. 2017;129(17):2437-2442. 128. Chan TSY, Li J, Loong F, et al. PD1 blockade with low-dose nivolumab in NK/T cell lymphoma failing L-asparaginase: efficacy and safety. Ann Hematol. 2018;97(1):193-196. 129. Huang H, Tao R, Hao S, et al. Sugemalimab monotherapy for patients with relapsed or refractory extranodal natural killer/T-cell lymphoma (GEMSTONE-201): results from a single-arm, multicenter, phase II study. J Clin Oncol. 2023;41(16):3032-3041. 130. Huang H, Tao R, Zou L, et al. Preliminary results from a multicenter, single-arm, phase 2 study of CS1001, an antiprogrammed death-ligand 1 (PD-L1) human monoclonal antibody (mAb), in patients (pts) with relapsed or refractory extranodal natural killer/T cell lymphoma (rr-ENKTL). Blood. 2019;134(Suppl_1):2833. 131. Kim SJ, Lim JQ, Laurensia Y, et al. Avelumab for the treatment of relapsed or refractory extranodal NK/T-cell lymphoma: an openlabel phase 2 study. Blood. 2020;136(24):2754-2763. 132. Bachy E, Savage KJ, Huang H, et al. Treating relapsed/refractory mature T- and NK-cell neoplasms with tislelizumab: a multicenter open label phase 2 study. Blood Adv. 2023;7(16):4435-4447. 133. Tao R, Fan L, Song Y, et al. Sintilimab for relapsed/refractory extranodal NK/T cell lymphoma: a multicenter, single-arm, phase 2 trial (ORIENT-4). Signal Transduct Target Ther. 2021;6(1):365. 134. Ratner L, Waldmann TA, Janakiram M, Brammer JE. Rapid progression of adult T-cell leukemia–lymphoma after PD-1 inhibitor therapy. N Engl J Med. 2018;378(20):1947-1948. 135. Haverkos BM, Alpdogan O, Baiocchi R, et al. Nanatinostat

R. Stuver et al.

(Nstat) and valganciclovir (VGCV) in relapsed/refractory (R/R) Epstein-Barr virus-positive (EBV +) lymphomas: final results from the phase 1b/2 VT3996-201 study. Blood. 2021;138(Suppl 1):623. 136. Kim WS, Oki Y, Kim SJ, et al. Autologous EBV-specific T cell treatment results in sustained responses in patients with advanced extranodal NK/T lymphoma: results of a multicenter study. Ann Hematol. 2021;100(10):2529-2539. 137. Pro B, Advani R, Brice P, et al. Brentuximab vedotin (SGN-35) in patients with relapsed or refractory systemic anaplastic large-cell lymphoma: results of a phase II study. J Clin Oncol. 2012;30(18):2190-2196. 138. Horwitz SM, Advani RH, Bartlett NL, et al. Objective responses in relapsed T-cell lymphomas with single-agent brentuximab vedotin. Blood. 2014;123(20):3095-3100. 139. O’Connor OA, Horwitz S, Masszi T, et al. Belinostat in patients with relapsed or refractory peripheral T-cell lymphoma: results of the pivotal phase II BELIEF (CLN-19) study. J Clin Oncol. 2015;33(23):2492-2499. 140. Piekarz RL, Frye R, Prince HM, et al. Phase 2 trial of romidepsin in patients with peripheral T-cell lymphoma. Blood. 2011;117(22):5827-5834. 141. Coiffier B, Pro B, Prince HM, et al. Results from a pivotal, openlabel, phase II study of romidepsin in relapsed or refractory peripheral T-cell lymphoma after prior systemic therapy. J Clin Oncol. 2012;30(6):631-636. 142. O’Connor OA, Pro B, Pinter-Brown L, et al. Pralatrexate in patients with relapsed or refractory peripheral T-cell lymphoma: results from the pivotal PROPEL study. J Clin Oncol. 2011;29(9):1182-1189. 143. Evens AM, Rosen ST, Helenowski I, et al. A phase I/II trial of bortezomib combined concurrently with gemcitabine for relapsed or refractory DLBCL and peripheral T-cell lymphomas. Br J Haematol. 2013;163(1):55-61. 144. Tan D, Phipps C, Hwang WYK, et al. Panobinostat in combination with bortezomib in patients with relapsed or refractory peripheral T-cell lymphoma: an open-label, multicentre phase 2 trial. Lancet Haematol. 2015;2(8):e326-333. 145. Zinzani PL, Pellegrini C, Broccoli A, et al. Lenalidomide monotherapy for relapsed/refractory peripheral T-cell lymphoma not otherwise specified. Leuk Lymphoma. 2011;52(8):1585-1588. 146. Morschhauser F, Fitoussi O, Haioun C, et al. A phase 2, multicentre, single-arm, open-label study to evaluate the safety and efficacy of single-agent lenalidomide (Revlimid) in subjects with relapsed or refractory peripheral T-cell non-Hodgkin lymphoma: the EXPECT trial. Eur J Cancer. 2013;49(13):2869-2876. 147. Toumishey E, Prasad A, Dueck G, et al. Final report of a phase 2 clinical trial of lenalidomide monotherapy for patients with T-cell lymphoma. Cancer. 2015;121(5):716-723. 148. Zinzani PL, Zain J, Mead M, et al. P1172: duvelisib in patients with relapsed/refractory peripheral T-cell lymphoma from the phase 2 PRIMO trial: updated expansion phase analysis. Hemasphere. 2022;6:1058-1059. 149. Moskowitz AJ, Ghione P, Jacobsen E, et al. A phase 2 biomarkerdriven study of ruxolitinib demonstrates effectiveness of JAK/STAT targeting in T-cell lymphomas. Blood. 2021;138(26):2828-2837.

Haematologica | 108 December 2023

3260

REVIEW SERIES

Biology and genetics of extranodal mature T-cell and NKcell lymphomas and lymphoproliferative disorders Natasha E. Lewis,1 Ting Zhou2 and Ahmet Dogan1

Correspondence: N.E. Lewis natashaelewis@gmail.com

Hematopathology Service, Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY and 2Molecular Diagnostic Laboratory, Department of Hematopathology, MD Anderson Cancer Center, Houston, TX, USA 1

A. Dogan dogana@mskcc.org Received: Accepted:

June 15, 2023. August 28, 2023.

Abstract The extranodal mature T-cell and NK-cell lymphomas and lymphoproliferative disorders represent a unique group of rare neoplasms with both overlapping and distinct clinicopathological, biological, and genomic features. Their predilection for specific sites, such as the gastrointestinal tract, aerodigestive tract, liver, spleen, and skin/soft tissues, underlies their classification. Recent genomic advances have furthered our understanding of the biology and pathogenesis of these diseases, which is critical for accurate diagnosis, prognostic assessment, and therapeutic decision-making. Here we review clinical, pathological, genomic, and biological features of the following extranodal mature T-cell and NK-cell lymphomas and lymphoproliferative disorders: primary intestinal T-cell and NK-cell neoplasms, hepatosplenic T-cell lymphoma, extranodal NK/T-cell lymphoma, nasal type, and subcutaneous panniculitis-like T-cell lymphoma.

Introduction

Refractory celiac disease

Mature T-cell and NK-cell lymphomas and lymphoproliferative disorders are uncommon and heterogeneous with variable clinical presentations, histopathology, genomic alterations, biological foundations, and clinical behavior. A rare subset has a predilection for extranodal sites and demonstrates several unique clinicopathological, genomic, and biological features. Complementing the companion review by Stuver et al.,1 focused on the clinical management of rare extranodal mature T-cell and NK-cell neoplasms that may require systemic therapy, here we summarize the clinicopathological and genomic features of the same group of disorders, emphasizing the pathobiological insights provided by recent genomic advances. While most of the entities reviewed here are classified similarly by the 2017 revised 4th edition of the World Health Organization (WHO) classification, the 2022 5th edition WHO, and the 2022 International Consensus Classification (ICC) systems, some updates to certain diseases have been made, which are noted. For simplicity, 2022 ICC terminology will be utilized when referring to specific entities.2 Select key biological, immunophenotypic, and genomic features of these diseases are summarized in Table 1.

Clinicopathological features Celiac disease (CD) is an immune-mediated condition that arises in genetically predisposed individuals in whom gluten ingestion triggers small bowel damage. Refractory celiac disease (RCD) is a rare, long-term complication of CD and is defined as the persistence of gastrointestinal symptoms and small bowel villous atrophy despite adhering to a strict gluten-free diet for 6-12 months.3 RCD is classified into two types. Type I RCD (RCD-I) is histologically similar to classic CD, demonstrating villous atrophy and increased polyclonal intraepithelial lymphocytes with a normal immunophenotype, and follows a relatively benign course. Type II RCD (RCD-II) morphologically resembles active CD and RCD-I, but is distinguished by clonal, immunophenotypically aberrant intraepithelial lymphocytes. The abnormal intraepithelial lymphocytes demonstrate dual T- and NK-cell traits, expressing cytoplasmic CD3 and the NK receptor NKp464 typically without surface CD3, CD5, CD8, or T-cell receptor (TCR) expression,5 but carrying clonal TCR gene rearrangements. The intraepithelial lymphocytes also typically express CD103 (αE integrin), a receptor for E-cadherin that is thought to promote ad-

Haematologica | 108 December 2023

3261

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL hesion of intraepithelial lymphocytes to epithelial cells.5 The clonal T cells in RCD-II commonly disseminate within the intestine (stomach, colon), to the peripheral blood and bone marrow, as well as to extra-intestinal solid organs, including skin, lung, and mesenteric lymph nodes.6 The prognosis of patients with RCD-II is poor, with there being a 30-50% chance of transformation into enteropathy-associated T-cell lymphoma (EATL) within 5 years.6,7 This risk and the detection of shared TCR gene rearrangements and other somatic genomic alterations among individual RCDII and their corresponding EATL support RCD-II as the neoplastic precursor of EATL.8 As such, RCD-II was newly added as a distinct entity to the 2022 ICC of mature lymphoid neoplasms.2 Genomic and biological features Abnormal intraepithelial lymphocytes in RCD-II are believed to arise from lymphoid precursors differentiating in the gut epithelium but are subsequently reprogrammed toward an NK/T innate-like fate in response to interleukin-15 (IL-15).9,10 IL-15 exerts this effect by switching off NOTCH-dependent T-cell differentiation and diverting the lymphoid precursors from adaptive to innate-like cell differentiation. The chromosomal aberrations in RCD-II include trisomy 1q, which is highly prevalent (90%), and recurrent 4q and 6q losses.6,8 The JAK-STAT pathway, known to regulate intraepithelial lymphocyte function and play pathogenic roles in several T/NK-cell neoplasms, is the most frequently mutated pathway. Indeed, 85% of cases show at least one somatic gain-of-function mutation in JAK1 or STAT3, with the JAK1 p.G1097 hotspot mutation being particularly prevalent (~50%).5,8 Deleterious mutations in negative JAK-STAT regulators (e.g., SOCS1, SOCS3) are common in patients without JAK1 or STAT3 mutations.8 NF-κB signaling is the second most affected pathway. Loss-of-function mutations in negative regulators of NF-κB, such as TNIP3 and TNFAIP3/A20, are detected in ~20%, although the prevalence increases (90%) when abnormal intraepithelial lymphocytes are purified. Recurrent somatic events affecting epigenetic regulators (TET2, KMT2D), DNA damage repair proteins (POT1), and the translational regulator DDX3X are also reported.5,8

Enteropathy-associated T-cell lymphoma Clinicopathological features Three types of aggressive primary intestinal T-cell lymphomas are recognized: EATL, monomorphic epitheliotropic intestinal T-cell lymphoma (MEITL), and intestinal T-cell lymphoma, not otherwise specified (ITCL-NOS).

N.E. Lewis et al.

EATL is the most common and has a strong association with CD, arising in 1% in this population. MEITL is less common and has no association with CD while ITCL-NOS is a diagnosis of exclusion when EATL, MEITL, and other defined T/NK-cell lymphomas have been ruled out. ITCLNOS is likely a heterogeneous category with poorly characterized biological and genetic features and thus will not be covered further in this review. EATL occurs almost exclusively in adults (median age >60 years) with a male predominance. A minority of patients present without a clinical history of CD but demonstrate histological or serological evidence of celiac enteropathy, suggesting subclinical or latent CD. The small bowel is most commonly affected while involvement of colon or stomach is rare (8%).11 Dissemination to extra-intestinal sites is common, and, in up to one-third of RCD-II patients, the initial lymphoma develops at an extra-intestinal site, such as mesenteric lymph nodes, spleen, liver, lung, bone marrow, or skin.6,12 Such extra-intestinal presentations of EATL may arise from pre-existing extra-intestinal RCD-II clonal T cells. The clinical course is unfavorable, with a median survival of ~10 months.13 Tumor cells are typically medium-sized to large and pleomorphic, occasionally with anaplastic or bizarre multinucleated cells (Figure 1A). A polymorphic inflammatory background is often present, occasionally obscuring the lymphoma cells. Angiocentricity and angioinvasion, extensive necrosis, and a high mitotic rate are common. CDrelated pathological features in remote intestinal mucosa are a diagnostic aid (Figure 1B). The tumor cells are immunophenotypically similar to the intraepithelial lymphocytes in RCD-II, including frequent CD103 expression11 (Figure 1C-H), with the exception of CD30, which is commonly expressed in EATL but rare in RCD-II in which it is considered a sign of progression to EATL.14 Genomic and biological features The pathogenesis of EATL remains incompletely understood; however, multiple lines of evidence point to RCDII as its precursor. In addition to shared TCR gene rearrangements, their genetic profiles overlap significantly, with EATL also commonly harboring trisomy 1q and mutations in JAK-STAT and NF-κB pathway genes (e.g., JAK1, STAT3, TNFAIP3/A20), particularly the JAK1 p.G1097 hotspot (68%). KMT2D, TET2, and DDX3X are among other common mutations in both conditions. In contrast to RCD-II, however, EATL exhibits higher genomic complexity, consistent with disease evolution. Multiple chromosomal imbalances are frequent in EATL but rare in RCD-II, and included gains at 9q33-34 and 5q34-35, and losses involving 8p22-23.2, 9p21.2-p21.3, 11q14.1-q14.2, 13q31, and 16q21.1. Multiple JAK1 or STAT3 mutations are observed in ~40% of EATL but only 10% of RCD-II,8 suggesting JAK1STAT3 mutations act as founders as well as drivers of

Haematologica | 108 December 2023

3262

I-NKLPD-GI

3263

Haematologica | 108 December 2023

Chromosomal abnormalities (%)

Gains: 1q (90)

-: CD4, CD5, CD30, EBER

-: CD30, CD56, granzyme B, TCRγδ, EBER

±: CD2, CD20, CD103, granzyme ±: CD7, CD103 B, perforin, MYC

Losses: 7p14.1 (29) 8p22-23.2 (27) 9p21.2-21.3 (13) 11q14.1–q14.2 (13) 16q12.1 (33)

Losses: 7p14.1 (75) 8p22-23.2 (47) 9p21.2-21.3 (33) 11q14.1–q14.2 (27) 16q12.1 (13)

Gains: Gains: CD4+ cases: 1q (20) 1q (73) STAT3::JAK2 (80) 5q34-35.2 (20) 5q34-35.2 (80) 7q33-34 (60) 7q33-34 (33) CD8+ cases: 8q24 (MYC, 27-40) 8q24 (MYC, 60-73) IL2 structural alter9q31.3-qter (73) ations (50) 9q (67)

-: sCD3, CD4, -: sCD3, CD4, CD5, CD56, CD5, CD56, CD30, TCRαβ, TCRγδ, TCRγδ, EBER EBER

Immunophenotype ±: CD8, granzyme ±: CD8, CD30, B, perforin, TCRαβ granzyme B, per(%) (20) forin

-: sCD3, CD4, CD5, TCRαβ, TCRγδ, EBER

±: CD8

+: CD3, CD7, CD8, +: CD2, CD3, CD4 +: CD2, cCD3, CD56, TIA1, MATK or CD8, CD5, TIA1 +: CD2, cCD3, CD7, TIA1, (CD8+ cases), CD7, CD103, TIA1 (>85), TCRγδ > granzyme B TCRαβ TCRαβ

+: CD2, cCD3, CD7, CD103, TIA1, NKp46

Celiac disease

Small bowel, colon Stomach, small > upper GI tract bowel > colon

IC-TLPD-GI

Associated conditions

Small bowel > colon, stomach

MEITL

Small bowel

Small bowel > colon, stomach

EATL

Localization

RCD-II SPTCL

i(7q) (25-80) Trisomy 8 (10-50)

Losses: 1p (100) 2p (83) 5p (83) 9q (83) 10q (83) 11q (83) 12q (83) 16 (83) 20 (100) 22 (100)

Gains: 2q (83) 4q (83)

-: CD4, CD56, CD30, TCRγδ, EBER

Continued on following page.

Losses: 6q21-25 (~40)

CD279 alterations (23)

-: CD4, CD5, -: CD4, CD16, granzyme B, perCD57 forin, CD57, EBER

±: CD8

+: CD3, CD8, TIA1, granzyme B, perforin, TCRαβ

+: CD2, CD3, CD7, +: CD2, cCD3, CD56, TIA1, CD56, TIA1, granzyme M, granzyme B, perTCRγδ > TCRαβ forin, EBER

±: CD5, CD7, CD8, CD30, TCRαβ, ±: CD2, CD5, CD7 TCRγδ

Autoimmune disease (e.g., SLE), HAVCR2 germline mutations

Upper aerodigesSubcutaneous aditive tract > skin, GI pose tissue tract, others

ENKTL

Immune compromise, IL18RAP, HLApossibly thiopurine, DRB1, HLA-DPB1 TNF-α inhibitor germline SNP therapy

Spleen, liver, BM

HSTCL

Table 1. Key biological, phenotypic, and genomic features of select mature extranodal T-cell and NK-cell lymphomas and lymphoproliferative disorders.

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL N.E. Lewis et al.

3264

Haematologica | 108 December 2023

DDX3X (20) PRDM1 (6)

DDX3X (32) PRDM1 (16)

GNAI2 (9-21) DDX3X (3)

TP53 (35) ATM (11)

SETD2 (>90) BCOR (11) KMT2D (6) TET2 (6)

I-NKLPD-GI

CD4+ cases: TET2 DNMT3A KMT2D (together 67)

CD4+ cases: STAT3 (50) JAK3 (30) SOCS1 deletion (17)

IC-TLPD-GI

UBR5 (5-10) IDH2 (5-10)

PIK3CD (9)

TP53 (5-10)

SETD2 (25) INO80 (21) TET3 (15) SMARCA2 (10)

STAT5B (30) STAT3 (10)

HSTCL

DDX3X (14) MGA (8)

TP53 (10)

KMT2D (13) BCOR (11) TET2 (9) KMT2C (8) EP300 (7) ARID1A (7)

STAT3 (18) JAK3 (9) STAT5B (<5)

ENKTL

HAVCR2 (25-85) NAV3 mutation/loss (44) PLCG2 (15) CBL (5) IDH1 (5)

PIK3CD (10) MTOR (5)

CDC27 (25) TET2 (15) KMT2C (15) KMT2D (15) ASXL1 (10) BAZ2A (10)

JAK3 (10)

SPTCL

RCD-II: type II refractory celiac disease; EATL: enteropathy-associated T-cell lymphoma; MEITL: monomorphic epitheliotropic intestinal T-cell lymphoma; IC-TLPD-GI: indolent clonal T-cell lymphoproliferative disorder of the gastrointestinal tract; I-NKLPD-GI: indolent NK-cell lymphoproliferative disorder of the gastrointestinal tract; HSTCL: hepatosplenic T-cell lymphoma; ENKTL: extranodal NK/T-cell lymphoma, nasal type; SPTCL: subcutaneous panniculitis-like T-cell lymphoma; GI: gastrointestinal; BM: bone marrow; NA: not applicable; SNP: single nucleotide polymorphisms; SLE: systemic lupus erythematosus; cCD3: cytoplasmic CD3; sCD3: surface CD3.

Others

PI3K/AKT/ mTOR

POT1 (26) TP53 (5-10)

KMT2D (37) TET2 (32) BCOR (32) ASXL1 (16) ARID1A (16)

TET2 (30) KMT2D (22) KMT2C (10) BCOR (10) CREBBP (8) SETD2 (4)

BRAF (26) KRAS (20) NRAS (5)

BRAF (10) NRAS (5) KRAS (0-5)

TNIP3 and TNFAIP3/A20 (together 31)

STAT5B (60) JAK3 (35-50) SH2B3 (20) STAT3 (15) JAK1 (12)

MEITL

JAK1 (74) STAT3 (47) SH2B3 (16) SOCS1 (10) JAK3 (5) STAT5B (5)

EATL

TNIP3 and TNFAIP3/A20 (together 20)

DNA damage POT1 (20) repair

Epigenetic regulators

NF-κB

MAPK

JAK/STAT

JAK1 (48) STAT3 (38) JAK3 (11) SOCS1 (11) SOCS3 (8)

Mutations (%)

RCD-II

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL N.E. Lewis et al.

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL transformation. JAK1-STAT3 pathway targeting may offer a potential therapeutic approach to suppress growth of RCD-II cells, preventing progression to EATL. EATL also harbor additional mutations uncommon in RCD-II, including those in the MAPK pathway (e.g., KRAS, NRAS, BRAF) (20%) and TP53.8,15 Additional pathogenic mutations in EATL include SH2B3, BCOR, ARID1A, SETD1B, PTPRC, PRD, NF1, and NOTCH1, the value of which in predicting progression is unclear.8 It is postulated that EATL lymphomagenesis follows a multi-step process, initiated by CD-associated cytokines that trigger polyclonal expansion of intraepithelial lymphocytes. IL-15, which is upregulated in CD intestinal epithelium, triggers a powerful anti-apoptotic cascade

N.E. Lewis et al.

involving JAK3 and STAT5 phosphorylation, hindering Tcell elimination after activation.16 Subsequent acquisition of JAK1 or STAT3 mutations in an intraepithelial lymphocyte clone confers hyper-responsiveness to IL-15 and other cytokines and, in concert with alterations in negative regulators of NF-κB, provides a selective advantage and promotes clonal outgrowth.9,17 The synergy between cytokines and mutations also triggers autonomous cytokine production in abnormal intraepithelial lymphocytes and epithelial cytotoxicity, resulting in self-sustaining inflammation that leads to loss of response to gluten-free diets and progression to RCD-II. This process also creates a genotoxic inflammatory environment that fosters genomic instability, enabling the

Figure 1. Enteropathy-associated T-cell lymphoma. (A-H) This case was composed of intermediate-sized to large cells with high mitotic activity, abundant apoptotic debris, and angioinvasion (A, H&E). Although the patient had not reported a history of gastrointestinal symptoms prior to presentation, the small bowel distant from the mass showed villous blunting and marked intraepithelial lymphocytosis (B, H&E) while serological evaluation detected anti-gliadin IgA, anti-gliadin IgG, anti-tissue transglutaminase IgA, and anti-endomysial IgA antibodies, consistent with subclinical celiac disease. The neoplastic cells showed a typical phenotype of enteropathy-associated T-cell lymphoma, expressing CD30 (C), and CD103 (D), and lacking CD4 (E), CD8 (F), TCRβ (G), and TCRδ (H) expression. H&E: hematoxylin and eosin. Haematologica | 108 December 2023

3265

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL accumulation of additional genetic aberrations and ultimately leading to the development of EATL.

Monomorphic epitheliotropic intestinal T-cell lymphoma Clinicopathological features Monomorphic epitheliotropic intestinal T-cell lymphoma (MEITL) constitutes <5% of gastrointestinal lymphomas; however, it is the predominant form of primary intestinal T-cell lymphoma in people of Asian and Hispanic descent. The low incidence of CD in both populations indicates that MEITL is not related to CD. MEITL primarily affects older adults, has a male predominance, and typically localizes to small bowel (jejunum more frequently than ileum or duodenum) and rarely the colon or stomach. Dissemination to regional lymph nodes and distant organs can occur.11 The prognosis is dismal, with a median survival of 7-15 months.18 Tumors usually consist of monotonous, small to mediumsized lymphoid cells that show prominent epitheliotropism (Figure 2A). However, variant morphology, including cellular pleomorphism, larger cell size, or prominent nucleoli, has been described in a minority of cases.19,20 Unlike EATL, MEITL typically lacks angiotropism, necrosis, and extensive background inflammation, although these can occasionally occur.20 Villous blunting may be seen but is mostly confined to the peritumoral mucosa. The phenotype of MEITL differs from that of EATL in that the neoplastic cells are typically positive for CD8 and CD56 but negative for CD30, irrespectively of histological appearance11,19,20 (Figure 2B-E). Occasional cases with atypical immunophenotypes (negative for CD8 and/or CD56) are described.19,20 Aberrant CD20 expression is reported in 20%. MATK expression is reported as a characteristic marker, although it is not widely used in practice.21 TCR is generally expressed, TCRγδ more commonly than TCRαβ. A small subset is TCR-silent, and coexpression of both TCR isoforms is rare. One-third of cases express MYC, which partly reflects underlying MYC alterations (Figure 2F-G). Epstein-Barr virus (EBV) is absent in tumor cells. Genomic and biological features The genetic profile of MEITL is distinct from that of EATL. Compared to EATL, MEITL shows a significantly higher frequency of gains in the MYC locus (73% vs. 27%) (Figure 2G) and a lower frequency of gains in 1q and 5q, although the two conditions share several recurrent chromosomal abnormalities, such as gains of 9q34 and 7 and losses of 8p22-23, 11q14, and 16q12.15,22-24 Recent studies have suggested that a combination of epigenetic deregulation and cell signaling activation may play a central role in the pathogenesis of MEITL. Deleterious

N.E. Lewis et al.

mutations or deletions of SETD2, observed in nearly all cases (97%), have emerged as a genetic hallmark of MEITL.20,25,26 SETD2 is a lysine methyltransferase that is exclusively responsible for catalyzing the trimethylation of lysine 36 in histone H3 (H3K36me). H3K36me is a key modification that regulates gene expression at multiple levels (e.g., histone and DNA methylation, transcriptional activities, RNA splicing). Additionally, H3K36me plays a role in DNA damage repair, facilitating recruitment of DNA repair machinery. SETD2 loss of function results in reduced or absent H3K36me, presumably interrupting the above-mentioned activities. SETD2 also methylates non-histone substrates such as α-tubulin, whose methylation by SETD2 during mitosis and cytokinesis is essential for proper chromosome segregation and genomic stability. Additionally, SETD2 binds to TP53, enhances its stability, and upregulates expression of some of its target genes.27 Considering the broad and multilayered roles of SETD2, it is difficult to predict which genes and pathways are critically affected by SETD2 inactivation in lymphomagenesis. One phenotypic consequence is the expansion of γδ T cells, as demonstrated in a T-cell-specific SETD2 knockout mouse model, which may explain the dominance of γδ T-cell origin in MEITL.15 Given its role in DNA damage repair and TP53 stabilization, SETD2 inactivation conceivably increases genomic instability and facilitates the acquisition of collaborating genetic events, which predominantly affect cell signaling pathways. In common with EATL, MEITL frequently harbors mutations in the JAK-STAT pathway, albeit in different genes, namely STAT5B and JAK3.20,22 Genes in the MAPK pathway (e.g., BRAF, KRAS, NRAS) are altered in 30-50% and are mutually exclusive.25,28 GNAI2, which encodes a guanine nucleotide binding protein subunit, is mutated in 9%-21%.22 One study identified SYK overexpression, likely due to promoter hypomethylation, as a distinctive marker of MEITL (95% prevalence vs. 0% in EATL), suggesting enhanced TCR signaling29 (Figure 2E). Approximately one-third of cases exhibit TP53 mutations, which are frequently associated with atypical morphology, concurrent MYC aberrancies (40% of TP53-mutated cases), and particularly dismal outcome, suggesting that TP53 and MYC aberrations may cooperatively drive MEITL progression.20

Indolent clonal T-cell lymphoproliferative disorder of the gastrointestinal tract Clinicopathological features Indolent clonal T-cell lymphoproliferative disorder of the gastrointestinal tract (IC-TLPD-GI) was upgraded from a provisional entity in the 2017 revised 4th edition of the WHO

Haematologica | 108 December 2023

3266

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL

N.E. Lewis et al.

Figure 2. Monomorphic epitheliotropic intestinal T-cell lymphoma. (A-E) A dense infiltrate of monotonous small to intermediate-sized neoplastic cells with condensed chromatin and a rim of pale cytoplasm which infiltrates the small bowel epithelium is typical of monomorphic epitheliotropic intestinal T-cell lymphoma (A, hematoxylin and eosin). The cells expressed CD8 (B), CD56 (C), and TCRδ (D), and overexpressed SYK (E). (F, G) Another case showed increased MYC expression (F) and gain of the MYC locus on chromosome 8q24, as determined by fluorescence in situ hybridization (G).

classification to a definite entity in both the 2022 5th edition of the WHO and ICC systems.2,30 The 5th edition of the WHO classification, however, changed the terminology from “lymphoproliferative disorder” to “lymphoma” due to the disease’s significant morbidity and ability to disseminate.30 It is most common among 50- to 60-year-olds and has a slight male predominance. It typically affects the small bowel and colon and is often multifocal. Mesenteric lymph node involvement is uncommon and distant dissemination is rare. While most patients experience chronic persistent/relapsing disease without progression for up to 14 years, death due to large cell transformation, occurring 10-27 years after diagnosis, has been reported.31 Limited data suggest that CD4+ cases may have a higher risk of progression.32 Histologically, IC-TLPD-GI shows a dense, non-destructive

proliferation of bland, monomorphic, small to intermediate-sized lymphocytes in the lamina propria with displacement of the intestinal epithelium without invasion. Mitotic figures and apoptosis are scarce, and vascular invasion and necrosis are absent. The tumor cells express CD3 and CD2 and may downregulate CD5 and/or CD7. Equal numbers of CD4+ and CD8+ cases are reported, with rare occurrences of double-negative or double-positive immunophenotypes. IC-TLPD-GI expresses TCRαβ and occasionally CD103. The Ki-67 proliferation index is low (< 5%) and EBV is absent. Genomic and biological features Despite limited published data, evidence suggests that ICTLPD-GI has a lower burden of genetic aberrations compared to EATL or MEITL.15,25,33 CD4+ and CD8+ cases exhibit distinct molecular signatures, which may partly explain

Haematologica | 108 December 2023

3267

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL their different risks of progression. Recurrent alterations in the CD4+ cases include JAK/STAT pathway alterations (e.g., STAT3-JAK2 fusion, STAT3 mutation, SOCS1 deletion) and loss-of-function mutations in epigenetic modifiers (e.g., TET2, DNMT3A, KMT2D).34 Conversely, CD8+ cases are enriched with IL2 structural alterations.33

Indolent NK-cell lymphoproliferative disorder of the gastrointestinal tract Clinicopathological features Indolent NK-cell lymphoproliferative disorder of the gastrointestinal tract (I-NKLPD-GI) is an entity newly added to both the 2022 5th edition WHO and ICC systems.2,30 It encompasses cases previously designated as NK-cell enteropathy or lymphomatoid gastropathy. I-NKLPD-GI is extremely rare with fewer than 80 cases having been documented.35 Most occur in middle-aged to older individuals and affect the stomach or duodenum. I-NKLPD-GI follows a protracted but indolent clinical course, with some lesions undergoing spontaneous resolution while others persist for years despite treatment. Unlike IC-TLPDGI, I-NKLPD-GI is not known to progress to a more aggressive disease or to disseminate to lymph nodes or other organs. Histological examination reveals a diffuse infiltrate of medium-sized to large lymphoid cells expanding the lamina propria, which may exhibit epitheliotropism, ulceration, glandular destruction, or accompanying acute inflammation but typically lacks angiocentricity and necrosis. Tumor cells demonstrate an activated NK-cell immunophenotype and variable Ki-67 proliferative index. EBV is absent, distinguishing it from extranodal NK/T-cell lymphoma, nasal type (ENKTL).11 Genomic and biological features Despite initial debates, the identification of somatic mutations in a subset of I-NKLPD-GI supports its neoplastic nature and justifies its classification as a lymphoproliferative disorder.2,36,37 A recent study identified a recurrent, somatic, small, in-frame deletion in exon 12 of JAK3. Nonrecurrent mutations involving PTPRS, AURKB, AXL, ERBB4, IGF1R, PIK3CB, CUL3, CHEK2, RUNX1T1, CIC, SMARCB1, and SETD5 are also reported.37 Clonal TCR gene rearrangements are absent.

Hepatosplenic T-cell lymphoma Clinicopathological features Hepatosplenic T-cell lymphoma (HSTCL) is an aggressive lymphoma of mature, cytotoxic T cells with a predilection

N.E. Lewis et al.

for spleen, liver, and bone marrow. It accounts for ~1-2% of T/NK-cell lymphomas, occurring more commonly in North America and Europe than in Asia,38,39 and has a dismal prognosis.40-43 It shows a male preponderance and most commonly affects young adults (median age early to mid 30s) but can develop across a wide age range. Approximately 20% occur in immunocompromised individuals (e.g., those with autoimmune disease, inflammatory bowel disease, prior solid organ or stem cell transplant, or who take immunosuppressive drugs).43,44 Most patients present with hepatosplenomegaly, B symptoms, cytopenias, and bone marrow involvement without significant lymphadenopathy. While peripheral blood involvement is common, lymphocytosis at initial diagnosis is rare.40-43 Hemophagocytic syndrome is a well-recognized but uncommon complication.43 Neoplastic T cells are usually small to intermediate in size with mature chromatin, inconspicuous nucleoli, and moderate amounts of pale agranular cytoplasm (Figure 3A). However, variable cytomorphology, including large cell size, cellular pleomorphism, or blastic appearance with dispersed chromatin resembling acute leukemia, can be seen.40-43 Lymphoma cells typically involve cords and sinusoids of the splenic red pulp as well as sinusoids of the bone marrow and liver. They typically express the T-cell markers CD2, CD3 and CD7 and aberrantly lack CD5 (Figure 3B). They usually lack both CD4 and CD8, but a minor subset expresses CD8. CD56 is positive in ~70% while CD57 is usually negative (Figure 3C).45,46 Most cases show a nonactivated cytotoxic T-cell phenotype, expressing T-cell intraellular antigen (TIA1) and granzyme M without perforin or granzyme B (Figure 3D). EBV is typically absent.45,46 Most HSTCL express TCRγδ (~75%) (Figure 3E), typically with Vδ1 gene usage,47 but variants expressing TCRαβ (~20%) or lacking TCR expression (~5%) occur.46 TCRαβ+ cases are clinically and pathologically similar to TCRγδ+ cases but have been associated with female sex, older age, and poorer outcomes in some studies.43,48 Genomic and biological features The most common cytogenetic alterations include isochromosome 7q [i(7q)] and trisomy 8, which often cooccur and have been reported in ~25-80% and 10-50% of cases, respectively, as determined by karyotype or fluorescence in situ hybridization analysis40-43,45,48,49 (Figure 3FI). Less frequent alterations include ring chromosome 7, losses in 4p, 10p, and 10q, and gains in 1q and 17q.49,50 Isochromosome 7q likely represents a primary event while other alterations, including trisomy 8, are postulated to occur secondarily.51 The pathogenic role of i(7q), which results in loss in 7p and gain in 7q, is still unclear. The 7p loss has been associated with enhanced CHN2 expression and its encoded signal transduction protein β2-chimerin, which may downmodulate the NFAT pathway and enhance

Haematologica | 108 December 2023

3268

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL

N.E. Lewis et al.

Figure 3. Hepatosplenic T-cell lymphoma. (A-F) Bone marrow biopsy in this patient with hepatosplenic T-cell lymphoma showed an infiltrate of intermediate-sized atypical lymphoid cells with condensed chromatin and pale cytoplasm involving and expanding the sinusoids (A, hematoxylin and eosin, black arrows). The cells expressed CD3 (B), CD56 (C), TIA1 (D), and TCRδ (E). Karyotyping detected multiple characteristic chromosomal abnormalities, including isochromosome 7q (red arrow) and trisomy 8 (black arrow) (F). (G-I) In another case of hepatosplenic T-cell lymphoma, both isochromosome 7q and trisomy 8 were detected by single nucleotide polymorphism array analysis (G, red arrow indicates 7p loss, purple arrow indicates 7q gain, black arrow indicates gain of 8) and fluorescence in situ hybridization (H, 3 signals of 7q31 [red probe] and 2 signals of CEP 7 [green probe] indicate gain of 7q; I, 3 signals of CEP 8 [red probe] indicates gain of 8). BAF: B-allele frequency.

cell proliferation, while the 7q gain is associated with increased expression of several genes at that locus implicated in tumorigenesis, such as RUNDC3B, PPP1R9A and ABCB1, which may impart an intrinsic growth advantage and chemoresistance.50 Gene expression profiling has demonstrated that HSTCL cluster separately from other T-cell lymphoma types.52 TCRγδ+ and TCRαβ+ HSTCL cluster together and show highly similar gene signatures.52 Genes upregulated in HSTCL as compared to peripheral T-cell lymphoma, not

otherwise specified (PTCL-NOS) and ENKTL include S1PR5, involved in homing of NK cells to the spleen (potentially contributing to tumor cell localization within spleen and marrow sinusoids), and ABCB1 (alias MDR1), which encodes a P-glycoprotein multidrug transporter (potentially contributing to tumor chemoresistance by extruding drugs from tumor cells), while the tumor suppressor AIM1 is underexpressed. In addition, genes encoding NK-cell associated molecules are overexpressed in HSTCL compared to PTCL-NOS, while genes involved in immunomodulation

Haematologica | 108 December 2023

3269

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL and CD5 are underexpressed. Several categories of genes are differentially expressed as compared to normal γδ T cells, suggesting their importance in pathogenesis.52,53 Overexpressed genes involve NK-cell-associated molecules (e.g., KIR2DL2, KIR2DL3) and those related to oncogenes (e.g., FOS, VAV3), cell adhesion (e.g., VCAM1), tyrosine kinases (e.g., SYK), signal transduction (e.g., SPRY2), the sonic hedgehog and WNT pathways, and S1PR5, while underexpressed genes include those associated with cytotoxicity (e.g., GZMB), cytokines, AIM1, and CD5. Pre-treatment tumor cells have been shown to demonstrate genetic heterogeneity, despite derivation from a common ancestral clone, with differential chemoresponsiveness.53 Chemoresistance was associated with increased expression of genes associated with tumor survival (e.g., IL32, TOX2) and drug resistance (e.g., AIF1, AKAP12), suggesting potential mechanisms of treatment failure.53 Somatic mutations occur in several gene types. Mutations of chromatin-modifying genes occur in ~62% of cases, the most common being inactivating mutations of SETD2 followed by mutations of INO80, TET3, and SMARCA2.49 Activating mutations of the signaling pathway genes STAT5B, STAT3, and PIK3CD are reported in nearly half of cases.45,49,54 Recurrent mutations in other driver genes, such as TP53, UBR5, and IDH2, are less common.49 Recurrent DNA methylation changes in HSTCL preferentially affect regulatory regions such as promotors and enhancers. These include hypermethylation of AIM1, BCL11B, CD5, CXCR6, GIMAP7, LTA, SEPT9, UBAC2, and UXS1, some of which have been implicated in the pathobiology of Tcell neoplasms and associated with aberrantly absent protein expression (e.g., CD5). Recurrently hypomethylated genes include ADARB1, NFIC, NR1H3, and ST3GAL3.52,55 Some of these alterations, such as chromatin-modifying gene mutations and AIM1 promotor methylation, STAT3/5B mutations, PIK3CD mutations, and SYK overexpression may represent therapeutic targets, potentially amenable to epigenetic modifiers, JAK/STAT pathway inhibitors, PI3K inhibitors, and SYK inhibitors, respectively.49,52,54 Given the association with immunocompromise, it has been postulated that immune suppression/dysregulation may play an etiological role, potentially by reducing an individual’s ability to clear pathogens leading to chronic antigenic stimulation of T cells, particularly γδ T cells that have more limited antigenic specificity.56 The resultant polyclonal γδ T-cell outgrowth may subsequently acquire genomic alterations leading to clonal expansion and malignant transformation. Approximately 10% of HSTCL arise in patients treated with thiopurines and/or tumor necrosis factor-α inhibitors for inflammatory bowel disease.44 While a causal role of these drugs in HSTCL development has been suggested, this remains controversial. The risk of developing lymphoma, including HSTCL, is reported to increase following treatment with these drugs.57 Tumor ne-

N.E. Lewis et al.

crosis factor-α inhibitor therapy has been associated with clonal expansion of γδ T cells in patients with high baseline γδ T-cell counts and can induce γδ T-cell proliferation in vitro in a dose-dependent manner, suggesting that it may contribute to early pre-neoplastic clone development.58 However, tumor necrosis factor-α inhibitors are not essential for the development of HSTCL among patients with immunodysregulatory disorders, suggesting that other factors may be more pathogenically important in such patients, such as other immunosuppressive drugs, genetic predisposition, and chronic antigenic stimulation.59

Extranodal NK/T-cell lymphoma, nasal type Clinicopathological features Extranodal NK/T-cell lymphoma (ENKTL) is an aggressive EBV-associated lymphoma derived from NK cells or T cells. Of note, the qualifier “nasal-type” was dropped from the name in the 5th edition of the WHO classification as the disease can present at extra-nasal sites.30 It accounts for ~10% of T/NK-cell lymphomas worldwide but is significantly more common in Asia and Latin America than in western countries.38,39,60 It typically arises in middle-age and most commonly affects the upper aerodigestive tract, with other sites of involvement including the gastrointestinal tract, skin, and testes. Secondary lymph node involvement, bone marrow infiltration, B symptoms, and hemophagocytic syndrome occur in a subset of patients.61 The prognosis is generally poor but variable, with worse outcomes reported for patients with non-nasal presentation.61 Tumors show a diffuse infiltration of pleomorphic cells with irregular nuclei, frequently with an angiocentric and angiodestructive growth pattern and necrosis (Figure 4A). They typically express cytoplasmic CD3, CD2, CD56, and cytotoxic markers (TIA1, granzyme B, perforin) and lack CD4 (Figure 4B, C). Those of NK-cell lineage lack surface CD3 expression and a clonal TCR gene rearrangement while T-cell-derived cases typically express surface CD3, may variably express CD5 or CD8, and show monoclonal TCR gene rearrangements. EBV is detected in most tumor cells by in situ hybridization for EBV-encoded RNA (EBER) (Figure 4D). While unique clinicopathological features can typically distinguish ENKTL among other EBV+ diseases, such as chronic active EBV disease and aggressive NK-cell leukemia, when widespread dissemination occurs, discrimination may not be possible. Genomic and biological features Recurrent gains involving chromosomes 1q, 2q, 7q, 17q and 20q and losses involving 6q, 11q, 13q and 17p are re-

Haematologica | 108 December 2023

3270

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL

N.E. Lewis et al.

Figure 4. Extranodal NK/T-cell lymphoma, nasal type. (A-E) This extranodal NK/T-cell lymphoma, nasal type showed sheets of atypical lymphoid cells ranging from small to large in size with irregular nuclei (A, hematoxylin and eosin). The cells were diffusely positive for CD3 (B), CD56 (C), and EBER (D). High PD-L1 expression was also present (E).

ported.62-65 Multiple potentially pathogenically important tumor suppressor genes lie within the most commonly deleted 6q21-25 region, including PRDM1, HACE1, PTPRK, FOXO3, ATG5, and AIM1.63,64,66,67 The gene expression profiles of both NK- and T-cell-derived ENKTL cluster together, separate from PTCL, NOS, and are characterized by upregulation of genes of the JAK/STAT, NF-κB and Notch pathways and MYC.63,68 Mutations most commonly affect the JAK/STAT pathway (e.g., STAT3, JAK3, STAT5B), tumor suppressors (e.g., TP53, DDX3X, MGA), and epigenetic modifiers (e.g., KMT2D, KMT2C, BCOR, TET2).62,69,70 JAK/STAT pathway activation, occurring through gene mutations or phosphorylation of JAK3 or STAT3, is pathogenically critical in ENKTL and a potential therapeutic target.63,71 RAS/MAPK (e.g., NRAS, KRAS, BRAF, MAP3K5), Notch (e.g., NOTCH1/2), NF-κB (e.g., ECSIT, BIRC3), and immune surveillance (e.g., CIITA, HLA-A) pathway mutations are less frequent.62,69,70 ENKTL commonly exhibit global promoter hypermethylation, including that of pathologically important tumor suppressor genes (e.g., BCL2L11, DAPK1, PTPN6, TET2).72 They also commonly express programmed cell death li-

gand-1 (PD-L1) (Figure 4E), which is mediated by CD274 alterations (amplification or 3’-untranslated region truncation) or upregulation driven by LMP1 or STAT3.73-76 It is still unclear whether PD-L1 expression or CD274 alterations can predict response to anti-PD-1/PD-L1 therapy, a treatment used with some success in the relapsed/refractory setting.75,77 Several genomic subtypes of ENKTL are reported. A multiomics study described three molecular subtypes associated with differential EBV transcriptional patterns and sensitivities to targeted therapies: (i) TSIM (alterations in tumor suppressor and immune modulator genes), characterized by JAK-STAT pathway activation, NK-cell origin, and PD-L1 overexpression; (ii) MB (MGA mutations and BRDT loss of heterozygosity), associated with MYC overexpression and poor outcomes; (iii) HEA (HDAC9, EP300, and ARID1A mutations), defined by epigenetic alterations, NF-κB activation and T-cell origin.78 Through consensus clustering analysis of mutations and copy number alterations, another study identified seven genetic clusters (C1C7) with differential survival outcomes.62 Patients in the C6

Haematologica | 108 December 2023

3271

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL group (characterized by RAS/RAF/MAPK pathway, JAK3, BCOR, and TP53 aberrations, and chromosome 1 and 7 copy number alterations) had inferior outcomes while improved survival was associated with groups C5 (gains of JAK3 and chromosome 19q/q13) and C7 (TET2 loss and ARID1B mutations). Using gene expression profiling and immunohistochemistry, another study identified four immune microenvironmental subtypes (immune tolerance, immune evasion-A, immune evasion-B, immune silenced) which were associated with differential clinical outcomes, PD-L1 expression, and response to anti-PD-1 therapeutics.79 The identification of specific germline polymorphisms associated with risk and outcomes along with the ethnic/geographic bias of ENKTL suggest that genetic predisposition with or without environmental factors may play a pathogenic role. An increased risk of ENKTL development has been associated with single nucleotide polymorphisms in IL18RAP (which encodes the interleukin-18 receptor accessory β-subunit), HLA-DRB1, and HLA-DPB1 in Asian individuals.80,81 Specific single nucleotide polymorphisms have also been associated with differential survival.82 Aside from minor differences in frequencies of genomic alterations, the genomic landscape is largely similar in Asian and Hispanic populations, suggesting similar oncogenic mechanisms.62,70 EBV invariably plays a critical pathogenic role given the strong association of ENKTL with the virus; however, the mechanisms are incompletely understood. The driver function of EBV is supported by the lower mutational burden in ENKTL and other EBV+ neoplasms compared to other aggressive tumors (e.g., diffuse large B-cell lymphoma).78 Most ENKTL harbor type A EBV with a 30 bp deletion in LMP1, although this is less common in Latin America, potentially due to geographic variation of EBV strains.70,83 Tumor cells usually show type II latency with clonal episomal EBV, although type I latency and some integration of EBV DNA into the host genome can occur.78,84 Viral LMP1-mediated activation of signaling pathways, such as NF-κB and MAPK, and epigenetic changes via modulation of host epigenetic machinery and EBV-encoded microRNA have been suggested as mechanisms of oncogenesis.85,86 Small indels and long-fragment deletions of the EBV genome as well as integration of EBV fragments into the host genome which disrupt transcription of important host genes, such as NHEJ1, may also promote oncogenesis.

Subcutaneous panniculitis-like T-cell lymphoma Clinicopathological features Subcutaneous panniculitis-like T-cell lymphoma (SPTCL)

N.E. Lewis et al.

is an adipotropic lymphoproliferative disorder of TCRαβ+, CD8+, cytotoxic T cells that primarily involves subcutaneous tissue. It accounts for ~1% of T/NK-cell lymphomas,38,39 affects a wide age range (median age 30-40 years), and occurs more commonly in females.87-90 SPTCL is typically limited to subcutaneous adipose tissue with rare reports of involvement of extracutaneous fat-rich sites.87-92 Patients typically present with multiple subcutaneous nodules or plaques, usually without ulceration, most commonly on the extremities or trunk. B symptoms and/or laboratory abnormalities are seen in over half of patients and hemophagocytic syndrome occurs in ~2030%.87,89,90 Approximately 20-40% of patients have an associated autoimmune disease, most commonly systemic lupus erythematosus,87,89,90 although autoantibodies are reported in ~60% of patients, including in those without a history of autoimmune disease.89,93 Given some overlapping clinical and morphological features, distinguishing SPTCL from lupus erythematosus panniculitis can be challenging. Although relapses are common, the clinical course is typically indolent with survival rates of >70%,87-90,93 but worse outcomes in those who develop hemophagocytic syndrome are reported.90 Historically, many patients were treated with combination chemotherapy, however, recent studies have demonstrated clinical responses to immunomodulatory drugs.89,90,93 Histologically, atypical lymphoid cells infiltrate subcutaneous fat lobules with sparing of the septa, superficial dermis, and epidermis.87-90 The neoplastic cells are predominantly small to intermediate in size with irregular, hyperchromatic nuclei, which characteristically rim individual adipocytes (Figure 5A). Admixed small reactive T cells, histiocytes, karyorrhectic debris and fat necrosis are common while there are typically few background B cells, plasma cells, granulocytes, and plasmacytoid dendritic cells. The neoplastic cells express CD3, CD8, and cytotoxic markers (TIA1, granzyme B, perforin), show variable loss of CD2, CD5 and/or CD7, and typically lack CD4, CD30 and CD56 expression (Figure 5B, C). They express TCRαβ and lack expression of TCRγδ and EBV (Figure 5D). The Ki-67 proliferation index is often high (Figure 5E). Genomic and biological features Clonal TCR gene rearrangements are detected in most cases.88-90 Chromosomal copy number alterations have been identified in isolated SPTCL cells, with the most common including losses in 1p, 2p, 5p, 9q, 10q, 11q 12q, 16, 20, and 22 and gains in 2q and 4q.94 Gene expression profiling studies have suggested that inflammatory pathways and immune escape may be etiologically important in SPTCL. By gene expression profiling, cases of SPTCL group together and apart from cases of lupus erythematosus panniculitis, suggesting distinct biological backgrounds.95 However, a subgroup of cases of lupus erythematosus

Haematologica | 108 December 2023

3272

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL

N.E. Lewis et al.

Figure 5. Subcutaneous panniculitis-like T-cell lymphoma. (A-E) A panniculitic lymphoid infiltrate composed of intermediatesized atypical lymphoid cells with irregular nuclei and hyperchromatic chromatin are seen rimming individual adipocytes in this case of subcutaneous panniculitis-like T-cell lymphoma (A, hematoxylin and eosin). Foamy macrophages are present in the background. The atypical lymphocytes expressed CD3 (B), CD8 (C), and TCRα (D), and showed a high Ki-67 proliferative index (E). (F) Another similar case showed atypical cytoplasmic granular/paranuclear dot-like TIM3 staining, instead of the normal membranous/cytoplasmic staining pattern, raising suspicion of misfolded TIM3 protein, although HAVCR2 mutational testing was not performed.

panniculitis has been shown to share some gene expression features with SPTCL, suggesting a potential molecular relationship. Genes reported to be overexpressed in SPTCL involve cytotoxicity (e.g., PRF1, NKG7, GZMB), cytokines and chemokines (e.g., IFNG, CXCR3, CXCR6, CXCL9, CXCL10, CXCL11, CCL5, CCR5), T-cell exhaustion/immune checkpoints (e.g., IL10, LAG3, CD27, TIGIT, CTLA4, EOMES, TBX21, PDCD1), and the immunotolerance-inducing enzyme indoleamine 2,3-dioxygenase 1 (IDO1), some of which have been associated with autoimmunity.95-97 Whole-exome and targeted sequencing studies identified a variety of somatic mutations, including those involving epigenetic modifiers (e.g., CDC27, TET2, KMT2C, KMT2D, ASXL1, BAZ2A, ARID1B), the PI3K/AKT/mTOR signaling path-

way (e.g., MTOR, PIK3CB, PIK3CD), the JAK/STAT pathway (e.g., JAK3, STAT3), and other immune response pathways (e.g., PLCG2, CBL, IDH1).98-102 Loss (deletion or loss of heterozygosity) and mutations of the tumor suppressor NAV3 have been reported in 44%94 and 10-15% of cases,100,101 respectively. Recent studies demonstrated a high frequency of predominantly biallelic missense mutations in HAVCR2, which encodes the protein T-cell immunoglobulin mucin 3 (TIM3).93,99,100,102 The incidence among SPTCL patients ranged from 25% in a European study93 to 85% within an Asian cohort.102 Variants include p.Y82C (most common and enriched in Asian individuals), p.I97M, and p.T101I. These mutations were germline (mostly homozygous or com-

Haematologica | 108 December 2023

3273

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL pound heterozygous) among patients in whom the germline could be assessed.92,99,102 While HAVCR2-mutated (HAVCR2MUT) SPTCL patients have typical clinical and histological features, they are reported to present at a younger age, suffer from more severe disease, including higher rates of hemophagocytic syndrome, and require more intensive therapy.93,99,100 Testing for HAVCR2 mutation in the clinical setting has been recommended to potentially identify patients at higher risk of aggressive disease and/or hemophagocytic syndrome who may benefit from more definitive therapy.87,93 TIM3 is a transmembrane receptor expressed in certain innate immune cells, including subsets of T/NK cells and macrophages. It acts as a negative immune checkpoint, terminating immune responses through interactions with ligands. HAVCR2 mutations result in TIM3 protein misfolding which impairs its normal localization to the cell surface, a phenotype that can be seen with immunohistochemistry or flow cytometry92,93,99 (Figure 5F). It is suggested that loss of normal TIM3 function leads to uncontrolled immune activation and excessive cytokine release, potentially promoting SPTCL along with development of hemophagocytic syndrome. Compared to cases with wild-type HAVCR2 (HAVCR2WT), HAVCR2Y82C SPTCL is enriched in genes involved in inflammation-associated cellular pathways, including IL6-JAKSTAT3 signaling and tumor necrosis factor-α signaling via NF-κB, consistent with enhanced inflammatory responses.100 HAVCR2WT SPTCL demonstrates upregulation of genes associated with lymphocyte homing (CCR4, GPR183) and autoimmunity (STAB2) and more frequently harbors several gene mutations (ASXL1, CAPN1, UNC13D, PIAS3, PIK3CD, KMT2D, BRD2), some of which (e.g., PIAS3) may function to deregulate immune pathways in the absence of deleterious HAVCR2 mutations. HAVCR2WT SPTCL also shows increased CCR4 expression, a chemokine receptor that regulates T regulatory cell homing in skin, and higher numbers of CCR4+ and FoxP3+ cells in the microenvironment, suggesting that loss of intact CCR4-mediated T regulatory cell activity may propagate unchecked inflammation within HAVCR2MUT SPTCL. Despite the available data, the pathogenesis of SPTCL remains unclear. The concomitance of SPTCL with clinical, serological, and/or histological features of autoimmune disease, such as systemic lupus erythematosus, and/or indeterminate or overlapping histopathological features in some patients led to speculation that both disorders may co-occur, with either autoimmune disease predisposing patients to malignancy via immune dysregulation or SPTCL

N.E. Lewis et al.

inducing autoimmune phenomena, or lie along a biological spectrum.87,103-105 Reports of SPTCL arising after other immune-activating events, such as infection and vaccination, further suggests that immune stimulation may trigger Tcell dysregulation leading to the development of SPTCL.89,102,106 Studies showing frequent expression of immune pathways, including those seen in autoimmune disorders, as well as the disease’s indolent course that largely lacks extracutaneous dissemination and responds to immunomodulation have further linked the pathobiology of immune dysregulation and SPTCL. The discovery of frequent HAVCR2 germline mutations in SPTCL patients at rates significantly higher than in the general population revealed a strong germline risk and furthered the understanding of SPTCL pathogenesis towards a disease of abnormal immune activation that fails to control a clonal T-cell outgrowth.87,99 However, given the report of an unaffected patient with a known homozygous germline HAVCR2 mutation,99 these mutations alone may not be sufficient for disease development. Additionally, the detection of additional genomic alterations in significant proportions of HAVCR2MUT tumors as well as the development of SPTCL in HAVCR2WT individuals, some of whom have underlying diseases such as systemic lupus erythematosus or infections, suggests that additional genomic alterations and/or biological triggers may be involved in the pathogenesis.

Conclusion The rarity of mature extranodal T-cell and NK-cell lymphomas and lymphoproliferative disorders makes their diagnosis, study, and biological understanding challenging. However, through the use of ever-advancing genomic and single-cell analytical techniques, insight into their pathogenesis continues to grow, and with it better opportunities to effectively diagnose, treat, and hopefully cure this unique group of challenging diseases. Disclosures AD reports having received personal fees for consultancy from Incyte, Loxo and EUS Pharma and research support from Roche and Takeda. NEL and TG have no conflicts of interest to disclose. Contributions NEL, TZ, and AD designed the manuscript, which NEL and TZ wrote. AD revised the manuscript. All authors approved the final version.

References 1. Stuver R, Epstein-Peterson Z, Horwitz S. Few and far between:

clinical management of rare extranodal subtypes of mature T-

Haematologica | 108 December 2023

3274

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL cell and NK-cell lymphomas. Haematologica. 2023;108(12)3244-3260. 2. Campo E, Jaffe ES, Cook JR, et al. The International Consensus Classification of Mature Lymphoid Neoplasms: a report from the Clinical Advisory Committee. Blood. 2022;140(11):1229-1253. 3. Rubio-Tapia A, Hill ID, Semrad C, et al. American College of Gastroenterology guidelines update: diagnosis and management of celiac disease. Am J Gastroenterol. 2023;118(1):59-76. 4. Cheminant M, Bruneau J, Malamut G, et al. NKp46 is a diagnostic biomarker and may be a therapeutic target in gastrointestinal T-cell lymphoproliferative diseases: a CELAC study. Gut. 2019;68(8):1396-1405. 5. Soderquist CR, Lewis SK, Gru AA, et al. Immunophenotypic spectrum and genomic landscape of refractory celiac disease type II. Am J Surg Pathol. 2021;45(7):905-916. 6. Malamut G, Afchain P, Verkarre V, et al. Presentation and longterm follow-up of refractory celiac disease: comparison of type I with type II. Gastroenterology. 2009;136(1):81-90. 7. Ilus T, Kaukinen K, Virta LJ, et al. Refractory coeliac disease in a country with a high prevalence of clinically-diagnosed coeliac disease. Aliment Pharmacol Ther. 2014;39(4):418-425. 8. Cording S, Lhermitte L, Malamut G, et al. Oncogenetic landscape of lymphomagenesis in coeliac disease. Gut. 2022;71(3):497-508. 9. Ettersperger J, Montcuquet N, Malamut G, et al. Interleukin-15dependent T-cell-like innate intraepithelial lymphocytes develop in the intestine and transform into lymphomas in celiac disease. Immunity. 2016;45(3):610-625. 10. Tack GJ, van Wanrooij RL, Langerak AW, et al. Origin and immunophenotype of aberrant IEL in RCDII patients. Mol Immunol. 2012;50(4):262-270. 11. Bhagat G, Isaacson P. Enteropathy-associated T-cell lymphoma and other primary intestinal T-cell lymphomas. In: Jaffe ES, Arber DA, Campo E, Harris NL, Quintanilla-Martinez L, eds. Hematopathology. 2nd edition. Philadelphia (PA): Elsevier; 2016. p. 693-711. 12. Malamut G, Chandesris O, Verkarre V, et al. Enteropathy associated T cell lymphoma in celiac disease: a large retrospective study. Dig Liver Dis. 2013;45(5):377-384. 13. Sieniawski M, Angamuthu N, Boyd K, et al. Evaluation of enteropathy-associated T-cell lymphoma comparing standard therapies with a novel regimen including autologous stem cell transplantation. Blood. 2010;115(18):3664-3670. 14. Farstad IN, Johansen FE, Vlatkovic L, et al. Heterogeneity of intraepithelial lymphocytes in refractory sprue: potential implications of CD30 expression. Gut. 2002;51(3):372-378. 15. Moffitt AB, Ondrejka SL, McKinney M, et al. Enteropathyassociated T cell lymphoma subtypes are characterized by loss of function of SETD2. J Exp Med. 2017;214(5):1371-1386. 16. Malamut G, El Machhour R, Montcuquet N, et al. IL-15 triggers an antiapoptotic pathway in human intraepithelial lymphocytes that is a potential new target in celiac disease-associated inflammation and lymphomagenesis. J Clin Invest. 2010;120(6):2131-2143. 17. Mention JJ, Ben Ahmed M, Bègue B, et al. Interleukin 15: a key to disrupted intraepithelial lymphocyte homeostasis and lymphomagenesis in celiac disease. Gastroenterology. 2003;125(3):730-745. 18. Tse E, Gill H, Loong F, et al. Type II enteropathy-associated Tcell lymphoma: a multicenter analysis from the Asia Lymphoma Study Group. Am J Hematol. 2012;87(7):663-668. 19. Hang JF, Yuan CT, Chang KC, et al. Targeted next-generation sequencing reveals a wide morphologic and immunophenotypic spectrum of monomorphic epitheliotropic intestinal T-cell

N.E. Lewis et al.

lymphoma. Am J Surg Pathol. 2022;46(9):1207-1218. 20. Veloza L, Cavalieri D, Missiaglia E, et al. Monomorphic epitheliotropic intestinal T-cell lymphoma comprises morphologic and genomic heterogeneity impacting outcome. Haematologica. 2023;108(1):181-195. 21. Tan SY, Ooi AS, Ang MK, et al. Nuclear expression of MATK is a novel marker of type II enteropathy-associated T-cell lymphoma. Leukemia. 2011;25(3):555-557. 22. Nairismägi ML, Tan J, Lim JQ, et al. JAK-STAT and G-proteincoupled receptor signaling pathways are frequently altered in epitheliotropic intestinal T-cell lymphoma. Leukemia. 2016;30(6):1311-1319. 23. Deleeuw RJ, Zettl A, Klinker E, et al. Whole-genome analysis and HLA genotyping of enteropathy-type T-cell lymphoma reveals 2 distinct lymphoma subtypes. Gastroenterology. 2007;132(5):1902-1911. 24. Ko YH, Karnan S, Kim KM, et al. Enteropathy-associated T-cell lymphoma--a clinicopathologic and array comparative genomic hybridization study. Hum Pathol. 2010;41(9):1231-1237. 25. Roberti A, Dobay MP, Bisig B, et al. Type II enteropathyassociated T-cell lymphoma features a unique genomic profile with highly recurrent SETD2 alterations. Nat Commun. 2016;7:12602. 26. Huang D, Lim JQ, Cheah DMZ, et al. Whole-genome sequencing reveals potent therapeutic strategy for monomorphic epitheliotropic intestinal T-cell lymphoma. Blood Adv. 2020;4(19):4769-4774. 27. Molenaar TM, van Leeuwen F. SETD2: from chromatin modifier to multipronged regulator of the genome and beyond. Cell Mol Life Sci. 2022;79(6):346. 28. Nicolae A, Xi L, Pham TH, et al. Mutations in the JAK/STAT and RAS signaling pathways are common in intestinal T-cell lymphomas. Leukemia. 2016;30(11):2245-2247. 29. Mutzbauer G, Maurus K, Buszello C, et al. SYK expression in monomorphic epitheliotropic intestinal T-cell lymphoma. Mod Pathol. 2018;31(3):505-516. 30. Alaggio R, Amador C, Anagnostopoulos I, et al. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: lymphoid neoplasms. Leukemia. 2022;36(7):1720-1748. 31. Margolskee E, Jobanputra V, Lewis SK, Alobeid B, Green PH, Bhagat G. Indolent small intestinal CD4+ T-cell lymphoma is a distinct entity with unique biologic and clinical features. PLoS One. 2013;8(7):e68343. 32. Jaffe ES, Chott A, Ott G, et al. Indolent T-cell lymphoproliferative disorder of the gastrointestinal tract. In: Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri S, Stein H, et al., eds. WHO classification of Tumours of Haematopoietic and Lymphoid Tissues. Revised 4th edition. Lyon (France): International Agency for Research on Cancer; 2017. p. 379-380. 33. Soderquist CR, Patel N, Murty VV, et al. Genetic and phenotypic characterization of indolent T-cell lymphoproliferative disorders of the gastrointestinal tract. Haematologica. 2020;105(7):1895-1906. 34. Sharma A, Oishi N, Boddicker RL, et al. Recurrent STAT3-JAK2 fusions in indolent T-cell lymphoproliferative disorder of the gastrointestinal tract. Blood. 2018;131(20):2262-2266. 35. Dargent JL, Tinton N, Trimech M, de Leval L. Lymph node involvement by enteropathy-like indolent NK-cell proliferation. Virchows Arch. 2021;478(6):1197-1202. 36. Mansoor A, Pittaluga S, Beck PL, Wilson WH, Ferry JA, Jaffe ES. NK-cell enteropathy: a benign NK-cell lymphoproliferative disease mimicking intestinal lymphoma: clinicopathologic features and follow-up in a unique case series. Blood.

Haematologica | 108 December 2023

3275

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL 2011;117(5):1447-1452. 37. Xiao W, Gupta GK, Yao J, et al. Recurrent somatic JAK3 mutations in NK-cell enteropathy. Blood. 2019;134(12):986-991. 38. Foss FM, Horwitz SM, Civallero M, et al. Incidence and outcomes of rare T cell lymphomas from the T Cell Project: hepatosplenic, enteropathy associated and peripheral gamma delta T cell lymphomas. Am J Hematol. 2020;95(2):151-155. 39. Vose J, Armitage J, Weisenburger D. International peripheral Tcell and natural killer/T-cell lymphoma study: pathology findings and clinical outcomes. J Clin Oncol. 2008;26(25):4124-4130. 40. Belhadj K, Reyes F, Farcet JP, et al. Hepatosplenic gammadelta T-cell lymphoma is a rare clinicopathologic entity with poor outcome: report on a series of 21 patients. Blood. 2003;102(13):4261-4269. 41. Falchook GS, Vega F, Dang NH, et al. Hepatosplenic gammadelta T-cell lymphoma: clinicopathological features and treatment. Ann Oncol. 2009;20(6):1080-1085. 42. Weidmann E. Hepatosplenic T cell lymphoma. A review on 45 cases since the first report describing the disease as a distinct lymphoma entity in 1990. Leukemia. 2000;14(6):991-997. 43. Yabe M, Medeiros LJ, Tang G, et al. Prognostic factors of hepatosplenic T-cell lymphoma: clinicopathologic study of 28 cases. Am J Surg Pathol. 2016;40(5):676-688. 44. Thai A, Prindiville T. Hepatosplenic T-cell lymphoma and inflammatory bowel disease. J Crohns Colitis. 2010;4(5):511-522. 45. Nicolae A, Xi L, Pittaluga S, et al. Frequent STAT5B mutations in γδ hepatosplenic T-cell lymphomas. Leukemia. 2014;28(11):2244-2248. 46. Yabe M, Miranda RN, Medeiros LJ. Hepatosplenic T-cell lymphoma: a review of clinicopathologic features, pathogenesis, and prognostic factors. Hum Pathol. 2018;74:5-16. 47. Przybylski GK, Wu H, Macon WR, et al. Hepatosplenic and subcutaneous panniculitis-like γ/δ T cell lymphomas are derived from different Vδ subsets of γ/δ T lymphocytes. J Mol Diagn. 2000;2(1):11-19. 48. Macon WR, Levy NB, Kurtin PJ, et al. Hepatosplenic alphabeta Tcell lymphomas: a report of 14 cases and comparison with hepatosplenic gammadelta T-cell lymphomas. Am J Surg Pathol. 2001;25(3):285-296. 49. McKinney M, Moffitt AB, Gaulard P, et al. The genetic basis of hepatosplenic T-cell lymphoma. Cancer Discov. 2017;7(4):369-379. 50. Finalet Ferreiro J, Rouhigharabaei L, Urbankova H, et al. Integrative genomic and transcriptomic analysis identified candidate genes implicated in the pathogenesis of hepatosplenic T-cell lymphoma. PLoS One. 2014;9(7):e102977. 51. Alonsozana EL, Stamberg J, Kumar D, et al. Isochromosome 7q: the primary cytogenetic abnormality in hepatosplenic gammadelta T cell lymphoma. Leukemia. 1997;11(8):1367-1372. 52. Travert M, Huang Y, de Leval L, et al. Molecular features of hepatosplenic T-cell lymphoma unravels potential novel therapeutic targets. Blood. 2012;119(24):5795-5806. 53. Song W, Zhang H, Yang F, et al. Single cell profiling of γδ hepatosplenic T-cell lymphoma unravels tumor cell heterogeneity associated with disease progression. Cell Oncol (Dordr). 2023;46(1):211-226. 54. Küçük C, Jiang B, Hu X, et al. Activating mutations of STAT5B and STAT3 in lymphomas derived from γδ-T or NK cells. Nat Commun. 2015;6:6025. 55. Bergmann AK, Fataccioli V, Castellano G, et al. DNA methylation profiling of hepatosplenic T-cell lymphoma. Haematologica. 2019;104(3):e104-e107. 56. Tripodo C, Iannitto E, Florena AM, et al. Gamma-delta T-cell

N.E. Lewis et al.

lymphomas. Nat Rev Clin Oncol. 2009;6(12):707-717. 57. Deepak P, Sifuentes H, Sherid M, Stobaugh D, Sadozai Y, Ehrenpreis ED. T-cell non-Hodgkin's lymphomas reported to the FDA AERS with tumor necrosis factor-alpha (TNF-α) inhibitors: results of the REFURBISH study. Am J Gastroenterol. 2013;108(1):99-105. 58. Kelsen J, Dige A, Schwindt H, et al. Infliximab induces clonal expansion of γδ-T cells in Crohn's disease: a predictor of lymphoma risk? PLoS One. 2011;6(3):e17890. 59. Yabe M, Medeiros LJ, Daneshbod Y, et al. Hepatosplenic T-cell lymphoma arising in patients with immunodysregulatory disorders: a study of 7 patients who did not receive tumor necrosis factor-α inhibitor therapy and literature review. Ann Diagn Pathol. 2017;26:16-22. 60. Laurini JA, Perry AM, Boilesen E, et al. Classification of nonHodgkin lymphoma in Central and South America: a review of 1028 cases. Blood. 2012;120(24):4795-4801. 61. Hong M, Lee T, Young Kang S, Kim SJ, Kim W, Ko YH. Nasal-type NK/T-cell lymphomas are more frequently T rather than NK lineage based on T-cell receptor gene, RNA, and protein studies: lineage does not predict clinical behavior. Mod Pathol. 2016;29(5):430-443. 62. Dong G, Liu X, Wang L, et al. Genomic profiling identifies distinct genetic subtypes in extra-nodal natural killer/T-cell lymphoma. Leukemia. 2022;36(8):2064-2075. 63. Huang Y, de Reyniès A, de Leval L, et al. Gene expression profiling identifies emerging oncogenic pathways operating in extranodal NK/T-cell lymphoma, nasal type. Blood. 2010;115(6):1226-1237. 64. Iqbal J, Kucuk C, Deleeuw RJ, et al. Genomic analyses reveal global functional alterations that promote tumor growth and novel tumor suppressor genes in natural killer-cell malignancies. Leukemia. 2009;23(6):1139-1151. 65. Nakashima Y, Tagawa H, Suzuki R, et al. Genome-wide arraybased comparative genomic hybridization of natural killer cell lymphoma/leukemia: different genomic alteration patterns of aggressive NK-cell leukemia and extranodal Nk/T-cell lymphoma, nasal type. Genes Chromosomes Cancer. 2005;44(3):247-255. 66. Chen YW, Guo T, Shen L, et al. Receptor-type tyrosine-protein phosphatase κ directly targets STAT3 activation for tumor suppression in nasal NK/T-cell lymphoma. Blood. 2015;125(10):1589-1600. 67. Karube K, Nakagawa M, Tsuzuki S, et al. Identification of FOXO3 and PRDM1 as tumor-suppressor gene candidates in NK-cell neoplasms by genomic and functional analyses. Blood. 2011;118(12):3195-3204. 68. Ng SB, Selvarajan V, Huang G, et al. Activated oncogenic pathways and therapeutic targets in extranodal nasal-type NK/T cell lymphoma revealed by gene expression profiling. J Pathol. 2011;223(4):496-510. 69. Kim H, Ko YH. The pathologic and genetic characteristics of extranodal NK/T-cell lymphoma. Life (Basel). 2022;12(1):73. 70. Montes-Mojarro IA, Chen BJ, Ramirez-Ibarguen AF, et al. Mutational profile and EBV strains of extranodal NK/T-cell lymphoma, nasal type in Latin America. Mod Pathol. 2020;33(5):781-791. 71. Bouchekioua A, Scourzic L, de Wever O, et al. JAK3 deregulation by activating mutations confers invasive growth advantage in extranodal nasal-type natural killer cell lymphoma. Leukemia. 2014;28(2):338-348. 72. Küçük C, Hu X, Jiang B, et al. Global promoter methylation analysis reveals novel candidate tumor suppressor genes in natural killer cell lymphoma. Clin Cancer Res.

Haematologica | 108 December 2023

3276

REVIEW SERIES - Biology and genetics of extranodal mature TNKCL 2015;21(7):1699-1711. 73. Bi XW, Wang H, Zhang WW, et al. PD-L1 is upregulated by EBVdriven LMP1 through NF-κB pathway and correlates with poor prognosis in natural killer/T-cell lymphoma. J Hematol Oncol. 2016;9(1):109. 74. Kataoka K, Miyoshi H, Sakata S, et al. Frequent structural variations involving programmed death ligands in Epstein-Barr virus-associated lymphomas. Leukemia. 2019;33(7):1687-1699. 75. Lim JQ, Huang D, Tang T, et al. Whole-genome sequencing identifies responders to pembrolizumab in relapse/refractory natural-killer/T cell lymphoma. Leukemia. 2020;34(12):3413-3419. 76. Song TL, Nairismägi ML, Laurensia Y, et al. Oncogenic activation of the STAT3 pathway drives PD-L1 expression in natural killer/T-cell lymphoma. Blood. 2018;132(11):1146-1158. 77. Kim SJ, Lim JQ, Laurensia Y, et al. Avelumab for the treatment of relapsed or refractory extranodal NK/T-cell lymphoma: an open-label phase 2 study. Blood. 2020;136(24):2754-2763. 78. Xiong J, Cui BW, Wang N, et al. Genomic and transcriptomic characterization of natural killer T cell lymphoma. Cancer Cell. 2020;37(3):403-419.e6. 79. Cho J, Kim SJ, Park WY, et al. Immune subtyping of extranodal NK/T-cell lymphoma: a new biomarker and an immune shift during disease progression. Mod Pathol. 2020;33(4):603-615. 80. Li Z, Xia Y, Feng LN, et al. Genetic risk of extranodal natural killer T-cell lymphoma: a genome-wide association study. Lancet Oncol. 2016;17(9):1240-1247. 81. Lin GW, Xu C, Chen K, et al. Genetic risk of extranodal natural killer T-cell lymphoma: a genome-wide association study in multiple populations. Lancet Oncol. 2020;21(2):306-316. 82. Tian XP, Ma SY, Young KH, et al. A composite single-nucleotide polymorphism prediction signature for extranodal natural killer/T-cell lymphoma. Blood. 2021;138(6):452-463. 83. Suzumiya J, Ohshima K, Takeshita M, et al. Nasal lymphomas in Japan: a high prevalence of Epstein-Barr virus type A and deletion within the latent membrane protein gene. Leuk Lymphoma. 1999;35(5-6):567-578. 84. Peng RJ, Han BW, Cai QQ, et al. Genomic and transcriptomic landscapes of Epstein-Barr virus in extranodal natural killer Tcell lymphoma. Leukemia. 2019;33(6):1451-1462. 85. de Mel S, Hue SS, Jeyasekharan AD, Chng WJ, Ng SB. Molecular pathogenic pathways in extranodal NK/T cell lymphoma. J Hematol Oncol. 2019;12(1):33. 86. Li L, Ma BBY, Chan ATC, Chan FKL, Murray P, Tao Q. Epstein-Barr virus-induced epigenetic pathogenesis of viral-associated lymphoepithelioma-like carcinomas and natural killer/T-cell lymphomas. Pathogens. 2018;7(3):63. 87. Guitart J, Mangold AR, Martinez-Escala ME, et al. Clinical and pathological characteristics and outcomes among patients with subcutaneous panniculitis-like T-cell lymphoma and related adipotropic lymphoproliferative disorders. JAMA Dermatol. 2022;158(10):1167-1174. 88. Kong YY, Dai B, Kong JC, et al. Subcutaneous panniculitis-like Tcell lymphoma: a clinicopathologic, immunophenotypic, and molecular study of 22 Asian cases according to WHO-EORTC classification. Am J Surg Pathol. 2008;32(10):1495-1502. 89. Michonneau D, Petrella T, Ortonne N, et al. Subcutaneous panniculitis-like T-cell lymphoma: immunosuppressive drugs induce better response than polychemotherapy. Acta Derm Venereol. 2017;97(3):358-364. 90. Willemze R, Jansen PM, Cerroni L, et al. Subcutaneous panniculitis-like T-cell lymphoma: definition, classification, and prognostic factors: an EORTC Cutaneous Lymphoma Group Study of 83 cases. Blood. 2008;111(2):838-845.

N.E. Lewis et al.

91. Gao J, Gauerke SJ, Martinez-Escala ME, et al. Bone marrow involvement by subcutaneous panniculitis-like T-cell lymphoma: a report of three cases. Mod Pathol. 2014;27(6):800-807. 92. Wegehaupt O, Groß M, Wehr C, et al. TIM-3 deficiency presenting with two clonally unrelated episodes of mesenteric and subcutaneous panniculitis-like T-cell lymphoma and hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer. 2020;67(6):e28302. 93. Sonigo G, Battistella M, Beylot-Barry M, et al. HAVCR2 mutations are associated with severe hemophagocytic syndrome in subcutaneous panniculitis-like T-cell lymphoma. Blood. 2020;135(13):1058-1061. 94. Hahtola S, Burghart E, Jeskanen L, et al. Clinicopathological characterization and genomic aberrations in subcutaneous panniculitis-like T-cell lymphoma. J Invest Dermatol. 2008;128(9):2304-2309. 95. Machan S, Rodríguez M, Alonso-Alonso R, et al. Subcutaneous panniculitis-like T-cell lymphoma, lupus erythematosus profundus, and overlapping cases: molecular characterization through the study of 208 genes. Leuk Lymphoma. 2021;62(9):2130-2140. 96. Li Z, Wang H, Dong R, et al. Single-cell RNA-seq reveals characteristics of malignant cells and immune microenvironment in subcutaneous panniculitis-like T-cell lymphoma. Front Oncol. 2021;11:611580. 97. Maliniemi P, Hahtola S, Ovaska K, et al. Molecular characterization of subcutaneous panniculitis-like T-cell lymphoma reveals upregulation of immunosuppression- and autoimmunity-associated genes. Orphanet J Rare Dis. 2014;9:160. 98. Fernandez-Pol S, Costa HA, Steiner DF, et al. High-throughput sequencing of subcutaneous panniculitis-like T-cell lymphoma reveals candidate pathogenic mutations. Appl Immunohistochem Mol Morphol. 2019;27(10):740-748. 99. Gayden T, Sepulveda FE, Khuong-Quang DA, et al. Germline HAVCR2 mutations altering TIM-3 characterize subcutaneous panniculitis-like T cell lymphomas with hemophagocytic lymphohistiocytic syndrome. Nat Genet. 2018;50(12):1650-1657. 100. Koh J, Jang I, Mun S, et al. Genetic profiles of subcutaneous panniculitis-like T-cell lymphoma and clinicopathological impact of HAVCR2 mutations. Blood Adv. 2021;5(20):3919-3930. 101. Li Z, Lu L, Zhou Z, et al. Recurrent mutations in epigenetic modifiers and the PI3K/AKT/mTOR pathway in subcutaneous panniculitis-like T-cell lymphoma. Br J Haematol. 2018;181(3):406-410. 102. Polprasert C, Takeuchi Y, Kakiuchi N, et al. Frequent germline mutations of HAVCR2 in sporadic subcutaneous panniculitislike T-cell lymphoma. Blood Adv. 2019;3(4):588-595. 103. Bosisio F, Boi S, Caputo V, et al. Lobular panniculitic infiltrates with overlapping histopathologic features of lupus panniculitis (lupus profundus) and subcutaneous T-cell lymphoma: a conceptual and practical dilemma. Am J Surg Pathol. 2015;39(2):206-211. 104. Magro CM, Crowson AN, Kovatich AJ, Burns F. Lupus profundus, indeterminate lymphocytic lobular panniculitis and subcutaneous T-cell lymphoma: a spectrum of subcuticular Tcell lymphoid dyscrasia. J Cutan Pathol. 2001;28(5):235-247. 105. Pincus LB, LeBoit PE, McCalmont TH, et al. Subcutaneous panniculitis-like T-cell lymphoma with overlapping clinicopathologic features of lupus erythematosus: coexistence of 2 entities? Am J Dermatopathol. 2009;31(6):520-526. 106. Kreher MA, Ahn J, Werbel T, Motaparthi K. Subcutaneous panniculitis-like T-cell lymphoma after COVID-19 vaccination. JAAD Case Rep. 2022;28:18-20.

Haematologica | 108 December 2023

3277

ARTICLE - Acute Lymphoblastic Leukemia

Effect of two additional doses of intrathecal methotrexate during induction therapy on serious infectious toxicity in pediatric patients with acute lymphoblastic leukemia Janina Heilmann,1* Simon Vieth,1* Anja Möricke,1 Andishe Attarbaschi,2 Draga Barbaric,3,4 Nicole Bodmer,5 Antonella Colombini,6 Luciano Dalla-Pozza,7 Sarah Elitzur,8 Shai Izraeli,8 Georg Mann,2 Felix Niggli,5 Daniela Silvestri,6,9 Jan Stary,10 Carmelo Rizzari,6 Maria Grazia Valsecchi,9 Ester Zapotocka,10 Martin Zimmermann,11 Gunnar Cario,1# Martin Schrappe1# and Valentino Conter6# Department of Pediatrics, Christian-Albrechts-University Kiel and University Medical Center Schleswig-Holstein, Kiel, Germany; 2Department of Pediatrics, St. Anna Children's Cancer Research Institute and St. Anna Children's Hospital, Medical University School, Vienna, Austria; 3Children's Cancer Institute Australia, University of New South Wales, Lowy Cancer Centre, Randwick, Australia; 4Kids Cancer Centre, Sydney Children's Hospital, Randwick, Australia; 5Department of Pediatric Oncology, University Children's Hospital, Zurich, Switzerland; 6Pediatric Clinic, University Milano-Bicocca-Fondazione MBBM / San Gerardo Hospital, Monza, Italy; 7The Children's Hospital at Westmead, Department of Oncology, Westmead, Australia; 8Pediatric Hematology-Oncology, Schneider Children's Medical Center and Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel; 9University of MilanoBicocca, Center of Biostatistics for Clinical Epidemiology, Department of Health Science, Milan, Italy; 10University Hospital Motol, Department of Pediatric Hematology/Oncology, Prague, Czech Republic and 11Hannover Medical School, Pediatric Hematology and Oncology, Hannover, Germany

Correspondence: Simon Vieth simon.vieth@uksh.de Received: Accepted: Early view:

August 19, 2022. March 30, 2023. April 6, 2023.

JH and SV contributed equally as first authors. GC, MS and VC contributed equally as senior authors.

Abstract Although initial central nervous system (CNS) involvement is rarely detected in childhood acute lymphoblastic leukemia (ALL), risk-adapted CNS-directed therapy is essential for all patients. Treatment intensity depends on the initial CNS status. In the AIEOP-BFM ALL 2009 trial, patients with cytomorphologic detection of leukemic blasts in initial cerebrospinal fluid were classified as CNS2 or CNS3 and received five intrathecal doses of methotrexate (MTX) in induction therapy compared to patients with CNS1 status (no blasts detected) who received three doses. The impact of additional intrathecal (IT) MTX on systemic toxicity in induction therapy is unknown. Between June 1st 2010 and February 28th 2017, a total of 6,136 ALL patients aged 1-17 years were enrolled onto the AIEOP-BFM ALL 2009 trial. The effect of three versus five doses of IT MTX during induction therapy on the incidence of severe infectious complications was analyzed. Among 4,706 patients treated with three IT MTX doses, 77 (1.6%) had a life-threatening infection during induction as compared to 59 of 1,350 (4.4%) patients treated with five doses (P<0.001; Odds Ratio 2.86 [95% Confidence Interval 1.99-4.13]). In a multivariate regression model, treatment with additional IT MTX proved to be the strongest risk factor for life-threatening infections (Odds Ratio 2.85 [1.96-4.14]). Fatal infections occurred in 16 (0.3%) and 38 (1.6%) patients treated with three or five IT MTX doses, respectively (P<0.001). As the relevance of additional intrathecal MTX in induction for relapse prevention in CNS2 patients is unclear, doses of intrathecal therapy have been reduced for these patients. (Clinicaltrials.gov identifiers: NCT01117441 and NCT00613457).

Introduction Involvement of the central nervous system (CNS) can be detected in about 3-5% of patients at initial diagnosis of acute lymphoblastic leukemia (ALL) and in 30-40% of pa-

tients at ALL relapse.1,2 Without CNS-directed therapy, relapses originating from the CNS can be expected in up to 75%.3 For a sustainable therapy for children with ALL a stratified prophylactic and therapeutic CNS-directed therapy is, therefore, indispensable. Introduction of cranial

Haematologica | 108 December 2023

3278

ARTICLE - Toxicity in intrathecal MTX in pediatric ALL

J. Heilmann et al.

radiotherapy (CRT) and intrathecal drugs have improved the outcome of CNS disease.4 Due to the long-term toxicity of CRT and steadily improving cure rates in ALL, CRT has mainly been replaced by systemic and intrathecal CNS-directed chemotherapy.5-9 In the AIEOP-BFM ALL 2009 trial, CNS status was determined by the number of nucleated cells and the presence of blasts in initial CSF before start of chemotherapy, and the presence of clinical and imaging findings of CNS disease. Patients with cytomorphologic detection of blasts in the initial CSF cytospin were classified as CNS2 status if the number of nucleated cells was ≤5/μL, whereas patients with higher cell counts and detection of blasts in the initial CSF cytospin were classified as CNS3. Due to the observation of a higher risk of relapse in patients with CNS2 status at the time of initial diagnostic workup, the study group from St. Jude Children’s Research Hospital postulated the need for a more intensive intrathecal therapy for these patients.10,11 Based on the data of Mahmoud et al., intensified therapy with two additional intrathecal doses of methotrexate (MTX) in the induction phase was introduced for patients with CNS2 status in the ALL-BFM 95 study of the Berlin-Frankfurt-Münster ALL group compared to patients with CNS1 status who received three doses. Analysis of the ALL-BFM 95 trial demonstrated that patients with CNS2 status had the same event-free survival (EFS) as patients with CNS1 status.12 Thus, patients with CNS2 as well as those with CNS3 status continued to receive two additional intrathecal doses of IT MTX in induction therapy in subsequent AIEOP-BFM ALL trials. There are few data on the effect of additional intrathecal treatment in induction chemotherapy on adverse events,

especially on severe infectious complication,13 and, to our knowledge, no data have been published on the effect of two additional IT MTX doses in induction on adverse infectious events. Therefore, we performed a retrospective analysis to investigate the toxic effects of two additional IT MTX doses in induction chemotherapy in the AIEOPBFM ALL 2009 trial.

Methods Patients and study design Between June 1st 2010 and February 28th 2017, a total of 6,136 ALL patients under 18 years of age were enrolled onto the AIEOP-BFM ALL 2009 trial (registered at EudraCT n. 2007004270-43). Patients were treated in participating study centers in Austria, Australia, the Czech Republic, Germany, Israel, Italy and Switzerland. Informed consent was obtained from the guardians of all patients. Eighty of the patients enrolled onto the study were not evaluable for analysis for the following reasons: only one additional dose of IT MTX in induction was administered (n=40), no information about additional IT MTX was available (n=24), or patients died before day 19, the day on which the first additional dose of IT MTX was scheduled per protocol (n=16) (Figure 1). Cytomorphologic evaluation of CSF cytospin preparations was carried out at the local treatment center; an additional central review of the cytospin preparation was performed in 56% of the cases. Patients without blasts in CSF were classified as having CNS1 status independent of the CSF cell count. Patients with non-traumatic lumbar puncture and detection of blasts on the diagnostic CSF cytospin were classified as

Figure 1. Consolidated standards for reporting of trials (CONSORT) diagram. add. : additional; IT: intrathecal; MTX: methotrexate; n: number. Haematologica | 108 December 2023

3279

ARTICLE - Toxicity in intrathecal MTX in pediatric ALL

J. Heilmann et al.

either CNS2a if the number of nucleated cells was ≤5/μL or CNS3a if the number of nucleated cells was >5/μL. Patients with traumatic lumbar puncture and detection of blasts were classified as having CNS2b status if the nucleated cell count was ≤5/μL or as either CNS2c or CNS3b (depending on the estimated level of blood contamination) if the nucleated cell count was >5/μL. (For details see the Online Supplementary Appendix). The study protocol was approved by the competent ethics committees of the national co-ordinating centers and by local ethics committees where required. Treatment Patients were treated according to the AIEOP-BFM ALL 2009 protocol. The treatment plan of the induction phase Protocol IA is shown in Table 1. According to the protocol, IT MTX was administered on days 1, 12 and 33 for patients classified as CNS1, and on days 1, 12, 19, 26 and 33 for patients classified as CNS2 or CNS3 during induction phase Protocol IA. Documentation of life-threatening infections Data on serious adverse events were regularly collected in the AIEOP-BFM ALL 2009 study according to the regulatory requirements and were classified as infectious events based on the information provided by the investigators. The assessment of life-threatening status was made by the investigator and centrally reviewed by qualified persons in the national study co-ordinating centers of the participating groups according to the ICH Harmonised Tripartite Guideline. An adverse event was considered as life-threatening if its occurrence placed the patient at immediate risk of death. A severe adverse event that might have caused death

if it had occurred in a more severe form, was not considered as life-threatening, as long as the patient was not at immediate risk of death. Details of the definition of lifethreatening are provided in Online Supplementary Table S2. Statistical analysis All analyses were performed on the basis of the number of IT MTX doses the patient had actually received. Differences in the distribution of individual parameters among patient 2 subsets were analyzed using the χ or Fisher’s exact test for categorized variables and the Mann-Whitney U test for continuous variables. The association between rates of adverse events and prognostic factors was examined by univariate and multivariate logistic regression analysis to calculate Odds Ratios (OR) and their 95% Confidence Intervals (CI). To describe the impact of CNS status on relapse risk, Cox proportional hazard model was used for uni- and multivariate analysis. Differences in the distribution of categorical variables were analyzed using the Fisher’s exact test. Analyses were carried out using SAS version 9.4.

Results Patients' characteristics A total of 6,056 patients were eligible for this analysis. Of these patients, CNS status was not evaluable in 335, mostly due to lack of sufficient diagnostic material or because the diagnostic spinal tap was delayed >72 hours after start of prednisone according to protocol. Among evaluable patients, 4,516 had CNS1 (74.5%), 1,071 CNS2 (17.6%), and 134 CNS3 (2.2%) status. IT MTX treatment was administered in deviation from the protocol in 41 patients

Table 1. Induction treatment in the AIEOP-BFM ALL 2009 study. Treatment phase/drug Prephase: Prednisone (PO/IV) Methotrexate (IT) Induction: Protocol IA Prednisone/prednisolone (PO/IV) or Dexamethasone (PO/IV)c Vincristine (IV) Daunorubicin (PI over 1 h) Cyclophosphamide (PI over 1 h)f PEG-L-asparaginase (PI over 2 h) Methotrexate (IT)

Single or daily dose

Days of application per phasea

60 mg/m2/d Age-adjustedb

1-7 1

60 mg/m2/d

8-28d

10 mg/m2/d 1.5 mg/m2/dose (max. 2 mg) 30 mg/m2/dose 1000 mg/m2/dose 2500 IU/m2/dose (max. 3750 IU) Age-adjustedb

8-28d 8, 15, 22, 29 8, 15, 22, 29e 10 12, 26 12, 33g

Adjustments of time schedule were allowed if clinical condition and bone marrow recovery were inadequate. bAge-adjusted dose: 1 to <2 years, 8 mg; 2 to <3 years, 10 mg; ≥3 years, 12 mg. cDexamethasone only for patients with T-cell acute lymphoblastic leukemia (T-ALL) and prednisone good response (PGR). dSteroids were tapered over 9 additional days. eRandomization: only two doses on days (d) 8 and 15 were given to patients randomized into the experimental arm in randomization R1. fOnly for T-ALL / prednisone-poor response patients. gAdditional intrathecal (IT) methotrexate (MTX) therapy on day 19 and 26 was administered to patients with CN2 and CNS3 status. PO: oral administration; IV: intravenous; PI: infusion; d: day; h: hour.

Haematologica | 108 December 2023

3280

ARTICLE - Toxicity in intrathecal MTX in pediatric ALL

J. Heilmann et al.

with CNS1 status and 33 patients with CNS2 or CNS3 status; therefore, 4,706 patients were treated with three and 1,350 with five doses of IT MTX in induction (Figure 1). Table 2 shows the initial patients' characteristics and the type of Protocol IA according to the intrathecal doses given. There was a significantly higher proportion of patients who received additional IT MTX among male patients, older patients (>10 years), T-lineage, ETV6-RUNX1 negativity, with high initial white blood cell (WBC) count (≥50x109/L), and patients with ≥10% blasts in FCM-MRD at day (d) 15. Accordingly, it was higher in patients who received one of the two T-ALL protocol variants, i.e., “IA Dexa” (dexamethasone instead of prednisolone) or protocol variant “IA-CPM” (with additional cyclophosphamide for high-risk T-ALL patients). Incidences of life-threatening and fatal infections according to number of intrathecal methotrexate doses Incidences of life-threatening and fatal infections in induction phase are shown in Table 3 with reference to patients' characteristics and the number of IT MTX doses in induction. Females showed a higher incidence of lifethreatening infections (2.7%) compared to males (1.9%), although this did not reach statistical significance (P=0.053). A statistically significantly higher incidence of

life-threatening infections was seen in older patients (3.6% in patients aged ≥10 years vs. 1.8% in patients <10 years; P<0.001), in patients with T-lineage ALL (3.6% vs. 2.0% in B-lineage ALL; P=0.005), ETV6-RUNX1 negativity (2.5% vs. 1.4% in patients with ETV6-RUNX1 positivity; P=0.012) and in patients assigned to the treatment with Protocol IA-Dexa (4.4% vs. 2.0% or 2.2% in patients assigned to Protocol IA/IA’ or IA-CPM, respectively; P=0.004). Patients aged ≥10 years of age in addition had a significantly higher incidence of fatal infections (1.5%) compared to younger patients (0.3%; P<0.001). Patients who received five doses of IT MTX had a statistically significantly higher incidence of life-threatening infections (4.4%) compared to patients with three doses of MTX (1.6%; P<0.001). Fatal infections appeared in 1.6% in patients with additional IT MTX versus 0.3% without additional IT MTX (P<0.001). Patients with CNS2 and CNS3 status had a statistically significantly higher incidence of life-threatening infections in induction (4.4% and 3.0%, respectively) compared to patients with CNS1 status (1.6%; P<001). There was no significant difference in the incidence of life-threatening infections between patients with CNS2 and those with CNS3 status (P=0.45). In addition, we analyzed the incidence of life-threatening infections in the consolidation phase which followed in-

Table 2. Proportion of patients with additional intrathecal methotrexate in induction related to different initial patients' characteristics and (as assigned) type of Protocol IA. No additional IT MTX* N (%)

Additional IT MTX* N (%)

Gender Male Female

2,651 (75.9) 2,055 (80.1)

840 (24.1) 510 (19.9)

<0.001

Age in years < 10 ≥ 10

3,590 (78.3) 1,116 (75.8)

993 (21.7) 357 (24.2)

0.04

Initial WBC, x109/L < 50 ≥ 50

3,989 (80.9) 715 (63.5)

939 (19.1) 411 (36.5)

<0.001

Immunophenotype B-lineage T-lineage

4,151 (80.1) 538 (63.3)

1,034 (19.9) 312 (36.7)

<0.001

ETV6-RUNX1 Negative Positive

3,538 (75.9) 1,098 (83.2)

1,125 (24.1) 222 (16.8)

<0.001

FCM-MRD d15 <10% blasts ≥10% blasts

4,010 (78.7) 570 (74.1)

1,085 (21.3) 199 (25.9)

0.005

Type of Protocol IA IA-Dexa IA/IA’ IA-CPM

373 (68.2) 4,151 (80.1) 165 (54.5)

174 (31.8) 1,034 (19.9) 138 (45.5)

<0.001

*Data refer to patients with successful investigation of the respective criteria. IT: intrathecal; MTX: methotrexate; N: number; WBC: white blood count; FCM-MRD: minimal residual disease by flow cytometry; d: day; Dexa: dexamethasone; IA/IA’ : prednisone/prednisolone with 4 or 2 doses (IA’) of daunorubicin in Protocol IA; CPM: cyclophosphamide.

Haematologica | 108 December 2023

3281

ARTICLE - Toxicity in intrathecal MTX in pediatric ALL

J. Heilmann et al.

duction Protocol IA. Treatment in this phase was independent of the CNS status. Incidences of life-threatening infection in consolidation phase were 0.78% (35/4516), 0.84% (9/1071), and 0.75% (1/134) in patients with CNS1, CNS2 and CNS3 status, respectively. In a multivariate logistic regression model including gender, age, WBC, ETV6-RUNX1 status, minimal residual disease by flow cytometry (FCM-MRD) d15, type of Protocol IA and the intrathecal MTX doses in induction as covariates, female gender, age ≥10 years, Protocol IA-Dexa, and the treatment with additional IT MTX showed independent significance on the risk of life-threatening infection in induction, with the highest effect for the treatment with additional IT MTX doses among the parameters analyzed (OR 2.85 [95% CI 1.96-4.14]; P<0.001) (Table 4). Detailed data of subgroups with life-threatening and fatal

infections in induction therapy according to the number of intrathecal doses of IT MTX related to different patients' characteristics are provided in Online Supplementary Table S3. To analyze the impact of CNS status on relapse risk, multivariate cause-specific Cox regression analyses on relapse incidence including gender, age (</≥10 years), WBC (</≥ 50x109/L), risk group (high-risk [HR] / non-HR), ETV6-RUNX1 rearrangement (for precursor B-cell ALL [pB-ALL] only), and CNS status (CNS1 [reference) / CNS2/CNS3] as co-variates were performed separately for pB-ALL and T-cell ALL (TALL). Hazard ratios for CNS2 status were 1.13 (95% CI: 0.921.39; P=0.25; n=4,919) or 1.10 (0.66-1.84; P=0.71; n=738) in patients with pB-ALL or T-ALL, respectively. The hazard ratios for CNS3 status were 1.59 (95% CI: 0.85-3.0; P=0.15) in B-ALL and 2.65 (95% CI: 1.56-4.51; P<0.001) in T-ALL.

Table 3. Life-threatening and fatal infections in induction therapy related to patients' characteristics, type of Protocol IA (as assigned), and number of intrathecal methotrexate administrations. Life-threatening infections in induction Total

Yes

Fatal infections in induction P

Total

6,056

100

5,920

97.8

136

2.2

Gender Male Female

3,491 2,565

100 100

3,424 2,469

98.1 97.3

67 69

Age in years <10 ≥10

4,583 1,473

100 100

4,500 1,420

98.2 96.4

Initial WBC, x109/L <50 ≥50

4,928 1,126

100 100

4,823 1,095

CNS status CNS1 CNS2 CNS3

4,516 1,071 134

100 100 100

Immunophenotype B-lineage T-lineage

5,185 850

ETV6-RUNX1 Negative Positive

Yes

6,018

99.4

0.6

1.9 2.7

0.053

3,470 2,548

99.4 99.3

21 17

0.6 0.7

0.87

83 53

1.8 3.6

<0.001

4,567 1,451

99.7 99.5

16 22

0.3 1.5

<0.001

98.1 97.2

105 31

2.1 2.8

0.22

4,901 1,115

99.5 99.0

27 11

0.5 1.0

0.14

4,442 1,024 130

98.4 95.6 97.0

74 47 4

1.6 4.4 3.0

<0.001

4,500 1,055 132

99.6 98.5 98.5

16 16 2

0.4 1.5 1.5

<0.001

100 100

5,081 819

98.0 96.4

104 31

2.0 3.6

0.005

5,155 842

99.4 99.1

30 8

0.6 0.9

0.24

4,663 1,320

100 100

4,546 1,302

97.5 98.6

117 18

2.5 1.4

0.012

4,630 1,315

99.3 99.6

33 5

0.7 0.4

0.24

FCM-MRD d15 <10% blasts ≥10% blasts

5,095 769

100 100

4,989 751

97.9 97.7

106 18

2.1 2.3

0.58

5,068 763

99.5 99.2

27 6

0.5 0.8

0.42

Type of Protocol IA IA-Dexa IA/IA’ IA-CPM

547 5,185 303

100 100 100

523 5,081 269

95.6 98.0 97.7

24 104 7

4.4 2.0 2.3

0.004

543 5,155 299

99.3 99.4 98.7

4 30 4

0.7 0.6 1.3

0.18

Add. IT MTX in P IA No Yes

4,706 1,350

100 100

4,629 1,291

98.4 95.6

77 59

1.6 4.4

<0.001

4,690 1,328

99.7 98.4

16 22

0.3 1.6

<0.001

N: number; WBC: white blood count; CNS: central nervous system; FCM-MRD: minimal residual disease by flow cytometry; d: day; Dexa: dexamethasone; IA/IA’: prednisone/prednisolone with 4 or 2 doses (IA’) of daunorubicin in Protocol CPM: cyclophosphamide; IT: intrathecal; MTX: methotrexate; Add.: additional. Haematologica | 108 December 2023

3282

ARTICLE - Toxicity in intrathecal MTX in pediatric ALL

J. Heilmann et al.

Discussion Our data show a highly significant impact of two additional IT MTX doses in induction therapy on the incidence of lifethreatening and fatal infections. These additional doses were indicated to be given to patients with CNS2 or CNS3 status at initial diagnosis and were omitted in patients with CNS1 status. Almost all patients had eventually received the IT therapy in induction in accordance with the protocol. This high correlation between the CNS status and IT therapy made it difficult to discriminate between whether the higher risk of severe infection is caused by the intensified IT therapy or is a specific feature of the CNS involvement itself. The multivariate logistic regression analysis did not allow inclusion of the status due to the high correlation, but it showed that the adverse effect of additional IT MTX was largely independent of the other relevant patient-, leukemia-, and therapy-related parameters included in the model. We conclude from the data that it is most likely that the additional IT therapy is the decisive risk factor rather than the feature “leukemia with CNS involvement”. This conclusion was supported by the finding of similar incidences of life-threatening infections in patients with CNS1, CNS2 and CNS3 status in the subsequent consolidation phase.

There were no significant differences in interim analyses on relapse incidence between CNS1 and CNS2 status. However, the vast majority of patients with CNS2 status had received two additional doses of IT MTX. The erstwhile introduction of intensified IT MTX in induction for CNS2 in ALL-BFM therapy was based on observations in protocols other than the ALL-BFM protocol and, like other therapy modifications, was continued in subsequent ALLBFM protocols. Given the data now available on adverse events, we have to question the impact of the additional IT MTX on ALL-BFM therapy. Severe infectious complications are the main cause for early death, mortality, and treatment delay in induction therapy in childhood ALL.14,15 Risk factors for infectious complications have been described14-17 and can be divided into factors relating to patients' characteristics (e.g., age), to disease (e.g., high initial WBC), or to treatment (e.g., dexamethasone therapy). So far, single therapeutic interventions like two additional IT MTX applications have not been included in risk factor analysis. Published data on the impact of 2-4 additional intrathecal treatments (MTX/cytarabine/hydrocortisone) in induction by the St. Jude Study group did not show any significant difference in the rates of grade 4 or 5 infections or grade 2-4 seiz-

Table 4. Uni- and multivariate logistic regression analyses on life-threatening infections in Protocol IA. Univariate analysis

Multivariate analysis

Odds Ratio

95% CI

Odds Ratio

95% CI

Gender Male Female

1 1.39

0.97-1.97

0.075

1 1.65

1.14-2.37

0.008

Age in years <10 ≥10

1 1.99

1.38-2.89

<0.001

1 1.80

1.21-2.66

0.003

Initial WBC, x109/L <50 ≥50

1 1.23

0.81-1.9

0.33

1 0.86

0.54-1.37

0.51

Immunophenotype B-lineage T-lineage

1 1.63

1.08-2.63

0.019

ETV6-RUNX1 Negative Positive

1 0.53

0.31-0.89

0.018

1 0.69

0.40-1.21

0.201

FCM-MRD d15 <10% blasts ≥10% blasts

1 1.14

0.69-1.91

0.57

1 0.99

0.57-1.65

0.96

Protocol IA-Dexa No Yes

1 2.12

1.31-3.41

0.002

1 1.75

1.05-2.97

0.034

Add. IT MTX in P IA No Yes

1 2.86

1.99-4.13

<0.001

1 2.85

1.96-4.14

< 0.001

Excluded from model*

WBC: white blood count; FCM-MRD: minimal residual disease by flow cytometry; Dexa: dexamethasone; IT: intrathecal; MTX: methotrexate; CI: Confidence Interval; d: day; Add.: additional. *Immunophenotype was not included in multivariate model due to high correlation with type of Protocol IA (P IA). Haematologica | 108 December 2023

3283

ARTICLE - Toxicity in intrathecal MTX in pediatric ALL

J. Heilmann et al.

ures.13 In contrast to treatment under the AIEOP-BFM ALL 2009 study protocol, patients in the St. Jude study received triple IT treatment (ITT) and leucovorin rescue (5 mg/m2 per dose, max. 5 mg) which was administered orally at 24 and 30 hours after each ITT during induction. Besides the smaller size of the cohort and other differences in treatment and stratification compared to the AIEOP-BFM ALL 2009 protocol, the leucovorin rescue may have protected patients against an additional risk of infection. Moreover, the overall number of grade 4 and 5 infections was slightly higher in the St. Jude study compared to the incidence of infections defined as lifethreatening and fatal infections in the AIEOP-BFM 2009 trial (8.6% vs. 5.9%, respectively). Evidently there is a systemic therapeutic effect of IT MTX in childhood ALL. This was demonstrated for response to prednisone pre-phase plus IT-MTX, with a significantly higher rate of good response to prednisone in patients with pB-ALL who received IT MTX on d1 as compared to those who did not receive IT MTX before d7.18,19 Only limited data are available on systemic side effects of IT MTX, but IT MTX administration results in greater systemic exposure compared to oral administration of the same dose.20 The CSF, with its blood-brain barrier, seems to act as an MTX reservoir with the ability to prolong systemic MTX exposure in a bi-exponential manner.21-23 In addition, IT MTX can cause acute tumor lysis syndrome after a single administration.24-26 Together, these observations lead to the widely accepted assumption that IT MTX has a systemic effect, which in turn may be associated with severe adverse events, as our study has now clearly demonstrated. When discussing the consequences of the observation that two additional IT MTX doses in patients with CNS2 status severely increase the risk of severe and lifethreatening infections, several aspects have to be taken into account. 1) The use of systemic leucovorin rescue in IT MTX overdose turned out to be beneficial, as described in case series.27,28 However, data are scarce on the impact of systemic leucovorin rescue in regular IT MTX administration, not only with respect to toxicity, but also on a plausible systemic antileukemic effect. There is a lack of systematic study data for leucovorin rescue in IV and IT MTX therapy that could answer this question. In general, the frequency of IV MTX-induced oral mucositis was described to decrease when the leucovorin dose was increased, and that response was related to the dose.29 2) Cytomorphology as the sole method of CSF assessment is still insufficient, as it is prone to preanalytical error and observer-derived vagueness.30,31 Interim analyses of data of the AIEOP-BFM ALL 2009 trial revealed relevant differences in the proportion of patients with CNS2 status between the participating groups (3.3-35.1%; data not shown). The reason for these differences is not

entirely clear. Different approaches regarding centralized or local cytospin assessment along with various technical reasons may contribute to the variation observed. It is, therefore, questionable whether CNS2 status with its assumed heterogeneity is a good parameter on which to justify treatment modification. A more precise leukemia detection in the CSF at diagnosis and during treatment is needed, e.g., by applying highly sensitive molecular genetic or flow cytometric detection methods. 3) In order to describe the prognostic potential of these new methods, treatment outcome should not be biased by treatment modification such as the application of two additional IT MTX doses. 4) In most patients, lumbar puncture is carried out in some kind of analog-sedation or anesthesia. This procedure in itself harbors the risk of various minor and major adverse events,32 including infection and cardio-respiratory incidents, entailing the need for further medical interventions with their own rates of adverse events, including infections. 5) The acute and late neurotoxicity and cognitive impairment associated with and caused by IT MTX has to be mentioned33-35 as a further reason to scrutinize the number of IT MTX treatments. Nevertheless, the ultimate goal of contemporary protocols should be the avoidance of CRT wherever possible. Therefore, intrathecal therapy in induction was de-escalated in the subsequent currently ongoing trial AIEOP-BFM ALL 2017: patients with CNS2 status are no longer treated with additional IT MTX in induction, whereas the approach was not changed for patients with CNS3. An increased risk of relapse in CNS2 patients cannot be completely ruled out after omitting the two intrathecal administrations in induction. This potential risk may also vary depending on the underlying treatment protocol, particularly, but not exclusively, with respect to CNS-directed treatment elements such as the total number of intrathecal administrations, high-dose MTX and cranial radiotherapy. Ultimately, we have to weigh an unknown benefit of the two additional IT MTX administrations in terms of relapse incidence against the strong evidence of a significantly increased risk of serious and potentially fatal infectious complications. Another potential risk might be an increase in the incidence of non-remission due to resistant CNS disease at the end of induction as a consequence of reduced IT induction therapy. In the AIEOP-BFM ALL 2009 study, this was an extremely rare event concerning less than 1 out of 5,000 patients. Additional IT doses, given individually to patients who still have evidence of leukemic blasts in the CSF at the time of the second therapeutic lumbar puncture on d12 may be considered in order to minimize the risk of higher incidence of resistant CNS disease. The uniform CNS-directed treatment of patients with CNS1 and CNS2 status forms a basis for prospective studies evaluating the prognostic relevance of the CNS2 status and of additional methods that are more reliable

Haematologica | 108 December 2023

3284

ARTICLE - Toxicity in intrathecal MTX in pediatric ALL

J. Heilmann et al.

than the cytomorphological CSF assessment. The ongoing AIEOP-BFM ALL 2017 trial addresses these questions. Disclosures AM has provided consultancy services for Clinigen and has received consultancy honoraria from BTG. AA has received honoraria for lectures, consultancy or advisory board participation from JazzPharma, Amgen, Novartis, MSD, Pfizer and Gilead, and compensation for travel expenses from JazzPharma and Sehas, has provided consultancy services for Amgen, and has received honoraria from Novartis and Medison Pharma. CR has received honoraria for lectures, presentations, or educational events from Amgen, Jazz, Servier, Clinigen and Serb. GC and/or study group have received research support from Amgen, JazzPharma, Novartis and Servier. GC has received travel support or speaker‘s honoraria from Amgen, JazzPharma, Novartis and Servier. MS and/or study group have received research support from SHIRE, JazzPharma, Servier, SigmaTau and Novartis. MS has received honoraria from Servier, Novartis, and JazzPharma. All of the other authors have no conflicts of interest to disclose. Contributions MS, VC, MZ, MGV, AB, GM and FN were involved in designing and planning the study. AM, JH, SV, GC and MS wrote the manuscript. JH, SV, AA, DB, NB, AC, LD, SE, SI, GM, FN, DS, JS, EZ and MS helped in collecting the data and provided patients for the study. MGV and MZ were the study statis-

ticians. MGV, MZ, AM, JH, SV and DS oversaw data checking and reporting during the study period, and analyzed the data. All authors contributed to the data interpretation, drafting, and revision of the manuscript, and gave final approval to submit the manuscript for publication. Acknowledgments The authors would like to thank all patients and families who participated in this trial, the physicians and nurses of all hospitals for their contribution in performing this study, and the study committees for productive discussions during the development and progress of the trial. We thank the partners in the reference laboratories and all the technicians for their expert work in cytology, genetics and MRD diagnostics, and the data managers for their careful study management. Funding For AIEOP, conduct of the study was supported by Comitato ML Verga, and Fondazione Tettamanti (Monza), Fondazione Città della Speranza, Fondazione Cariparo (Padova). For BFM, conduct of the study was supported by Deutsche Krebshilfe e.V., Bonn, Germany (grant 50-2698 Schr1 and grant 50-2410 Ba7), Oncosuisse/Krebsforschung Schweiz (grant OCS 1230-02-2002), and St. Anna Kinderkrebsforschung Austria. Data-sharing statement Queries regarding data sharing should be addressed to the corresponding author.

References 1. Frishman-Levy L, Izraeli S. Advances in understanding the pathogenesis of CNS acute lymphoblastic leukaemia and potential for therapy. Br J Haematol. 2017;176(2):157-167. 2. Pui CH, Howard SC. Current management and challenges of malignant disease in the CNS in paediatric leukaemia. Lancet Oncol. 2008;9(3):257-268. 3. Evans AE, Gilbert ES, Zandstra R. The increasing incidence of central nervous system leukemia in children. (Children's Cancer Study Group A). Cancer. 1970;26(2):404-409. 4. Hustu HO, Aur RJ, Verzosa MS, Simone JV, Pinkel D. Prevention of central nervous system leukemia by irradiation. Cancer. 1973;32(3):585-597. 5. Reiter A, Schrappe M, Ludwig, et al. Chemotherapy in 998 unselected childhood acute lymphoblastic leukemia patients. Results and conclusions of the multicenter trial ALL-BFM 86. Blood. 1994;84(9):3122-3133. 6. Conter V, Schrappe M, Arico M, et al. Role of cranial radiotherapy for childhood T-cell acute lymphoblastic leukemia with high WBC count and good response to prednisone. Associazione Italiana Ematologia Oncologia Pediatrica and the BerlinFrankfurt-Munster groups. J Clin Oncol. 1997;15(8):2786-2791. 7. Schrappe M, Reiter A, Henze G, et al. Prevention of CNS recurrence in childhood ALL: results with reduced radiotherapy combined with CNS-directed chemotherapy in four consecutive

ALL-BFM trials. Klin Padiatr. 1998;210(4):192-199. 8. Conter V, Arico M, Valsecchi MG, et al. Long-term results of the Italian Association of Pediatric Hematology and Oncology (AIEOP) acute lymphoblastic leukemia studies, 1982-1995. Leukemia. 2000;14(12):2196-2204. 9. Moricke A, Reiter A, Zimmermann M, et al. Risk-adjusted therapy of acute lymphoblastic leukemia can decrease treatment burden and improve survival: treatment results of 2169 unselected pediatric and adolescent patients enrolled in the trial ALL-BFM 95. Blood. 2008;111(9):4477-4489. 10. Mahmoud HH, Rivera GK, Hancock ML, et al. Low leukocyte counts with blast cells in cerebrospinal fluid of children with newly diagnosed acute lymphoblastic leukemia. N Engl J Med. 1993;329(5):314-319. 11. Gajjar A, Harrison PL, Sandlund JT, et al. Traumatic lumbar puncture at diagnosis adversely affects outcome in childhood acute lymphoblastic leukemia. Blood. 2000;96(10):3381-3384. 12. Burger B, Zimmermann M, Mann G, et al. Diagnostic cerebrospinal fluid examination in children with acute lymphoblastic leukemia: significance of low leukocyte counts with blasts or traumatic lumbar puncture. J Clin Oncol. 2003;21(2):184-188. 13. Jeha S, Pei D, Choi J, et al. Improved CNS control of childhood acute lymphoblastic leukemia without cranial irradiation: St

Haematologica | 108 December 2023

3285

ARTICLE - Toxicity in intrathecal MTX in pediatric ALL

J. Heilmann et al.

Jude Total Therapy Study 16. J Clin Oncol. 2019;37(35):3377-3391. 14. Moricke A, Zimmermann M, Valsecchi MG, et al. Dexamethasone vs prednisone in induction treatment of pediatric ALL: results of the randomized trial AIEOP-BFM ALL 2000. Blood. 2016;127(17):2101-2112. 15. O'Connor D, Bate J, Wade R, et al. Infection-related mortality in children with acute lymphoblastic leukemia: an analysis of infectious deaths on UKALL2003. Blood. 2014;124(7):1056-1061. 16. Li MJ, Chang HH, Yang YL, et al. Infectious complications in children with acute lymphoblastic leukemia treated with the Taiwan Pediatric Oncology Group protocol: a 16-year tertiary single-institution experience. Pediatr Blood Cancer. 2017;64(10). 17. Inaba H, Pei D, Wolf J, et al. Infection-related complications during treatment for childhood acute lymphoblastic leukemia. Ann Oncol. 2017;28(2):386-392. 18. Hasegawa D, Manabe A, Ohara A, et al. The utility of performing the initial lumbar puncture on day 8 in remission induction therapy for childhood acute lymphoblastic leukemia: TCCSG L99-15 study. Pediatr Blood Cancer. 2012;58(1):23-30. 19. Thyss A, Suciu S, Bertrand Y, et al. Systemic effect of intrathecal methotrexate during the initial phase of treatment of childhood acute lymphoblastic leukemia. The European Organization for Research and Treatment of Cancer Children's Leukemia Cooperative Group. J Clin Oncol. 1997;15(5):1824-1830. 20. Bostrom BC, Erdmann GR, Kamen BA. Systemic methotrexate exposure is greater after intrathecal than after oral administration. J Pediatr Hematol Oncol. 2003;25(2):114-117. 21. Bleyer WA. The clinical pharmacology of methotrexate: new applications of an old drug. Cancer. 1978;41(1):36-51. 22. Shapiro WR, Young DF, Mehta BM. Methotrexate: distribution in cerebrospinal fluid after intravenous, ventricular and lumbar injections. N Engl J Med. 1975;293(4):161-166. 23. Goldie JH, Price LA, Harrap KR. Methotrexate toxicity: correlation with duration of administration, plasma levels, dose and excretion pattern. Eur J Cancer. 1972;8(4):409-414. 24. Simmons ED, Somberg KA. Acute tumor lysis syndrome after intrathecal methotrexate administration. Cancer. 1991;67(8):2062-2065. 25. Benekli M, Gullu IH, Savas MC, et al. Acute tumor lysis syndrome following intrathecal methotrexate. Leuk Lymphoma. 1996;22(3-4):361-363. 26. Konuma T, Ooi J, Takahashi S, et al. Fatal acute tumor lysis

syndrome following intrathecal chemotherapy for acute lymphoblastic leukemia with meningeal involvement. Intern Med. 2008;47(22):1987-1988. 27. Malbora B, Ozyurek E, Kocum AI, Ozbek N. Delayed recognition of intrathecal methotrexate overdose. J Pediatr Hematol Oncol. 2009;31(5):352-354. 28. Riva L, Conter V, Rizzari C, Jankovic M, Sala A, Milani M. Successful treatment of intrathecal methotrexate overdose with folinic acid rescue: a case report. Acta Paediatr. 1999;88(7):780-782. 29. Van der Beek JN, Oosterom N, Pieters R, de Jonge R, van den Heuvel-Eibrink MM, Heil SG. The effect of leucovorin rescue therapy on methotrexate-induced oral mucositis in the treatment of paediatric ALL: a systematic review. Crit Rev Oncol Hematol. 2019;142:1-8. 30. Del Principe MI, Buccisano F, Cefalo M, et al. High sensitivity of flow cytometry improves detection of occult leptomeningeal disease in acute lymphoblastic leukemia and lymphoblastic lymphoma. Ann Hematol. 2014;93(9):1509-1513. 31. Crespo-Solis E, Lopez-Karpovitch X, Higuera J, Vega-Ramos B. Diagnosis of acute leukemia in cerebrospinal fluid (CSF-acute leukemia). Curr Oncol Rep. 2012;14(5):369-378. 32. Patel MM, Kamat PP, McCracken CE, Simon HK. Complications of deep sedation for individual procedures (lumbar puncture alone) versus combined procedures (lumbar puncture and bone marrow aspirate) in pediatric oncology patients. Hosp Pediatr. 2016;6(2):95-102. 33. Cheung YT, Sabin ND, Reddick WE, et al. Leukoencephalopathy and long-term neurobehavioural, neurocognitive, and brain imaging outcomes in survivors of childhood acute lymphoblastic leukaemia treated with chemotherapy: a longitudinal analysis. Lancet Haematol. 2016;3(10):e456-e466. 34. Duffner PK, Armstrong FD, Chen L, et al. Neurocognitive and neuroradiologic central nervous system late effects in children treated on Pediatric Oncology Group (POG) P9605 (standard risk) and P9201 (lesser risk) acute lymphoblastic leukemia protocols (ACCL0131): a methotrexate consequence? A report from the Children's Oncology Group. J Pediatr Hematol Oncol. 2014;36(1):8-15. 35. Bhojwani D, Sabin ND, Pei D, et al. Methotrexate-induced neurotoxicity and leukoencephalopathy in childhood acute lymphoblastic leukemia. J Clin Oncol. 2014;32(9):949-959.

Haematologica | 108 December 2023

3286

ARTICLE - Acute Lymphoblastic Leukemia

Impact of central nervous system involvement in adult patients with Philadelphia-negative acute lymphoblastic leukemia: a GRAALL-2005 study Corentin Orvain,1,2,3 Sylvain Chantepie,4 Xavier Thomas,5 Martine Escoffre-Barbe,6 Françoise Huguet,7 Yohan Desbrosses,8 Gaelle Guillerm,9 Madalina Uzunov,10 Thibaut Leguay,11 Sarah Barbieux,12 Norbert Vey,13 Patrice Chevallier,14 Jean-Valère Malfuson,15 Stéphane Lepretre,16 Michael Baumann,17,18 Murat Aykut,18,19 Abdelaziz Chaib,20 Magalie Joris,21 Hacène Zerazhi,22 Georg Stussi,18,23 Jacques Chapiro,24 Céline Berthon,25 Caroline Bonmati,26 Eric Jourdan,27 Diana Carp,28 Ambroise Marcais,29 Maria-Pilar Gallego-Hernanz,30 Iona Vaida,31 Karin Bilger,32 Alban Villate,33 Florence Pasquier,34 Yves Chalandon,18,35 Sébastien Maury,36 Véronique Lheritier,37 Norbert Ifrah,1,2,3 Hervé Dombret,38 Nicolas Boissel38# and Mathilde Hunault-Berger1,2,3# on behalf of the Group for Research on Adult Acute Lymphoblastic Leukemia (GRAALL, including the former France-Belgium Group or Lymphoblastic Acute Leukemia in Adults, LALA), the French WesternEastern Group for Lymphoblastic Acute Leukemia (GOELAL), and the Swiss Group for Clinical Cancer Research (SAKK)

Correspondence: M. Hunault-Berger mahunault@chu-angers.fr Received: Accepted: Early view:

October 25, 2022. February 28, 2023. March 9, 2023.

Maladies du Sang, CHU d’Angers, Angers, France; 2Fedération Hospitalo-Universitaire GrandOuest Acute Leukemia, FHU-GOA, France; 3Université d'Angers, Inserm UMR 1307, CNRS UMR 6075, Nantes Université, CRCI2NA, F-49000 Angers, France; 4Institut d’Hématologie, CHU de Caen, Caen, France; 5Hématologie Clinique, HCL, Centre Hospitalier Lyon Sud, Pierre Bénite, France; 6Hématologie Clinique, CHU de Rennes, Rennes, France; 7Hématologie, Centre Hospitalo-Universitaire de Toulouse, Institut Universitaire du Cancer de Toulouse-Oncopole, Toulouse-Oncopole, Toulouse, France; 8Hématologie Clinique, CHU de Besançon, Besançon, France; 9Hématologie Clinique, CHRU de Brest, Brest, France; 10Hématologie, Hôpital de la Pitié - Salpêtrière, Paris, France; 11Hématologie Clinique, Hôpital du Haut-Lévêque, CHU de Bordeaux, Pessac, France; 12Hématologie Clinique, Centre Hospitalier de Dunkerque, Dunkerque, France; 13 Hématologie Clinique, Institut Paoli-Calmettes, Marseille, France; 14Hématologie Clinique, CHU de Nantes, Nantes, France; 15Hématologie Clinique, Hôpital d’Instruction des Armées, Percy, France; 16Département d’Hématologie, Centre Henri-Becquerel, Rouen, France; 17Klinik für Med. Onkologie und Hämatologie, Kantonsspital St. Gallen, St. Gallen, Switzerland; 18Swiss Group for Clinical Cancer Research (SAKK), Bern, Switzerland; 19Klinik für Medizinische Onkologie und Hämatologie, Universitätsspital Zürich, Zürich, Switzerland; 20Hémato-Oncologie et Médecine Interne, Centre Hospitalier du Pays d’Aix, Aix-en-Provence, France; 21Hématologie Clinique, CHU d’Amiens, Amiens, France; 22Hématologie Clinique, Centre Hospitalier Henri Duffaut, Avignon, France; 23Clinica di Ematologia, Istituto Oncologico della Svizzera Italiana, Bellinzona, Switzerland; 24Onco-Hématologie, Hôpitaux Civils de Colmar, Colmar, France; 25Maladies du Sang, CHU de Lille, Lille, France; 26Service d’Hématologie, CHRU de Nancy, Nancy, France; 27 Hématologie Clinique, CHU de Nîmes, Nîmes, France; 28Oncologie Médicale, Centre Hospitalier d’Orléans, Orléans, France; 29Hématologie Clinique, Hôpital Necker, AP-HP, Paris, France; 30 Hématologie Clinique, CHU de Poitiers, Poitiers, France; 31Onco-Hématologie, Centre Hospitalier René-Dubos, Pontoise, France; 32Oncologie et Hématologie, Institut de Cancérologie Strasbourg Europe (ICANS), Strasbourg, France; 33Hématologie et Thérapie Cellulaire, CHRU de Tours, Tours, France; 34Département d'Hématologie, Gustave Roussy, Université Paris-Saclay, Villejuif, France; 35Department of Oncology, Hematology Division, University Hospital of Geneva and Faculty of Medicine of Geneva, Geneva, Switzerland; 36Département d’Hématologie, Assistance Publique-Hôpitaux de Paris (AP-HP), Hôpital Henri Mondor, Créteil, France; 37 Coordination du groupe GRAALL, Centre Hospitalier Lyon Sud, Pierre Bénite, France and 38 Hématologie Adulte, Hôpital Saint-Louis, AP-HP, Paris, France #

NB and MH-B contributed equally as senior authors.

Abstract Whereas the prognosis of adult patients with Philadelphia-negative acute lymphoblastic leukemia (ALL) has greatly improved since the advent of pediatric-inspired regimens, the impact of initial central nervous system (CNS) involvement has not been formerly re-evaluated. We report here the outcome of patients with initial CNS involvement included in the Haematologica | 108 December 2023

3287

ARTICLE - Initial CNS involvement in adults with ALL

C. Orvain et al.

pediatric-inspired prospective randomized GRAALL-2005 study. Between 2006 and 2014, 784 adult patients (aged 18-59 years) with newly diagnosed Philadelphia-negative ALL were included, of whom 55 (7%) had CNS involvement. In CNSpositive patients, overall survival was shorter (median 1.9 years vs. not reached, HR=1.8 [1.3-2.6], P<0.001). While there was no statistical difference in cumulative incidence of relapse between CNS+ and CNS- patients (HR=1.5 [0.9-2.5], P=0.11), non-relapse mortality was significantly higher in those with initial CNS disease (HR=2.1 [1.2-3.5], P=0.01). This increase in toxicity was mostly observed in patients randomized to the high-dose cyclophosphamide arm and in those who received allogeneic stem cell transplantation. Exploratory landmark analyses did not show any association between either cranial irradiation or allogeneic stem cell transplantation and outcome. Despite improved outcome in young adult ALL patients with pediatric-inspired protocols, CNS involvement is associated with a worse outcome mainly due to excess toxicity, without improved outcome with allogeneic SCT.

Introduction Acute lymphoblastic leukemia (ALL) often involves the central nervous system (CNS), which is considered a sanctuary for leukemic cells. It has been recently shown that ALL cells could use neural migratory pathways to invade the CNS and adapt to the CNS microenvironment by modifying metabolic pathways.1,2 As relapse can develop from occult CNS disease, CNS-directed therapies such as high-dose methotrexate and intrathecal chemotherapy are mandatory in all patients with ALL whether they have CNS involvement or not.3 At diagnosis, 4-11% of patients with ALL have CNS involvement.4-8 These patients are more likely to have other extramedullary site involvement, T-cell ALL, and a higher white blood cell (WBC) count at diagnosis.4,6 In adults with ALL, the prognosis of initial CNS involvement is controversial, but has been associated with higher rates of primary resistant disease and early death in one series, and with decreased overall survival in another.4,6 The role of allogeneic stem cell transplantation (SCT) was also controversial in these two historical series.4,6 Whereas more recent studies have shown that CNS involvement is associated with a worse outcome in children with ALL,9,10 no study has re-evaluated the impact of CNS involvement in adults in more recent years during which time they are increasingly receiving pediatric-inspired protocols.11 In comparison to conventional adult protocols, these pediatric-inspired regimens, which include higher doses of corticosteroids, vincristine, asparaginase, and increased CNS-directed therapy with lower doses of other cytotoxic chemotherapy, have greatly improved the outcome of adult patients.7,8,11-15 Meanwhile, indications for allogeneic SCT have changed, relying more on measurable residual disease (MRD).16,17 We report here the association of CNS involvement with survival in adults with ALL included in the pediatric-inspired prospective GRAALL-2005 study.

Methods Patients and settings The Group for Research on Adult Acute Lymphoblastic

Leukemia 2005 trial (GRAALL-2005) was conducted between 2006 and 2014 at 57 French, 8 Belgian, and 8 Swiss centers. This randomized trial evaluated the impact of hyperfractionated cyclophosphamide and rituximab in adult patients aged 18-59 years with ALL without BCR-ABL.14,18 The full GRAALL 2005 protocol is available in Online Supplementary Figure S1. Historical definitions from the Children’s Oncology Group were used to classify the initial CNS status based on cerebrospinal fluid (CSF) conventional cytospin (CC) (Online Supplementary Table S1).9 Flow cytometry was not routinely used for CSF evaluation at diagnosis. All patients received a 5-drug induction therapy including native Escherichia coli L-Asparaginase (L-Asp). Patients in complete morphological remission (CR) received two consolidation courses including high-dose methotrexate and high-dose cytarabine. MRD was based on patientspecific Ig/TCR gene rearrangement monitoring using standardized quantitative real-time polymerase chain reaction (qRT-PCR) with a sensitivity of at least 10-4, centrally performed on bone marrow (BM) samples after the first induction course. High-risk patients aged <55 years, defined in the Online Supplementary Methods, were eligible for allogeneic SCT in first CR with a conditioning regimen including cyclophosphamide and total body irradiation (TBI). Patients in persistent CR who did not proceed to allogeneic stem cell transplantation (SCT) received a late intensification followed by one consolidation course and a 2-year maintenance. CNS-directed prophylaxis included one methotrexate intrathecal injection during steroid pre-phase followed by six triple intrathecal injections. CNS irradiation was recommended for all patients (i.e., 15 grays for patients proceeding to allogeneic SCT and 24 grays for other patients). Besides CNS irradiation, CNS-positive patients received an additional eight triple intrathecal therapy during induction and were eligible for allogeneic SCT in first CR. During induction therapy, they received fewer LAsp injections (5 instead of 8) to reduce the risk of CNS thrombosis. Informed consent was obtained from all patients at entry into this trial, which was conducted in accordance with

Haematologica | 108 December 2023

3288

ARTICLE - Initial CNS involvement in adults with ALL

C. Orvain et al.

the Declaration of Helsinki and approved by the InstituResults tional Ethics Committee Ile-de-France VI, France. This trial was registered at www.clinicaltrials.gov (NCT00327678). Incidence of initial central nervous system involvement and clinical presentation Statistical analysis Between 2006 and 2014, 784 adult patients with newly diCategorical variables were presented as numbers with agnosed Ph-negative ALL were included of whom 55 (7%) proportions and compared using the χ2 test or the had initial CNS involvement. The first lumbar puncture was Fisher’s exact test, as appropriate. Continuous variables performed during pre-phase steroid therapy in most pawere presented as medians with interquartile range tients (672/784 patients, 86%). CNS-positive patients were (IQR) and compared using the Mann and Whitney and more likely to have T-cell ALL (P=0.004), leukocytes ≥30x Kruskal-Wallis tests, as appropriate. Univariable and 109/L (P=0.016), and higher hemoglobin levels (P=0.02) multivariable Cox regression models were used to ana- (Table 1). Most CNS-positive patients (n=47, 85%) were lyze the association between patients' characteristics classified as CNS-3 (>5 white blood cells/μl and a positive and overall survival (OS). The Kaplan-Meier method was CC and/or clinical signs) whereas 7 patients (13%) were used to estimate median OS and survival curves were classified as CNS-2 (<5 white blood cells/μl and a positive compared using the log-rank test. Cumulative incidence CC) (Table 2). Initial presentation was heterogeneous since of relapse (CIR; with non-relapse mortality [NRM] as a 27 patients (49%) had clinical symptoms (9/27, 33% with competing event) and NRM (death without prior relapse concurrent positive CC) and 24 patients (44%) had only with relapse as a competing risk) were summarized CSF findings. The most prominent clinical sign was triusing cumulative incidence estimates and compared geminal anesthesia that was reported in 41% of patients using Gray’s test. Cox proportional hazards regression with clinical symptoms. Imaging (mostly magnetic resonmodels were used to estimate cause-specific Hazard ance imaging), performed in 26 patients (47%), was only Ratio (HR). Landmark analyses were performed to study positive in 5: 4 with clinical symptoms and one with CSF the impact of cranial irradiation and allogeneic SCT.19 All findings. tests were two-sided. P<0.05 was considered statistically significant. Statistical analyses were performed Association between central nervous system with R (R Foundation for Statistical Computing, Vienna, involvement and outcome Austria; http://www.r-project.org). A patient flowchart is shown in Figure 1. First CR rate, in-

Table 1. Characteristics of patients with or without central nervous system involvement.

Characteristic Age, median (IQR), years Female, N (%) BMI, median (IQR), kg/m2 Phenotype, N (%) B-cell T-cell

All, N=784 36 (25-48) 312 (40) 23.6 (21.1-27.2)

CNS-negative, N=729 37 (25-48) 294 (40) 23.7 (21.1-27.3)

CNS-positive, N=55 30 (24-44) 18 (33) 23.5 (21.1-26.2)

P 0.15 0.27 0.75 0.004

523 (67) 261 (33)

496 (68) 233 (32)

27 (49) 28 (51)

WBC at diagnosis, median (IQR), x109/L

12 (4-42)

11 (4-41)

23 (9-66)

0.15

Hb at diagnosis, median (IQR), g/dL PLT at diagnosis, median (IQR), x109/L Poor early PB blast clearance, N (%) Not evaluable Poor early BM blast clearance, N (%) Not evaluable CR, N (%) Induction death, N (%) MRD1 negativity, N (%) Not evaluable Allogeneic SCT in first CR, N (%)

10.2 (8.2-12.3) 72 (33-154) 187 (24) 3 306 (39) 40 722 (92) 44 (6) 126 (16) 445 278 (35)

10.1 (8.2-12.2) 72 (32-154) 172 (24) 3 285 (39) 36 672 (92) 41 (6) 119 (16) 416 249 (34)

11.1 (8.8-13) 78 (36-137) 15 (27)

0.02 0.16 0.50

21 (38) 4 50 (91) 3 (6) 7 (13) 29 30 (55)

0.58 0.79 >0.99 0.26 0.002

BM: bone marrow; BMI: body mass index; CNS: central nervous system; CR: complete remission; Hb: hemoglobin; IQR: interquartile range; MRD1: measurable residual disease after induction; N: number; PB: peripheral blood; PLT: platelet count; SCT: stem cell transplantation; WBC: white blood cell count. Haematologica | 108 December 2023

3289

ARTICLE - Initial CNS involvement in adults with ALL

C. Orvain et al.

duction death, and achievement of negative MRD after induction (Ig/TCR <10-4) were similar between CNSnegative and CNS-positive patients (92% vs. 91%, P=0.79; 6% vs. 6%, P=1; 38% vs. 27% in patients in whom MRD was evaluated, P=0.26, respectively) (Table 1). OS was, however, shorter in CNS-positive patients (median 1.9 years vs. not reached, HR=1.8, [1.3-2.6], P<0.001) (Figure 2A). OS was similar when patients who received allogeneic SCT in first CR were censored at the time of transplant (P=0.01) (Online Supplementary Figure S2A). CNS involvement at diagnosis was not associated with a statistically different CIR (HR=1.5 [0.9-2.5], P=0.11), but it was significantly associated with higher NRM (HR=2.1 [1.2-3.5], P=0.01) (Figure 2B, C). Causes of NRM were similar between CNS-positive and CNS-negative patients, including infection (69% vs. 52%), transplant-related (13% vs. 15%), and thrombosis (6% vs. 5%) while two were undetermined in CNS-positive patients (13%). Other causes of NRM in CNS-negative patients included bleeding (7%), second cancer (4%), and other causes (4%), while 15 (13%) were undetermined. At three years, 34% (20-47%) of CNS-positive patients versus 26% (23-29%) of CNSnegative patients relapsed. CNS-positive patients had a non-significant higher risk of CNS relapse (6% [0-13%] vs. 2% [1-3%] at 3 years, P=0.095) whereas combined and isolated BM relapses were similar to CNS-negative patients (4% [0-10%] vs. 2% [1-3%], P=0.41 and 21% [10-33%] vs. 20% [17-23%] at 3 years, P=0.79, respectively) (Online Supplementary Figure S3). However, after censoring patients at allogeneic SCT, the CIR was higher in CNS-positive patients (HR=2.2 [1.2 -3.9], P=0.01) whereas there was no difference in NRM between the two groups (HR=1.4 [0.7-3.1], P=0.4) (Online Supplementary Figure S2B, C). Because the GRAALL-2005 study randomized patients to standard (Standard-C arm) and hyperfractionated (hyper-C arm) cyclophosphamide, we evaluated the impact of the randomization arm in patients with initial CNS involvement. The characteristics of CNS-positive patients were similar according to the randomization arm (Online Supplementary Table S2). The proportion of patients receiving cranial irradiation (57% vs. 40% for the standard and hyper-C arms, respectively, P=0.28) and allogeneic SCT (53% vs. 56%, P=1) was also similar. CNS-positive patients randomized to the hyper-C arm had shorter OS (median 1.1 years vs. not reached, P<0.001) whereas CNSnegative patients had similar outcomes (Online Supplementary Figure S4A). CIR was similar according to the randomization arm but there was a significant increase in NRM in CNS-positive patients randomized to the hyper-C arm (Online Supplementary Figure S4B, C).

Table 2. Initial clinical and cerebrospinal fluid presentation of patients with central nervous system involvement.

CNS-positive, N=55 CNS status, N (%) CNS-2 CNS-3 NA Presentation, N (%) CSF only Clinical signs only CSF + clinical signs NA Clinical signs, N (%) Trigeminal anesthesia Facial paralysis Paresthesia: extremities Visual signs Meningeal syndrome Motor deficit Confusion Radiological signs, N (%) Present Absent NA

7 (13) 47 (85) 1 24 (44) 18 (33) 9 (16) 4 27 (49) 11 (20) 4 (7) 4 (7) 3 (5) 2 (4) 2 (4) 2 (4) 5 (9) 21 (38) 29

CNS: central nervous system; CSF: cerebrospinal fluid; NA: not available.

kocytes ≥ 30x109/L (HR=1.4 [1.1-1.7], P=0.1), CNS involvement (HR=1.3 [1.8-2.6], P=0.001), and poor early BM blast clearance (HR=1.3 [1.1-1.7], P=0.02) were associated with lower OS (Table 3). The significant association between CNS involvement at diagnosis and lower OS was also observed in a multivariable Cox regression model (HR=2.1 [1.4-3], P<0.001). Age (HR=1.3 [1.2-1.4], P<0.001) and leukocytes ≥ 30x109/L (HR=1.4 [1-1.8], P=0.03) were the only two other factors associated with OS in this multivariable model (Table 3). After adjustment for randomization arm in a Cox regression model, initial CNS involvement was still associated with worse OS (HR=1.8 [1.3-2.6], P=0.002). In CNS-positive patients, factors associated with OS were female gender (HR=3.5 [1.5-8.3], P=0.005) and poor early BM blast clearance (HR=4.4 [1.8-11], P<0.001) after multivariable analysis. Neither age nor ALL subtype were associated with OS after adjustment (HR=1 [0.7-1.4], P=0.8 and HR=1.1 [0.5-2.5], P=0.9, respectively) (Online Supplementary Figure S5).

Post-remission treatment Twenty-seven (49%) CNS-positive patients received cranial irradiation as part of consolidation therapy, after a median of 160 days, including 18 patients before allogeneic Analysis of factors associated with overall survival SCT. As recommended by the study protocol, patients with In a univariable Cox regression model, age (HR=1.3 [1.2-1.4], CNS involvement were more likely to receive allogeneic P<0.001), body-mass index (HR=1.4 [1.2-1.8], P<0.001), leu- SCT (55% vs. 34%, P=0.002), after a median interval of 170 Haematologica | 108 December 2023

3290

ARTICLE - Initial CNS involvement in adults with ALL

C. Orvain et al.

Figure 1. Flowchart of patients with central nervous system involvement at diagnosis. C: cyclophosphamide; CNS: central nervous system; CR1: first complete remission; SCT: stem cell transplantation.

days (vs. 142 days for patients without initial CNS involvement proceeding to allogeneic SCT, P=0.003). Twenty-five CNS-positive patients did not undergo allogeneic SCT, including 11 who did not have a suitable donor, seven who relapsed before allogeneic SCT could be carried out, six who died from toxicity before allogeneic SCT, and one who was deemed unfit for allogeneic SCT. We therefore performed two landmark analyses for these patients: one at day 160 to study cranial irradiation and one at day 170 to study allogeneic SCT and their association with outcome. After excluding 14 CNS-positive patients who experienced

an event (relapse, death) before the day 160 landmark point, there was no difference in OS (HR=0.7 [0.3-1.6], P=0.4) between patients who received cranial irradiation and those who did not (Figure 3A). Of nine CNS-positive patients who received cranial radiation therapy without allogeneic SCT, only one death resulting from toxicity was observed. There were 11 relapses, four in those who received cranial radiation (2 BM and 2 CNS relapses) and seven in those who did not (5 BM and 2 combined relapses). In addition, after excluding 16 CNS-positive patients who had an event before the day 170 landmark

Haematologica | 108 December 2023

3291

ARTICLE - Initial CNS involvement in adults with ALL

C. Orvain et al.

Figure 2. Outcomes according to central nervous system status at diagnosis. (A) Kaplan-Meier curves for overall survival, (B) cumulative incidence of relapse, and (C) cumulative incidence of nonrelapse mortality according to (CNS) involvement at diagnosis. Haematologica | 108 December 2023

3292

Haematologica | 108 December 2023

3293

1.4 (1-1.9)

1.8 (1.3-2.4)

0.7 (0.5-0.99)

0.9 (0.6-1.4)

2.1 (1.2-3.5)

1.1 (0.7-1.6)

1.2 (0.8-1.7)

Female

BMI/10

T-cell phenotype

WBC count at diagnosis x109/L

CNS involvement

Poor early PB blast clearance

Poor early BM blast clearance 0.3

0.7

0.01

0.7

0.04

<0.001

0.06

<0.001

1.2 (0.8-1.8)

1.2 (0.7-2)

2.8 (1.6-4.7)

0.9 (0.5-1.4)

0.9 (0.6-1.4)

1.4 (1-2)

1.3 (0.9-1.9)

1.5 (1.3-1.8)

0.3

0.5

<0.001

0.6

0.1

0.2

<0.001

1.5 (1.1-1.9)

1.4 (1.1-1.9)

1.5 (0.9-2.5)

1.7 (1.3-2.2)

1 (0.7-1.3)

1.1 (0.9-1.5)

1 (0.7-1.3)

1.1 (1-1.2)

0.01

0.02

0.1

<0.001

0.8

0.4

0.7

0.1

Univariable analysis HR

1.4 (1-1.9)

1.1 (0.8-1.6)

1.6 (0.9-2.7)

1.7 (1.3-2.4)

0.8 (0.6-1.1)

1.1 (0.8-1.6)

0.9 (0.7-1.2)

0.03

0.6

0.08

<0.001

0.2

0.4

0.6

0.1

Multivariable analysis

1.1 (1-1.2)

Relapse

Univariable analysis

Multivariable analysis

Overall survival

0.4

0.01

0.2

1.3(1.1-1.7)

1.3 (1-1.6)

0.02

0.05

1.8 (1.3-2.6) < 0.001

1.4 (1.1-1.7)

0.8 (0.7-1.1)

1.4 (1.2-1.8) < 0.001

1.1 (0.9-1.4)

1.3 (1-1.7)

1.1 (0.8-1.5)

2.1 (1.4-3)

1.3 (1-1.7)

1.2 (1-1.5)

0.04

0.5

< 0.001

0.07

0.1

1.3 (1.2-1.4) < 0.001 1.3 (1.2-1.4) < 0.001

BM: bone marrow; BMI: body mass index; CNS: central nervous system; HR: Hazard Ratio; PB: peripheral blood; WBC: white blood cell count.

1.6 (1.4-1.9)

Multivariable analysis

Univariable analysis

Age/10

Factors

Non-relapse mortality

Table 3. Univariable and multivariable analysis of factors associated with non-relapse mortality, relapse, and overall survival by Cox regression.

ARTICLE - Initial CNS involvement in adults with ALL C. Orvain et al.

ARTICLE - Initial CNS involvement in adults with ALL

C. Orvain et al.

point, allogeneic SCT was not associated with OS (HR=0.8 [0.3-2.3], P=0.7) (Figure 3B). In a sensitivity analysis, at day 220, when 90% of patients had received allogeneic SCT, similar results were observed (data not shown).

Discussion In adults with Ph-negative ALL, the impact of CNS disease at presentation is controversial with no recent large-scale series, especially for patients receiving pediatric-inspired approaches.4,6 Treatment strategy usually includes CNSdirected therapy such as intrathecal chemotherapy, highdose systemic methotrexate, or cranial radiation therapy

and/or allogeneic SCT.4,6 For such patients included in the prospective randomized GRAALL-2005 trial, we observed an adverse outcome which seems to be mostly driven by an increased risk of toxicity, whether due to the hyperfractionated cyclophosphamide arm or to allogeneic SCT, rather than an increased risk of relapse. As in previous reports, we confirm that CNS involvement is a rare event in adults at diagnosis (7% of patients).4-8 In our study, most patients (85%) were classified as CNS-3, which contrasts with previous reports in children in which there was a more balanced distribution or, in some studies, more patients with CNS-2 disease.9,10,20-22 We cannot exclude the possibility that some patients with CNS-2 disease could have been under-reported in our study while

Figure 3. Outcomes of patients with central nervous system disease at diagnosis. (A) Overall survival of patients who received cranial radiation therapy by a 160-day landmark analysis and (B) overall survival of patients who received allogeneic stem cell transplantation (SCT) by a 170-day landmark analysis. Haematologica | 108 December 2023

3294

ARTICLE - Initial CNS involvement in adults with ALL

C. Orvain et al.

the poor outcome of children with CNS-2 is a question of debate.9,10,20 CSF flow cytometry, which can increase the likelihood of identifying patients from 6-13% to 18-25%,23-26 was not routinely available at a multicentric level at the time of the GRAALL-2005 study since it requires either special conditioning or rapid processing.23,24 CNS involvement has been associated with an increased risk of relapse in some23-25,27 but not all studies.26 Garcia et al. reported that additional intrathecal therapy (twice weekly until the CSF and symptoms had cleared, then once weekly for 4 weeks) did not mitigate the worse outcome, despite the rapid clearance of blast cells in the CSF.26 In our cohort, the CR rate of CNS-positive patients was over 90%, similar to historical series, meaning that patients with initial CNS disease at diagnosis did not have a more resistant disease.4 Moreover, CNS involvement neither modified induction death rates nor MRD status after induction, despite patients receiving fewer L-asp injections (5 vs. 8 in patients without CNS disease) to reduce the rate of CNS thrombosis due to increased intrathecal treatment. This contrasts with both the MRC UKALL XII/ECOG E2993 trial, in which patients with CNS involvement had both higher mortality rates in remission and an increased risk of relapse,4 and with a previous study of our group (GET-LALA) conducted before the pediatric-inspired treatment era, where the induction death rate was twice as high (10%) in these patients.6 Patients included in the MRC UKALL XII/ECOG E2993 trial received intrathecal therapy and cranial radiation as CNS-directed therapy and were likely to proceed to allogeneic SCT.4 Their systemic CNS-directed therapy was, however, less intense, with high-dose methotrexate administered later during the treatment course.4 In contrast to this study, our previous GET-LALA trial had showed that patients with initial CNS involvement receiving autologous or allogeneic SCT had better outcomes than patients receiving chemotherapy alone. However, the global results of this older protocol were far inferior to modern regimens, with few patients receiving systemic chemotherapy with good CNS penetration or cranial radiation therapy.6 Moreover, the rate of CNS relapse remained high after allogeneic SCT, including TBI or TBI-free conditioning regimens, in patients with previous CNS involvement.28-31 Although the number of patients is limited, this questions whether allogeneic SCT is the most efficient way to reduce relapse in patients with CNS disease at diagnosis, as it might for patients with other high-risk features such as MRD.16 The inferior outcome in our patients with initial CNS disease was more likely due to toxicity, since higher mortality in remission was observed in patients who received hyperfractionated cyclophosphamide and/or proceeded to allogeneic SCT. Due to the small number of patients with initial CNS disease, these conclusions are, however, specu-

lative. There is no clear explanation for increased toxicity of hyperfractionated cyclophosphamide in CNS patients as it was not observed in the entire cohort of randomized patients.14 Higher NRM was no longer observed after censoring patients who proceeded to allogeneic SCT, suggesting that many patients suffered from transplant-related toxicity. On the other hand, the cumulative incidence of relapse was higher in patients who did not proceed to allogeneic SCT. This largely explains why there was no difference in OS between patients receiving SCT or not in the landmark analysis. It might be argued that, while some relapses may be prevented with allogeneic SCT, there was no benefit in OS due to increased toxicity. Despite our small number of patients with CNS involvement, we found no association between cranial irradiation during consolidation and outcome in the landmark analysis. As in other studies, cranial irradiation was recommended for patients with CNS involvement in many clinical trials but was seldom performed (27/55 patients in the GRAALL2005 study).4 In children with initial CNS involvement, the omission of cranial irradiation was associated with a slight increase in risk of CNS relapse in some studies but did not affect overall survival when both intrathecal and systemic CNS-directed chemotherapy were intensified.10,22,32-35 As cranial irradiation does not completely mitigate the risk of CNS relapse but may induce neurocognitive impairment, many co-operative groups have now abandoned cranial irradiation in all children, including those with initial CNS involvement.22,33-42 In one study analyzing the outcome of 467 adult patients with ALL who did not receive cranial irradiation during first-line treatment, the risk of CNS relapse was not increased in the 18 patients with initial CNS involvement.5 To our knowledge, no other study has evaluated the omission of cranial irradiation in adult patients with overt CNS disease. We could not evaluate the association between cranial irradiation and outcomes in CNSnegative patients as this procedure was not exhaustively recorded in CNS-negative patients. In conclusion, despite improved outcome in adults with ALL included in pediatric-inspired protocols, CNS involvement is still associated with reduced survival, mainly due to excessive toxicity. The historical use of allogeneic SCT did not improve outcome. Because of the rarity of CNS involvement, it is unlikely that different treatment approaches will be tested in these patients in prospective controlled trials. However, it will be important to evaluate the outcome of these patients in the GRAALL-2014 protocol that no longer retains CNS involvement as an indication for allogeneic SCT. Whereas the treatment strategy has remained unchanged for CNS-positive patients, including cranial radiation, our approach to prophylaxis has been modified in CNS-negative patients, with higher doses of systemic methotrexate during consolidation for patients under the age of 45 and an additional seven triple intrathecal injections during therapy,

Haematologica | 108 December 2023

3295

ARTICLE - Initial CNS involvement in adults with ALL

C. Orvain et al.

while cranial radiation is only recommended for patients undergoing allogeneic SCT. Disclosures CO acted as a consultant to Incyte and Novartis. MHB was a member of the advisory board of Erytech© and acted as a consultant to Jazz Pharmaceuticals. YC acted as a consultant to MSD, Novartis, Incyte, BMS, Pfizer, Abbvie, Roche, Jazz Pharmaceuticals, Gilead, Amgen, Astra Zeneca, and Servier, and received travel support from MSD, Roche, Gilead, Amgen, Incyte, Abbvie, Janssen, Astra Zeneca, and Jazz Pharmaceuticals.

data analysis and interpretation. All authors are responsible for the data collection and assembly and writing the paper, and approved the final version for publication. Funding The study was sponsored by the Regional Clinical Research Office, Paris, France, and supported by grants from the Programme Hospitalier de Recherche Clinique (PHRC), the French Ministry of Health, and the Institut National du Cancer in France (PHRC n. AOM 04144 – P040429), and the Swiss State Secretariat for Education, Research, and Innovation (SERI), Switzerland.

Contributions Data-sharing statement PC, FH, NI, HD, NB and MHB are responsible for study con- The datasets analyzed during the current study are available ception and design. CO, NB and MHB are responsible for from the corresponding author upon reasonable request.

References 1. Yao H, Price TT, Cantelli G, et al. Leukaemia hijacks a neural mechanism to invade the central nervous system. Nature. 2018;560(7716):55-60. 2. Savino AM, Fernandes SI, Olivares O, et al. Metabolic adaptation of acute lymphoblastic leukemia to the central nervous system microenvironment depends on stearoyl-CoA desaturase. Nat Cancer. 2020;1(10):998-1009. 3. Omura G, Moffitt S, Vogler W, Salter M. Combination chemotherapy of adult acute lymphoblastic leukemia with randomized central nervous system prophylaxis. Blood. 1980;55(2):199-204. 4. Lazarus HM, Richards SM, Chopra R, et al. Central nervous system involvement in adult acute lymphoblastic leukemia at diagnosis: results from the international ALL trial MRC UKALL XII/ECOG E2993. Blood. 2006;108(2):465-472. 5. Sancho J-M, Ribera J-M, Oriol A, et al. Central nervous system recurrence in adult patients with acute lymphoblastic leukemia: frequency and prognosis in 467 patients without cranial irradiation for prophylaxis. Cancer. 2006;106(12):2540-2546. 6. Reman O, Pigneux A, Huguet F, et al. Central nervous system involvement in adult acute lymphoblastic leukemia at diagnosis and/or at first relapse: results from the GET-LALA group. Leuk Res. 2008;32(11):1741-1750. 7. DeAngelo DJ, Stevenson KE, Dahlberg SE, et al. Long-term outcome of a pediatric-inspired regimen used for adults aged 18-50 years with newly diagnosed acute lymphoblastic leukemia. Leukemia. 2015;29(3):526-534. 8. Stock W, Luger SM, Advani AS, et al. A pediatric regimen for older adolescents and young adults with acute lymphoblastic leukemia: results of CALGB 10403. Blood. 2019;133(14):1548-1559. 9. Winick N, Devidas M, Chen S, et al. Impact of Initial CSF findings on outcome among patients with National Cancer Institute standard- and high-risk B-cell acute lymphoblastic leukemia: a report from the Children’s Oncology Group. J Clin Oncol. 2017;35(22):2527-2534. 10. Sirvent N, Suciu S, De Moerloose B, et al. CNS-3 status remains an independent adverse prognosis factor in children with acute lymphoblastic leukemia (ALL) treated without cranial irradiation: results of EORTC Children Leukemia Group study 58951. Arch Pediatr. 2021;28(5):411-416. 11. Boissel N, Auclerc M-F, Lhéritier V, et al. Should adolescents

with acute lymphoblastic leukemia be treated as old children or young adults? Comparison of the French FRALLE-93 and LALA94 trials. J Clin Oncol. 2003;21(5):774-780. 12. Huguet F, Leguay T, Raffoux E, et al. Pediatric-inspired therapy in adults with Philadelphia chromosome-negative acute lymphoblastic leukemia: the GRAALL-2003 study. J Clin Oncol. 2009;27(6):911-918. 13. Ram R, Wolach O, Vidal L, Gafter-Gvili A, Shpilberg O, Raanani P. Adolescents and young adults with acute lymphoblastic leukemia have a better outcome when treated with pediatricinspired regimens: systematic review and meta-analysis. Am J Hematol. 2012;87(5):472-478. 14. Huguet F, Chevret S, Leguay T, et al. Intensified therapy of acute lymphoblastic leukemia in adults: report of the randomized GRAALL-2005 clinical trial. J Clin Oncol. 2018;36(24):2514-2523. 15. Advani AS, Larsen E, Laumann K, et al. Comparison of CALGB 10403 (Alliance) and COG AALL0232 toxicity results in young adults with acute lymphoblastic leukemia. Blood Adv. 2021;5(2):504-512. 16. Dhédin N, Huynh A, Maury S, et al. Role of allogeneic stem cell transplantation in adult patients with Ph-negative acute lymphoblastic leukemia. Blood. 2015;125(16):2486-2496. 17. Ribera J-M, Morgades M, Ciudad J, et al. Chemotherapy or allogeneic transplantation in high-risk Philadelphia chromosome-negative adult lymphoblastic leukemia. Blood. 2021;137(14):1879-1894. 18. Maury S, Chevret S, Thomas X, et al. Rituximab in B-lineage adult acute lymphoblastic leukemia. N Engl J Med. 2016;375(11):1044-1053. 19. Delgado J, Pereira A, Villamor N, Lopez-Guillermo A, Rozman C. Survival analysis in hematologic malignancies: recommendations for clinicians. Haematologica. 2014;99(9):1410-1420. 20. Burger B, Zimmermann M, Mann G, et al. Diagnostic cerebrospinal fluid examination in children with acute lymphoblastic leukemia: significance of low leukocyte counts with blasts or traumatic lumbar puncture. J Clin Oncol. 2003;21(2):184-188. 21. Levinsen M, Taskinen M, Abrahamsson J, et al. Clinical features and early treatment response of central nervous system involvement in childhood acute lymphoblastic leukemia: CNS

Haematologica | 108 December 2023

3296

ARTICLE - Initial CNS involvement in adults with ALL

C. Orvain et al.

involvement in childhood ALL. Pediatr Blood Cancer. 2014;61(8):1416-1421. 22. Tang J, Yu J, Cai J, et al. Prognostic factors for CNS control in children with acute lymphoblastic leukemia treated without cranial irradiation. Blood. 2021;138(4):331-343. 23. Del Principe MI, Buzzatti E, Piciocchi A, et al. Clinical significance of occult central nervous system disease in adult acute lymphoblastic leukemia. A multicenter report from the Campus ALL Network. Haematologica. 2019;106(1):39-45. 24. Thastrup M, Marquart HV, Levinsen M, et al. Flow cytometric detection of leukemic blasts in cerebrospinal fluid predicts risk of relapse in childhood acute lymphoblastic leukemia: a Nordic Society of Pediatric Hematology and Oncology study. Leukemia. 2020;34(2):336-346. 25. Córdova-Serrano RD, Almanza-Huante E, Fernández-Sánchez E, Hernández-Alcántara A, Espinosa-Bautista K. Central nervous system (CNS) involvement has an adverse impact on survival in newly diagnosed adult acute lymphoblastic leukemia (ALL) assessed by flow cytometry. Leuk Lymphoma. 2021;62(13):3264-3270. 26. Garcia KA, Cherian S, Stevenson PA, et al. Cerebrospinal fluid flow cytometry and risk of central nervous system relapse after hyperCVAD in adults with acute lymphoblastic leukemia. Cancer. 2022;128(7):1411-1417. 27. Ranta S, Nilsson F, Harila-Saari A, et al. Detection of central nervous system involvement in childhood acute lymphoblastic leukemia by cytomorphology and flow cytometry of the cerebrospinal fluid: diagnosis of CNS leukemia in children by flow cytometry. Pediatr Blood Cancer. 2015;62(6):951-956. 28. Chantepie SP, Mohty M, Tabrizi R, et al. Treatment of adult ALL with central nervous system involvement at diagnosis using autologous and allogeneic transplantation: a study from the Société Française de Greffe de Moelle et de Thérapie Cellulaire. Bone Marrow Transplant. 2013;48(5):684-690. 29. Aldoss I, Al Malki MM, Stiller T, et al. Implications and management of central nervous system involvement before allogeneic hematopoietic cell transplantation in acute lymphoblastic leukemia. Biol Blood Marrow Transplant. 2016;22(3):575-578. 30. Shigematsu A, Kako S, Mitsuhashi K, et al. Allogeneic stem cell transplantation for adult patients with acute lymphoblastic leukemia who had central nervous system involvement: a study from the Adult ALL Working Group of the Japan Society for Hematopoietic Cell Transplantation. Int J Hematol. 2017;105(6):805-811. 31. Esfandbod M, Enshaei M, Monzavi SM, Kabootari M, Behfar M, Hamidieh AA. Radiation-free myeloablative allogeneic hematopoietic stem cell transplantation for adult acute lymphoblastic leukemia: a comparison of outcomes between patients with and without central nervous system involvement.

Leuk Res. 2021;111:106703. 32. Wilejto M, Giuseppe GD, Hitzler J, Gupta S, Abla O. Treatment of young children with CNS-positive acute lymphoblastic leukemia without cranial radiotherapy: treating children with CNS ALL without CRT. Pediatr Blood Cancer. 2015;62(11):1881-1885. 33. Vora A, Andreano A, Pui C-H, et al. Infuence of cranial radiotherapy on outcome in children with acute lymphoblastic leukemia treated with contemporary therapy. J Clin Oncol. 2016;34(9):919-926. 34. Taskinen M, Oskarsson T, Levinsen M, et al. The effect of central nervous system involvement and irradiation in childhood acute lymphoblastic leukemia: lessons from the NOPHO ALL-92 and ALL-2000 protocols. Pediatr Blood Cancer. 2017;64(2):242-249. 35. Pui C-H, Bowman WP, Ribeiro RC, et al. Treating childhood acute lymphoblastic leukemia without cranial irradiation. N Engl J Med. 2009;360(26):2730-2741. 36. Jeha S, Pei D, Choi J, et al. Improved CNS control of childhood acute lymphoblastic leukemia without cranial irradiation: St Jude Total Therapy Study 16. J Clin Oncol. 2019;37(35):3377-3391. 37. Spiegler BJ, Kennedy K, Maze R, et al. Comparison of long-term neurocognitive outcomes in young children with acute lymphoblastic leukemia treated with cranial radiation or highdose or very high-dose intravenous methotrexate. J Clin Oncol. 2006;24(24):3858-3864. 38. Zhou C, Zhuang Y, Lin X, Michelson AD, Zhang A. Changes in neurocognitive function and central nervous system structure in childhood acute lymphoblastic leukaemia survivors after treatment: a meta-analysis. Br J Haematol. 2020;188(6):945-961. 39. Zając-Spychała O, Pawlak M, Karmelita-Katulska K, et al. Antileukemic treatment-induced neurotoxicity in long-term survivors of childhood acute lymphoblastic leukemia: impact of reduced central nervous system radiotherapy and intermediateto high-dose methotrexate. Leuk Lymphoma. 2018;59(10):2342-2351. 40. Marwaha RK, Kulkarni KP, Bansal D, Trehan A. Central nervous system involvement at presentation in childhood acute lymphoblastic leukemia: management experience and lessons. Leuk Lymphoma. 2010;51(2):261-268. 41. Waber DP, Turek J, Catania L, et al. Neuropsychological outcome from a randomized trial of triple intrathecal chemotherapy compared with 18 Gy cranial radiation as CNS treatment in acute lymphoblastic leukemia: findings from Dana-Farber Cancer Institute ALL Consortium Protocol 95-01. J Clin Oncol. 2007;25(31):4914-4921. 42. Jacola LM, Krull KR, Pui C-H, et al. Longitudinal assessment of neurocognitive outcomes in survivors of childhood acute lymphoblastic leukemia treated on a contemporary chemotherapy protocol. J Clin Oncol. 2016;34(11):1239-1247.

Haematologica | 108 December 2023

3297

ARTICLE - Bone Marrow Failure

Conditional survival and standardized mortality ratios of patients with severe aplastic anemia surviving at least one year after hematopoietic cell transplantation or immunosuppressive therapy Ryotaro Nakamura,1* Bhavisha A. Patel,2* Soyoung Kim,3,4 F. Lennie Wong,5 Saro H. Armenian,5 Emma M. Groarke,2 Daniel Keesler,4 Kyle M. Hebert,4 Michael Heim,4 Mary Eapen6# and Neal S.

Correspondence: R. Nakamura

Young2#

rnakamura@coh.org

Received: Accepted: Early view:

Department of Hematology/Hematopoietic Cell Transplantation, City of Hope National

Medical Center, Duarte, CA; 2Hematology Branch, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MD; 3Division of Biostatistics, Institute for Health 4

and Equity, Medical College of Wisconsin, Milwaukee, WI; Center for International Blood and Marrow Transplant Research, Department of Medicine, Medical College of Wisconsin,

January 21, 2023. May 23, 2023. June 1, 2023.

Milwaukee, WI; 5Department of Population Sciences, City of Hope National Medical Center, Duarte, CA and 6Division of Hematology/Oncology, Department of Medicine, Medical College of Wisconsin, Milwaukee, WI, USA *

RN and BAP contributed equally as first authors.

ME and NSY contributed equally as senior authors.

Abstract Immunosuppressive treatment (IST) and hematopoietic cell transplant (HCT) are standard therapies for severe aplastic anemia (SAA). We report on conditional survival and standardized mortality ratios (SMR), which compare the mortality risk with the general population adjusted for age, gender, and race/ethnicity, in patients with SAA alive for at least 12 months after treatment with IST or HCT between 2000 and 2018. Given changes to treatment regimens and differences in length of follow-up, two treatment periods were defined a priori: 2000-2010 and 2011-2018. The SMR of patients treated during the period 2000-2010 and who survived one year were 3.50 (95% confidence interval [CI]: 2.62-4.58), 4.12 (95% CI: 3.20-5.21), and 8.62 (95% CI: 6.88-10.67) after IST, matched related donor HCT, and alternative donor HCT, respectively. For the period 2011-2018, the corresponding SMR were 2.89 (95% CI: 1.54-4.94), 3.12 (95% CI: 1.90-4.82), and 4.75 (95% CI: 3.45-6.38), respectively. For IST patients, their mortality risk decreased over time, and became comparable to the general population by five years. For patients who underwent HCT during 2000-2010 and 2011-2018, their mortality risk became comparable to the general population after ten years and after five years, respectively. Thus, 1-year survivors after IST or HCT can expect their longevity beyond five years to be comparable to that of the general US population.

Introduction Severe aplastic anemia (SAA) is caused by T-cell-mediated autoimmune destruction of hematopoietic progenitor cells associated with severe pancytopenia and a hypocellular marrow.1 Hematopoietic cell transplantation (HCT) can be curative and is increasingly available with donor options other than HLA matched siblings. For patients who are older, unfit, or who lack a suitable donor, immunosuppressive treatment (IST), primarily anti-thymocyte globulin (ATG) in combination with cyclosporine (CSA), remains the standard

of care.2 Over the last decades, overall survival for SAA has improved with continued refinement of treatment protocols including addition of eltrombopag, a thrombopoetin receptor agonist,3,4 and improved supportive care in both the IST and HCT settings.5,7 However, patients treated with IST can have incomplete or no hematologic recovery; approximately 30-40% experience relapse by five years.3,8 In addition, development of late clonal hematopoietic disorders such as myelodysplastic syndrome (MDS) or acquisition of isolated chromosomal aberration is a well-known complication in patients treated with IST.9,10 HCT is also associated with

Haematologica | 108 Mese 2023

3298

ARTICLE - Conditional survival after HCT or IST in SAA

R. Nakamura et al.

risks, including acute and chronic graft-versus-host disease (GvHD) and late complications such as cancer (seen in approximately 10% of survivors).10 These complications, along with persistent disease, have an impact on quality of life of SAA survivors.11,12 While outcomes after initial IST or HCT have been well-described, and complications after therapy are also better understood, limited data are available regarding the temporal variations of mortality risks on survivors of SAA after IST or HCT. It is known that HCT survivors continue to have substantially higher mortality rates compared with the general population.13-15 Socie and colleagues reported the data from the Center for International Blood and Marrow Transplantation Research (CIBMTR), which included 1029 patients with SAA transplanted between 1980 and 1993, mostly from a matched sibling donor (90%). Their analysis showed a 6% (95% confidence interval [CI]: 4-7%) mortality incidence at seven years in SAA patients who had survived at least two years after HCT.15 A subsequent registry study by Wingard and colleagues included 2171 patients with SAA transplanted between 1980 and 2003, again mostly from a matched sibling donor (81%), and reported the relative mortality risk over time, showing increased risks (approx. 5-fold) over more than ten years in these 2-year HCT survivors.14 A single institution analysis at City of Hope by Wong and colleagues also included patients with SAA (n=180), demonstrating that the standardized mortality ratios (SMR) for 1-year, 5-year, 10-year survivors were 9.0, 4.0, and 2.2, respectively.16 However, the above-mentioned registry studies did not have sufficient data on conditional survival and SMR after alternative donor HCT (matched/mismatched unrelated donors, mismatched related donors). Moreover, no studies have evaluated conditional survival or SMR in SAA after IST. As short-term outcomes after HCT and IST improve, the data on ongoing mortality risks in surviving patients are increasingly important and relevant for decision-making and risk-stratification for personalization of treatment. In this study, we enjoyed the benefit of using the resources of the CIBMTR and National Institutes of Health (NIH) to determine the conditional survival rates and SMR in patients with SAA treated with HCT or IST between 2000-2018, conditioned by survival time since treatment. A better understanding of conditional survival and SMR would allow risk-adapted interventions tailored to time-varying mortality risk in survivors to be implemented.

Methods Patients with SAA who received their first allogeneic HCT (n=3571) and reported to the CIBMTR or received IST at the Clinical Center, NIH (n=395) in the United States between January 2000 and December 2018 were eligible. Patients who survived less than 12 months after their treatment (n=677

[19%] after HCT and n=16 [4%] after IST) were excluded, resulting in 2894 patients who underwent allogeneic HCT and 379 patients who received IST for the current analysis. All IST recipients received anti-thymocyte globulin (equine ATG) with the following agents: cyclosporine (n=100), cyclosporine with mycophenolate (n=90), cyclosporine with sirolimus (n=35), and cyclosporine with eltrombopag (n=155; 2012 to 2018). Of the 2894 HCT recipients, n=1428 received grafts from HLA matched siblings, n=256 from HLA mismatched relatives, n=848 from HLA matched, and n=362 from HLA mismatched unrelated donors. All patients provided written informed consent. The study was approved by the Institutional Review Board of the National Marrow Donor Program. The outcome of interest was mortality in patients with SAA who were alive for at least 12 months from onset of treatment. Patients were analyzed in their intended treatment group: HCT and IST. Seventy-nine of 379 (21%) patients who received IST also received an HCT for recurrent disease after IST. All patients were followed through until death or last contact regardless of any treatment intervention in their assigned group. Patients in the HCT group were not censored at second HCT, and patients in the IST group were not censored for HCT or for recurrent disease. Death from any cause was considered an event. Surviving patients were censored at last follow-up. As this study spanned almost two decades (2000-2018), two treatment periods were assigned a priori: 2000-2010 and 2011-2018; each period was studied separately. Within each treatment period, patients were divided into five treatment groups: IST, HLA matched sibling, HLA mismatched relative, HLA matched unrelated, and HLA mismatched unrelated donor transplantations. Conditional survival estimates were calculated using the Kaplan-Meier method for the sub-cohort that had survived a given length of time (x) after transplantation. The standardized relative mortality risk (SMR) was used to quantify all-cause mortality risk after treatment, and compared to the age-, race-, and sexmatched general population.17 Person-years at risk were calculated from an anchor date after treatment until date of death or date last known alive. The equality of SMR was tested and two-sided 95% confidence intervals (CI) were calculated. Risk factors for mortality were examined using the Cox proportional hazards model.18 Variables considered included age at diagnosis (<18, 18-39, ≥ 40 years), sex (male vs. female), race/ethnicity (White non-Hispanic vs. Black non-Hispanic vs. Asian vs. Hispanic vs. other), and disease recurrence modelled as time-dependent co-variate. In subset analysis limited to HCT recipients, the effect of grade II-IV and grade II acute GvHD and chronic GvHD on mortality was examined. Variables that met a significance level ≤0.05 were included in the final Cox model. All variables met the assumption of proportional hazards and there were no first order interactions between the variables held in the final models. All P-values are two-sided,

Haematologica | 108 Mese 2023

3299

ARTICLE - Conditional survival after HCT or IST in SAA

R. Nakamura et al.

and analyses were carried out using SAS version 9.4 (Cary, NC, USA).

Results Patients’ and treatment characteristics A total of 379 patients who received IST (n=224 in 2000-2010, n=155 in 2011-2018) and 2894 patients who underwent HCT (n=1282 in 2000-2010, n=1612 in 2011-2018) were eligible for analysis. Patients’ characteristics are shown in Tables 1 and 2 according to their treatment period. Median age of patients during 2000-2010 was 30 years (range 2-82 years) for IST, 17.3 years (range 0.1-70.2 years) for matched sibling HCT, and 15.9 years (range 0.1-65.9 years) for alternative donor HCT. The corresponding median age for the period 2011-2018 were 29 years (range 3-82 years), 18.5 years (range 0-75.5 years), and 17.9 years (range 0-73.6 years), respectively. Sex was similarly distributed across the groups. For the period 2000-2010, non-Hispanic Whites represented 51% of the IST cohort and 74% of the HCT cohort. For the period 2011-2018, non-Hispanic Whites represented 57% of the IST cohort and 57% of the HCT cohort. The use of donor types was consistent with clinical practice in the periods studied. In the earlier period, 2000-2010, and in the later period, 2011-2018, matched siblings were the predominant donors and accounted for 735 of 1282 (57%) and 693 of 1612 (43%) transplantations, respectively. There was an increase in the use of other related donors in the period 2000-2010: 188 of 1612 (12%) compared to 68 of 1282 (5%) in the period 2011-2018. Bone marrow was the predominant graft source. There was no difference in the proportion of patients with acute and chronic GvHD according to transplant period. In the IST group (n=379), 100 patients (26%) received ATG/cyclosporine, 90 (24%) received ATG/cyclosporine/mycophenolate, 35 (9%) received ATG/ cyclosporine/sirolimus, and the remaining 155 (41%) patients received ATG/cyclosporine/eltrombopag. Fifty-nine patients in the IST group received HCT for recurrent disease and another 20 patients in the IST group received HCT for clonal hematologic features (MDS or cytogenetic abnormalities). Standardized mortality ratio, conditional survival and risk factors for mortality in 1-year survivors Treatment period 2000-2010 The SMR and the 5-year conditional survival at the 1-year, 5-year and 10-year landmark timepoints for the treatment period 2000-2010 are summarized in Table 3 and Figure 1. One-year SMR was significantly higher than for an age-, sex-, race-/ethnicity-matched US population regardless of treatment group. However, SMR for those surviving five and ten years after IST were comparable to that of the US population. In HCT recipients, SMR decreased over time and were comparable to that of the US population

at ten years. Unlike IST recipients, the 5-year SMR after HLA matched sibling and alternative donor HCT remained higher than that of the US population. Our models tested treatment type (IST, matched sibling and alternative donor HCT) with adjustment for age, sex, race/ethnicity and disease recurrence in 1-year survivors to identify mortality risks. All alternative donor HCT were considered as a single entity as there were no differences in SMR between HLA mismatched relative, HLA matched, and HLA mismatched unrelated donor HCT recipients (data not shown). After IST and HLA matched sibling HCT, older age at diagnosis (≥40 years) was associated with higher mortality risk (Table 4). After alternative donor HCT, mortality was higher in those aged ≥18 years (Table 4). Mortality risks were lower in females who received IST and alternative donor transplantation (Table 4). Sex was not associated with mortality after HLA matched sibling HCT. Disease recurrence was associated with higher mortality after IST (Hazard Ratio [HR] 1.89, 95% CI: 1.083.30; P=0.027), HLA matched sibling HCT (HR 3.45, 95% CI: 2.04-5.86; P<0.0001), and alternative donor HCT (HR 5.14, 95% CI: 2.89-9.16; P<0.0001). In multivariable analysis limited to recipients of HLA matched sibling HCT, a history of grade II-IV acute GvHD (HR 2.32, 95% CI: 1.36-3.97; P=0.002) and chronic GvHD (HR 3.26, 95% CI: 1.97-5.46; P<0.0001) was associated with higher mortality. Similarly, grade II-IV acute (HR 1.62, 95% CI: 1.04-2.53; P=0.0335) and chronic (HR 2.37, 95% CI: 1.48-3.79; P=0.0003) GvHD were associated with a higher risk for mortality after alternative donor HCT. Grade II acute GvHD was not associated with mortality after HLA matched sibling (HR 1.49, 95% CI: 0.75-2.94; P=0.25) or alternative (HR 0.83, 95% CI: 0.49-1.40; P=0.48) donor HCT. Treatment period 2011-2018 The SMR and the 5-year conditional survival at landmark timepoints 1- and 5-years for the treatment period 20112018 are summarized in Table 5 and Figure 2. One-year SMR was higher than for an age-, sex- race-/ethnicity-matched US population regardless of treatment group. However, by five years, SMR after IST for HLA matched sibling and alternative donor HCT were comparable to that of the US population. Disease recurrence was the only factor associated with higher mortality after IST (HR 4.25, 95% CI: 1.1515.69; P=0.029) and alternative donor HCT (HR 7.06, 95% CI: 3.87-12.88; P<0.0001). Although the risks were higher after HLA matched sibling HCT, this did not meet the level of significance (HR 2.71, 95% CI: 0.95-7.71; P=0.063). Age, sex and race/ethnicity were not associated with mortality risks (data not shown). Multivariate analysis of the subset of recipients of HLA matched sibling HCT, a history of grade II-IV acute GvHD (HR 5.29, 95% CI: 2.19-12.78; P=0.0002) and chronic GvHD (HR 4.85, 95% CI: 2.00-11.74; P=0.0005) was associated with higher mortality. Similarly, mortality risks were higher after alternative donor HCT in those with

Haematologica | 108 Mese 2023

3300

ARTICLE - Conditional survival after HCT or IST in SAA

R. Nakamura et al.

Table 1. Characteristics of patients undergoing immunosuppressive treatment or hematopoietic cell transplantation for treatment period 2000-2010.

Characteristic

IST N (%)

Matched sibling N (%)

Other related* N (%)

Matched unrelated N (%)

Mismatched unrelated N (%)

N of patients

224

735

301

178

Age in years at diagnosis < 18 18-39 ≥ 40

47 (21) 96 (43) 81 (36)

396 (54) 231 (31) 108 (15)

36 (53) 24 (35) 8 (12)

157 (52) 107 (36) 37 (12)

118 (66) 41 (23) 19 (11)

Sex Male Female

131 (58) 93 (42)

430 (59) 305 (41)

43 (63) 25 (37)

167 (55) 134 (45)

108 (61) 70 (39)

Race White, non-Hispanic Black, non-Hispanic Hispanic Asian Other/not reported

115 (51) 49 (22) 44 (20) 16 (7) 0 (0)

456 (62) 56 (8) 99 (13) 54 (7) 70 (10)

41 (60) 10 (15) 3 (4) 4 (6) 10 (15)

233 (77) 9 (3) 34 (11) 12 (4) 13 (4)

84 (47) 21 (12) 46 (26) 13 (7) 14 (8)

N/A N/A N/A

492 (67) 137 (19) 106 (14)

30 (44) 22 (32) 16 (24)

31 (10) 126 (42) 144 (48)

17 (10) 67 (38) 94 (53)

N/A 224 (100)

469 (64) 266 (36)

28 (41) 40 (59)

30 (10) 271 (90)

22 (12) 156 (88)

Graft type Bone marrow Peripheral blood Umbilical cord blood

N/A N/A N/A

596 (81) 133 (18) 6 (1)

42 (62) 21 (31) 5 (7)

232 (77) 69 (23) 0 (0)

99 (56) 31 (17) 48 (27)

Conditioning regimen Cy ± ATG Cy/fludarabine ± ATG TBI regimens 1000 cGy + other agents TBI regimens 200-400 cGy Cy Bu + other (Cy, fludarabine or melphalan) Fludarabine/melphalan ± thiotepa Not reported

N/A N/A N/A N/A N/A N/A N/A

551 (75) 37 (5) 25 (3) 28 (3) 33 (4) 11 (1) 51 (7)

29 (42) 8 (12) 4 (6) 13 (19) 5 (7) 2 (3) 7 (10)

22 (7) 41 (14) 37 (12) 169 (56) 17 (6) 10 (3) 5 (2)

7 (4) 18 (11) 21 (12) 99 (56) 17 (10) 13 (8) 3 (2)

GvHD prophylaxis, N (%) Calcineurin-containing Ex vivo T-cell depletion/CD 34 selection Post-transplant Cy-containing Not reported

N/A N/A N/A N/A

663 (90) 5 (1) 2 (<1) 65 (9)

50 (74) 8 (11) 0 (0) 10 (15)

286 (95) 9 (3) 0 (0) 6 (2)

166 (94) 8 (5) 0 (0) 4 (2)

Acute GvHD grade II-IV No Yes Not reported

N/A N/A N/A

593 (81) 96 (13) 46 (6)

36 (53) 18 (26) 14 (21)

200 (66) 85 (28) 16 (5)

109 (61) 66 (37) 3 (2)

Acute GvHD grade III-V No Yes Not reported

N/A N/A N/A

657 (89) 32 (4) 46 (6)

46 (68) 8 (12) 14 (21)

264 (88) 21 (7) 16 (5)

154 (87) 21 (12) 3 (2)

Chronic GvHD No Yes Not reported

N/A N/A N/A

563 (77) 129 (18) 43 (6)

35 (51) 20 (29) 13 (19)

187 (62) 94 (31) 20 (7)

86 (48) 86 (48) 6 (3)

118 (13-236)

112 (12-249)

120 (38-237)

120 (13-244)

117 (12-242)

Interval from diagnosis to transplant ≤ 3 months 4 -12 months >12 months IST None Yes

Follow-up in months, median (range)

IST: immunosuppressive therapy; Cy: cyclophosphamide; ATG: anti-thymocyte globulin; Bu: bulsuphan; TBI: total body irradiation; GvHD: graftversus-host disease; N/A: not available. *Matched related (not sibling): N=38. Mismatched related: N=30; %: due to rounding, totals do not correspond to 100%.

Haematologica | 108 Mese 2023

3301

ARTICLE - Conditional survival after HCT or IST in SAA

R. Nakamura et al.

Table 2. Characteristics of patients undergoing immunosuppressive treatment or hematopoietic cell transplantation for treatment period 2011-2018.

Characteristic

IST N (%)

Matched sibling N (%)

Other related* N (%)

Matched unrelated N (%)

Mismatched unrelated N (%)

N of patients

155

693

188

547

184

Age in years at diagnosis < 18 18-39 ≥ 40

40 (26) 59 (38) 56 (36)

335 (48) 255 (37) 103 (15)

92 (49) 64 (34) 32 (17)

248 (45) 167 (31) 132 (24)

124 (67) 39 (21) 21 (11)

Sex Male Female

76 (49) 79 (51)

388 (56) 305 (44)

114 (61) 74 (39)

292 (53) 255 (47)

100 (54) 84 (46)

Race White, non-Hispanic Black, non-Hispanic Hispanic Asian Other/not reported

89 (57) 33 (21) 6 (4) 16 (10) 11 (7)

360 (52) 76 (11) 140 (20) 49 (7) 68 (10)

67 (36) 49 (26) 44 (23) 12 (6) 16 (9)

407 (74) 27 (5) 74 (14) 19 (3) 20 (4)

82 (45) 33 (18) 40 (22) 7 (4) 22 (12)

N/A N/A N/A

460 (66) 155 (22) 78 (11)

48 (26) 60 (32) 80 (43)

84 (15) 239 (44) 224 (41)

17 (9) 79 (43) 88 (48)

N/A 155 (100)

450 (65) 243 (35)

46 (24) 142 (76)

79 (14) 468 (86)

22 (12) 162 (88)

N/A N/A N/A

619 (89) 70 (10) 4 (1)

142 (76) 31 (16) 15 (8)

455 (83) 92 (17) 0 (0)

122 (66) 30 (16) 32 (17)

N/A N/A N/A N/A N/A N/A N/A

475 (68) 134 (19) 4 (<1) 44 (6) 20 (3) 19 (3) 3 (<1)

21 (11) 6 (4) 0 (0) 150 (80) 3 (2) 8 (5) 0 (0)

31 (6) 115 (21) 3 (1) 345 (64) 23 (4) 30 (6) 0 (0)

11 (6) 30 (16) 2 (1) 116 (63) 7 (4) 18 (10) 0 (0)

GvHD prophylaxis, N (%) Calcineurin-containing Ex vivo T-cell depletion/CD 34 selection Post-transplant Cy-containing Not reported

N/A N/A N/A N/A

671 (97) 3 (<1) 9 (1) 10 (1)

50 (27) 17 (9) 119 (63) 2 (1)

502 (92) 19 (3) 25 (5) 1 (<1)

157 (85) 14 (7) 10 (5) 3 (2)

Acute GvHD grade II-IV No Yes Not reported

N/A N/A N/A

597 (86) 85 (12) 11 (2)

148 (79) 9 (5) 5 (3)

402 (73) 128 (23) 17 (3)

130 (71) 52 (28) 2 (1)

Acute GvHD grade III-V No Yes Not reported

N/A N/A N/A

658 (95) 24 (3) 11 (2)

174 (93) 9 (5) 5 (3)

505 (92) 25 (5) 17 (3)

163 (89) 19 (10) 2 (1)

Chronic GvHD No Yes Not reported

N/A N/A N/A

552 (80) 133 (19) 8 (1)

131 (70) 50 (27) 7 (4)

400 (73) 137 (25) 10 (2)

120 (65) 54 (29) 10 (5)

50 (13-98)

49 (12-120)

38 (12-106)

48 (12-114)

56 (12-115)

Interval from diagnosis to transplant < 3 months 4 -12 months > 12 months Immunosuppressive therapy None Yes Graft type Bone marrow Peripheral blood Umbilical cord blood Conditioning regimen Cy ± ATG Cy/fludarabine ± ATG TBI regimens 1000 cGy + other agents TBI regimens 200-400 cGy Cy Bu + other (Cy, fludarabine or melphalan) Fludarabine/melphalan ± thiotepa Not reported

Follow-up in months, median (range) *

Matched related (not sibling): N=25. Mismatched related: N=163. Cy: cyclophosphamide; ATG: anti-thymocyte globulin; Bu: bulsuphan; TBI: total body irradiation; GvHD: graft-versus-host disease; N/A: not available; %: due to rounding, totals do not correspond to 100%.

Haematologica | 108 Mese 2023

3302

ARTICLE - Conditional survival after HCT or IST in SAA

R. Nakamura et al.

a history of grade II-IV acute (HR 2.74, 95% CI: 1.50-4.98; P=0.0010) and chronic (HR 2.43, 95% CI: 1.30-4.56; P=0.0056) GvHD. Grade II acute GvHD was associated with mortality after alternative donor (HR 1.95, 95% CI: 1.03-3.70; P=0.04) but not after HLA matched sibling (HR 2.13, 95% CI: 0.716.38; P=0.18) HCT. There were 66 deaths among the 379 patients who received IST. The two predominant causes of death were infection (21%) and death from an HCT-related complication (21%) in those who failed IST and thereafter received HCT. Other causes include: myeloid malignancy (9%), solid tumor (3%), bleeding (3%), and other causes (9%). Cause of death was not available for 21 patients (32%). There were 219 deaths among the 2894 patients who underwent HCT. The predominant causes of death include:

infection (17%), organ failure (20%), GvHD (12%), and recurrent disease (graft failure: 11%). Other causes include myeloid malignancy (3%), solid tumor (2%), bleeding (3%), Epstein-Barr virus-associated lymphoproliferative disease (1%), and other causes (2%). Cause of death was not available for 64 patients (29%).

Discussion In this study, we evaluated SMR in patients who received HCT and IST for SAA and were alive for at least one year after their treatment; an intent-to-treat approach was adopted. In both treatment periods studied, that together spanned two decades, long-term survival was over 80%

Figure 1. Standardized mortality ratio: treatment period 2000-2010. IST: immunosuppressive therapy; HCT: hematopoietic cell transplant.

Table 3. Conditional survival probability and standardized mortality ratio by landmark timepoints (2000-2010). Survival landmark

Treatment group

Evaluable

Events

5-year survival

SMR (95% CI)

Overall SMR (95% CI)

1-year

IST HLA matched sibling HCT Alternative donor HCT

224 735 547

53 70 85

0.81 0.94 0.89

3.50 (2.62-4.58) 4.12 (3.20-5.21) 8.62 (6.88-10.67)

4.95 (4.29-5.67)

5-year

IST HLA matched sibling HCT Alternative donor HCT

154 558 413

17 33 36

0.89 0.96 0.92

1.30 (0.76-2.09) 2.11 (1.45-2.96) 3.92 (2.73-5.46)

2.26 (1.81-2.80)

10-year

IST HLA matched sibling HCT Alternative donor HCT

78 294 220

3 15 10

0.97 0.93 0.92

0.33 (0.07-0.95) 1.35 (0.76-2.23) 1.62 (0.78-2.98)

1.06 (0.70-1.53)

SMR: standardized mortality ratio; CI: confidence interval; IST: immunosuppressive therapy; HCT: hematopoietic cell transplant.

Haematologica | 108 Mese 2023

3303

ARTICLE - Conditional survival after HCT or IST in SAA

R. Nakamura et al.

tients treated during the period 2000-2011. The association between older age (≥ 40 years) and higher late mortality has been previously reported by others.14,15 Females fared better than males in this analysis. We did not identify an association between biologic variables and survival during the period 2011-2018. In all treatment groups, disease recurrence with or without salvage treatment was associated with higher mortality.19 A history of grade III-IV acute and chronic GvHD was also associated with higher mortality regardless of donor type and transplant period. Causes of death in patients with SAA surviving one year after IST and HCT were consistent with the expected mix in this population: infections and transplant-related causes for those treated with IST, and infection, GvHD, organ failure, and graft failures after HCT. Deaths attributed to myeloid neoplasm after IST was 9% and after HCT, 3%. IST recipients were older, and the observed evolution to myeloid

in patients who survived for at least one year after IST or HCT. Consistent with other reports on SMR for SAA after HCT, we also observed higher SMR in the earlier years after treatment, but the risks decreased over time. In the earlier treatment period studied (2000-2010), SMR of 5-year survivors after HCT from either HLA matched sibling or alternative donor were greater than the age-, sex- and race-/ ethnicity-matched US population. In contrast, there was no difference in SMR between 5-year HCT survivors and an age-, sex- and race-/ethnicity-matched US population in the later treatment period (2011-2018), likely explained by advances in transplant conditioning regimens, GvHD prophylaxis, and supportive care for HCT. Importantly, SMR after IST were comparable to an age-, sex- and race-/ethnicity-matched US population in the periods 2000-2010 and 2011-2018. In risk factor analyses, two biologic variables, age and sex, adversely affected long-term survival of pa-

Table 4. Multivariate analysis for prognostic factors for treatment period 2000-2010. Variables

Evaluable

Events

Hazard Ratio (95% CI)

IST Age in years at diagnosis < 18 18-39 ≥ 40 years Sex Male Female

47 96 81

4 16 33

1.00 1.87 (0.62-5.60) 5.68 (2.00-16.14)

<0.0001 0.27 0.001

131 93

38 15

1.00 0.40 (0.22-0.73)

0.002 0.003

HLA matched sibling HCT Age in years at diagnosis < 18 18-39 ≥ 40 years

396 231 108

22 18 30

1.00 1.29 (0.69-2.41) 4.41 (2.53-7.66)

<0.00001 0.42 <0.0001

311 172 64

33 30 22

1.00 1.79 (1.07-2.97) 4.59 (2.59-8.13)

<0.0001 0.026 <0.0001

318 229

59 26

1.00 0.58 (0.37-0.93)

0.02 0.023

Alternative donor HCT Age in years at diagnosis < 18 18-39 ≥ 40 years Sex Male Female

CI: Confidence Interval; IST: immunosuppressive therapy; HCT: hematopoietic cell transplant.

Table 5. Conditional survival probability and standardized mortality ratio by landmark timepoints (2011-2018). Survival landmark

Treatment group

Evaluable

Events

5-year survival

SMR (95% CI)

Overall SMR (95% CI)

1-year

IST HLA matched sibling HCT Alternative donor HCT

155 693 919

13 20 44

0.85 0.97 0.93

2.89 (1.54-4.94) 3.12 (1.90-4.82) 4.75 (3.45-6.38)

3.82 (3.01-4.77)

5-year

IST HLA matched sibling HCT Alternative donor HCT

55 256 312

3 3 6

0.82 0.97 0.95

1.27 (0.26-3.73)

1.17 (0.60-2.04)

SMR: standardized mortality ratio; CI: confidence interval; IST: immunosuppressive therapy; HCT: hematopoietic cell transplant.

Haematologica | 108 Mese 2023

3304

ARTICLE - Conditional survival after HCT or IST in SAA

R. Nakamura et al.

neoplasm could be explained in part by age and in part by treatment with disease-modifying agents. Five of 7 HCT recipients who developed myeloid malignancy were transplanted at a median age of 58 years (range 33-65 years). Probable etiologies include the challenge in distinguishing SAA from MDS in the absence of cytogenetic abnormalities, the effect of radiation-containing conditioning regimen,20 and the possibility of donor-derived myeloid malignancy. Donor hematopoiesis in the context of HCT could be prone to clonal hematopoiesis of indeterminate potential and evolution to MDS from stressed hematopoiesis after HCT.21 Two children aged three years at HCT who developed acute myeloid leukemia may have had an inherited bone marrow failure syndrome without classical phenotype. The data sources we used rely on the diagnosis assigned by the transplant center and we cannot comment on the diagnostic tests that had confirmed SAA. Historically, approximately two-thirds of patients achieved a response to IST after treatment with equine ATG combined with cyclosporine,22-24 and the robustness of hematologic response was found to be positively associated with longterm survival.7 More recently, eltrombopag has been shown to have efficacy in patients with IST-refractory SAA25-27 and in previously untreated patients with SAA with an overall response rate 80-95% and 2-year survival >95%.4 With longer follow-up (4 years), recurrent disease and clonal evolution were comparable to that in a historic cohort treated with IST except in patients with high-risk clonal evolution.3 A phase III European trial comparing IST with or without eltrombopag also confirmed that the rate, rapidity, and

strength of hematologic response were better with IST and eltrombopag in previously untreated patients with SAA.28 Alternative donor HCT using matched/mismatched unrelated,29,30 umbilical cord,31,32 and haploidentical donors33-35 are increasingly offered to those without a matched sibling, and are associated with significant risks of late morbidity and mortality due to GvHD, poor immune reconstitution, cardiopulmonary disease, and/or secondary non-hematologic neoplasms, which remain serious medical challenges for HCT survivors.10,14-16 For both IST and HCT, there is a clinical need to understand how mortality risk changes with survival time. The use of conditional survival allows for a dynamic evaluation of risks during survivorship, whereas the conventionally calculated survival probability from the time of IST or HCT does not reflect changes in prognosis over time. Our study was not designed to compare the outcomes between IST and HCT as a first-line therapy. Instead, our primary objective was to estimate the conditional survival and SMR in patients who have survived the early phases of treatment. Therefore, the observations in this study commenced at HCT in the HCT cohort and at IST in the IST cohort, not from the time of diagnosis or from first-line therapy, since many had received ATG/CSA +/- eltrombopag prior to HCT. There are limitations in our study. By design, the data are based on a single comprehensive center for IST while HCT data are based on the registry with the contribution of data from 40-50 centers. The transplant dataset may include biases: although we aim to provide a complete report, we may have missed late events as patients tend

Figure 2. Standardized mortality ratio: treatment period 2011-2018. IST: immunosuppressive therapy; HCT: hematopoietic cell transplant. Haematologica | 108 Mese 2023

3305

ARTICLE - Conditional survival after HCT or IST in SAA

R. Nakamura et al.

to transfer care with relocation or change in insurance coverage. Our study is also limited by a lack of detailed data on patient-reported outcomes such as social functions or quality of life metrics. We note the modest follow-up of the more recent cohort that should be followed for several more years for late mortality data. Furthermore, the relatively few events in 5- and 10-year survivors resulted in SMR with wide confidence boundaries. We believe a more robust interpretation of SMR is one that considers all treatment groups together with findings in keeping with those observed when treatment groups were compared separately. However, despite the limitations, these data significantly further our knowledge on conditional survival and SMR for SAA after HCT and IST in the recent era. Although long-term survival improves over time for 1-year survivors after either IST or HCT, the persistence of mortality risks greater than the general population warrant studies in cause-specific mortality. Disclosures No conflicts of interest to disclose. Contributions RN, BAP, SK, LW, SHA, KMH, ME and NSY designed the study. BAP, DAK, KMH and MH assembled the data. SK performed the analysis. RN, BAP, SK, LW, SHA, EMG, DAK, KMH, MH, ME and NSY interpreted the data. RN drafted the manuscript. BAP, SK, LW, SHA, EMG, DAK, KMH, MH, ME and NSY critically reviewed and approved the manuscript. Funding This work was supported in part by grant U24-CA076518, National Institute of Health and contract HHSH 25021200016C from the Health Resources and Services Administration/ Department of Health and Human Services (HRSA/DHHS) to the Center for International Blood and Marrow Transplant Research, the Intramural Research Program of the National Heart, Lung, and Blood Institute, and a

Cooperative and Research Development Agreement with Novartis/GSK to the Intramural Research Program of the National Heart, Lung, and Blood Institute. The views expressed in this article do not reflect the official policy or position of the National Institute of Health, HRSA/DHHS or any other agency of the United States government. Other support to the Center for International Blood and Marrow Transplant Research include: Grant N00014-21-1-2954 and N00014-23-1-2057 from the Office of Naval Research; Be the Match Foundation, the Medical College of Wisconsin, the National Marrow Donor Program, and from following commercial entities: AbbVie; Actinium Pharmaceuticals Inc.; Adaptimmune; Adaptive Biotechnologies Corporation; ADC Therapeutics; Adienne SA; Allogene; Allovir Inc.; Amgen Inc.; Angiocrine; Anthem; Astellas Pharma US; AstraZeneca; Atara Biotherapeutics; BeiGene; bluebird bio Inc.; Bristol Myers Squibb Co.; CareDx Inc.; CRISPR; CSL Behring; CytoSen Therapeutics Inc.; Eurofins Viracor, DBA Eurofins Transplant Diagnostics; Gamida-Cell Ltd.; Gilead; GlaxoSmithKline; HistoGenetics; Incyte Corporation; Iovance; Janssen Research & Development LLC; Janssen/Johnson & Johnson; Jasper Therapeutics; Jazz Pharmaceuticals Inc.; Kadmon; Karius; Kiadis Pharma; Kite, a Gilead Company; Kyowa Kirin; Legend Biotech; Magenta Therapeutics; Mallinckrodt Pharmaceuticals; Medexus Pharma; Merck & Co.; Mesoblast; Millennium, the Takeda Oncology Co.; Miltenyi Biotec Inc.; MorphoSys; Novartis Pharmaceuticals Corporation; Omeros Corporation; OptumHealth; Orca Biosystems Inc.; Ossium Health Inc.; Pfizer Inc.; Pharmacyclics LLC, an AbbVie Company; Pluristem; PPD Development LP; Sanofi; Sanofi-Aventis U.S. Inc.; Sobi Inc.; Stemcyte; Takeda Pharmaceuticals; Talaris Therapeutics; Terumo Blood and Cell Technologies; TG Therapeutics; Vertex Pharmaceuticals; Vor Biopharma Inc.; Xenikos BV. Data-sharing statement Data are available at https://www.cibmtr.org/ReferenceCenter/PubList/PubDsDownload/Pages/default.aspx

References 1. Young NS. Aplastic anemia. N Engl J Med. 2018;379(17):1643-1656. 2. Locasciulli A, Oneto R, Bacigalupo A, et al. Outcome of patients with acquired aplastic anemia given first line bone marrow transplantation or immunosuppressive treatment in the last decade: a report from the European Group for Blood and Marrow Transplantation (EBMT). Haematologica. 2007;92(1):11-18. 3. Patel BA, Groarke EM, Lotter J, et al. Long-term outcomes in patients with severe aplastic anemia treated with immunosuppression and eltrombopag: a phase 2 study. Blood. 2022;139(1):34-43. 4. Townsley DM, Scheinberg P, Winkler T, et al. Eltrombopag added to standard immunosuppression for aplastic anemia. N Engl J Med. 2017;376(16):1540-1550. 5. Bacigalupo A, Giammarco S, Sica S. Bone marrow

transplantation versus immunosuppressive therapy in patients with accquired severe aplastic anemia. Int J Hematol. 2016;104(2):168-174. 6. Valdez JM, Scheinberg P, Nunez O, Wu CO, Young NS, Walsh TJ. Decreased infection-related mortality and improved survival in severe aplastic anemia in the past two decades. Clin Infect Dis. 2011;52(6):726-735. 7. Georges GE, Doney K, Storb R. Severe aplastic anemia: allogeneic bone marrow transplantation as first-line treatment. Blood Adv. 2018;2(15):2020-2028. 8. Rosenfeld S, Follmann D, Nunez O, Young NS. Antithymocyte globulin and cyclosporine for severe aplastic anemia: association between hematologic response and long-term outcome. JAMA. 2003;289(9):1130-1135.

Haematologica | 108 Mese 2023

3306

ARTICLE - Conditional survival after HCT or IST in SAA

R. Nakamura et al.

9. Socie G, Rosenfeld S, Frickhofen N, Gluckman E, Tichelli A. Late clonal disease of treated aplastic anemia. Semin Hematol. 2000;37(1):91-101. 10. Socie G, Henry-Amar M, Bacigalupo A, et al. Malignant tumors occurring after treatment of aplastic anemia. European Bone Marrow Transplantation-Severe Aplastic Anaemia Working Party. N Engl J Med. 1993;329(16):1152-1157. 11. Viollier R, Passweg J, Gregor M, et al. Quality-adjusted survival analysis shows diferences in outcome after immunosuppression or bone marrow transplantation in aplastic anemia. Ann Hematol. 2005;84(1):47-55. 12. Buchbinder D, Nugent DJ, Brazauskas R, et al. Late effects in hematopoietic cell transplant recipients with acquired severe aplastic anemia: a report from the late effects working committee of the center for international blood and marrow transplant research. Biol Blood Marrow Transplant. 2012;18(12):1776-1784. 13. Martin PJ, Counts GW, Appelbaum FR, et al. Life expectancy in patients surviving more than 5 years after hematopoietic cell transplantation. J Clin Oncol. 2010;28(6):1011-1016. 14. Wingard JR, Majhail NS, Brazauskas R, et al. Long-term survival and late deaths after allogeneic hematopoietic cell transplantation. J Clin Oncol. 2011;29(16):2230-2239. 15. Socie G, Stone JV, Wingard JR, et al. Long-term survival and late deaths after allogeneic bone marrow transplantation. N Engl J Med. 1999;341(1):14-21. 16. Wong FL, Teh JB, Atencio L, et al. Conditional survival, causespecific mortality, and risk factors of late mortality after allogeneic hematopoietic cell transplantation. J Natl Cancer Inst. 2020;112(11):1153-1161. 17. Everitt B, Skrondal A. Standardized mortality rate (SMR). The Cambridge Dictionary of Statistics. 3rd ed. New York: Cambridge University Press; 2022. 18. Cox DR. Regression models and life-tables. J R Stat Soc Series B Stat Methodol. 1972;34(2):187-220. 19. Vaht K, Goransson M, Carlson K, et al. Incidence and outcome of acquired aplastic anemia: real-world data from patients diagnosed in Sweden from 2000-2011. Haematologica. 2017;102(10):1683-1690. 20. Baker KS, Leisenring WM, Goodman PJ, et al. Total body irradiation dose and risk of subsequent neoplasms following allogeneic hematopoietic cell transplantation. Blood. 2019;133(26):2790-2799. 21. Engel N, Rovo A, Badoglio M, et al. European experience and risk factor analysis of donor cell-derived leukaemias/MDS following haematopoietic cell transplantation. Leukemia. 2019;33(2):508-517. 22. Frickhofen N, Kaltwasser JP, Schrezenmeier H, et al. Treatment of aplastic anemia with antilymphocyte globulin and methylprednisolone with or without cyclosporine. The German

Aplastic Anemia Study Group. N Engl J Med. 1991;324(19):1297-1304. 23. Rosenfeld SJ, Kimball J, Vining D, Young NS. Intensive immunosuppression with antithymocyte globulin and cyclosporine as treatment for severe acquired aplastic anemia. Blood. 1995;85(11):3058-3065. 24. Scheinberg P, Nunez O, Weinstein B, Scheinberg P, Wu CO, Young NS. Horse versus rabbit antithymocyte globulin in acquired aplastic anemia. N Engl J Med. 2011;365(5):430-438. 25. Lengline E, Drenou B, Peterlin P, et al. Nationwide survey on the use of eltrombopag in patients with severe aplastic anemia: a report on behalf of the French Reference Center for Aplastic Anemia. Haematologica. 2018;103(2):212-220. 26. Olnes MJ, Scheinberg P, Calvo KR, et al. Eltrombopag and improved hematopoiesis in refractory aplastic anemia. N Engl J Med. 2012;367(1):11-19. 27. Desmond R, Townsley DM, Dumitriu B, et al. Eltrombopag restores trilineage hematopoiesis in refractory severe aplastic anemia that can be sustained on discontinuation of drug. Blood. 2014;123(12):1818-1825. 28. Peffault de Latour R, Kulasekarataj A, Iacobelli S, et al. Eltrombopag added to immunesuppression in severe aplastic anemia. N Engl J Med. 2022;386(1):11-23. 29. Rice C, Eikema DJ, Marsh JCW, et al. Allogeneic hematopoietic cell transplantation in patients aged 50 years or older with severe aplastic anemia. Biol Blood Marrow Transplant. 2019;25(3):488-495. 30. Anderlini P, Wu J, Gertsen I, et al. Cyclophosphamide conditioning in patients with severe aplastic anaemia given unrelated marrow transplantation: a phase 1-2 dose deescalation study. Lancet Haematol. 2015;2(9):e367-375. 31. Kuwatsuka Y, Kanda J, Yamazaki H, et al. A comparison of outcomes for cord blood transplantation and unrelated bone marrow transplantation in adult aplastic anemia. Biol Blood Marrow Transplant. 2016;22(10):1836-1843. 32. Peffault de Latour R, Rocha V, Socie G. Cord blood transplantation in aplastic anemia. Bone Marrow Transplant. 2013;48(2):201-202. 33. Zhang Y, Huo J, Liu L, et al. Comparison of hematopoietic stem cell transplantation outcomes using matched sibling donors, haploidentical donors, and immunosuppressive therapy for patients with acquired aplastic anemia. Front Immunol. 2022;13:837335. 34. DeZern AE, Zahurak ML, Symons HJ, et al. Haploidentical BMT for severe aplastic anemia with intensive GVHD prophylaxis including posttransplant cyclophosphamide. Blood Adv. 2020;4(8):1770-1779. 35. DeZern AE, Eapen M, Wu J, et al. Haploidentical bone marrow transplantation in patients with relapsed or refractory severe aplastic anaemia in the USA (BMT CTN 1502): a multicentre, single-arm, phase 2 trial. Lancet Haematol. 2022;9(9):e660-e669.

Haematologica | 108 Mese 2023

3307

ARTICLE - Hematopoiesis

Clonal hematopoiesis with DNMT3A and PPM1D mutations impairs regeneration in autologous stem cell transplant recipients Patrick Stelmach,1,2,3* Sarah Richter,1* Sandra Sauer,1* Margarete A. Fabre,4,5,6,7* Muxin Gu,4,5,6 Christian Rohde,1 Maike Janssen,1 Nora Liebers,1,8 Rumyana Proynova,1 Niels Weinhold,1 Marc S. Raab,1 Hartmut Goldschmidt,1 Birgit Besenbeck,1 Petra Pavel,9 Sascha Laier,9 Andreas Trumpp,2,3,10,11 Sascha Dietrich,1,12# George S. Vassiliou4,5,6# and Carsten Müller-Tidow1,8,12# Department of Medicine V, Heidelberg University Hospital, Heidelberg, Germany; 2Division of Stem Cells and Cancer, German Cancer Research Center (DKFZ) and DKFZ-ZMBH Alliance, Heidelberg, Germany; 3Heidelberg Institute for Stem Cell Technology and Experimental Medicine (HI-STEM GmbH), Heidelberg, Germany; 4Wellcome-MRC Cambridge Stem Cell Institute, University of Cambridge, Cambridge, UK; 5Department of Hematology, University of Cambridge, Cambridge, UK; 6Wellcome Sanger Institute, Wellcome Genome Campus, Cambridge, UK; 7Center for Genomics Research, Discovery Sciences, BioPharmaceuticals R&D, AstraZeneca, Cambridge, UK; 8National Center for Tumor Diseases (NCT), Heidelberg, Germany; 9Stem Cell Laboratory, Institute of Clinical Transfusion Medicine and Cell Therapy Heidelberg GmbH, Heidelberg, Germany; 10Faculty of Biosciences, Heidelberg University, Heidelberg, Germany; 11German Cancer Consortium (DKTK), Heidelberg, Germany and 12 Molecular Medicine Partnership Unit, European Molecular Biology Laboratory (EMBL), Heidelberg, Germany 1

Correspondence: C. Müller-Tidow carsten.mueller-tidow@med.uni-heidelberg.de G. S. Vassiliou

gsv20@cam.ac.uk Received: Accepted: Early view:

February 21, 2023 June 19, 2023. June 29, 2023.

PS, SR, SS and MAF contributed equally as first authors. SD, GSV and CM-T contributed equally as senior authors.

Abstract Clonal hematopoiesis (CH) is an age-related condition driven by stem and progenitor cells harboring recurrent mutations linked to myeloid neoplasms. Currently, potential effects on hematopoiesis, stem cell function and regenerative potential under stress conditions are unknown. We performed targeted DNA sequencing of 457 hematopoietic stem cell grafts collected for autologous stem cell transplantation (ASCT) in myeloma patients and correlated our findings with high-dimensional longitudinal clinical and laboratory data (26,510 data points for blood cell counts/serum values in 25 days around transplantation). We detected CHrelated mutations in 152 patients (33.3%). Since many patients (n=54) harbored multiple CH mutations in one or more genes, we applied a non-negative matrix factorization (NMF) clustering algorithm to identify genes that are commonly co-mutated in an unbiased approach. Patients with CH were assigned to one of three clusters (C1-C3) and compared to patients without CH (C0) in a gene specific manner. To study the dynamics of blood cell regeneration following ASCT, we developed a time-dependent linear mixed effect model to validate differences in blood cell count trajectories amongst different clusters. The results demonstrated that C2, composed of patients with DNMT3A and PPM1D single and co-mutated CH, correlated with reduced stem cell yields and delayed platelet count recovery following ASCT. Also, the benefit of maintenance therapy was particularly strong in C2 patients. Taken together, these data indicate an impaired regenerative potential of hematopoietic stem cell grafts harboring CH with DNMT3A and PPM1D mutations.

Introduction Hematopoietic stem cells (HSC) acquire somatic mutations in an age-dependent manner.1 Some mutations confer a Darwinian fitness advantage to the cell and fuel clonal outgrowth.2 The presence of driver mutations in blood cells from otherwise healthy individuals characterizes the common age-related phenomenon termed clonal

hematopoiesis (CH). These clonal populations predispose individuals to an increased risk of developing blood cancer (0.5-1% per year in unselected cohorts) and cardiovascular disease.3,4 Clonal expansion results from mutations in a restricted set of leukemia-associated genes.5,6 DNMT3A (DNA methyltransferase 3A) mutations are the most common drivers of this state7, 8. TET2 (Tet methylcytosine dioxygenase 2) and PPM1D (protein phosphatase Mn2+/Mg2+-de-

Haematologica | 108 December 2023

3308

ARTICLE - CH impairs regenerative potential of stem cells

P. Stelmach et al.

pendent 1D) are also among the most frequently mutated genes.7,9 TET2 mutations have been linked to a pro-inflammatory phenotype mediated by mutated progenitor cells that can contribute to atherogenesis10 and there is evidence that an enhanced inflammatory response in TET2 mutated mice correlates with disease progression of myeloid neoplasms.11 Mutations in the PPM1D gene have been associated with prior exposure to cytotoxic therapy.12-15 Multiple myeloma (MM) is a plasma cell neoplasm and the standard of care therapy for newly diagnosed and eligible patients includes induction therapy followed by high-dose chemotherapy with melphalan and subsequent autologous stem cell transplantation (ASCT). Triplet combinations including the proteasome inhibitor bortezomib, dexamethasone and an immunomodulatory drug demonstrated efficacy with high response rates prior to ASCT.16 The standard of care also includes maintenance therapy.17-19 ASCT provides an excellent opportunity to compare the regenerative potential of CH versus wild-type HSC. Previously, the presence of CH-related driver mutations in autologous stem cell grafts and bone marrow samples of MM patients was found to be associated with inferior survival rates and an increased risk for myeloma progression.20,21 Currently, the effects of individual CH mutations on regenerative potential of HSC and effects on specific cell lines are unknown, although recent data indicate a longer time to neutrophil and platelet engraftment following ASCT in lymphoma patients with CH.22 Here, we analyzed a large cohort of ASCT patients and high-dimensional data warehouse data to identify associations between CH and blood cell regeneration in patients with MM.

University of Heidelberg (reference no. S-850/2021). A total of 165 patients were treated with high-dose chemotherapy and ASCT as part of a clinical trial. Among these, six patients were treated in the GMMG-HD3 phase III trial (clinicaltrials gov. Identifier: NCT00028886), 60 patients in the GMMG-HD4 phase III trial (clinicaltrialsregister.eu; EudraCT No. 2004-000944-26), 62 patients in the GMMGMM5 phase III trial (EudraCT No. 2010-019173-16) and 37 patients in the GMMG-HD6 phase III trial (clinicaltrials gov. Identifier: NCT02495922).

Methods Patients and stem cell grafts Mobilized stem cell products (peripheral blood with CD34+ cells) from 457 MM patients were harvested by leukapheresis at the University Hospital Heidelberg between 2004 and 2019. The inclusion criteria were: diagnosis of MM, patients were harvested for their first ASCT and did not receive granulocyte colony-stimulating factor (G-CSF) after re-infusion. In patients with tandem ASCT, data on blood cell count recovery refer to the first transplantation. Clinical data was analyzed through May, 2022. Stem cell mobilization was performed with G-CSF (filgrastim) combined with either cyclophosphamide monotherapy (2 g/m2) or a combination of cyclophosphamide (2 g/m2), doxorubicin (60 mg/m2) and dexamethasone (80 mg). Cytogenetic high risk was defined as the detection of one of the following genetic aberrations by fluorescence in situ hybridization (FISH): t(4;14), t(14;16) or del(17p) or chromosome 1q21 gain/amplification. All patients provided written informed consent to the use of their biomaterial and clinical data. The project was approved by the Ethics Committee of the

Isolation of genomic bulk DNA and targeted bulk DNA sequencing In order to obtain cellular material for isolation of genomic DNA, frozen material was scratched off from frozen apheresis products on dry ice to obtain a cell suspension with a volume of almost 300 μL. Automated DNA isolation was performed using the ReliaPrepTM Large Volume HT genomic DNA Isolation System (Promega, WI, USA) according to manufacturer’s instructions. Genomic DNA was isolated in Tris/EDTA (TE) buffer and used to prepare sequencing libraries. We performed targeted DNA sequencing (Illumina Novaseq) of 457 bulk stem cell products after target enrichment for 56 genes implicated in myeloid malignancies (Agilent SureSelect ELID 3156971; Online Supplementary Table S2). Data warehouse The data warehouse (Department of Hematology and Oncology, Heidelberg University Hospital) centralizes department-specific data from the hospital information system and peripheral systems in an automated way. To date, it contains clinical data of approximately 120,000 patients dating back to 2004. In addition to basic data, it also includes diagnostic results and treatment-related data. Since every patient contact is stored together with the collected information, detailed and dense longitudinal follow-up data are included. Non-negative matrix factorization clustering and statistical analyses The detailed description of non-negative matrix factorization (NMF) clustering and all statistical analyses can be found in the Online Supplementary Appendix. All P values were two-sided and significance levels set at P<0.05. Calculations were done using R version 4.0.1 R (Foundation for Statistical Computing, Vienna, Austria). The waterfall plot visualizing the mutational landscape and the lollipop plots were created using the Maftools package in R23.

Results Mutational landscape of clonal hematopoiesis We performed targeted sequencing of 56 genes impli-

Haematologica | 108 December 2023

3309

ARTICLE - CH impairs regenerative potential of stem cells

P. Stelmach et al.

Continued on following page. Haematologica | 108 December 2023

3310

ARTICLE - CH impairs regenerative potential of stem cells

P. Stelmach et al.

Figure 1. Mutational landscape of clonal hematopoiesis at the time of leukapheresis and clusters identified by the non-negative matrix factorization clustering algorithm. Scheme of study design. (A) Scheme of study design. (B) Density of data points for platelet counts in the myeloma cohort in 25 days around transplant (from day -4 to day +20) sourced from our data warehouse. The conditioning chemotherapy with high-dose melphalan was given on day -3 and day -2 and re-transfusion of stem cells was performed on day 0. (C) Co-mutation plot illustrating all mutations detected in N=152 patients. Each column represents a single patient. The bar graph on the right shows the percentage of the different variant classes for each gene out of all detected mutations. The top bar graph summarizes the total mutation burden (TMB) in each patient subdivided by variant class. (D) Clusters identified by the non-negative matrix factorization (NMF) clustering algorithm including patients with mutations in the 8 most frequently mutated genes (N=129). The method clusters mutations that often occur together based on a co-occurrence matrix. The number of patients included in the respective cluster is indicated above. (E) The heatmap illustrates the affiliation of each patient to the 3 different clusters (C1-3) based on mixture coefficients and each column represents a single patient. The continuous color scale indicates the affiliation to the respective cluster. Finally, the patients have been hard clustered based on the cluster with the highest affiliation. A total of 129 patients were included in the three NMF clusters. VAF: variant allele frequency; ASCT: autologous stem cell transplan-

cated in myeloid malignancies (Online Supplementary Tables S1 and S2) on DNA purified from unsorted MM stem cell grafts and correlated our findings with high dimensional longitudinal clinical and laboratory data (Figure 1A, B; Online Supplementary Tables S1 and S4). Overall, we detected mutations at a variant allele frequency (VAF) ≥0.01 in 152 patients (33.3%), with their prevalence increasing with age (Online Supplementary Figure S1A, B; Online Supplementary Table S3). In 98 patients a single mutation was detected, whereas 54 patients harbored >1 mutation either in a different or the same gene (Figure 1C). In line with published data, DNMT3A was the most frequently mutated gene with 78 mutations, of which 56 were missense mutations including six at the R882 hotspot (Online Supplementary Figure S1C). The median VAF of DNMT3A mutations was 0.02 (range, 0.01-0.30). TET2 and PPM1D were the next most frequently mutated genes (Figure 1C). PPM1D mutations (n=15) were localized in exons 5 and 6 (Online Supplementary Figure S1D) and had a median VAF of 0.02 (range, 0.01-0.23). Non-negative matrix factorization identified three mutational clusters (C1-C3) Different mutations detected in the same patient may have different individual effects, which could also depend on their VAF. We applied a NMF clustering algorithm on the co-occurrence matrix of mutations, a mathematical dimensionality reduction method to identify clusters of genes that often co-mutated in the same sample (Figure 1D, E).24 This allowed an unbiased clustering of mutations unaffected by prior knowledge on biological functions or clinical associations and also considered patients with a more complex mutational profile. Three clusters were identified and each of the 129 patients harboring mutations in the eight most frequently mutated genes in our cohort was assigned to one of the clusters (C1-C3, Figure 1D, E). Thereby, a comparison between patients with (C1C3, n=129) and without (C0, n=305) CH by cluster was facilitated. A total of 23 patients could not be assigned to a cluster as they harbored mutations in infrequently mutated genes.

DNMT3A and PPM1D mutations (C2) were associated with reduced stem cell yields and lower pre-transplant blood platelet counts C2 was composed of patients with DNMT3A and PPM1D single and co-mutated CH. We observed that, compared to patients without CH (C0), patients in C2 had a lower number of CD34+ stem cells harvested (median for C0 vs. C2, 7.50 [range, 0.59-44.90] vs. 4.65 [range 0.35-23.00] x106 CD34+ stem cells /kg; P=0.009; Figure 2A; Table 1). Further, multivariate median regression confirmed C2 (P=0.04), patient’s age (P<0.001) and the application of plerixafor (P<0.001) as independent adverse predictors of the CD34+ stem cell number (Figure 2B). This observation suggested that HSC function and hematopoiesis were impaired in patients harboring DNMT3A- and PPM1D-mutated hematopoietic stem and progenitor cells (HSPC). Overall, patients mobilized with combination chemotherapy (cyclophosphamide, doxorubicin and dexamethasone [CAD]) harvested more CD34+ stem cells (Figure 2C). We further investigated if blood cell counts for leukocytes, neutrophils and platelets as well as hemoglobin levels prior to conditioning chemotherapy with melphalan differed between the clusters (Figures 2E, F; Online Supplementary Figure S1E). We observed that patients in C2 had significantly lower blood platelet counts before ASCT (median platelet count/nL for C0 vs. C2, 257 [range, 82785] vs. 226 [range, 37-412]; P= 0.0016; Figure 2E; Table 1). This result, confirmed by multivariate regression analysis adjusted for age, MM cytogenetic risk and therapy response before ASCT (C2: P=0.0008; Figure 2F), is in line with the impaired stem cell yield. In contrast, there was no association between C2 mutations and pre-transplant hemoglobin values, leukocyte or neutrophil counts (Online Supplementary Figure S1E). Since these models did not consider the VAF of individual mutations, we analyzed patients assigned to C2 and used the highest VAF mutation per patient. Interestingly, Figure 2G is indicative for an anti-proportional correlation between VAF and pretransplant platelet counts in C2 (blue line) and the VAF correlated with lower pretransplant platelets in a multivariate linear regression analysis (P=0.04; Figure 2H). But, the VAF was not associated with the

Haematologica | 108 December 2023

3311

ARTICLE - CH impairs regenerative potential of stem cells

P. Stelmach et al.

number of harvested stem cells in C2 patients (Online Supplementary Figure S4C). Further, in a multivariate Tobit regression model, C2 patients displayed significantly higher serum CRP levels prior to melphalan conditioning chemotherapy (P=0.018; Figure 2D) compared to patients without CH, suggesting increased levels of inflammation in these patients at baseline.

tients´ age and transplanted CD34+ cell numbers. C2 mutations were associated with a lower platelet nadir following transplant (median platelet nadir/nL C0 vs. C2, 11 [range, 2-46] vs. 9 [range, 3-27]; P=0.001, Figure 3E) and this was validated in a multivariate linear regression model (P=0.009; Figure 3F). In contrast, there was no difference for leukocyte counts, neutrophils or hemoglobin values (Online Supplementary Figures S3A, B; S4A, B and S11A). We also analyzed the number of platelet transfusions within 20 days after transplant (Figure 3G). C2 patients compared to C0 patients were projected to have received 1.41 (range, 1.04-1.87; P=0.022; Figure 3H) times as many platelet transfusion units. This result was confirmed when considering only patients with DNMT3A single-mutated CH (Online Supplementary Figure S10). In contrast, we did not observe significant differences in the number of red blood cell transfusions (data not shown). Finally, we also analyzed the effect of VAF on post-transplant platelet counts. But, the VAF was not found to influence the regeneration of peripheral blood platelet counts after transplant (Online Supplementary Figure S4D, E). Collectively, our data provide evidence that there

DNMT3A and PPM1D mutations (C2) associated with a delayed regeneration of blood platelet counts after transplant In the 25 days around ASCT (from day -4 to day +20), our data warehouse contained 26,510 blood cell counts/serum analyte values. These high-dimensional data enabled us to model the impact of CH mutations on the dynamics of blood cell regeneration over time following transplant (Figure 3A; Online Supplementary Figure S2A-C). Therefore, we analyzed the trajectories of peripheral blood cell counts over time after ASCT with respect to different mutational clusters (Figure 3A) and developed a time-dependent linear mixed effect model (Figure 3B, D). This model validated a delayed platelet count engraftment in C2 patients (P=0.02; Figure 3B, C), independent of pa-

Continued on following page. Haematologica | 108 December 2023

3312

ARTICLE - CH impairs regenerative potential of stem cells

P. Stelmach et al.

Figure 2. C2 mutations were associated with reduced stem cell yields and lower pretransplant platelet counts in the peripheral blood. (A) Box plot illustrating the number of harvested stem cells (normalized to the days of leukapheresis, see method description) per cluster. C0: patients in whom no clonal hematopoiesis (CH) mutation was detected. The respective median value is indicated. (B) Forest plot visualizing the output of a quantile (median) regression model for the number of harvested CD34+ stem cells/kg (normalized to the days of apheresis) including the specified independent variables. The model was applied since the data is skewed and bimodal distributed. The reference for the mutated patients (C1-C3) are the patients without CH mutations (C0). The reference for the shown remission states prior to autologous stem cell transplantation (ASCT) is the patient group that achieved a complete remission (CR) after induction chemotherapy. The forest plot visualizes the estimates/coefficients (slope of the regression curve) and their respective confidence intervals. Statistical significance is indicated if the estimate is flagged with one or more stars. (C) Box plot illustrating the number of harvested stem cells in patients mobilized with combination chemotherapy (cyclophosphamide, doxorubicin and dexamethasone [CAD]) and patients mobilized with cyclophosphamide monotherapy. The respective median value is indicated. (D) Tobit regression model of C-reactive protein (CRP) values on day -4. This regression model was applied since CRP serum values are censored <2 mg/dL, the data are extremely skewed and have an excess of zeros.42 (E) Box plot visualizing platelet counts before transplantation per cluster. The respective median value is indicated. (F) Forest plot visualizing the output of a linear regression model for peripheral blood platelet counts before transplant including the specified independent variables. The platelet counts have been logtransformed to obtain a normal distribution. The plot illustrates the estimates/coefficients and their respective confidence intervals and statistical significance is indicated if the value is flagged with 1 or more stars. The reference for patients in C13 is the patient group without CH mutations (C0). The reference for the shown remission states prior to ASCT is the patient group that achieved CR after induction chemotherapy. (G) Scatter plot of pretransplant platelets and maximum variant allele frequency (VAF) with regression lines per cluster. Although the regression line for C2 (blue) is indicative for an anti-proportional correlation, Spearman’s correlation indicates no significant relationship between platelets and VAF per cluster (R: Spearman correlation coefficient). The respective P value for the correlation per cluster is indicated. (H) Forest plot visualizing the output of a linear regression model for peripheral platelet counts prior to ASCT including the specified independent variables. Maximum VAF (C2): mutation with the highest VAF per patient within C2. The VAF was included as a continuous covariate. The platelet values have been log-transformed to obtain a normal distribution. Cytogenetic NA: cytogenetic data not available; SD: stable disease; PD: progressive disease; PR: partial response.

Haematologica | 108 December 2023

3313

ARTICLE - CH impairs regenerative potential of stem cells

P. Stelmach et al.

is an altered regenerative potential in C2 patients irrespective of the clone size.

tients and those without CH (C2 vs. C0, median OS 6.39 years vs. not reached [NR]; P=0.048; Online Supplementary Figure S6A). This difference disappeared for patients treated with maintenance therapy (Online Supplementary Figure S6B), even when comparing C2 and C3 (TET2, SMC3 and SF3B1 single and co-mutated) patients who displayed the greatest difference in the absence of maintenance (C2 vs. C3, no maintenance therapy, median OS 6.39 years vs. 11.87 years; P=0.013; Online Supplementary Figure S6A,B). Further, among C2 patients, maintenance therapy resulted in a significant difference in PFS and OS (maintenance therapy no vs. yes, median progression-

Patients with DNMT3A and PPM1D mutated clonal hematopoiesis (C2) benefited from maintenance therapy Comparing survival probabilities between overall CH (all mutations) and patients without CH, there was no difference in PFS or OS (Online Supplementary Figure S5A,B), however, the remission status before ASCT influenced the OS (Online Supplementary Figure S9). Considering only patients not treated with maintenance therapy, there was a difference in OS between C2 pa-

Table 1. Patient characteristics per cluster (C1-C3). C0 N=305

C1 N=20

C2 N=62

C3 N=47

All N=434

56.9 (8.73) 58.0 (28.0, 72.0)

62.6 (6.48) 64.5 (51.0, 72.0)

61.2 (7.60) 63.0 (39.0, 72.0)

59.6 (7.29) 61.0 (47.0, 71.0)

58.1 (8.52) 59.0 (28.0, 72.0)

Sex, N (%) Male Female

192 (63.0) 113 (37.0)

12 (60.0) 8 (40.0)

32 (51.6) 30 (48.4)

26 (55.3) 21 (44.7)

262 (60.4) 172 (39.6)

MM cytogenetic high-risk, N (%) No Yes NA

120 (39.3) 104 (34.1) 81 (26.6)

10 (50.0) 2 (10.0) 8 (40.0)

25 (40.3) 19 (30.6) 18 (29.0)

17 (36.2) 17 (36.2) 13 (27.7)

172 (39.6) 142 (32.7) 120 (27.6)

Mobilization chemotherapy, N (%) CAD Cyclophosphamide mono Other NA

176 (57.7) 13 (4.3) 4 (1.3) 112 (36.7)

12 (60.0) 0 (0) 1 (5.0) 7 (35.0)

40 (64.5) 1 (1.6) 2 (3.2) 19 (30.6)

27 (57.4) 0 (0) 2 (4.3) 18 (38.3)

255 (58.8) 14 (3.2) 9 (2.1) 156 (35.9)

Characteristic Age in years at apheresis Mean (SD) Median (min, max)

Harvested CD34+ stem cells x106/kg Mean (SD) Median (min, max)

8.47 (6.12) 7.39 (5.79) 6.70 (5.72) 8.44 (6.41) 8.17 (6.10) 7.50 (0.590, 44.9) 5.89 (0.900, 19.0) 4.65 (0.350, 23.0) 7.80 (1.06, 32.1) 7.12 (0.350, 44.9)

Platelet transfusions post-ASCT to day 20, N Mean (SD) Median (min, max)

0.692 (0.817) 1.00 (0, 5.00)

1.15 (1.42) 1.00 (0, 6.00)

1.06 (1.19) 1.00 (0, 6.00)

0.766 (0.914) 1.00 (0, 5.00)

0.774 (0.932) 1.00 (0, 6.00)

Hospitalization, N of days Mean (SD) Median (min, max)

16.4 (2.78) 16.0 (10.0, 32.0)

17.9 (3.37) 17.0 (14.0, 28.0)

16.7 (2.85) 17.0 (12.0, 24.0)

16.6 (3.77) 15.0 (12.0, 31.0)

16.6 (2.95) 16.0 (10.0, 32.0)

Maintenance therapy, N (%) No Yes

98 (32.1) 207 (67.9)

9 (45.0) 11 (55.0)

25 (40.3) 37 (59.7)

17 (36.2) 30 (63.8)

149 (34.3) 285 (65.7)

Platelet nadir post-ASCT /nL Mean (SD) Median (min, max)

12.1 (6.38) 11.0 (2.00, 46.0)

9.65 (5.29) 8.00 (4.00, 22.0)

9.58 (4.69) 9.00 (3.00, 27.0)

11.6 (5.05) 11.0 (2.00, 29.0)

11.5 (6.04) 10.0 (2.00, 46.0)

Platelet count before ASCT /nL Mean (SD) Median (min, max) NA, N (%)

267 (78.5) 257 (82.0, 785) 3 (1.0)

250 (84.2) 240 (121, 414) 0 (0)

232 (74.1) 226 (37.0, 412) 0 (0)

256 (89.3) 237 (107, 498) 0 (0)

260 (80.1) 249 (37.0, 785) 3 (0.7)

C-reactive protein before ASCT, mg/dL Mean (SD) Median (min, max) NA, N (%)

3.76 (4.50) 2.00 (2.00, 35.4) 55 (18.0)

5.21 (6.47) 2.00 (2.00, 21.5) 7 (35.0)

8.80 (23.8) 2.00 (2.00, 130) 13 (21.0)

3.20 (2.62) 2.00 (2.00, 13.8) 4 (8.5)

4.44 (9.82) 2.00 (2.00, 130) 79 (18.2)

MM: multiple myeloma; SD: standard deviation; min: minimum; max: maximum; NA: not available; ASCT: autologous stem cell transplantation; CAD: cyclophosphamide, doxorubicin and dexamethasone. Haematologica | 108 December 2023

3314

ARTICLE - CH impairs regenerative potential of stem cells

P. Stelmach et al.

free survival [PFS] 1.79 vs. 7.26 years; P=0.00079; median OS 6.39 years vs. 10.76 years; P=0.013; Figure 4A; Online Supplementary Figure S7). A multivariate cox regression model including C0 and C2 patients confirmed that C2 patients benefited significantly from maintenance therapy regarding PFS independent of their cytogenetic risk status (hazard ratio [HR]: 0.42; P=0.013; Online Supplementary Figure S4B). Given that maintenance strategies used during the recent two decades differ profoundly in their efficacy, conclusions drawn from this have to consider the heterogenous maintenance types in our cohort (Online Supplementary Figure S6E, F). Therefore, we validated these findings in the patient group treated with lenalidomide as a standard of care in MM maintenance therapy (Online Supplementary Figure S8). Taken together, our data indicate that C2 patients particularly benefit from maintenance therapy. There were four cases of therapy-related myeloid neoplasms (t-MN) in total reported in our cohort (0.7%). Moreover, two patients were diagnosed with B-ALL. In all but one of these patients, CH mutations were detected in the stem cell product.

Here, we provide evidence that the presence of DNMT3A and PPM1D mutations in MM stem cell grafts is associated with an impaired HSC function. This was evident in the reduced numbers of harvested HSC and lower platelet counts in the peripheral blood. Since low platelet counts before stem cell mobilization are also a known risk factor for poor mobilizer, both observations are presumably related. The findings indicate that the presence of CH mutations signifies the presence of stressed hematopoiesis around ASCT in myeloma patients. Compared to prior studies investigating CH, our results show that mutated genes in CH obviously harbor different significance with regard to regenerative potential in transplant-related stress hematopoiesis. While C2 mutant HSPC respond to hematopoietic stress, there is no evidence related to mutations in other genes. The loss of Dnmt3a in murine HSC allows regeneration over successive transplants in mice.25 However, while simultaneously expanding HSC numbers in the bone marrow,

Discussion

Continued on following page. Haematologica | 108 December 2023

3315

ARTICLE - CH impairs regenerative potential of stem cells

P. Stelmach et al.

Figure 3. C2 mutations were associated with delayed regeneration of platelet counts in the peripheral blood after transplant. The trajectories (grey lines) show the effect of time on platelet counts in the peripheral blood within 20 days after transplant. The bold colored lines show the average of platelet counts per cluster. (B) Time-dependent linear mixed effect model to day 50 including the specified independent variables. The effect estimates and their respective confidence intervals are shown in the forest plot and statistical significance is indicated if the value is flagged with one or more stars. (C) Effect plot illustrating the effect of time on platelet counts from day 0 to day 50 after transplant for patients without clonal hematopoiesis (CH) mutations (C0, red line) and those affiliated with C2 (grey line). (D) Effect of time on peripheral platelet counts from day 0 to day 50 depending on the number of retransfused CD34+ cells /kg normalized to the days of leukapheresis. (E) Box plot illustrating the platelet nadir following transplant per cluster. C0: patients in whom no CH mutation was detected. (F) Forest plot visualizing a linear regression model for the lowest platelet count (nadir) following high-dose chemotherapy and transplant including the specified independent variables. The reference for C1-3 patients are the patients without CH mutations (C0) and the reference for the shown remission states prior to autologous stem cell transplantation (ASCT) is the patient group that achieved a complete remission (CR) after induction chemotherapy. (G) Box plot showing the sum of platelet transfusions within 20 days post-ASCT per cluster. (H) Forest plot visualizing a Poisson regression for the number of platelet transfusions within 20 days following ASCT including the specified independent variables. The regression model was applied since the data corresponded to a Poisson distribution and constituted count data with a high number of zero values. The platelet values have been log-transformed to obtain a normal distribution. The log-mean values and the respective confidence intervals are shown. Cytogenetic NA: cytogenetic data not available; SD: stable disease; PD: progressive disease; PR: partial response.

a loss of Dnmt3a in mice also impairs HSC differentiation over serial transplantation.26 Together, there is evidence that certain DNMT3A variants confer a repopulating advantage with transplantation in mice and humans.13,25 Pre-

viously, it has been shown that DNMT3A and PPM1D clones respond in the opposite way in post-transplant regenerative hematopoiesis of lymphoma patients.13 While DNMT3A mutant clones often expanded, PPM1D mutant

Haematologica | 108 December 2023

3316

ARTICLE - CH impairs regenerative potential of stem cells

P. Stelmach et al.

clones decreased in size, indicating that DNMT3A variants confer a repopulation advantage with transplantation.13 In contrast, NSG mice transplanted with bone marrow of individuals with mutant DNMT3A showed stable kinetics of DNMT3A mutant clones over several months.27 In our previous study, we showed that Dnmt3a R882H-mutated murine HSC are impaired in hematopoietic potential after transplant and azacytidine treatment.28 Furthermore, a recent study demonstrated that DNMT3A R882 mutations in stem cell grafts of patients with multiple myeloma skew a megakaryocytic-erythroid biased differentiation.29 Together, there may be species-related differences, however, our findings suggest an impaired HSC function related to C2 mutations. Based on prior in vitro studies, the preferential adverse effects on platelet counts are not fully explained. Further investigations might elucidate the underlying mechanisms. The expression of PPM1D is induced by genotoxic stress such as ionizing radiation and correlates with upregulation of the tumor suppressor protein p53.30-32 Upon genotoxic stress from cytotoxic therapy, PPM1D-and TP53-mutated clones expand tremendously.13,33 The biological consequences of PPM1D mutations in hematopoietic cells and how mutations in exon 6 of PPM1D confer a fitness advantage to these cells in presence of chemotherapy have been comprehensively studied and there is strong evidence that PPM1D mutations improve HSC survival and result in clonal expansion.14,31 Considering our results, the

impact of C2 mutations might be indirectly associated with poor hematopoietic reserve, particularly given the small VAF of the PPM1D clones. Interestingly, C2 patients showed higher baseline CRP blood values before ASCT. Increased CRP values may indicate the connection between C2 mutations, inflammation, stress hematopoiesis and impaired HSC functions. So far, a pro-inflammatory signature has been associated with TET2-mutated HSPC. Mechanistic studies and more extensive cytokine data may further elucidate the link between pro-inflammatory signatures and impaired regenerative potential of stem cells. The VAF was related to pre-transplant platelet counts, however, there was no correlation with VAF after transplant. Collectively, these findings indicate an altered regenerative potential beyond the size of C2 clones suggesting that the mutation effects are secondary mediated by inflammation and stress. Of note, the analyses were based on limited numbers of specimens in each cluster. Thus, with larger numbers, smaller effects might become more apparent. DNMT3A mutations have been detected in isolated myeloma plasma cells.34,35 Thus, one limitation of our study is that we cannot exclude that in some patients the somatic mutations detected by bulk sequencing result from a contamination of the graft with residual myeloma cells. However, DNMT3A mutations in myeloma cells are rare and the mutational landscape of MM is heterogenous.36,37 Our results are indicative that C2 patients benefit from mainten-

Continued on following page. Haematologica | 108 December 2023

3317

ARTICLE - CH impairs regenerative potential of stem cells

P. Stelmach et al.

Figure 4. The benefit of maintenance therapy was particularly strong in C2 patients. (A) Progression-free survival (PFS) and overall survival (OS) from the day of transplant for C0 and C2 stratified by maintenance therapy (grey: no maintenance therapy, orange: maintenance therapy) for C0 and C2 patients. C0: patients in whom no clonal hematopoiesis (CH) mutation was detected. The respective P value calculated by log-rank test is indicated. (B) Multivariate Cox regression model (calculated from day +90 to overcome immortal time bias) including C2 patients and patients without CH mutations (C0). C0 patients are the reference and are, therefore, not shown. The model showed a statistically significant interaction between C2 patients and maintenance therapy. C2 patients may benefit from maintenance therapy particularly strong regarding PFS. This difference in the cumulative incidence of myeloma relapse among patients in C0 and C2 is shown below. The presence of cytogenetic high-risk lesions was included as a covariate, thereby correcting for cytogenetic high-risk status. NA: not analyzed; cytogenetic NA: cytogenetic data not available; SD: stable disease; PD: progressive disease; PR: partial response.

ance therapy. A hypothesis for the impact of maintenance therapy in C2 patients could be that their native hematopoiesis is exhausted or unfit, thus providing reduced competition to MM. For PPM1D clones this can be argued, as they thrive in such an environment. For DNMT3A, we showed recently that these clones lose fitness in old age.38 There is a growing body of evidence, that clonal selection of pre-existing mutant HSC occurs under the stress of cytotoxic therapy and somatic mutations in genes involved in the DNA damage response (DDR) pathway are enriched in

the blood of patients formerly exposed to chemo- and radiation therapies.12,39 Recently, it was shown that lenalidomide treatment provides a selective advantage to TP53 mutant HSC thereby promoting therapy-related neoplasms.40 Hematological toxicity during lenalidomide treatment is a major obstacle and the presence of large CH clones in the peripheral blood of lymphoma patients before treatment was associated with development of severe hematological toxicity.41 Treatment caused hematopoietic clones with TP53 mutations expand over time, while clones with DNMT3A

Haematologica | 108 December 2023

3318

ARTICLE - CH impairs regenerative potential of stem cells

P. Stelmach et al.

mutations are more stable.41 In a smaller cohort of lymphoma patients, larger CH clones (VAF ≥5%) were related to lower hemoglobin levels and platelet counts and these data are in line with our current findings.41 Stress hematopoiesis occurs in multiple clinical settings. In conditions such as sepsis and multiple organ failure, hematopoiesis is frequently impaired, although the causal mechanisms are often unknown. The effects of CH mutations in stress hematopoiesis are clinically relevant and might thus also affect blood cell homeostasis in other clinical conditions. Thus, ASCT as a routine clinical intervention may be considered as a model system for the effects of human hematopoiesis under severe stress conditions. Taken together, our data demonstrated mutation-specific effects of stress hematopoiesis upon ASCT in CH carriers. These findings may contribute to better assess patients´ risks and presumed benefits of ASCT and stress hematopoiesis in general.

samples and edited the manuscript. NL and RP performed data retrieval from the data warehouse and edited the manuscript. NW performed and interpreted analyses and edited the manuscript. MSR, AT and HG interpreted results and edited the manuscript. BB, PP and SL processed patient samples. SD designed and supervised statistical analyses and edited the manuscript. GSV designed the study, supervised sequencing analyses, supervised the project and wrote the manuscript. CMT designed the study, supervised the project, acquired funding and wrote the manuscript.

Disclosures MAF is an employee and stockholder of AstraZeneca. All other authors have no conflicts of interest to disclose. Contributions PS and SR designed the study, performed and interpreted analyses and wrote the manuscript. SS designed the study, processed patient samples and edited the manuscript. MAF performed sequencing and analyses, interpreted results and edited the manuscript. MG performed sequencing and analyses and edited the manuscript. CR performed analyses and edited the manuscript. MJ processed patient

Funding PS is funded by a fellowship of the DKFZ Clinician Scientist Program, supported by the Dieter Morszeck Foundation. MAF was funded by a Wellcome Clinical Research Fellowship (WT098051). NL was supported by a Heidelberg School of Oncology (HSO2) fellowship from the National Center for Tumor Diseases (NCT) Heidelberg. NW, MSR and HG were supported by the Dietmar-Hopp Foundation. GSV is funded by a Cancer Research UK Senior Cancer Fellowship (C22324/A23015) and work in his laboratory is also funded by the European Research Council, Leukemia and Lymphoma Society, Rising Tide Foundation for Clinical Cancer Research, Kay Kendall Leukemia Fund, Blood Cancer UK and Wellcome Trust. This study was supported by research funding (MU1328/23-1) from the German Research Foundation (DFG). Data-sharing statement For original data, please contact the corresponding author.

References 1. Jaiswal S, Fontanillas P, Flannick J, et al. Age-related clonal hematopoiesis associated with adverse outcomes. N Engl J Med. 2014;371(26):2488-2498. 2. Bowman RL, Busque L, Levine RL. Clonal hematopoiesis and evolution to hematopoietic malignancies. Cell Stem Cell. 2018;22(2):157-170. 3. Genovese G, Kahler AK, Handsaker RE, et al. Clonal hematopoiesis and blood-cancer risk inferred from blood DNA sequence. N Engl J Med. 2014;371(26):2477-2487. 4. Jaiswal S, Natarajan P, Silver AJ, et al. Clonal hematopoiesis and risk of atherosclerotic cardiovascular disease. N Engl J Med. 2017;377(2):111-121. 5. Steensma DP, Bejar R, Jaiswal S, et al. Clonal hematopoiesis of indeterminate potential and its distinction from myelodysplastic syndromes. Blood. 2015;126(1):9-16. 6. Jaiswal S, Ebert BL. Clonal hematopoiesis in human aging and disease. Science. 2019;366(6465):eaan4673. 7. Challen GA, Goodell MA. Clonal hematopoiesis: mechanisms driving dominance of stem cell clones. Blood. 2020;136(14):1590-1598. 8. Huang YH, Chen CW, Sundaramurthy V, et al. Systematic profiling of DNMT3A variants reveals protein instability mediated by the DCAF8 E3 ubiquitin ligase adaptor. Cancer

Discov. 2022;12(1):220-235. 9. Husby S, Favero F, Nielsen C, et al. Clinical impact of clonal hematopoiesis in patients with lymphoma undergoing ASCT: a national population-based cohort study. Leukemia. 2020;34(12):3256-3268. 10. Fuster JJ, MacLauchlan S, Zuriaga MA, et al. Clonal hematopoiesis associated with TET2 deficiency accelerates atherosclerosis development in mice. Science. 2017;355(6327):842-847. 11. Yeaton A, Cayanan G, Loghavi S, et al. The impact of inflammation-induced tumor plasticity during myeloid transformation. Cancer Discov. 2022;12(10):2392-2413. 12. Coombs CC, Zehir A, Devlin SM, et al. Therapy-related clonal hematopoiesis in patients with non-hematologic cancers is common and associated with adverse clinical outcomes. Cell Stem Cell. 2017;21(3):374-382. 13. Wong TN, Miller CA, Jotte MRM, et al. Cellular stressors contribute to the expansion of hematopoietic clones of varying leukemic potential. Nat Commun. 2018;9(1):455. 14. Hsu JI, Dayaram T, Tovy A, et al. PPM1D mutations drive clonal hematopoiesis in response to cytotoxic chemotherapy. Cell Stem Cell. 2018;23(5):700-713.

Haematologica | 108 December 2023

3319

ARTICLE - CH impairs regenerative potential of stem cells

P. Stelmach et al.

15. Bolton KL, Ptashkin RN, Gao T, et al. Cancer therapy shapes the fitness landscape of clonal hematopoiesis. Nat Genet. 2020;52(11):1219-1226. 16. McCaughan GJ, Gandolfi S, Moore JJ, Richardson PG. Lenalidomide, bortezomib and dexamethasone induction therapy for the treatment of newly diagnosed multiple myeloma: a practical review. Br J Haematol. 2022;199(2):190-204. 17. Jackson GH, Davies FE, Pawlyn C, et al. Lenalidomide maintenance versus observation for patients with newly diagnosed multiple myeloma (Myeloma XI): a multicentre, openlabel, randomised, phase 3 trial. Lancet Oncol. 2019;20(1):57-73. 18. McCarthy PL, Holstein SA, Petrucci MT, et al. Lenalidomide maintenance after autologous stem-cell transplantation in newly diagnosed multiple myeloma: a meta-analysis. J Clin Oncol. 2017;35(29):3279-3289. 19. de Tute RM, Pawlyn C, Cairns DA, et al. Minimal residual disease after autologous stem-cell transplant for patients with myeloma: prognostic significance and the impact of lenalidomide maintenance and molecular risk. J Clin Oncol. 2022;40(25):2889-2900. 20. Mouhieddine TH, Sperling AS, Redd R, et al. Clonal hematopoiesis is associated with adverse outcomes in multiple myeloma patients undergoing transplant. Nat Commun. 2020;11(1):2996. 21. Wudhikarn K, Padrnos L, Lasho T, et al. Clinical correlates and prognostic impact of clonal hematopoiesis in multiple myeloma patients receiving post-autologous stem cell transplantation lenalidomide maintenance therapy. Am J Hematol. 2021;96(5):E157-E162. 22. Lackraj T, Barouch SB, Medeiros JJF, et al. Clinical significance of clonal hematopoiesis in the setting of autologous stem cell transplantation for lymphoma. Am J Hematol. 2022;97(12):1538-1547. 23. Mayakonda A, Lin DC, Assenov Y, Plass C, Koeffler HP. Maftools: efficient and comprehensive analysis of somatic variants in cancer. Genome Res. 2018;28(11):1747-1756. 24. Brunet JP, Tamayo P, Golub TR, Mesirov JP. Metagenes and molecular pattern discovery using matrix factorization. Proc Natl Acad Sci U S A. 2004;101(12):4164-4169. 25. Jeong M, Park HJ, Celik H, et al. Loss of Dnmt3a immortalizes hematopoietic stem cells in vivo. Cell Rep. 2018;23(1):1-10. 26. Challen GA, Sun D, Jeong M, et al. Dnmt3a is essential for hematopoietic stem cell differentiation. Nat Genet. 2011;44(1):23-31. 27. Midic D, Rinke J, Perner F, et al. Prevalence and dynamics of clonal hematopoiesis caused by leukemia-associated mutations in elderly individuals without hematologic disorders. Leukemia. 2020;34(8):2198-2205. 28. Scheller M, Ludwig AK, Gollner S, et al. Hotspot DNMT3A

mutations in clonal hematopoiesis and acute myeloid leukemia sensitize cells to azacytidine via viral mimicry response. Nat Cancer. 2021;2(5):527-544. 29. Nam AS, Dusaj N, Izzo F, et al. Single-cell multi-omics of human clonal hematopoiesis reveals that DNMT3A R882 mutations perturb early progenitor states through selective hypomethylation. Nat Genet. 2022;54(10):1514-1526. 30. Husby S, Hjermind Justesen E, Gronbaek K. Protein phosphatase, Mg(2+)/Mn(2+)-dependent 1D (PPM1D) mutations in haematological cancer. Br J Haematol. 2021;192(4):697-705. 31. Kahn JD, Miller PG, Silver AJ, et al. PPM1D-truncating mutations confer resistance to chemotherapy and sensitivity to PPM1D inhibition in hematopoietic cells. Blood. 2018;132(11):1095-1105. 32. Fiscella M, Zhang H, Fan S, et al. Wip1, a novel human protein phosphatase that is induced in response to ionizing radiation in a p53-dependent manner. Proc Natl Acad Sci U S A. 1997;94(12):6048-6053. 33. Eskelund CW, Husby S, Favero F, et al. Clonal hematopoiesis evolves from pretreatment clones and stabilizes after end of chemotherapy in patients with MCL. Blood. 2020;135(22):2000-2004. 34. Pawlyn C, Kaiser MF, Heuck C, et al. The spectrum and clinical impact of epigenetic modifier mutations in myeloma. Clin Cancer Res. 2016;22(23):5783-5794. 35. Walker BA, Mavrommatis K, Wardell CP, et al. Identification of novel mutational drivers reveals oncogene dependencies in multiple myeloma. Blood. 2018;132(6):587-597. 36. Robiou du Pont S, Cleynen A, Fontan C, et al. Genomics of multiple myeloma. J Clin Oncol. 2017;35(9):963-967. 37. Rasche L, Chavan SS, Stephens OW, et al. Spatial genomic heterogeneity in multiple myeloma revealed by multi-region sequencing. Nat Commun. 2017;8(1):268. 38. Fabre MA, de Almeida JG, Fiorillo E, et al. The longitudinal dynamics and natural history of clonal haematopoiesis. Nature. 2022;606(7913):335-342. 39. Gibson CJ, Lindsley RC, Tchekmedyian V, et al. Clonal hematopoiesis associated with adverse outcomes after autologous stem-cell transplantation for lymphoma. J Clin Oncol. 2017;35(14):1598-1605. 40. Sperling AS, Guerra VA, Kennedy JA, et al. Lenalidomide promotes the development of TP53-mutated therapy-related myeloid neoplasms. Blood. 2022;140(16):1753-1763. 41. Husby S, Baech-Laursen C, Eskelund CW, et al. Clonal hematopoiesis is associated with hematological toxicity during lenalidomide-based therapy for MCL. Leukemia. 2022;36(12):2912-2916. 42. McElduff F, Cortina-Borja M, Chan SK, Wade A. When t-tests or Wilcoxon-Mann-Whitney tests won't do. Adv Physiol Educ. 2010;34(3):128-133.

Haematologica | 108 December 2023

3320

ARTICLE - Myeloproliferative Disorders

Haploidentical donor hematopoietic cell transplantation for myelodysplastic/myeloproliferative overlap neoplasms: results from a North American collaboration Tania Jain,1 Hua-Ling Tsai,2 Hany Elmariah,3 Pankit Vachhani,4 Theodoros Karantanos,1 Sarah A. Wall,5 Lukasz P. Gondek,1 Asad Bashey,6 Alla Keyzner,7 Roni Tamari,8 Michael R. Grunwald,9 Sameem Abedin,10 Kalyan V. G. Nadiminti,11 Madiha Iqbal,12 Aaron T. Gerds,13 Auro Viswabandya,14 Shannon R. McCurdy,15 Monzr M. Al Malki,16 Ravi Varadhan,2 Haris Ali,16 Vikas Gupta,14 Richard J. Jones1 and Salman Otoukesh16 Division of Hematological Malignancies and Bone Marrow Transplantation, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University, Baltimore, MD, USA; 2Division of Biostatistics and Bioinformatics, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD, USA; 3Department of Bone Marrow Transplant and Cellular Immunotherapy, H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL, USA; 4Division of Hematology and Oncology, O’Neal Comprehensive Cancer Center, University of Alabama, Birmingham, AL, USA; 5Division of Hematology, The Ohio State University - James Comprehensive Cancer Center, Columbus, OH, USA; 6Blood and Marrow Transplant Program, Northside Hospital, Atlanta, GA, USA; 7The Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, New York, NY, USA; 8Adult Bone Marrow Transplantation Service, Department of Medicine, Memorial Sloan Kettering Cancer Center, and Department of Medicine, Weill Cornell Medical College, New York, NY, USA; 9Department of Hematologic Oncology and Blood Disorders, Levine Cancer Institute, Atrium Health, Charlotte, NC, USA; 10Division of Hematology/Oncology, Medical College of Wisconsin, Milwaukee, WI, USA; 11Division of Hematology, Medical Oncology and Palliative Care, Department of Medicine, University of Wisconsin School of Medicine and Public Health, Madison, WI, USA; 12Department of Hematology-Oncology, Mayo Clinic, Jacksonville, FL, USA; 13Department of Hematology and Medical Oncology, Cleveland, OH, USA; 14Princess Margaret Cancer Center, University of Toronto, Toronto, Ontario, Canada; 15University of Pennsylvania, Philadelphia, PA, USA and 16Department of Hematology and Hematopoietic Cell Transplantation, City of Hope, Duarte, CA, USA

Correspondence: T. Jain tjain2@jhmi.edu Received: Accepted: Early view:

April 25, 2023. June 29, 2023. July 6, 2023.

Abstract Haploidentical donors offer a potentially readily available donor, especially for non-White patients, for hematopoietic cell transplantation (HCT). In this North American collaboration, we retrospectively analyzed outcomes of first HCT using haploidentical donor and post-transplantation cyclophosphamide (PTCy) in myelodysplastic syndrome/myeloproliferative neoplasm (MDS/MPN) overlap neoplasms (MDS/MPN). We included 120 consecutive patients who underwent HCT using a haploidentical donor for MDS/MPN across 15 centers. Median age was 62.5 years and 38% were of non-White/Caucasian ethnicity. The median follow-up was 2.4 years. Graft failure was reported in seven of 120 (6%) patients. At 3 years, nonrelapse mortality (NRM) was 25% (95% confidence interval [CI]: 17-34), relapse 27% (95% CI: 18-36), grade 3-4 acute graftversus-host disease 12% (95% CI: 6-18), chronic graft-versus-host disease requiring systemic immunosuppression 14% (95% CI: 7-20), progression-free survival (PFS) 48% (95% CI: 39-59), and overall survival (OS) 56% (95% CI: 47-67). On multivariable analysis, NRM was statistically significantly associated with advancing age at HCT (per decade increment, subdistribution hazard ratio [sdHR] =3.28; 95% CI: 1.30-8.25); relapse with the presence of mutation in EZH2/RUNX1/SETBP1 (sdHR=2.61; 95% CI: 1.06-6.44); PFS with advancing age at HCT (per decade increment, HR=1.98, 95% CI: 1.13-3.45); and OS with advancing age at HCT (per decade increment, HR=2.01; 95% CI: 1.11-3.63) and splenomegaly at HCT/prior splenectomy (HR=2.20; 95% CI: 1.04-4.65). Haploidentical donors are a viable option for HCT in MDS/MPN, especially for those disproportionately represented in the unrelated donor registry. Hence, donor mismatch should not preclude HCT for patients with MDS/MPN, an otherwise incurable malignancy. In addition to patient age, disease-related factors including splenomegaly and high-risk mutations dominate outcomes following HCT.

Haematologica | 108 December 2023

3321

ARTICLE - Haplo-HCT in MDS/MPN

T. Jain et al.

Introduction

Methods

Myelodysplastic/myeloproliferative overlap neoplasms (MDS/MPN) are a group of clonal myeloid neoplasms and per the 2022 World Health Organization (WHO) classification include the diagnoses of chronic myelomonocytic leukemia (CMML), MDS/MPN with neutrophilia (or atypical chronic myeloid leukemia or aCML per 2016 WHO classification), MDS/MPN with SF3B1 mutation and thrombocytosis, MDS/MPN not otherwise specified (MDS/MPN-NOS).1-3 Over the years, several mostly small retrospective studies have established the curative potential of allogeneic hematopoietic cell transplantation (HCT) in MDS/MPN,4-7 but none of these incorporated related HLA-haploidentical donors in a meaningful way. In the current era, the use of haploidentical donors in the broad scope of HCT has evolved significantly since the advent of post-transplantation cyclophosphamide (PTCy).8-11 The role of haploidentical donors in MDS/MPN offers the potential advantage of available donors in a timely manner for an otherwise incurable malignancy. Additionally, finding a fully matched donor in the donor registry can be challenging for non-White patients due to the lower diversity of donors from these populations. Therefore, haploidentical donors can often be suitable donor options for patients who may be ethnically underrepresented in the donor registry. On the other hand, theoretical concerns of delayed engraftment or graft failure have been raised with haploidentical donor HCT owing to disease-related marrow fibrosis and splenomegaly. Hence, we conducted this study via our North American collaboration to systematically evaluate the clinical outcomes in MDS/MPN after haploidentical donor-PTCy HCT. Genomic landscape plays a notable role in the prognostication of all MDS/MPN entities with worse prognoses attributed to higher number of mutations and specific high-risk mutations.12-15 Prior work has demonstrated that mutation(s) in EZH2, RUNX1, or SETBP1 (E/R/S) is associated with a lower likelihood of response to hypomethylating agents, a commonly used non-transplantation therapeutic approach in MDS/MPN.12 Therefore, in this study, we sought to explore the role of genomic landscape, including a specific evaluation of E/R/S mutations, in determining outcomes of haploidentical donor HCT. Since the prevalence of these diseases is low and morphological distinction for these individual entities is often obscure, we grouped the various MDS/MPN entities in this study. At the same time, features of dysplasia, as well as proliferation, remain a unifying feature of MDS/MPN entities. Furthermore, the prognosis of these individual entities especially with advanced or high-risk disease is poor, unless remission is achieved followed by consolidation with HCT.12

Patient selection and multi-institutional collaboration This study leverages an ongoing multi-institutional collaboration of HCT centers across the USA and Canada to evaluate the role of HCT in rare myeloid malignancies. Fifteen institutions participated in this retrospective study, with Johns Hopkins University, (Baltimore, MD, USA) as the coordinating site (IRB00292283, approved on September 19, 2021). Each participating institution obtained approval from its respective Institutional Review Board and data was transferred to Johns Hopkins University upon completion of data-sharing agreements with each participating site. The study was designed in keeping with the tenets of the Declaration of Helsinki. This study was designed prior to the publication of the 2022 update of World Health Organization (WHO) and International Consensus Classification definitions.1,2 Hence, the diagnosis of CMML, MDS/MPN with neutrophilia, MDS/MPN with SF3B1 mutation and thrombocytosis, and MDS/MPN-NOS was in accordance with the 2016 WHO classification for MDS/MPN.3 Bone marrow biopsy reports of all included patients were reviewed by the participating site investigator as well as the coordinating site investigator (TJ) for adjudication of MDS/MPN diagnosis. Additional inclusion criteria for all centers were: (i) adult (age ≥18 years) patients who underwent a first HCT, (ii) HCT using haploidentical donor defined as family donor mismatched for haplotype, and PTCy-based graft-versushost disease (GvHD) platform, and (iii) HCT timeline between January 2011 and December 2021. Patients who had a transformation to blast phase (>20% blasts in blood or marrow) at any point in the disease course and those who underwent haploidentical donor cord blood HCT were excluded. All patients at all the collaborating institutions who met these criteria were included in the analysis. Definitions Revised International Prognostic Scoring System (R-IPSS), clinical/molecular CMML-specific prognostic scoring system (CPSS-mol), and MDS/MPN responses to therapy were assessed as previously published.16-18 Spleen size was measured by imaging or physical exam. Given the variability of this measurement, we labeled spleen size of <12 cm on imaging or non-palpable on physical exam as normal, and ≥12 cm on imaging or palpable below costal margin on physical exam was considered enlarged. Time to neutrophil engraftment was defined as days from the day of HCT to the first of the 3 consecutive days when the absolute neutrophil count was ≥500/µL, while time to platelet engraftment was defined as days from the day of HCT to the first of the 3 consecutive data of platelets >20,000/μL in the absence of platelet transfusions for 7

Haematologica | 108 December 2023

3322

ARTICLE - Haplo-HCT in MDS/MPN

T. Jain et al.

consecutive days.19 Graft failure was defined as a lack of donor hematopoietic cell engraftment following HCT (<5% donor chimerism) at any time following HCT, without evidence of disease relapse.20 Non-relapse mortality (NRM) was death from any cause in the absence of disease relapse. Acute and chronic GvHD were graded per standard criteria.21,22 Day 0 of HCT was used as the reference day for time-to-event outcomes. Overall survival (OS) was defined from the date of HCT (day 0) to the date of death from any cause or censored at the last follow-up date for alive patients. The events of progression-free survival (PFS) included relapse or death, whichever occurred first.

Table 1. Baseline patient, disease, and hematopoietic cell transplantation details (N=120). Characteristic Patient details Median age at HCT in years (range) Median months from diagnosis to HCT (range)

Statistical analysis For outcomes subject to competing events, cumulative incidences were reported and the distribution differences between groups were compared via Gray’s K-sample tests.26 When estimating the cumulative incidence function of relapse, NRM was the competing event and vice versa. When estimating the cumulative incidence of GvHD, the competing events included graft failure and death without graft failure and without the corresponding GvHD event. OS and PFS were estimated via KaplanMeier method, and the distribution differences between groups were compared via log-rank test. Patients who did not relapse or die were censored on the date of last follow-up. Cox proportional hazards model was applied in univariate and multivariable analyses to estimate the hazard ratio of OS and PFS.27 Fine-Gray subdistribution hazards model was used univariate and multivariable analyses of relapse, NRM, or GvHD outcomes.28 Covariates in multivariable analyses were selected based on clinical relevance and statistical significance noted on univariate analysis. All hypothesis testing was two-sided based on a significance level of 0.05 without considering multiplicity. Analyses were conducted in R version 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria).

62.5 (18-75) 10.35 (1.1-399.2)

Male sex, N (%)

77 (64.2)

Race/ethnicity, N (%) Caucasian African American Hispanic Asian Alaskan Native

74 (61.7) 21 (17.5) 13 (10.8) 11 (9.2) 1 (0.8)

Diagnosis, N (%) CMML MDS/MPN-NOS MDS/MPN with neutrophilia MDS/MPN with SF3B1 and thrombocytosis HCT-CI, N (%) 0-1 ≥2

Cytogenetic and somatic mutation data Cytogenetic results were deemed “high-risk” per those included in intermediate, high, and very high-risk categories of R-IPSS.17 Next-generation sequencing (NGS) was used to obtain somatic mutation data at individual participating institutions and results from these respective tests were used for analysis. NGS was obtained prior to HCT in all patients on whom the data is available, either at diagnosis or with the pre-HCT evaluations. High-risk mutations on NGS included mutations in NRAS, SETBP1, RUNX1, EZH2, TP53, ASXL1, and STAG2 as previously described.14,15,23-25 NGS was done at individual participating institutions and included commonly reported mutations in myeloid malignancies.

N=120

61 (50.8) 48 (40.0) 5 (4.2) 6 (5.0) 65 (54.2) 55 (45.8)

Disease-related details High-risk cytogenetics, N (%)*

25 (20.8)

Number of high-risk mutations on NGS (total N=90), N (%) 0 1 ≥2

32 (35.6) 26 (28.9) 32 (35.6)

Mutations in E/R/S present, total N=90, N (%) R-IPSS risk category, N (%) Very low/ low Intermediate/ high/ very high Spleen size at HCT, N (%) Normal Enlarged Splenectomy Marrow blasts at HCT, N (%) <10% ≥10%

28 (31.1) 48 (40) 72 (60) 67 (55.8) 50 (41.7) 3 (2.5) 113 (94.2) 7 (5.8)

HCT-related details HCT year, N (%) 2011-2018 2019-2021 Recipient CMV seropostive, N (%) Donor age at HCT in years <30 30-45 >45 Conditioning regimen intensity, N (%) Myeloablative Reduced-intensity Non-myeloablative

63 (52.5) 57 (47.5) 80 (66.7) 33 (27.5) 62 (51.7) 25 (20.8) 22 (18.3) 44 (36.7) 54 (45.0)

GvHD prophylaxis (with PTCy), N (%) Tacrolimus MMF Sirolimus MMF ATG-cyclosporin

97 (80.8) 19 (15.8) 4 (3.3)

Graft source, N(%) Bone marrow Peripheral blood

25 (20.8 95 (79.2)

Median CD34+ cell dose x106/kg (range)

5 (0.86-23.8)

Per R-IPSS, del(7q), +8, +19, i(17q), -7, inv(3)/t(3q)/del(3q), double including -7/del(7q), Complex: 3 or more abnormalities. CMV: cytomegalovirus; E/R/S: EZH2/RUNX1/SRSF2; HCT: hematopoietic cell transplant; HCT-CI: HCT comorbidity index; PTCy: post-transplantation cyclophosphamide; R-IPSS: Revised-International Prognostic Scoring System; GvHD: graft-versus-host disease; MDS/MPN: myelodysplastic/myeloproliferative overlap neoplasm; NGS: next-generation sequencing; MDS/MPN-NOS: MDS/MPN not otherwise specified; MMF: mycophenolate mofetil, CMML: chronic myelomonocytic leukemia. *

Haematologica | 108 December 2023

3323

ARTICLE - Haplo-HCT in MDS/MPN

T. Jain et al.

Results Baseline patient and hematopoietic cell transplantation details We identified 120 patients across the 15 participating institutions who underwent a first haplo-HCT for MDS/MPN using PTCy-based GvHD prophylaxis. A descriptive summary of these patients is shown in Table 1. Patients were more commonly of male sex (64%) and over one third (37%) were ≥65 years of age, in keeping with the male predominance and older median age of diagnosis of MDS/MPN.15,29 Forty-six (38%) patients were of non-White/Caucasian race/ethnic background, who are disproportionately represented on the donor registry. Karnofsky performance score was <90 in 46 (39%) patients at the time of HCT. R-IPSS was low or very low in 40% of patients who underwent HCT, most commonly due to the presence of high-risk somatic mutations and/or younger age. Cytogenetic analysis revealed normal karyotype in 75 (63%) patients, as is often the case in MDS/MPN. NGS data was available in 90 patients and 87 had at least one detectable mutation on the NGS panel with 45 (50%) harboring >3 mutations, while 58 (64%) had one or more high-risk somatic mutations (Figure 1). Consistent with prior reports, men had a higher

median number of mutations than women (4 vs. 3; P=0.015), but the proportion of men with high-risk somatic mutations or those with E/R/S mutations, while higher, was not statistically significantly different than women (70% vs. 56%; P=0.19; 32% vs. 29%, P=0.79; Online Supplementary Table S1).15 As anticipated, bridging treatment prior to HCT varied significantly across all patients and hypomethylating agents were the most common agent for bridging used in 88 (73%) patients, hydroxyurea only in five (4%), and induction chemotherapy (including venetoclaxbased regimens) in 19 (16%) patients. Of the 106 patients who underwent bridging therapy, 40 (38%) achieved a complete response/partial response (CR/PR) per international consortium criteria prior to HCT.18 Seven (6%) patients had 10% or more blasts in the bone marrow at the time of HCT. Three patients had undergone splenectomy prior to HCT, two due to MDS/MPN, and one for a different malignancy. Consistent with established institutional protocols, a diverse range of specific conditioning regimens were used as detailed in Online Supplementary Table S2. Clinical outcomes The median follow-up on this study is 2.4 years after HCT, based on the reverse Kaplan-Meier method. Online Sup-

Figure 1. Mutational landscape of subset of patients undergoing haploidentical hematopoietic cell transplantation for myelodysplastic/myeloproliferative overlap neoplasm (N=90). CMML: chronic myelomonocytic leukemia; MDS/MPN: myelodysplastic/myeloproliferative overlap neoplasm; MDS/MPNNOS: MDS/MPN-not otherwise specified; MDS/MPN-SF3B1-T: MDS/MPN with SF3B1 mutation and thrombocytosis. Haematologica | 108 December 2023

3324

ARTICLE - Haplo-HCT in MDS/MPN

T. Jain et al.

plementary Table S3 summarizes engraftment, NRM, relapse, acute and chronic GvHD, PFS and OS. Figure 2 provides Kaplan-Meier analysis of all clinical outcomes. The OS and PFS for CMML (n=61/120, 51%) and MDS/MPN-NOS (n=48, 40%), which comprised over 90% of patients, were not statistically different (hazard ratio [HR] =0.94; 95% CI: 0.54-1.66; P=0.84 for OS; and HR=1.06; 95% CI: 0.62-1.81; P=0.82 for PFS). These HR are suggestive of very similar outcomes. Next, a comparison of CMML, MDS/MPN-NOS, MDS/MPN with neutrophilia, and MDS/MPN with SF3B1 mutations and thrombocytosis as separate entities also found no statistically significant difference (Figure 3A). Given this finding in addition to the known low prevalence of these diagnoses and overall clinicopathological overlap, we combined the four entities for subsequent analyses.

31 (IQR=22.5-41) days (Online Supplementary Table S3). Seven (6%) patients had graft failure, all of whom had received reduced intensity conditioning (RIC)/non-myeloablative conditioning (NMAC), six had used a peripheral blood graft, and two received anti-thymocyte globulin and PTCy for GvHD prophylaxis. Three of these seven patients (43%) died before day +100 due to infections. The remaining four underwent a second HCT, and two of those are alive at the last follow-up at day +481 and day +2,337, respectively. One of the second HCT was done using the same donor (peripheral blood graft instead of marrow graft) and the remaining three were done using a different donor. Of note, there were a total four patients who received anti-thymocyte globulin and PTCy which accounts for a notable 50% graft failure with this regimen, while acknowledging the small size of this subset.

Engraftment and outcomes after graft failure Median time to neutrophil engraftment was 18 (interquartile range [IQR] =16-22) days and platelet engraftment was

Non-relapse mortality and relapse The cumulative incidence of NRM was 20% (95% CI: 1328) at 1-year and 25% (95% CI: 17-34) at 3 years (Figure

Figure 2. Clinical outcomes for entire cohort (N=120). (A) Cumulative incidence of non-relapse mortality (NRM) and relapse; (B) acute graft-versus-host disease (aGvHD) grades 2-4 and grades 3-4; (C) chronic GvHD (cGVHD) all grade and cGvHD requiring systemic immunosuppression; and (D) Kaplan Meier estimates of overall survival (OS) and progression-free survival (PFS). Gr: grade; Req. system: requiring systemic immunosuppression; Relap: relapse.

Haematologica | 108 December 2023

3325

ARTICLE - Haplo-HCT in MDS/MPN

T. Jain et al.

2A; Online Supplementary Table S3). The cause of death among these patients were infection in 17 (14%), GvHD in five (4%), organ toxicity in six (5%), another malignancy in two (2%), and unknown in three patients (3%). The cumulative incidence of relapse was 20% (95% CI: 13-27) at 1 year and 27% (95% CI: 18-36) at 3 years. In total, 30 (25%) patients had relapsed of whom 24 (20%) had died as a result of relapsed disease, by the last follow-up. Seven patients underwent donor lymphocyte infusion (DLI), most commonly for molecular relapse, of whom two restored full donor chimerism while one had only a transient improvement in chimerism. Four patients underwent a second HCT after relapse, two of whom had received a prior DLI also. All four of these patients are deceased at the last follow-up from persistent disease.

Acute and chronic graft-versus-host disease Most patients who experienced acute GvHD had highest grade of grade 2. At 1 year, the cumulative incidence of acute GvHD grade 2-4 was 35% (95% CI: 27-44) and grade 3-4 acute GvHD was 12% (95% CI: 6-18) (Figure 2B; Online Supplementary Table S3). The skin and gut were the most commonly involved organs in 27 of 42 (64%) patients. The cumulative incidence of chronic GvHD at 3 years was 33% (95% CI: 24-42) and chronic GvHD requiring systemic therapy was 14% (95% CI: 7-20) (Figure 2C; Online Supplementary Table S3). Progression-free survival and overall survival At 1 year and 3 years, the probability of PFS was 60% (95% CI: 51-69) and 48% (95% CI: 39-59) and the probability of OS was 70% (95% CI: 62-79) and 56% (95% CI: 47-67), re-

Figure 3. Difference in overall survival by diagnosis, age, spleen size and marrow blasts in the total cohort of 120 patients. (A) Overall survival (OS) by diagnosis entities of chronic myelomonocytic leukemia (CMML), myelodysplastic/myeloproliferative overlap neoplasm (MDS/MPN) with neutrophilia (MDS/MPN-N), MDS/MPN not otherwise specified (MDS/MPN-NOS), and MDS/MPN with SF3B1 mutation and thrombocytosis (MDS/MPN-SF3B1-T). (B) OS by patient age at hematopoietic cell transplantation (HCT). (C) OS by spleen size at HCT. (D) OS by bone marrow blast percentage.

Haematologica | 108 December 2023

3326

ARTICLE - Haplo-HCT in MDS/MPN

T. Jain et al.

Table 2. Univariate analysis for non-relapse mortality, relapse, progression-free survival and overall survival (N=120). NRM sdHR (95% CI)

Relapse P

sdHR (95% CI)

PFS P

HR (95% CI)

OS P

HR (95% CI)

Age at HCT (10-year increment)

1.40 (0.88-2.23) 0.15 1.51 (1.00-2.28) 0.05 1.63 (1.15-2.02) 0.006 1.53 (1.07-2.18) 0.02

Sex (female vs. male)

0.80 (0.38-1.67) 0.55 0.72 (0.34-1.55) 0.40 0.70 (0.41-1.19) 0.19 0.72 (0.41-1.27) 0.25

Race/ethnicity Caucasian Ref Ref Ref Ref African American 0.75 (0.26-2.16) 0.59 1.30 (0.52-3.24) 0.57 1.04 (0.51-2.11) 0.91 0.96 (0.44-2.09) 0.92 Hispanic/Asian/Alaskan Native American 1.58 (0.73-3.42) 0.24 1.20 (0.51-2.80) 0.68 1.53 (0.84-2.79) 0.16 1.70 (0.92-3.11) 0.09 Years from diagnosis to HCT

1.08 (1.02-1.14) 0.01 0.86 (0.70-1.05) 0.15 1.03 (0.96-1.11) 0.39 1.05 (0.97-1.13) 0.21

HCT year (2019-2021 vs. 2011-2018)

0.41 (0.18-0.92) 0.04 0.64 (0.30-1.38) 0.26 0.47 (0.27-0.82) 0.008 0.44 (0.23-0.82) 0.01

Diagnosis CMML MDS/MPN-NOS MDS/MPN-N or MDS/MPN-SF3B1-T

Ref Ref Ref Ref 1.30 (0.62-2.71) 0.48 0.93 (0.46-1.90) 0.85 1.06 (0.62-1.81) 0.82 0.94 (0.54-1.66) 0.84 2.98 (1.07-8.34) 0.04 0.33 (0.05-2.27) 0.26 1.48 (0.61-3.59) 0.39 1.84 (0.75-4.51) 0.18

KPS (≥90 vs. <90)

0.96 (0.48-1.93) 0.91 0.50 (0.25-1.02)

0.65 (0.39-1.07) 0.09 0.83 (0.48-1.41) 0.48

HCT-CI (≥2 vs. 0-1)

1.32 (0.67-2.62) 0.42 0.88 (0.43-1.78)

1.07 (0.65-1.76) 0.79 1.16 (0.69-1.96) 0.58

R-IPSS (intermediate/high/very high vs. very low/low)

1.84 (0.85-3.98) 0.12 0.94 (0.46-1.95)

1.42 (0.82-2.43) 0.21 1.40 (0.79-2.46) 0.25

Bridging therapy None Hydrea only HMA/ JAK inhibitor/IMiD Induction

Ref Ref Ref Ref 2.14 (0.32-14.16) 0.43 No relapse 0.93 (0.19-4.63) 0.93 1.15 (0.22-5.99) 0.86 1.16 (0.36-3.73) 0.80 1.35 (0.42-4.33) 0.61 1.33 (0.56-3.14) 0.51 1.38 (0.54-3.53) 0.50 1.58 (0.46-5.40) 0.46 1.32(0.32-5.42) 0.70 1.51 (0.57-4.05) 0.41 1.55 (0.54-4.49) 0.42

CR/PR prior to HCT

0.79 (0.38-1.63) 0.52 1.88 (0.81-4.34) 0.14 1.20 (0.69-2.09) 0.52 2.38 (1.01-5.59) 0.047

Spleen size at HCT (enlarged/splenectomy 2.10 (1.04-4.25) 0.04 1.68 (0.82-3.45) 0.15 2.17 (1.31-3.61) <0.005 0.90 (0.47-1.75) 0.76 vs. normal) Marrow blasts at HCT (≥10% vs. >10%)

1.60 (0.55-4.70) 0.39 1.86 (0.58-5.95) 0.29 1.97 (0.84-4.60) 0.12 0.99 (0.49-1.98) 0.98

High-risk cytogenetics (presence vs. absence)

1.06 (0.47-2.41) 0.88 0.68 (0.27-1.73) 0.42 0.83 (0.44-1.56) 0.57 1.23 (0.63-2.41) 0.54

High-risk NGS (presence vs. absence)

0.70 (0.29-1.72) 0.44 2.65 (0.87-8.12) 0.09 1.34 (0.69-2.63) 0.39 1.18 (0.59-2.34) 0.64

Number of high-risk mutations (≥2 vs. 0-1) 0.61 (0.23-1.60) 0.32 2.70 (1.20-6.05) 0.02 1.51 (0.81-2.83) 0.19 1.21 (0.97-1.51) 0.09 E/R/S mutation (presence vs. absence)

0.56 (0.19-1.60) 0.28 3.33 (1.47-7.52) <0.005 1.72 (0.92-3.23) 0.09 0.87 (0.46-1.65) 0.68

Donor age (10-year increment)

1.20 (0.88-1.63) 0.25 1.12 (0.84-1.50) 0.45 1.21 (0.98-1.49) 0.07 1.21 (0.97-1.51) 0.09

Graft source (blood vs. marrow)

0.85 (0.37-1.95) 0.70 0.91 (0.38-2.14) 0.82 0.85 (0.47-1.55) 0.60 0.87 (0.46-1.65) 0.68

GvHD prophylaxis Tacrolimus Sirolimus ATG cyclosporine

Ref Ref 0.36 (0.08-1.62) 0.18 2.02(0.95-4.33) 4.16 (1.29-13.48) 0.02 No relapse

Conditioning intensity MAC RIC/NMAC

Ref Ref Ref 1.77 (0.70-4.49) 0.23 3.42 (0.81-14.39) 0.09 2.90 (1.35-6.76)

CMV reactivation requiring intervention

1.37 (0.69-2.68) 0.37 0.43 (0.16-1.15) 0.09 0.76 (0.43-1.33) 0.33 0.87(0.49-1.54) 0.63

Ref Ref 0.07 1.07 (0.54-2.11) 0.85 0.87 (0.41-1.85) 0.71 2.07 (0.64-6.68) 0.22 2.26 (0.70-7.33) 0.17 Ref 2.44 (1.04-5.71) 0.04

NRM: non-relapse mortality; OS: overall survival; PFS: progression-free survival; sdHR: subdistribution hazard ratio; CI: confidence interval; ATG: anti-thymocyte globulin; CMV: cytomegalovirus; E/R/S: EZH2/RUNX1/SRSF2; GvHD: graft-versus-host disease; HCT: hematopoietic cell transplant; HCT-CI: HCT comorbidity index; HMA: hypomethylating agent; ImiD: immunomodulatory drugs; JAK: Janus kinase; KPS: Karnofsky performance score; R-IPSS: Revised-International Prognostic Scoring System; MAC: myeloablative conditioning; MDS/MPN: myelodysplastic/myeloproliferative overlap neoplasm; MDS/MPN-NOS: MDS/MPN not otherwise specified; MDS/MPN-N: MDS/MPN with neutrophilia; MDS/MPN-SF3B1-T: MDS/MPN with SF3B1 mutation and thrombocytosis; NGS: next-generation sequencing; NMAC: non-myeloablative conditioning; RIC: reduced-intensity conditioning; Ref: reference.

Haematologica | 108 December 2023

3327

ARTICLE - Haplo-HCT in MDS/MPN

T. Jain et al.

spectively, as tabulated in the Online Supplementary Table S3. The respective Kaplan-Meier curves for PFS and OS are shown in Figure 2D. Univariate analysis The univariate analysis including patient, disease, and HCT variables for NRM, relapse, PFS, and OS is detailed in Table 2. In the univariate analysis for OS, advanced patient age at HCT (HR=1.53 per decade increase in age, 95% CI: 1.072.18; P=0.02; Figure 3B), year of HCT prior to 2019 (HR=0.44; 95% CI: 0.23-0.82; P=0.01), splenomegaly at HCT /prior splenectomy (HR=2.57; 95% CI: 1.48-4.44; P<0.005; Figure 3C), ≥10% blasts in marrow at HCT (HR=2.38; 95% CI: 1.01-5.59; P=0.046; Figure 3D), and RIC/NMAC (HR=2.44; 95% CI: 1.04-5.71; P=0.04) were associated with inferior

OS. The presence of ≥2 high-risk mutations (subdistribution HR [sdHR] =2.70; 95% CI: 1.20-6.05; P=0.02; Figure 4A), and E/R/S mutations (sdHR=3.33; 95% CI: 1.47-7.51; P<0.005; Figure 4B), were associated with a significantly higher risk of relapse following HCT. Multivariable analysis Patient age at HCT, year of HCT, RIPSS, presence of E/R/S mutation, splenomegaly at HCT, donor age, and intensity of conditioning regimen were included in the multivariable analysis. Blast percentage was not included in the multivariable analysis because only seven patients had blasts over 10%. A complete multivariable analysis is shown in Table 3. Advanced age at HCT (sdHR=3.28 for every 10 years increment in age; 95% CI: 1.30-8.25; P=0.01) was as-

Figure 4. Difference in relapse by high-risk mutations in the cohort of 90 patients with next-generation sequencing data. (A) Relapse by number of high-risk mutations. (B) Relapse by EZH2, RUNX1, or SETBP1 mutations (Mut). NGS: next-generation sequencing.

Table 3. Multivariable analysis for non-relapse mortality, relapse, progression-free survival and overall survival (N=90). NRM sdHR (95% CI)

Relapse P

sdHR (95% CI)

PFS P

HR (95% CI)

OS P

HR (95% CI)

Age at HCT (10-year increment)

3.28 (1.30-8.25) 0.01 1.11 (0.67-1.81) 0.69 1.98 (1.13-3.45) 0.02 2.01 (1.11-3.63) 0.02

HCT year (2019-2021 vs. 2011-2018)

0.39 (0.10-1.47) 0.16 0.71 (0.28-1.80) 0.48 0.55 (0.28-1.10) 0.09 0.53 (0.25-1.13) 0.10

R-IPSS (intermediate/high/very high vs. very low/low)

2.53 (0.84-7.68) 0.10 0.82 (0.32-2.12) 0.68 1.44 (0.74-2.81) 0.28 1.35 (0.66-2.75) 0.42

E/R/S mutation (presence vs. absence)

0.43 (0.14-1.33) 0.14 2.61 (1.06-6.44) 0.04 1.24 (0.64-2.40) 0.52 0.88 (0.42-1.82) 0.73

Spleen size at HCT (enlarged/splenectomy 1.19 (0.34-4.14) 0.78 1.70 (0.74-3.87) 0.21 1.78 (0.91-3.51) 0.09 2.20 (1.04-4.65) 0.04 vs. normal) Donor age (10-year increment)

1.36 (0.87-2.12) 0.18 1.09 (0.68-1.77) 0.71 1.17 (0.85-1.63) 0.34 1.14 (0.79-1.62) 0.49

Conditioning intensity (RIC/NMAC vs. MAC)

0.37 (0.12-1.16) 0.09 3.90 (1.32-11.49) 0.01 1.51 (0.75-3.05) 0.25 1.20 (0.57-2.52) 0.64

NRM: non-relapse mortality; PFS: progression-free survival; OS: overall survival; sdHR: subdistribution hazard ratio; ATG: anti-thymocyte globulin; E/R/S: EZH2/RUNX1/SRSF2; HCT: hematopoietic cell transplant; R-IPSS: revised-International Prognostic Scoring System; MAC: myeloablative conditioning; NMAC: non-myeloablative conditioning; RIC: reduced-intensity conditioning.

Haematologica | 108 December 2023

3328

ARTICLE - Haplo-HCT in MDS/MPN

T. Jain et al.

sociated with higher NRM. RIC/NMAC and presence of E/R/S were significantly associated with higher relapse rate (sdHR=3.90; 95% CI: 1.32-11.49; P=0.01 for RIC/NAMC and HR=2.61; 95% CI: 1.06-6.44; P=0.04 for E/R/S mutations). Inferior PFS and OS were noted with advanced age at HCT (HR=1.98; 95% CI: 1.13-3.45; P=0.02 for PFS and HR=2.01; 95% CI: 1.11-3.63; P=0.02 for OS), and splenomegaly at HCT or a splenectomy prior to HCT (HR=1.78; 95% CI: 0.91-3.51; P=0.09 for PFS and HR=2.20; 95% CI: 1.04-4.65; P<0.04 for OS). As a result of the counterpoise of lower NRM and higher relapse, conditioning intensity did not show a significant association with PFS or OS (HR=1.51; 95% CI: 0.753.05; P=0.25 for PFS and HR=1.20; 95% CI: 0.57-2.52; P=0.64 for OS). Notably, the choice of myeloablative conditioning, over RIC/NMAC, was statistically significantly correlated with younger age at HCT in this analysis, corroborating observations from clinical practice (Online Supplementary Figure S1).

Discussion Our study provides a comprehensive description of the outcomes of haplo-HCT in MDS/MPN in a cohort of 120 patients in this multi-institutional collaboration. The potentially curative role of HCT in high-risk CMML was recently elucidated in comparison to non-HCT options.6 We demonstrate that haploidentical donors can be used with HCT outcomes similar to what has been historically reported with matched donors, in the rare diagnosis of MDS/MPN. This is particularly important for populations who are less likely to find a fully matched donor in the unrelated donor registry. Graft failure rate was under 10% and OS was 70% at 1 year and 56% at 3 years in our study. In the recent international analysis of CMML patients without AML transformation, HCT resulted in OS of about 30-35% at 3 years, with a majority (~75%) of donors being HLA-matched siblings or unrelated donors.6 Japanese nationwide registry data reported an OS of 48.5% at 3 years in MDS/MPN-NOS, using a variety of related, unrelated, and cord blood donors.30 Notably, 40% patients in this study were under 50 years of age at HCT. In a Mayo Clinic cohort of 17 CMML and eight MDS/MPN-NOS patients without antecedent blast transformation, HCT with matched donors resulted in a graft failure of 6% and 0%, and OS of 47% and 41%, at 2 years, respectively.7 Among 14 patients with MDS/MPN with neutrophilia, the Japanese registry study reported 54% OS at 1 year, using predominantly matched donors and select cord blood donors.4 While the timeline of HCT in all these studies varies, outcomes in our study are comparable to the limited reports presented above, underscoring that donor availability should not preclude consideration of HCT for patients with MDS/MPN whose

outcomes remain poor in the absence of the HCT. This study also explores modifiable disease-related features, in the form of spleen size control and blast reduction, which can possibly be optimized prior to HCT to allow for superior disease control and survival following HCT. We previously demonstrated the role of enlarged spleen size in negatively impacting relapse outcomes in 9,31 As demonstrated previously, splenomyelofibrosis. megaly is an indicator of aggressive disease biology and is not addressed with splenectomy, which did not appear to correlate with improved outcomes.31 JAK inhibitors have shown meaningful spleen size reduction in myelofibrosis3234 and have an emerging role in the management of CMML by targeting JAK-STAT dependent GM-CSF signaling in CMML.35,36 Hence, JAK inhibitors may address spleen size reduction prior to HCT in MDS/MPN as in myelofibrosis, an evaluation warranted in future studies. While MDS/MPN (except CMML) were not included in the pivotal VIALE-A trial, retrospective studies have demonstrated disease control with a combination of hypomethylating agents with BCL-2 inhibitor, venetoclax, in select patients with elevated blasts in MDS/MPN.37 We cannot identify an optimal bridging therapy in this study due to the variable availability of drugs over the years and various factors guiding bridging therapy selection in the real-world, including individual center practices. However, a systematic evaluation of the role of JAK inhibitors and BCL-2 inhibitors for disease control and as a bridge to HCT in MDS/MPN is warranted. Intensity of conditioning regimen is often a matter of discussion in planning HCT, especially in chronic myeloid malignancies where average age of diagnosis or HCT is often over 60 years. As is noted in our study, decisions on conditioning intensity are commonly driven by age and the comorbidity status of an individual patient, in that younger or fitter patients are enriched in the myeloablative cohort. In myelofibrosis, MDS and other myeloid malignancies, retrospective studies of higher intensity conditioning demonstrate the possibility of better disease control but at the expense of higher NRM,38-41 similar to what we note in this study. Ultimately, OS is not statistically different. Hence, patient selection remains a critical confounder to consider when interpreting the role of intensity of conditioning in a retrospective manner. A growing body of evidence has uncovered the role of the genomic landscape in the overall prognosis as well as response to hypomethylating agent therapy in MDS/MPN.12,14,42 Our study further elaborates on the role of somatic mutations in MDS/MPN, including previously defined high-risk mutations, in prognosticating outcomes of haploidentical donor HCT. In two cohorts of CMML patients, mutations in ASXL1, CBL, RUNX1, NRAS, and SETBP1 were associated with adverse survival.16,42 We also demonstrated higher prevalence of high-risk mutations, specifically EZH2, in

Haematologica | 108 December 2023

3329

ARTICLE - Haplo-HCT in MDS/MPN

T. Jain et al.

men with MDS/MPN, which may be responsible for inferior overall outcomes when compared to women.15 E/R/S, and ASXL1 mutations have also been associated with a lower risk of response to non-HCT therapy, especially hypomethylating agents, in MDS/MPN.12,43 In the context of HCT, somatic mutation data has historically been limited to CMML. Mutations in DNMT3A, TP53, ASXL1, and NRAS correlated with inferior survival in two different studies.13,44 The presence of high-risk mutations, specifically E/R/S, or the presence of ≥2 high-risk mutations significantly increased the risk of relapse in our study. Collectively, these data suggest that MDS/MPN harboring high-risk mutations identified in the non-transplantation context also influence relapse following HCT. Notably, the difference was significant for relapse, but not OS in our study, which possibly suggests that some of these patients can be salvaged after relapse following BMT with DLI or other novel therapeutic strategies. MDS/MPN is a rare, yet consequential, disease entity. This extensive report of haploidentical donor HCT in MDS/MPN was feasible due to our robust multi-institutional collaboration. The present analysis nevertheless remains limited by its retrospective nature and heterogenous practice across centers. We combined the four entities within MDS/MPN for this analysis given the high-risk nature of all four entities when considered for HCT and because our initial analysis demonstrated no statistical difference in OS among CMML, MDS/MPN-NOS, MDS/MPN with neutrophilia, and MDS/MPN with SF3B1 mutation and thrombocytosis. The strategy of combining diseases for analyses increases the sample size but introduces the confounding effect of variable nuances of these entities. Conclusion We demonstrate feasibility and comparable outcomes with haplo-HCT and PTCy in MDS/MPN with reference to previously published data, in a multi-institution study. Given the otherwise incurable diagnosis of MDS/MPN, this study provides a rationale for expanding the potential donor pool to include haploidentical donors in patients undergoing transplant evaluation. Optimization of disease-related factors such as spleen size and blast percentage reduction prior to HCT should be explored in future studies. The mutation landscape associated with MDS/MPN can guide outcomes with HCT. Future studies are needed to explore pre-HCT therapies and their impact on modifying mutational burden and post-transplantation outcomes. Disclosures TJ has received institutional research support from CTI Biopharma, Kartos Therapeutics and Incyte; discloses advisory board participation with Care Dx, Bristol Myers Squibb, Incyte, Abbvie, CTI, Kite, Cogent Biosciences, Blueprint Medicine, Telios. PV discloses a consulting or advisory role at Abbvie, Amgen, Blueprint Medicines, Cogent Biosciences, Incyte, CTI

BioPharma Corp, Daiichi Sankyo, GlaxoSmith Kline, Novartis, Pfizer, Genentech, Servier, Stemline, MorphoSys; is part of the Speakers' Bureau of Incyte, CTI BioPharma Corp, Blueprint Medicines; has institutional research support from Seattle Genetics, Amgen, Astex Pharmaceuticals, Incyte, Blueprint Medicines, Kartos Therapeutics, Gilead/Forty Seven, Constellation Pharmaceuticals, AbbVie, CTI BioPharma Corp and Takeda. SAW has received institutional research support from CTI Biopharma, Kartos, Telios, Incyte and Abbvie; discloses advisory board participation with Abbvie. ATG discloses a consultancy role at PharmaEssentia, Imago Biosciences, Morphosys, CTI Biopharma, Sierra Oncology/Glaxo Smith Kline, Telios, Bristol Myers Sqibb and AbbVie. MRG has received consulting fees from AbbVie, Agios/Servier, Amgen, Astellas Pharma, Blueprint Medicines, Bristol Myers Squibb, Cardinal Health, CTI BioPharma, Daiichi Sankyo, Gamida Cell, Genentech, Gilead Sciences, GSK/Sierra Oncology, Incyte Corporation, Invitae, Jazz, Karius, Novartis, Ono Pharmaceutical, Pfizer, Pharmacosmos, Premier Pharma, and Stemline Therapeutics; has received research support from Incyte Corporation and Janssen; discloses stock ownership of Medtronic. KVGN discloses advisory board participation with Morphosys, Abbvie Inc; has received research support from Abbvie Inc. MMA has received research support from NexImmune, Gilead and Incyte; has received consulting fees from NexImmune, Gilead, Incyte and CareDx. HA received research support from Incyte; consulting fee from Incyte, GSK, Karyopharm, Abbvie. VG discloses scientific advisory board and consultancy for Novartis, Incyte, BMS-Celgene, Sierra oncology, Pfizer, Abbvie, Morphosys, GSK, CTI Biopharma; has received research support from Novartis and Incyte. The remaining authors have no conflicts of interest to disclose. Contributions TJ conceptualized the study, acquired data, interpreted the data analysis, wrote the first draft of the manuscript, and approved the final draft. HLT and RV conducted the data analysis, edited the manuscript draft, and approved the final draft. HE, PV, TK, SAW, LPG, AB, AK, RT, MRG, SA, KVGN, MI, ATG, AV, SRM, MAM, DA, VG and SO acquired data, interpreted the data analysis, edited the manuscript draft, and approved the final draft. RJJ conceptualized the study, interpreted the data analysis, edited the manuscript draft, and approved the final draft. Funding This work was supported by National Institutes of Health National Cancer Institute grants P01 CA225618 and P30 CA006973. Data-sharing statement De-identified patient data will be made available upon a reasonable request to the corresponding author after publication.

Haematologica | 108 December 2023

3330

ARTICLE - Haplo-HCT in MDS/MPN

T. Jain et al.

References 1. Arber DA, Orazi A, Hasserjian RP, et al. International Consensus Classification of Myeloid Neoplasms and Acute Leukemia: integrating morphological, clinical, and genomic data. Blood. 2022;140(11):1200-1228. 2. Khoury JD, Solary E, Abla O, et al. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: myeloid and histiocytic/dendritic neoplasms. Leukemia. 2022;36(7):1703-1719. 3. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood. 2016;127(20):2391-2405. 4. Itonaga H, Ota S, Ikeda T, et al. Allogeneic hematopoietic stem cell transplantation for the treatment of BCR-ABL1-negative atypical chronic myeloid leukemia and chronic neutrophil leukemia: a retrospective nationwide study in Japan. Leuk Res. 2018;75:50-57. 5. Park S, Labopin M, Yakoub-Agha I, et al. Allogeneic stem cell transplantation for chronic myelomonocytic leukemia: a report from the Societe Francaise de Greffe de Moelle et de Therapie Cellulaire. Eur J Haematol. 2013;90(5):355-364. 6. Robin M, de Wreede LC, Padron E, et al. Role of allogeneic transplantation in chronic myelomonocytic leukemia: an international collaborative analysis. Blood. 2022;140(12):1408-1418. 7. Sharma P, Shinde SS, Damlaj M, et al. Allogeneic hematopoietic stem cell transplant in adult patients with myelodysplastic syndrome/myeloproliferative neoplasm (MDS/MPN) overlap syndromes. Leuk Lymphoma. 2017;58(4):872-881. 8. Ambinder A, Jain T, Tsai HL, Horowitz MM, Jones RJ, Varadhan R. HLA-matching with PTCy: a reanalysis of a CIBMTR dataset with propensity score matching and donor age. Blood Adv. 2022;6(14):4335-4346. 9. Kunte S, Rybicki L, Viswabandya A, et al. Allogeneic blood or marrow transplantation with haploidentical donor and posttransplantation cyclophosphamide in patients with myelofibrosis: a multicenter study. Leukemia. 2022;36(3):856-864. 10. Luznik L, O'Donnell PV, Symons HJ, et al. HLA-haploidentical bone marrow transplantation for hematologic malignancies using nonmyeloablative conditioning and high-dose, posttransplantation cyclophosphamide. Biol Blood Marrow Transplant. 2008;14(6):641-650. 11. Webster JA, Luznik L, Tsai HL, et al. Allogeneic transplantation for Ph+ acute lymphoblastic leukemia with posttransplantation cyclophosphamide. Blood Adv. 2020;4(20):5078-5088. 12. Karantanos T, Tsai HL, Gondek LP, et al. Genomic landscape of myelodysplastic/myeloproliferative neoplasm can predict response to hypomethylating agent therapy. Leuk Lymphoma. 2022;63(8):1942-1948. 13. Mei M, Pillai R, Kim S, et al. The mutational landscape in chronic myelomonocytic leukemia and its impact on allogeneic hematopoietic cell transplantation outcomes: a Center for Blood and Marrow Transplantation Research (CIBMTR) analysis. Haematologica. 2023;108(1):150-160. 14. Palomo L, Meggendorfer M, Hutter S, et al. Molecular landscape and clonal architecture of adult myelodysplastic/myeloproliferative neoplasms. Blood. 2020;136(16):1851-1862. 15. Karantanos T, Gondek LP, Varadhan R, et al. Gender-related differences in the outcomes and genomic landscape of patients with myelodysplastic syndrome/myeloproliferative neoplasm overlap syndromes. Br J Haematol. 2021;193(6):1142-1150.

16. Elena C, Galli A, Such E, et al. Integrating clinical features and genetic lesions in the risk assessment of patients with chronic myelomonocytic leukemia. Blood. 2016;128(10):1408-1417. 17. Greenberg PL, Tuechler H, Schanz J, et al. Revised international prognostic scoring system for myelodysplastic syndromes. Blood. 2012;120(12):2454-2465. 18. Savona MR, Malcovati L, Komrokji R, et al. An international consortium proposal of uniform response criteria for myelodysplastic/myeloproliferative neoplasms (MDS/MPN) in adults. Blood. 2015;125(12):1857-1865. 19. Kharfan-Dabaja MA, Kumar A, Ayala E, et al. Standardizing definitions of hematopoietic recovery, graft rejection, graft failure, poor graft function, and donor chimerism in allogeneic hematopoietic cell transplantation: a report on behalf of the American Society for Transplantation and Cellular Therapy. Transplant Cell Ther. 2021;27(8):642-649. 20. Lowsky R, Messner H. Mechanisms and treatment of graft failure. In: Forman SJ, Negrin RS, Antin JH, Appelbaum FR, editors. Thomas’ Hematopoietic Cell Transplantation: Stem Cell Transplantation, I, Fifth Edition. Hoboken (NJ): Wiley-Blackwell; 2015. p. 944-958. 21. Przepiorka D, Weisdorf D, Martin P, et al. 1994 Consensus Conference on Acute GVHD Grading. Bone Marrow Transplant. 1995;15(6):825-828. 22. Rowlings PA, Przepiorka D, Klein JP, et al. IBMTR Severity Index for grading acute graft-versus-host disease: retrospective comparison with Glucksberg grade. Br J Haematol. 1997;97(4):855-864. 23. Patnaik MM, Vallapureddy R, Lasho TL, et al. EZH2 mutations in chronic myelomonocytic leukemia cluster with ASXL1 mutations and their co-occurrence is prognostically detrimental. Blood Cancer J. 2018;8(1):12. 24. Thota S, Viny AD, Makishima H, et al. Genetic alterations of the cohesin complex genes in myeloid malignancies. Blood. 2014;124(11):1790-1798. 25. Tsai SC, Shih LY, Liang ST, et al. Biological activities of RUNX1 mutants predict secondary acute leukemia transformation from chronic myelomonocytic leukemia and myelodysplastic syndromes. Clin Cancer Res. 2015;21(15):3541-3551. 26. Gray RJ. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat. 1988;16:1141-1154. 27. Cox DR. Regression models and life tables (with discussion). J R Statist Soc B. 1972;34(2):187-220. 28. Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Statist Assoc. 1999;94:496-509. 29. Rollison DE, Howlader N, Smith MT, et al. Epidemiology of myelodysplastic syndromes and chronic myeloproliferative disorders in the United States, 2001-2004, using data from the NAACCR and SEER programs. Blood. 2008;112(1):45-52. 30. Kurosawa S, Shimomura Y, Tachibana T, et al. Outcome of allogeneic hematopoietic stem cell transplantation in patients with myelodysplastic/myeloproliferative neoplasmsunclassifiable: a retrospective nationwide study of the Japan Society for Hematopoietic Cell Transplantation. Biol Blood Marrow Transplant. 2020;26(9):1607-1611. 31. Jain T, Tsai HL, DeZern AE, et al. Post-transplantation cyclophosphamide-based graft-versus-host disease prophylaxis with nonmyeloablative conditioning for blood or marrow transplantation for myelofibrosis. Transplant Cell Ther.

Haematologica | 108 December 2023

3331

ARTICLE - Haplo-HCT in MDS/MPN

T. Jain et al.

2022;28(5):259. 32. Harrison C, Kiladjian JJ, Al-Ali HK, et al. JAK inhibition with ruxolitinib versus best available therapy for myelofibrosis. N Engl J Med. 2012;366(9):787-798. 33. Harrison CN, Schaap N, Vannucchi AM, et al. Janus kinase-2 inhibitor fedratinib in patients with myelofibrosis previously treated with ruxolitinib (JAKARTA-2): a single-arm, open-label, non-randomised, phase 2, multicentre study. Lancet Haematol. 2017;4(7):e317-e324. 34. Verstovsek S, Mesa RA, Gotlib J, et al. A double-blind, placebocontrolled trial of ruxolitinib for myelofibrosis. N Engl J Med. 2012;366(9):799-807. 35. Padron E, Dezern A, Andrade-Campos M, et al. A multiinstitution phase I trial of ruxolitinib in patients with chronic myelomonocytic leukemia (CMML). Clin Cancer Res. 2016;22(15):3746-3754. 36. Padron E, Painter JS, Kunigal S, et al. GM-CSF-dependent pSTAT5 sensitivity is a feature with therapeutic potential in chronic myelomonocytic leukemia. Blood. 2013;121(25):5068-5077. 37. Sanber K, Ye K, Tsai HL, et al. Venetoclax in combination with hypomethylating agent for the treatment of advanced myeloproliferative neoplasms and acute myeloid leukemia with extramedullary disease. Leuk Lymphoma. 2023;64(4):846-855. 38. Jain T, Alahdab F, Firwana B, Sonbol MB, Almader-Douglas D, Palmer J. Choosing a reduced-intensity conditioning regimen for allogeneic stem cell transplantation, fludarabine/busulfan versus fludarabine melphalan: a systematic review and meta-

analysis. Biol Blood Marrow Transplant. 2019;25(4):728-733. 39. Jain T, Kunze KL, Temkit M, et al. Comparison of reduced intensity conditioning regimens used in patients undergoing hematopoietic stem cell transplantation for myelofibrosis. Bone Marrow Transplant. 2019;54(2):204-211. 40. McLornan D, Szydlo R, Koster L, et al. Myeloablative and reduced-intensity conditioned allogeneic hematopoietic stem cell transplantation in myelofibrosis: a retrospective study by the Chronic Malignancies Working Party of the European Society for Blood and Marrow Transplantation. Biol Blood Marrow Transplant. 2019;25(11):2167-2171. 41. Robin M, Porcher R, Wolschke C, et al. Outcome after transplantation according to reduced-intensity conditioning regimen in patients undergoing transplantation for myelofibrosis. Biol Blood Marrow Transplant. 2016;22(7):1206-1211. 42. Padron E, Garcia-Manero G, Patnaik MM, et al. An international data set for CMML validates prognostic scoring systems and demonstrates a need for novel prognostication strategies. Blood Cancer J. 2015;5(7):e333. 43. Duchmann M, Yalniz FF, Sanna A, et al. Prognostic role of gene mutations in chronic myelomonocytic leukemia patients treated with hypomethylating agents. EBioMedicine. 2018;31:174-181. 44. Gagelmann N, Badbaran A, Beelen DW, et al. A prognostic score including mutation profile and clinical features for patients with CMML undergoing stem cell transplantation. Blood Adv. 2021;5(6):1760-1769.

Haematologica | 108 December 2023

3332

ARTICLE - Non-Hodgkin Lymphoma

Long non-coding RNA mitophagy and ALK-negative anaplastic lymphoma-associated transcript: a novel regulator of mitophagy in T-cell lymphoma Valentina Mularoni,1 Benedetta Donati,1 Annalisa Tameni,1 Veronica Manicardi,1 Francesca Reggiani,1 Elisabetta Sauta,2 Magda Zanelli,3 Marco Tigano,4 Emanuele Vitale,1,5 Federica Torricelli,1 Stefano Ascani,6 Giovanni Martino,6,7 Giorgio Inghirami,8 Francesca Sanguedolce,9 Alessia Ruffini,10 Alberto Bavieri,10 Stefano Luminari,10 Marco Pizzi,11 Angelo Paolo Dei Tos,11 Cinzia Fesce,12 Antonino Neri,13 Alessia Ciarrocchi1 and Valentina Fragliasso1 Laboratory of Translational Research, Azienda USL-IRCCS di Reggio Emilia, Reggio Emilia, Italy; 2IRCCS Humanitas Clinical and Research Center, Milan, Italy; 3Pathology Unit, Department of Oncology, Oncology, Azienda USL-IRCCS di Reggio Emilia, Reggio Emilia, Italy; 4Sidney Kimmel Medical College, Thomas Jefferson University, Philadelphia, PA, USA; 5 Clinical and Experimental Medicine Ph.D. Program, University of Modena and Reggio Emilia, Modena, Italy; 6Pathology Unit, Azienda Ospedaliera Santa Maria di Terni, University of Perugia, Terni, Italy; 7Institute of Hematology and CREO, University of Perugia, Perugia, Italy; 8 Department of Pathology and Laboratory Medicine, Weill Cornell Medicine, New York, NY, USA; 9Pathology Unit, Policlinico Riuniti, University of Foggia, Foggia, Italy; 10Hematology Unit, Azienda USL-IRCCS di Reggio Emilia, Reggio Emilia, Italy; 11Surgical Pathology & Cytopathology Unit, Department of Medicine-DIMED, University of Padova, Padova, Italy; 12 Hematology Unit, University Hospital, Foggia, Italy and 13Scientific Directorate, Azienda USL-IRCCS di Reggio Emilia, Reggio Emilia, Italy

Correspondence: V. Fragliasso Valentina.Fragliasso@ausl.re.it Received: Accepted: Early view:

December 14, 2022. June 20, 2023. June 29, 2023.

Abstract Long non-coding RNA (lncRNA) are emerging as powerful and versatile regulators of transcriptional programs and distinctive biomarkers of progression of T-cell lymphoma. Their role in the aggressive anaplastic lymphoma kinase-negative (ALK–) subtype of anaplastic large cell lymphoma (ALCL) has been elucidated only in part. Starting from our previously identified ALCL-associated lncRNA signature and performing digital gene expression profiling of a retrospective cohort of ALCL, we defined an 11 lncRNA signature able to discriminate among ALCL subtypes. We selected a not previously characterized lncRNA, MTAAT, with preferential expression in ALK– ALCL, for molecular and functional studies. We demonstrated that lncRNA MTAAT contributes to an aberrant mitochondrial turnover restraining mitophagy and promoting cellular proliferation. Functionally, lncRNA MTAAT acts as a repressor of a set of genes related to mitochondrial quality control via chromatin reorganization. Collectively, our work demonstrates the transcriptional role of lncRNA MTAAT in orchestrating a complex transcriptional program sustaining the progression of ALK– ALCL.

Introduction T-cell lymphoma (TCL) is a complex and heterogeneous group of neoplasms with different biology and outcome.1 Its diagnostic classification is still a challenge, making its treatment suboptimal. Anaplastic lymphoma kinasenegative (ALK–) anaplastic large cell lymphoma (ALCL) is one of the most aggressive subtypes of TCL and is characterized by a dismal prognosis and high mortality.2-4 The molecular details of the pathogenesis of ALK– ALCL are largely unknown, thus limiting the development of targeted strategies.5 Therefore, clarifying the mechanisms underlying this disease is of utmost importance.

The fast and massive implementation of deep sequencing technologies has highlighted the role of the non-coding genome in regulating cancer proliferation, especially in complex tumors without a clear genetic driver.6,7 Particular attention has been paid to a family of transcripts of about 200 bp with no coding potential, known as long noncoding RNA (lncRNA).8 LncRNA possess the striking ability to interact with protein-coding and non-coding transcripts regulating and integrating a complex network of processes simultaneously.9-12 Thanks to their peculiar and versatile features, lncRNA are able to influence – in a context-dependent manner – several aspects of cancer biology, such as cellular proliferation, apoptosis, metabolic reprogram-

Haematologica | 108 December 2023

3333

ARTICLE - Transcriptional role of lncRNA MTAAT in lymphoma ming, genomic instability, drug resistance, invasion, and metastasis.13,14 Thus, deregulating the expression of certain lncRNA can directly influence the pathogenesis and/or progression of different types of cancers, including TCL.15 Previously, we investigated the contribution of lncRNA to ALCL pathogenesis by performing deep transcriptomic profiling of a cohort of ALCL. We identified a unique set of 18 lncRNA overexpressed in neoplastic T-lymphocytes compared to normal T-lymphocytes.16 We were able to show that among those, the lncRNA BlackMamba acts as a major transcriptional regulator of neoplastic T-lymphocytes in the ALK– ALCL subtype, influencing lymphoma progression.16,17 Therefore, the identified lncRNA are not only distinctive biomarkers of disease but also play relevant functional roles in the biogenesis of ALCL. In this work, we expand our knowledge regarding the functional relevance of lncRNA in the development and progression of ALCL. Starting from our previously identified lncRNA signature and integrating the gene expression profiles of a large cohort of ALCL patients with clinical information, we defined a powerful lncRNA signature that discriminates between ALK– and ALK+ subtypes. We then focused on the XLOC_211989 transcript – which resulted specifically associated with the ALK– ALCL subtype – and characterized its function by using multiple functional approaches. Our data suggest that this non-coding transcript coordinates the expression of a set of genes involved in mitochondrial homeostasis and mitophagy promoting lymphoma cell survival. We named this novel lncRNA MiTophagy and ALK– Anaplastic Lymphoma Associated Transcript (MTAAT).

Methods Patients’ specimens Fresh and viable cryopreserved cells and formalin-fixed paraffin-embedded (FFPE) sections of two retrospective cohorts of ALCL were isolated from diagnostic/relapsed primary lymphoma biopsies. Diagnoses were assigned according to the World Health Organization classification.1 Tissues used for expression analyses were selected for their high tumor cell content (>50%). The FFPE cohort of samples of ALCL consisted of 54 cases whereas the freshly frozen cohort of samples consisted of 18 ALCL cases. Regarding clinicopathological characteristics of the FFPE cohort eligible for the analysis (n=44), among ALK– ALCL, 17/29 (59%) patients were male and the median age was 70 years (range, 31-88). Regarding ALK+ ALCL, 8/15 (53%) patients were male and the median age was 39 years (range, 21-75). Considering follow-up data available for 29/44 patients, the median follow-up of the cohort was 44 months (range, 2.2-152). The 4-year progression-free survival rate was 70.6% (95% confidence interval: 55.2-

V. Mularoni et al.

90.3). The freshly frozen cohort of ALCL and immunophenotypic features of resting and donor CD4+ T-lymphocytes have been described previously.16 The study was approved through institutional human ethics review boards of the Ethical Committee AVEN and AUSL-IRCCS of Reggio Emilia (287/2018/OSS/IRCCSRE), and patients provided written informed consent in accordance with the Declaration of Helsinki. RNA extraction and quantitative polymerase chain reaction Total RNA from cells was extracted by TRIzol (Thermo Fisher Scientific) according to the manufacturer’s instructions. One microgram of total RNA was retrotranscribed using the iScript cDNA kit, (Biorad). The amplified transcript level of each specific gene was normalized on the CHMP2A housekeeping gene. The ΔΔCt quantification method was used for quantitative polymerase chain reaction (RT-qPCR) analyses. The list of primers used is provided in Online Supplementary Table S1. Antisense LNA GapmeR transfection MAC2A and TLBR-2 cells (1×106) were transfected with 50 nM Antisense LNA GapmeR for a single knockdown (KD). Antisense LNA GapmeR transfections were performed using the Cell Line Nucleofector Kit SF and Amaxa 4D Nucleofector (program DS-130 for TLBR-2; FI115 for MAC2A). Twenty-four hours after transfection, cells were harvested and plated at 2.5×105 cells/mL. Antisense LNA GapmeR Negative Control (Cat. N/ID: 33951, Qiagen) was used as a negative control. For lncRNA_211989/MTAAT we used two different GapmeRs (Cat. N/ID: 339511, Qiagen); their sequences are provided in Online Supplementary Table S2. Mitochondrial staining Mitochondria were stained with Mitotracker (Thermo Fisher Scientific) according to the manufacturer’s instructions. Cells were then harvested, fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS) 1X for 10 min at room temperature, and spotted on glass slides using Cytospin (Thermo Scientific) as previously reported.18 Dots were washed in PBS 1X three times and nuclei were stained with DAPI. Tetramethylrhodamine methyl ester perchlorate (TMRM) staining was performed according to the manufacturer’s instructions (Thermo Fisher Scientific), without substantial changes. Cells were then harvested and washed once in RPMI medium without serum. Membrane potential was immediately measured by flow cytometry with a FACSCanto™ II Cell Analyzer (BD Biosciences). Statistical analysis Statistical analyses were performed using GraphPad Prism software (GraphPad). Statistical significance was deter-

Haematologica | 108 December 2023

3334

ARTICLE - Transcriptional role of lncRNA MTAAT in lymphoma mined using a Student t test. Each experiment was replicated multiple times (>3 up to 6). All analyses were performed using R software version 4.1.3.

Results XLOC_211989 is a novel biomarker that stratifies patients with ALK– anaplastic large cell lymphoma We assessed whether the previously generated ALCL-associated lncRNA signature16 might be used to distinguish ALCL subtypes in a retrospective cohort of 54 ALCL cases. Gene expression analyses were performed by digital multiplexing profiling using a custom panel of probes targeting the 17 previously identified lncRNA and four additional

V. Mularoni et al.

genes used for the molecular classification of ALCL subtypes.19 Out of the total of 54 samples, 44 ALCL gene expression profiles passed stringent quality controls and resulted eligible for the analysis (Figure 1A). Of these 44 samples, 15 (34%) were classified as ALK+ ALCL and 29 (66%) scored as ALK– ALCL (Figure 1A, Online Supplementary Figure S1A, B, Online Supplementary Table S4). Focusing on lncRNA, we confirmed that 17/17 (100%) lncRNA were expressed in the ALCL cohort. Principal component analysis showed that lncRNA expression profiles segregated well with ALK+ and ALK– ALCL samples (Figure 1B) resulting in 11 of the 17 (65%) lncRNA significantly deregulated between the two subtypes (Figure 1C, Online Supplementary Figure S1C). In particular: seven lncRNA were overexpressed in ALK– ALCL

Continued on following page. Haematologica | 108 December 2023

3335

ARTICLE - Transcriptional role of lncRNA MTAAT in lymphoma

V. Mularoni et al.

Figure 1. XLOC_211989 stratifies patients with ALK– anaplastic large cell lymphoma. (A) Outline of the study workflow for the formalin-fixed paraffin-embedded (FFPE) training cohort of samples to classify patients with anaplastic large cell lymphoma (ALCL) based on transcriptomic profiles. Figure created with BioRender.com. (B) Principal component analysis shows the variance between ALK+ ALCL (red dots) and ALK– ALCL (gray dots) samples of the FFPE training cohort explained by the expression level of the 17 long non-coding RNA (lncRNA). (C) Dot plots show the distributions of gene expression counts of the 11 statistically significant lncRNA among ALK+ ALCL (red dots) and ALK– ALCL (black dots) of the FFPE training cohort. Comparisons were considered statistically significant for P≤0.05 (*). (D) An alluvial plot shows the workflow for the identification and validation of XLOC_211989. (E) Box plot representing the expression of XLOC_211989 among ALK+ (red) and ALK– (gray) ALCL evaluated by quantitative polymerase chain reaction in the validation cohort. The comparison was considered statistically significant for P≤0.05 (*). (F) Correlation plot between ALK and XLOC_211989 normalized and log2 transformed gene counts in the FFPE training cohort (left panel). Correlation plot between BlackMamba and XLOC_211989 normalized and log2-transformed gene counts in the training cohort (right panel). (G) Receiver operating characteristic curve showing the performance of the XLOC_211989-based scoring system in the training cohort of ALCL patients. The curve is colored according to XLOC_211989 adjusted values, the relation between colors and different values is represented by the bar on the right. The formula was generated by a generalized linear regression model.

patients while four lncRNA were overexpressed in ALK+ ALCL patients (Figure 1C). In accordance with our previous observations,16 lncRNA BlackMamba showed a significant, negative correlation with ALK expression (Figure 1C). We focused our attention on the six uncharacterized lncRNA that were significantly associated with the ALK– subtype (Figure 1D). First, we confirmed their association with the ALK– subtype by targeted RT-qPCR performed in a previously published cohort of 15 freshly frozen ALCL.16 From this analysis, XLOC_211989 – herein named lncRNA MTAAT – showed the strongest and most significant association with ALK– ALCL (Figure 1D, E, Online Supplementary Figure S1D). No detectable MTAAT expression was observed in donor resting or activated CD4+ cells (Online Supplementary Figure S1E), further confirming ALCL-restricted MTAAT expression.

To strengthen these observations, we explored the correlation between MTAAT and ALK expression in the retrospective cohort of ALCL samples. Linear regression analysis showed that MTAAT was inversely correlated with ALK expression and positively correlated with lncRNA BlackMamba (Figure 1F). The receiver operating characteristic curve for ALK subtype classification showed that MTAAT has a high capacity (70%) to discriminate between ALK– and ALK+ patients (area under the curve=0.70; 95% confidence interval: 0.54-0.85) (Figure 1G, Online Supplementary Figure S1F). The MTAAT promoter is bound by RNA polymerase II and enriched for active histone marks Genomic annotation showed that the MTAAT sequence matches an uncharacterized intergenic transcript en-

Haematologica | 108 December 2023

3336

ARTICLE - Transcriptional role of lncRNA MTAAT in lymphoma coded on the plus strand of chromosome 3 with an estimated transcript length of 7,189 bp and no predicted alternative isoforms (Figure 2A). In silico analysis predicted four potential open reading frames with irrelevant coding potential within the MTAAT sequence, confirming the non-coding nature of this transcript (Online Supplementary Table S5). Given the specificity of MTAAT expression in the ALK– ALCL subtype, we sought to define in vitro the molecular mechanisms that control its expression. For this analysis, we chose the two ALK– ALCL cell lines, TLBR-2 and MAC2A, displaying the highest levels of MTAAT expression (Online Supplementary Figure S2A). To identify the promoter of MTAAT, we first analyzed RNA polymerase II (RNAPII) genomic occupancy and histone 3 trimethyl lysine 4 (H3K4me3) profile using chromatin immunoprecipitation (ChIP) followed by sequencing in TLBR-2 cells. A high-density distribution of RNAPII within a 2000 bp region spanning the putative transcription start site (TSS) of MTAAT (P3-P6) was observed. This region was also enriched in H3K4me3, confirming the promoter-like nature of its sequence (Figure 2A). We confirmed the findings by ChIP-qPCR in both TLBR-2 and MAC2A cell lines (Figure 2B, C, Online Supplementary Figure S2B). We also showed that this region is marked by a high level of histone H3 acetyl-lysine 27 (H3K27ac) confirming that this locus is transcriptionally active in these cellular models (Figure 2D, Online Supplementary Figure S2B). Notably, ChIP-qPCR analysis did not show RNAPII or histone modification enrichment in a cell line, CUTLL1, negative for MTAAT expression (Online Supplementary Figure S2C), validating the specificity of our observations. To assess whether the putative promoter of MTAAT is able to transactivate transcription, we cloned the 2000 bp DNA sequence spanning from −1,500 bp to +500 bp of the MTAAT-TSS upstream of a luciferase reporter cassette. In this “promoter-like” configuration, high luciferase activity was detected in both MAC2A and TLBR-2 cells (Figure 2E). Next, we searched for the signaling pathways underlying MTAAT expression. Based on the genomic profiles observed in ChIP-sequencing, we selected a 500 bp region spanning the TSS of MTAAT and performed a motif search analysis to identify potential transcription factor binding sites. For this, we used the FIMO analysis pipeline20 and identified 117 hypothetical transcription factors (Online Supplementary Table S6). Notably, some transcription factors, including STAT, GATA, and IRF family members, are pertinent to signaling pathways known to be active and deregulated in ALK– ALCL5 (Figure 2F). Specifically, we found a significant enrichment of several pathways related to the cellular response to cytokines, interferon, interleukins, and regulation of Tcell differentiation.

V. Mularoni et al.

MTAAT is a chromatin-associated long non-coding RNA essential for the transcriptional control of mitochondrial processes To examine the biological role of MTAAT in TCL, we first studied MTAAT cellular localization performing subcellular fractionation. We found that MTAAT was enriched in the nucleus and strongly associated with the chromatin fraction of lymphoma cells (Figure 3A, Online Supplementary Figure S3A) suggesting a putative role in chromatin organization and gene expression regulation. To investigate the role of this lncRNA in regulating lymphoma transcription, we silenced MTAAT expression by targeting different regions, single or in combination, with gapmer technology. MTAAT expression was measured by RT-qPCR and the delivery of multiple gapmers by electroporation resulted in effective knockdown (KD) (>50%) of MTAAT across all ALK– ALCL cell lines tested (Figure 3B). We then used next-generation RNA sequencing to evaluate the genome-wide transcriptional changes triggered by MTAAT silencing (MTAATKD). TLBR-2 cells were subjected to MTAATKD and RNA was collected 24 h after. In parallel and as a control, a scrambled gapmer was also delivered. RNA-sequencing analysis revealed 2,217 differentially expressed genes in MTAATKD compared to the gapmer control. Among these, 67.5% (1,497 genes) were protein-coding. Specifically, we detected 524 downregulated and 937 upregulated genes upon lncRNA MTAATKD (false discovery rate <0.1) (Figure 3C). These findings suggest a role for MTAAT in both the activation and the repression of transcription. Notably, the genomic regions of MTAAT targets are far beyond chromosome 3 (Figure 3D). This suggests that MTAAT could regulate ALK– ALCL transcriptional programs in trans and at a genome-wide level. Gene-set enrichment analysis revealed diverse biological processes associated with deregulated genes. Specifically, downregulated transcripts showed significant enrichment in several gene sets related to mitochondrial respiratory chain complexes, DNA damage response, and chromatin organization. In contrast, upregulated transcripts are mostly implicated in immune response, glycolytic process, integrated stress response as well as regulation of mitochondrion organization (Figure 3E). Using RT-qPCR, we validated a representative set of upregulated genes confirming the RNA-sequencing results (Figure 3F). To exclude off-target effects of the gapmers designed for MTAATKD, we decided to corroborate the results by silencing MTAAT with a CRISPR-interference system, by using doxycycline-inducible dCas9-KRAB and two different single-guide RNA targeting the MTAAT promoter. Following lentiviral transduction of ALK– ALCL cells, we induced dCas9-KRAB for 48 h with doxycycline and evaluated MTAAT expression. RT-qPCR confirmed that both singleguide RNA repressed the level of MTAAT by >60% (Online Supplementary Figure S3B-D). Importantly, the expression

Haematologica | 108 December 2023

3337

ARTICLE - Transcriptional role of lncRNA MTAAT in lymphoma

V. Mularoni et al.

Figure 2. MTAAT is a novel long non-coding RNA. (A) Schematic representation of the locus and structure of MTAAT showing the position and enrichment level of active marks of transcription assessed by chromatin immunoprecipitation (ChIP) sequencing in TLBR-2 cells. (B-D) ChIP-quantitative polymerase chain reaction detection of RNAPII (B), H3K4me3 (C), and H3K27Ac (D) markers on MTAAT fragments in ALK- anaplastic large cell lymphoma (ALCL) cell lines. The GAPDH promoter was used as a positive control (CTR+) whereas a non-coding intergenic region served as a negative control (CTR−). The values are representative of three independent experiments. (E) Luciferase activity of the MTAAT fragment in a pGL3 vector. Data are represented as a normalized ratio of firefly-Renilla luciferase activities and are expressed as mean values ± standard deviation (N=3). **P≤0.01. (F) Graphs show the most representative enriched transcription factor-related pathways. For each pathway, the transcription factor with the major number of interactions is colored in yellow. These graphs were created using cytoskape software. Haematologica | 108 December 2023

3338

ARTICLE - Transcriptional role of lncRNA MTAAT in lymphoma of 10/12 (83%) gene targets after dCas9-KRAB-mediated MTAAT silencing was consistent with gapmer MTAATKD, ruling out any off-target effects (Online Supplementary Figure S3E). Collectively, the transcriptional changes that we observed upon MTAATKD indicated that this lncRNA acts as a repressor of a set of genes related to mitochondrial quality control. MTAAT represses BNIP3 and BNIP3L via histone modifications To understand how the lncRNA MTAAT regulates the expression of mitochondria-related genes, we evaluated changes in the chromatin organization triggered upon MTAATKD investigating, by ChIP, the distribution of H3K4Me3, H3K27Ac, and RNAPII on BCL2 Interacting Protein 3 (BNIP3) and BCL2 Interacting Protein 3 Like (BNIP3L also known as NIX). These proteins resulted in target genes of lncRNA MTAAT and their loss has been implicated in the accumulation of dysfunctional mitochondria in the hematopoietic system.21,22 After the depletion of MTAAT,

V. Mularoni et al.

H3K4Me3 and H3K27Ac levels increased significantly in BNIP3 and BNIP3L promoters (Figure 4A-C). Likewise, RNAPII was found to be dramatically enriched around the TSS of both genes upon MTAATKD (Figure 4D). Similar changes were observed in additional MTAAT target genes such as Activating Transcriptional Factor 4 (ATF4) and X-Box Binding Protein 1 (XBP1) (Online Supplementary Figure S4A-D). Concordantly with the gene expression profile, no changes were observed in Optineurin (OPTN) gene (Online Supplementary Figure S4A-E). Furthermore, in silico analysis performed with catRAPID23 showed a high interaction propensity of MTAAT with H3K27 methylation complex (Online Supplementary Figure S4F). Collectively, these data confirm that changes in chromatin markers are directly linked to the activity of MTAAT on its target genes. To strengthen the clinical relevance of MTAAT regulation on these genes, we investigated the expression of BNIP3 and BNIP3L in the retrospective cohort of ALCL included in this study. In line with the repressive effect of MTAAT, BNIP3 expression was lower in ALK– ALCL than in ALK+

Continued on following page. Haematologica | 108 December 2023

3339

ARTICLE - Transcriptional role of lncRNA MTAAT in lymphoma

V. Mularoni et al.

Figure 3. MTAAT knockdown results in transcriptional changes. (A) Relative expression of MTAAT in the cytoplasm, nucleus, and chromatin fraction of ALK– anaplastic large cell lymphoma (ALCL) cell lines (average of three independent experiments ± standard deviation). One-tailed t test, **P≤0.01. (B) Quantitative polymerase chain reaction (RT-qPCR) analysis of MTAAT expression 24 h after gapmer nucleofection in MAC2A and TLBR-2 cells. One-tailed t test, **P≤0.01. (C) The heatmap depicts hierarchical clustering based on 2,217 differentially expressed genes, whose read counts are Z-score normalized. Unsupervised hierarchical clustering was performed between gapmer control and gapmer #1+#2 samples (as indicated by the colored bar on columns) with a complete linkage method. Color intensity for each gene shows Z-score values ranging from red for upregulation to green for downregulation (left panel). The graph shows the number and the category of deregulated genes after RNA sequencing in TLBR-2 nucleofected with gapmers #1+#2 compared to gapmer control. (D) The circular plot displays the genomic location of MTAAT (highlighted with a red line) and the protein-coding genes differentially expressed upon its knockdown. Red links connect the long non-coding (lncRNA) MTAAT to upregulated genes, green links connect the lncRNA MTAAT to downregulated genes (right panel). (E) Most significantly enriched pathways (adjusted P<0.05) are represented showing the number of up- and down-regulated differentially expressed genes mapped in each considered pathway (left panel). The heatmap depicts validated significantly downregulated genes. The orange-red color bar shows the fold difference on the log2 scale calculated between doxycycline-treated and control samples. Lighter orange represents the most downregulated genes (right panel). (F) RT-qPCR validation of significantly upregulated genes obtained from RNA sequencing in TLBR-2 cells after nucleofection with gapmers (24 h). Each datum represents the mean ± standard error of mean (N=3). Two-tailed t test, *P<0.05; **P<0.01.

ALCL patients (Figure 4E). A similar gene expression correlation was observed in a panel of non-TCL cell lines (Online Supplementary Figure S4G). In contrast, no significant differences were observed between ALCL subtypes for BNIP3L expression (Figure 4E). MTAAT sustains the growth of anaplastic large cell lymphoma by regulating mitophagy The transcriptional changes observed upon MTAATKD are suggestive of specific disruptions in mitochondrial homeostasis, such as an aberrant increase in mitochondrial density or changes in mitochondrial morphology. We wondered

whether MTAAT promotes ALCL progression by controlling mitochondrial clearance. First, we investigated whether mitochondrial abundance changes in TCL upon dCas9-KRABinducible MTAATKD. We evaluated mitochondrial mass by Mitotracker staining and cytofluorimetric analysis. This analysis showed a time-dependent reduction in Mitotracker intensity signal upon MTAATKD (Figure 5A), whereas doxycycline treatment alone did not lead to changes (data not shown). Along the same line of evidence, we observed a strong reduction in mitochondrial DNA copy number, as determined by RT-qPCR analysis of the mitochondrial gene ND1 (Figure 5B). Furthermore, the steady-state level of sev-

Haematologica | 108 December 2023

3340

ARTICLE - Transcriptional role of lncRNA MTAAT in lymphoma

V. Mularoni et al.

Figure 4. MTAAT knockdown modifies active transcriptional markers on regulatory elements of gene targets of MTAAT. (A) Graphs show the transcription start site (TSS) and fragments around the TSS of BNIP3 and BNIP3L. (B-D) Chromatin immunoprecipitation and quantitative polymerase chain reaction detection of H3K4me3 (B), H3K27Ac (C), and RNAPII (D) on MTAAT-gene target fragments in the TLBR-2 MTAATKD cell line. The values are representative of three independent experiments. (E) Box plots show the expression level of BNIP3 and BNIP3L in the cohort of formalin-fixed paraffin-embedded samples of anaplastic large cell lymphoma. One-tailed t test, *P≤0.01. DOX: doxycycline; CTR: control.

eral mitochondrial proteins (cytochrome c oxidase subunit IV, superoxide dismutase, cytochrome, and prohibitin 1), assessed by western blot, confirmed these findings (Figure 5C). Since oxidative function is strictly linked to mitochondrial network dynamics, we evaluated the mitochondrial morphology of Mitotracker-stained cells using immunofluorescence. In a basal condition, ALK– ALCL cells showed the mitochondrial network predominantly distributed around the perinuclear region. Upon depletion of MTAAT, mitochondria displayed a more apical/basal localization which is indicative of a less active mitochondrial state24 (Figure 5D). Concordantly, mitochondrial membrane potential, evaluated by TMRM staining, was reduced upon MTAATKD, suggesting mitochondrial dysfunction (Figure 5E). In line with these data, ALK– ALCL patients with high expression of MTAAT showed a high intensity and diffuse staining for superoxide dismutase compared to those expressing low levels of MTAAT (Online Supplementary Figure S5A). Growing evidence points to a strong relationship between BNIP3, which acts as an adaptor for tethering mitochondria

to nascent autophagosomes, and the activation of a selective form of macroautophagy known as mitophagy.25 Having observed the overexpression of BNIP3 upon MTAATKD, we asked whether the observed changes in mitochondrial mass were due to the activation of mitophagy. First, we assessed whether canonical mitophagy markers can be detected upon MTAAT silencing. Notably, MTAATKD induced a time-dependent decrease of LC3 and of the autophagy receptor SQSTM1/p62 in both cell lines, suggestive of increased autophagy flux. Supporting this, treatment with chloroquine, an autophagy inhibitor that blocks the fusion of autophagosomes with lysosomes, blocked the MTAAT-dependent increase of autophagic flux (Online Supplementary Figure S5B). To strengthen these results, we analyzed the co-localization of specific mitophagy adaptors with autophagosomes. We co-transfected TLBR-2 cells with plasmids encoding LC3-GFP and BNIP3-Flag and analyzed their behavior upon MTAATKD. In control cells, both markers showed diffuse and homogeneous staining across the cytoplasm (Figure 6A). By contrast, upon MTAATKD, both LC3 and BNIP3

Haematologica | 108 December 2023

3341

ARTICLE - Transcriptional role of lncRNA MTAAT in lymphoma

V. Mularoni et al.

Figure 5. MTAAT inactivation impairs mitochondrial homeostasis. (A) Histograms of mean fluorescence intensity (MFI) assessed by flow cytometry analysis show a left shift of Mitotracker red fluorescence in MTAAT knockdown (MTAATKD) cell lines (red and yellow) compared to control (CTR, grey) at 48 and 72 h after doxycycline (DOX) treatment. Bar graphs show the MFI of cells stained with Mitotracker red and analyzed by FACS. Each datum represents the mean ± standard error of mean (SEM) (N=3). Two-tailed t test. *P<0.05; **P<0.01 relative to the CTR. (B) Copy number analysis of the mitochondrial gene ND1 normalized on nuclear gene β-actin (48 and 72 h after DOX treatment). Each datum represents the mean ± SEM (N=3). Two-tailed t test, *P<0.05; **P<0.01 relative to CTR. (C) Western blots show the expression of a set of mitochondrial proteins after MTAAT KD in TLBR-2 and MAC2A cells (72 h). (D) Immunofluorescence shows the localization of mitochondria in MTAATKD cells. Cells were stained with Mitotracker-red dye and DAPI. The white scale bar represents 10 μm. The adjacent bar graphs indicate the percentage of Mitotracker-red stained cells with apical/basal mitochondria (N=500 cells). Each datum represents the mean ± SEM (N=3). Two-tailed t test, *P<0.05; **P<0.01 relative to CTR. (E) Histograms of FACS analysis show a right shift of tetramethylrhodamine methyl ester perchlorate (TMRM) fluorescence in MTAATKD cell lines (red and yellow) compared to the CTR (gray). Bar graphs show the MFI of cells stained with TMRM and analyzed by FACS. Each datum represents the mean ± SEM (N=3). Two-tailed t test, *P<0.05; **P<0.01 relative to CTR. PHB1: prohibitin 1; SOD1: superoxide dismutase; COX IV: cytochrome c oxidase subunit IV; CytC: cytochrome C; α-TUB: αtubulin. Haematologica | 108 December 2023

3342

ARTICLE - Transcriptional role of lncRNA MTAAT in lymphoma accumulated into bright cytoplasmic puncta suggestive of LC3 lipidation and BNIP3 recruitment. Importantly, LC3 and BNIP3 puncta co-localize upon MTAATKD (Figure 6A), indicating active mitophagy. We hypothesized that tumor cells specifically block mitophagy to increase mitochondrial mass and sustain proliferation. We, therefore, asked whether MTAATKD impacts ALK– ALCL cell viability. Notably, growth curve analysis and viability assays showed that depleting MTAAT significantly reduced cellular proliferation in both TLBR-2 and MAC2A cell lines (Figure 6B). No changes were recorded in cell cycle profiles and apoptosis was not induced upon MTAATKD, con-

V. Mularoni et al.

sistent with energy deprivation, rather than cell death (Online Supplementary Figure S5C, D). Collectively, our data support a model in which the lncRNA MTAAT exerts its function by stimulating an increase in mitochondrial mass – and energy output – which is used by ALK– ALCL cells to sustain cell proliferation (Figure 6C).

Discussion The implementation of digital gene expression profiling paved the way for application of a transcriptomic approach

Figure 6. MTAAT knockdown promotes mitophagy and reduces cellular proliferation. (A) Immunofluorescence images of TLBR-2 MTAATKD cells after 48 h of doxycycline induction. Cells were stained with DAPI and FLAG antibodies for BNIP3 detection. The white scale bar represents 10 μm. (B) Growth curves show the proliferation of cells after MTAAT depletion. Each datum represents the mean ± standard error of mean (N=3). Twotailed t test. **P<0.01 relative to the control. (C) Graphical model of MTAAT function in lymphoma. Figure created with BioRender.com. DOX: doxycycline; CTR: control; KD: knockdown.

Haematologica | 108 December 2023

3343

ARTICLE - Transcriptional role of lncRNA MTAAT in lymphoma to the classification of TCL, increasing the precision of diagnosis over conventional methods.26 Progress toward understanding the transcriptional complexity of tumors revealed how coding genes are not the only drivers of cancer progression, with non-coding transcripts, such as lncRNA, regulating essential transcriptional cascades during tumorigenesis.9,27,28 However, how lncRNA drive cellular and clinical phenotypes of aggressive TCL subtypes remains unknown. In this study, we identified a set of lncRNA that act as molecular classifiers to distinguish ALK+ and ALK– ALCL. We also report the role of one of these lncRNA, which we renamed MTAAT, in regulating mitochondrial turnover and progression of aggressive ALK– ALCL. We identified MTAAT as significantly associated with the ALK– ALCL phenotype in two independent cohorts of ALCL patients: first from a cohort of FFPE diagnostic biopsies analyzed by digital expression profiling with the Nanostring nCounter platform and subsequently in a cohort of frozen tissues by RT-qPCR. We also demonstrated the high accuracy of MTAAT in predicting ALK– subtypes in ALCL classification. Although the use of lncRNA as biomarkers is still in its infancy, our results strongly suggest that digital lncRNA profiling could be integrated into diagnostic panels to improve the accuracy and precision of ALCL stratification. We selected the lncRNA MTAAT for functional studies based on its association with ALK– ALCL. Analysis of the regulatory elements of MTAAT showed its intrinsic ability to regulate transcription. Transcriptional regulation by lncRNA appears to be a mechanism widely used by hematologic malignancies to control the transcription of selective pathways tuning aberrant proliferation and survival of B and T cells.29-32 By performing RNA sequencing on MTAATKD cell lines, we highlighted how the transcriptional program supported by MTAAT converges on the regulation of mitochondrial pathways. Mechanistic investigations showed that the loss of MTAAT is linked to a unique phenotype characterized by increased mitochondrial turnover through positive mitophagy stimulation, accompanied by a reduction in cell proliferation. Remarkably, lymphomas are characterized as oxidative tumors, indicating a requirement of mitochondrial function for tumor progression.33,34 Mitochondria work as metabolic hubs to support cell growth and proliferation, and act as sensors of intracellular stresses that could threaten survival.35,36 Although the role of mitophagy in lymphoid malignancies is still debated, some evidence indicates that the constitutive repression of autophagy/mitophagy contributes to lymphomagenesis.37-41 The increase in the mitochondrial pool in tumor cells is also emerging as a key factor in the success of immunotherapeutic treatments,42–44 such as chimeric antigen receptor-expressing T cells.45 These chimeric antigen receptor-expressing cells

V. Mularoni et al.

represent an incredibly promising strategy for the treatment of several malignancies and for this reason, further investigations aimed at elucidating the role of MTAAT are warranted. Among the downstream targets of MTAAT, we identified BNIP3 whose expression is upregulated upon MTAATKD via chromatin reorganization. BNIP3 was originally reported to function as a BH3-only protein which induced programmed cell death.46,47 More recently, it has been shown to function as a stress-induced mitophagy receptor that interacts directly with LC3 to promote the turnover of otherwise healthy mitochondria.48,49 Although various human solid cancers overexpress BNIP3 as they become hypoxic,50 inactivation of BNIP3 via promoter hypermethylation is a common feature of aggressive and advancedstage cancers such as triple-negative breast cancer, hematologic malignancies and advanced-stage pancreatic cancer.51-54 In these tumors, epigenetic silencing of BNIP3 correlates with high cancer cell proliferation, poor prognostic features, and chemoresistance.55,56 In accordance with this observation, we found a significant reduction of BNIP3 expression in a cohort of ALK– ALCL patients. This finding suggests a tumor-suppressive function of BNIP3 also in the context of ALCL. We speculate that the loss of BNIP3 associated with reduced mitophagy may create a more aggressive tumor phenotype and contribute, at least in part, to the chemoresistance observed in this malignancy. This evidence paves the way for the implementation of targeted therapeutic strategies able to re-express BNIP3 in this lymphoid malignancy. In conclusion, we have characterized the novel lncRNA MTAAT as a new potential biomarker in the stratification of ALCL patients. Functionally, MTAAT acts as a transcriptional brake on mitophagy, promoting the accumulation of mitochondria and supporting lymphoma progression. These findings corroborate our previous data showing a key role of lncRNA in the control of different transcriptional programs in ALK– ALCL. Disclosures No conflicts of interest to disclose. Contributions VM and AT performed experiments and analyzed data, BD performed gene expression and Nanostring analyses, VM, FT, ES, and EV performed RNA-sequencing, ChIP-sequencing, and bioinformatics analyses. FR performed FACS analyses and cell sorting experiments, MZ, SA, GI, FS, CF, MP, AR, GM, SL, APDS, and AB provided tissue samples and lymphoma diagnosis. MT and AN provided important experimental and analytic support. AC interpreted the results and helped to discuss the results. VF designed the project, interpreted the results, and wrote the manuscript. All the authors read and approved the final version of the manuscript.

Haematologica | 108 December 2023

3344

ARTICLE - Transcriptional role of lncRNA MTAAT in lymphoma Acknowledgments The authors are grateful to Marina Grassi, Ione Tamagnini, Gloria Venturini, and Riccardo Fuoco for technical help and to all the members of the laboratory for helpful discussion. Funding VM and AT were supported by Fondazione AIRC per la Ricerca sul Cancro (AIRC). FR was supported by Fondazione Umberto Veronesi. VM and AR were supported by Fondazione GRADE Onlus. This study was funded by the Italian Ministry of Health through Ricerca Finalizzata (N. GR-201602364298, to VF), Bando per la Valorizzazione della Ricerca Istituzionale 2021- fondi 5 per Mille 2020 (to VF) and Fondazione AIACE (to VF). The study was also partially sup-

V. Mularoni et al.

ported by the Italian Ministry of Health-Ricerca Corrente Annual Program 2024. Data-sharing statement All data generated and/or analyzed in this study are included in this article and its Online Supplementary Appendix. The MTAAT sequence has been deposited in the GenBank database with accession number OM642832. Gene expression profile data are available at the Gene Expression Omnibus (GEO) repository (accession number: GSE217426). RNA-sequencing raw data in fastq.gz format are available in the ArrayExpress repository, dataset EMTAB-12462 (https://www.ebi.ac.uk/arrayexpress/experiments/E-MTAB-12462).

References 1. Swerdlow SH, Campo E, Pileri SA, et al. The 2016 revision of the World Health Organization classification of lymphoid neoplasms. Blood. 2016;127(20):2375-2390. 2. Ferreri AJM, Govi S, Pileri SA, Savage KJ. Anaplastic large cell lymphoma, ALK-negative. Crit Rev Oncol Hematol. 2013;85(2):206-215. 3. Parrilla Castellar ER, Jaffe ES, Said JW, et al. ALK-negative anaplastic large cell lymphoma is a genetically heterogeneous disease with widely disparate clinical outcomes. Blood. 2014;124(9):1473-1480. 4. Savage KJ, Harris NL, Vose JM, et al. ALK− anaplastic large-cell lymphoma is clinically and immunophenotypically different from both ALK+ ALCL and peripheral T-cell lymphoma, not otherwise specified: report from the International Peripheral TCell Lymphoma Project. Blood. 2008;111(12):5496-5504. 5. Fiore D, Cappelli LV, Broccoli A, Zinzani PL, Chan WC, Inghirami G. Peripheral T cell lymphomas: from the bench to the clinic. Nat Rev Cancer. 2020;20(6):323-342. 6. Rheinbay E, Nielsen MM, Abascal F, et al for the PCAWG Drivers and Functional Interpretation Working Group, PCAWG Structural Variation Working Group, and PCAWG Consortium. Analyses of non-coding somatic drivers in 2,658 cancer whole genomes. Nature. 2020;578(7793):102-111. 7. van Galen P. Decoding the noncoding cancer genome. Cancer Discov. 2020;10(5):646-647. 8. Kapranov P, Cheng J, Dike S, et al. RNA maps reveal new RNA classes and a possible function for pervasive transcription. Science. 2007;316(5830):1484-1488. 9. Statello L, Guo C-J, Chen L-L, Huarte M. Gene regulation by long non-coding RNAs and its biological functions. Nat Rev Mol Cell Biol. 2021;22(2):96-118. 10. Marchese FP, Raimondi I, Huarte M. The multidimensional mechanisms of long noncoding RNA function. Genome Biol. 2017;18(1):206. 11. Goff LA, Rinn JL. Linking RNA biology to lncRNAs. Genome Res 2015;25(10):1456-1465. 12. Ransohoff JD, Wei Y, Khavari PA. The functions and unique features of long intergenic non-coding RNA. Nat Rev Mol Cell Biol. 2018;19(3):143-157. 13. Huarte M. The emerging role of lncRNAs in cancer. Nat Med. 2015;21(11):1253-1261. 14. Anastasiadou E, Jacob LS, Slack FJ. Non-coding RNA networks

in cancer. Nat Rev Cancer. 2018;18(1):5-18. 15. Iannello A, Ciarrocchi A, Fragliasso V, Vaisitti T. Lift the curtain on long non-coding RNAs in hematological malignancies: pathogenic elements and potential targets. Cancer Lett. 2022;536:215645. 16. Fragliasso V, Verma A, Manzotti G, et al. The novel lncRNA BlackMamba controls the neoplastic phenotype of ALK− anaplastic large cell lymphoma by regulating the DNA helicase HELLS. Leukemia. 2020;34(11):2964-2980. 17. Tameni A, Sauta E, Mularoni V, et al. The DNA-helicase HELLS drives ALK− ALCL proliferation by the transcriptional control of a cytokinesis-related program. Cell Death Dis. 2021;12(1):130. 18. Fragliasso V, Chiodo Y, Ferrari-Amorotti G, et al. Phosphorylation of serine 21 modulates the proliferation inhibitory more than the differentiation inducing effects of C/EBPα in K562 cells. J Cell Biochem. 2012;113(5):1704-1713. 19. Agnelli L, Mereu E, Pellegrino E, et al. Identification of a 3-gene model as a powerful diagnostic tool for the recognition of ALKnegative anaplastic large-cell lymphoma. Blood. 2012;120(6):1274-1281. 20. Grant CE, Bailey TL, Noble WS. FIMO: scanning for occurrences of a given motif. Bioinformatics. 2011;27(7):1017-1018. 21. Sandoval H, Thiagarajan P, Dasgupta SK, et al. Essential role for Nix in autophagic maturation of erythroid cells. Nature. 2008;454(7201):232-235. 22. O’Sullivan TE, Johnson LR, Kang HH, Sun JC. BNIP3- and BNIP3L-mediated mitophagy promotes the generation of natural killer cell memory. Immunity. 2015;43(2):331-342. 23. Bellucci M, Agostini F, Masin M, Tartaglia GG. Predicting protein associations with long noncoding RNAs. Nat Methods. 2011;8(6):444-445. 24. Facucho-Oliveira JM, St. John JC. The relationship between pluripotency and mitochondrial DNA proliferation during early embryo development and embryonic stem cell differentiation. Stem Cell Rev Rep. 2009;5(2):140-158. 25. Youle RJ, Narendra DP. Mechanisms of mitophagy. Nat Rev Mol Cell Biol. 2011;12(1):9-14. 26. Amador C, Bouska A, Wright G, et al. Gene expression signatures for the accurate diagnosis of peripheral T-cell lymphoma entities in the routine clinical practice. J Clin Oncol. 2022;40(36):4261-4275. 27. Palazzo AF, Koonin EV. Functional long non-coding RNAs evolve

Haematologica | 108 December 2023

3345

ARTICLE - Transcriptional role of lncRNA MTAAT in lymphoma from junk transcripts. Cell. 2020;183(5):1151-1161. 28. Schmitt AM, Chang HY. Long noncoding RNAs in cancer pathways. Cancer Cell. 2016;29(4):452-463. 29. Doose G, Haake A, Bernhart SH, et al. MINCR is a MYC-induced lncRNA able to modulate MYC’s transcriptional network in Burkitt lymphoma cells. Proc Natl Acad Sci U S A. 2015;112(38):E5261-5270. 30. Zhao P, Ji M-M, Fang Y, et al. A novel lncRNA TCLlnc1 promotes peripheral T cell lymphoma progression through acting as a modular scaffold of HNRNPD and YBX1 complexes. Cell Death Dis. 2021;12(4):321. 31. Zhao C-C, Jiao Y, Zhang Y-Y, et al. Lnc SMAD5-AS1 as ceRNA inhibit proliferation of diffuse large B cell lymphoma via Wnt/βcatenin pathway by sponging miR-135b-5p to elevate expression of APC. Cell Death Dis. 2019;10(4):252. 32. Cui Y, Xu H, Yang Y, et al. The regulation of miR-320a/XBP1 axis through LINC00963 for endoplasmic reticulum stress and autophagy in diffuse large B-cell lymphoma. Cancer Cell Int. 2021;21(1):305. 33. Li M, Teater MR, Hong JY, et al. Translational activation of ATF4 through mitochondrial anaplerotic metabolic pathways is required for DLBCL growth and survival. Blood Cancer Discov. 2022;3(1):50-65. 34. Bhalla K, Jaber S, Nahid MN, et al. Role of hypoxia in diffuse large B-cell lymphoma: metabolic repression and selective translation of HK2 facilitates development of DLBCL. Sci Rep. 2018;8(1):744. 35. Giacomello M, Pyakurel A, Glytsou C, Scorrano L. The cell biology of mitochondrial membrane dynamics. Nat Rev Mol Cell Biol. 2020;21(4):204-224. 36. Eisner V, Picard M, Hajnóczky G. Mitochondrial dynamics in adaptive and maladaptive cellular stress responses. Nat Cell Biol. 2018;20(7):755-765. 37. Bertolo C, Roa S, Sagardoy A, et al. LITAF , a BCL6 target gene, regulates autophagy in mature B-cell lymphomas. Br J Haematol. 2013;162(5):621-630. 38. Wang Z, Xu F, Yuan N, et al. Rapamycin inhibits pre-B acute lymphoblastic leukemia cells by downregulating DNA and RNA polymerases. Leuk Res. 2014;38(8):940-947. 39. Cheng C, Wang T, Song Z, et al. Induction of autophagy and autophagy-dependent apoptosis in diffuse large B-cell lymphoma by a new antimalarial artemisinin derivative, SM1044. Cancer Med. 2018;7(2):380-396. 40. Nahimana A, Attinger A, Aubry D, et al. The NAD biosynthesis inhibitor APO866 has potent antitumor activity against hematologic malignancies. Blood. 2009;113(14):3276-3286. 41. Torossian A, Broin N, Frentzel J, et al. Blockade of crizotinibinduced BCL2 elevation in ALK-positive anaplastic large cell lymphoma triggers autophagy associated with cell death. Haematologica. 2019;104(7):1428-1439. 42. Sainero-Alcolado L, Liaño-Pons J, Ruiz-Pérez MV, Arsenian-

V. Mularoni et al.

Henriksson M. Targeting mitochondrial metabolism for precision medicine in cancer. Cell Death Differ. 2022;29(7):1304-1317. 43. Missiroli S, Perrone M, Genovese I, Pinton P, Giorgi C. Cancer metabolism and mitochondria: finding novel mechanisms to fight tumours. eBioMedicine 2020;59:102943. 44. Porporato PE, Filigheddu N, Pedro JMB-S, Kroemer G, Galluzzi L. Mitochondrial metabolism and cancer. Cell Res. 2018;28(3):265-280. 45. Zhang L, Zhang W, Li Z, et al. Mitochondria dysfunction in CD8+ T cells as an important contributing factor for cancer development and a potential target for cancer treatment: a review. J Exp Clin Cancer Res. 2022;41(1):227. 46. Chen G, Ray R, Dubik D, et al. The E1B 19K/Bcl-2–binding protein Nip3 is a dimeric mitochondrial protein that activates apoptosis. J Exp Med. 1997;186(12):1975-1983. 47. Ray R, Chen G, Vande Velde C, et al. BNIP3 heterodimerizes with Bcl-2/Bcl-XL and induces cell death independent of a Bcl-2 homology 3 (BH3) domain at both mitochondrial and nonmitochondrial sites. J Biol Chem. 2000;275(2):1439-1448. 48. Hanna RA, Quinsay MN, Orogo AM, Giang K, Rikka S, Gustafsson ÅB. Microtubule-associated protein 1 light chain 3 (LC3) interacts with Bnip3 protein to selectively remove endoplasmic reticulum and mitochondria via autophagy. J Biol Chem. 2012;287(23):19094-19104. 49. Glick D, Zhang W, Beaton M, et al. BNip3 regulates mitochondrial function and lipid metabolism in the liver. Mol Cell Biol. 2012;32(13):2570-2584. 50. Sowter HM, Ferguson M, Pym C, et al. Expression of the cell death genes BNip3 and NIX in ductal carcinoma in situ of the breast; correlation of BNip3 levels with necrosis and grade. J Pathol. 2003;201(4):573-580. 51. Chourasia AH, Tracy K, Frankenberger C, et al. Mitophagy defects arising from BNip3 loss promote mammary tumor progression to metastasis. EMBO Rep. 2015;16(9):1145-1163. 52. Koop EA, van Laar T, van Wichen DF, de Weger RA, van der Wall E, van Diest PJ. Expression of BNIP3 in invasive breast cancer: correlations with the hypoxic response and clinicopathological features. BMC Cancer. 2009;9(1):175. 53. Murai M, Toyota M, Satoh A, et al. Aberrant DNA methylation associated with silencing BNIP3 gene expression in haematopoietic tumours. Br J Cancer. 2005;92(6):1165-1172. 54. Okami J, Simeone DM, Logsdon CD. Silencing of the hypoxiainducible cell death protein BNIP3 in pancreatic cancer. Cancer Res. 2004;64(15):5338-5346. 55. Akada M, Crnogorac-Jurcevic T, Lattimore S, et al. Intrinsic chemoresistance to gemcitabine is associated with decreased expression of BNIP3 in pancreatic cancer. Clin Cancer Res. 2005;11(8):3094-3101. 56. Erkan M, Kleeff J, Esposito I, et al. Loss of BNIP3 expression is a late event in pancreatic cancer contributing to chemoresistance and worsened prognosis. Oncogene. 2005;24(27):4421-4432.

Haematologica | 108 December 2023

3346

ARTICLE - Hodgkin Lymphoma

B-cell receptor reactivity against Rothia mucilaginosa in nodular lymphocyte-predominant Hodgkin lymphoma Lorenz Thurner,1 Natalie Fadle,1 Evi Regitz,1 Sophie Roth,2 Onur Cetin,1 Igor Age Kos,1 Simon Mauro Hess,1 Julia Bein,3 Rainer Maria Bohle,4 Martine Vornanen,5 Christer Sundström,6 Laurence de Leval,7 Enrico Tiacci,8 Peter Borchmann,9 Andreas Engert,9 Viola Poeschel,1 Gerhard Held,10 Eva C. Schwarz,11 Frank Neumann,1 Klaus-Dieter Preuss,1 Markus Hoth,11 Ralf Küppers,12 Karola Lehmann,13 Martin-Leo Hansmann,14,15 Sören L. Becker,2 Moritz Bewarder1# and Sylvia Hartmann3# José Carreras Center for Immuno- and Gene Therapy and Internal Medicine I, Saarland University Medical School, Homburg/Saar, Germany; 2Institute of Medical Microbiology and Hygiene, Saarland University, Homburg, Germany; 3Dr. Senckenberg Institute of Pathology, Goethe University of Frankfurt am Main, Frankfurt am Main, Germany; 4Saarland University Medical School, Institute of Pathology, Homburg/Saar, Germany; 5Department of Pathology, Tampere University Hospital and University of Tampere, Tampere, Finland; 6Department of Immunology, Genetics and Pathology, Uppsala University Hospital, Uppsala, Sweden; 7 Department of Laboratory Medicine and Pathology, Lausanne University Hospital and Lausanne University, Lausanne, Switzerland; 8Institute of Hematology, Ospedale S. Maria della Misericordia, and the Department of Medicine, University of Perugia, Perugia, Italy; 9 University of Cologne, First Department of Internal Medicine, Center for Integrated Oncology Aachen Bonn Cologne Dusseldorf, Cologne, Germany; 10Department of Internal Medicine 1, Westpfalz-Klinikum, Kaiserslautern, Germany; 11Department of Biophysics, Center for Integrative Physiology and Molecular Medicine, Medical Faculty, Saarland University, Homburg, Germany; 12Institute of Cell Biology (Cancer Research), University of Duisburg-Essen, Medical School, Essen, and Deutsches Konsortium für Translationale Krebsforschung (DKTK), Germany; 13Proteome Factory AG, Berlin, Germany; 14Frankfurt Institute for Advanced Studies, Frankfurt am Main, Germany and 15Institute of Pathology and Molecular Pathology, Helios University Hospital, Wuppertal, Germany 1

Correspondence: Lorenz Thurner lorenz.thurner@uks.eu Received: Accepted: Early view:

January 23, 2023. April 24, 2023. May 4, 2023.

https://doi.org/10.3324/haematol.2023.282698 Published under a CC BY license

MB and SH contributed equally as senior authors

Abstract Nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL) is a Hodgkin lymphoma expressing functional B-cell receptors (BCR). Recently, we described a dual stimulation model of IgD+ lymphocyte-predominant cells by Moraxella catarrhalis antigen RpoC and its superantigen MID/hag, associated with extralong CDR3 and HLA-DRB1*04 or HLADRB1*07 haplotype. The aim of the present study was to extend the antigen screening to further bacteria and viruses. The fragment antibody-binding (Fab) regions of seven new and 15 previously reported cases were analyzed. The reactivity of non-Moraxella spp.-reactive Fab regions against lysates of Rothia mucilaginosa was observed in 5/22 (22.7%) cases. Galactofuranosyl transferase (Gltf) and 2,3-butanediol dehydrogenase (Bdh) of R. mucilaginosa were identified by comparative silver- and immuno-staining in two-dimensional gels, with subsequent mass spectrometry and validation by western blots and enzyme-linked immunosorbent assay. Both R. mucilaginosa Gltf and Bdh induced BCR pathway activation and proliferation in vitro. Apoptosis was induced by recombinant Gltf/ETA’-immunotoxin conjugates in DEV cells expressing recombinant R. mucilaginosa-reactive BCR. Reactivity against M. catarrhalis RpoC was confirmed in 3/7 newly expressed BCR (total 10/22 reactive to Moraxella spp.), resulting in 15/22 (68.2%) cases with BCR reactivity against defined bacterial antigens. These findings strengthen the hypothesis of bacterial trigger contributing to subsets of NLPHL.

Introduction Nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL) accounts for 5-16% of Hodgkin lymphomas.1 The disease-defining lymphocyte-predominant (LP) tumor cells

have a late germinal center B-cell phenotype. NLPHL variants with a diffuse growth pattern and histiocyte-rich microenvironment are closely related to T-cell/histiocyterich large B-cell lymphoma (THRLBCL).2,3 When NLPHL was first described, a frequently indolent course was noted and

Haematologica | 108 December 2023

3347

ARTICLE - BCR of NLPHL detect antigens of R. mucilaginosa a potential misguided immune response to infection was speculated.4 In contrast to Hodgkin-Reed-Sternberg cells in classical Hodgkin lymphoma, LP cells are usually negative for Epstein-Barr virus and have a preserved B-cell phenotype, including B-cell receptor (BCR) expression.5 NLPHL is genotypically distinct from classical Hodgkin lymphoma, with frequent BCL6 translocations and recurrent mutations in SGK1, DUSP2 and JUNB.2,6-11 The intraclonal immunoglobulin variable (IgV) region gene diversification,8 strong expression of BCL612 and AID,13,14 and a histological picture that resembles a germinal center suggest an ongoing immune response.15 In the recently published classification of the International Consensus Conference NLPHL was renamed to nodular lymphocyte predominant B-cell lymphoma (NLPBL),16 but in the simultaneously published current 5th version of the World Health Organization classification the term NLPHL was retained.17 Activation of the nuclear factor (NF)-κB pathway is observed in LP cells;2 however, NF-κB activity in LP cells is not caused by mutations in NFKBIA or TNFAIP318 or by latent Epstein-Barr virus5 infection. It may therefore represent a consequence of chronic BCR stimulation. In line with this hypothesis, we demonstrated that dynamic immunological synapses are formed between LP cells and the surrounding rosetting follicular T helper (TFH) cells.19 IgD+ NLPHL had been reported to have a strong male predominance and to present primarily with cervical manifestations.20 We reported the reactivities of mostly IgD+ LP cells against the specific antigens of Moraxella spp. associated with a specific IGVH genotype, a presumably permissive HLA class II haplotype (HLA-DRB01*04 or HLA-DRB01*07), and the presence of light chain-restricted serum antibodies against M. catarrhalis RpoC. Moreover, we proposed a BCR co-stimulation model of IgD+ LP cells by M. catarrhalis RpoC via the fragment antibody-binding (Fab) region and the IgD-binding superantigen MID/hag via the Fc domain.21,22 In another study, similar immunoglobulin genes were found in nonIgD+ NLPHL cases.23 Having identified M. catarrhalis as a trigger for a subset of NLPHL, we hypothesized that there may be further, possibly commensal bacteria eliciting NLPHL. The aim of the present study was, therefore, to extend the screening of the recombinant Fab regions of LP cells to other microbial antigens, particularly those that are part of the resident flora of the oral cavity, upper respiratory tract and gut, such as Rothia spp, Enterococcus spp., and Lactobacillus spp. A second aim of the study was to confirm the reactivity against M. catarrhalis RpoC by analysis of new cases.

Methods Study samples The study was approved by the institutional review boards

L. Thurner et al.

of the University Cancer Center (UCT), the Ethics Committee at the University Hospital Frankfurt (project number: SHN-06-2018), and the local ethics committee of the Ärztekammer des Saarlandes, Saarbrücken (Ha 147/17). Frozen tissue sections of NLPHL specimens were obtained from the Department of Pathology of the UCT Frankfurt, Tampere University, Tampere, Finland, from the Institute of Pathology, CHUV Lausanne, Switzerland, and from the Department of Pathology, Uppsala University Hospital, Sweden. All NLPHL cases were negative for Epstein-Barr virus. Informed consent was obtained from the patients in accordance with the Declaration of Helsinki. The sera of patients with classical Hodgkin lymphoma or NLPHL were obtained from the HD13 (ISRCTN63474366), HD14 (ISRCTN04761296), HD15 (ISRCTN32443041), HD16 (NCT00736320), HD17 (NCT01356680), and HD18 (NCT00515554) trials of the German Hodgkin Study Group.24-27 Sera from patients with THRLBCL were obtained from the RICOVER (NCT00052936), Hi-CHOEP (NCT00129090), UNFOLDER (NCT00278408), FLYER (NCT00278421), and CHOP-R-ESC (NCT00290667) trials, and sera from patients with primary mediastinal B-cell lymphoma were obtained from the UNFOLDER (NCT00278408) trial of the German High-Grade NonHodgkin Lymphoma Study Group.28-31 Fab region screening in bacterial lysates and virome array Laser microdissection of LP cells from cryospecimens and IgV gene multiplex seminested polymerase chain reactions were performed, as described by Küppers et al.32 This also applies to the prokaryotic expression cloning of the recombinant Fab regions in Escherichia coli TG1 (Online Supplementary Methods).33-35 To screen for the potential reactivity of Fab regions against bacterial antigens, heat-inactivated lysates of different bacterial strains or patients’ isolates, including Rothia aeria and Rothia mucilaginosa, were provided by the Institute of Medical Microbiology and Hygiene of Saarland University, Homburg/Saar, Germany (Online Supplementary Methods). To verify the specificity of the reactivities of NLPHL-derived Fab regions against the components of R. mucilaginosa, recombinant Fab regions derived from primary central nervous system lymphoma, diffuse large B-cell lymphoma, mantle cell lymphoma, primary mediastinal Bcell lymphoma and chronic lymphocytic leukemia were screened against these recombinant FLAG-tagged antigens of R. mucilaginosa by enzyme-linked immunosorbent assay.35-37 In addition, the recombinant NLPHL-Fab regions were screened on a custom-made VIROME microarray representing 2,206 viral genes chosen from 158 viruses (ASU, Biodesign Institute, Arizona State University, AZ, USA) (Online Supplementary Methods).

Haematologica | 108 December 2023

3348

ARTICLE - BCR of NLPHL detect antigens of R. mucilaginosa Generation of DEV cells expressing Rothia spp.-reactive B-cell receptors followed by activation, proliferation, and apoptosis assays after stimulation with Rothia antigens and immunotoxins DEV cells expressing Rothia-reactive BCR were generated as previously described (Online Supplementary Methods).20,22 DEV cells either expressing a BCR with reactivity against Rothia mucilaginosa Gltf or Moraxella catarrhalis RpoC, and of IgD or IgG class, were subjected to western blot for activation of the BCR pathway, proliferation and apoptosis assays after stimulation with the respective bacterial antigens or incubation with bacterial antigens conjugated with an immunotoxin (Online Supplementary Methods). HLA genotyping HLA-DRB1 and HLA-DQ typing was performed for all patients via sequencing (Labor Thiele, Kaiserslautern, Germany).

L. Thurner et al.

Statistical analysis Normality was evaluated using the Shapiro-Wilk and Kolmogorov-Smirnov tests. Statistical significance was calculated using unpaired, two-tailed t tests for comparison of parametric distributions, without adaptation for multiple comparisons (Prism 9, GraphPad).

Results Patient, HLA-DRB1, and IgV gene characteristics The previous expression-cloned cohort of the Fab regions of 15 NLPHL cases21,22 was extended by Ig heavy and light chain gene amplification from microdissected LP cells with an additional seven NLPHL cases (Table 1; IgV gene characteristics are presented in Table 2). The median age of the patients with successfully amplified

Table 1. Characteristics of the patients with nodular lymphocyte-predominant Hodgkin lymphoma included in this study

Case

Sex

Age yrs

Primary diagnosis/relapse

Localization

LP cells IgD+

Pattern

Reactivity against M. cat. RpoC or M. osl. succinate-CoA ligase sub α

Relapse

Axillary

Relapse

Supraclavicular

Yes*

2nd relapse

Cervical

Yes

Yes*

First manifestation

Cervical

First manifestation

Retroperitoneal

Relapse

Cervical

Yes

Yes*

2nd relapse

Axillary

A+D

Primary progression

Abdominal

D+E

Relapse

Inguinal

Yes

Yes*

2nd relapse

Inguinal

Yes

Yes*

First manifestation

Cervical

Yes

Yes*

First manifestation

Cervical

Yes*

First manifestation

Cervical

Yes

Yes*

First manifestation

Parotid

Yes

E, B + D

Yes*

First manifestation

Axillary

Yes

Yes*

Relapse

Axillary

Yes*

First manifestation

Abdominal

Yes

Yes*

First manifestation

Cervical

Yes

Yes*

First manifestation

Inguinal

Yes*

Relapse

Cervical

Yes

Relapse

Infraclavicular

Yes*

First manifestation

Inguinal

Reactivity against R. mucilaginosa

IgD+ nodular lymphocyte-predominant (LP) Hodgkin lymphoma; cases 7 and 8: (a) nodular lymphocyte-predominant Hodgkin lymphoma, (b) histological transformation into diffuse large B-cell lymphoma in the same lymph node. M: male; F: female; yrs: years; M. cat.: Moraxella catarrhalis; M. osl.: Moraxella osloensis; R. mucilaginosa: Rothia mucilaginosa; NA: not available. Additional information on HLA subtype and stage is available in Online Supplementary Table S1. *

Haematologica | 108 December 2023

3349

ARTICLE - BCR of NLPHL detect antigens of R. mucilaginosa rearranged IgV genes was 28 (9-65) years, with a male predominance (18/22, 82%). IgD+ LP cells were identified in 11 patients with a median age of 21 (12–42) years (male:female 9:2). B-cell receptors derived from lymphocyte-predominant cells react with lysates of M. catarrhalis and R. mucilaginosa Three of seven of the new recombinant Fab regions bound to lysates of Moraxella spp. In combination with seven Moraxella-reactive Fab regions from the first 15 cases, 10/22 cases were Moraxella spp.-reactive, and nine of them were derived from IgD+ positive cases (Figure 1A). In contrast, the pooled NLPHL Fab regions reacted in dotblots against lysates of R. mucilaginosa. Following individual testing of these Fab regions, 5/22 NLPHL-derived BCR (#2, #12, #15, #16 and #19, with 1/5 IgD+) reacted against R. mucilaginosa lysates (Figure 1A). None of these five Fab regions

L. Thurner et al.

cross-reacted with M. catarrhalis lysates. In western blots of R. mucilaginosa lysates using the NLPHL-derived Fab regions as primary antibodies, two bands of approximately 35-40 kDa and 75 kDa were detected (Figure 1B; Online Supplementary Figures S3 and S4). Following two-dimensional gel-based isolation of single spots (Figure 1C), mass spectrometry was used to identify glycerol-3-phopshate dehydrogenase/oxidase (WP_44150604), Gtf)/Gltf (WP_012903226), type I glyceraldehyde-3-phosphate dehydrogenase (WP_012903599), 1deoxy-D-xylulose-5-phosphate synthase (WP_044150996), 2-oxoglutarate dehydrogenase, E2 component, dihydrolipoamide succinyltransferase (WP_044150317), 2,3-butanediol dehydrogenase (WP_044150456), and iron-sulfur cluster-binding protein (WP_012902908) of R. mucilaginosa as probable candidate antigens. Specific binding of the R. mucilaginosa-reactive recombinant BCR in the R. mucilaginosa lysate against recombinant Gltf

Table 2. IgV heavy chain gene analysis of lymphocyte-predominant cells and identified antigenic targets. Homology DH gene %

CDR3 JH/JL length gene AA

Case

VH gene

Junction

Identified antigen

VH1-3-01

91.3

D3-10*01

JH4*02

CAREVRPPRIIMIWGVGLLDFW

Human RPS27a (autoantigen)

VH1-46*01

87.5a

D3-16*02

JH2*01

YYCARDEGDIRRYFDLW

R. mucilaginosa Gltf

VH3-11*01

95.1

D3-3*01

JH6*03

CARVAGAAGRNYNYWSGYWEDYYFMDVW

M. catarrhalis RpoC

VH4-31*01

99.2

D3-22

JH3*02

CARGPPPYDSSGYYSHGLDIW

VH4-39*01

100

D6-19*01

JH6*04

CASMGAVAGMMFGMDVW

Human PC3 (autoantigen)

VH3-07*01

95.5

D3-3*01

JH6*04

CAREVLRWGGSYDFWSNYYEDYFALDVW

M. catarrhalis RpoC

VH1-69

86.4

D6-19*01

JH5*02

CARDYSRGVCGPRYGMDVW

VH3-30

89.9

D2-15*01

JH6*02

CARKGGDPVLALFVPNFAMDVW

VH3-48*03

82.0

D3-3*01

JH6*02

CAKSVLTAKSGKSYKFWNNYHEDYHYYLMDVW

M. catarrhalis RpoC

VH4-59*01

86.4b

D3-3*01

JH6*03

CATVDPTVVEGRVKYYDFWSGYYGTDQRYYYMDVW

M. catarrhalis RpoC

VH3-11*01

95.5

D3-3*01

JH6*02

CARLLTSEGSRKYYDFWSNYWEGYQYYTMDVW

M. osloensis SUCLG1

VH4-34

100

D3-10*01

JH5*02

CARGPYLWFGERGWFDPW

R. mucilaginosa Bdh

VH3-11*01

83.0

D3-3*01

JH6*03

CARLCAAGGRSYDFWSGYYENYFYMEVW

M. catarrhalis RpoC

VH3-11*01

87.6

D3-3*01

JH6*02

CARLIEAGGVGKHYDFWSGYYTVDYYYGMDVW

M. catarrhalis RpoC

VH3-30*04

89.8

D3-3*01

JH6*02

CARTTWVGVVGRIKYYDFWSGYHGTGMEYYTMDVW

R. mucilaginosa Gltf

VH23*04

90.3

DH6-19*01 JH2*02

CAKLPLRPQWLYRGYFDLW

R. mucilaginosa Gltf

VH3-11*01

87.5

DH6-19*01 JH4*02

CARDIHHQWLNPVINPHWVDPVDYFDYW

M. catarrhalis RpoC

VH1-69*13

99.6

DH5-12*01 JH3*02

CASPGRYSGYGYDAFDIW

M. catarrhalis RpoC

VH3-7*01

94.4

DH3-3*01

JH6*03

CATIVLDSIKGSVRYYDFWSGHHGLSYYYYYMDVW

R. mucilaginosa Gltf

VH2-5*02

96.6

DH3-9*01

JH3*02

CAHTHEDILTGSDALDIW

VH1-2*02

88.9

DH3-22*01 JH4*02

CARDRIEDFFDSSGYVYW

M. catarrhalis RpoC

VH1-18*01

97.6

DH2-15*01 JH3*02

CARVPWFGLCSGGSCYEDAFDIW

VH5-10-1*01

99.6

DH3-10*01 JH5*02

CARSLYYGSGPRFDPW

Additional information on light chains is available in Online Supplementary Table S2. All cases had functional IgV genes. For case 15 two functional light chains were amplified. aAnd insertion in FR3 (AGAAAT). bWith deletion of 3 nt in CDR2. AA: amine acids; R. mucilaginosa: Rothia mucilaginosa; M. catarrhalis: Moraxella catarrhalis; M. osloensis: Moraxella osloensis. Haematologica | 108 December 2023

3350

ARTICLE - BCR of NLPHL detect antigens of R. mucilaginosa as the antigen or band/spot of 75 kDa and against Bdh as the antigen or band/spot of 39 kDa was verified by enzymelinked immunosorbent assay (Figure 1D). In contrast to NLPHL, none of the recombinant BCR derived from primary central nervous system lymphoma, diffuse large B-cell lymphoma, mantle cell lymphoma, or chronic lymphocytic leukemia cases reacted with the R. mucilaginosa lysate (Online Supplementary Figure S3). Furthermore, no Fab reactivity was detected against the total R. mucilaginosa lysate or against any of the other investigated bacteria. Virome screening of the pooled non-Moraxella spp.-reactive

L. Thurner et al.

Fab regions resulted in 13 candidate antigens with normalized values ≥2: CbCD00959563 of coxsackievirus B1, MsCD00595052 of measles, strain Ichinose WT, CagA_884_2.1, and CagA_FL_2.1 of Helicobacter pylori 26695, HrCD00959476 of human rhinovirus A1, HsCD00959820 and HsCD00959751 of H1N1 subtype, CbCD00594880 of coxsackievirus B4 (strain E2), CaCD00959834 of coxsackievirus A9, HrCD00956402 of human rubulavirus 2, HcCD00959896 of human coronavirus NL63, and HsCD00959755 of H3N2 subtype (Online Supplementary Table S3; Online Supplementary Figure S1). After expression cloning of the obtained

Continued on following page. Haematologica | 108 December 2023

3351

ARTICLE - BCR of NLPHL detect antigens of R. mucilaginosa

L. Thurner et al.

Figure 1. Reactivity of recombinant IgD+ lymphocyte-predominant cell-derived Fab regions with bacterial lysates. (A) Representative immuno-dot blots of bacterial lysates, showing reactivity against lysate of M. catarrhalis from previously reported Fab regions, and from Fab regions of new cases as validation. In addition there are nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL)-derived recombinant Fab regions of five cases (#2, #12, #15, #16, #19) bound to lysates of R. mucilaginosa. (B) Representative western blot of polyacrylamide gel electrophoresis of lysate of R. mucilaginosa using pooled and individual recombinant Fab regions of NLPHL, which had shown reactivity against R. mucilaginosa as primary antibody. Two bands of approximately 3540 kDa and 75 kDa were detected. (C) Silver staining of lysate of R. mucilaginosa for 45 min of blotting gel (Proteome factory AG, Berlin, Germany), and immunostain of antigens as dots in a two-dimensional gel blot using the recombinant pooled NLPHL Fab region as primary antibody. (D) Enzyme-linked immunosorbent assay with recombinant C-terminally FLAG-tagged potential antigens of R. mucilaginosa identified by mass spectrometry. The respective Fab region at a concentration of 10 μg/mL (positive control = rec. the Fab region of patient #11) showed a reactivity against galactofuranosyl transferase and 2,3-butanediol dehydrogenase of R. mucilaginosa.

plasmids, non-reactivity against these candidate antigens was confirmed on dot-plots with the pooled NLPHL Fab regions (data not shown). Serum antibodies against galactofuranosyl transferase and 2,3-butanediol dehydrogenase of R. mucilaginosa in patients with nodular lymphocyte-predominant Hodgkin lymphoma Antibodies with relevant titers against R. mucilaginosa Gltf were found in 2/98 (2%) and against R. mucilaginosa Bdh

in 3/98 (3%) of patients with NLPHL enrolled in the clinical trials of the German Hodgkin Study Group (Figure 2A), which is representative of the general population of NLPHL patients. All antibodies against R. mucilaginosa Gltf and Bdh were of IgG class and predominantly of the IgG1 subclass, and they were kappa light-chain restricted (Figure 2C, D). The titers ranged from 1:400 to 1:800 (Figure 2B). No patient with classical Hodgkin lymphoma (0/98) (Figure 2A), THRLBCL (0/50), or primary mediastinal B-cell lymphoma (0/80) (Online Supplementary Figure S4) pres-

Haematologica | 108 December 2023

3352

ARTICLE - BCR of NLPHL detect antigens of R. mucilaginosa

L. Thurner et al.

Figure 2. Serological responses against Gltf and Bdh of R. mucilaginosa. (A) Sera of patients with nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL) and classical Hodgkin lymphoma (cHL) (diluted 1:100) were tested for antibodies against galactofuranosyl transferase (Gltf) and 2,3-butanediol dehydrogenase (Bdh) of R. mucilaginosa. (B) Titers of anti-Gltf antibodies and anti-2,3-Bdh antibodies in seropositive patients with NLPHL (left) or cHL (right). The curves represent the optical density (OD) at 490 nm of different serum dilutions. Patients with NLPHL had titers between 1:200 and 1:800. (C) Analysis of Ig classes, and IgG subclasses of serum-antibodies against Gltf and 2,3-Bdh of R. mucilaginosa in patients with NLPHL. The antiGltf antibodies and anti-2,3-Bdh antibodies were exclusively of the IgG class, mostly IgG1. (D) Light chains of serum-antibodies against Gltf and 2,3-Bdh of R. mucilaginosa in patients with NLPHL.

ented serum antibodies against Gltf and Bdh of R. mucilaginosa. In summary, antibodies against Gltf and Bdh of R. mucilaginosa were found in NLPHL patients but not in patients with related lymphomas. R. mucilaginosa galactofuranosyl transferase induces stimulation of lymphocyte-predominant cells Because we hypothesized that specific stimulation of the LP-cell BCR by R. mucilaginosa Gltf contributes to lymphomagenesis in a subset of NLPHL, we functionally investigated patient-derived Gltf-specific BCR. For this purpose, cells of the only available NLPHL cell line (DEV) were transfected to express, in a doxycycline-dependent manner, recombinant BCR derived from patients #2 and #3 with constant regions Cγ or Cδ. Gltf stimulation resulted in strong activation of the BCR signaling pathway in DEV cells that stably expressed Gltf-

reactive BCR, as determined by phosphorylation of key BCR signaling factors (pTyr525/526 spleen tyrosine kinase [SYK], pTyr96 B-cell linker [BLNK], pTyr759 phosphoinositide-specific phospholipase C [PLC]γ2, and pTyr223 Bruton tyrosine kinase [BTK]) (Figure 3A). Activation of the BCR pathway was associated with a significant increase in MYC expression (Figure 3A). Stimulation with recombinant R. mucilaginosa Gltf resulted in significantly increased proliferation of DEV cells that stably expressed Gltf-reactive BCR (Figure 3B). Furthermore, the recombinant immuno-conjugate of R. mucilaginosa Gltf conjugated to a truncated form of the exotoxin A of Pseudomonas aeruginosa (Gltf/ETA’) inhibited proliferation of DEV cells stably expressing Gltfspecific BCR but had no effect on non-transfected DEV cells or on DEV cells that expressed M. catarrhalis RpoCreactive BCR (Figure 3B). An increase in the number of

Haematologica | 108 December 2023

3353

ARTICLE - BCR of NLPHL detect antigens of R. mucilaginosa

L. Thurner et al.

Continued on following page. Haematologica | 108 December 2023

3354

ARTICLE - BCR of NLPHL detect antigens of R. mucilaginosa

L. Thurner et al.

Figure 3. Stimulation of different types of lymphocyte-predominant cells by Gltf of R. mucilaginosa. (A) Activation of the B-cell receptor (BCR) signaling pathway. Western blot analysis of components of the BCR signaling pathway shows activation by R. mucilaginosa galactofuranosyl transferase (Gltf) and by M. catarrhalis RpoC in DEV cells transfected and expressing R. mucilaginosa-Gltf-reactive recombinant BCR of case #2 with stronger bands of the activated isoforms pTyr525/526 SYK, pTyr96 BLNK, pTyr759 PLCγ2 and pTyr223 BTK and of MYC after incubation with R. mucilaginosa Gltf. Similarly, reaction to the BCR pathway after incubation with M. catarrhalis RpoC to DEV cells with RpoC-reactive BCR of case #3. (B) Tetrazolium proliferation assay with the transfected DEV cells expressing recombinant nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL)-derived BCR with reactivity against Gltf of R. mucilaginosa or RpoC of M. catarrhalis. Stimulation with Gltf of R. mucilaginosa increased proliferation in DEV cells transfected to express Gltf-reactive BCR but not in DEV cells expressing RpoC-reactive BCR. Incubation by ETA’ toxin-conjugated to Gltf of R. mucilaginosa or to RpoC/ETA’ of M. catarrhalis on DEV cells expressing either recombinant NLPHL-derived BCR (#2) with reactivity against R. mucilaginosa Gltf or against M. catarrhalis RpoC-reactive (#3) led to a decrease in proliferation. *P<0.05; **P≤0.01, ***P≤0.001, ****P≤0.0001. (C) Apoptosis induced by ETA’ toxin-conjugated to Gltf of R. mucilaginosa or RpoC of M. catarrhalis is dependent on doxycycline-induced expression of Gltf- or RpoC-reactive BCR on lymphocyte-predominant cells. Characterization of DEV cells transfected to express doxycycline-inducible IgD+ recombinant BCR with (from case #2) reactivity against Gltf of R. mucilaginosa or RpoC of M. catarrhalis (from case #3) by annexin-V/FITC and propidium iodide staining after 24 h culture in the presence of RpoC/ETA’, Gltf/ETA’ or staurosporin. Dox: doxycycline.

apoptotic cells was detected via annexin V/propidium iodide staining after incubation with R. mucilaginosa Gltf/ETA’ in the DEV cells expressing a recombinant BCR with the respective reactivity, but not in DEV cells with different reactivity of recombinant BCR - i.e. against M. catarrhalis RpoC - or without doxycycline induction of recombinant BCR at all (Figure 3C).

Discussion The present study strengthens the evidence for the contribution of chronic antigenic simulation by common bacteria to subsets of NLPHL.38 Two specific antigens of R. mucilaginosa were identified as targets for BCR in 5/22 NLPHL cases. In addition, the reactivity against M. catarrhalis RpoC antigens was verified for recombinant BCR from 3/7 newly analyzed NLPHL cases, resulting in a total reactivity of 10/22 cases against Moraxella spp. In addition to R. mucilaginosa, the recombinant lymphoma BCR of most cases (15/22; 68.2%) were reactive against specific bacterial antigens. As seen before for M. catarrhalis RpoC, light chain-restricted antibodies against R. mucilaginosa Gltf and 2,3Bdh, predominantly of the IgG1 subclass, were observed in NLPHL sera. While serum antibodies for M. catarrhalis RpoC were relatively frequently identified in NLPHL patients, antibodies against R. mucilaginosa were less frequently observed. Both light chain-restricted antibodies against Rpoc and R. mucilaginosa are probably the result of a very strong immune reaction against the specific antigen. However, not all patients with such lymphoma BCR reactivity have high titers of serum antibodies; this discrepancy was also seen for lymphoma BCR reactivity against Ars2 in ABC-type diffuse large B-cell lymphoma and for LRPAP1 in mantle cell lymphoma.39-41 Similar to previous observations of M. catarrhalis RpoC,22 no antibodies against R. mucilaginosa Gltf or Bdh were detected in the sera of patients with THRLBCL, the NLPHL-related aggressive B-cell lymphoma subtype,42,43 suggesting a dis-

tinct mechanism in these lymphomas. R. mucilaginosa is a commensal, Gram-positive, coagulase-negative coccus and a part of the resident flora of the oral cavity. It typically resides in the anatomical drainage area of cervical lymph nodes.44 However, in immunocompromised patients, R. mucilaginosa can be pathogenic, and bacteremia can occur.38,39,41-44 Regarding the previously described association of ultralong CDR3 with reactivity against M. catarrhalis RpoC, the extended cohort of NLPHL Fab regions revealed that reactivity against M. catarrhalis RpoC was not restricted to BCR with ultralong CDR3, and some cases with ultralong CDR3 were reactive to antigens of R. mucilaginosa. Therefore, ultralong CDR3 and skewed IGHD-IGHJ usage are likely not specific to reactivity against a specific antigen but instead represent a relatively frequent structural combination, suggesting that LP cells frequently stem from a B-cell population with ultralong CDR3 that regularly exists in healthy individuals. Recently, in a VH3-23 model of natural antibodies, Wang et al. described skewed usage of IGHD2-IGHJ6 and IGHD3-IGHJ6 rearrangements connected to the presence of ultralong CDR3 loops that stabilized the long CDR3 loop rather than contributing to the paratop.20,45 The clustering of HLA-DRB1:04 and DRB1:07 haplotypes in Moraxella-reactive NLPHL22 suggests cognate T-cell involvement in the immune reaction that finally leads to lymphoma development. Further studies demonstrated the formation of immunological synapses between TFH and LP cells, which further supports this idea.19 No HLA restriction was observed in NLPHL cases with R. mucilaginosa-directed BCR, with the limitation of a low number of cases. In addition to the lower pathogenicity compared with M. catarrhalis, R. mucilaginosa is not known to express a superantigen like MID/hag of M. catarrhalis.46-48 Considering its potential as a chronic antigenic trigger, it should be noted that R. mucilaginosa is known to form biofilms. In vitro, both R. mucilaginosa Gltf and Bdh induced growth by activating the BCR pathway. A main limitation of the study is that our selection of viral

Haematologica | 108 December 2023

3355

ARTICLE - BCR of NLPHL detect antigens of R. mucilaginosa and bacterial pathogens represents only a fraction of the environmental pathogens; therefore, we likely did not consider all the potential variables. In the current study, we selected R. mucilaginosa and R. aeria as they occur in a pattern that is comparable to that of Moraxella spp. in oropharyngeal samples and other samples of the upper respiratory tract. This necessitates further and extended systematic screening for microbial antigens, including comprehensive microbiome banks. In addition to NLPHL, certain lymphomas, such as duodenal or pediatric follicular lymphoma, different types of marginal zone lymphomas, and more aggressive lymphomas should be studied in this context. The findings lead to the question of why and under which circumstances immune responses against common bacteria - even a component of the oral resident flora trigger lymphoma.46-48 From the present study it is difficult to conclude how far bacterial antigens actively support growth of NLPHL once the lymphoma has developed and how far this may be targetable by antibiotic treatment. As eradication of the infectious trigger has been proven successful and has even become standard therapy for some advanced indolent lymphoma entities,49-53 this might serve as a therapeutic model to be investigated in future clinical trials. However, heterogeneity of LP cells, their active somatic hypermutation of BCR genes, and also transformation into high-grade lymphomas would represent challenges for an antibiotic-based therapeutic approach. On the other hand, the rarity of NLPHL poses a serious challenge to the feasibility of a meaningful clinical trial. Possible proposals for clinical trials might initially investigate the relapsed, advanced setting with an exclusive low-grade manifestation and detection of corresponding antibodies against M. catarrhalis or R. mucilaginosa. Given recent reports from retrospective analyses regarding less intensive treatments and active surveillance

L. Thurner et al.

strategies for NLPHL,54 the results of the present study should stimulate discussions on clinical trials investigating antibiotic therapy or vaccination strategies against bacteria-associated NLPHL subgroups. Disclosures LT has received travel grants from Abbvie, Janssen and EUSA-Pharm, and consultancy fees from Takeda, AstraZeneca, Merck and EUSA-pharm (less than $10,000 each). The other authors declare no competing interests. Contributions LT, SLB, and SH designed the study, performed experiments, analyzed data and wrote the manuscript. NF, ER, KL, FN, and NS performed experiments and analyzed data. SR, JB, YJK, RMB, AN, MV, CS, LdL, PB, AE, VP, and GH supplied essential material and interpreted data. RK, ET, KDP, SMH, OC, IAK, ES, MH, MLH, and MB designed the study, interpreted data, and revised the manuscript. Acknowledgments We thank Claudia Schormann and the DSHNHL team and Bernhard Thurner for constant support, advice, and critical reading of the manuscript. Funding MP, LT, K-DP, and FN have been supported by the Wilhelm Sander Stiftung. LT was supported by a University of Saarland fellowship and by the NanoBioMed young investigator program of Saarland University. SH is supported by the Deutsche Forschungsgemeinschaft (grant HA 6145/7-1). Data-sharing statement All relevant data are available in the Article and Online Supplementary Information or from the corresponding author upon reasonable request.

References 1. Saarinen S, Pukkala E, Vahteristo P, Mäkinen MJ, Franssila K, Aaltonen LA. High familial risk in nodular lymphocytepredominant Hodgkin lymphoma. J Clin Oncol. 2013;31(7):938-943. 2. Brune V, Tiacci E, Pfeil I, et al. Origin and pathogenesis of nodular lymphocyte-predominant Hodgkin lymphoma as revealed by global gene expression analysis. J Exp Med. 2008;205(10):2251-2268. 3. Hartmann S, Doring C, Jakobus C, et al. Nodular lymphocyte predominant Hodgkin lymphoma and T cell/histiocyte rich large B cell lymphoma--endpoints of a spectrum of one disease? PLoS One. 2013;8(11):e78812. 4. Jackson H Jr, Parker F. Hodgkin’s disease. II. Pathology. N Engl J Med. 1944;231:35-44. 5. Anagnostopoulos I, Hansmann ML, Franssila K, et al. European Task Force on Lymphoma project on lymphocyte predominance Hodgkin disease: histologic and immunohistologic analysis of submitted cases reveals 2 types of Hodgkin disease with a

nodular growth pattern and abundant lymphocytes. Blood. 2000;96(5):1889-1899. 6. Hartmann S, Schuhmacher B, Rausch T, et al. Highly recurrent mutations of SGK1, DUSP2 and JUNB in nodular lymphocyte predominant Hodgkin lymphoma. Leukemia. 2015;40(9):2278-2285. 7. Kuppers R, Rajewsky K, Braeuninger A, Hansmann ML. L&H cells in lymphocyte-predominant Hodgkin’s disease. N Engl J Med. 1998;338(11):763-764. 8. Marafioti T, Hummel M, Anagnostopoulos I, et al. Origin of nodular lymphocyte-predominant Hodgkin’s disease from a clonal expansion of highly mutated germinal-center B cells. N Engl J Med. 1997;337(7):453-458. 9. Wlodarska I, Nooyen P, Maes B, et al. Frequent occurrence of BCL6 rearrangements in nodular lymphocyte predominance Hodgkin lymphoma but not in classical Hodgkin lymphoma. Blood. 2003;101(2):706-710. 10. Mottok A, Renné C, Willenbrock K, Hansmann ML, Bräuninger A. Somatic hypermutation of SOCS1 in lymphocyte-predominant

Haematologica | 108 December 2023

3356

ARTICLE - BCR of NLPHL detect antigens of R. mucilaginosa Hodgkin lymphoma is accompanied by high JAK2 expression and activation of STAT6. Blood. 2007;110(9):3387-3390. 11. Braeuninger A, Kuppers R, Strickler JG, Wacker HH, Rajewsky K, Hansmann ML. Hodgkin and Reed-Sternberg cells in lymphocyte predominant Hodgkin disease represent clonal populations of germinal center-derived tumor B cells. Proc Natl Acad Sci U S A. 1997;94(17):9337-9342. 12. Falini B, Bigerna B, Pasqualucci L, et al. Distinctive expression pattern of the BCL-6 protein in nodular lymphocyte predominance Hodgkin’s disease. Blood. 1996;87(2):465-471. 13. Greiner A, Tobollik S, Buettner M, et al. Differential expression of activation-induced cytidine deaminase (AID) in nodular lymphocyte-predominant and classical Hodgkin lymphoma. J Pathol. 2005;205(5):541-547. 14. Muramatsu M, Sankaranand VS, Anant S, et al. Specific expression of activation-induced cytidine deaminase (AID), a novel member of the RNA-editing deaminase family in germinal center B cells. J Biol Chem. 1999;274(26):18470-18476. 15. Fan Z, Natkunam Y, Bair E, Tibshirani R, Warnke RA. Characterization of variant patterns of nodular lymphocyte predominant Hodgkin lymphoma with immunohistologic and clinical correlation. Am J Surg Pathol. 2003;27(10):1346-1356. 16. Campo E, Jaffe ES, Cook JR, et al. The International Consensus Classification of Mature Lymphoid Neoplasms: a report from the Clinical Advisory Committee. Blood. 2022;140(11):1229-1253. 17. Alaggio R, Amador C, Anagnostopoulos I, et al. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: Lymphoid Neoplasms. Leukemia. 2022;36(7):1720-1748. 18. Schumacher MA, Schmitz R, Brune V, et al. Mutations in the genes coding for the NF-kappaB regulating factors IkappaBalpha and A20 are uncommon in nodular lymphocytepredominant Hodgkin’s lymphoma. Haematologica. 2010;95(1):153-157. 19. Bein J, Thurner L, Hansmann ML, Hartmann S. Lymphocyte predominant cells of nodular lymphocyte predominant Hodgkin lymphoma interact with rosetting T cells in an immunological synapse. Am J Hematol. 2020;95(12):1495-1502. 20. Prakash S, Fountaine T, Raffeld M, Jaffe ES, Pittaluga S. IgD positive L&H cells identify a unique subset of nodular lymphocyte predominant Hodgkin lymphoma. Am J Surg Pathol. 2006;30(5):585-592. 21. Thurner L, Kemele M, Regitz E, et al. Moraxella catarrhalis-a bacterial involvement in the origin of nodular lymphocyte predominant Hodgkin lymphoma of IGD type? Haematologica. 2015;100(s1):106-107. 22. Thurner L, Hartmann S, Fadle N, et al. Lymphocyte predominant cells detect Moraxella catarrhalis-derived antigens in nodular lymphocyte-predominant Hodgkin lymphoma. Nat Commun. 2020;11(1):2465. 23. Paschold L, Willscher E, Bein J, et al. Evolutionary clonal trajectories in nodular lymphocyte-predominant Hodgkin lymphoma with high risk of transformation. Haematologica. 2021;106(10):2654-2666. 24. Behringer K, Goergen H, Hitz F, et al. Omission of dacarbazine or bleomycin, or both, from the ABVD regimen in treatment of early-stage favourable Hodgkin’s lymphoma (GHSG HD13): an open-label, randomised, non-inferiority trial. Lancet. 2015;385(9976):1418-1427. 25. Engert A, Borchmann P, Pluetschow A, et al. Dose-escalation with BEACOPP escalated is superior to ABVD in the combinedmodality treatment of early unfavorable Hodgkin lymphoma: final analysis of the German Hodgkin Study Group (GHSG) HD14 trial. Blood. 2010;116(21):765.

L. Thurner et al.

26. Borchmann P, Goergen H, Kobe C, et al. EBEACOPP with or without rituximab in interim-PET-positive advanced-stage Hodgkin lymphoma: updated results of the international, randomized phase 3 GHSG HD18 trial. Hematol Oncol. 2017;35(S2):65. 27. Borchmann P, Goergen H, Kobe C, et al. PET-guided treatment in patients with advanced-stage Hodgkin’s lymphoma (HD18): final results of an open-label, international, randomised phase 3 trial by the German Hodgkin Study Group. Lancet. 2017;390(10114):2790-2802. 28. Pfreundschuh M, Schubert J, Ziepert M, et al. Six versus eight cycles of bi-weekly CHOP-14 with or without rituximab in elderly patients with aggressive CD20+ B-cell lymphomas: a randomised controlled trial (RICOVER-60). Lancet Oncol. 2008;9(2):105-116. 29. Schmitz N, Nickelsen M, Ziepert M, et al. Conventional chemoimmunotherapy (R-CHOEP-14) or high-dose therapy (RMega-CHOEP) for young, high-risk patients with aggressive B-cell lymphoma: final results of the randomized Mega-CHOEP trial of the German High-Grade Non-Hodgkin Lymphoma Study Group (DSHNHL). J Clin Oncol. 2011;29(15_suppl):8002. 30. Pfreundschuh M, Murawski N, Ziepert M, et al. Radiotherapy (RT) to bulky (B) and extralymphatic (E) disease in combination with 6xR-CHOP-14 or R-CHOP-21 in young good-prognosis DLBCL patients: results of the 2x2 randomized UNFOLDER trial of the DSHNHL/GLA. J Clin Oncol. 2018;36(15_suppl):7574. 31. Poeschel V, Held G, Ziepert M, et al. Excellent outcome of young patients (18-60 years) with favourable-prognosis diffuse large B-cell lymphoma (DLBCL) treated with 4 cycles CHOP plus 6 applications of rituximab: results of the 592 patients of the Flyer trial of the DSHNHL/GLA. Blood. 2018;132(Suppl 1):781. 32. Kuppers R, Schneider M, Hansmann ML. Laser-based microdissection of single cells from tissue sections and PCR analysis of rearranged immunoglobulin genes from isolated normal and malignant human B cells. Methods Mol Biol. 2013;971:49-63. 33. Voswinkel J, Kerkdijk AJ, Mueller A, Assmann G, Pfreundschuh M, Held G. A novel system to test for specificity of B cell receptors from tissue of Wegener’s granulomatosis patients. Clin Exp Rheumatol. 2008;26(3 Suppl 49):S90-S96. 34. de Haard HJ, van NN, Reurs A, et al. A large non-immunized human Fab fragment phage library that permits rapid isolation and kinetic analysis of high affinity antibodies. J Biol Chem. 1999;274(26):18218-18230. 35. Zwick C, Fadle N, Regitz E, et al. Autoantigenic targets of B-cell receptors derived from chronic lymphocytic leukemias bind to and induce proliferation of leukemic cells. Blood. 2013;121(23):4708-4717. 36. Thurner L, Nimmesgern A, Neumann F, et al. LRPAP1 is a frequent proliferation-inducing antigen of BCRs of mantle cell lymphomas and can be used for specific therapeutic targeting. Leukemia. 2019;33(1):148-158. 37. Thurner L, Preuss K-D, Bewarder M, et al. Hyper N-glycosylated SAMD14 and neurabin-I as driver CNS autoantigens of PCNSL. Blood. 2018;132(26):2744-2753. 38. Kuppers R. Mechanisms of B-cell lymphoma pathogenesis. Nat Rev Cancer. 2005;5(4):251-262. 39. Thurner L, Hartmann S, Bewarder M, et al. Identification of the atypically modified autoantigen Ars2 as the target of B-cell receptors from activated B-cell-type diffuse large B-cell lymphoma. Haematologica. 2021;106(8):2224-2232. 40. Thurner L, Hartmann S, Fadle N, et al. LRPAP1 is a frequent proliferation-inducing antigen of BCRs of mantle cell lymphomas and can be used for specific therapeutic targeting. Leukemia. 2019;33(1):148-158.

Haematologica | 108 December 2023

3357

ARTICLE - BCR of NLPHL detect antigens of R. mucilaginosa 41. Thurner L, Fadle N, Bittenbring JT, et al. LRPAP1 autoantibodies in mantle cell lymphoma are associated with superior outcome. Blood. 2021;137(23):3251-3258. 42. Hartmann S, Döring C, Jakobus C, et al. Nodular lymphocyte predominant Hodgkin lymphoma and T cell/histiocyte rich large B cell lymphoma - endpoints of a spectrum of one disease? PLoS One. 2013;8(11):e78812. 43. Hartmann S, Döring C, Vucic E, et al. Array comparative genomic hybridization reveals similarities between nodular lymphocyte predominant Hodgkin lymphoma and T cell/histiocyte rich large B cell lymphoma. Br J Haematol. 2015;169(3):415-422. 44. Masters BR. Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases, Eighth Edition (2015) Eds: John E. Bennett, Raphael Dolin, Martin J. Blaser. ISBN: 13-978-1-45574801-3, Elsevier Saunders. Graefes Arch Clin Exp Ophthalmol. 2016;254:2285-2287. 45. Wang H, Yan K, Wang R, et al. Antibody heavy chain CDR3 length-dependent usage of human IGHJ4 and IGHJ6 germline genes. Antib Ther. 2021;4(2):101-108. 46. Forsgren A, Grubb AO. Many bacterial species bind human IgD. J Immunol. 1979;122(4):1468-1472. 47. Perez Vidakovics ML, Riesbeck K. Virulence mechanisms of Moraxella in the pathogenesis of infection. Curr Opin Infect Dis. 2009;22(3):279-285.

L. Thurner et al.

48. Samuelsson M, Jendholm J, Amisten S, Morrison SL, Forsgren A, Riesbeck K. The IgD CH1 region contains the binding site for the human respiratory pathogen Moraxella catarrhalis IgD-binding protein MID. Eur J Immunol. 2006;36(9):2525-2534. 49. Zucca E, Arcaini L, Buske C, et al. Marginal zone lymphomas: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2020;31(1):17-29. 50. Wotherspoon AC, Ortiz-Hidalgo C, Falzon MR, Isaacson PG. Helicobacter pylori-associated gastritis and primary B-cell gastric lymphoma. Lancet. 1991;338(8776):1175-1176. 51. Wotherspoon AC, Doglioni C, Diss TC, et al. Regression of primary low-grade B-cell gastric lymphoma of mucosaassociated lymphoid tissue type after eradication of Helicobacter pylori. Lancet. 1993;342(8871):575-577. 52. Chanudet E, Zhou Y, Bacon CM, et al. Chlamydia psittaci is variably associated with ocular adnexal MALT lymphoma in different geographical regions. J Pathol. 2006;209(3):344-351. 53. Ferreri AJM, Govi S, Pasini E, et al. Chlamydophila psittaci eradication with doxycycline as first-line targeted therapy for ocular adnexae lymphoma: final results of an international phase II trial. J Clin Oncol. 2012;30(24):2988-2994. 54. Borchmann S, Joffe E, Moskowitz CH, et al. Active surveillance for nodular lymphocyte-predominant Hodgkin lymphoma. Blood. 2019;133(20):2121-2129.

Haematologica | 108 December 2023

3358

ARTICLE - Plasma Cell Disorders

Amyloidogenic light chains impair plasma cell survival Marjorie Pick,1 Eyal Lebel,1 Sharona Elgavish,2 Hadar Benyamini,2 Yuval Nevo,2 Rachel Hertz,3 Jacob Bar-Tana,3 Paola Rognoni,4 Giampaolo Merlini4 and Moshe E. Gatt1

Correspondence: M. Pick marjorie@cc.huji.ac.il

Department of Hematology, Hadassah Medical Center, Faculty of Medicine, Hebrew University of Jerusalem, Jerusalem, Israel; 2Info-CORE, I-CORE Bioinformatics Unit of the Hebrew University of Jerusalem, Jerusalem, Israel; 3School of Public Health, Hebrew University of Jerusalem, Jerusalem, Israel and 4Amyloidosis Research and Treatment Center, Fondazione IRCCS Policlinico San Matteo and University of Pavia, Pavia, Italy. 1

Received: Accepted: Early view:

November 29, 2022. June 20, 2023. June 29, 2023.

Abstract Systemic light chain amyloidosis (AL) is a clonal plasma cell disorder characterized by the deposition of misfolded immunoglobulin light chains (LC) as insoluble fibrils in organs. The lack of suitable models has hindered the investigation of the disease mechanisms. Our aim was to establish AL LC-producing plasma cell lines and use them to investigate the biology of the amyloidogenic clone. We used lentiviral vectors to generate cell lines expressing LC from patients suffering from AL amyloidosis. The AL LC-producing cell lines showed a significant decrease in proliferation, cell cycle arrest, and an increase in apoptosis and autophagy as compared with the multiple myeloma LC-producing cells. According to the results of RNA sequencing the AL LC-producing lines showed higher mitochondrial oxidative stress, and decreased activity of the Myc and cholesterol pathways. The neoplastic behavior of plasma cells is altered by the constitutive expression of amyloidogenic LC causing intracellular toxicity. This observation may explain the disparity in the malignant behavior of the amyloid clone compared to the myeloma clone. These findings should enable future in vitro studies and help delineate the unique cellular pathways of AL, thus expediting the development of specific treatments for patients with this disorder.

Introduction Systemic light chain amyloidosis (AL) is a rare monoclonal plasma cell (PC) disorder characterized by the systemic deposition of misfolded immunoglobulin light chains (LC) as insoluble fibrils in organs.1 Without specific therapies patients who present with amyloidogenic LC that target vital organs will ultimately die.2 Furthermore, AL amyloidosis may be found in 10% of patients with multiple myeloma, in whom it confers a worse prognosis.3-6 Unlike in multiple myeloma (MM), in which the clinical picture is dominated by the hyper-proliferative MM clone, in AL severe organ dysfunction is usually caused by a small PC clone producing the amyloidogenic LC.7 Treatment approaches to AL are derived from MM protocols and aim at suppressing the clone with chemotherapy and novel PC targeted agents.8,9 Lessons learned in the clinic reveal that reducing the concentration of the circulating amyloidogenic free LC improves cardiac function and prolongs survival.10,11 This indicates that it is not solely the deposition of the mass of the amyloid in the tissues which causes organ injury1,12,13 and that toxicity of the LC

is a fundamental contributor.7 In AL, the PC clones are usually small (median 10% of bone marrow cells) and, interestingly, the free LC levels in AL may be 10 logs lower than in MM irrespective of serious organ damage. Amyloid LC precursors are likely to mediate cellular toxicity through a mechanism that causes oxidative stress and activates the apoptotic pathway.1,14 Much research has been performed to determine the molecular factors that make a particular LC protein amyloidogenic, and to elucidate the mechanism of amyloid fibril formation and even to characterize the amyloid formation in vitro, with some success at unveiling the process.1,12,14,15 To date, there are only two characterized amyloidogenic cell lines, which were produced from the same patient.16 Their limits are the lack of an analogous control non-amyloidogenic line, and their ability to produce only one type of LC. Both MM and AL originate from a neoplastic PC clone; however, AL has an attenuated proliferative physiology and different clinical behavior. Several lines of evidence indicate that the amyloidogenic clone is at the crossroads between monoclonal gammopathy of undetermined sig-

Haematologica | 108 December 2023

3359

ARTICLE - AL plasma cell survival

M. Pick et al.

nificance and MM, with less mutation burden,17 more stable genetic status,18 and lower intraclonal heterogeneity19 than MM. Previous reports indicate that amyloid LC determine endoplasmic reticulum oxidative stress in the PC that produce them, sensitizing the clone to the action of proteasome inhibitors and thus providing a molecular basis for the exquisite sensitivity to these agents observed in the clinic.20 The same mechanism has been shown in PC producing LC that also aggregate causing LC deposition disease.21 Here we report the establishment of a model to study the molecular mechanisms underlying amyloid LC proteotoxicity. We used lentiviral vectors that contain cloned LC sequences from patients with AL amyloidosis with cardiac or renal involvement and from patients with MM to enable stable expression of LC in MM cell lines. The purpose of our study was to investigate the effects of amyloid LC on PC proliferation and metabolism, two key aspects that were not explored in the previous two studies that focused on endoplasmic reticulum stress and protein degradation pathways.20,21

Methods Cell lines The cell lines used were: KMS11 and NCI-H929 (MM cell lines), HL60 (an acute promyelocytic leukemia cell line), 721.221 (an HLA-negative B-cell line) and 293T or HEK293T (a human embryonic kidney cell line that expresses a mutant version of the SV40 large T antigen) all kindly donated by Prof Ben-Yehuda (Hadassah Medical Organization, Israel); JJN3 (a MM cell line) kindly donated by Dr Katia Beider (Hebrew University, Israel); HEK293 (a human embryonic kidney cell line) kindly donated by Prof Ariella Oppenheim (Hebrew University, Israel); normal human fibroblasts, kindly donated by Dr Scibienski (Stockholm University, Sweden); and HC92 (a rat cardiomyocyte cell line) kindly donated by Prof Ronen Beni (Hebrew University, Israel). Foreskin fibroblasts, HC92, HEK293 and 293T cells were grown and maintained in Dulbecco modified Eagle medium (Thermo Fisher, USA) while the other cell lines were grown and maintained in Roswell Park Memorial Institute medium (Thermo Fisher, USA). All media were supplemented with 10% fetal calf serum (Thermo Fisher, USA), L-glutamine (29.2 μg/mL), and penicillin (1,000 U/mL)/streptomycin (10 mg/mL) (Biological Industries, Israel) and incubated at 37°C in 5% CO2. All cell lines were passaged bi-weekly and tested for mycoplasma. Generation of cell lines expressing light chains of myeloma and AL amyloidosis patients In order to establish a new model to study amyloidogen-

esis and its toxic effects in living human cells we used lentiviral vectors to stably express the production of amyloidogenic LC in MM cell lines. The 600 base pair LC gene was cloned within a pCR 2.1-TOPO plasmid as previously described22 (sequence in the Online Supplementary File). The same 600 base pair LC sequence was generated via polymerase chain reaction from this plasmid and inserted into the multiple cloning site of the minusDsRed-GFP expressing lentivirus (kindly donated by Prof. O. Mandelboim). Thus, we generated lentivirus plasmids cloned with amyloidogenic λ LC cDNA sequences from a patient who suffered from cardiac AL [H], another LC cDNA sequence from a second patient who suffered from kidney - nephrotic AL [K] and a LC sequence from a third patient who did not have AL and was denoted non-amyloidogenic MM [M]. All plasmids contained a reporter sequence of green fluorescent protein (GFP) expression. Using 293T cells we generated infectious viral particles containing λ LC and stably transduced three MM cell lines which have the intracellular machinery not only to produce but also to secrete LC proteins. Thus, we were able to create novel MM cell lines that produce, in addition to their own MM k LC,2325 an amyloidogenic λ LC and GFP. Two separate clones of each AL LC were generated (e.g., M2.3 and M2.4, or K1.4 and K1.10, or H3.8 and H3.5) and then at least three infections of each separate clone were performed on the MM cell lines JJN3, KMS11 and NCI-H292. The number of RNA transcripts for JJN3 and three separate transfections were assessed by RNA sequencing (Figure 1A). Total λ protein in KMS11 lysate (Figure 1B) and GFP expression, determined by flow cytometry (Figure 1C), detected the presence of the LC. Six biological repeats for every LC were tested in this study. These MM cell lines all produced κ LC naturally allowing the λ LC-producing plasmid to be detected and distinguished clearly from the κ production of the cell lines themselves (Figure 1D). The proliferation rate of the three lines was between 36 and 50 h. Additionally, we transduced non-MM cell lines (HL60, 293T, 722.221) with two clones of each AL LC [H] and [K] and MM LC [M] and two separate infections as a further control to determine intracellular toxicity of the amyloidogenic LC on non-PC lines. GFP and λ expression was assessed by flow cytometry (FACSCalibur, BD Bioscience, USA) in order to determine the efficiency of infection (Figure 1C). No decrease in viability or increase in apoptosis was found in the non-PC lines transduced with the AL and MM LC (Online Supplementary Figure S1A, C). However, addition of exogenouse AL LC was toxic (Online Supplementary Figure S1B). In the majority of cases more than 95% of cells were GFP-positive (Figure 1C). Using an enzymelinked immunosorbent assay (Bethyl, TX, USA), we were able to detect the AL LC intracellularly in both MM and non-MM cell lines (Figure 1A, B), but secreted AL and MM LC only in the MM cell lines (Figure 1D). Supernatant was

Haematologica | 108 December 2023

3360

ARTICLE - AL plasma cell survival

M. Pick et al.

collected from transfected MM cells that secreted their LC into the medium while growing. After 5 days of growth the supernatant was collected and incubated for 48 h with MM cells, primary kidney and heart cells, 721.221 and HL60, which were then assessed for apoptosis by flow cytometry following exposure to the medium containing LC. Significant increases in apoptosis (determined by propidium iodide and annexin V staining) were detected

in the primary cells and MM cells when incubated with supernatant containing cardiotoxic [H] and nephrotoxic [K] LC, but not with supernatant containing non-amyloidogenic MM LC [M] (Figure 1E-G). Cell proliferation Ten thousand cells were seeded in wells and every 3 days a sample of cells was removed and enumerated using try-

Continued on following page. Haematologica | 108 December 2023

3361

ARTICLE - AL plasma cell survival

M. Pick et al.

Figure 1. Assessment of intracellular and secreted λ light chains in transduced cell lines. (A) The table shows the raw counts and number of insertions, from RNA sequencing (each a separate infection of the 3 repetitions performed) of the JJN3 cell line JJN3 infected with only the green fluorescent protein (GFP) reporter or lentivirus containing cardiotoxic [H], nephrotoxic [K] amyloidogenic and multiple myeloma (MM) [M] λ light chains (LC). (B) Representative western blot illustrating the expression of λ protein in KMS11 cell lines: control (GFP [G]), AL LC ([H] and [K]), and MM LC [M] including intensity of λ bands (N=4, for separate transfections average for both JJN3 and KMS11 transduced lines). (C) Representative dot plot from flow cytometry analysis detecting monoclonal λ in the transduced KMS11 cell lines (N=6). The table shows the percent GFP and λ -positive cells from four cell lines infected (2 separate transfections). (D) Enzyme-linked immunosorbent assay to quantitate the secretion of λ LC was performed on the supernatants from which the various cell lines were grown after they had been infected. Supernatants containing secreted LC were collected 6 days after JJN3, KMS11 (MM cell lines), 721.221 and HL60 (non-MM cell lines) had been transduced with the various LC lentiviruses. GFP: empty control; [M]: multiple myeloma; [H] cardiotoxic or [K] nephrotoxic amyloidogenic LC lentivirus (N=6, 3 separate transfections in duplicate). *P<0.05 compared to GFP control. (E) Supernatants collected from MM cell lines (KMS11 and JJN3) that secreted LC and were found to contain λ LC were incubated with both MM and nonMM cell lines (JJN3, foreskin fibroblasts, heart cells [HC92] and kidney cells [HEK293]) and then the cells were assessed for increases in apoptosis (by propidium iodide staining, sub-G1 peak) as compared to the control. *P<0.05 compared to the GFP control (N=4, 2 experiments from KMS11 and 2 from JJN3 combined). (F) Cell cycle analysis of cell lines from figure (E). No significant differences in cell cycle were observed, only increases in apoptosis. (G) Supernatants that were collected from MM (JJN3) cell lines that secreted LC and were found to contain λ LC were incubated with both MM and non-MM cell lines (JJN3, KMS11, 721.221, and HL60) and cells were assessed for increases in annexin V expression as compared to control. *P<0.05 compared to the GFP control (N=4, 2 experiments from JJN3 combined).

pan blue exclusion and a hemocytometer to assess cell growth. Additionally, cell viability and proliferation were assessed using the CellTiter-Glo® kit (Promega, MI, USA), which measures viability and proliferation of cells by quantitating luminescence of ATP. Ten thousand cells were seeded in either quadruplicate or quintuplicate and after 5 days the proliferation was assessed by read-out of luminescence. Cell cycle and apoptosis analysis A re-suspended pellet of 1x106 cells was incubated over-

night at 4°C in 1 mL of 100% ethanol (AR. Gadot, Israel) for fixation and to minimize clumping. The cells were then washed with 1 mL phosphate-buffered saline (PBS) and treated with 50 μL of 100 μg/mL RNase A (Sigma-Aldrich, USA) for 30 min at 37°C to remove traces of RNA from the samples. The cells were then washed in PBS and 25 μL of propidium iodide (1 mg/mL) were added. Samples were analyzed by flow cytometry and gated on the FL2-Width FL2-Area to identify clumps and doublets. Ten thousand events were acquired and apoptotic cells were detected in the ‘SUB G1’ area of the graph by a FACScalibur cyto-

Haematologica | 108 December 2023

3362

ARTICLE - AL plasma cell survival

M. Pick et al.

meter (BD Bioscience, USA). Annexin V binding experiments were performed using the APC Annexin V Apoptosis Detection Kit with propidium iodide (Biolegends, CA, USA). Briefly, 1x106 cells were washed with annexin binding buffer and then incubated with annexin V APC for 15 min at room temperature. Further washing with binding buffer was performed and 5 mL (0.5 mg/mL) of propidium iodide were added before acquisition on the FACScalibur cytometer. Western blots Washed and pelleted cells were lysed in cold RIPA buffer (1 mL/100 mg) containing 10 mM phenylmethylsulfonyl fluoride (Thermo Scientific), a serine protease inhibitor, and centrifuged at ~14,000 x g for 15 min. Supernatant was collected for western blot analysis. The samples were run in 3-(N-morpholino) propane sulfonic acid on acrylamide gels using a Sea Blue marker for size, then transferred to a polyvinylidene fluoride membrane. After transfer, the samples were blocked for 90 min with 5% skim milk solution in PBS-Tween. The membrane was incubated with primary antibody overnight at 4°C and then placed in the secondary antibody solution for 30 min. Electroluminescence was used to detect bands. The antibodies used were purchased from: Abcam (UK) – λ (1:5,000, ab109247 - polyclonal), κ – (1:20,000, ab134083 - polyclonal), c-Myc – (1:10,000, ab152146 - polyclonal), – p62 (1:1000, ab56416 - SQSTM1); Santa Cruz Biotechnology (TX, USA) – Cyclin E (1:200 of 200 mg/mL, SC-247-HE12), – GAPDH (1:400 of 200 mg/mL, SC-25778 -FL335), – Actin (1:5000, 41185 C4) or Cell Signaling Technology (MA, USA) – Caspase-3 (1:1000, 9665-8G10), – PARP/PARP1 (1:1000, 9542). Exogenous light chain assay Purified LC were obtained as previously described22 from patients' 24 h urine collections. Briefly, human urine samples combined with 0.1% sodium azide (w/v) were centrifuged and ammonium sulphate was added to precipitate the LC which were then solubilized in 20 mM sodium phosphate, pH 7.0, and dialyzed against the same buffer. LC were purified by anion exchange chromatography. MM and non-MM cell lines were seeded in 96-well plates at 10,000 cells/well, in pentaplicate, and LC which were resuspended in medium were added with a pipette at the designated concentrations to medium containing cells, and cultured for 5 days. CellTiter-Glo® (Promega, USA) reagent was added and the luminescence of the plates was read using a Berthold illuminometer (Titertek, Germany). RNA extraction for RNA sequencing We used a Qiagen RNEasy mini kit to extract RNA and tested the quality of the RNA with 6000 Nano Bio analyzer (Agilent Technologies, CA. USA). Transcriptome libraries

were prepared using the Illumina TruSeq RNA Library Preparation Kit (Illumina #RS-122-2001), according to the manufacturer’s recommended protocol, starting with approximately 1.2 mg of total RNA. The amplified indexed libraries were quantified using an Invitrogen Qubit fluorometer and pooled equally according to the pool design. Pooled libraries were run on a 4% agarose gel and DNA approximately 270 base pairs long (the length of RNA inserts plus the 3′ and 5′ adaptors) was size-selected and recovered in 15 µL elution buffer (QIAGEN). Size-selected libraries were then quantified again using the Qubit Fluorometer. Size was verified using High Sensitive DNA gels on an Agilent 2200 TapeStation instrument. Libraries were sequenced on a NextSeq 500 instrument using the NextSeq 500 High Output V2 Sequencing Kit (FC-404-2005), in a single-end configuration, reading 80 base pairs. Sequencing and differential expression analysis Raw reads were quality-trimmed at both ends, using inhouse Perl scripts, with a quality threshold of 32. Following quality-trimming, adapter sequences were removed with cutadapt (version 1.9.1, http://cutadapt.readthe docs.org/en/stable/),22 filtering out reads that became shorter than 15 nt (-m parameter). The remaining reads were further filtered to remove very low quality reads, using the fastq_quality_filter program of the FASTX package (version 0.0.14, http://hannonlab.cshl.edu/fastx_toolkit/), with a quality threshold of 20 at 90% or more of the reads’ positions. The processed reads were aligned with up to five mismatches per read to the human transcriptome and genome using TopHat (v2.0.14).23 The genome version was GRCh38, with annotations from Ensembl release 84. The genome was slightly modified to include the different transgenic sequences (as additional chromosomes). Raw counts were obtained with the Cufflinks package (v2.2.1),24,25 using the cuffquant program with the genome bias correction (-b parameter) and the multi-mapped reads assignment algorithm (-u parameter), followed by cuffnorm. Normalization and differential expression were achieved with the DESeq2 package (version 1.12.4),26 using default parameters. Functional enrichment analysis Downstream analysis of the expression data was done using two approaches. The first approach used a cut-off (i.e., threshold-dependent). Significantly differentially expressed genes (Padj<0.1) were subjected to pathway and molecular function enrichment analysis using Ingenuity Pathway Analysis (IPA®) (QIAGEN Inc., https://digitalinsights.qiagen.com/products-overview/discovery-insightsportfolio/content-exploration-and-databases/qiagen-ipa/ . The second approach consisted of whole data analysis. Whole differential expression data from amyloidogenic

Haematologica | 108 December 2023

3363

ARTICLE - AL plasma cell survival

M. Pick et al.

[H] and [K] versus non-amyloidogenic [M] LC-expressing cell lines were subjected to gene set enrichment analysis (GSEA).26 GSEA uses all differential expression data (cutoff independent) to determine whether sets of genes, defined a priori, show statistically significant, concordant differences between two biological states. GSEA was run against the hallmark gene set collection from the molecular signatures database (mSigDB). Signals with normalized enrichment scores of more than 3 or less than -3 were chosen for experimental validation. Oxidative stress measured by flow cytometry One million cells of the JJN3 cell line and its clones were treated with either 3.3 μg/mL MitoSOX (Thermofisher Scientific, USA) or 5 μM dihydroethidium (DHE, Sigma, USA) for 30 min in phenol-free medium. Tetramethylrhodamine methylester perchlorate (TMRM, Molecular Probes, USA) was added to the cell culture (25 nM final concentration) for the last hour of incubation. Following all incubations, cells were washed twice with PBS and then 50,000 cells were acquired and analyzed using a FACScalibur cytometer. MitoSOX measures mitochondrial superoxide while DHE measures total cellular reactive oxygen species. TMRM measures mitochondrial membrane potential. Cholesterol assay The Amplex® Red Cholesterol Assay Kit (Thermofisher Scientific, USA) measures the concentration of both free cholesterol and cholesteryl esters by a coupled enzyme assay, which results in a colorimetric (570 nm)/fluorometric (λex=535/λem=587 nm) product, proportional to the cholesterol present. The assay was performed according to the kit's instructions. MM cells from [K], [H] and [M] lines were seeded at 2 x 10 6 cells/50 mL of reaction mixture and assayed for free and total cholesterol levels. Cytometric bead assay The BDTM Cytometric Bead Array (BD Bioscience, USA) is a flow cytometry application that allows quantification of multiple cytokines simultaneously. The Cytometric Bead Array system was used to capture cytokines secreted into the supernatant by MM cells, seeded at 1x105/mL, which had been transfected with [K], [H] and [M] LC. The assay was performed according to the kit’s instructions. Events were acquired on a FACScalibur and analyzed with Simplify Analysis with FCAP Array v3.0 software and Excel. Statistics A standard t test was used to determine statistically significant differences in all experiments not involving RNA assessment and RNA sequencing.

Results Amyloidogenic light chains are toxic to plasma cells Amyloidogenic and non-amyloidogenic MM cell lines were produced from constructs encoding cDNA from 600 patient-specific base pair-sequenced LC that were generated from patients suffering from cardiotoxic [H] and nephrotoxic [K] AL and non-amyloidogenic MM ([M]). LC sequences (Online Supplementary Data File) and functional changes in PC biology and growth were assessed. Multiple transfections were performed from two clones of each LC: [K], [H], and [M]. As an added control, we transduced the MM cell line with a lentiviral vector encoding only GFP [G]. We found that the MM cells (JJN3, KMS11 and NCIH929) which contained cardiotoxic [H] LC and/or nephrotoxic [K] LC had a significantly lower proliferation rate (Figure 2A, data not shown) and significantly more apoptosis/autophagy (Figure 2B-D, data not shown) than the control cell lines ([M] and [G]). The MM cells transduced with nephrotoxic [K] LC had significantly more cells in G0/G1 of the cell cycle and expressed higher levels of the autophagic protein p62 (Figure 2B, E) compared to control cell lines. We assessed the expression of autophagic protein p62 to determine concordance with previously published data. However, the levels of proteins involved in the cell cycle, such as Cyclin E, a protein required for the transition of cells from G1 to S, showed no significant differences between the cell lines (Online Supplementary Figure S2A). Nevertheless, the levels of proteins involved in apoptosis, such as cleaved caspase 3, were increased in the MM cell lines transduced with constructs encoding amyloid LC (Online Supplementary Figure S2A) and these cells showed significantly lower viability, as measured by lower ATP consumption using the CellTiter-Glo® assay (Figure 2D and F, asterisks). Moreover, poly (ADP-ribose) polymerase (PARP), a family of proteins involved in a number of cellular processes such as DNA repair, genomic stability, and programmed cell death, was shown to be highly increased in the MM cells transduced with [K] LC (Online Supplementary Figure S2B, C). To understand the selective mechanism involved in the cell proliferation and death of the cells transduced with amyloidogenic LC, RNA-sequencing analysis was performed on the cells to compare the biological effects of AL LC versus MM LC in MM cell lines. RNA-sequencing analysis of genes involved in autophagy and the intrinsic and extrinsic apoptotic pathways generated heatmaps displaying significant changes in RNA levels when the three cell lines were compared (Online Supplementary Figure S2D), showing that the cell lines containing AL LC had increases in death pathways in line with increases in autophagic protein p62 and apoptosis (Figure 2). FAS ligand RNA levels were upregulated in the JJN3 cell lines

Haematologica | 108 December 2023

3364

ARTICLE - AL plasma cell survival

M. Pick et al.

Figure 2. Cardiotoxic and, nephrotoxic amyloidogenic light chains are toxic to multiple myeloma cell lines. (A) Cell counts of multiple myeloma (MM) cell lines (JJN3 and KMS11) over 9 days following transduction with MM [M], cardiotoxic [H] or nephrotoxic [K] amyloidosis (AL) light chain (LC) constructs or with only the green fluorescent protein (GFP) reporter (5 separate infections for each cell line). (B) Cell cycle analysis of transduced MM cell lines (KMS11 and JJN3) with the previous four lentivirus constructs. Table of summary of percent cells in each stage of the cell cycle. *P<0.05, **P<0.01 (N=6; separate transfections 3 times JJN3 and 3 times KMS11). (C) Representative May-Grünwald staining of the MM cell line KMS11: untreated, infected with empty control lentivirus (GFP), with MM or with cardiotoxic (AL) LC (N=4 separate transfections). (D) Assessment of ATP uptake (intensity) as an indicator of viability of MM cell lines (JJN3 and KMS11) that were uninfected (control, C) or infected with the same lentiviral vectors (3 separate transfections for each MM cell line). (E) Representative western blot analysis of JJN3 lysate of p62 (an autophagy substrate that is used as a reporter of autophagy activity) with (+) and without (-) bafilomycin, a potent inhibitor of cellular autophagy (6 replicate experiments, duplicate for each MM cell line). The table shows the mean ± standard deviation of the six replicates. (F) Purified LC were isolated from patients’ urine and added to the media of growing MM JJN3 cells and viability was assessed by ATP uptake (intensity) using the CellTiter-Glo® assay. LC from MM patients’ urine (M7 and M8), LC from patients with cardiotoxic AL (H7, H15 and H18) or LC from a patient with nephrotoxic AL (K5) were incubated for 5 days with naïve JJN3 cells. Triangles indicate increases in LC concentration, i.e., 0, 50, 100, 200 and 400 mg/mL, (N=3, 4-5 replicates per experiment). Haematologica | 108 December 2023

3365

ARTICLE - AL plasma cell survival

M. Pick et al.

transduced with MM LC, whereas caspase-3 and -6 showed trends of RNA upregulation in the cell lines transduced with [K] and [H]. When IPA® Map Activator Prediction analysis was performed, it was found that key genes in the death receptor signaling pathway were either significantly experimentally upregulated or predicted to be upregulated, as calculated from the expression of the measured genes of this pathway by both the AL LC cell lines as compared to [M] (Online Supplementary Figure S2E). The involvement of the autophagic pathway in AL LC toxicity has been reported elsewhere and our results concord with these findings.20 To determine whether the toxicity of the amyloidogenic LC that was seen in the MM cell lines was specific to PC only, the same lentivirus combinations were transduced multiple times into 293T (kidney), HL60 (acute myeloid leukemia) and 721.221 (B lymphocyte) cell lines. We did not observe changes in viability, measured by ATP consumption, or decreases in cell cycle when the LC were produced within the cells (S and G2M, Online Supplementary Figure S1A-C). Transduced amyloidogenic light chains cause a significant increase in stress pathways in multiple myeloma cells Further RNA-sequencing analysis was performed on the cells containing AL LC versus MM LC in order to compare biological effects in the MM cell lines. A gene expression heatmap (Online Supplementary Figure S3A) showed significantly upregulated or downregulated expression of various genes in the MM cells containing AL LC. An IPA®

heatmap of the cell death and survival category shows functions that are involved in cell viability and cell death for which significantly enriched differentially expressed genes were found in MM cells that contained amyloid LC (Padj<0.1, Benjamini-Hochberg P<0.05) (Figure 3A, Online Supplementary Table S1). Furthermore, it is interesting to note that functions that are involved in cell death and apoptosis are predicted to be significantly upregulated, while those that are involved in cell viability are predicted to be downregulated, as reflected by IPA® z-scores that are higher than 2 in the former and lower than -2 in the latter. These results validate the phenotypic results shown earlier (Figure 2). In the IPA® heatmap representing small molecule biochemistry, three molecular pathways, ‘concentration of ATP’, ‘metabolism of cholesterol’ and ‘concentration of Dglucose’, are predicted to decrease significantly in the AL LC-containing cell lines (Figure 3B, Online Supplementary Table S2) as compared to MM LC-containing cell lines. Interestingly, 11 significantly differentiated upregulated genes which participate in the oxidative phosphorylation enriched canonical pathway (Benjamini-Hochberg, P<0.05) are upregulated in MM LC-containing cells as compared to the cell lines expressing AL LC (Figure 4A, depicted with pink outline, Online Supplementary Table S3). However, IPA® does not predict activation states of pathways and therefore the upregulation of these 12 genes does not necessarily signify activation of the pathway. We therefore validated this pathway in vitro. Mitochondrial superoxide was measured by flow cytometry in MM cell lines containing AL LC and MM LC. Similarly to data generated by sec-

Continued on following page. Haematologica | 108 December 2023

3366

ARTICLE - AL plasma cell survival

M. Pick et al.

Figure 3. RNA-sequencing analysis demonstrates a significant increase in apoptotic and cell death pathways in JJN3 cells infected with amyloidogenic light chains. (A, B) QIAGEN Ingenuity Pathway Analysis® (IPA®) biological function or disease heatmap of the categories cell death and survival (A) and small molecule biochemistry (B). Each box represents a specific biological function or disease that is connected to the two functional categories above. The size of the box correlates with the gene enrichment and the color of the box indicates the predicted increase (orange, positive z-score) or decrease (blue, negative z-score). Gray indicates no change in the biological process/disease activity. Functions with IPA® z-scores >2 or < -2 are predicted to be significantly upregulated/downregulated, respectively. IPA® calculates the P value and the activation z-score independently and, therefore, functions with moderate enrichment can have a significant activation z-score due to hallmark genes that contribute highly to the activation state. Thirty-seven molecular functions and diseases passed the absolute z-score of 2, of which ten are shown in the tables. For all experiments, three separate transductions in JJN3 cells, P<0.05.

ondary RNA sequencing (Figure 4A), cells expressing AL LC ([H] and [K]) had significantly increased mitochondrial oxidative stress compared to MM cells containing nonamyloidogenic LC [M] (Figure 4B). No change was observed in mitochondrial membrane potential, as analyzed by TMRM. Another well-established method to identify classes of genes whose expression is significantly altered between different states is GSEA. We used GSEA to compare the significant changes in gene set expression from JJN3 MM cells lines that expressed AL LC ([K] and [H]) versus cell lines that contained non-AL MM LC. The analysis found four gene sets with significant differences between the states. Cholesterol homeostasis, TNFα signaling via NFκB and hypoxia were significantly enriched in the downregulated genes, while the Myc-target gene set was significantly enriched in the upregulated genes in AL LC ([H] and [K]) cells compared to non-amyloidogenic LC [M] (Figure 5A-C). Cholesterol homeostasis was significantly downregulated in the AL LC JJN3 MM cells and significantly higher levels of both free and total cholesterol were found in the cells transduced with AL LC ([H] and [K]) (Figure 5A). This result indicates a decrease in cholesterol metabolism and homeostasis, confirming the results seen in GSEA. Genes known to be Myc targets or involved in the Myc pathway were shown to be upregulated in the AL LC

MM cells and higher levels of Myc protein and RNA itself were found in these cells (Figure 5B, Online Supplementary Figure S3B). TNFα enhances cell invasion via the NFκB pathway and NFκB controls DNA replication, cytokine production and cell survival. Genes involved in TNFα signaling via NFκB were downregulated in the AL LC cells as compared to JJN3 MM LC cells, indicating decreased cell survival and cytokine production of the inflammatory pathway (Figure 5C, Online Supplementary Table S4). Secretion of various cytokines involved in the NFκB pathway was measured in the supernatant of AL LC versus JJN3 MM LC cells using the the Cytometric Bead Array kit. Six cytokine levels were measured simultaneously (IL-8, IL-1β, IL-6, IL-10, TNFα and IL-12p70). Concentrations of IL-1β, TNFα and IL-12p70 were below the detectable limit of the assay. However, for IL10, an anti-inflammatory cytokine, concentrations were significantly increased in AL [H] LC cells as compared to MM LC cells. In contrast, levels of the pro-inflammatory cytokines IL-8 and IL-6 were significantly lower in the AL LC-containing cells (P<0.02-0.0002) (Figure 5C). It is worth noting that IL-6 is known to be involved in MM disease progression. These results indicate that the AL LC MM cells have decreased NFκB activity, which could explain the decrease in cell viability, with the interplay balanced between various inflammatory mediators as well.

Haematologica | 108 December 2023

3367

Haematologica | 108 December 2023

3368

Figure 4. RNA-sequencing analysis demonstrates a significant increase in the oxidative stress pathway in JJN3 cells infected with amyloidogenic light chains. (A) Ingenuity Pathway Analysis® enriched canonical pathway for oxidative phosphorylation (Benjamini-Hochberg procedure, P=0.046) with pink lines showing upregulation of significantly changed genes that participate in this pathway. (B) Validation using fluorescence of reactive oxygen species, measured by flow cytometry. Multiple myeloma (MM) cells containing amyloidogenic light chains ([H] and [K]) have increased mitochondrial oxidative stress as compared to MM cells containing non-amyloidogenic light chains [M]. For all experiments, three separate transductions in JJN3 cells, *P<0.05. TMRM: tetramethylrhodamine methylester perchlorate.

ARTICLE - AL plasma cell survival M. Pick et al.

ARTICLE - AL plasma cell survival

M. Pick et al.

Figure 5. Gene set enrichment analysis of JJN3 cells infected with amyloidogenic versus non-amyloidogenic light chains. In the gene set enrichment analysis (GSEA), pathways represented had a normalized enrichment score (NES) of more than 3 or less than -3 and a false discovery rate (FDR) of 0.02. (A) GSEA graph of the cholesterol pathway (left) with validation using an enzyme-linked immunosorbent assay (right) to detect the presence of cholesterol in JJN3 cells expressing either amyloidogenic (AL) or multiple myeloma (MM) light chains (LC). Note the significantly higher levels of both free and total cholesterol in the cells transduced with AL LC ([H] and [K]) as compared to cells transduced with MM LC [M], N= 5, *P<0.005, **P<0.002. (B) GSEA graph of the Myc target pathway (left) and representative western blot analysis showing high levels of Myc in the cells transduced with AL LC ([K] and [H]) as compared to cells transduced with MM [M] or GFP alone [G] or untreated MM cells [J] (right). Values below are the intensities of the bands. (C) GSEA graph of TNFα signaling via NFκB (left) and concentrations of the cytokines IL-10, IL-8 and IL-6, as determined by Cytometric Bead Assay (right). Note the significant decreases of IL-8 and IL-6 in the cells transduced with AL LC ([K] and [H]) *P<0.020.0002. (D) GSEA graph of the hypoxia pathway. For all experiments, three separate transfections in JJN3 cells.

Finally, the gene set associated with hypoxia was found to be significantly downregulated in the AL LC MM cells (Figure 5D). Hypoxia occurs when cells are deprived of adequate levels of oxygen. If the AL LC cells are continuously

hypoxic this could contribute to the increased cell death observed in these cells. This effect may be due to the increase in reactive oxygen species causing activation of the hypoxia genes as a secondary stress response.

Haematologica | 108 December 2023

3369

ARTICLE - AL plasma cell survival

M. Pick et al.

Discussion The amyloid PC clone shows a selective high sensitivity to proteasome inhibitor-based therapy, resulting in a high rate of rapid, complete responses.27 In previous studies this exquisite sensitivity was attributed to the oxidative endoplasmic reticulum stress induced by amyloid LC.20,21 Furthermore, the time to next therapy was reported to be ≥7 years in 60% of patients achieving complete remission after bortezomib-based therapy,27 indicating a low proliferative rate of the amyloid clone, as also documented by early studies analyzing the PC labeling index.28 However, the molecular basis of the indolence of the amyloid clone remained undetermined. We hypothesized that the clone size and malignant behavior may be partially influenced by the internal LC proteotoxicity to the diseased PC itself. The amyloidogenic PC in our study were phenotypically different with significant growth arrest compared to their non-amyloidogenic controls (Figure 2). Not only were they less proliferative but, additionally, the cell lines containing AL LC had higher apoptosis and autophagy (Figure 2, Online Supplementary Figure S2) which is compatible with the findings of previous studies.22,29,30 This apoptotic phenomenon was also shown to be a major mechanism of sensitivity to bortezomib in AL cells20 and altering LC balance within the cells may affect AL PC survival.31 In this study, we demonstrate that this sensitivity was a specific internal PC effect and it did not occur in other types of non-PC lines producing AL LC. To further delineate the underlying mechanism of the LC toxicity we subjected the transduced cells to RNA-sequencing analysis. This first revealed that the cells containing AL LC had decreased viability and increased apoptosis (Figure 3, Online Supplementary Tables S1-3, Online Supplementary Figure S3B), which could be due to major pathways involved in the metabolic stability of these cells. Pathways whose genes were significantly changed included the cholesterol pathway, Myc pathway, NFκB signaling, hypoxia and oxidative stress (Figures 3-5). We validated the RNA-sequencing results by measuring the activation of these pathways in vitro and indeed found these to be more active in MM cells transduced with MM LC than in cells transduced with AL LC (Figures 3-5). This is in line with previous studies showing pro-apoptotic cascades and high oxidative stress caused by amyloido-

genic LC.20,30,32 RNA levels of key regulators of apoptosis and control of proliferation were found to be significantly increased during the RNA-sequencing analysis. The most significant pathways were validated herein, but many others were not investigated further in this study. Our future plans are to confirm some of these gene functions and effects of exogenous compounds in our novel model in order to delineate even more precisely the significant differences in function between a LC generated by PC from MM versus AL patients, and the amyloidogenic PC phenotype. Here we have shown that a novel model of MM cells expressing AL LC in vitro simulated toxicity in PC and that this toxicity leads to increased apoptosis and decreased proliferation due to the malfunction of key biological processes, thus supporting our hypothesis that the AL clone size and low-grade malignant behavior may be partially influenced by internal LC proteotoxicity to the diseased PC itself. Our novel transduction system may enable future in vitro studies to delineate detailed AL-unique cellular pathways and devise specific targeted treatments for AL amyloidosis. Disclosures No conflicts of interest to disclose. Contributions EL, RH, JB, PR, and MP conducted experiments and analyzed the data. MP, SE, HB, YN, GM, and MEG designed the research and analyzed data. MP and MEG wrote the paper. All authors have read and revised the manuscript. Acknowledgments The authors thank Dr. Riki Perlman for editing the final version of the article. Funding This work was supported by project grants from the Israel Cancer Association, Bertha Bekhor, Hematology Research Project, Pfizer Global Medical Grant and the Adele and David Brown bequest fund in the name of the Fineberg family Data-sharing statement All raw RNA-sequencing data can be found online.

References 1. Merlini G, Bellotti V. Molecular mechanisms of amyloidosis. N Engl J Med. 2003;349(6):583-596. 2. Obici L, Perfetti V, Palladini G, et al. Clinical aspects of systemic amyloid diseases. Biochim Biophys Acta. 2005;1753(1):11-22. 3. Desikan KR, Dhodapkar MV, Hough A, et al. Incidence and impact of light chain associated (AL) amyloidosis on the

prognosis of patients with multiple myeloma treated with autologous transplantation. Leuk Lymphoma. 1997;27(3-4):315-319. 4. Madan S, Dispenzieri A, Lacy MQ, et al. Clinical features and treatment response of light chain (AL) amyloidosis diagnosed in patients with previous diagnosis of multiple myeloma. Mayo

Haematologica | 108 December 2023

3370

ARTICLE - AL plasma cell survival

M. Pick et al.

Clin Proc. 2010;85(3):232-238. 5. Rajkumar SV, Gertz MA, Kyle RA. Primary systemic amyloidosis with delayed progression to multiple myeloma. Cancer. 1998;82(8):1501-1505. 6. Rajkumar SV, Gertz MA, Kyle RA. Prognosis of patients with primary systemic amyloidosis who present with dominant neuropathy. Am J Med. 1998;104(3):232-237. 7. Merlini G, Stone MJ. Dangerous small B-cell clones. Blood. 2006;108(8):2520-2530. 8. Gatt ME, Palladini G. Light chain amyloidosis 2012: a new era. Br J Haematol. 2013;160(5):582-598. 9. Palladini G, Merlini G. Current treatment of AL amyloidosis. Haematologica. 2009;94(8):1044-1048. 10. Palladini G, Campana C, Klersy C, et al. Serum N-terminal probrain natriuretic peptide is a sensitive marker of myocardial dysfunction in AL amyloidosis. Circulation. 2003;107(19):2440-2445. 11. Palladini G, Lavatelli F, Russo P, et al. Circulating amyloidogenic free light chains and serum N-terminal natriuretic peptide type B decrease simultaneously in association with improvement of survival in AL. Blood. 2006;107(10):3854-2858. 12. Merlini G, Westermark P. The systemic amyloidoses: clearer understanding of the molecular mechanisms offers hope for more effective therapies. J Intern Med. 2004;255(2):159-178. 13. Phipps JE, Kestler DP, Foster JS, et al. Inhibition of pathologic immunoglobulin-free light chain production by small interfering RNA molecules. Exp Hematol. 2010;38(11):1006-1013. 14. Bellotti V, Chiti F. Amyloidogenesis in its biological environment: challenging a fundamental issue in protein misfolding diseases. Curr Opin Struct Biol. 2008;18(6):771-779. 15. Baden EM, Sikkink LA, Ramirez-Alvarado M. Light chain amyloidosis - current findings and future prospects. Curr Protein Pept Sci. 2009;10(5):500-508. 16. Arendt BK, Ramirez-Alvarado M, Sikkink LA, et al. Biologic and genetic characterization of the novel amyloidogenic lambda light chain-secreting human cell lines, ALMC-1 and ALMC-2. Blood. 2008;112(5):1931-1941. 17. Rossi A, Voigtlaender M, Janjetovic S, et al. Mutational landscape reflects the biological continuum of plasma cell dyscrasias. Blood Cancer J. 2017;7(2):e537. 18. Huang XF, Jian S, Lu JL, et al. Genomic profiling in amyloid light-chain amyloidosis reveals mutation profiles associated with overall survival. Amyloid. 2020;27(1):36-44. 19. Bochtler T, Merz M, Hielscher T, et al. Cytogenetic intraclonal heterogeneity of plasma cell dyscrasia in AL amyloidosis as compared with multiple myeloma. Blood Adv.

2018;2(20):2607-2618. 20. Oliva L, Orfanelli U, Resnati M, et al. The amyloidogenic light chain is a stressor that sensitizes plasma cells to proteasome inhibitor toxicity. Blood. 2017;129(15):2132-2142. 21. Bender S, Ayala MV, Bonaud A, et al. Immunoglobulin light-chain toxicity in a mouse model of monoclonal immunoglobulin lightchain deposition disease. Blood. 2020;136(14):1645-1656. 22. Diomede L, Rognoni P, Lavatelli F, et al. A Caenorhabditis elegans-based assay recognizes immunoglobulin light chains causing heart amyloidosis. Blood. 2014;123(23):3543-3552. 23. Gatt ME, Takada K, Mani M, et al. TRIM13 (RFP2) downregulation decreases tumour cell growth in multiple myeloma through inhibition of NF kappa B pathway and proteasome activity. Br J Haematol. 2013;162(2):210-220. 24. Dutta-Simmons J, Zhang Y, Gorgun G, et al. Aurora kinase A is a target of Wnt/beta-catenin involved in multiple myeloma disease progression. Blood. 2009;114(13):2699-2708. 25. Mani M, Carrasco DE, Zhang Y, et al. BCL9 promotes tumor progression by conferring enhanced proliferative, metastatic, and angiogenic properties to cancer cells. Cancer Res. 2009;69(19):7577-7586. 26. Subramanian A, Tamayo P, Mootha VK, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005;102(43):15545-15550. 27. Manwani R, Cohen O, Sharpley F, et al. A prospective observational study of 915 patients with systemic AL amyloidosis treated with upfront bortezomib. Blood. 2019;134(25):2271-2280. 28. Gertz MA, Kyle RA, Greipp PR. The plasma cell labeling index: a valuable tool in primary systemic amyloidosis. Blood. 1989;74(3):1108-1111. 29. Witzig TE, Timm M, Larson D, et al. Measurement of apoptosis and proliferation of bone marrow plasma cells in patients with plasma cell proliferative disorders. Br J Haematol. 1999;104(1):131-137. 30. Diomede L, Rognoni P, Lavatelli F, et al. Investigating heartspecific toxicity of amyloidogenic immunoglobulin light chains: a lesson from C. elegans. Worm. 2014;3(3):e965590. 31. Zhou P, Ma X, Iyer L, et al. One siRNA pool targeting the lambda constant region stops lambda light-chain production and causes terminal endoplasmic reticulum stress. Blood. 2014;123(22):3440-3451. 32. Mishra S, Joshi S, Ward JE, et al. Zebrafish model of amyloid light chain cardiotoxicity: regeneration versus degeneration. Am J Physiol Heart Circ Physiol. 2019;316(5):H1158-H1166.

Haematologica | 108 December 2023

3371

ARTICLE - Plasma Cell Disorders

IL6Myc mouse is an immunocompetent model for the development of aggressive multiple myeloma Michael D. Pisano,1,2 Fumou Sun,3 Yan Cheng,3 Deepak Parashar,2 Vivian Zhou,2 Xuefang Jing,4 Ramakrishna Sompallae,5 Jenica Abrudan,6 Michael T. Zimmermann,6 Angela Mathison,6 Siegfried Janz2 and Miles A. Pufall7 1

Interdisciplinary Graduate Program in Immunology, University of Iowa, Iowa City, IA; 2Division of Hematology and Oncology, Department of Medicine, Medical College of Wisconsin, Milwaukee, WI; 3Myeloma Center, Department of Internal Medicine and Winthrop P. Rockefeller Cancer Institute, University of Arkansas for Medical Sciences, Little Rock, AR; 4Department of Pathology, University of Iowa Roy J. and Lucille A. Carver College of Medicine, Iowa City, IA; 5 Iowa Institute for Genetics, University of Iowa Roy J. and Lucille A. Carver College of Medicine, Iowa City, IA; 6Genomic Sciences and Precision Medicine Center, Medical College of Wisconsin, Milwaukee, WI and 7Department of Biochemistry and Molecular Biology, the University of Iowa Roy J. and Lucille A. Carver College of Medicine, Holden Comprehensive Cancer Center, Iowa City, IA, USA

Correspondence: M.A. Pufall mpufall@uiowa.edu Received: Accepted: Early view:

December 12, 2022. July 4, 2023. July 13, 2023.

Abstract Multiple Myeloma (MM) is a plasma cell neoplasm originating in the bone marrow and is the second most common blood cancer in the United States. One challenge in understanding the pathogenesis of MM and improving treatment is a lack of immunocompetent mouse models. We previously developed the IL6Myc mouse that generates plasmacytomas at 100% penetrance that phenotypically resemble aggressive MM. Using comprehensive genomic analysis, we found that the IL6Myc tumors resemble aggressive MM by RNA and protein expression. We also found that IL6Myc tumors accumulated fusions and missense mutations in genes that overlap significantly with human myeloma, indicating that the mouse is good model for studying disease etiology. Lastly, we derived cell lines from IL6Myc tumors that express cell surface markers typical of MM and readily engraft into mice, home to the bone marrow, and induce osteolytic disease. The cell lines may be useful in developing immunotherapies directed against BAFF-R and TACI, though not BCMA, and may also be a good model for studying dexamethasone resistance. These data indicate that the IL6Myc model is useful for studying development of aggressive MM and for developing new treatments against such forms of the disease.

Introduction Multiple myeloma (MM) is the second most common hematological malignancy in the United States. Although immunotherapies show promise in curing MM, most patients receive conventional treatments that are temporary in efficacy and subject to eventual relapse with more aggressive disease.1-3 Better treatments are urgently needed. One challenge in developing new treatments is the availability of MM models. Several mouse models have been developed, two of which have come into wide use, Vk*Myc and 5TMM,4,5 that faithfully recapitulate aspects of the disease. Both models are immunocompetent but are driven differently: Vk*Myc by relying on AID-dependent activation of Myc4 and 5TMM by relying on transfer of a 5T cell line to C57BL/KaLwRij mice.4,5 Both models have made important progress in addressing the role of the immune system in MM. 5TMM models have helped to elucidate the role of the Th1:Th2 axis in MM6 and anti-PD-L1

immunotherapy7 and Vk*Myc mice have elucidated roles of natural killer (NK) and CD8 T-cell surveillance8 and the role of interleukin 17 (IL17), and Th17 cells, in disease progression.9 The 5TMM models of MM have proven useful in studying bone marrow homing and myeloma-like bone disease (MBD).10 The Vk*Myc mouse has also proven useful in tracking gene expression changes that underlie the progression from monoclonal gammopathy of undetermined significance (MGUS) to MM and the accumulation of genetic lesions in the form of copy number variations (CNV) that are associated with disease and intratumoral heterogeneity.11 In addition to helping elucidate the etiology of MM, both models have been useful in developing treatments. In our laboratory we developed an immunocompetent model that appears to mimic aggressive or high-risk disease; the IL6Myc mouse. Global overexpression of IL6 is a proliferative signal that contributes to the differentiation of B cells into plasma cells. Myc is overexpressed by an

Haematologica | 108 December 2023

3372

ARTICLE - IL6Myc mouse models aggressive multiple myeloma iMycEμ transgene in B cells inserted upstream of the intronic Igh enhancer, which recapitulates the t(8;14)(q24;q32) translocation from Burkitt lymphoma. IL6Myc mice develop plasmacytomas quickly, in 45 to 175 days, with 100% penetrance, with average survival of 91.6 days.12 Importantly, IL6Myc plasmacytomas phenotypically resemble human MM, including a peripheral blood Mspike,12 lytic bone disease, and malignant plasma cell infiltration in the peripheral lymph nodes, kidney, and liver.13,14 The development of extramedullary disease in the spleen and mesenteric lymph nodes that is characteristic of the IL6Myc mouse is rare in MM, but is associated with MM patients that have progressed to plasma cell leukemia.15 M-spikes in IL6Myc mice are initially more oligoclonal, in contrast with monoclonal disease observed in human MM, but become less polyclonal as the mice age.13 The striking penetrance, consistency, and phenotypic resemblance to aggressive MM disease suggest that the IL6Myc mouse is a useful model for testing treatments and studying MM development. One model for MM development is Myc addiction. Myc variants occur in >40% of newly diagnosed myelomas, with Myc overexpressed in 67% of frank MM.16 This suggests that Myc expression sets the conditions for MM development as B cells become “addicted to Myc”.17 Myc overexpression alone is insufficient to drive myelomagenesis, indicating that other genetic lesions are required for transition to MM. Expression of IL6 and Myc may provide the conditions, as in the Vk*Myc mouse, to accumulate gene variants that drive development of aggressive MM.4 In order to molecularly characterize the IL6Myc mouse, and to determine which clinical subtype(s) IL6Myc most resembles we conducted comprehensive genomic profiling. In order to determine which malignancies IL6Myc tumors most resemble and to identify genes proteins that are potentially driving the disease, we performed RNA sequencing (RNA-seq) and reverse phase protein arrays (RPPA). In order to determine whether the IL6Myc mouse generates genetic lesions that are consistent with development of MM we analyzed RNA-seq data for gene fusions and performed whole-exome sequencing (WES) to identify protein-coding mutations. Lastly, inspired by the generation of 5T cells, we derived IL6Myc cell lines which we characterized for engraftment, homing, cell surface makers, and drug sensitivity to determine their suitability for developing models to test better treatments.

Methods Mice IL6Myc mice were bred by crossing BALB/c.IL6 and iMycEμ mice as described previously. Experiments conducted under MCW Animal Use Application #AUA00006541.

M.D. Pisano, et al.

RNA sequencing IL6Myc tumor, naïve, and activated B cells were isolated, RNA extracted (Qiagen RNeasy) from specimens, sequenced, processed (TrimGalore), and mapped (Salmon), gene set enrichment analysis (GSEA), and subjected to fusion gene analysis (STAR-Fusion). Principal component analysis (PCA), similarity, and differential gene expression (DESeq2) were conducted in RStudio 4.1.1 (see attached KnitedScript.pdf). Quantitative polymerase chain reaction (qPCR) was performed using iScript™ and SYBR® Green Master Mix (Bio-Rad). Reverse-phase proteomic array Forty-four mouse MLN tumors and three wild-type (WT) splenic B cells were isolated from BALB/c and analyzed by the MD Anderson Functional Proteomics RPPA Facility as described previously.18 Reverse-phase proteomic arra (RPPA) data was further analyzed in R. Whole-exome sequencing DNA was extracted from 45 IL6Myc MLN tumor samples and five BALB/c Kidney samples, exonic DNA captured (SureSelectXT Mouse All Exon), sequenced, and analyzed in R/Bioconductor. Overlap between sets of variants was assayed by hypergeometric test. Derivation of IL6Myc cell lines IL6Myc cell lines were derived from splenic (IL6Myc-1) and mesenteric lymph node (IL6Myc-2,3,4,5) tumors. Cells were single-cell cloned and cultured in IMDM supplemented plus 15% fetal bovine serum (FBS). Flow cytometry Cells were isolated from culture, washed, and blocked with 1 ng/μL Fc blocking antibody for 10 minutes (min), incubated with antibodies for 30 min, washed, resuspended in phosphate-buffered saline (PBS) with 2% FBS and run on a BD Accuri™ C6 Plus Flow Cytometer. Data were analyzed using FlowJo. X-ray microscopy and microtomography imaging of mice with IL6Myc-1 adoptive transfer IL6Myc-1 cells were engrafted by adoptive transfer. For Xray microscopy and X-ray microtomography (μCT) imaging mouse carcasses were fixed in formalin, fit in tubes, and imaged using dual-energy 3D X-ray microscopy (Zeiss Xradia 520 Versa). Projections were collected through 360° rotations. Images were analyzed using the BoneJ (v. 1.3.14, ImageJ v. 1.49 m). Histological analysis and tartrate-resistant acid phosphatase staining Bone samples were fixed in formaldehyde, decalcified, rinsed with PBS, fixed in cold 70% ethanol, and embedded

Haematologica | 108 December 2023

3373

ARTICLE - IL6Myc mouse models aggressive multiple myeloma in paraffin. Tissue sections were stained for tartrate-resistant acid phosphatase (TRAP) activity using an acid phosphatase leukocyte kit (Sigma). Specimens were counterstained with eosin and mounted for imaging and histomorphometry. Osteoclasts were enumerated in five high-power microscopy fields per slide.

M.D. Pisano, et al.

IL6Myc tumors (SPL, MLN) to activated B cells, which are similar to normal plasma cells. About half of all expressed genes (9,713 vs. 18,551 total) were significantly differentially expressed (q≤0.05) in tumors compared to controls and strongly skewed toward increased expression (6,411 up vs. 3,302 down) (Figure 1C; IL6Myc_v_activated_b_DGE.xlsx). This pattern resembles other Myc overexpression studies19 BAFF and APRIL stimulation and is indicative of a global change in gene expression. 5 Cells were harvested from culture, seeded at 1x10 cells Under these circumstances, differential expression of inper well in 100 uL IMDM plus 1% FBS and treated with ve- dividual genes could not be confidently discerned by conhicle (1% FBS IMDM), 100 ng/mL APRIL, or 100 ng/mL BAFF ventional methods. Instead, we performed GSEA to for 72 hours (h). Viability was measured by incubating with identify patterns and genes and reverse-phase protein MTS reagent and read at 490 nm (SpectraMax i3x). (RPPA) to validate misexpression of cancer-related proteins. Drug treatment Cells were seeded in IMDM plus 15% FBS in 96-well plates. IL6Myc gene expression pattern is consistent with Stocks of drugs were diluted, titrated into wells, and in- high-risk disease cubated for 24-72 h. Viability was measured by Promega In order to identify sets of genes associated with transReal-Time Glo reagent with fluorescence (SpectraMax i3x). formation and MM we performed GSEA, which is less senData was analyzed using GraphPad Prism. sitive to global changes because it compares rank order instead of expression levels of genes. Of the top 100 enriched gene sets (by normalized enrichment score) generated by comparing IL6Myc tumor and activated B cells, Results solid cancers (n=22), hematological malignancies (n=13), Gene expression patterns of IL6Myc tumors are similar extracellular matrix gene sets (n=16), and stemness (n=10) to human multiple myeloma were highly represented. Of solid cancers, gene sets from In order to determine whether the tumors molecularly re- prostate and breast cancer were the most represented semble human MM, we performed RNA-seq. PCA compar- (Figure 1D; Online Supplementary Table S1). In order to ing gene level counts of seven splenic (SPL) and seven identify which genes contribute to enrichment of these mesenteric lymph node (MLN) tumors to five lipolysac- top gene sets, we performed leading edge analysis (Online charide (LPS)-activated (plasmablast) and four quiescent Supplementary Figure S3A, B). This identified 274 genes BALB/c B-cell controls showed that the tumors were dis- that drove enrichment of >5 gene sets (Online Supplementinct from controls but did not cluster by mouse or dis- tary Table S2). We validated three of these genes by qPCR ease location (Figure 1A). PCA followed by centroid due to their association with oncogenesis in previous litanalysis of human myeloma, diffuse large B-cell lym- erature: CLU,20 CTSB,21 and TGFBI.22 Although each was upphoma (DLBCL), acute myeloid leukemia, acute lympho- regulated as expected, none were significant due to high blastic leukemia (ALL) and Burkitt lymphoma, as well as between sample variance (Online Supplementary Figure prostate cancer and breast cancer downloaded from the S3C). Cancer Genome Atlas (TCGA), indicates that IL6Myc tu- Three MM gene sets were highly enriched; mors most closely resemble MM (Figure 1B). Single-linkage ZHAN_MULTIPLE_MYELOMA_MS_UP gene set, which is hierarchical clustering (Online Supplementary Figure S1) associated with high-risk MMSET (multiple myeloma indicated that IL6Myc tumors were most similar to DLBCL, SET domain)-activating translocations, 23 with human MM next. These results indicate that IL6Myc ZHAN_MULTIPLE_MYELOMA_CD1_DN, enrichment of tumors resemble mature B-cell malignancies including which means IL6Myc is dissimilar to the low-risk CD-1 MM. In order to determine whether IL6Myc tumors re- subtype; and ZHAN_MULTIPLE_MYELOMA_DOWN (Figsemble a subtype of MM, we compared IL6Myc tumors to ure 1E). In order to validate these genes sets we pertranslocation, disease stage and outcome subsets of formed qPCR in seven MLN tumors versus activated B-cell human MM using RNA-seq data from the Multiple Mye- controls on the top two genes for each set identified by loma Research Foundation (MMRF) CoMMpass database. leading-edge analysis (Online Supplementary Figure S3D). Although IL6Myc tumors clustered with MM, they did not qPCR validated upregulation of CTS3 and MITF for the resemble any specific subtype (Online Supplementary Fig- ZHAN_MULTIPLE_MYELOMA_MS_UP gene set. Although ure S2). little is known about the role of either gene in MM, the In order to determine which genes are differentially ex- IL6Myc model provides a system to test their role in myepressed in tumors we compared expression levels in all lomagenesis. However, neither gene (DOK3, HES1) for Haematologica | 108 December 2023

3374

ARTICLE - IL6Myc mouse models aggressive multiple myeloma

M.D. Pisano, et al.

Continued on following page. Haematologica | 108 December 2023

3375

ARTICLE - IL6Myc mouse models aggressive multiple myeloma

M.D. Pisano, et al.

Figure 1. IL6Myc tumors resemble human multiple myleloma. (A) Gene expression was measured by RNA sequencing (RNA-seq) for IL6Myc tumors and activated and quiescent wild-type (WT) B cells from BALB/c mice. Principle component analysis (PCA) was performed to determine the similarity between tumor origin (mesenteric lymph nodes or spleen) and type of B cell (activated or quiescent). (B) Comparison of the tumor types using PCA of RNA-seq gene expression data. Density plots for the PC of each tumor type are plotted on the X and Y axis. Centroids and error bars are indicated in tan, with a table indicating the distance between IL6Myc centroid and each cancer. (C) Volcano plot of differential expression comparing mesenteric lymph node tumor to activated B-cell (plasmablast) control by RNA-seq. Red indicates genes that are significantly (q≤0.05) differentially expressed, with no log2 fold change cutoff. (D) Summary of the top 100 enriched gene sets from gene set enrichment analysis (GSEA) of IL6Myc tumors compared to activated B cells. (E) Top multiple myeloma gene sets from GSEA for IL6Myc tumor versus activated B cells. (F) Volcano plot of differential protein expression from RPPA of tumor versus normal B cells. Red; P<0.01, no log2 fold change cutoff. (G) Correlation of protein (RPPA) and RNA (RNA-seq) log2 fold change for significantly differentially expressed genes/proteins comparing tumor versus normal B cells. Blue line, best fit linear regression (r=0.27; P≤0.001). ALL: acute lymphobalstic leukemia; AML: acute myeloid leukemia; MM: multiple myeloma; DLBCL: diffuse large B-cell lymphoma.

ZHAN_MULTIPLE_MYELOMA_CD1_DN and only ITGA2B for the ZHAN_MULTIPLE_MYELOMA_DOWN were upregulated as expected. These data support that the gene expression patterns of IL6Myc tumors resemble MMSET and high-risk MM but lack other defining expression features. In order to determine which oncoproteins are misexpressed in IL6Myc tumors, and to validate expression by RNA-seq, we performed RPPA. When IL6Myc tumors were compared to WT splenic B cells isolated from BALB/c mice, 144 of the 333 proteins tested were differentially expressed (P<0.01) (Figure 1F). In order to validate RNA-seq expression changes with RPPA, we performed linear regression analysis. This showed a modest, though significant correlation between expression patterns (r=0.27; P≤0.001), with the expression a substantial subset correlating well (Figure 1G). Among correlating genes, three associated with MM were notable: EPHA2 and CD274 (upregulated) and STAT5A (downregulated) (Figure 1G). EPHA2 and CD274 (PD-L1) are both targets of anti-MM therapy. Anti-EPHA2 compounds are in clinical trials24 and PD-L1 is a target for immunotherapy due to its success in other malignancies and elevated expression in relapse-refractory patients.25 STAT5A decreases proliferation of MM cells and is a target for degradation of YTHDF2 which is upregulated in MM patients with poor outcomes.26 These results show that IL6Myc tumors have features of poor prognosis MM and perhaps immune evasion.

with normal B cells having fewer Ig-coding fusions per sample (Figure 2A; Online Supplementary Table S3). A significant (P<0.002) number (n=10) of the IL6Myc genes involved in fusions overlapped with genes involved in fusions identified in human MM from the MMRF data set (Figure 2B). One fusion, Snd1-Braf, was between the same two genes and is associated with several cancers28-30 (Online Supplementary Table S3). A more in-depth examination of fusions shared between IL6Myc and human MM may reveal a role in myelomagenesis. In order to determine whether IL6Myc mice accumulate missense mutations in genes associated with Myc-driven MM, whole-exome sequencing (WES) was performed on 45 MLN whole-tumor samples. In order to eliminate background variation, WES was performed on five WT kidney samples from BALB/c mice and identified variants were subtracted. We eliminated other common variants using single-nucleotide polymorphism from the Mouse Genome Project. In order to ensure high-confidence variants that result in protein mutation, we considered only missense and nonsense mutations and used a baseline 5% allelic frequency, leaving 1,351 variant-containing genes (Figure 2C; Online Supplementary Table S4). This IL6Myc gene list overlapped significantly (324 of 1,351 genes; P<6.4x10-22) with the most common human variants (top 1% of genes >10 variants human MM from the MMRF database; Figure 2D). Perhaps more importantly, a significant (P<0.0003) number of IL6Myc variant-containing genes (n=12) overlapped with the 80 genes with known driver mutations in human MM.31 This list included PIM1, which accumulated mutations in the kinase domain of the PIM1 gene in both IL6Myc tumors and in human MM (Figure 2E). Overall, these data show that IL6Myc tumors exhibit an accumulation of variants in many of the same genes with genetic lesions in human MM, suggesting that the IL6Myc mouse is able to acquire the many of same mutations that drive human MM.

Genes with fusions and mutations in IL6Myc tumors overlap with those in human multiple myeloma The penetrance of MM in the IL6Myc mouse suggests that driving IL6 and Myc creates conditions for genetic lesions to arise that drive myelomagenesis. Because recurrent Ig fusions with non-Ig protein coding genes are a common feature of MM,27 we used STAR-Fusion on the tumor RNAseq datasets to identify 78 fusion genes in IL6Myc tumors. Excluding the intra-Ig rearrangements typical of B-lineage cells, 30 genes were involved in coding-coding (non-Ig protein-coding genes) fusions and the remaining ten were Gene expression of IL6Myc cell lines resemble human between Ig loci and other protein-coding genes. This ratio multiple myeloma of coding-coding and Ig-coding is similar to human MM, We derived five cell lines from IL6Myc splenic (IL6Myc-1) Haematologica | 108 December 2023

3376

ARTICLE - IL6Myc mouse models aggressive multiple myeloma and mesenteric lymph node (IL6Myc2-5) tumors. Cell lines adapted readily to culture, with similar doubling times ranging from 33 (IL6Myc-3) to 48 h (IL6Myc-2,5) (Figure 3A; Online Supplementary Figure S4A). In order to determine their similarity to primary tumors and human MM, we performed RNA-seq and RPPA. The RNA expression profiles in IL6Myc cell lines were similar to each other but distinct from primary tumor and normal primary B cells

M.D. Pisano, et al.

(Online Supplementary Figure S4B, C). However, comparison of the top 100 GSEA enriched sets between primary tumor and cell lines identified 17 overlapping gene sets (Online Supplementary Table S5) suggesting that although some expression patterns are maintained, IL6Myc cell lines and primary tumors are transcriptionally distinct. Protein expression levels were more similar between IL6Myc tumors and cell lines (Online Supplementary Figure

Figure 2. Genetic lesions in IL6Myc tumors significantly overlap with those observed in human multiple myeloma. (A) Gene fusions were identified in IL6Myc and wild-type (WT) B cells from RNA-sequencing (RNA-seq) data. Excluding inter-immunglobulin (Ig) rearrangements, most fusions were identified between non-Ig protein coding genes (coding-coding). Different from human multiple myeloma (MM), most Ig fusions were between the IgH locus and non-Ig protein coding genes. Blue: WT BALB/c B cells. Red: isolated B cells from IL6Myc tumors. Blue: fusions identified MM from RNA-seq data (ref). (B) Overlap of genes involved in fusions between IL6Myc MM (Multiple Myeloma Research Foundation [MMRF] data) is significant (P≤0.002). (C) Process for filtering variants. (D) Filtered variants identified by whole-exome sequencing in in IL6Myc (N=1351) overlap significantly (N=324; P<6.4e-22) with the top 1% most common variants identified from the MMRF database. (E) Variants from IL6Myc tumors (top) and the MMRF data set (bottom) were mapped onto the PIM1 gene. All variants localize to the N-terminal half. Statistical significance of overlap was determined by hypergeometric test (http://nemates.org/MA/progs/overlap_stats.html). Haematologica | 108 December 2023

3377

ARTICLE - IL6Myc mouse models aggressive multiple myeloma

M.D. Pisano, et al.

Continued on following page. Haematologica | 108 December 2023

3378

ARTICLE - IL6Myc mouse models aggressive multiple myeloma

M.D. Pisano, et al.

Figure 3. IL6Myc cell lines are IgA producing and have cell surface markers largely resembling human multiple myeloma. (A) Proliferation of IL6Myc cell lines measured by cell viability at 3 time points. (B) Immunglobulin (Ig) isotyping enzyme-linked immunorbant assay heatmap. Red indicates a higher signal, and green a weaker signal. Positive control Ig and negative control are indicated as “positive” and “negative”. Flow cytometry for the faction of cells staining positive for: (C) markers of clinical multiple myeloma (MM). Data was gathered in triplicate and error bars represent standard deviation. (D) Markers correlating with disease state in MM and/or those with value as immunotherapeutic targets. Data was gathered in triplicate and error bars represent standard deviation. (E) Survival and proliferation cell surface proteins in cell lines. Data was gathered in triplicate and error bars represent standard deviation. (F) Survival and proliferation cell surface proteins in mouse spleen. (G) Survival and proliferation cell surface proteins in bone marrow. (H) Growth of IL6Myc cell lines after 72 hours with addition of APRIL or BAFF compared to negative control. Significance was determined by t test. *P<0.05; ***P<0.0001. OD: optical density.

S4D), indicating that the cells lines are similar to the tumor. IL6Myc cell lines also most closely resembled human MM compared to other cancers (Online Supplementary Figure S4E, F). This indicates that despite differences in expression, the IL6Myc cell lines are similar to IL6Myc primary tumors and still model human MM. IL6Myc cell lines resemble malignant plasma cells by surface markers In order to subtype the IL6Myc cell lines we evaluated Ig isotype expression and cell surface markers. IL6Myc cell lines production of IgA antibodies outstripped other isotypes (Figure 3B). IgA-producing MM is the second most common form of MM, comprising approximately 20% of human MM and is associated with worse survival compared to the more common IgG-producing MM.32 IL6Myc cell lines were consistent with plasma cells, exhibiting surface expression of high CD138, CD54 and SLAMF7, low CD20,33 and were negative for CD19 and B220. However, the cell lines lacked CD56, a marker of human malignant plasma cells (Figure 3C). Positive staining of CD24, a marker of MM stemness,34 combined with CD81+CD117-, which correlates with high-risk disease, suggests that, despite the lack of CD56, IL6Myc cells resemble high-risk MM.35 In order to evaluate whether IL6Myc cell lines contain targets for existing chimeric antigen receptor (CAR) T cells, we performed flow cytometry for CD48, NKG2D, and BCMA (Figure 3D, E). Variable, but positive, staining was observed for CD48 and NKG2D, but negative for BCMA. This contrasts with primary IL6Myc splenic tumors and bone marrow from IL6Myc mice which stained positive for cell surface BCMA (Figure 3F). It also contrasts with RNA-seq from the IL6Myc cell lines which shows low, but significant expression of BCMA at the mRNA level (Online Supplementary Figure S5), indicating that BCMA expression is post-transcriptionally regulated in IL6Myc cell lines. Thus, although tumors in intact IL6Myc mice would be useful for studying BCMA CART therapies the cell lines are likely not (Figure 3F, G). BCMA, BAFF-R, and TACI provide important survival and proliferative messages to B lineage cells, have shared ligands, and signal through similar pathways. BAFF-R is typically expressed early in plasma cell development, and then lost in favor of BCMA and TACI.36,37 Primary IL6Myc tumors from spleen and bone marrow (Figure 3F, G) express

all three, to varying degrees, but IL6Myc cell lines retaining only BAFF-R and TACI (Figure 3E). Although BAFF-R positivity has been noted in some subsets of MM, it is not a common feature.38 These data suggest that BAFF-R and TACI may provide important survival and growth signals to IL6Myc cell lines. In order to test this, cell lines were stimulated with BAFF, a ligand of BAFF-R and TACI, or APRIL, a ligand of TACI and BCMA. Although BAFF ligand significantly (P<0.05) stimulated growth of three of the five cell lines (IL6Myc1, 2, and 5) and APRIL only one (IL6Myc-1), a trend toward increased growth is evident for both ligands in all cell lines (Figure 3H). These results indicate that BAFF-R, and sometimes TACI, stimulate enhanced growth of IL6Myc cell lines while BCMA does not. These results also indicate that IL6Myc cell lines would be useful for testing the efficacy of CAR T cells targeting TACI and/or BAFF-R, which has been evaluated in conjunction with BCMA targeting in MM.39 These targets may become more relevant in patients who relapse after BCMA targeted therapy and the approximately 5.7% of patients that are BCMA negative.40 IL6Myc-1 cells home to the bone marrow and induce osteolytic disease Particularly given their derivation (from splenic and mesenteric tumors), we sought to determine whether IL6Myc cell lines would serve as useful models by engrafting them into both immunocompromised (NOD-SCID) and immunocompetent (BALB/c) mice. We injected 2x106 cells intravenously (IV) into BALB/c mice following priming by lethal whole-body irradiation (10 Gy). Lethal host priming required hematopoietic stem cell rescue using bone marrow transplantation (3x106 cells). We also injected 2x106 cells IV into immunocompromised NOD-SCID mice without irradiation (Figure 4A). Several host mice developed hind limb paralysis suggesting plasma cell infiltration of the spine. Histopathological analysis of the spine showed extramedullary tumor growth that damaged spinal nerve roots and intramedullary tumor growth that compressed the spinal cord directly (Figure 4B). In order to test whether these mice develop myeloma-like bone disease (MBD), we completed μCT analysis. This demonstrated severe bone loss in L2-4 spine in both BALB/c and NODSCID mice (Figure 4C, D). Changes in serum biomarkers of

Haematologica | 108 December 2023

3379

ARTICLE - IL6Myc mouse models aggressive multiple myeloma

M.D. Pisano, et al.

Figure 4. Osteolytic disease in IL6Myc-1 autografted host Study design. mice. (A) IL6Myc-1 cells were derived from the spleen of a PCT-bearing IL6Myc mouse. Two million tumor cells were transferred intravenously (IV) to lethally irradiated BLAB/c mice rescued by co-transfer of hematopoietic stem cells obtained from untreated BALB/c mice (top). Adoptive transfer of IL6Myc-1 cells to untreated NOD-SCID mice was used as control (bottom). Tumor manifestation required ~3 months in BLAB/c and ~3 weeks NODSCID mice. More than half of the mice in both groups developed weakness or full-fledged paralysis of hind limbs indicated by 2 photographs on the right. (B) Two types of tumor growth (yellow asterisks) may underlie hind leg paralysis. Extramedullary tumor expansion (center panel) can damage spinal nerves and nerve roots but leaves vertebral bone (containing normal bone marrow) and spinal cord intact (yellow arrows). Intramedullary tumor expansion (right) can compress the spinal cord after destroying bone and infiltrating bone marrow (yellow arrows) but leaves spinal nerves (yellow arrowheads) unscathed. Transversal tissue sections stained according to hematoxylin & eosin in all 3 cases, including the normal mouse shown in the left panel. Black horizontal lines denote a scale of 2 mm. (C) Representative transversal X-ray mircotomography (μCT) images of the proximal femur from tumor-bearing NOD-SCID or BALB/c mice adoptively transferred with IL6Myc-1 cells. A normal femur used as control is shown for comparison. (D) Bar diagrams indicating bone loss in L2-4 spine of adoptively transferred BALB/c mice (blue) and NOD-SCID mice (red) compared to normal mice (white). (E) Bar diagrams indicating changes in serum biomarkers of osteolytic disease. Mean values and standard deviation of the mean are plotted (*P<0.05; **P<0.01; ***P< 0.001). ELISA: enzyme-linked immunorbant assay; n.s.: not significant.

Haematologica | 108 December 2023

3380

ARTICLE - IL6Myc mouse models aggressive multiple myeloma

M.D. Pisano, et al.

osteolytic disease were also higher in NOD-SCID than BALB/c hosts, e.g., soluble RANKL was twice as high (324±78.5 vs. 152±59.2 pg/mL) and OPG was twice as low (2.65±0.972 vs. 5.73±1.60 ng/mL). Soluble TRAP was also highest in NOD-SCID (28.5±9.76 U/L) followed by BALB/c (20.4±5.38 U/L) with both significantly elevated compared to controls (6.87±7.24 U/L) (Figure 4E). These results suggest that IL6Myc-1 cells result in MM-like disease like that observed in transgenic mice and, importantly, homes to typical sites of MM in the bone marrow.

doses in three of five cell lines (Figure 5; Online Supplementary Figure S6B, G, H). The cell lines varied in their sensitivities to each drug, but largely followed similar drug response patterns. Resistance to pomalidomide treatment was expected due to its reduced activity in mice, but resistance to dexamethasone indicates that the IL6Myc cell lines are associated with aggressive, poor prognosis MM.41

IL6Myc cell lines are resistant to dexamethasone In order to determine how IL6Myc cell lines respond to commonly used MM pharmacological agents, we treated cell lines with bortezomib, dexamethasone, panobinostat, melphalan, and 4-hydroxyperoxy cyclophosphamide (activated cyclophosphamide). All cell lines were sensitive to bortezomib (Figure 5; Online Supplementary Figure S6A, H) and panobinostat (Figure 5; Online Supplementary Figure S6D, H), but only died with the highest concentrations of pomalidomide (Figure 5; Online Supplementary Figure S6C, H), melphalan (Figure 5; Online Supplementary Figure S6E, H) and 4-hydroxyperoxy cyclophosphamide (Figure 5; Online Supplementary Figure S6F, H). Cell lines were insensitive to dexamethasone, with >50% of cells surviving the highest

The results of comprehensive genomic analysis reflect the phenotypic observation that the IL6Myc mouse develops high-risk, aggressive MM. Phenotypically, the tumors in IL6Myc mice resemble human MM including bone marrow infiltrating malignant plasma cells, MM-like bone disease, serum M-spike, and spleen, MLN, peripheral lymph nodes, kidney, and liver involvement.12-14 Although RNA-seq data was challenging to interpret, there was some evidence from GSEA that IL6Myc tumors also resemble high-risk MM through enrichment of the high-risk MMSET gene set23 (Figure 1). Protein expression analyses provided further evidence of high-risk disease including exclusive expression of IgA32 (Figure 3) reduced expression of STAT5 and increased expression of PD-L1, both of which are linked to increased dis-

Discussion

Figure 5. IL6Myc response to anti-multiple myeloma drugs. Summary of area under the curve (AUC) values of 4-hydroxyperoxy cyclophosphamide (4-HP-Cyclophosphamide), bortezomib, dexamethasone, melphalan, panobinostat, and pomalidomide. Data was collected in triplicate and error bars represent standard deviation. Haematologica | 108 December 2023

3381

ARTICLE - IL6Myc mouse models aggressive multiple myeloma ease severity in human MM25,26 (Figure 1). High-risk disease was also indicated by CD24, a marker of stemness,34 and by CD81+CD117-, a diagnostic marker of poor prognosis MM35 (Figure 3). Thus, the IL6Myc mouse tumors and cell lines have many of the molecular hallmarks of high-risk or aggressive MM. The IL6Myc mouse is also a promising model for understanding the development of MM. Our data show an accumulation of variants and fusions in IL6Myc tumor genes that overlap significantly with genes, including transforming genes, with genetic lesions in human MM (Figure 2). This strongly suggests that IL6Myc mice favor acquisition of transforming mutations. The common location of IL6Myc and human MM variants in the kinase domain of PIM1 was particularly striking. PIM kinase inhibitors have shown promise in treatment of MM with several open clinical trials including in relapse/refractory MM.42 Although PIM2 is often upregulated in MM, inhibitors are pan-PIM kinase42 and may hit other PIM kinases. The IL6Myc mouse is a useful model to test the role of PIM1 and other PIM kinases in myelomagenesis and treatment. IL6Myc mice are also useful for developing immunotherapies against MM. The tumors of intact mice display BCMA and are, therefore, useful in testing CAR T cells directed against this target. The IL6Myc cell lines do not display BCMA and are thus less useful in this regard. However, IL6Myc cell lines allow study of how the BCMA, TACI and BAFF-R axis can be exploited to better treat MM. These receptors provide important proliferative and survival signals to MM cells. Although BCMA and TACI positivity is found in most forms of MM, BAFF-R positivity is less common. IL6Myc cell lines expressed cell surface TACI and surprisingly, BAFF-R. Stimulation experiments confirmed that BAFF-R, and likely TACI, can stimulate growth in IL6Myc cell lines while BCMA does not (Figure 3). This pattern mirrors non-remission patients who also express BAFF-R and TACI.43 This expression pattern accommodates the high circulating BAFF levels found in MM that tend to increase with disease progression.44 Immuno-targeting of BAFF-R and TACI would therefore be useful in BAFF-R+TACI+ MM. Indeed, recently developed CAR T cells displaying BAFF ligand can kill MM cells displaying BAFF-R, TACI, or BCMA.39 Because of a paucity of BAFF-R-positive cell lines, IL6Myc cell lines and primary tu-

M.D. Pisano, et al.

mors are uniquely poised to the examine the role of BAFF-R in MM both in vitro and, due to their homing patterns (Figure 4), in vivo. Further studies of these receptors in IL6Myc may elucidate novel therapies on this axis that otherwise would be difficult to address in other MM cell lines. Disclosures No conflicts of interest to disclose. Contributions MDP was involved in research design, collected, and analyzed data, and assisted in writing the paper. FS, YC, VZ and XJ collected and analyzed data, and derived IL6Myc cell lines. RS, JA, MTZ and AM analyzed data, performed bioinformatics work, and assisted in writing the paper. SJ was involved in research design, analyzed data, and helped to write the paper. DP helped to analyze data and to write the paper. MP oversaw the work, analyzed data, and helped to design and to write the paper. Acknowledgements Thank you to the Linda T. and John A. Mellowes Center for Genomic Sciences and Precision Medicine at the Medical College of Wisconsin for help with STAR-fusion and cell line RNA sequencing. Thank you to the University of Iowa Genomics division for primary tumor RNA sequencing. Thank you to Dr. Michael Tomasson, who helped us to get to core questions in the manuscript, and to Dr. Gail Bishop who provided support reviewing the paper. Funding SJ was supported by NCI R01CA151354; the William G. Schuett, Jr., Multiple Myeloma Research Endowment; and a generous gift from the Riney Family Multiple Myeloma Research Initiative. MAP was supported by a grant from the Aiming for a Cure Foundation. Data-sharing statement The authors are committed to open sharing of data. All sequencing data has been deposited in the Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo/, accession number: GSE220828). All other data will be provided upon request to the corresponding author.

References 1. Majithia N, Rajkumar SV, Lacy MQ, et al. Early relapse following initial therapy for multiple myeloma predicts poor outcomes in the era of novel agents. Leukemia. 2016;30(11):2208-2213. 2. Teoh PJ, Chng WJ. CAR T-cell therapy in multiple myeloma: more room for improvement. Blood Cancer J. 2021;11(4):84. 3. Yamamoto L, Amodio N, Gulla A, Anderson KC. Harnessing the immune system against multiple myeloma: challenges and opportunities. Front Oncol. 2020;10:606368. 4. Chesi M, Robbiani DF, Sebag M, et al. AID-dependent activation of

a MYC transgene induces multiple myeloma in a conditional mouse model of post-germinal center malignancies. Cancer Cell. 2008;13(2):167-180. 5. Croese JW, Vas Nunes CM, Radl J, van den Enden-Vieveen MH, Brondijk RJ, Boersma WJ. The 5T2 mouse multiple myeloma model: characterization of 5T2 cells within the bone marrow. Br J Cancer. 1987;56(5):555-560. 6. Hong S, Qian J, Yang J, Li H, Kwak LW, Yi Q. Roles of idiotypespecific t cells in myeloma cell growth and survival: Th1 and CTL

Haematologica | 108 December 2023

3382

ARTICLE - IL6Myc mouse models aggressive multiple myeloma cells are tumoricidal while Th2 cells promote tumor growth. Cancer Res. 2008;68(20):8456-8464. 7. Hallett WH, Jing W, Drobyski WR, Johnson BD. Immunosuppressive effects of multiple myeloma are overcome by PD-L1 blockade. Biol Blood Marrow Transplant. 2011;17(8):1133-1145. 8. Guillerey C, Ferrari de Andrade L, Vuckovic S, et al. Immunosurveillance and therapy of multiple myeloma are CD226 dependent. J Clin Invest. 2015;125(5):2077-2089. 9. Calcinotto A, Brevi A, Chesi M, et al. Microbiota-driven interleukin17-producing cells and eosinophils synergize to accelerate multiple myeloma progression. Nat Commun. 2018;9(1):4832. 10. Vanderkerken K, Asosingh K, Croucher P, Van Camp B. Multiple myeloma biology: lessons from the 5TMM models. Immunol Rev. 2003;194:196-206. 11. Croucher DC, Richards LM, Tsofack SP, et al. Longitudinal singlecell analysis of a myeloma mouse model identifies subclonal molecular programs associated with progression. Nat Commun. 2021;12(1):6322. 12. Rutsch S, Neppalli VT, Shin DM, et al. IL-6 and MYC collaborate in plasma cell tumor formation in mice. Blood. 2010;115(9):1746-1754. 13. Duncan K, Rosean TR, Tompkins VS, et al. (18)F-FDG-PET/CT imaging in an IL-6- and MYC-driven mouse model of human multiple myeloma affords objective evaluation of plasma cell tumor progression and therapeutic response to the proteasome inhibitor ixazomib. Blood Cancer J. 2013;3(11):e165. 14. Sun F, Cheng Y, Walsh SA, et al Osteolytic disease in IL-6 and Myc dependent mouse model of human myeloma. Haematologica. 2020;105(3):e111-e115. 15. Sher T, Miller KC, Deeb G, Lee K, Chanan-Khan A. Plasma cell leukaemia and other aggressive plasma cell malignancies. Br J Haematol. 2010;150(4):418-427. 16. Chng WJ, Huang GF, Chung TH, et al. Clinical and biological implications of MYC activation: a common difference between MGUS and newly diagnosed multiple myeloma. Leukemia. 2011;25(6):1026-1035. 17. Holien T, Vatsveen TK, Hella H, Waage A, Sundan A. Addiction to cMYC in multiple myeloma. Blood. 2012;120(12):2450-2453. 18. Parashar D, Geethadevi A, Aure MR, et al. miRNA551b-3p activates an oncostatin Signaling module for the progression of triplenegative breast cancer. Cell Rep. 2019;29(13):4389-4406. 19. Nie Z, Hu G, Wei G, et al. c-Myc is a universal amplifier of expressed genes in lymphocytes and embryonic stem cells. Cell. 2012;151(1):68-79. 20. Ming X, Bao C, Hong T, et al. Clusterin, a novel DEC1 target, modulates DNA damage-mediated cell death. Mol Cancer Res. 2018;16(11):1641-1651. 21. Cheriyath V, Kuhns MA, Kalaycio ME, Borden EC. Potentiation of apoptosis by histone deacetylase inhibitors and doxorubicin combination: cytoplasmic cathepsin B as a mediator of apoptosis in multiple myeloma. Br J Cancer. 2011;104(6):957-967. 22. Kaiser MF, Johnson DC, Wu P, et al. Global methylation analysis identifies prognostically important epigenetically inactivated tumor suppressor genes in multiple myeloma. Blood. 2013;122(2):219-226. 23. Zhan F, Huang Y, Colla S, et al. The molecular classification of multiple myeloma. Blood. 2006;108(6):2020-2028. 24. Wilson K, Shiuan E, Brantley-Sieders DM. Oncogenic functions and therapeutic targeting of EphA2 in cancer. Oncogene. 2021;40(14):2483-2495. 25. Tamura H, Ishibashi M, Sunakawa-Kii M, Inokuchi K. PD-L1-PD-1 Pathway in the Pathophysiology of Multiple Myeloma. Cancers (Basel). 2020;12(4):924. 26. Hua Z, Wei R, Guo M, et al. YTHDF2 promotes multiple myeloma

M.D. Pisano, et al.

cell proliferation via STAT5A/MAP2K2/p-ERK axis. Oncogene. 2022;41(10):1482-1491. 27. Cleynen A, Szalat R, Kemal Samur M, et al. Expressed fusion gene landscape and its impact in multiple myeloma. Nat Commun. 2017;8(1):1893. 28. Chu YH, Wirth LJ, Farahani AA, et al Clinicopathologic features of kinase fusion-related thyroid carcinomas: an integrative analysis with molecular characterization. Mod Pathol. 2020;33(12):2458-2472. 29. Jang JS, Lee A, Li J, et al. Common oncogene mutations and novel SND1-BRAF transcript fusion in lung adenocarcinoma from never smokers. Sci Rep. 2015;5:9755. 30. Klubickova N, Agaimy A, Hajkova V, et al. RNA-sequencing of myxoinflammatory fibroblastic sarcomas reveals a novel SND1::BRAF fusion and 3 different molecular aberrations with the potential to upregulate the TEAD1 gene including SEC23IP::VGLL3 and TEAD1::MRTFB gene fusions. Virchows Arch. 2022;481(4):613-620. 31. Oben B, Froyen G, Maclachlan KH, et al. Whole-genome sequencing reveals progressive versus stable myeloma precursor conditions as two distinct entities. Nat Commun. 2021;12(1):1861. 32. Nair B, Waheed S, Szymonifka J, Shaughnessy JD, Jr., Crowley J, Barlogie B. Immunoglobulin isotypes in multiple myeloma: laboratory correlates and prognostic implications in total therapy protocols. Br J Haematol. 2009;145(1):134-137. 33. Ajaz B, Akhtar A, Chang CC, Solh M, Tangonan K, Khaled Y. Plasma cell CD20 expression: primary aberrant expression or receptor upregulation. Leuk Lymphoma. 2014;55(2):444-446. 34. Gao M, Bai H, Jethava Y, et al. Identification and characterization of tumor-initiating cells in multiple myeloma. J Natl Cancer Inst. 2020;112(5):507-515. 35. Chen F, Hu Y, Wang X, Fu S, Liu Z, Zhang J. Expression of CD81 and CD117 in plasma cell myeloma and the relationship to prognosis. Cancer Med. 2018;7(12):5920-5927. 36. Kampa M, Notas G, Stathopoulos EN, Tsapis A, Castanas E. The TNFSF Members APRIL and BAFF and their receptors TACI, BCMA, and BAFFR in oncology, with a special focus in breast cancer. Front Oncol. 2020;10:827. 37. Lee L, Bounds D, Paterson J, et al. Evaluation of B cell maturation antigen as a target for antibody drug conjugate mediated cytotoxicity in multiple myeloma. Br J Haematol. 2016;174(6):911-22. 38. Novak AJ, Darce JR, Arendt BK, et al. Expression of BCMA, TACI, and BAFF-R in multiple myeloma: a mechanism for growth and survival. Blood. 2004;103(2):689-694. 39. Wong DP, Roy NK, Zhang K, et al. A BAFF ligand-based CAR-T cell targeting three receptors and multiple B cell cancers. Nat Commun. 2022;13(1):217. 40. Salem DA, Maric I, Yuan CM, et al. Quantification of B-cell maturation antigen, a target for novel chimeric antigen receptor Tcell therapy in Myeloma. Leuk Res. 2018;71:106-111. 41. Xu J, Su Y, Xu A, et al. miR-221/222-mediated inhibition of autophagy promotes dexamethasone resistance in multiple myeloma. Mol Ther. 2019;27(3):559-570. 42. Wu J, Chu E, Kang Y. PIM kinases in multiple myeloma. Cancers (Basel). 2021;13(17):4304. 43. Pan J, Sun Y, Zhang N, et al. Characteristics of BAFF and APRIL factor expression in multiple myeloma and clinical significance. Oncol Lett. 2017;14(3):2657-2662. 44. Alexandrakis MG, Roussou P, Pappa CA, et al. Relationship between circulating BAFF serum levels with proliferating markers in patients with multiple myeloma. Biomed Res Int. 2013;2013:389579.

Haematologica | 108 December 2023

3383

ARTICLE - Plasma Cell Disorders

Cancer-specific mortality in multiple myeloma: a population-based retrospective cohort study Arleigh McCurdy,1 Hsien Seow,2 Gregory P. Pond,2 Anastasia Gayowsky,3 Rajshekhar Chakraborty,4 Alissa Visram,1 Rayan Kaedbey,5 Anita D’Souza,6 Ghulam Rehman Mohyuddin,7 Tanya M. Wildes,8 Rafael Fonseca9 and Hira Mian2 The Ottawa Hospital Research Institute, Ottawa, Ontario, Canada; 2Department of Oncology, McMaster University, Hamilton, Ontario, Canada; 3ICES McMaster, McMaster University, Hamilton, Ontario, Canada; 4Columbia University, New York, NY, USA; 5Department of Medicine, Jewish General Hospital, Montreal, Quebec, Canada; 6Division of Hematology/Oncology, Medical College of Wisconsin, Milwaukee, WI, USA; 7Department of Hematology, Huntsman Cancer Institute, Salt Lake City, UT, USA; 8Division of Hematology/Oncology, University of Nebraska Medical Center, Omaha, NE, USA and 9Division of Hematology/Oncology, Mayo Clinic, Phoenix, AZ, USA 1

Correspondence: A. McCurdy amccurdy@toh.ca Received: Accepted: Early view:

March 7, 2023. July 4, 2023. July 13, 2023.

Abstract Survival has improved in patients diagnosed with multiple myeloma (MM) over the last two decades; however, there remains a paucity of data on the causes of death in MM patients and whether causes of death change during the disease trajectory. We conducted a retrospective population-based study to evaluate the rates of MM-specific versus non-MM cause of death and to identify factors associated with cause-specific death in MM patients, stratified into autologous stem cell transplant (ASCT) and non-ASCT cohorts. A total of 6,677 patients were included, 2,576 in the ASCT group and 4,010 in the non-ASCT group. Eight hundred and seventy-three (34%) ASCT patients and 2,787 (68%) non-ASCT patients died during the follow-up period. MM was the most frequent causes of death, causing 74% of deaths in the ASCT group and 67% in the non-ASCT group. Other cancers were the second leading causes of death, followed by cardiac and infectious diseases. Multivariable analysis demonstrated that a more recent year of diagnosis and novel agent use within 1 year of diagnosis were associated with a decreased risk of MM-specific death, whereas a history of previous non-MM cancer, older age, and the presence of CRAB criteria at diagnosis increased the risk of non-MM death. Our data suggests that despite improvement in MM outcomes in recent years, MM remains the greatest threat to overall survival for patients. Further advances in the development of effective MM therapeutic agents in both ASCT and non-ASCT populations and patient access to them is needed to improve outcomes.

Introduction Multiple myeloma (MM) is common hematologic malignancy, representing 2% of all cancers1 and affecting 27,000 patients per year in North America, with an age standardized incidence rate of 5.2/100,000.2 It is predominantly an illness of older adults, with a median age of 69 years at diagnosis.1 The treatment of patients with MM has changed dramatically over the last two decades, as therapy has evolved beyond the traditional backbone of corticosteroids and alkylating agents to combinations of ‘novel’ drugs including proteosome inhibitors (PI), immunomodulatory drugs (IMiD) and monoclonal antibodies.3, 4 As a result, the overall survival of MM patients has improved steadily, with recent updates reporting median survival times exceeding 10 years in select patient groups.5,6 Despite these improvements, MM remains incurable, with the vast majority of patients experiencing

disease relapse and remaining at risk for MM-specific death. There is a paucity of data on the proportion of MM-specific versus non-MM (COD) among MM patients in the novel agent era, and it is unclear whether this changes during the course of the illness. Previous populationbased studies have shown MM to be the most common cause of death in MM patients, with approximately 75% of patients experiencing MM-specific death.7-11 Two large USbased SEER analyses evaluating MM-specific survival between 1973-200810 and 1987-20138 showed improvements in MM-specific death rates in more recent years, however more modern patient cohorts with increasing novel agent use have not yet been evaluated. This information is becoming more important in the age of prolonged survival for MM patients, whereby patients may be at risk for nonMM complications and comorbidities, and therapeutic decisions may become even more complex.

Haematologica | 108 December 2023

3384

ARTICLE - Cause of death MM

A. McCurdy et al.

Understanding the relative rates of MM-specific versus non-MM cause of deaths is critical in guiding decisions regarding treatment options, monitoring strategies, supportive care, and patient counseling to enable shared decision-making. Thus, we conducted a population-based study with following objectives: i) to evaluate rates of MMspecific versus non-MM cause of death and ii) to identify prognostic factors associated with cause-specific death among patients with MM.

Methods We conducted a retrospective population-based study using data from ICES (formerly known as the Institute for Clinical Evaluative Sciences), a prospective administrative database that captures all health records in the publicly funded health care system in Ontario, Canada. Ontario has a universal, single payer, publicly funded system which provides access to health care expenditures including chemotherapy agents. Administrative records are maintained by the province’s health care system, which captures virtually all health care encounters, with a loss of follow-up of less than 0.5% per year.12 Linked administrative datasets used for this study included the Ontario Health Insurance Plan (OHIP), Registered Persons Database (RPDB), Ontario Cancer Registry (OCR) and the Ontario Registrar General Death database (ORGD), which receives information from death certificates completed by physicians (Form 16) and codes causes of death by ICD code.13 These datasets were linked using unique encoded identifiers from the provincial health insurance number and analyzed at ICES. This study was approved by the Hamilton Integrated Research Ethics Board, REB number 5887. Adult patients treated for newly diagnosed MM between 2007-2018 were identified using ICD-O-3 code 9732/3 (MM). Patients who did not receive MM treatment within 1 year following diagnosis were excluded to ensure that those with smoldering MM were not included in our analysis, consistent with prior population-based studies.14 The cohort was then stratified into autologous stem cell transplant (ASCT) and non-ASCT groups, with the ASCT group being those who had ASCT within 1 year of diagnosis. Treatment sequencing in Ontario is largely uniform given the funding guidelines.15 Briefly, all eligible patients (i.e., fit and age <70 years) undergo induction therapy with CYBORD (cyclophosphamide-bortezomib-dexamethasone) followed by ASCT in first line and lenalidomide maintenance (funded 2014) until disease progression. Due to the funding pathways in Ontario, for all eligible patients ASCT is largely done in first line (early-ASCT) as previously shown with nearly 70% of patients <65 years receiving an ASCT within 1 year of diagnosis.16 Non-ASCT patients undergo treatment with PI- and/or IMID-based therapies

with VMP (velcade-melphalan-prednisone) previously and Rd (lenalidomide-dexamethasone +/- bortezomib) in more recent years. At the time of relapse, patients have access to anti-CD38 (funded 2019), carfilzomib (funded 2018) and pomalidomide (funded 2015) based regimens. Baseline patient and disease characteristics at the time of MM diagnosis were collected. Comorbidities were recorded based on health service use in the 24 months preceding MM diagnosis using the Johns Hopkins Adjusted Clinical Group system score (calculated using The Johns Hopkins ACG® System Version 10), whereby sum of the 32 ACG® System Aggregated Diagnosis Groups (ADG) was categorized as high (≥10) or low (<10) comorbidity burden.17 Patient geographic and socioeconomic status (SES) was defined by neighborhood income quintile (rural, and among urban areas level 1 having the lowest SES and level 5 the highest SES status). MM CRAB features (hypercalcemia, renal failure, anemia, bone lesions) within 6 months before or after MM diagnosed were defined by respective ICD codes as published previously.18 Prior cancer diagnosis within 15 years of the index MM diagnosis was recorded. As treatment algorithms for MM are largely consistent across the province due to funding criteria, all funded cancer drugs (alkylating agents, PI, IMiD, and monoclonal antibodies) available to patients during the study period were extracted.19 Among patients that died during the study follow-up period, the cause of death was identified using the antecedent cause of death as reported in the ORGD by ICD code. We estimated the cumulative incidence of MMspecific and non-MM cause of death. In order to identify the association of prognostic factors on the probability of MM-specific and non-MM deaths, we performed competing risks regression and estimated the multivariate FineGray subdistribution hazard ratios adjusted for covariates. Results were reported as hazard ratio (HR) with 95% confidence interval (CI) and statistical significance defined as P<0.05. Analyses were conducted using Statistical analysis system (SAS version 9.4).

Results A total of 6,677 patients were identified in the newly diagnosed MM cohort, 2,576 in the ASCT group and 4,101 in the non-ASCT group. Baseline patient characteristics are shown in Table 1. The median age at diagnosis for the overall cohort was 68 years (58 and 74 years in the ASCT group and non-ASCT group, respectively). Forty-three percent of the overall cohort was female. There was a low burden of comorbidities as indicated by a low ADG score (<10) in 74% of all patients, a high comorbidity burden (ADG ≥10) was more common in the non-ASCT group than the ASCT group (31% vs. 19%; P<0.001). In total, 824 pa-

Haematologica | 108 December 2023

3385

ARTICLE - Cause of death MM

A. McCurdy al.

tients (12%) of the overall cohort had a diagnosis of another cancer within 15 years prior to the MM diagnosis, which was present in a higher proportion of patients in the non-ASCT group (15%) compared to the ASCT patients (8%). Common cancers present prior to MM diagnosis were prostate (4% non-ASCT; 2% ASCT) breast (1% for both non-ASCT and ASCT groups) and colorectal cancers (2% non-ASCT cohort; <1% ASCT cohort). The majority of patients (87%) of the cohort were from urban centers, and SES by neighborhood income quintile was balanced across quintiles 1-5 with no significant difference between the ASCT and non-ASCT groups. Patients diagnosed in 2007-2013 comprised 56% of the overall cohort with 44% being diagnosed between 20142017. The majority of patients received treatment with a novel agent within 1 year of diagnosis, 75% in the ASCT group and 81% in the non-ASCT group. Eight hundred and seventy-three (34%) in the ASCT group and 2,787 (68%) in the non-ASCT group died during the study follow-up period, with median follow up time of 45 months (range, 25-73) and 27 months (range, 14-49) respectively. The median overall survival for the ASCT cohort was 8.4 years (95% confidence interval [CI]: 7.9-9.2) and for the non-ASCT cohort was 3.0 years (95% CI: 2.8-3.1). The cumulative incidence of MM-specific death was higher than non-MM death for both the ASCT (24.9%: MMspecific and 7.9 %: non-MM at 5 years) and non-ASCT cohort (47.6 %: MM-specific and 22.5%: non-MM at 5 years) throughout the disease trajectory (Figure 1). Cause of death stratified by time from MM diagnosis (<3

years, 3-5 years and >5 years) for both cohorts is shown in Figure 2. MM-specific death was higher overall in the ASCT cohort (74% vs. 67% in the non-ASCT cohort) and at each respective time point. Other non-MM cancers accounted for 7% of deaths in the ASCT group and 6% in the non-ASCT group. These cancers included cancer of the lung (11% non-ASCT and ASCT), acute myeloid leukemia (7% non-ASCT and 10% ASCT) and colorectal cancers (4% non-ASCT cohort and 3% ASCT cohort). Additionally, in our cohort 2.5% and 5.8% of patients died from heart disease and 6.2% and 4.3% of infectious diseases in each ASCT and non-ASCT group respectively. Detailed causes of death are presented in the Online Supplementary Table S1. Multivariable analysis showing factors associated with MM and non-MM cause of death is shown in Table 2. MMspecific mortality was decreased in the more recent 2014-2017 cohort in both ASCT hazard ratio [HR] =0.72; 95% CI: 0.65-0.80) groups, with similar effects observed for non-MM specific mortality. The use of novel agents was also associated with decreased MM-specific mortality for both groups, after adjusting for non-MM death as a competing risk: ASCT (HR=0.84; 95% CI: 0.70-1.00) and non-ASCT (HR=0.84; 95% CI: 0.75-0.94). CRAB features at diagnosis were associated with an increased risk of both MM-specific and non-MM causes of death across all groups, whereas a history of previous non-MM cancer was associated with an increased risk of non-MM cause of death among both the ASCT (HR=1.91; 95% CI: 1.292.82) and non-ASCT (HR=1.26; 95% CI: 1.05-1.50) groups.

Table 1. Baseline characteristics. Variable Age in years at diagnosis

Value

Non-ASCT Cohort N=4,101

ASCT Cohort N=2,576

Total N=6,677

Median (IQR)

75 (70-80)

59 (53-64)

69 (60-77)

2,319 (56.6)

1,478 (57.4)

3,797 (56.9)

Male sex, N (%) Total ADG score* (%)

Low comorbidities (ADG <10) High comorbidities (ADG ≥10)

2,816 (68.7) 1,285 (31.3)

2,091 (81.2) 485 (18.9)

4,907 (73.5) 1,770 (26.5)

Rural status and neighbourhood income quintile, N (%)

Rural Urban - quintile 1 (low income) Urban - quintile 2-4 Urban - quintile 5 (high income)

564 (13.8) 686 (16.7) 2120 (51.7) 719 (17.5)

301 (11.7) 328 (12.7) 1410 (54.7) 534 (20.7)

865 (13.0) 1,014 (15.2) 3,530 (52.9) 1,253 (18.8)

Year of diagnosis

2007-2013 2014-2017

2,339 (57.0) 1,762 (43.0)

1,371 (53.2) 1,205 (46.8)

3,710 (55.6) 2,967 (44.4)

CRAB at diagnosis^, N (%)

1,923 (46.9)

991 (38.5)

2,914 (43.6)

Previous cancer within 15 years of MM diagnosis, N (%)

608 (14.8)

216 (8.4)

824 (12.3)

Novel drugs within 1 year of diagnosis#, N (%)

3,307 (80.6)

1,937 (75.2)

5,244 (78.5)

Hypercalcemia, renal failure, anemia, bone lesions; *co-morbidities defined by John Hopkin aggregated diagnosis group (ADG) with high comorbidities defined as ≥ 10; #lenalidomide, thalidomide or bortezomib within 1 year of diagnosis; does not add to 100% due to missing values. ASCT: autologous stem cell trasnplant; IQR: interquartile range; MM: multiple myeloma.

Haematologica | 108 December 2023

3386

ARTICLE - Cause of death MM

A. McCurdy al.

Discussion This study represents one of the largest real-world cohort studies to examine MM-specific versus non-MM cause of death among MM patients. Our data suggests that even though MM-specific mortality has decreased in more recent years and with novel agents, MM remains the greatest threat to survival for MM patients, including among the older non-ASCT cohort. Our results show similar trends to those seen in a large cohort study evaluating MM-specific death using the US Surveillance, Epidemiology, and End Results Program (SEER) and Puerto Rico Central Cancer Registry (PRCCR) database between 1987-20138, showing 72% of patients having MM-specific death as compared to 69% in our co-

hort, with similar median ages at diagnosis of 69 and 68 years respectively. They also report a decrease in MM-specific death in more recent years, suggesting patients with access to novel therapeutics and resultant prolonged overall MM survival may have an increased risk of developing non-MM comorbidities with age and time. A second SEER analysis evaluating MM-survival by historical time cohorts and novel agent access further supported this observation, with markedly reduced MM-specific mortality following the advent of new treatments.10 It is important to note that while MM-specific mortality has decreased over time, overall survival for older adults with MM remains poor with a median overall survival noted to be around 3 years consistent with previously published series.20 The SEER/PRCCR study also reported MM as the most

Figure 1. Cumulative incidence by cause of death and transplant group. MM: multiple myeloma. Haematologica | 108 December 2023

3387

ARTICLE - Cause of death MM

A. McCurdy al.

common cause of death across time all time periods, with a slightly decreased risk of MM-specific death for patients alive at 2 (SEER) or 3 (PRCCR) years post MM diagnosis.8 We saw similar outcomes, with fewer patients having MM as the primary cause of death amongst patients alive 5 years post-diagnosis. There are several factors that could contribute to this finding, including an increased risk of non-MM illnesses with age, chemotherapy adverse effects, and/or favorable MM disease biology. Non-myeloma cancer was the second most common cause of death in our study, leading to 6% of deaths in the non-ASCT group and 7% in the ASCT group, followed closely by cardiac disease in 5.8% and 2.5% of patients respectively. Interestingly, in the aforementioned SEER/PRCCR study, patients in the PRCCR had an 11.2% mortality rate from other cancers followed by 5.3% from heart disease, whereas the SEER cohort had cardiac disease as the highest non-MM cause of death (11.5%) followed by closely by other cancers (5.2%).8 Little is known about the impact of a prior cancer diag-

nosis on survival in MM patients, and more work in this area is needed. While our study demonstrated that 12.8% of our cohort were diagnosed with a separate cancer within 15 years prior to the index MM diagnosis, we were not able to distinguish the relative contribution of these diagnoses versus secondary primary malignancies (SPM) acquired after the MM diagnosis to the overall non-MM cancer deaths in our cohort. SPM are a known risk in patients with MM,21 in particular associations with ASCT and lenalidomide have been reported.22-24 A SEER analysis examining SPM in patient with MM over time and treatment eras showing a small increase in SPM over time, independent of improved MMspecific survival, from 4.7% of patients in a 1995-99 cohort to 6.3% in a 2005-09 cohort, potentially driven partly by an increased risk of developing lymphoma and other hematologic malignancies.22 In the DETERMINATION study that compared lenalidomide, bortezomib and dexamethasone (RVD) for eight cycles versus RVD + ASCT/RVD consolidation, both followed by lenali-

Figure 2. Causes of death by autologous stem cell transplant group and time from multiple myeloma diagnosis. COD: cause of death; MM: multiple myeloma; ASCT: autologous stem cell transplant.

Haematologica | 108 December 2023

3388

ARTICLE - Cause of death MM

A. McCurdy al.

Table 2. Multivariable Fine-Gray model showing factors associated with multiple myeloma specific versus non-multiple myeloma cause of death (sub-hazard ratio presented with 95% confidence interval).^ ASCT <1 year following diagnosis Variable

No ASCT <1 year following diagnosis

Value

MM COD (95% CI)

Non-MM COD (95% CI)

MM COD (95% CI)

Non-MM COD (95% CI)

Age in years at diagnosis date

<50 (ref = 50-69) 70-79 (ref = 50-69) 80+ (ref = 50-69)

0.98 (0.79-1.20) 1.33 (0.81-2.18) 2.03 (0.64-6.39)

0.67 (0.44-1.02) 0.31 (0.08-1.23) 6.44 (2.19-18.99)

0.85 (0.52-1.39) 0.91 (0.81-1.02) 1.08 (0.94-1.23)

0.88 (0.43-1.78) 1.15 (0.96-1.37) 1.46 (1.20-1.76)

Sex

Male (ref = female)

1.20 (1.02-1.41)

1.31 (1.00-1.73)

0.99 (0.91-1.09)

1.12 (0.98-1.28)

Rural status and neighbourhood income quintile

Rural (ref = urban: 1) Urban: 2-4 (ref = urban: 1) Urban: 5 (ref = urban: 1)

1.00 (0.72-1.38) 1.09 (0.85-1.39) 0.85 (0.64-1.14)

1.12 (0.68-1.85) 0.84 (0.55-1.26) 1.14 (0.72-1.80)

1.13 (0.95-1.33) 1.12 (0.98-1.28) 1.00 (0.85-1.18)

0.88 (0.70-1.10) 0.77 (0.65-0.92) 0.73 (0.59-0.91)

Total ADG score*

High (ADG >=10) (ref = low [ADG <10])

1.06 (0.86-1.30)

1.12 (0.80-1.57)

1.06 (0.95-1.17)

1.11 (0.96-1.27)

Year of diagnosis

2014-2017 (ref = 2007-2013)

0.69 (0.55-0.86)

0.82 (0.57-1.18)

0.72 (0.65-0.80)

0.86 (0.74-1.00)

CRAB^

Yes (ref = No)

1.72 (1.48-2.01)

1.55 (1.19-2.02)

1.52 (1.38-1.66)

1.32 (1.16-1.51)

Cancer diagnosis in the previous 15 years

Yes (ref = No)

0.84 (0.61-1.15)

1.91 (1.29-2.82)

0.92 (0.80-1.06)

1.26 (1.05-1.50)

Novel drugs within 1 year of diagnosis#

Yes (ref = No)

0.84 (0.70-1.00)

0.92 (0.68-1.24)

0.84 (0.75-0.94)

0.85 (0.73-0.99)

Comorbidities defined by John Hopkin aggregated diagnosis group (ADG) with high comorbidities defined as >/= 10; #lenalidomide, thalidomide or bortezomib within 1 year of diagnosis; ^hypercalcemia, renal failure, anemia, bone lesions. COD: cause of death; MM: multiple myeloma; ASCT: autologous stem cell transplant; ref: reference. *

domide maintenance until progression, 10.4% of patients in the RVD group and 10.7% of patients in the RVD + ASCT group developed SPM.23 While secondary hematologic malignancies developed in 2.5% of the RVD group and 3.6% of the RVD + ASCT group, no RVD patients developed acute myeloid leukemia (AML) or a myelodysplastic syndrome as opposed to 2.7% RVD + ASCT patients (P=0.002).23 In the IFM 2009 trial, comparing RVD for eight cycles versus RVD + ASCT/RVD consolidation, both followed by maintenance lenalidomide for 1 year or until progression, SPM occurred in 6.0% of the RVD group and 7.1% of the RVD + ASCT group, with hematologic malignancies developing in one (0.3%) of RVD patients (MDS) and four (1.1%) RVD + ASCT patients (3 AML, 1 MDS).24 Interestingly, in our study AML was the second most common non-MM cancer cause of death in both cohorts, accounting for 7% of non-MM cancer deaths in the non-ASCT group and 10% in the ASCT group. Patients in both cohorts had exposure to alkylating agents, with the use of cyclophosphamide for induction and high dose melphalan for conditioning in the ASCT group, as well as standard dose melphalan used for induction in many of the non-ASCT patients. Patients in our study were also older, with median age of 69 years at diagnosis, as compared to 57-59 years in the RVD groups and 55-60 years in the RVD + ASCT groups for DETERMINATION23 and IFM200924 respectively. The association of cardiovascular disease in MM patients has been previously reported, is higher than a control co-

hort without cancer,25 and may occur due to both MM-related (treatment, anemia, renal failure) and MM-unrelated (age, obesity, diabetes)26 factors, supporting it as a common cause of death in MM patients.27 Two studies using the SEER database27,28 and a large French administrative cohort29 showed a decreased risk of cardiovascular mortality over the last several decades in MM patients, which we did not see over the time horizon in the present study. Our study has several strengths including the large study population, comprehensive database, and standardized provincial treatment algorithms that may minimize heterogeneity in treatment that could impact MM-specific death. However, there are also several limitations. We used the antecedent (underlying) cause of death by ICD10 as captured in the ORGD, which is populated from death certificates filled out by physicians. Misclassification of some patients is possible, though a high level of agreement on cause of death between the ORGD and a large prospective cohort of cancer patients with rigorous clinical follow up has been previously reported.30 Additionally, determining solely one underlying cause of death may not always be possible particularly among causes such as infections were untreated MM, treatment for MM as well as other underlying comorbidities may all contribute to infections. We were unable to capture some specific patient and disease characteristics that could impact patient outcomes, such as high-risk cytogenetics. Our population also represent a more homogenous group of patients

Haematologica | 108 December 2023

3389

ARTICLE - Cause of death MM

A. McCurdy al.

treated within a publicly funded health care system with relatively uniform treatment approaches and, therefore, our results may not be generalizable to all treatment settings. Finally, we are unable to evaluate outcomes in patients receiving the most modern therapies (i.e., monoclonal antibodies) as these therapies were not funded in the upfront treatment setting during the study period. Our study demonstrated that MM remains the most likely cause of death in patients diagnosed with MM, despite impressive improvements in overall survival in recent years with a growing armamentarium of novel therapeutics. This data can be helpful for patients and clinicians in guiding shared treatment decision-making and estimating risk benefit of therapies. It strongly supports the notion that optimization of MM-directed care is paramount for patients, whose survival ultimately remains at greatest risk from MM. However, we must remain mindful of the risks associated with MM-directed treatment such as secondary malignancies with our past and currently available treatments, and significant infections with the advent of T-cell redirecting therapies.31-33 Further advances in the delivery of safe and effective MM therapeutic agents in both ASCT and non-ASCT cohort is needed to further improve outcomes in this disease. Disclosures AM discloses consultancy/honoraria fees from BMS, Takeda, Janssen, Amgen, Sanofi, Forus, GSK and Pfizer; has received research funding from BMS. GP discloses to currently hold individual stocks of Roche Canada; discloses consultancy fees from Astra-Zeneca, Merck and Profound Medical; discloses membership on advisory committee for Takeda. RC discloses consultancy/advisory board fees from Janssen, Sanofi, and Adaptive Biotech. AV discloses honoraria/advisory fees from Janssen, Sanofi, Apotex and Pfizer. RK discloses honoraria/advisory fees fom Janssen, BMS, FORUS, Amgen, Sanofi and Pfizer. AD has received institutional research funding from Sanofi, Takeda, TeneoBio, Caelum and Prothena; discloses consultancy/advisory board fees fom Janssen, Prothena, Imbrium, Pfizer and BMS. TW has received consultancy fees from Janssen, Carevive, Seattle Genetics and Sanofi. RF discloses consultancy/honoraria fees AbbVie, Amgen, Bayer, BMS, GSK, H3 Therapeutics, Janssen, Juno, Karyopharm, Kite, Merck, Novartis, Oncopeptides, On-

coTracker, Pfizer, Pharmacyclics, Regeneron, Sanofi and Takeda; is part of the scientific advisory board of Adaptive Biotechnologies, Caris Life Sciences, OncoMyx and OncoTracker. HM discloses consultancy/honoraria fees from BMS, Takeda, Janssen, Amgen, Sanofi, Forus, GSK and Pfizer; has receieved research funding Janssen. All other authors have no conflicts of interest to disclose. Contributions HM, AM, GP, HS and AG developed the concept and designed the study. AG and HM collected data. HM, AM, GP, HS and AG analyzed and interpreted data. All authors wrote the manuscript and approved the final version of the article. Acknowledgments This document used data adapted from the Statistics Canada Postal CodeOM Conversion File, which is based on data licensed from Canada Post Corporation, and/or data adapted from the Ontario Ministry of Health Postal Code Conversion File, which contains data copied under license from ©Canada Post Corporation and Statistics Canada. Parts of this material are based on data and/or information compiled and provided by CIHI, the Ontario Ministry of Health, and Ontario Health (OH). The analyses, conclusions, opinions, and statements expressed herein are solely those of the authors and do not reflect those of the funding or data sources; no endorsement is intended or should be inferred. Parts of this report are based on Ontario Registrar General (ORG) information on deaths, the original source of which is ServiceOntario. The views expressed therein are those of the authors and do not necessarily reflect those of ORG or the Ministry of Public and Business Service Delivery. We thank IQVIA Solutions Canada Inc. for use of their Drug Information File. Funding HM is currently funded by the Hamilton Health Sciences Early Career Award. This study was supported by ICES, which is funded by an annual grant from the Ontario Ministry of Health and the Ministry of Long-Term Care. Data-sharing statement The data that support the findings of this study are available at ICES but restrictions apply to the availability of these data.

References 1. Cancer Stat Facts: Myeloma. National Cancer Institute Surveillance, Epidemiology and End Results program. https://seer.cancer.gov/statfacts/html/mulmy.html Accessed 12 June 2023. 2. Cowan AJ, Allen C, Barac A, et al. Global burden of multiple myeloma: a systematic analysis for the Global Burden of Disease Study 2016. JAMA Oncol. 2018;4(9):1221-1227.

3. Turesson I, Bjorkholm M, Blimark CH, Kristinsson S, Velez R, Landgren O. Rapidly changing myeloma epidemiology in the general population: Increased incidence, older patients, and longer survival. Eur J Haematol. 2018;101(2):237-244. 4. Costa LJ, Brill IK, Omel J, Godby K, Kumar SK, Brown EE. Recent trends in multiple myeloma incidence and survival by age, race, and ethnicity in the United States. Blood Adv. 2017;1(4):282-287.

Haematologica | 108 December 2023

3390

ARTICLE - Cause of death MM

A. McCurdy al.

5. Mian H, Reece D, Masih-Khan E, et al. Survival and outcomes of newly diagnosed multiple myeloma patients stratified by transplant status 2007-2018: retrospective analysis from the Canadian Myeloma Research Group Database. Clin Lymphoma Myeloma Leuk. 2022;22(8):608-617. 6. Joseph NS, Kaufman JL, Dhodapkar MV, et al. Long-term follow-up results of lenalidomide, bortezomib, and dexamethasone induction therapy and risk-adapted maintenance approach in newly diagnosed multiple myeloma. J Clin Oncol. 2020;38(17):1928-1937. 7. Mohammadi M, Cao Y, Glimelius I, Bottai M, Eloranta S, Smedby KE. The impact of comorbid disease history on all-cause and cancer-specific mortality in myeloid leukemia and myeloma - a Swedish population-based study. BMC Cancer. 2015;15:850. 8. Castañeda-Avila MA, Ortiz-Ortiz KJ, Torres-Cintrón CR, Birmann BM, Epstein MM. Trends in cause of death among patients with multiple myeloma in Puerto Rico and the United States SEER population, 1987-2013. Int J Cancer. 2020;146(1):35-43. 9. Bhatt VR, Loberiza FR, Jr., Jing H, et al. Mortality patterns among recipients of autologous hematopoietic stem cell transplantation for lymphoma and myeloma in the past three decades. Clin Lymphoma Myeloma Leuk. 2015;15(7):409-415.e1. 10. Libby E, Garcia D, Quintana D, et al. Disease-specific survival for patients with multiple myeloma: significant improvements over time in all age groups. Leuk Lymphoma. 2014;55(12):2850-2857. 11. Eisfeld C, Kajüter H, Möller L, Wellmann I, Shumilov E, Stang A. Time trends in survival and causes of death in multiple myeloma: a population-based study from Germany. BMC Cancer. 2023;23(1):317. 12. Silver SA, Harel Z, McArthur E, et al. Causes of death after a hospitalization with AKI. J Am Soc Nephrol. 2018;29(3):1001-1010. 13. Ontario AoPHEi. https://www.apheo.ca/data-vital-statsmortality. Accessed January 31, 2023. 14. Mohyuddin GR, Sinnarajah A, Gayowsky A, Chan KKW, Seow H, Mian H. Quality of end-of-life care in multiple myeloma: a 13year analysis of a population-based cohort in Ontario, Canada. Br J Haematol. 2022;199(5):688-695. 15. Canada’s Drug and Health Technology Agency. Multiple myeloma sequencing guidelines. https://www.cadth.ca/multiple-myeloma Accessed May 25, 2023. 16. Mian HS, Seow H, Wildes TM, et al. Disparities in treatment patterns and outcomes among younger and older adults with newly diagnosed multiple myeloma: a population-based study. J Geriatr Oncol. 2021;12(4):508-514. 17. Starfield B, Weiner J, Mumford L, Steinwachs D. Ambulatory care groups: a categorization of diagnoses for research and management. Health Serv Res. 1991;26(1):53-74. 18. Fiala MA, Dukeman J, Tuchman SA, Keller M, Vij R, Wildes TM. Development of an algorithm to distinguish smoldering versus symptomatic multiple myeloma in claims-based data sets. JCO Clin Cancer Inform. 2017;1:CCI.17.00089. 19. Newell AM, VanSwearingen JM, Hile E, Brach JS. The modified

Gait Efficacy Scale: establishing the psychometric properties in older adults. Phys Ther. 2012;92(2):318-328. 20. Mian H, Wildes TM, Vij R, Pianko MJ, Major A, Fiala MA. Dynamic frailty risk assessment among older adults with multiple myeloma: a population-based cohort study. Blood Cancer Journal. 2023;13(1):76. 21. Musto P, Anderson KC, Attal M, et al. Second primary malignancies in multiple myeloma: an overview and IMWG consensus. Ann Oncol. 2017;28(2):228-245. 22. Palumbo A, Bringhen S, Kumar SK, et al. Second primary malignancies with lenalidomide therapy for newly diagnosed myeloma: a meta-analysis of individual patient data. Lancet Oncol. 2014;15(3):333-342. 23. Richardson PG, Jacobus SJ, Weller EA, et al. Triplet therapy, transplantation, and maintenance until progression in myeloma. N Engl J Med. 2022;387(2):132-147. 24. Attal M, Lauwers-Cances V, Hulin C, et al. Lenalidomide, bortezomib, and dexamethasone with transplantation for myeloma. N Engl J Med. 2017;376(14):1311-1320. 25. Armenian SH, Xu L, Ky B, et al. Cardiovascular disease among survivors of adult-onset cancer: a community-based retrospective cohort study. J Clin Oncol. 2016;34(10):1122. 26. Bringhen S, Milan A, Ferri C, et al. Cardiovascular adverse events in modern myeloma therapy - Incidence and risks. A review from the European Myeloma Network (EMN) and Italian Society of Arterial Hypertension (SIIA). Haematologica. 2018;103(9):1422-1432. 27. Yin X, Fan F, Zhang B, Hu Y, Sun C. Cardiovascular-specific mortality among multiple myeloma patients: a populationbased study. Ther Adv Hematol. 2022;13:20406207221086755. 28. Costa LJ, Godby KN, Chhabra S, Cornell RF, Hari P, Bhatia S. Second primary malignancy after multiple myeloma-population trends and cause-specific mortality. Br J Haematol. 2018;182(4):513-520. 29. Cottin Y, Boulin M, Doisy C, et al. Mortality and major cardiovascular events among patients with multiple myeloma: analysis from a Nationwide French Medical Information Database. Cancers (Basel). 2022;14(13):3049. 30. Brenner DR, Tammemägi MC, Bull SB, Pinnaduwaje D, Andrulis IL. Using cancer registry data: agreement in cause-of-death data between the Ontario Cancer Registry and a longitudinal study of breast cancer patients. Chronic Dis Can. 2009;30(1):16-19. 31. Moreau P, Garfall AL, van de Donk N, et al. Teclistamab in relapsed or refractory multiple myeloma. N Engl J Med. 2022;387(6):495-505. 32. Munshi NC, Anderson LD, Jr., Shah N, et al. Idecabtagene vicleucel in relapsed and refractory multiple myeloma. N Engl J Med. 2021;384(8):705-716. 33. Trudel S, Cohen AD, Krishnan AY, et al. Cevostamab monotherapy continues to show clinically meaningful activity and manageable safety in patients with heavily pre-treated relapsed/refractory multiple myeloma (RRMM): updated results from an ongoing phase I study. Blood. 2021;138(Suppl 1):S157.

Haematologica | 108 December 2023

3391

ARTICLE - Plasma Cell Disorders

Disease associations with monoclonal gammopathy of undetermined significance can only be evaluated using screened cohorts: results from the population-based iStopMM study Aðalbjörg Ýr Sigurbergsdóttir,1 Sæmundur Rögnvaldsson,1,2 Sigrún Thorsteinsdóttir,1,3 Ingigerður Sverrisdóttir,1,4 Guðrún Ásta Sigurðardóttir,1 Brynjar Viðarsson,2 Páll Torfi Önundarson,1,2 Bjarni A. Agnarsson,2 Margrét Sigurðardóttir,2 Ingunn Þorsteinsdóttir,2 Ísleifur Ólafsson,2 Ásdís Rósa Þórðardóttir,1 Gauti Kjartan Gíslason,1 Andri Ólafsson,1 Malin Hultcrantz,5 Brian G. M. Durie,6 Stephen Harding,7 Ola Landgren,8 Thorvarður Jón Löve1,2 and Sigurður Yngvi Kristinsson1,2 University of Iceland, Reykjavík, Iceland; 2Landspítali - the National University Hospital of Iceland, Reykjavík, Iceland; 3Rigshospitalet, Copenhagen, Denmark; 4Sahlgrenska University Hospital, Gothenburg, Sweden; 5Memorial Sloan Kettering Cancer Center, New York, NY, USA; 6Cedars-Sinai Samuel Oschin Cancer Center, Los Angeles, CA, USA; 7The Binding Site, Birmingham, UK and 8Sylvester Comprehensive Cancer Center, Miami, FL, USA 1

Correspondence: S.Y. Kristinsson sigyngvi@hi.is Received: Accepted: Early view:

March 22, 2023. June 30, 2023. July 13, 2023.

Abstract Monoclonal gammopathy of undetermined significance (MGUS) is an asymptomatic precursor condition that precedes multiple myeloma and related disorders but has also been associated with other medical conditions. Since systematic screening is not recommended, MGUS is typically diagnosed due to underlying diseases and most cases are not diagnosed. Most previous studies on MGUS disease associations have been based on clinical cohorts, possibly resulting in selection bias. Here we estimate this selection bias by comparing clinically diagnosed and screened individuals with MGUS with regards to demographics, laboratory features, and comorbidities. A total of 75,422 participants in the Iceland Screens, Treats, or Prevents Multiple Myeloma (iStopMM) study were screened for MGUS by serum protein electrophoresis, immunofixation and free light chain assay (clinicaltrials gov. Identifier: NCT03327597). We identified 3,352 individuals with MGUS, whereof 240 had previously been clinically diagnosed (clinical MGUS), and crosslinked our data with large, nationwide registries for information on comorbidities. Those with clinical MGUS were more likely to have at least one comorbidity (odds ratio=2.24; 95% confidence interval: 1.30-4.19), and on average had more comorbidities than the screened MGUS group (3.23 vs. 2.36, mean difference 0.68; 95% confidence interval: 0.46-0.90). They were also more likely to have rheumatological disease, neurological disease, chronic kidney disease, liver disease, heart failure, or endocrine disorders. These findings indicate that individuals with clinical MGUS have more comorbidities than the general MGUS population and that previous studies have been affected by significant selection bias. Our findings highlight the importance of screening data when studying biological and epidemiological implications of MGUS.

Introduction Monoclonal gammopathy of undetermined significance (MGUS) is an asymptomatic precursor condition that consistently precedes multiple myeloma and related disorders. It is present in approximately 4.5% of the general population over the age of 50 years.1-6 MGUS is asymptomatic and is, therefore, typically diagnosed incidentally during clinical workup for various medical conditions, while most patients remain undiagnosed.1 Although asymptomatic, MGUS has been associated with over 100

medical disorders, that are either assumed to cause or result from MGUS. These include thrombosis,7 infections,8 autoimmune diseases,9,10 fractures,11,12 neuropathies,13 as well as excess mortality.14 In some cases, there exists histopathologic evidence of a causative relationship between MGUS and disease, usually with the identification of the monoclonal immunoglobulin in the affected tissue (for example glomerulus or skin lesion).15 However, the majority of previously reported disease associations with MGUS have been found in retrospective epidemiological studies. Since systematic screening for MGUS is currently not rec-

Haematologica | 108 December 2023

3392

ARTICLE - Clinical versus screened MGUS disease associations ommended, most prior studies have been based on clinical cohorts of individuals diagnosed with MGUS as part of routine workup for other clinical signs or symptoms. Therefore, the diagnosis of MGUS is usually made due to the diagnosis of another disease, including autoimmune disorders or kidney diseases. This leads to a potential bias in the selection of individuals with more medical conditions into previously studied MGUS cohorts and, therefore, overestimation of the associations between MGUS and disease. This is supported by the only prior screening study on disease associations with MGUS, where in a screened cohort in Olmsted County, Minnesota, (n=17,398) only 14 of 75 previously reported associations were confirmed.16 However, the study included numerous distinct and specific diagnostic codes with unclear clinical correlation and thus, the observed number of cases with underlying conditions were small. This decreased statistical power and necessitated the adjustments for multiple comparisons, limiting the possibility of detecting true subtle associations.17 The Iceland Screens, Treats, or Prevents Multiple Myeloma (iStopMM) study is a large Icelandic screening study for MGUS.18 It included both individuals with incidentally diagnosed (clinical) MGUS and people diagnosed through screening, providing a unique opportunity to explore the differences between clinical and screened MGUS with regards to selection bias in clinical cohorts. The aim of this study was to systematically estimate possible selection bias in studies on MGUS by comparing the association of demographic, clinical, and laboratory factors with clinical MGUS with their association with MGUS diagnosed by systematic screening.

Methods iStopMM study The iStopMM study is a population-based, nationwide screening study and randomized controlled trial of followup strategies that aims to evaluate the benefits and potential harms of screening for MGUS (clinicaltrials gov. Identifier: NCT03327597). The study has previously been described in detail.18 Briefly, 80,759 individuals, more than half of the Icelandic population aged ≥40 years in 2016, provided informed consent and 75,422 (>93%) of them were subsequently screened for MGUS using serum protein electrophoresis (SPEP), immunofixation and free light chain (FLC) assay. Those who had abnormal screening results, in absence of a more advanced disease, were randomized into one of three study arms, where arm 1 continued as they had never undergone screening while arm 2 and 3 followed different follow-up strategies.18 The study protocol and all information material from the iStopMM study has been approved by the Icelandic National

A. Sigurbersdóttir et al.

Bioethics Committee and the Icelandic Data Protection Authority. Access to national healthcare registries has been approved by the Icelandic Directorate of Health and the Icelandic Cancer Society. Study cohort and data Participants who had an M-protein on SPEP or an abnormal FLC ratio and were randomized into one of our three study arms were eligible for this study. Participants who had M-protein levels ≥3.0 g/dL or an FLC ratio ≥100 or ≤0.01 were excluded as they, by definition, have smoldering myeloma or multiple myeloma.19 Data from the Icelandic Cancer Registry and laboratory records from Landspítali - the National University Hospital of Iceland (NUHI) and Læknasetrið, the only laboratories performing SPEP in Iceland, were crosslinked to the study cohort. Those who had M-protein previously identified in the clinical setting, were defined as clinical MGUS. M-protein concentration, MGUS isotype, and FLC ratio were acquired from the study screening samples. Finally, comorbidity data was acquired from two national registries: the Hospital Discharge Register and Register of Primary Health Care Contacts, where the accuracy of chronic disease diagnoses has been found to be >95%.20 International Classification of Diseases, 10th Revision (ICD-10) codes from the registries were grouped together into relevant disease categories (Online Supplementary Table S1) and their prevalence compared between clinical and screened MGUS.21 Statistical analyses All statistical analyses compared clinical MGUS to screened MGUS. Demographic features (age, sex, and residence) were evaluated by t test and χ2 test for continuous and binary variables, respectively. When comparing laboratory features, linear regression was used to assess continuous variables (M-protein concentration, number of risk factors for progression), and logistic regression to assess binary variables (M-protein isotype, abnormal serum FLC ratio [<0.26 or >1.65]). Participants’ combined number of comorbidities was compared between the two groups by linear regression. Logistic regression was then used to compare the prevalence of each comorbidity category between the groups. All analyses were adjusted for age and sex, except for those regarding demographic differences. A P value of <0.05 was considered indicative of statistical significance. R Statistical Software (version 3.6.2; R Core Team 2021) was used for all statistical analyses.22

Results A total of 75,422 individuals were screened for MGUS in the iStopMM study, of whom 3,352 were diagnosed with

Haematologica | 108 December 2023

3393

ARTICLE - Clinical versus screened MGUS disease associations MGUS based on M-proteins detected by SPEP. Of these individuals, 240 had previously been diagnosed with clinical MGUS (7.2%) (Table 1). Demographic features The median age of individuals with clinical MGUS was significantly higher than in the screened MGUS group (73.5 vs. 69 years, respectively; P<0.001). The clinical MGUS group was also more likely to reside in Iceland’s capital area compared with those with screened MGUS (68.3% vs. 60.6%; P=0.02). No differences in sex distribution were detected between the two groups (Table 1). Laboratory findings The clinical MGUS group had a 0.17 g/dL higher mean Mprotein concentration than those with screened MGUS (0.51 clinical vs. 0.34 g/dL screened, mean difference 0.17 g/dL, 95% confidence interval [CI]: 0.13-0.22; P<0.001), after adjusting for age and sex (Table 1). There were no statistically significant differences in the prevalence of each risk factor for progression from MGUS to MM or related lymphoproliferative disease, as defined by the International Myeloma Working Group (M-protein isotype [non-IgG MGUS], M-protein concentration ≥1.5 g/dL, and an abnormal FLC ratio [<0.26 or >1.65]), nor in the mean number of risk factors for progression (0.80 clinical vs. 0.74 screened) (Table 1).23 Comorbidities Individuals with clinical MGUS were significantly more likely to have at least one comorbidity compared with those with MGUS diagnosed by screening (n=227 [94.6%]

A. Sigurbersdóttir et al.

clinical vs. n=2,704 [86.9%] screened; odds ratio [OR] =2.24; 95% CI: 1.30-4.19; P<0.01), and had a 37% higher mean number of comorbidities (3.23 clinical vs. 2.36 screened, mean difference 0.68; 95% CI: 0.46-0.90; P<0.001), in models adjusting for age and sex. Furthermore, those with clinical MGUS were more likely to have rheumatological disease (OR=2.98; 95% CI: 2.084.20; P<0.001), neurological disease (OR=2.77; 95% CI: 1.84-4.06; P<0.001), chronic kidney disease (OR=2.55; 95% CI 1.65-3.85; P<0.001), liver disease (OR=2.39; 95% 95% CI: 1.21-4.35; P<0.01), heart failure (OR=2.37; 95% CI:1.57-3.50; P<0.001), and endocrine disorder (OR=1.61; 95% CI: 1.142.24; P<0.01), after adjusting for age and sex. No significant differences in the prevalence of other disease categories were detected (Table 2; Figure 1).

Discussion In this large population-based nationwide screening study with over 75,000 participants and detailed nationwide information on comorbidities, we analyzed the differences between individuals with clinically diagnosed and screened MGUS. We found that individuals with a clinical diagnosis of MGUS had a 37% higher number of comorbidities compared with those found to have MGUS as part of our screening study and were two- to three-fold more likely to have been diagnosed with certain medical conditions. Our study shows that individuals diagnosed with MGUS during work-up for other medical conditions have more comorbidities than individuals diagnosed through systematic screening. This shows that studies comparing

Table 1. Demographic and laboratory features of clinical monoclonal gammopathy of undetermined significanc (MGUS) and screened MGUS. Clinical MGUS N=240

Screened MGUS N=3,112

Median age in years (range)a

73.5 (42-96)

69 (41-100)

***

Male sex, N (%)b

122 (50.8)

1,671 (53.7)

Residence in capital area, N (%)b

164 (68.3)

1,886 (60.6)

* OR (95% CI)

Non-IgG MGUS, N (%)

75 (31.2)

1,044 (33.5)

0.87 (0.65-1.15)

Pathological FLC ratio, N (%)c

86 (35.8)

908 (29.2)

1.29 (0.98-1.70)

M-protein ≥1.5 g/dL, N (%)c

8 (3.3)

45 (1.4)

1.92 (0.83-3.92) Δ (95% CI)

Mean number of risk factors of progression (SD) Mean M-protein concentration, g/dL (SD)d

0.80 (0.78)

0.74 (0.73)

0.04 (-0.06-0.14)

0.51 (0.44)

0.34 (0.33)

0.17 (0.13-0.22) ***

An odds ratio (OR) >1.00 and Δ >0.00 indicates increased risk for the clinical monoclonal gammopathy of undetermined significanc (MGUS) group. A P value of <0.05 was considered indicative of statistical significance. An FLC ratio <0.26 or >1.65 was considered pathological. aAssessed by t test. bAssessed by χ2 test. cAssessed by logistic regression, adjusted for age and sex. dAssessed by linear regression, adjusted for age and sex. *Indicates P<0.05, **P<0.01, ***P<0.001. CI: confidence interval; Ig: immunoglobulin; FLC: free light chain; M: monoclonal; SD: standard deviation. Haematologica | 108 December 2023

3394

ARTICLE - Clinical versus screened MGUS disease associations

A. Sigurbersdóttir et al.

Table 2. Prevalence of comorbidities in clinical monoclonal gammopathy of undetermined significanc (MGUS) and screened MGUS. Clinical MGUS N=240

Screened MGUS N=3,112

Mean N (SD)

Δ (95% CI)

3.23 (2.05)

2.36 (1.78)

0.68 (0.46-0.90)

***

N (%)

OR (95% CI)

Any comorbidityb

227 (94.6)

2,704 (86.9)

2.24 (1.30-4.19)

Rheumotological diseaseb

47 (19.6)

225 (7.2)

2.98 (2.08-4.20)

***

Neurological diseaseb

35 (14.6)

169 (5.4)

2.77 (1.84-4.06)

***

Chronic kidney diseaseb

31 (12.9)

154 (4.9)

2.55 (1.65-3.85)

***

Liver diseaseb

12 (5.0)

66 (2.1)

2.39 (1.21-4.35)

Heart failureb

37 (15.4)

198 (6.4)

2.37 (1.57-3.50)

***

Dementiab

8 (3.3)

40 (1.3)

2.16 (0.91-4.55)

Endocrine diseaseb

52 (21.7)

426 (13.7)

1.61 (1.14-2.24)

Arrhythmiab

59 (24.6)

535 (17.2)

1.38 (0.99-1.90)

Hypertensionb

163 (67.9)

1,784 (57.3)

1.34 (1.00-1.82)

Chronic lung diseaseb

69 (28.8)

705 (22.7)

1.31 (0.97-1.75)

Diabetesb

32 (13.3)

329 (10.6)

1.24 (0.83-1.81)

Pancreatic diseaseb

1 (0.4)

9 (0.3)

1.22 (0.07-6.60)

Cerebrovascular diseaseb

22 (9.2)

223 (7.2)

1.12 (0.68-1.75)

Cancerb

47 (19.6)

503 (16.2)

1.11 (0.78-1.54)

Peripheral vascular diseaseb

17 (7.1)

179 (5.8)

1.10 (0.63-1.81)

Ischemic heart diseaseb

56 (23.3)

624 (20.1)

1.06 (0.76-1.47)

Psychological diseaseb

85 (35.4)

1,122 (36.1)

0.99 (0.75-1.31)

Inflammatory bowel diseaseb

2 (0.8)

40 (1.3)

0.66 (0.11-2.19)

Comorbiditiesa

An odds ratio (OR) >1.00 and Δ >0.00 indicates increased risk for the clinical monoclonal gammopathy of undetermined significance (MGUS) group. A P value of <0.05 was considered indicative of statistical significance. aAssessed by linear regression, adjusted for age and sex. bAssessed by logistic regression, adjusted for age and sex. *P<0.05, **P<0.01, ***P<0.001. CI: confidence interval

disease associations with MGUS require data from screened populations to limit selection bias. Individuals with clinical MGUS were on average 4.5 years older and had 0.17 g/dL higher M-protein concentration than those with screened MGUS. Since, by definition, those with clinical MGUS had had MGUS for a longer time, and because M-protein generally increases over time, these results were expected.24 Both groups had a relatively low mean M-protein concentration (0.51 g/dL clinical vs. 0.34 g/dL screened), a difference that, while statistically significant, is unlikely to be clinically important. Furthermore, no statistically significant differences in the prevalence of individual risk factors of MGUS progression, or in the cumulated number of risk factors, were detected in our study. According to previous studies, the risk of MGUS progression to malignancy is low at 0.5-1.5% per year.2,6,24,25 The findings of this study indicate that previous estimates of MGUS progression have not been affected by selection bias and that clinically diagnosed

MGUS is similar to MGUS diagnosed by screening. The clinical MGUS group had statistically a significantly higher number of comorbidities compared to the screened MGUS group and was more likely to have at least one comorbidity, suggesting higher disease burden in general in clinically diagnosed individuals compared with the total MGUS population. This has previously been hypothesized but evidence has lacked until now.16 This important finding suggests selection bias in clinical cohorts of MGUS. In particular, chronic kidney disease, endocrine disorders, neurological disease, liver disease and rheumatological disease were more prevalent among those with clinically diagnosed MGUS. This is clinically important since various disorders of these disease categories have previously been associated with MGUS, including glomerulonephritis,26,27 immune deposition kidney diseases,28 hyperparathyroidism,29 peripheral neuropathies,13 hepatitis C,30,31 and rheumatoid arthritis.9 Due to these previously reported associations with MGUS, and often indistinct clinical signs

Haematologica | 108 December 2023

3395

ARTICLE - Clinical versus screened MGUS disease associations

A. Sigurbersdóttir et al.

Figure 1. Comorbidity prevalence of the study cohort. A P value of <0.05 was considered indicative of statistical significance. Assessed by logistic regression models. All analyses were adjusted for age and sex. *P<0.05, **P<0.01, ***P<0.001. MGUS: monoclonal gammopathy of undetermined significance.

and symptoms, affected individuals are more likely to undergo serum protein blood testing (e.g., SPEP, immunofixation, and FLC assays) during clinical workup, leading to the diagnosis of MGUS. For some of these conditions, SPEP is even advised during workup, according to clinical guidelines.32,33 This might also explain the higher prevalence of heart failure in the clinical MGUS group, as individuals suspected of amyloidosis causing their symptoms usually undergo SPEP and FLC assays.34 These differences suggest that at least some of the previously reported disease associations with MGUS have been overestimated as the evidence was based on clinical MGUS cohorts. Further studies to estimate true associations between MGUS and other diseases are needed on screened populations. Furthermore, multifaceted data, including information on laboratory markers and medications, in addition to careful, complex data analyses suiting each association, are essential to estimate causal relationships. Our study has several strengths. It is based on the largest screening study on MGUS to date, and it is the first one that is both population-based and nationwide. The high participation rate in the iStopMM study (54.3% of the Icelandic population ≥40 years of age) makes the study cohort highly representative of the general population. Extensive information was gathered on all participants, including M-protein concentration, M-protein isotype, and

FLC ratios, which was used to confirm all MGUS diagnoses. By crosslinking our data to large national registries, where disease diagnoses are recorded prospectively with very high completeness and accuracy, we also gathered high-quality information on participants’ comorbidities, including cancer diagnoses. In addition, data was collected prospectively and in the same manner for all participants regardless of exposure status. Our study also has some important limitations. Relatively few individuals had some of the comorbidities assessed, which may have affected the statistical power to ascertain potential association for those diseases. Additionally, since there were multiple statistical tests done in our study, some associations found may have been spurious. However, due to the hypothesis-generating nature of the study, we did not adjust for multiple testing.17 Finally, the Icelandic population is ethnically and genetically homogenous. Furthermore, the Icelandic health care system is universal, which makes access to full range health care services available for the entire population. However, this should not affect the results of the higher number of comorbidities among those with MGUS diagnosed in the clinical setting compared with those diagnosed through screening, although the prevalence of certain comorbidities may differ between countries and different health care systems.

Haematologica | 108 December 2023

3396

ARTICLE - Clinical versus screened MGUS disease associations In summary, we have shown that there is no meaningful difference in the severity of the precursor condition in individuals with a diagnosis of MGUS in the clinical setting compared with those diagnosed with MGUS in our screening study. However, there is a significant difference with regards to the presence of comorbidities. Those with a clinical diagnosis of MGUS have a higher mean number of comorbidities and are more likely to have been diagnosed with certain medical conditions. Our findings emphasize the fact that MGUS cohorts, based on clinically diagnosed populations, are inherently biased towards individuals with more comorbidities, compared to cohorts consisting of screened individuals. Furthermore, our findings support that selection bias has affected the results of many previous studies reporting on MGUS and various medical associations, and that at least some associations may not be biologically true or have been significantly overestimated. Going forward, it is imperative that screened MGUS cohorts are used to evaluate the epidemiological and biological implications of MGUS and that studies based on clinical MGUS cohorts should be interpreted with caution. Disclosures BGMD discloses consultancy for Amgen, Janssen, Celgene and Takeda. SH is the director of the Binding Site. OL has received research funding from National Institutes of Health, National Cancer Institute, US Food and Drug Administration, Multiple Myeloma Research Foundation, International Myeloma Foundation, Leukemia and Lymphoma Society, Perelman Family Foundation, Rising Tide Foundation, Amgen, Celgene, Janssen, Takeda, Glenmark, Seattle Genetics and Karyopharm; has received honoraria/advisory board fees from Adaptive, Amgen, Binding Site, BMS, Celgene, Cellectis, Glenmark, Janssen, Juno and Pfizer; and serves on Independent Data Monitoring Committees for clinical trials by Takeda, Merck, Janssen and Theradex. SYK has received research funding from the International Myeloma Foundation, European Research Council, Icelandic Center for Research, Amgen, Celgene.

A. Sigurbersdóttir et al.

The remaining authors have no conflicts of interest to disclose. Contributions AYS, SR, ST, and SYK designed the study. AYS and SR performed the analyses. AYS, SR, ST, TLJ, and SYK were involved in the interpretation of the results. AYS, SR, and SYK wrote the manuscript, with additional input from the other co-authors. All authors approved the submission of the final manuscript. Acknowledgments Screening tests were performed by the Binding Site. Crosslinking of study data to national registries was performed by the Icelandic Directorate of Health and the Icelandic Cancer Society. Special thanks go to the thousands of Icelanders who provided their informed consent for participation in the study. Funding The iStopMM study is funded by the Black Swan Research Initiative by the International Myeloma Foundation and the Icelandic Center for Research (grant no. 173857), the European Research Council (ERC) under the European Union‘s Horizon 2020 research and innovation programme (grant no. 716677). Additional funding was provided by the University of Iceland, Landspítali University Hospital, and the Icelandic Cancer Society. Data-sharing statement The iStopMM study has been approved by the Icelandic National Bioethics Committee. According to Icelandic law on ethics in research, and due to data privacy regulations and per informed consent from participants in this study, we are not allowed to share patient-level data used for this study. We, however, encourage researchers or parties interested in collaborations for non-commercial use to apply to the corresponding author. All requests will be reviewed by the iStopMM team to verify whether data sharing is within the restrictions of the study‘s ethical approval.

References 1. Kyle RA, Therneau TM, Rajkumar SV, et al. Prevalence of monoclonal gammopathy of undetermined significance. N Engl J Med. 2006;354(13):1362-1369. 2. Dispenzieri A, Katzmann JA, Kyle RA, et al. Prevalence and risk of progression of light-chain monoclonal gammopathy of undetermined significance: a retrospective population-based cohort study. Lancet. 2010;375(9727):1721-1728. 3. Murray D, Kumar SK, Kyle RA, et al. Detection and prevalence of monoclonal gammopathy of undetermined significance: a study utilizing mass spectrometry-based monoclonal immunoglobulin rapid accurate mass measurement. Blood Cancer J. 2019;9(12):102.

4. Weiss BM, Hebreo J, Cordaro DV, et al. Increased serum free light chains precede the presentation of immunoglobulin light chain amyloidosis. J Clin Oncol. 2014;32(25):2699-2704. 5. Landgren O, Kyle RA, Pfeiffer RM, et al. Monoclonal gammopathy of undetermined significance (MGUS) consistently precedes multiple myeloma: a prospective study. Blood. 2009;113(22):5412-5417. 6. Kyle RA, Therneau TM, Rajkumar SV, et al. Long-term follow-up of IgM monoclonal gammopathy of undetermined significance. Blood. 2003;102(10):3759-3764. 7. Kristinsson SY, Pfeiffer RM, Björkholm M, et al. Arterial and venous thrombosis in monoclonal gammopathy of

Haematologica | 108 December 2023

3397

ARTICLE - Clinical versus screened MGUS disease associations undetermined significance and multiple myeloma: a population-based study. Blood. 2010;115(24):4991-4998. 8. Kristinsson SY, Tang M, Pfeiffer RM, et al. Monoclonal gammopathy of undetermined significance and risk of infections: a population-based study. Haematologica. 2012;97(6):854-858. 9. Lindqvist EK, Goldin LR, Landgren O, et al. Personal and family history of immune-related conditions increase the risk of plasma cell disorders: a population-based study. Blood. 2011;118(24):6284-6291. 10. Sigurbergsdóttir A, Love TJ, Kristinsson SY. Autoimmunity, infections, and the risk of monoclonal gammopathy of undetermined significance. Front Immunol. 2022;13:876271. 11. Kristinsson SY, Tang M, Pfeiffer RM, et al. Monoclonal gammopathy of undetermined significance and risk of skeletal fractures: a population-based study. Blood. 2010;116(15):2651-2655. 12. Thorsteinsdottir S, Lund SH, Lindqvist EK, et al. Bone disease in monoclonal gammopathy of undetermined significance: results from a screened population-based study. Blood Adv. 2017;1(27):2790-2798. 13. Rögnvaldsson S, Steingrímsson V, Turesson I, Björkholm M, Landgren O, Kristinsson SY. Peripheral neuropathy and monoclonal gammopathy of undetermined significance: a population-based study including 15,351 cases and 58,619 matched controls. Haematologica. 2020;105(11):2679-2681. 14. Kristinsson SY, Björkholm M, Andersson TM, et al. Patterns of survival and causes of death following a diagnosis of monoclonal gammopathy of undetermined significance: a population-based study. Haematologica. 2009;94(12):1714-1720. 15. Fermand JP, Bridoux F, Dispenzieri A, et al. Monoclonal gammopathy of clinical significance: a novel concept with therapeutic implications. Blood. 2018;132(14):1478-1485. 16. Bida JP, Kyle RA, Therneau TM, et al. Disease associations with monoclonal gammopathy of undetermined significance: a population-based study of 17,398 patients. Mayo Clin Proc. 2009;84(8):685-693. 17. Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology. 1990;1(1):43-46. 18. Rögnvaldsson S, Love TJ, Thorsteinsdottir S, et al. Iceland screens, treats, or prevents multiple myeloma (iStopMM): a population-based screening study for monoclonal gammopathy of undetermined significance and randomized controlled trial of follow-up strategies. Blood Cancer J. 2021;11(5):94. 19. Rajkumar SV, Dimopoulos MA, Palumbo A, et al. International Myeloma Working Group updated criteria for the diagnosis of multiple myeloma. Lancet Oncol. 2014;15(12):e538-e548. 20. Rögnvaldsson S, Long TE, Thorsteinsdottir S, Love TJ, Kristinsson SY. Validity of chronic disease diagnoses in Icelandic healthcare registries. Scand J Public Health. 2023;51(2):173-178. 21. WHO. International statistical classification of diseases and

A. Sigurbersdóttir et al.

related health problems, 10th revision (ICD-10). World Health Organization; 2016. 22. R Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2019. https://www.r-project.org. Accessed April 20, 2023. 23. Kyle RA, Durie BGM, Rajkumar SV, et al. Monoclonal gammopathy of undetermined significance (MGUS) and smoldering (asymptomatic) multiple myeloma: IMWG consensus perspectives risk factors for progression and guidelines for monitoring and management. Leukemia. 2010;24(6):1121-1127. 24. Kyle RA, Therneau TM, Rajkumar SV, et al. A long-term study of prognosis in monoclonal gammopathy of undetermined significance. N Engl J Med. 2002;346(8):564-569. 25. Turesson I, Kovalchik SA, Pfeiffer RM, et al. Monoclonal gammopathy of undetermined significance and risk of lymphoid and myeloid malignancies: 728 cases followed up to 30 years in Sweden. Blood. 2014;123(3):338-345. 26. Nasr SH, Satoskar A, Markowitz GS, et al. Proliferative glomerulonephritis with monoclonal IgG deposits. J Am Soc Nephrol. 2009;20(9):2055-2064. 27. Leung N, Bridoux F, Hutchison CA, et al. Monoclonal gammopathy of renal significance: when MGUS is no longer undetermined or insignificant. Blood. 2012;120(22):4292-4295. 28. Nasr SH, Valeri AM, Cornell LD, et al. Renal monoclonal immunoglobulin deposition disease: a report of 64 patients from a single institution. Clin J Am Soc Nephrol. 2012;7(2):231-239. 29. Arnulf B, Bengoufa D, Sarfati E, et al. Prevalence of monoclonal gammopathy in patients with primary hyperparathyroidism: a prospective study. Arch Intern Med. 2002;162(4):464-467. 30. Andreone P, Zignego AL, Cursaro C, et al. Prevalence of monoclonal gammopathies in patients with hepatitis C virus infection. Ann Intern Med. 1998;129(4):294-298. 31. Brown LM, Gridley G, Check D, Landgren O. Risk of multiple myeloma and monoclonal gammopathy of undetermined significance among white and black male United States veterans with prior autoimmune, infectious, inflammatory, and allergic disorders. Blood. 2008;111(7):3388-3394. 32. KDIGO 2021 Clinical practice guideline for the management of glomerular diseases. Kidney Int. 2021;100(4s):S1-S276. 33. England JD, Gronseth GS, Franklin G, et al. Practice parameter: the evaluation of distal symmetric polyneuropathy: the role of laboratory and genetic testing (an evidence-based review). Report of the American Academy of Neurology, the American Association of Neuromuscular and Electrodiagnostic Medicine, and the American Academy of Physical Medicine and Rehabilitation. PM R. 2009;1(1):5-13. 34. Gertz MA. Immunoglobulin light chain amyloidosis: 2022 update on diagnosis, prognosis, and treatment. Am J Hematol. 2022;97(6):818-829.

Haematologica | 108 December 2023

3398

ARTICLE - Plasma Cell Disorders

Changing trends in the risk factors for second primary malignancies after autologous stem cell transplantation for multiple myeloma before and after the introduction of proteasome inhibitors and immunomodulatory drugs Hiroyuki Takamatsu,1 Tomohiro Matsuda,2 Shohei Mizuno,3 Tsutomu Takahashi,4 Shin-ichi Fuchida,5 Ichiro Hanamura,3 Keisuke Kataoka,6,7 Nobuhiro Tsukada,8 Morio Matsumoto,9 Akira Hangaishi,10 Noriko Doki,11 Naoyuki Uchida,12 Masashi Sawa,13 Yumiko Maruyama,14 Shingo Kurahashi,15 Koji Nagafuji,16 Yoriko Harazaki,17 Shinichi Kako,18 Shinsuke Iida,19 Tatsuo Ichinohe,20 Yoshinobu Kanda,18,21 Yoshiko Atsuta22,23 and Kazutaka Sunami24 Department of Hematology, Kanazawa University, Kanazawa; 2Division of International Health Policy Research, National Cancer Center Institute for Cancer Control, Tokyo; 3Division of Hematology, Department of Internal Medicine, Aichi Medical University, Nagakute; 4 Department of Hematology, Shimane University Hospital, Izumo; 5Department of Hematology, Japan Community Health care Organization Kyoto Kuramaguchi Medical Center, Kyoto; 6Division of Molecular Oncology, National Cancer Center Research Institute, Tokyo; 7Division of Hematology, Department of Medicine, Keio University School of Medicine, Tokyo; 8Division of Hematology, Japanese Red Cross Medical Center, Tokyo; 9Department of Hematology, National Hospital Organization Shibukawa Medical Center, Shibukawa; 10 Department of Hematology, National Center for Global Health and Medicine, Tokyo; 11 Hematology Division, Tokyo Metropolitan Cancer and Infectious Diseases Center, Komagome Hospital, Tokyo; 12Department of Hematology, Federation of National Public Service Personnel Mutual Aid Associations, Toranomom Hospital, Tokyo; 13Department of Hematology and Oncology, Anjo Kosei Hospital, Anjo; 14Department of Hematology, University of Tsukuba Hospital, Tsukuba; 15Division of Hematology and Oncology, Toyohashi Municipal Hospital, Toyohashi; 16Division of Hematology and Oncology, Department of Medicine, Kurume University Hospital, Kurume; 17Division of Hematology, Miyagi Cancer Center, Natori; 18Division of Hematology, Jichi Medical University Saitama Medical Center, Saitama; 19Division of Hematology and Oncology, Nagoya City University Hospital, Nagoya; 20 Department of Hematology and Oncology, Research Institute for Radiation Biology and Medicine, Hiroshima University, Hiroshima; 21Division of Hematology, Department of Medicine, Jichi Medical University, Shimotsuke; 22Japanese Data Center for Hematopoietic Cell Transplantation, Nagakute; 23Department of Registry Science for Transplant and Cellular Therapy, Aichi Medical University School of Medicine, Nagakute and 24Department of Hematology, National Hospital Organization Okayama Medical Center, Okayama, Japan

Correspondence: H. Takamatsu takamaz@staff.kanazawa-u.ac.jp Received: Accepted: Early view:

March 27, 2023. July 10, 2023. July 20, 2023.

Abstract The incidence of second primary malignancies (SPM) in long-term survivors of multiple myeloma (MM) is increasing because of increased life expectancy. We retrospectively analyzed the risk factors for SPM in patients with MM after autologous stem cell transplantation (ASCT) before and after the introduction of proteasome inhibitors and immunomodulatory drugs (IMiDs). In total, 2,340 patients newly diagnosed with MM who underwent ASCT between 1995 and 2016 were enrolled in this study. Forty-three patients developed SPM (29 solid, 12 hematological, and 2 unknown tumors), with cumulative incidence rates of 0.8% and 2.5% at 24 and 60 months, respectively. The cumulative incidence rates of hematological and solid SPM at 60 months were 0.8% and 1.8%, respectively. The overall survival (OS) rate at 60 months after ASCT was 62.9% and the OS rates after the diagnosis of SPM at 24 months were 72.2% for hematological SPM and 70.9% for solid SPM. Multivariate analysis revealed that the use of IMiDs (P=0.024) and radiation (P=0.002) were significant independent risk factors for SPM. The probabilities of developing SPM and death due to other causes (mainly MM) at 60 months were 2.5% and 36.5%, respectively, indicating that the risk of SPM was lower than that of death from MM. Furthermore, SPM between the pre-novel and novel agent eras (ASCT between 2007 and 2016) groups significantly increased (1.9% vs. 4.3% at 60 months; P=0.022). The early occurrence of SPM after ASCT should be monitored cautiously.

Haematologica | 108 December 2023

3399

ARTICLE - Risk factors for SPM after ASCT for MM

H. Takamatsu et al.

Despite the development of various novel agents, such as proteasome inhibitors (PI) and immunomodulatory drugs (IMiDs), autologous stem cell transplantation (ASCT) is the gold standard treatment for transplant-eligible (TE) patients with multiple myeloma (MM).1-4 Because patients with MM survive longer, the incidence rate of second primary malignancies (SPM) in long-term survivors of MM is increasing. To date, only a few studies have evaluated SPM in real-world patients,5-9 particularly in Asian patients with MM.10-12 In this study, we analyzed Japanese patients with MM who underwent ASCT using large registry data.

were used to calculate hazard ratios (HR) with 95% confidence intervals (CI) for all variables. Multivariate analyses were performed by entering all variables associated with survival into the Cox proportional hazards model or variables associated with SPM into the Fine-Gray proportional hazards model. Owing to the small incidence rate of SPM, the factors of P value <0.1 were entered into multivariate analysis. All statistical analyses were performed using the EZR software package (Saitama Medical Center, Jichi Medical University, Saitama, Japan),17 which is a graphical user interface for R version 3.4.0 (R Foundation for Statistical Computing, Vienna, Austria). P<0.05 was considered statistically significant.

Methods

Results

Data source In this retrospective observational study, ASCT data from the Registry of the Japanese Society for Transplantation and Cellular Therapy (JSTCT) and the Japanese Data Center for Hematopoietic Cell Transplantation were collected and analyzed. More than 99% of all transplant centers in Japan report and update their outcomes annually. As the registry data comprised anonymized clinical information, patient consent was not required for registration. This study was approved by the Data Management Committee of the JSTCT and the Institutional Review Board of Kanazawa University (no. 2019-118 [3163]). In total, 2,340 patients with newly diagnosed MM who underwent ASCT between 1995 and 2016 and whose clinical data were sufficient for this analysis were enrolled in this study. Patients with a history of solid tumors were excluded from the analysis. Cancer incidence and survival data in the Japanese general population were estimated using population-based cancer registries through the Monitoring of Cancer Incidence in Japan project conducted by the Japan Cancer Surveillance Research Group.13 Cancer type was classified according to the International Classification of Diseases, tenth revision.

Patients’ characteristics A total of 2,340 (males, 1,329 [56.8%]; females, 1,011 [43.2%]) patients newly diagnosed with MM were extracted from the database. The median age of the patients was 58 (range, 22–72) years at ASCT. Immunoglobulin (Ig) G, IgA, IgD, IgE, IgM, Bence Jones protein, non-secreting, and unknown antibodies were observed in 1,340 (57.3%), 452 (19.3%), 63 (2.7%), three (0.1%), six (0.3%), 416 (17.8%), 38 (1.6%), and 22 (1.0%) patients, respectively. International Staging System (ISS) stages 1, 2, and 3 were observed in 774 (33.1%), 825 (35.3%), and 455 (19.4%) patients, respectively. ISS staging was not assessed in 286 (12.2%) patients. In total, 1,908 (81.5%) and 432 (18.5%) patients received single melphalan 200 mg/m2 (MEL 200) and double MEL 200, respectively, as a conditioning regimen before ASCT. Moreover, 771 (32.9%) and 1,569 (67.1%) patients underwent ASCT between 1995 and 2006 (pre-novel agent era) and between 2007 and 2016 (novel agent era), respectively. In total, 1,562 (66.8%), 977 (41.8%), and 904 (38.6%) patients received PI (bortezomib [n=1562], carfilzomib [n=14], and ixazomib [n=11]), IMiDs (thalidomide [n=185], lenalidomide [n=852] and pomalidomide [n=41]), and both PI and IMiDs, respectively. Meanwhile, 131 (5.6%) patients received radiation treatment, and 50 (2.1%) received allogeneic stem cell transplantation post-ASCT. The disease statuses at ASCT were as follows: 366 (15.6%), stringent complete response (CR)/CR; 608 (26.0%), very good partial response; 1,122 (48.0%), partial response; 164 (7.0%), stable disease; 55 (2.4%), progressive disease; and 25 (1.1%), unknown (Table 1).

Introduction

Statistical analyses Categorical and continuous variables were compared using Fisher’s exact test and the Mann–Whitney U test, respectively. Overall survival (OS) was calculated from the time of ASCT or diagnosis of SPM using the Kaplan-Meier method and compared between groups using a log-rank test. The probabilities of SPM and death were estimated based on cumulative incidence methods and compared between groups using the Gray test, considering death without SPM or SPM without death as competing events.14,15 Multivariate analysis for OS was performed using the Cox proportional hazards model, whereas multivariate analysis for SPM was performed using the Fine-Gray regression model.16 The Cox proportional and Fine-Gray proportional hazards models

Incidence rates and types of second primary malignancies The median follow-up period after ASCT was 24 (range, 0–218) months (Online Supplementary Figure S1). Fortythree patients in this cohort developed SPM, with cumulative incidence rates of 0.8% (95% CI: 0.4-1.2) and 2.5% (95% CI: 1.7-3.6) at 24 and 60 months, respectively, and

Haematologica | 108 December 2023

3400

ARTICLE - Risk factors for SPM after ASCT for MM

H. Takamatsu et al.

Table 1. Baseline characteristics and outcomes of 2,340 transplant-eligible patients. Patients with SPM N=43 unknown type, N=2

Patients without SPM N=2297

Hematologic malignancies, N=12

Non-hematological cancers, N=29

Median age in years (range) at ASCT

60.5 (54-69)

58 (42-70)

58 (22-72)

0.094

>65 years of age at diagnosis, N (%)

1 (8.3)

3 (10.3)

220 (9.6)

1.000

Male, N (%)

5 (41.7)

18 (62.1)

1304 (56.8)

0.878

M-protein isotype, N (%) IgG IgA IgD IgE IgM Light chain only Non-secreting Unknown

11 (91.7) 1 (8.3) 0 0 0 0 0 0

17 (58.6) 7 (24.1) 0 0 0 4 (13.8) 1 (3.4) 0

1312 (57.1) 444 (19.3) 63 (2.7) 3 (0.1) 6 (0.3) 410 (17.8) 37 (1.6) 22 (1.0)

Light chain type, N (%) κ λ Unknown

9 (75.0) 3 (25.0) 0

21 (72.4) 6 (20.7) 2 (6.9)

1384 (60.3) 828 (36.0) 85 (3.7)

ISS at diagnosis, N (%) I II III Unknown

5 (41.7) 2 (16.7) 3 (25.0) 2 (16.7)

10 (34.5) 10 (34.5) 3 (10.3) 6 (20.7)

759 (33.0) 813 (35.4) 449 (19.5) 276 (12.0)

0.878

0.173

0.154

Conditioning regimen Single melphalan 200 mg/m2 Tandem melphalan 200 mg/m2

10 (83.3) 2 (16.7)

20 (69.0) 9 (31.0)

1878 (81.8) 419 (18.2)

ASCT year between 1995 and 2006 between 2007 and 2016

3 (25.0) 9 (75.0)

18 (62.1) 11 (37.9)

748 (32.6) 1549 (67.4)

PI and/or IMiDs Tx during induction and post-ASCT, N (%) PI without IMiDs based IMiDs without PI based PI + IMiDs based

9 (75.0) 5 (41.7) 0 4 (33.3)

13 (44.8) 3 (10.3) 5 (17.2) 5 (17.2)

1612 (70.2) 650 (28.3) 67 (2.9) 895 (39.0)

0.028 0.175 0.002 0.017

Radiation Tx

2 (16.7)

4 (13.8)

125 (5.4)

0.030

1 (3.4)

49 (2.1)

0.608

Allo-SCT post-ASCT

0.071

0.005

Pre-ASCT response, N (%) sCR/CR VGPR PR SD PD Unknown

0.596 0 3 (25.0) 8 (66.7) 1 (8.3) 0 0

4 (13.8) 7 (24.1) 13 (44.8) 3 (10.3) 1 (3.4) 1 (3.4)

362 (15.8) 597 (26.0) 1100 (47.9) 160 (7.0) 54 (2.4) 24 (1.0)

Cause of death, N (%) MM SPM Others

3 (100) 0 3 (100) 0

14 (100) 5 (35.7) 8 (57.1) 1 (7.1)

583 (100) 495 (84.9) 0 88 (15.1)

<0.0001

SPM: second primary malignancy; ASCT: autologous stem cell transplantation; ISS: International Staging System; Tx: therapies; PI: proteasome inhibitor; IMiDs: immunomoduratory drugs; sCR: stringent complete response; CR: complete response; VGPR: very good partial response; PR: partial response; SD: stable disease; PD: progressive disease; MM: multiple myeloma. *Patients with SPM vs. without SPM. Haematologica | 108 December 2023

3401

ARTICLE - Risk factors for SPM after ASCT for MM

H. Takamatsu et al.

these patients had no history of solid cancer before ASCT. Twenty-nine solid (7, lung; 4, stomach; 3, breast; 2, liver; 2, pancreas; 2, colon; 1, uterus; 1, thyroid gland; 2, bladder; 1, tongue; 2, sarcoma; 1 vulvar; and 1, basal cell carcinoma [BCC]), 12 hematological (8, myelodysplastic syndrome [MDS]; 1, acute leukemia; and 3, lymphoproliferative disorders), and two unknown tumors were observed (Table 2). The cumulative incidence rates of hematological and solid SPM at 60 months were 0.8% and 1.8%, respectively. Risk factors for second primary malignancy The risk factors for SPM were analyzed (Figure 1; Table 3), including age at ASCT (≤65 or >65 years), sex, PI/IMiD treatment, use of radiation, single/double ASCT, and period of ASCT (1995–2006 or 2007–2016). Because bortezomib, thalidomide, and lenalidomide were released for relapsed/refractory MM treatment in Japan in December 2006, February 2009, and July 2010, respectively, we categorized the patients into two treatment cohorts: the prenovel (1995–2006) and novel (2007–2016) agent eras. Univariate analysis showed that the risk of SPM was increased in the novel agent era (Figure 1A; 1.9% vs. 4.3% at 60 months; P=0.022), and IMiDs without PI treatment (Figure 1E; Table 3; P=0.029) and the use of radiation (Figure 1F; Table 3; P=0.003) were also significant risk factors for SPM. Multivariate analysis also revealed that IMiDs without PI treatment (HR=2.738; 95% CI: 1.142–6.568; P=0.024) and the use of radiation (HR=4.151; 95% CI: 1.687–10.210; P=0.002) were significant independent risk factors for SPM. In contrast, PI without IMiD treatment were not a risk factor for SPM (Figure 1D; Table 3). Survival rates and causes of death after autologous stem cell transplantation and diagnosis of second primary malignancies The OS rate at 60 months after ASCT was 62.9% (Figure 2A), and the OS rate between the pre-novel (1995-2006) and novel agent (2007-2016) era groups significantly improved (59.2% vs. 69.5% at 60 months; P<0.0001; Figure 2A) after ASCT. Furthermore, the use of PI and/or IMiDs improved OS (Online Supplementary Figure S2C). However, there were no differences in OS regarding age, sex, and single/tandem ASCT (Online Supplementary Figure S2A, B, D). Multivariate analysis revealed that the use of PI and/or IMiDs and tandem ASCT significantly improved OS (Table 3). The probabilities of developing SPM and death due to other causes (mainly MM) at 60 months were 2.5% and 36.5%, respectively (Figure 1A), indicating that the risk of SPM was lower than that of death from MM. Furthermore, Figure 2B shows a comparison of the OS rates between patients with and without SPM. There were no significant differences in the OS rates among hematological SPM, solid SPM, and no SPM cases.

Table 2. Second primary malignancy by type (N=43). Variable

Hematologic malignancy

Myelodysplastic syndrome

Acute leukemia

Post-transplant lymphoproliferative disorder

MTX-related lymphoproliferative disorders

Solid tumor

Lung

Stomach

Breast

Bladder

Colon

Liver

Pancreas

Sarcoma

Uterus

Thyroid gland

Tongue

Vulvar

Basal cell carcinoma

Type not reported

MTX: methotrexate.

The OS rates after the diagnosis of SPM at 24 months were 72.2% for hematological and 70.9% for solid SPM (median follow-up period, 23 months; Figure 2C). There was no significant difference in the OS rates post-SPM diagnosis between solid and hematological cancers (P=0.762). Comparison of adjusted survival probabilities of solid and hematological cancers between patients with second primary malignancy post-autologous stem cell transplantation and the general population in Japan In order to compare the OS rates between SPM post-ASCT and primary solid cancers (PSC) and primary hematological cancers (PHC) in Japan’s general population, we conducted a study where patients with PSC or PHC were matched with patients with SPM post-ASCT at a 20:1 ratio based on sex, age, and cancer type. We analyzed the OS rate following the diagnosis. Lymphoproliferative disorders and carcinoma in situ were excluded because there were no data in the population-based cancer registries. The OS rate in patients with solid SPM post-ASCT was comparable with that of PSC in the general Japanese population (Figure 2C, D). Owing to the small number of hematological SPM, it was difficult to compare the OS rates between hematological

Haematologica | 108 December 2023

3402

ARTICLE - Risk factors for SPM after ASCT for MM

H. Takamatsu et al.

SPM and PHC. Five primary MM, 13 SPM, and one bacterial infection were identified as the causes of death in 19 patients with SPM in this cohort.

Discussion The incidence rate of SPM in patients with MM in Japan in our study (0.8% [95% CI: 0.4-1.2] and 2.5% [95% CI: 1.73.6] at 24 and 60 months) was somewhat lower than that (5.3% [95% CI: 4.4-6.3] at 72 months post-ASCT) reported by Sahebi et al.6 Although the risk of SPM increased in the novel agent era group, the mortality rate of SPM was lower than that of other causes (primarily MM). Considering the

increase in the number of long-term survivors of MM, the occurrence of SPM should be monitored cautiously. Our multivariate analysis revealed that IMiDs without PI treatment (P=0.024) and the use of radiation (P=0.002) were significant independent risk factors for SPM. To the best of our knowledge, no studies have demonstrated that the use of radiation is a significant risk factor for SPM. However, radiation is a risk factor for cancer. Data on atomic bomb survivors demonstrated a linear relationship between cancer development and radiation dose,18 and patients with Hodgkin lymphoma usually receive chemotherapy and/or radiation therapy, which induce secondary cancer.19 Therefore, our result of radiation on SPM seems to be reasonable. However, to date, several reports have

Haematologica | 108 December 2023

3403

ARTICLE - Risk factors for SPM after ASCT for MM

H. Takamatsu et al.

Figure 1. Cumulative incidence rate of developing second primary malignancies (SPM) (black and red curves) versus death from all causes without occurrence of SPM (green and blue curves). According to (A) years of transplant, development of SPM: 1.9% (1995–2006) vs. 4.3% (2007–2016) at 60 months (P=0.022); (B) age at transplant; (C) sex; (D) proteasome inhibitors (PI) not immunomodulatory drugs (IMiDs) vs. other than PI not IMiDs; (E) IMiDs not PI vs. other than IMiDs not PI; (F) radiation, development of SPM: 2.1% (radiation [−]) vs. 13.8% (radiation [+]) at 60 months (P<0.001).

Table 3. Risk factors of second primary malignancy and predictor of overall survival. SPM

Overall survival

HR (95% CI) for SPM from univariate analysis

P for univariate analysis

HR (95% CI) for SPM from multivariate analysis

P for multivariate analysis

HR (95% CI) for OS from univariate analysis

P for univariate analysis

Age in years, ≤65 vs. 65<

1.591 (0.564-4.488)

0.380

1.160 (0.833-1.614)

0.381

1.247 (0.891-1.745)

0.198

Sex, female vs. male

1.097 (0.600-2.006)

0.760

1.122 (0.954-1.319)

0.164

1.144 (0.973-1.345)

0.105

Treatments of PI and IMiDs 1 PI (-) and IMiDs (-) 1.352 (0.594-3.081) PI (+) and IMiDs (-) 2.694 (1.108 - 6.550) PI (-) and IMiDs(+) 1.096 (0.483-2.486) PI (+) and IMiDs (+)

NA 0.470 0.029 0.830

1 1.183(0.495-2.828) 2.738 (1.142-6.568) 0.840 (0.368-1.917)

NA 0.710 0.024 0.680

1 0.543 (0.411-0.716) 0.713 (0.502-1.012) 0.785 (0.628-0.981)

NA <0.0001 0.059 0.034

1 0.479(0.360-0.638) 0.660 (0.462-0.941) 0.681 (0.537-0.863)

NA <0.000001 0.022 0.001

Local radiation therapy

3.657 (1.542-8.672)

0.003

4.151 (1.687-10.210)

0.002

1.413(0.989-2.019)

0.057

1.462 (1.017-2.100)

0.040

Single vs. tandem ASCT

0.849 (0.431-1.671)

0.630

0.904 (0.760-1.076)

0.258

0.788 (0.656-0.947)

0.011

Independent variables

HR P for (95% CI) multivariate for OS from analysis multivariate analysis

OS: overall survival; PI: proteasome inhibitors; IMiDs: immunomoduratory drugs; (+): administered; (-): not administered; HR: hazard ratio; CI: confidence interval; ASCT: autologous stem cell transplantation; NA: not applicable.

demonstrated that the incidence rate of SPM increases among patients with MM who receive lenalidomide.1,20-22 Recently, using the Myeloma XI trial data, Jones et al. reported that TE patients receiving lenalidomide maintenance had an SPM incidence rate of 12.2% at 7 years, compared with 5.8% in those being observed (P=0.003). They also demonstrated that the SPM incidence rate was higher in patients treated with lenalidomide at both induction and maintenance compared with single exposure

or no exposure,23 suggesting a dose-dependent risk of lenalidomide on SPM. Hematological SPM was almost all confined to lenalidomide-treated patients using the Myeloma XI data.23 Regarding patients with hematological SPM (n=12) in our cohort, only four of the 12 (33%) patients had received IMiDs before SPM diagnosis. Regarding the types of SPM, we mainly observed MDS in hematological SPM and lung, stomach, and breast cancers in solid cancer SPM. In contrast, Jones et al. demon-

Haematologica | 108 December 2023

3404

ARTICLE - Risk factors for SPM after ASCT for MM

H. Takamatsu et al.

strated that hematological malignancies (MDS or acute myeloid leukemia [AML]) and non-melanoma skin cancer (BCC or squamous cell carcinoma) were mostly confined to patients receiving lenalidomide maintenance.23 The incidence rates of solid tumors were similar between the lenalidomide maintenance and observation groups. The

types of solid tumors were prostate (incidence rate, 2%) and breast (incidence rate, 1%) cancers, but not gastric cancer. Engelhardt et al. also did not observe any gastric cancer in SPM cases.5 Although the trend in hematological SPM type (MDS/AML) in our cohort is almost the same as that reported in Western countries, the types of solid

Figure 2. Influence of second primary malignancies on overall survival rates. (A) Comparison of overall survival (OS) rates between pre-novel (1995–2006) and novel (2007–2016) agent eras. (B) OS rates according to second primary malignancy (SPM) occurrence after autologous stem cell transplantation. (C) OS rates of patients with SPM post-SPM occurrence. (D) OS rates of Japanese patients with primary cancer who were matched to patients with SPM in this cohort by age, sex, and cancer type. Haematologica | 108 December 2023

3405

ARTICLE - Risk factors for SPM after ASCT for MM

H. Takamatsu et al.

cancer SPM seem to differ from those in Western countries, suggesting a reflection of ethnic and environmental background. For example, stomach cancer is not as common in Western countries as in Asian countries, possibly owing to the infection rate of Helicobacter pylori in the stomach, and skin cancer is relatively rarer in Japan than in Western countries. Tzeng et al. also reported a high incidence rate of stomach cancer and low incidence rate of skin cancer among Taiwanese patients with MM.10 Compared with reports from Western countries,5,6,24-29 there are only a few reports on SPM in Asian countries.10-12 Ailawadhi et al. reported that the risk of developing SPM among patients varied depending on the patient’s ethnic background.7 Owing to the genetic and environmental differences between Asians and Westerners, SPM may be different. Reports from Western countries showed that the incidence rate of SPM at 60 months was between 4% and 11%,5,25,27-29 which is higher than that in our study (0.8% [95% CI: 0.4-1.2] and 2.5% [95% CI: 1.7-3.6%] at 24 and 60 months), particularly for hematological cancers (0.8% at 60 months). Conversely, the reports from Taiwan demonstrated that the incidence rate of SPM was 1.8 % (0.9% solid cancers and 0.9% hematological cancers) at 23 months10 and 1.8% (1.5% solid cancers and 0.3% hematological cancers) at 24 months;12 similarly, reports from Japan showed that the incidence rate of SPM was 4.7% (3.7% solid cancers and 1.0% hematological cancers)30 and 5.6% (4.0% solid cancers and 1.7% hematological cancers) at 60 months.11 According to these data, the incidence rate of SPM might be less among Asian patients with MM. Yamasaki et al. analyzed the risk factors for SPM among Japanese patients with MM. Multivariate analysis identified a history of high-dose cyclophosphamide use for peripheral blood stem cell harvest in TE patients with MM and aged >65 years at diagnosis or a history of adriamycin, lenalidomide, or thalidomide use in transplant-ineligible patients with MM as independent risk factors for SPM (P<0.001), and lenalidomide did not facilitate SPM development in TE patients with MM, all of whom received oral or high-dose melphalan.11 Of the 211 TE patients analyzed, only 18 received lenalidomide. Given the small number of cases, it would be challenging to evaluate the impact of lenalidomide on SPM. Interestingly, Liu et al. reported that contemporary treatment regimens using novel agents (mainly bortezomib) were associated with a lower risk of SPM than chemotherapy alone. However, in their cohort, thalidomide-containing regimens were used, lenalidomide was not used, and its effect on SPM was not revealed. Therefore, our study is the first to demonstrate that the use of IMiDs is a risk factor for SPM in TE Asian patients with MM. The prognosis of patients with MM after SPM diagnosis has also been reported. Cooper et al. reported the survival rate of 2,837 patients with MM diagnosed with SPM using ASCO CancerlinQTM analysis.31 They showed that patients with sec-

ondary AML had the worst prognosis, followed by those with lung cancer. In our data, we could not analyze the prognosis of patients with SPM based on the type of secondary cancer because of the small incidence rates. However, there was no evident difference in OS rates between patients with non-SPM and SPM when SPM were divided into hematological and solid SPM (Figure 2B). Our results are consistent with those of a previous study.8 There have been few reports that describe the prognostic comparisons of patients with SPM post-ASCT and the general population. Barth et al. examined the OS and causespecific survival and cumulative incidence function of cancer-related death among patients with MM with SPM of the breast, prostate, lung, colon/rectum, or bladder or melanoma using the population-based Surveillance, Epidemiology, and End Results Registry (2004-2015). For all studied cancers, except those of the lung, overall mortality was significantly higher among patients with MM than among controls (HR=1.84-2.81). However, the cumulative incidence function of cancer-related death did not differ (subHR=0.84-0.99).9 Tzeng et al. analyzed patients with MM in Taiwan using data from population-based insurance claims and demonstrated that the overall incidence rate of secondary malignancy was lower in the MM cohort than in the comparison cohort (93.6 vs. 104.5 per 10,000 person-years, incidence rate ratio=0.90; 95% CI: 0.78-1.04). However, the incidence rate of hematological malignancies was 11-fold greater in patients with MM (47.2 vs. 4.09 per 10,000 person-years) with an adjusted HR of 13.0 (95% CI: 7.79-21.6) compared with the comparison cohort. The relative risk of secondary malignancy was also high for myeloid leukemia (21.2 vs. 1.36 per 10,000 person-years).10 In our study, the OS rate in patients with SPM post-ASCT was compatible with that of primary cancers in the general Japanese population. This study had some limitations. First, we could not confirm the duration of induction and post-ASCT therapy, including the doses of medications and radiation. Second, we could not completely exclude the existence of a primary malignancy that led to SPM because there were no strict entry criteria to exclude primary malignancy at the entry of the patient. Third, the median follow-up period of 24 months after ASCT was relatively short for monitoring SPM. Not all occurrences of SPM were possibly captured within this timeframe in the registration because of a long latency before the development of SPM. Therefore, we must follow up for a longer period and estimate the risk of SPM. Fourth, we could not compare the patients who received ASCT with high-dose melphalan and those who were not exposed to high-dose melphalan. Recently, Richardson et al. reported that 5-year cumulative incidences of SPM were 10.4% in the triplet therapy (lenalidomide, bortezomib, and dexamethasone [RVD])-alone group and 10.7% in the transplantation group, and the incidence was similar in the two groups. Hematologic SPM

Haematologica | 108 December 2023

3406

ARTICLE - Risk factors for SPM after ASCT for MM

H. Takamatsu et al.

occurred in 2.5% in the RVD-alone group and 3.6% in the transplantation group, with AML or MDS reported in none of the patients in the RVD-alone group, as compared with 2.7% in the ASCT group (P=0.002).27 Finally, we could not obtain information on treatments for SPM to analyze their prognosis after SPM occurred. In conclusion, although the risk of SPM increased among patients who received IMiDs (P=0.024) and used radiation (P=0.002), the mortality rate of SPM was lower than that of other causes (primarily MM). Considering the increase in the number of long-term survivors of MM, the occurrence of SPM should be monitored cautiously. Disclosures HT received consulting fees from SRL; honoraria from Janssen Co. Ltd., Ono Pharmaceutical Co., Sanofi Pharmaceutical Co., and Bristol-Myers Squibb and research funding from Bristol-Myers Squibb. KK received honoraria from Ono Pharmaceutical, Eisai, Astellas Pharma, Novartis, Chugai Pharmaceutical, AstraZeneca, Sumitomo Pharma, Kyowa Kirin, Janssen Pharmaceutical, Takeda Pharmaceutical, Otsuka Pharmaceutical, SymBio Pharmaceuticals, Bristol Myers Squibb, Pfizer, Nippon Shinyaku, Daiichi Sankyo, Alexion Pharmaceuticals, AbbVie, Meiji Seika Pharma, Sanofi, Sysmex, Mundipharma, Incyte Corporation, and Kyorin Pharmaceutical; has equity ownership of Asahi Genomics and a patent for Genetic alterations, and received research funding from Otsuka Pharmaceutical, Chordia Therapeutics, Chugai Pharmaceutical, Takeda Pharmaceutical, Meiji Seika Pharma and a scholarship from Asahi Kasei Pharma, Eisai, Otsuka Pharmaceutical, Ono Pharmaceutical, Kyowa Kirin, Shionogi, Takeda Pharmaceutical, Sumitomo Dainippon Pharma, Chugai Pharmaceutical, Teijin Pharma, Japan Blood Products Organization, Mochida Pharmaceutical, JCR Pharmaceuticals, Nippon Shinyaku. MM received honoraria from Sanofi. S. Kako received honoraria from Chugai Pharmaceutical Co., Ltd. SI received honoraria from Janssen, Bristol-Myers Squibb, Sanofi, Takeda and Ono and Grant

support from Janssen, Bristol-Myers Squibb, Sanofi, Takeda, Ono, Chugai, Pfizer, Abbvie, Novartis, Daichi Sankyo, Celgene, Amgen, Kyowa Kirin, Glaxo SmithKlein and Alexion Pharma. YK received honoraria from Bristol-Myers Squibb, Novartis, Chugai, Pfizer, Sanofi, Janssen, SymBio, and Sumitomo; and research grant support from Chugai, Sumitomo, Kyowa Kirin, Otsuka, Eisai, Asahi Kasei, Takeda and Ono. KS received honoraria from Celgene, Bristol-Myers Squibb, Takeda and Sanofi; and research grant from Ono, MSD, Celgene, Abbvie Takeda, Sanofi, Bristol-Myers Squibb, Daichi Sankyo, Alexion Pharma, Glaxo SmithKlein, Otsuka, Novartis, Amgen, Janssen, Chugai, Kyowa Kirin and Pfizer. The remaining authors have no conflicts of interest to disclose. Contributions HT designed the research study, performed research, collected, analyzed, interpreted data, and wrote the manuscript. SM, TT, SF, IH and KS designed the research study, performed research, and collected data. TM analyzed the survival of solid and hematological cancers in the general population in Japan. KK, NT, MM, AH, ND, NU, MS, YM, S. Kurahashi, KN, YH, S. Kako, SI, TI, YK and YA performed research and collected data. Acknowledgments We thank Dr Koji Kawamura of Tottori University for the consultation of statistics and all physicians and staff members of the collaborating institutes of the Japan Society for Transplantation and Cellular Therapy. Funding This study was supported by funding from Kanazawa University. Data-sharing statement Qualified researchers can request access to data and related study documents, including statistical analysis plan and dataset specifications.

References 1. Palumbo A, Cavallo F, Gay F, et al. Autologous transplantation and maintenance therapy in multiple myeloma. N Engl J Med. 2014;371(10):895-905. 2. Attal M, Lauwers-Cances V, Hulin C, et al. Lenalidomide, bortezomib, and dexamethasone with transplantation for myeloma. N Engl J Med. 2017;376(14):1311-1320. 3. Cavo M, Gay F, Beksac M, et al. Autologous haematopoietic stemcell transplantation versus bortezomib-melphalan-prednisone, with or without bortezomib-lenalidomide-dexamethasone consolidation therapy, and lenalidomide maintenance for newly diagnosed multiple myeloma (EMN02/HO95): a multicentre, randomised, open-label, phase 3 study. Lancet Haematol. 2020;7(6):e456-e468. 4. Gay F, Musto P, Rota-Scalabrini D, et al. Carfilzomib with

cyclophosphamide and dexamethasone or lenalidomide and dexamethasone plus autologous transplantation or carfilzomib plus lenalidomide and dexamethasone, followed by maintenance with carfilzomib plus lenalidomide or lenalidomide alone for patients with newly diagnosed multiple myeloma (FORTE): a randomised, open-label, phase 2 trial. Lancet Oncol. 2021;22(12):1705-1720. 5. Engelhardt M, Ihorst G, Landgren O, et al. Large registry analysis to accurately define second malignancy rates and risks in a well-characterized cohort of 744 consecutive multiple myeloma patients followed-up for 25 years. Haematologica. 2015;100(10):1340-1349. 6. Sahebi F, Iacobelli S, Sbianchi G, et al. Incidence of second primary malignancies after autologous transplantation for

Haematologica | 108 December 2023

3407

ARTICLE - Risk factors for SPM after ASCT for MM

H. Takamatsu et al.

multiple myeloma in the era of novel agents. Biol Blood Marrow Transplant. 2018;24(5):930-936. 7. Ailawadhi S, Swaika A, Razavi P, Yang D, Chanan-Khan A. Variable risk of second primary malignancy in multiple myeloma patients of different ethnic subgroups. Blood Cancer J. 2014;4(9):e243. 8. Fei F, Reddy V, Rosenblum F. Secondary primary malignancies in patients with multiple myeloma: a single institution experience. Hematol Oncol. 2021;39(5):674-679. 9. Barth P, Castillo JJ, Olszewski AJ. Outcomes of secondary solid tumor malignancies among patients with myeloma: a population-based study. Cancer. 2019;125(4):550-558. 10. Tzeng HE, Lin CL, Tsai CH, et al. Time trend of multiple myeloma and associated secondary primary malignancies in Asian patients: a Taiwan population-based study. PLoS One. 2013;8(7):e68041. 11. Yamasaki S, Yoshimoto G, Kohno K, et al. Risk of secondary primary malignancies in multiple myeloma patients with or without autologous stem cell transplantation. Int J Hematol. 2019;109(1):98-106. 12. Liu Y, Hou HA, Qiu H, Tang CH. Is the risk of second primary malignancy increased in multiple myeloma in the novel therapy era? A population-based, retrospective cohort study in Taiwan. Sci Rep. 2020;10(1):14393. 13. Matsuda T, Ajiki W, Marugame T, et al. Population-based survival of cancer patients diagnosed between 1993 and 1999 in Japan: a chronological and international comparative study. Jpn J Clin Oncol. 2011;41(1):40-51. 14. Gray RJ. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat. 1988;16(3):1141-1154. 15. Gooley TA, Leisenring W, Crowley J, Storer BE. Estimation of failure probabilities in the presence of competing risks: new representations of old estimators. Stat Med. 1999;18(6):695-706. 16. Fine JP. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999;94(446):496-509. 17. Kanda Y. Investigation of the freely available easy-to-use software 'EZR' for medical statistics. Bone Marrow Transplant. 2013;48(3):452-458. 18. Pierce DA, Preston DL. Radiation-related cancer risks at low doses among atomic bomb survivors. Radiat Res. 2000;154(2):178-186.

19. Hudson M, L.S. C. Hodgkin's lymphoma: Lippincott Williams & Wilkins, 2005. 20. McCarthy PL, Owzar K, Hofmeister CC, et al. Lenalidomide after stem-cell transplantation for multiple myeloma. N Engl J Med. 2012;366(19):1770-1781. 21. Attal M, Lauwers-Cances V, Marit G, et al. Lenalidomide maintenance after stem-cell transplantation for multiple myeloma. N Engl J Med. 2012;366(19):1782-1791. 22. Palumbo A, Bringhen S, Kumar SK, et al. Second primary malignancies with lenalidomide therapy for newly diagnosed myeloma: a meta-analysis of individual patient data. Lancet Oncol. 2014;15(3):333-342. 23. Jones JR, Cairns D, Menzies T, et al. Second primary malignancy incidence in patients receiving lenalidomide at induction and maintenance; long-term follow up of 4358 patients enrolled to the Myeloma XI trial. Blood. 2022;140(Suppl 1):S1823-1825. 24. Jonsdottir G, Lund SH, Bjorkholm M, et al. The impact of prior malignancies on second malignancies and survival in MM patients: a population-based study. Blood Adv. 2017;1(25):2392-2398. 25. Poh C, Keegan T, Rosenberg AS. Second primary malignancies in multiple myeloma: A review. Blood Rev. 2021;46:100757. 26. Musto P, Anderson KC, Attal M, et al. Second primary malignancies in multiple myeloma: an overview and IMWG consensus. Ann Oncol. 2017;28(2):228-245. 27. Richardson PG, Jacobus SJ, Weller EA, et al. Triplet therapy, transplantation, and maintenance until progression in myeloma. N Engl J Med. 2022;387(2):132-147. 28. Rosenberg AS, Brunson A, Tuscano J, et al. Effect of autologous hematopoietic stem cell transplant on the development of second primary malignancies in multiple myeloma patients. Blood Cancer J. 2021;11(1):5. 29. Jagannath S, Abonour R, Durie BGM, et al. Impact of post-ASCT maintenance therapy on outcomes in patients with newly diagnosed multiple myeloma in Connect MM. Blood Adv. 2018;2(13):1608-1615. 30. Kosugi S, Shibayama H, Nakatani E, et al. [Second primary malignancies among patients with myeloma-related-diseases in the KMF database]. Rinsho Ketsueki. 2016;57(7):839-847. 31. Cooper JD, Thornton JA, Gibson SJ, Pham K, Sunderland K, DeStefano CB. Survival of patients with multiple myeloma diagnosed with second primary malignancies: an ASCO Cancerlinq Analysis. Blood. 2022;140(Suppl 1):S10039-10040.

Haematologica | 108 December 2023

3408

ARTICLE - Red Cell Biology & its Disorders

Early splenectomy in a large cohort of children with sickle cell anemia: risks and consequences Aimen Mechraoui,1 Ghislaine Ithier,1 Justine Pages,2 Zinedine Haouari,1 Liza Ali,3 Arnaud Bonnard,3 Malika Benkerrou,1,4 Florence Missud,1 Berengère Koehl,1,5 Laurent Holvoet,1 Enora Le Roux2,4 and Valentine Brousse1,5 1

Centre de Référence MCGRE, Service d’Hématologie-Immunologie, AP-HP, Hôpital Robert Debré, F-75019 Paris; 2Clinical Epidemiology Unit, Robert Debré University Hospital, Inserm, CIC 1426, F-75019 Paris; 3Service de Chirurgie, APHP, Hôpital Universitaire Robert Debré, F-75019 Paris; 4Inserm, UMR-1123 ECEVE, Université Paris Cité, Paris and 5Université Paris Cité and Université des Antilles, Inserm, BIGR, F-75015 Paris, France

Correspondence: V. Brousse valentine.brousse@gmail.com Received: Accepted: Early view:

December 12, 2022. May 15, 2023. May 25, 2023.

Abstract In children with sickle cell anemia (SCA), early splenic complications can require splenectomy, but the benefit-to-risk ratio and the age at which splenectomy may be safely performed remain unclear. To address this question, we analyzed the rate of post-splenectomy events in children with SCA splenectomized between 2000-2018 at the Robert Debré University Hospital, Paris, France. A total of 188 children underwent splenectomy, including 101 (11.9%) from our newborn cohort and 87 referred to our center. Median (Q1-Q3) age at splenectomy was 4.1 years (range 2.5-7.3 years), with 123 (65.4%) and 65 (34.6%) children splenectomized at ≥3 years of age or <3 years of age, respectively. Median postsplenectomy follow-up was 5.9 years (range 2.7-9.2 years) yielding 1192.6 patient-years (PY) of observation. Indications for splenectomy were mainly acute splenic sequestration (101 [53.7%]) and hypersplenism (75 [39.9%]). All patients received penicillin prophylaxis; 98.3% received 23-valent polysaccharic pneumococcal (PPV-23) vaccination, and 91.9% a median number of 4 (range 3-4) pneumococcal conjugate vaccine shots prior to splenectomy. Overall incidence of invasive bacterial infection and thrombo-embolic events were 0.005 / PY (no pneumococcal infections) and 0.003 / PY, respectively, regardless of age at splenectomy. There was an increased proportion of children with cerebral vasculopathy in children splenectomized <3 years of age (0.037 / PY vs. 0.011 / PY; P<0.01). A significantly greater proportion of splenectomized than non-splenectomized children were treated with hydroxycarbamide (77.2% vs. 50.1%; P<0.01), suggesting a more severe phenotype in children who present spleen complications. If indicated, splenectomy should not be delayed in children, provided recommended pneumococcal prophylaxis is available. Spleen complications in childhood may serve as a marker of severity.

Introduction In children with sickle cell disease, spleen function may be altered early in life. This alteration has been shown to occur as early as 3-6 months of age and to affect 87% of infants with sickle cell anemia (Hb SS or Hb S β° thalassemia) at a mean age of 12.9 months.1,2 In the majority of children, this process is clinically silent. Over time, the spleen becomes fibrotic and loses all its main functions, namely blood filtration and immune defense against blood-borne pathogens, resulting in functional asplenia and, ultimately, to the complete loss of the organ, a process called auto-splenectomy.3,4 However, in a small proportion of children, the spleen may cause serious clinical complications such as acute splenic sequestration (ASS) or hypersplenism. ASS refers to a rapid trapping of red blood cells (RBC) in an acutely enlarged

spleen, resulting in hypovolemia and severe anemia. Its prevalence is around 10-25% in children with SCA, with a median age at first occurence of 1.4 years.5 ASS is generally unpredictable, although a low fetal hemoglobin (HbF) level in the first 3-6 months of life has been correlated to its occurrence.6,7 Importantly, ASS further alters spleen function.1 Modalities of prevention of ASS are still under debate because neither chronic transfusion nor hydroxycarbamide (the major disease-modifying treatment in SCA) have demonstrated efficacy in decreasing its recurrence.8 The other frequent splenic complication is hypersplenism, a condition defined by chronic enlargement of the spleen and subsequent cytopenia.9 Its onset is generally progressive and its prevalence is ill-defined. Hypersplenism can result in failure to thrive, severe chronic anemia, and abdominal pain. Both ASS crises and hypersplenism result in, or coexist with, reduced spleen function.

Haematologica | 108 December 2023

3409

ARTICLE - Early splenectomy in children with SCA

A. Mechraoui, et al.

The major consequence of hyposplenism is an increased susceptibility to infection from encapsulated bacteria, particularly to pneumococcal invasive infections.10,11 This risk is particularly high in very young infants, as in infancy (≤2 years of age) the spleen has still not fully matured.12 In children with SCA, this increased risk was well-established both before and after the initiation of systematic prophylactic therapy with penicillin and the introduction of pneumococcal vaccines.13-17 While these preventive measures have dramatically reduced the risk of infection, pneumococcal invasive infections remain frequent and severe, are the leading cause of death in young children with SCA in high-income countries, and presumably contribute in large measure to the high rate of mortality in children <5 years of age in low-income countries.18,19 Surgical splenectomy is the radical treatment of acute splenic events in children with SCA, but its consequences in terms of additional infectious risks are difficult to understand in the absence of a baseline measurement of the residual splenic function at the time of surgery. In other words, whether surgical splenectomy further increases the risk of pneumococcal infection in children with SCA is still an open question. In addition, the age at which splenectomy is best performed remains a subject of debate. Early publications on splenectomy in children with hemolytic disorders showed a very high risk of overwhelming sepsis in young infants.20,21 While very few studies have evaluated the risks of splenectomy with age, in many centers the lower age limit for splenectomy has been fixed at 4-5 years of age, although the evidence supporting such a recommendation is poor.16,22 Splenectomy has also been associated with an increased thrombo-embolic risk in patients with hemolytic anemias, notably in patients with non-transfusion-dependent thalassemia.23,24 SCA is a hypercoagulable state at baseline, with contributing factors including chronic inflammation, increased circulation of micro particles and thrombocytosis.25-28 At the same time, a higher incidence of venous thromboembolism has been evidenced in patients with SCA.29-31 Nevertheless, it is still unclear whether the surgical removal of the spleen further increases the thromboembolic risk. In general, the benefit-to-risk ratio of splenectomy can be challenging to define in young children with SCA. Recent studies have indicated there is no increased risk, but these have concerned either small numbers of patients with SCD, or a later age at splenectomy in relation to a large proportion of patients with Sβ+ genotypes.17,32 As a result, the lower limit of age at which splenectomy may be safely performed remains to be determined. The main objective of this study was, therefore, to analyze the occurrence of infectious complications related to encapsulated bacteria as well as of thrombo-embolic events in a large cohort of children with SCA who underwent sple-

nectomy during childhood. The secondary objectives were to analyze the effect of age at splenectomy on the occurrence of medical events in the post-splenectomy period, and to compare the rate of sickle cell-related events with a non-splenectomized population.

Methods The Robert Debré University Hospital is a tertiary hospital in Paris, France, with expertise in the care of children with SCA and a national reference center for this disease population. The center manages the follow-up of one of the largest cohorts of children with this diagnosis in Europe. This study included patients from the new born cohort as well as patients referred to the surgical department of our center. The new born cohort includes all patients who have been followed-up at our institution since diagnosis after new born screening. The non-splenectomized patients of the new born cohort served as a comparison group for the secondary objectives of the study. This retrospective study included all children with SCA (homozygous sickle cell anemia or Sβ° thalassemia) splenectomized at our center between January 1st 2000 and December 31st 2018. The observation period began at the date of splenectomy and ended either the day of the last known visit to our center or the date of hematopoietic stem cell transplant or the date of death or July 1st 2021, whichever came first. During the study period, all children received daily penicillin prophylaxis (50-100,000 UI/kg/day) until 10 years of age, as well as a 23-valent polysaccharic pneumococcal vaccination after 2 years of age (with booster shots every 3-5 years), as part of our institutional protocol. In addition, after 2000, children were immunized with a 7-valent conjugated pneumococcal vaccine, and this was replaced by a 13-valent conjugated pneumococcal vaccine after 2010. Treatment with hydroxycarbamide was initiated in those children who after 2 years presented with recurrent painful crises or acute chest syndrome, as recommended in the national guidelines at the time of study.33 All medical events treated in our hospital and/or reported in the local medical file prior to and following surgical splenectomy were recorded, including indication for splenectomy. Primary outcomes of interest (invasive infectious episode and thrombo-embolic events) were recorded post splenectomy. An invasive infectious episode was defined as a septic episode with a positive bacterial identification (culture or PCR).34-36 Episodes of pneumonia without bacterial identification were not considered invasive infections. Thrombo-embolic events included deep vein thrombosis and/or pulmonary embolism diagnosed by ultrasound (US) or computed tomography (CT) scan. The occurrence of medical events related to SCA such as

Haematologica | 108 December 2023

3410

ARTICLE - Early splenectomy in children with SCA

A. Mechraoui, et al.

vaso-occlusive events (VOE), acute chest syndrome (ACS), delayed post-transfusional reaction (DHTR), or cerebral vasculopathy (CV) were also collected post splenectomy. The definition of CV included the occurrence of an abnormal transcranial Doppler and/ or a vascular stenosis of a major cerebral artery on MRA and/or an overt stroke. Events that occurred within one month of surgery were further categorized as post-operative complications. This study was approved by the local ethics committee (CEER-RD n. 2021-552) and was registered at the Assistance Publique des Hôpitaux de Paris (AP-HP) (n. 20210408110925).

(range 7.5-16.5 years), with a median post-splenectomy follow-up of 5.9 years (range 2.7-9.2 years), altogether yielding 1192.6 PY. Median age at splenectomy was 4.1 years (range 2.5-7.3 years), including 3 children splenectomized before 2 years of age: at 1.8, 1.9, and 1.9 years of age, respectively (see the Online Supplementary Appendix for details). Median age of the cohort at the time of analysis was 14.4 years (range 9.9-18.3 years). Splenectomy was performed by laparoscopic procedure in 179 (95.2%) cases and laparotomy in 9 (4.8 %) cases. During follow-up, one death (0.5%) occurred, in a boy of 9 years of age who had undergone splenectomy at the age Statistical analysis of 29 months. He died from a posterior stroke that ocResults are expressed as numbers and percentages for cat- curred during a severe episode of ACS requiring mechegorical variables and as median (25th quartile; 75th quartile) anical ventilation. His death was considered unrelated to for continuous variables. Overall and age-specific incidence splenectomy. A total of 9 patients benefited from a hemaof events were calculated as the number of events divided topoietic stem cell transplant (HSCT) with an HLAby total patient-years (PY) at risk. The univariate compari- matched sibling donor. (Indications for HSCT are detailed son of percentages was performed using the χ2 test and in Online Supplementary Table S1). the Mann-Whitney test to compare the distribution of the continuous variables. Poisson regression analyses were Pneumococcal prophylaxis performed to compare incidence of medical events accord- All the children in this cohort (100%) were receiving peniing to the age at splenectomy (before or after 3 years of cillin prophylaxis at the time of the surgery. Regarding the age). This age threshold was set because it is increasingly pneumococcal conjugated vaccine (PCV 7 or 13), of 172 applied in clinical practice and because, in our center, (95.0%) patients with available data and eligible for PCV where a substantial number of children underwent sple- vaccination during the study period (PCV 7 licensure in nectomy between the ages of 2 and 3 years, it allowed for 2000 and PCV 13 licensure in 2010), 158 (91.9%) had rean acceptable sample size for comparative purposes.37,38 ceived at least one dose, while the median number of Statistical analyses were performed using SAS software doses of PCV vaccine received at the time of splenectomy version 9.4 (SAS Institute Inc., Cary, NC, USA). was 4 (range 3-4). Regarding the polysaccharidic pneumococcal vaccine (PPV-23), among 173 patients with available data and eligible for vaccination (age >2 years), 170 (98.3%) were appropriately vaccinated: 166 (95.9%) before spleResults nectomy and 4 (2.3%) shortly (<3 months) after splenecGeneral results tomy. Median age at immunization was 2.1 years (range During the study period, a total of 1167 children with sickle 1.8-3.3 years), and a median of 2 (range 1-3) doses were cell disease were followed-up at the Robert Debré Uni- given during follow-up. versity Hospital subsequent to new born screening, including 852 children with SCA (either HbSS or HbS β° Indication for splenectomy thalassemia) (Figure 1). A total of 101 (11.9%) children (all Indications for splenectomy are shown in Table 1. Prewith SCA) underwent splenectomy. In addition, 87 children vention of ASS recurrence was the main indication in 101 with SCA were referred to our hospital for splenectomy. (53.7%) cases. Median age at splenectomy was signifiAltogether, 188 splenectomized children were included in cantly younger in children splenectomized for ASS this analysis. All patients underwent total splenectomy. (P<0.01). Regarding splenomegaly and hypersplenism, in many cases, median age at diagnosis was not deterGeneral characteristics of the cohort mined, in relation with the progressive onset of these In this cohort, 55.3% were male. The majority of patients complications. Of note, in 5 patients who had presented were homozygous for HbS (164; 87.2%), 23 (12.2%) were prior ASS, splenectomy was performed as an emergency compound heterozygotes Hb Sβ°-thalassemia, and one procedure due to the occurrence of DHTR, suspected patient had an undetermined genotype (SS or Sβ°-thalas- because of an acute post-transfusion drop of hemoglosemia). G6PD status was known in 158 (84.0%) patients, bin concentration, a low level of post transfusion hemowith a deficiency evidenced in 27 (17.1%). globin A, signs of hyperhemolysis, and an acutely Median (Q1-Q3) follow-up of the cohort was 11.8 years enlarged spleen. Haematologica | 108 December 2023

3411

ARTICLE - Early splenectomy in children with SCA

A. Mechraoui, et al.

Regarding children whose indication for splenectomy was ASS, the first episode occurred at 1.7 years of age (range 0.9-3.1 years) with a median 3 episodes (range 2.0-4.0). The majority of children had experienced more than one episode before splenectomy, with a number of episodes of ASS before splenectomy of one, 2-5, and > 5 episodes in 4 (4.0%), 89 (88.1%), and 6 (5.9%) cases, respectively. All 4 patients splenectomized after one single episode of ASS presented with severe threatening acute anemia (Hb ranging from 1.5 to < 4.7g/dL). (Details of these cases are available in the Online Supplementary Appendix).

Incidence of infectious and thrombo-embolic complications Six cases of invasive bacterial infection occurred in 5 patients during follow-up, yielding an incidence of 0.005 PY. Of note, there was no infection with an encapsulated bacterium. These infections included 5 episodes of septicemia and one episode of osteoarticular infection. The bacteria isolated during these episodes were salmonella (n=3), Klebsiella (n=1), E. Coli (n=1), and Enterobacter cloacae (n=1). These events occurred with a median delay of 10.3 months (range 8.0-49.9 months) following splenectomy. Management A thrombo-embolic complication occurred in 4 cases durAll except one allo-immunized child benefited from ing follow-up, including one case in the post-operative either a 'top-up' transfusion or exchange transfusion period, yielding an incidence of 0.003 PY. These events (all prior to surgery. In 84 (45.2%) children, this was part of a deep vein thrombosis) occurred at Day 4, 12 months, 15 prophylactic chronic transfusion therapy initiated prior months, and 34 months after surgical splenectomy. to surgery, with a median duration of 15.0 months (range 8.0-22.0 months). Median ferritin level at the time of Clinical course post splenectomy splenectomy was available in 50 (59.5%) patients and During the post-operative period, 8 complications ocwas 395 ng/mL (range 159-783 ng/mL), with 12 patients curred in 6 patients (0.03%) despite prior exchange trans(14.6%) receiving iron chelation therapy. Only 36 patients fusion: VOE n=3, ACS n=2, and ACS + left subphrenic (19.1%) were treated with hydroxycarbamide before sple- hematoma + portal thrombosis in one patient. These comnectomy. plications were not associated with age at splenectomy.

Figure 1. Flow chart of the study. SCA: sickle cell anemia; N: number.

Haematologica | 108 December 2023

3412

ARTICLE - Early splenectomy in children with SCA

A. Mechraoui, et al.

Table 1. Indication for splenectomy in the cohort of children with sickle cell anemia.

Indication for splenectomy

N (%)

Acute splenic sequestration Hypersplenism Splenomegaly DHTR Spleen infarct

101 (53.7) 75 (39.9) 5 (2.7) 5 (2.7) 1 (0.5)

Median age at diagnosis in years (Q1-Q3) 1.7 (0.9-3.1) 3.8 (1.9-5.5)* Undetermined 5.5 (4.2-6.9) 11

Median age at splenectomy in years (Q1-Q3) 3.3 (2.4-5.4) 5.9 (2.9-8.5) 6.6 (3.7-6.9) 5.1 (4.0-6.3) 11.1

DHTR: delayed hemolytic transfusion reaction; Q: quartile. *Children with available data for median age at diagnosis N=25/75 (33.33%). Q: quartile.

Table 2. Description of main sickle cell anemia-related events following splenectomy.

Events

VOE ACS CV

Patients N (%)

Median age at splenectomy in years (Q1-Q3)

Median N of episodes (Q1-Q3)

129 (73.7)* 64 (37.0)# 28 (16.4)^

4.1 (2.5-6.6) 4.2 (2.8-6.8) 2.4 (2.2-3.2)

2.0 (0.0-5.0) 1.0 (1.0-3.0) NA -

Median delay Median age at after 1st episode in splenectomy in years (Q1-Q3) months (Q1-Q3) 5.6 11.4 (3.6-9.7) (4.6-30.5) 8.1 2.3 (4.9-12.1) (0.9-5.3) 5.6 26.6 (3.4-6.5) (12.1-43.7)

Overall incidence (PY) 0.69 0.1 0.02

N: number; PY: patient years; VOE: vaso-occlusive event; ACS: acute chest syndrome; CV: cerebral vasculopathy (including abnormal transcranial Doppler, stroke and cerebral stenosis). *N=175 patients with available data. #N=173 patients with available data. ^N=171 patients with available data. NA: not applicable; Q: quartile.

Occurrence of relevant SCA-related events (VOE, ACS, DHTR, CV) following splenectomy is detailed in Table 2. There was no difference in incidence of VOE, ACS or CV according to the indication for splenectomy. DHTR was recorded in 7 patients following splenectomy with a median (min-max) delay of 5.8 years (range 1.1-11.7 years). A total of 28 (16.4%) patients developed CV, yielding an overall incidence of 0.02 PY; delay post splenectomy ranged from 5 to 82 months. Cerebral vasculopathy consisted of abnormal transcranial Doppler (TCD) in 19, cerebral arterial stenosis on MRI in 8, and stroke in one child, 16 months after splenectomy. A total of 138 (78.41 %) children were treated with hydroxycarbamide at last follow-up, out of 176 patients for whom data were available. Median age at initiation was 5.4 years (range 3.5-8.3 years).

time period and overall: 78/101 (77.2%) versus 369/737 (50.1%) (P<0.01). Median age at hydroxycarbamide initiation was also significantly younger in splenectomized patients: 4.9 years (range 3.3-7.3 years) versus 6.7 years (range 4.010.3 years) (P=0.001). In order to take into account the effect of treatment, we compared the incidence of complications in treated and non-treated children according to spleen status (splenectomy vs. no splenectomy) (Tables 3 and 4). While the incidence of ACS did not vary, the incidence of VOE was higher in non-treated non-splenectomized children while it was greater in treated splenectomized children, suggesting a more severe profile in the latter. Regarding CV, there was a greater incidence of events in the splenectomized group, with a trend in untreated children, and a significant difference in treated children.

Comparison of clinical profile according to splenectomy In order to further investigate the profile of patients who underwent splenectomy and its association with clinical phenotype, we compared the frequency of events among splenectomized and non-splenectomized children who underwent follow-up at our institution. Both groups had a comparable median age: 12 years (range 9.0-17.8 years) versus 13.5 years (range 6.9-19.2 years) (P=0.59). While there was an increase in the use of hydroxycarbamide over time, there was a significantly greater proportion of patients treated in the splenectomized group both at each

Comparison of patients’ characteristics and events according to age at splenectomy General characteristics In the global cohort of 188 children, 65 (34.6%) underwent splenectomy before the age of 3 years, and 123 (65.4%) after the age of 3 years (Figure 1). Median (Q1-Q3) postsplenectomy follow-up was 5.0 (range 3.6-8.5) and 6.4 (range 2.5-9.7) years in the groups splenectomized <3 and ≥3 years of age, respectively. Of note, there was a significant difference in rates of splenectomy before 3 years of age between the different time periods of our study

Haematologica | 108 December 2023

3413

ARTICLE - Early splenectomy in children with SCA

A. Mechraoui, et al.

Table 3. Incidence of complications in untreated children according to splenectomy or no splenectomy.

Events VOE ACS CV

No splenectomy

Splenectomy

Overall incidence (PY)

2492/6962.93=0.36 212/6962.93=0.03 90/6962.93=0.013

216/741.94=0.29 24/741.94=0.03 12/741.94=0.02

P 0.0036 0.78 0.47

PY: patient years; VOE: vaso-occlusive event; ACS: acute chest syndrome; CV: cerebral vasculopathy (including abnormal transcranial Doppler, stroke and cerebral stenosis).

Table 4. Incidence of complications in children treated with hydroxycarbamide according to splenectomy or no splenectomy.

Events VOE ACS CV

No splenectomy Overall incidence (PY) 1134/2641.94=0.43 129/2641.94=0.05 17/2641.94=0.006

Splenectomy Overall incidence (PY) 298/517.88=0.58 34/517.88=0.07 10/517.88=0.02

P <0.0001 0.12 0.006

PY: patient years; VOE: vaso-occlusive event; ACS: acute chest syndrome; CV: cerebral vasculopathy (including abnormal transcranial Doppler, stroke and cerebral stenosis).

(P=0.046): 36.7%, 20%, 34.1%, and 51.5% during the 20002005, 2005-2010, 2010-2015, and 2015-2018 time periods, respectively. Indication for splenectomy before the age of 3 years was ASS in the majority of cases (n=44, 67.7%), a proportion significantly greater than in children splenectomized after the age of 3 years (n=57 [46.3]; P<0.01). This group of patients benefited from a transfusion program in a larger proportion than older children (70.8% vs. 31.4%; P<0.01). Similarly, this younger group was treated in a larger proportion with hydroxycarbamide (88.5% vs. 73.0%; P=0.02), at a younger age (3.5 [range 2.8-5.0] years vs. 7.3 [range 4.9-9.5] years; P<0.001), particularly when the indication for hydroxycarbamide initiation was anemia (38.9% vs. 7.1% in the older group; P<0.01). Of note, all children except three splenectomized before the age of 3 years were started on hydroxycarbamide following splenectomy, and not before, ruling out the possibility that treatment may have been the cause of spleen complications and the need for subsequent early splenectomy. Comparison of post-splenectomy events in children according to age at splenectomy There was no statistical difference between the two groups in the incidence of post-splenectomy invasive infections (0.005 PY in both groups; P=0.96) or thromboembolic events (0.003 PY before the age of 3 years vs. 0.004 PY after the age of 3 years; P=0.70) Regarding other events, while the median age at first episode of VOE was younger in those splenectomized earlier, the overall incidence of VOE did not differ according to age at splenectomy (0.68 PY vs. 0.69 PY, respectively;

P=0.93). Similarly, the median age at first episode of ACS was younger in those splenectomized before the age of 3 years but there was no difference in overall incidence between groups (0.09 / PY vs. 0.10 / PY, respectively; P=0.65). In contrast, there was a statistically increased frequency of cerebral vasculopathy in children splenectomized before the age of 3 years compared to the others (0.04 / PY vs. 0.01 / PY; P<0.01), despite a higher proportion of children treated with hydroxycarbamide.

Discussion This study is the largest report so far on splenectomy in children with SCA and related complications. Regarding the risk of overwhelming sepsis, the period of follow-up (1192.6 PY) enables us to confidently confirm its very low incidence, in agreement with recent reports in other settings.16,22,39 Arguably, this risk has been described as lifelong, leaving a possibility of delayed events not captured by this study; however, most events are known to occur in the two first years following splenectomy. Regarding this low infectious risk, it is important to remember that the level of vaccine coverage against pneumococcal risk was high in this cohort (>90%) for both PCV and polysaccharidic vaccines, a scenario that is not always the case, including in high-income countries.40 In line with this, all children were also receiving antibiotic prophylaxis at the time of splenectomy and presumably until at least 10 years of age, as recommended at our center.41 With such anti-pneumococcal measures, splenectomy in childhood with SCA appears to be safe and does not result in an in-

Haematologica | 108 December 2023

3414

ARTICLE - Early splenectomy in children with SCA

A. Mechraoui, et al.

creased risk of severe invasive infection. Consequently, the risk of overwhelming infection should no longer be a major argument against surgical splenectomy whenever indicated.17,32 One important finding of this study is the absence of an increased infectious risk when splenectomy is performed before 3 years of age, with a comparable very low incidence of invasive infections in those splenectomized before or after 3 years of age. However, a large majority of children were splenectomized between 2 and 3 years of age in the younger group, with only 3 children splenectomized under the age of 2. The absence of an increased infectious risk in these younger children may be associated to the early loss of spleen function, with a subsequent risk related to surgical splenectomy that is not modified by the procedure. It has, indeed, been demonstrated that ASS negatively impacts splenic function.1 Accordingly, the majority of children who were under 3 years old at splenectomy had experienced ASS, with a median number of 3 episodes. Whatever the explanation, this finding seems to suggest that it is better to proceed to splenectomy without delay in children aged between 2 and 3 years old who have an indication of splenectomy for ASS, although such a management strategy is not currently widely accepted. In many centers, prophylactic chronic transfusion therapy is usually initiated in order to decrease the risk of severe acute anemia in case of recurrence of ASS, and this is prolonged until splenectomy can be “safely” performed, generally around 4 or 5 years of age.37,42 However, chronic transfusion does not prevent the recurrence of ASS, while it may cause serious complications, including iron overload, as illustrated by an elevated mean ferritin level (395 ng/mL [range 159-783 ng/mL]) at a median age of 4.1 years in our cohort.8 Not only can a strategy based on early splenectomy avoid transfusion therapy, iron overload and alloimmunization, but also, in low-income countries, blood-transmitted viral infections.43 Nevertheless, early splenectomy is only to be considered if appropriate pneumococcal prophylaxis is both prescribed and available, and when parents have been educated about the risk of infection. This study also suggests that surgical splenectomy does not increase the risk of thrombo-embolic events and that age at splenectomy does not influence this risk.44 Admittedly, median age at last follow-up was 14.4 years (range 9.9-18.3 years) and it cannot be excluded that the risk increases as these patients grow older. Likewise, splenectomy-related complications such as pulmonary hypertension were not observed in this cohort, as these events are exceptional in young patients but may increase over time.45 Whether surgical splenectomy specifically can be considered an additional risk to the basal risk of auto splenectomy in older patients remains to be determined. Our findings raise an important new question as to the re-

lationship between spleen complications and disease severity, particularly when these occur early in life. Overall, a greater proportion of children who underwent splenectomy thereafter required hydroxycarbamide initiation compared to those who did not (77.2% vs. 50.1%). This finding cannot be attributable to differences in management or hydroxycarbamide prescription, as this comparison was performed within the same cohort during the same study period. Importantly, a greater incidence of cerebral vasculopathy was observed in splenectomized children, particularly in those splenectomized before the age of 3 years, a majority of whom presented episodes of ASS. At least two hypotheses may be advanced. The spleen may serve as a filter or a barrier protecting the cerebral vasculature from injury caused by dense cells or fragile red cells prone to hemolysis.28 The premature destruction of the splenic filter, particularly by ASS at an age of cerebral vascular vulnerability, may therefore predispose to cerebral vasculopathy. Whether the surgical procedure further aggravates the risk is, at this point, an unanswered question. Another explanation may be related to the hemolytic phenotype of patients who experience both cerebral vasculopathy and splenic complications, particularly ASS. The latter is usually wrongly viewed as being a vaso-occlusive complication, yet in the splenic filtering beds RBC flow freely out of vessels. ASS is associated with low HbF levels and a high rate of hemolytic markers.7 Consequently, early splenic complications may represent potential markers of disease severity. However, whether ASS in particular could serve as a surrogate marker of cerebrovascular risk remains to be confirmed. For this retrospective cohort, we used data already collected within a database, hence minimizing biases, including recall bias.46 Nevertheless, this study has the limitations of observational designs (i.e., missing data, bias, no demonstration of causality) which cannot be avoided in any study on this subject since a randomized controlled trial would not be ethical in this context. Nonetheless, the study was conducted in a single center over a period of time in which those clinical practices that had the potential to impact the occurrence of post-splenectomy events did not undergo any major change. Moreover, a comparison of patients’ characteristics was made in order to avoid potential confounding variables. Finally, all children with sufficient data were included in our study, and our sample is representative of our center's population. In the largest pediatric cohort study of splenectomized patients with SCA to date, we show that surgical splenectomy does not result in a significantly increased risk of invasive bacterial infections or thrombo-embolic complication. Splenectomy performed before 3 years of age does not further increase those risks. Splenectomy should, therefore, not be delayed in patients once there is an indication for surgical removal, provided that the children have received the rec-

Haematologica | 108 December 2023

3415

ARTICLE - Early splenectomy in children with SCA

A. Mechraoui, et al.

ommended vaccines and continue to receive prophylactic penicillin therapy. This study also points to a more severe phenotype in patients who present with spleen complications requiring splenectomy, as well as to the overlapping phenotype of patients who experience spleen complications (particularly ASS) and develop cerebral vasculopathy. The relationship between ASS and cerebral vascular vulnerability would be particularly interesting to investigate in a prospective study.

Committee of Forma Therapeutics. The other authors have no conflicts of interest to disclose. Contributions VB, AM and ELR designed the study. AM and ZH collected the data. VB, AM, ELR and JP analyzed the data. LA, AB, MB, FM, GI, BK, LH, AM and VB were in charge of patients. VB, AM and ELR wrote the manuscript. All authors revised and edited the manuscript.

Disclosures Data-sharing statement VB has provided consultancy services for Global Blood Thera- For original data, please contact Valentine Brousse: valpeutics and Beam Therapeutics, and sits on the Advisory entine.brousse@aphp.fr.

References 1. El Hoss S, Cochet S, Marin M, et al. Insights into determinants of spleen injury in sickle cell anemia. Blood Adv. 2019;3(15):2328-2336. 2. Rogers ZR, Wang WC, Luo Z, et al. Biomarkers of splenic function in infants with sickle cell anemia: baseline data from the BABY HUG Trial. Blood. 2011;117(9):2614-2617. 3. Nardo-Marino A, Glenthøj A, Brewin JN, et al. The significance of spleen size in children with sickle cell anemia. Am J Hematol. 2022;97(12):1520-1528. 4. Diggs LW. Siderofibrosis of the spleen in sickle cell anemia. JAMA. 1935;104(7):538-541. 5. Brousse V, Elie C, Bankerrou M, et al. Acute splenic sequestration crisis in sickle cell disease: cohort of 190 paediatric patients. Br J Haematol. 2012;156(5):643-648. 6. Emond AM, Collis R, Darvill D, Higgs DR, Maude GH, Serjeant GR. Acute splenic sequestration in homozygous sickle cell disease: natural history and management. J Pediatr. 1985;107(2):201-206. 7. Brousse V, El Hoss S, Bouazza N, et al. Prognostic factors of disease severity in infants with sickle cell anemia: a comprehensive longitudinal cohort study. Am J Hematol. 2018;93(11):1411-1419. 8. Owusu-Ofori S, Remmington T. Splenectomy versus conservative management for acute sequestration crises in people with sickle cell disease. Cochrane Database Syst Rev. 2017;11(11):CD003425. 9. Totowa NJ. The complete spleen: structure, function, and clinical disorders. Can J Surg. 2022;45(3):226. 10. Lee GM. Preventing infections in children and adults with asplenia. Hematol Am Soc Hematol Educ Prog. 2020;2020(1):328-335. 11. Yee ME, Lai KW, Bakshi N, et al. Bloodstream Infections in children with sickle cell disease: 2010-2019. Pediatrics. 2022;149(1):e2021051892. 12. Timens W, Boes A, Rozeboom-Uiterwijk T, Poppema S. Immaturity of the human splenic marginal zone in infancy. Possible contribution to the deficient infant immune response. J Immunol. 1989;143(10):3200-3206. 13. Rankine-Mullings AE, Owusu-Ofori S. Prophylactic antibiotics for preventing pneumococcal infection in children with sickle cell disease. Cochrane Database Syst Rev. 2021;2021(3):CD003427. 14. Halasa NB, Shankar SM, Talbot TR, et al. Incidence of invasive pneumococcal disease among individuals with sickle cell disease before and after the introduction of the

pneumococcal conjugate vaccine. Clin Infect Dis. 2007;44(11):1428-1433. 15. Adamkiewicz T, Thomas S, Tunali A, et al. Population-based surveillance of pneumococcal infections in children with sickle cell disease before and after Prevnar 7® and Prevnar 13® licensure: implications for expanded vaccination. Blood. 2021;138(Suppl 1):763. 16. Lesher AP, Kalpatthi R, Glenn JB, Jackson SM, Hebra A. Outcome of splenectomy in children younger than 4 years with sickle cell disease. J Pediatr Surg. 2009;44(6):1134-1138. 17. Pinto VM, Gianesin B, Piel FB, et al. Morbidity and mortality of sickle cell disease patients is unaffected by splenectomy: evidence from 3 decades follow-up in a high-income setting. Haematologica. 2023;108(4):1158-1162. 18. Desselas E, Thuret I, Kaguelidou F, et al. Mortality in children with sickle cell disease in mainland France from 2000 to 2015. Haematologica. 2020;105(9):e440-443. 19. Grosse SD, Odame I, Atrash HK, Amendah DD, Piel FB, Williams TN. Sickle cell disease in Africa: a neglected cause of early childhood mortality. Am J Prev Med. 2011;41(6 Suppl 4):S398-405. 20. King H, Shumacker HB. Splenic Studies: I. Susceptibility to infection after splenectomy performed in infancy. Ann Surg. 1952;136(2):239-242. 21. Eraklis AJ, Kevy SV, Diamond LK, Gross RE. Hazard of overwhelming infection after splenectomy in childhood. N Engl J Med. 1967;276(22):1225-1229. 22. Kalpatthi R, Kane ID, Shatat IF, Rackoff B, Disco D, Jackson SM. Clinical events after surgical splenectomy in children with sickle cell anemia. Pediatr Surg Int. 2010;26(5):495-500. 23. Borgna Pignatti C, Carnelli V, Caruso V, et al. Thromboembolic events in beta thalassemia major: an Italian multicenter study. Acta Haematol. 1998;99(2):76-79. 24. Cappellini MD, Poggiali E, Taher AT, Musallam KM. Hypercoagulability in β-thalassemia: a status quo. Expert Rev Hematol. 2012;5(5):505-511; quiz 512. 25. Brousse V, Allaf B, Benkerrou M. [Dépistage néonatal de la drépanocytose en France.] Med Sci (Paris). 2021;37(5):482-490. 26. Leuenberger M, Sartori C. [La rate : entre mystères et découvertes.] Rev Med Suisse. 2010;6(269):2080-2085. 27. Panigrahi I, Agarwal S. Thromboembolic complications in β-thalassemia: beyond the horizon. Thromb Res. 2007;120(6):783-789.

Haematologica | 108 December 2023

3416

ARTICLE - Early splenectomy in children with SCA

A. Mechraoui, et al.

28. Sundd P, Gladwin MT, Novelli EM. Pathophysiology of sickle cell disease. Ann Rev Pathol. 2019;14(1):263-292. 29. Stein PD, Beemath A, Meyers FA, Skaf E, Olson RE. Deep venous thrombosis and pulmonary embolism in hospitalized patients with sickle cell disease. Am J Med. 2006;119(10):897.e7-11. 30. van Hamel Parsons V, Gardner K, Patel R, Thein SL. Venous thromboembolism in adults with sickle cell disease: experience of a single centre in the UK. Ann Hematol. 2016;95(2):227-232. 31. Alkindi S, Al-Ghadani AR, Al-Zeheimi SR, et al. Predicting risk factors for thromboembolic complications in patients with sickle cell anaemia - lessons learned for prophylaxis. J Int Med Res. 2021;49(12):3000605211055385. 32. Yacobovich J, Barzilai-Birenboim S, Steinberg-Shemer O, Stark P, Pazgal I, Tamary H. Splenectomy in childhood for nonmalignant haematologic disorders - long-term follow-up shows minimal adverse effects. Br J Haematol. 2020;190(6):909-915. 33. [Prise en charge de la drépanocytose chez l’enfant et l’adolescent.] Haute Autorité de Santé. https://www.hassante.fr/jcms/c_272479/fr/prise-en-charge-de-la-drepanocytose -chez-l-enfant-et-l-adolescent Accessed October 25, 2019. 34. EPIBAC. https://www.santepubliquefrance.fr/maladies-ettraumatismes/maladies-et-infections-respiratoires/infections-a -pneumocoque/articles/epibac Accessed August 25, 2021. 35. Takamatsu A, Matsuzaka S, Kodama F. [Clinical characteristics of invasive pneumococcal disease of the mucoid phenotype: a case series]. Kansenshogaku Zasshi. 2017;91(2):127-131. 36. CDC. Manual for the surveillance of vaccine-preventable diseases. https://www.cdc.gov/vaccines/pubs/survmanual/index.html 2021 Accessed September 2, 2021. 37. Brousse V, Buffet P, Rees D. The spleen and sickle cell disease: the sick(led) spleen. Br J Haematol. 2014;166(2):165-176.

38. Lesher AP, Kalpatthi R, Glenn JB, Jackson SM, Hebra A. Outcome of splenectomy in children younger than 4 years with sickle cell disease. J Pediatr Surg. 2009;44(6):1134-1138; discussion 1138. 39. Wright JG, Hambleton IR, Thomas PW, Duncan ND, Venugopal S, Serjeant GR. Postsplenectomy course in homozygous sickle cell disease. J Pediatr. 1999;134(3):304-309. 40. Reeves SL, Jary HK, Gondhi JP, Kleyn M, Wagner AL, Dombkowski KJ. Pneumococcal vaccination coverage among children with sickle cell anemia, sickle cell trait, and normal hemoglobin. Pediatr Blood Cancer. 2018;65(10):e27282. 41. [ALD n° 10 - Syndromes drépanocytaires majeurs de l’enfant et de l’adolescent.] Haute Autorité de Santé. https://www.hassante.fr/jcms/c_938890/fr/ald-n-10-syndromes-drepanocytaires -majeurs-de-l-enfant-et-de-l-adolescent. Accessed November 4, 2022. 42. Iolascon A, Andolfo I, Barcellini W, et al. Recommendations regarding splenectomy in hereditary hemolytic anemias. Haematologica. 2017;102(8):1304-1313. 43. Dei-Adomakoh Y, Asamoah-Akuoko L, Appiah B, Yawson A, Olayemi E. Safe blood supply in sub-Saharan Africa: challenges and opportunities. Lancet Haematol. 2021;8(10):e770-e776. 44. Taher A, Isma’eel H, Mehio G, et al. Prevalence of thromboembolic events among 8,860 patients with thalassaemia major and intermedia in the Mediterranean area and Iran. Thromb Haemost. 2006;96(4):488-491. 45. Chan KH, Rizvi SH, De Jesus-Rojas W, et al. Pulmonary hypertension screening in children with sickle cell disease. Pediatr Blood Cancer. 2023;70(1):e29980. 46. Mann CJ. Observational research methods. Research design II: cohort, cross sectional, and case-control studies. Emerg Med J. 2003;20(1):54-60.

Haematologica | 108 December 2023

3417

ARTICLE - Red Cell Biology & its Disorders

Metabolic signatures of cardiorenal dysfunction in plasma from sickle cell patients as a function of therapeutic transfusion and hydroxyurea treatment Angelo D’Alessandro,1,2 S. Mehdi Nouraie,3 Yingze Zhang,3 Francesca Cendali,1 Fabia Gamboni,1 Julie A. Reisz,1 Xu Zhang,4 Kyle W. Bartsch,5,6 Matthew D. Galbraith,5,6 Joaquin M. Espinosa,5,6,7 Victor R. Gordeuk4 and Mark T. Gladwin8 1

Department of Biochemistry and Molecular Genetics, University of Colorado Denver – Anschutz Medical Campus, Aurora, CO; 2Division of Hematology, Department of Medicine, University of Colorado Denver – Anschutz Medical Campus, Aurora, CO; 3Division of Pulmonary, Allergy and Critical Care Medicine, Department of Medicine, University of Pittsburgh, PE; 4 Department of Medicine, University of Illinois at Chicago, Chicago, IL; 5Linda Crnic Institute for Down Syndrome, University of Colorado - Anschutz Medical Campus, Aurora, CO; 6Department of Pharmacology, University of Colorado - Anschutz Medical Campus, Aurora, CO; 7School of Medicine Information Services, University of Colorado - Anschutz Medical Campus, Aurora, CO and 8University of Maryland School of Medicine, University of Maryland, Baltimore, MD, USA

Correspondence: A. D’Alessandro angelo.dalessandro@cuanschutz.edu M.T. Gladwin mgladwin@som.umaryland.edu Received: Accepted: Early view:

April 5, 2023. June 30, 2023. July 13, 2023.

Abstract Metabolomics studies in sickle cell disease (SCD) have been so far limited to tens of samples, owing to technical and experimental limitations. To overcome these limitations, we performed plasma metabolomics analyses on 596 samples from patients with SCD enrolled in the WALK-PHaSST study (clinicaltrials gov. Identifier: NCT00492531). Clinical covariates informed the biological interpretation of metabolomics data, including genotypes (hemoglobin [Hb] SS, hemoglobin SC), history of recent transfusion (HbA%), response to hydroxyurea treatment (fetal Hb%). We investigated metabolic correlates to the degree of intravascular hemolysis, cardiorenal function, as determined by tricuspid regurgitation velocity (TRV), estimated glomerular filtration rate (eGFR), and overall hazard ratio (unadjusted or adjusted by age). Recent transfusion events or hydroxyurea treatment were associated with elevation in plasma-free fatty acids and decreases in acyl-carnitines, urate, kynurenine, indoles, carboxylic acids, and glycine- or taurine-conjugated bile acids. High levels of these metabolites, along with low levels of plasma S1P and L-arginine were identified as top markers of hemolysis, cardiorenal function (TRV, eGFR), and overall hazard ratio. We thus uploaded all omics and clinical data on a novel online portal that we used to identify a potential mechanism of dysregulated red cell S1P synthesis and export as a contributor to the more severe clinical manifestations in patients with the SS genotype compared to SC. In conclusion, plasma metabolic signatures - including low S1P, arginine and elevated kynurenine, acyl-carnitines and bile acids - are associated with clinical manifestation and therapeutic efficacy in SCD patients, suggesting new avenues for metabolic interventions in this patient population.

Introduction Sickle cell disease (SCD) encompasses a group of inherited hemoglobinopathies caused by mutations in the gene coding for the β globin chain (HBB) in the hemoglobin tetramer.1,2 These mutations result in the substitution of single amino acid residue, E6 to V in sickle hemoglobin (HbS) allele βS, or E6 to K in HbC. Under deoxygenated conditions, sickling of red blood cells (RBC) occurs following crystallization of mutant globins into macrofibers,3-5 which in turn disrupts erythrocyte morphology, rheology6 and metabolism.7 Clinically, these molecular events manifest themselves in a complex disease char-

acterized by hemolytic anemia, inflammation and recurrent vaso-occlusive episodes.8 While such manifestations tend to be more severe in patients with the SS genotype, vaso-occlusive complications are still present in patients with the SC genotype.2,9 Lifelong transfusion represent a mainstay in the management of SCD in children and adults.8 Increases in circulating levels of heme and iron secondary to the elevated hemolytic propensity of sickle RBC triggers recurrent episodes of vaso-occlusion and inflammation, which progressively cause damage to most organs, including the kidneys, lungs, and the cardiovascular system, and overall reduced life expectancy.8 Interventions based on treat-

Haematologica | 108 December 2023

3418

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

ment with hydroxyurea (hydroxycarbamide) have shown a certain degree of efficacy in upregulating the expression of fetal hemoglobin (HbF) and increasing mean corpuscular volume,10 while decreasing the frequency of pain episodes and other acute complications in adults and children with sickle cell anemia of SS or Sβºthal genotypes, reducing the incidence of life-threatening neurological events such as primary stroke by maintaining transcranial Doppler velocities.11 The circulating metabolome offers a window not just on chemical individuality under healthy conditions,12 but also onto dysregulation of systems homeostasis under pathological conditions, such as in the case of SCD.7 Studies on the circulating metabolome in SCD have identified potential new avenues for metabolic interventions. For example, arginine supplementation13 has been proposed as a strategy to boost nitric oxide (NO) synthesis and NO-dependent vasodilation, thus counteracting hemoglobin-dependent NO scavenging secondary to intravascular hemolysis, and release of RBC arginase which has been shown to catabolize L-arginine.13,14 Similarly, hemolysis-derived free hemoglobin, heme and iron have been associated with increased renal, cardiopulmonary and infectious complications in this and other patient populations.15,16 An alternative metabolic intervention, glutamine supplementation has been proposed as a strategy to boost glutathione synthesis, thus counteracting oxidant stress in the sickle RBC.17,18 More recently, metabolomics interventions have been designed to boost the rate of late, payoff steps enzymes of glycolysis such as pyruvate kinase. This strategy promotes the synthesis of adenosine triphosphate and energy metabolism in the sickle cell, while depleting the reservoirs of allosteric modulators like 2,3-diphosphoglycerate that promote sickling by stabilizing the tense deoxygenated state of hemoglobin.19 Other metabolomics studies have identified dysregulated adenosine signaling20,21 and sphingolipid metabolism22 as potential targets for therapeutic interventions, though studies have not progressed beyond promising preclinical animal models. One major limitations of almost all metabolomics studies of SCD plasma to date is that limited sample size and lack of extensively curated clinical data hampered the extrapolation of the translational relevance of the reported findings. To bridge this gap, here we performed a metabolomics study of plasma from 596 SCD patients, for which extensive information on patients’ demographics (sex, age, body mass index [BMI]), genotypes, hematological and clinical characteristics, as well as therapeutic interventions (transfusion events, as inferred from HbA%, hydroxyurea treatment) were available. Results from analyses in plasma were compared against the recently characterized RBC metabolome in this cohort,23 identifying compartment-specific metabolic signatures associated with potential novel strategies for metabolic interventions in SCD.

Methods An extended version of this section is provided in the Online Supplementary Appendix, and our previous metabolomics report on RBC samples from the same cohort.23 Ethical statement, blood collection, processing and storage A total of 596 SCD patients with available plasma samples were recruited as part of the Walk-PHaSST clinical trial (clinicaltrials gov identifier: NCT00492531 and BioLINCC study: HLB01371616a) under protocols approved by the University of Pittsburgh.9,24-28 Clinical covariates were collected prior to plasma metabolomics analysis, including intravascular hemolysis, tricuspid regurgitation velocity (TRV),29 parameters indicative of kidney function such as estimated glomerular filtration rate (eGFR), creatinine.9,24-28 Standard high-pressure liquid chromotography (HPLC) methods were used to determine the percentage of sickle HbS, HbC, HbF and normal HbA.9,24-28 Ultra-high pressure liquid chromatography-mass spectrometry metabolomics Metabolomics extractions,30,31 analyses32,33 and elaborations34 were performed as previously described. Statistical analyses Multivariate and multivariable analyses, including principal component analyses (PCA), hierarchical clustering analyses, two-way ANOVA, correlation analyses (Spearman) and calculation of receiver operating characteristic curves were performed using MetaboAnalyst 5.0.35 For survival analysis, we used time to right censorship (including death or last follow-up) as time to event and vital status (death or alive) was the studied outcome. PCA was used to derive a hemolytic component from lactate dehydrogenase, aspartate aminotransferase, total bilirubin and reticulocyte percent.27,36,37 We applied Cox proportional hazard models to calculate the hazard ratios and P value for each metabolite. Cox analysis and adjusted regressions of metabolites to clinical outcomes were performed in R (R Core Team (2022), https://www.r-project.org/). Sickle cell disease ShinyApp portal A portal for online sharing of all the data generated in this study, based on code from the COVID-Ome Explorer portal.38

Results Metabolomics of the WALK-PHASST sickle cell disease cohort Metabolomics analyses were performed on plasma samples from 596 sickle cell study participants in the

Haematologica | 108 December 2023

3419

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

screening phase of Walk-PHASST cohort (Figure 1A). Results are extensively reported in the Online Supplementary Table S1, which also includes anonymized patients’ biological data, including genotype, sex, age, BMI (Figure 1A). An unsupervised analysis of the plasma metabolomics results is provided in the Online Supplementary Figure S1. PCA and hierarchical clustering analyses (HCA) of all the metabolomics data showed a significant separation of plasma samples from individuals with SS genotypes compared to the other groups (Online Supplementary Figure S1A, B). As previously noted,23 despite double randomization of the samples and blinding of the analytical team, plasma creatinine levels measured via mass spectrometry significantly correlated with clinical measurement for this metabolite (Online Supplementary Figure S1C). Plasma metabolic profiles in SS versus SC patients who have not been recently transfused Individuals with SS genotypes represented approximetely 73% of the cohort in this study and showed distinct clinical9,24-28 and metabolic phenotypes (Online Supplementary Figure S1) compared to patients with other genotypes. As such, the first analysis we performed focused on the SS patient population. Clinically, many of these patients are routinely transfused, which could be herein assessed by screening of HbA percentages. We used this intervention to all for an in vivo assessment of the metabolome of plasma from patients with high levels of HbS compared with patients with SS genotype who have very high levels of transfused AA blood as a control. For this first analysis, other treatments (e.g., hydroxyurea - see below) were excluded. HCA of plasma metabolites significantly affected by HbA and HbS levels in SS patients identified a significant increase in the levels of free fatty acids (palmitoleic, myristoleic, myristic, stearic, docosapentaenoic, arachidonic, dihomogammalinolenic and adrenic acid) as a function of high levels of HbA, as opposed to declines in a series of acyl-carnitines (5:0, 5:0 OH, 14:0, 16:1; Figure 1B). Elevated HbS levels were correlated (Spearman; P<0.05) with increases in urate, gluconolactone phosphate (a metabolite of the oxidative phase of the pentose phosphate pathway [PPP]) and cystine, with decreases in the levels of pyruvate, choline and acetylcholine, on top of the aforementioned free fatty acids (Figure 1C). In order to further evaluate the impact of recent transfusion events on the plasma metabolome of SS patients, data were binned into extremes HbA phenotypes, with two groups identified based on “no recent transfusion events” (HbA <5%) or “recent transfusion events” (HbA >80%) that showed distinct metabolic phenotypes by partial least square-discriminant analysis (PLS-DA; Figure 1D). Volcano plots of significant metabolites in these two groups further identified transfusion-associated elevation in pyruvate and other pentose or hexose phosphate compounds (fructose bisphosphate

and ribose phosphate) - usually associated with hemolysis when observed in plasma, since they should only be detected in intracellular compartments - and decreases in indoles, acyl-carnitines (especially those derived from branched chain amino acid catabolism - 5:0 and 5:0 OH) and pyridoxamine (Figure 1E). Impact of hydroxyurea treatment on the plasma metabolome Of the 433 patients with SS genotypes enrolled in this study, 306 had not been recently transfused (HbA <20%), of which 147 had received treatment with hydroxyurea (Online Supplementary Figure S2A). Unsupervised analyses including PCA and HCA identified a modest impact of hydroxyurea treatment on the plasma metabolome, with the strongest impact on the levels of acyl-carnitines, urate, Larginine and amino acids (lower in treated patients compared to untreated ones; Online Supplementary Figure S2B, C). These effects were corroborated by correlation with HbF %, which increased after hydroxyurea treatment and was also positively associated with elevated tryptophan and indole metabolites (indole, indole acetate; Online Supplementary Figure S2D). Linear models, unadjusted or adjusted by HbF% confirmed a strong impact of hydroxyurea on plasma L-arginine, acyl-carnitines and amino acids (Online Supplementary Figure S2E). Acknowledging the limited signal emerged from this analysis, perhaps as a result of uncertain adherence to the hydroxyurea regimen, we performed an additional analysis by focusing on SS patients on hydroxyurea with mean corpuscular volumes >105 (Online Supplementary Figure S2F), owing to the well-established effect on this hematological parameter of hydroxyurea treatments.10 This analysis unmasked a signal related with hydroxyurea-dependent decrease in circulating levels of some long-chain acyl-carnitines, but not conjugated bile acids – still elevated in this subgroup of patients compared to the untreated group (Online Supplementary Figure S2G). Comparison of SS patients to less clinically severe SC genotypes show trends consistent with beneficial effects of transfusion and hydroxyurea We then performed a direct comparison of non-recently transfused SS patients to the metabolome of patients with less clinically severe genotype of SC (Figure 2A). Linear models identified alterations of sphingolipid metabolism (especially sphingosine 1-phosphate [S1P] and sphinganine 1-phosphate), methionine sulfoxide, phosphocreatine and PPP metabolites (gluconolactone phosphate and ribose diphosphate) as the top markers distinguishing SS and SC patients, even upon adjustment for HbA% (Figure 2B). These findings were corroborated by HCA, which also showed enrichment in SS plasma of acyl-carnitines, taurine- and glycine-conjugated bile acids (taurodexycho-

Haematologica | 108 December 2023

3420

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

late, taurocholate, taurochenodeoxycholate, taurolithocholate, glycocholate, glycochenodeoxycholate), carboxylic acids (succinate, malate, oxaloacetate), purine deamination products (urate, hydroxisourate), as opposed

to depletion in L-arginine and S1P (Figure 2C). These findings were further highlighted by the identification of metabolic correlates to HbC % (Figure 2D), the top of which presented in the form of violin plots in Figure 2E.

Figure 1. Plasma metabolomics of the WALK-PHASST SCD cohort. (A) The plasma metabolome of patients with SS genotype who did not receive other treatment (e.g., hydroxyurea) was evaluated as a function of recent transfusion events, as determined by hemoglobin A (HbA)% (N=123 males and N=120 females). In (B), hierarchical clustering analysis of plasma metabolomics data as a function of HbS and HbA% (top 50 significant metabolites are shown). In (C), smile plot of plasma metabolic correlates to HbS%. In (D), partial least-square discriminant analysis of plasma metabolomics data in SS patients with recent (HbA >80%) or non-recent (HbA <5%) transfusion events. In (E), volcano plot of plasma metabolites significantly higher in patients with no recent transfusion event (HbA <5%) or with recent exchange transfusion (HbA >80%). PC: principal component; FC: fold change; UHPLC-MS: ultra-high pressure liquid chromatography-mass spectometry. Haematologica | 108 December 2023

3421

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

Plasma metabolic markers of intravascular hemolysis in the SS sickle cell disease population Degree of intravascular hemolysis was most elevated in SS patients with the more severe clinical manifestations. Analysis of metabolomics data as a function of the degree of hemolysis were thus performed in non-recently transfused (HbA <20%) SS patients via PCA, HCA and correlation analysis (Figure 3A-C). Results indicate a strong positive correlation of hemolysis with circulating levels of conjugated bile acids, urate, acyl-carnitines, free fatty acids, pentose and hexose phosphate metabolites, carboxylic acids (succinate, malate) and a negative correla-

tion with free amino acids, especially L-arginine and tryptophan metabolites (e.g., kynurenine). Similar results were even more evident in the Online Supplementary Figure S3, in which we focused on the top 30 patients with highest and lowest overall hemolysis, without constraining for patients’ genotypes nor for recent transfusion events. Plasma metabolic correlates to Dopplerechocardiographic tricuspid regurgitation jet velocity in the WALK-PHaSST cohort Cardiorenal dysfunction secondary to elevated hemolysis are common comorbidities in SCD, especially in patients

D E

Figure 2. Plasma metabolic differences between sickle cell disease patients with SS and SC genotypes. (A) Plasma metabolomics analyses were performed on plasma from patients with SS (N=433) or SC genotype (N=107) who were not recently transfused (HbA <20% - N=306 and N=101 for SS and SC, respectively). (B) Linear models were used to identify the main differences between SS and SC patients, either unadjusted (x axis) or adjusted by HbA% (y axis). In (C), hierarchical clustering analysis of the top 50 metabolites by t test, and Spearman correlation analyses between plasma metabolites and HbC % (D). In (E), focus on selected metabolites, including sphingosine 1-phosphate (S1P) and pentose phosphate pathway metabolites that differed significantly between plasma of SC and SS patients. Haematologica | 108 December 2023

3422

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

with SS genotypes.13,28 Altered TRV is a hallmark of pulmonary dysfunction in this patient population, though plasma metabolic markers of aberrant TRV have not yet been reported, and is the focus of the analysis in Figure

4. Focusing on non-recently transfused (HbA <20%) SS patients, PCA, HCA, linear models (unadjusted, or adjusted for relevant covariates of patients’ age and sex) and correlation analyses all highlighted a significant cor-

Figure 3. Plasma metabolic markers of hemolysis in SS sickle cell disease patients who were not recently transfused (hemoglobin A <20%). In (A), principal component analysis (PCA) of the plasma metabolomics data from this cohort, color-coded based on the degree of hemolysis. In (B-D) heat map, linear model (unadjusted or adjusted by subject sex and age on the y axis) and volcano plot of the plasma metabolites most significantly associated with hemolytic propensity. In (E), volcano plot of the most significant plasma metabolic changes between the 25 patients with the highest and lowest degree of hemolysis (F). Hb: hemoglobin; RBC: red blood cell. Haematologica | 108 December 2023

3423

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

relation to TRV for plasma levels of S1P, L-arginine and tryptophan (depleted with high TRV) and their catabolites (citrulline, creatinine, kynurenine), acyl-carnitines (elevated with high TRV; Figure 4A-D). Similar results were even more evident in the Online Supplementary Figure S4,

in which we focused on the top 30 patients with highest and lowest overall TRV, without constraining for patients’ genotypes nor for recent transfusion events. A snapshot of this analysis is provided in the volcano plot in Figure 4E, which focuses on SS patients with high versus low

Figure 4. Plasma metabolic markers of cardiovascular dysfunction as gleaned by tricuspid jet velocity in SS patients. In (A), principal component (PC) analysis of the plasma metabolomics data from sickle cell (SS) patients with no recent transfusion event (hemoglobin A [HbA] <20%), color-coded based on the tricuspid regurgitation velocity (TRV). In (B-D), heat map, linear model (unadjusted or adjusted by subject sex and age on the y axis) and volcano plot of the metabolites most significantly associated with tricuspid regurgitation velocity (TRV). In (E), volcano plot and heat map of the most significant metabolic changes between the 25 patients with the highest and lowest degree of TRV. Haematologica | 108 December 2023

3424

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

TRV - further highlighting depletion of glutamine and 5- The sickle cell disease online portal for real time oxoproline (a product of the γ-glutamyl-cycle) in patients processing of plasma and red blood cell metabolomics with elevated TRV. data in sickle cell disease against clinical covariates, and cross-matrix correlations: impact on pain crises Plasma metabolic correlates to renal function in the Clinical and plasma metabolomics data in this study, and WALK-PHaSST cohort RBC metabolomics data from our previous study23 were Plasma metabolic markers of renal function in SS patients collated into an online portal for real time data processwho had not been transfused recently (HbA <20%) were ing and figure generation. The portal, open and accesshere evaluated as a function of eGFR, via PCA, HCA, linear ible at https://mirages.shinyapps.io/SCD2023/ serves as models (unadjusted, or adjusted by patients’ sex and age) a repository for the data generated as part of this study, and correlation analyses (Figure 5A-D). Results confirmed and as a tool to perform real time analyses and export of a significant negative correlation between eGFR and cre- results upon correlation analyses between metabolites atinine, with concomitant accumulation of acyl-carni- and clinical covariates, with the opportunity to modify tines, depletion of arginine and accumulation of its filters (e.g., age, sex) or cross-matrix (plasma vs. RBC) catabolites (creatinine, citrulline), elevation in the levels correlations. The portal offers the opportunity to correct of pyridoxal, n-acetylneuraminate, pro-inflammatory car- for covariates of interest, for example by adjusting by boxylic acids (succinate), kynurenine, choline, acetyl-cho- HbS, HbA, HbF and HbC% (Online Supplementary Figure line, urate, indoles. Similar results were even more S9). Leveraging this portal we performed cross-matrix evident in the Online Supplementary Figures S5 and S6, analysis of plasma and RBC metabolomics data from the in which we focused on the top 25 patients with highest Walk-PHASST study. An overview of the matrix of plasma and lowest overall eGFR as a function of patients’ age in and RBC intra- and cross-matrix metabolic correlates is SS patients only with no recent transfusion event (HbA provided in Figure 6D, where metabolite pairs (x axis) are <20% - volcano plot in Figure 5E and heat map in the On- sorted by Spearman correlation coefficients (from lowest line Supplementary Figure S5) or the top 30 patients by to highest). These analyses are further presented in the high versus low eGFR overall, without constraining for pa- correlation matrix in the Online Supplementary Figure tients’ genotypes nor for recent transfusion events (On- S10. Notably, most metabolic markers of cardiorenal dysline Supplementary Figure S6). function (especially creatinine, citrulline and tryptophan metabolites, serotonin and kynurenine) showed signifiPlasma metabolic markers of mortality, time to event cant (P<0.0001) positive correlations (r>0.5) between and hazard ratio in the WALK-PHaSST cohort plasma and RBC levels. Arginine and ornithine levels, like Finally, we performed biomarker analyses to identify pyruvate, poorly correlated between plasma and RBC, plasma metabolites associated with signatures of mor- suggesting cell-intrinsic mechanisms affecting the levels tality, as gleaned by volcano plot of dichotomy outcomes of these metabolites in SCD. However, these analyses (dead vs. alive; Figure 6A), receiver operating character- also identified a discrepancy between elevation of intraistics curve determination (Online Supplementary Figure cellular RBC S1P levels and depletion of circulating S1P S7) and a more formal hazard ratio of metabolomics data in plasma of patients with SS genotype compared to any adjusted by patients’ age (Figure 6B). Results confirmed other genotype tested in this study (Figure 6E; Online a strong association between conjugated bile acids, argi- Supplementary Figure S11). Similarly, increased arginine nine and tryptophan metabolism (creatinine, citrulline, uptake from plasma to RBC could explain the observed kynurenine, anthranilate, indoles), carboxylic acids (2-hy- compartment and genotype-specific (SS vs. SC) changes droxyglutarate, succinate), acyl-carnitines and untoward in arginine levels (Figure 6E; Online Supplementary Figure clinical outcomes. Correlation of clinical and metabolic S11). covariates were thus plotted in the form of a network Compartment-specific responses result in differential analysis in Figure 6C, which identified hubs of conjugated matrix-specific correlations to clinically-relevant outbile acids, free fatty acids (especially very-long chain comes, such as self-reported pain crises events in the poly- and highly-unsaturated fatty acids), glutamine me- year preceding the analysis. Correlation analysis of tabolites (glutamine, glutamate, 2-oxoglutarate), pan- plasma and RBC metabolomics data to clinical measuretothenate and acyl-carnitines associated with ments of pain perception showed a limited number of hematological (HbS, HbA, HbC, HbF %; RBC, white blood metabolites significantly associated with this parameter, cell [WBC] counts) and cardiorenal function parameters mostly plasma metabolites that were positively associ(eGFR, creatinine, siten, BMI, cardiovascular risk, hemoly- ated (Online Supplementary Figure S12A). Of note, AcCa sis). Further correlation analyses are reported for these 10:1 in plasma and AcCa 2:0 in RBC were the top positive clinical covariates to plasma metabolite levels in Online and negative correlates to pain crisis events (Online SupSupplementary Figure S8. plementary Figure S12B, C, respectively). In keeping with Haematologica | 108 December 2023

3425

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

compartment-specific levels of ornithine and S1P, plasma - but not RBC – ornithine ranked amongst the top positive correlates to pain (Online Supplementary Figure S12D). Plasma S1P and RBC S1P were positively and negatively associated to pain, respectively (Online Supplementary Figure S12E).

Discussion Here we performed an extensive plasma metabolomics analysis of a clinically well-curated cohort of 596 SCD patients. Availability of detailed clinical measurements of HbS, A, C and F percentages allowed us to identify the im-

E D

Figure 5. Plasma metabolic markers of renal dysfunction as gleaned by estimated glomerular filtration rate and creatinine in SS patients. In (A), principal component analysis of the metabolomics data from this cohort, color-coded based on the estimated glomerular filtration rate (eGFR). In (B-D), heat map, linear model (unadjusted or adjusted by subject sex and age on the y axis) and volcano plot of the metabolites most significantly associated with eGFR. In (E), volcano plot and heat map of the most significant metabolic changes between the 25 patients with the highest and lowest eGFR. Tr: tricuspid; Hb: hemoglobin. Haematologica | 108 December 2023

3426

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

pact of recent transfusion events, genotype (with an emphasis on the more severe SS vs. the less clinically severe SC genotype). As a result, we identified multiple signatures associated with cardiorenal dysfunction and hazard ratios, including alterations in tryptophan/kynurenine/indole metabolism, bile acid deconjugation, acyl-carnitine and S1P metabolism, glutaminolysis and mitochondrial dysfunction

and L-arginine metabolism to citrulline and creatinine (Figure 7). Results are relevant in that they confirm and expand upon previous findings in the literature, linking metabolic differences23 to clinically relevant outcomes and providing a metabolic window on clinical efficacy of common interventions like transfusion of packed RBC and treatment with hydroxyurea.39 For example, previous

Figure 6. Plasma metabolic signatures of mortality, hazard ratio and plasma metabolic correlates to previously published red blood cell metabolomics data in the WALK-PHaSST cohort. (A) Plasma metabolic correlates to mortality in the WALK-PHaSST cohort. In (B), results from the hazard ratio analysis of plasma metabolomics data adjusted by patients’ age. In (C), plasma metabolite levels from the present analysis of 596 samples were correlated with all available clinical and hematological covariates and plotted in the form of a network, where each node represents a covariate, and edges indicate correlation between two covariates (thickness proportional to the module and significance of this association). In (D), plasma metabolite levels from the WALK-PHaSST cohort were matched to red blood cell (RBC) metabolomics measurements (D’Alessandro et al.23). Results indicated a strong positive cross-matrix correlation for almost all metabolic markers of cardiorenal dysfunction, except for sphingosine 1-phosphate (S1P), whose levels were elevated in SS RBC and sickle cell (SC) plasma, suggesting an increased synthesis or decreased export in the more clinically severe SS genotype (E). Haematologica | 108 December 2023

3427

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

smaller scale studies had documented a beneficial effect of transfusion events in SCD in the circulating plasma and RBC metabolome,40 though the scale of these studies was limited to a handful of recipients with no characterization of clinical relevant covariates. Previous studies had identified dysregulated arginine metabolism41 via hemolysis-triggered increases in arginase activity as factor limiting nitric oxide synthesis and vasodilation in SCD patients, ultimately contributing to endothelial and cardiovascular dysfunction.13 Here we expand on these findings identifying a role for transfusion events or hydroxyurea treatment in the amelioration of dysregulated metabolic signatures in these pathways, effects that result

in the mitigation of arginine consumption and accumulation of creatinine, ornithine, citrulline and guanidinoacetate. Of note, arginine catabolism to ornithine by arginase 1 was here associated with higher incidence of pain crises events in the year preceding the analysis. Altogether, our findings are supportive of a role for the modulation of this pathway in the management of clinical manifestations in SCD, from pain crises to hemolysis-associated cardiovascular and renal dysfunction. Similarly, previous studies had focused on glutamine consumption in the context of SCD, and glutamine supplementation as a therapeutic strategy to fuel de novo glutathione synthesis as a strategy to counteract oxidant

Figure 7. Summary of the main metabolic pathways affected in plasma of sickle cell disease patients that correlate with relevant clinical outcomes. Plasma metabolomics analyses identified dysregulation of (A) tryptophan/kynurenine/indole metabolism, (B) conjugated bile acids, (C) elevation of acyl-carnitines, (D) mitochondrial dysfunction, (E) arginine metabolism as critically altered metabolic pathways associated with cardiorenal dysfunction and hazard ratios in this patient population. Haematologica | 108 December 2023

3428

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

stress in the sickle RBC, though findings are controversial as to whether glutamine supplementation aggravates or mitigates the incidence of pain and vaso-occlusive crises in this patient population.17,18,42 Our findings show that accumulation of carboxylic acids - especially succinate and malate - are observed in plasma concomitantly to glutamine consumption and generation of 5-oxoproline, a byproduct of glutathione catabolism via the γ-glutamyl-cycle and a metabolic endproduct of this cycle in the mature RBC, owing to the lack of the enzyme oxoprolinase. This observation is relevant in light of the role of these carboxylic acids, especially succinate, in the stabilization of hypoxia inducible factor 1α,43 and activation of its transcriptional targets, including interleukin 1β.44 Of note, accumulation of amino and carboxylic acids downstream to glutamine catabolism (glutamate, 2-oxoglutarate, 2-hydroxyglutarate, succinate - in particular) was linked to signatures of cardiorenal dysfunction and overall hazard ratio, suggesting that fueling of glutathione synthesis via oral supplementation not only promotes anaplerotic reactions, but also likely fuels catabolism towards the accumulation of pro-inflammatory small molecule mediators. Further studies with stable isotope-labeled tracers will be necessary to further test this hypothesis and the ensuing clinical ramifications. Glycine and taurine conjugated bile acids are synthesized in the liver and deconjugated by the gut microbiome, and their accumulation in plasma is consistent with microbiome dysbiosis.45,46 These metabolites have been implicated in the pro-inflammatory activation of 47 macrophages, as well as in coagulopathic complications (e.g., after trauma and hemorrhage),48 both common comorbidities in sickle cell patients. Conjugated bile acids had been previously identified as markers of cholecystectomy in sickle cell patients,42 though the impact of patients’ genotypes, transfusion events or hydroxyurea treatment on the circulating levels of these metabolites had not been reported. Of note, indole metabolites derived from tryptophan catabolism are also derived from microbial catabolism,49 and participate in systems homeostasis.50 Like kynurenines,51 other endogenous tryptophan metabolites, indoles participate in immunomodulatory effects on macrophages through triggering of aryl hydrocarbon receptors.52 Of note, kynurenine was here associated with high hemolysis, increased TRV and hazard ratios. This finding is interesting in that elevated kynurenine was previously reported in individuals suffering with elevated cardiopulmonary dysfunction, pulmonary hypertension and lung fibrosis, such as subjects with Down syndrome.53 Elevation of circulating levels of kynurenine is a hallmark of viral infection with pulmonary complications, such as COVID-19, whereby this metabolite correlates with interleukin 6 levels, clinical severity of SARS-CoV-2 infection and, ultimately, clinical outcomes.54,55 Of note, in both

cases elevation of kynurenine is associated with the activation of the interferon-indoleamine 2,3-dioxygenase axis, in the latter case stimulated by sensing of viral nucleosides by the cGAS-STING system.56 It is interesting to note that mitochondrial DNA and RNA (e.g., in the context of aging) have been reported to contribute to activation of the cGAS-STING responses.57 Since mature sickle RBC have been reported to contain up to five to six residual mitochondria per cell,58,59 in light of the elevated degree of sickle RBC hemolysis, it is tempting to speculate that the elevated kynurenine levels as a function of high degree of hemolysis reported here participate in a similar circulating mitochondrial DNA sensing cascade, which could pave the way for potential novel therapeutic interventions. Finally, it is worth noting that elevation of kynurenine and indoles subtracts a substrate for the synthesis of serotonin, which is an essential component of platelet dense granules60 and a neurotransmitter that could contribute to counteracting nociceptive signaling.61 Here we performed network analyses of combined metabolomics and clinical data, which showed a strong association between hematological (e.g., HbS%) and clinical parameters with all metabolites mentioned in this discussion, as well as free fatty acids and acyl-carnitines. Of note, elevated kynurenine levels had been previously associated with increased susceptibility to hemolysis in the context of obesity,62 whereby depletion in circulating free fatty acids (previously also reported in pediatric patients with SCD63) and elevation acyl-carnitines had been identified as a marker of increased membrane RBC lipid damage and remodeling. This observation was in keeping with previous reports on the activation of the so-called Lands cycle in the context of SCD64 (or even just sickle cell trait),65 or other osmotic, mechanical or oxidative stressors to RBC, such as blood storage,66 exercise,67 chronic68 or acute kidney dysfunction.69 Of note, our findings expand on these observations, by identifying acyl-carnitine accumulation (and correction thereof by transfusion more than hydroxyurea treatment) as a hallmark of cardiorenal dysfunction in the sickle cell patient population. Interventions with carnitine supplementation in the context of SCD70 may thus be beneficial - especially in combination with hydroxyurea71 - to fuel the Lands cycle and contribute to prevent or mitigate RBC membrane lipid damage. Here we report a significant association between S1P depletion in plasma and genotypes with distinct severity of clinical manifestations (SS vs. SC), including pain crisis events. In previous studies, we and others had reported an elevation of RBC levels of the sphingolipid S1P in association with disease severity22 or increased (normal and sickle) erythrocyte susceptibility to extravascular hemolysis.72 While the mechanism of action is unclear, it has been posited that S1P may contribute to 2,3-diphosphoglycerate-dependent modulation of oxygen off-loading.22 By

Haematologica | 108 December 2023

3429

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

contributing to deoxyhemoglobin stabilization, hypoxia-induced elevation of S1P73 may cooperate with DPG to impact the so-called oxygen-dependent metabolic modulation that involves glycolytic enzyme sequestration to the membrane via binding to band 3, which in turn regulates RBC capacity to activate the PPP to generate reducing equivalents to sustain antioxidant challenges.74,75 However, here we observe a significant depletion of circulating S1P in plasma from subjects with SS genotypes, especially upon correction for transfusion events or hydroxyurea treatment. These results are suggestive of a role for decreased export, rather than (or in combination with) elevated synthesis may be responsible for dysregulated S1P metabolism in SCD. These results could be perhaps explained by polymorphisms in the S1P transporter Mfsd2b76 - previously associated with elevated susceptibility to osmotic fragility in healthy blood donor volunteers and sickle cell patients.37 In brief, while focusing only on central carbon and nitrogen metabolism with limited coverage of the lipidome or exposome,77 our study expands on existing literature and provides an extensive overview of the plasma metabolome in SCD and its clinical correlates, to the extent they are affected by common interventions like blood transfusion or treatment with hydroxyurea. The generation of a novel online portal as a repository for all the data described in this study will facilitate further dissemination and elaborations of the data presented herein in the future.

licensed to Globin Solutions and Hope Pharmaceuticals. MTG is a principal investigator in this investigator-initiated research study with Bayer Pharmaceuticals to evaluate riociguate as a treatment for patients with SCD. MTG has previously served on Bayer HealthCare LLC’s Heart and Vascular Disease Research Advisory Board and previously on the Forma Therapeutics scientific advisory board. All the other authors disclose no conflicts of interest relevant to this study. JME has provided consulting services for Elli Lily and Gilead Sciences and serves in the advisory board of Perha Pharmaceuticals. Contributions SMN, YZ, VG, MTG designed and executed the WALKPHASST study, collected and stored the samples, performed clinical measurements. FC, FG, JAR and AD performed metabolomics analyses (untargeted and targeted quantitative). KB, MG and JME built the SCD metabolome portal. AD performed data analysis and prepared figures and tables. AD wrote the first draft of the manuscript, which was revised by all the other authors. All the authors contributed to finalizing the manuscript.

Funding AD was supported by funds by the National Institute of General and Medical Sciences (RM1GM131968), and from the National Heart, Lung, and Blood Institute (R01HL146442, R01HL149714, R01HL148151, R01HL161004). MTG receives research support from NIH grants 5R01HL098032, 2R01HL125886, 5P01HL103455, 5T32HL110849, and UH3HL143192. This work was supported by Disclosures AD is a founder of Omix Technologies Inc. and Altis Bio- UH3HL143192. JME was supported by the National Institute of sciences LLC. AD is also a consultant for Hemanext Inc and Allergy and Infectious Diseases (R01AI50305) and the Global Macopharma Inc. MTG is a co-inventor of patents and pat- Down Syndrome Foundation. The content is solely the responent applications directed to the use of recombinant neuro- sibility of the authors and does not necessarily represent the globin and heme-based molecules as antidotes for CO official views of the National Institutes of Health. poisoning, which have been licensed by Globin Solutions, Inc. MTG is a shareholder, advisor, and director in Globin Sol- Data-sharing statement utions, Inc. MTG is also co-inventor on patents directed to All the data generated in this study are available in the Online the use of nitrite salts in cardiovascular diseases, which were Supplementary Table S1 and interactively accessible through previously licensed to United Therapeutics, and is now the ad hoc data portal https://mirages.shinyapps.io/SCD2023/.

References 1. Kato GJ, Piel FB, Reid CD, et al. P. Sickle cell disease. Nat Rev Dis Primers. 2018;4(1):18010. 2. da Guarda CC, Yahouédéhou S, Santiago RP, et al. Sickle cell disease: a distinction of two most frequent genotypes (HbSS and HbSC). PLoS One. 2020;15(1):e0228399. 3. Rab MAE, van Oirschot BA, Bos J, et al. Rapid and reproducible characterization of sickling during automated deoxygenation in sickle cell disease patients. Am J Hematol. 2019;94(5):575-584. 4. Henry ER, Cellmer T, Dunkelberger EB, et al. Allosteric control of hemoglobin S fiber formation by oxygen and its relation to the pathophysiology of sickle cell disease. Proc Natl Acand Sci U S A. 2020;117(26):15018-15027.

5. McDade WA, Josephs R. On the formation and crystallization of sickle hemoglobin macrofibers. J Struct Biol. 1993;110(1):90-97. 6. Connes P, Alexy T, Detterich J, Romana M, Hardy-Dessources MD, Ballas SK. The role of blood rheology in sickle cell disease. Blood Rev. 2016;30(2):111-118. 7. Adebiyi MG, Manalo JM, Xia Y. Metabolomic and molecular insights into sickle cell disease and innovative therapies. Blood Adv. 2019;3(8):1347-1355. 8. Rees DC, Williams TN, Gladwin MT. Sickle-cell disease. Lancet. 2010;376(9757):2018-2031. 9. Saraf SL, Molokie RE, Nouraie M, et al. Differences in the clinical and genotypic presentation of sickle cell disease around the

Haematologica | 108 December 2023

3430

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

world. Paediatr Respir Rev. 2014;15(1):4-12. 10. Keikhaei B, Yousefi H, Bahadoram M. Hydroxyurea: clinical and hematological effects in patients with sickle cell anemia. Glob J Health Sci. 2015;8(3):252-256. 11. Rankine-Mullings AE, Nevitt SJ. Hydroxyurea (hydroxycarbamide) for sickle cell disease. Cochrane Database Syst Rev. 2022;9(9):CD002202. 12. Houten SM. Metabolomics: unraveling the chemical individuality of common human diseases. Ann Med. 2009;41(6):402-407. 13. Morris CR, Kato GJ, Poljakovic M, et al. Dysregulated arginine metabolism, hemolysis-associated pulmonary hypertension, and mortality in sickle cell disease. JAMA. 2005;294(1):81-90. 14. Kato GJ, Steinberg MH, Gladwin MT. Intravascular hemolysis and the pathophysiology of sickle cell disease. J Clin Invest. 2017;127(3):750-760. 15. Vallelian F, Buehler PW, Schaer DJ. Hemolysis, free hemoglobin toxicity, and scavenger protein therapeutics. Blood. 2022;140(17):1837-1844. 16. Buehler PW, Swindle D, Pak DI, et al. Murine models of sickle cell disease and beta-thalassemia demonstrate pulmonary hypertension with distinctive features. Pulm Circ. 2021;11(4):20458940211055996. 17. Niihara Y, Miller ST, Kanter J, et al. A phase 3 trial of l-glutamine in sickle cell disease. N Angl J Med. 2018;379(3):226-235. 18. Niihara Y, Zerez CR, Akiyama DS, Tanaka KR. Oral L-glutamine therapy for sickle cell anemia: I. subjective clinical improvement and favorable change in red cell NAD redox potential. American J Hematol. 1998;58(2):117-121. 19. Rab MAE, Van Oirschot BA, Kosinski PA, et al. AG-348 (Mitapivat), an allosteric activator of red blood cell pyruvate kinase, increases enzymatic activity, protein stability, and ATP levels over a broad range of PKLR genotypes. Haematologica. 2021;106(1):238-249. 20. Song A, Wen AQ, Wen YE, et al. p97 dysfunction underlies a loss of quality control of damaged membrane proteins and promotes oxidative stress and sickling in sickle cell disease. FASEB J. 2022;36(5):e22246. 21. Zhang Y, Dai Y, Wen J, et al. Detrimental effects of adenosine signaling in sickle cell disease. Nat Med. 2011;17(1):79-86. 22. Sun K, D'Alessandro A, Ahmed MH, et al. Structural and functional insight of sphingosine 1-phosphate-mediated pathogenic metabolic reprogramming in sickle cell disease. Sci Rep. 2017;7(1):15281. 23. D'Alessandro A, Nouraie SM, Zhang Y, et al. In vivo evaluation of the effect of sickle cell hemoglobin S, C and therapeutic transfusion on erythrocyte metabolism and cardiorenal dysfunction. Am J Hematol. 2023;98(7):1017-1028. 24. Gladwin MT, Barst RJ, Gibbs JS, et al. Risk factors for death in 632 patients with sickle cell disease in the United States and United Kingdom. PLoS One. 2014;9(7):e99489. 25. Machado RF, Barst RJ, Yovetich NA, et al. Hospitalization for pain in patients with sickle cell disease treated with sildenafil for elevated TRV and low exercise capacity. Blood. 2011;118(4):855-64. 26. Milton JN, Rooks H, Drasar E, et al. Genetic determinants of haemolysis in sickle cell anaemia. Br J Haematol. 2013;161(2):270-278. 27. Nouraie M, Lee JS, Zhang Y, et al. The relationship between the severity of hemolysis, clinical manifestations and risk of death in 415 patients with sickle cell anemia in the US and Europe. Haematologica. 2013;98(3):464-472. 28. Sachdev V, Kato GJ, Gibbs JS, et al. Echocardiographic markers of elevated pulmonary pressure and left ventricular diastolic dysfunction are associated with exercise intolerance in adults

and adolescents with homozygous sickle cell anemia in the United States and United Kingdom. Circulation. 2011;124(13):1452-1460. 29. Kato GJ. TRV: a physiological biomarker in sickle cell disease. Pediatr Blood Cancer. 2012;58(6):831-832. 30. Bertolone L, Shin HK, Stefanoni D, et al. ZOOMICS: comparative metabolomics of red blood cells from Old World monkeys and humans. Front Physiol. 2020;11:593841. 31. Stefanoni D, Shin HKH, Baek JH, et al. Red blood cell metabolism in Rhesus macaques and humans: comparative biology of blood storage. Haematologica. 2020;105(8):2174-2186. 32. Nemkov T, Hansen KC, D'Alessandro A. A three-minute method for high-throughput quantitative metabolomics and quantitative tracing experiments of central carbon and nitrogen pathways. Rapid Commun Mass Spectrom. 2017;31(8):663-673. 33. Nemkov T, Reisz JA, Gehrke S, Hansen KC, D'Alessandro A. Highthroughput metabolomics: isocratic and gradient mass spectrometry-based methods. Methods Mol Biol. 2019;1978:13-26. 34. D'Alessandro A, Culp-Hill R, Reisz JA, et al. Heterogeneity of blood processing and storage additives in different centers impacts stored red blood cell metabolism as much as storage time: lessons from REDS-III-Omics. Transfusion. 2019;59(1):89-100. 35. Pang Z, Chong J, Zhou G, et al. MetaboAnalyst 5.0: narrowing the gap between raw spectra and functional insights. Nucleic Acids Res. 2021;49(W1):W388-W396. 36. Minniti CP, Sable C, Campbell A, et al. Elevated tricuspid regurgitant jet velocity in children and adolescents with sickle cell disease: association with hemolysis and hemoglobin oxygen desaturation. Haematologica. 2009;94(3):340-347. 37. Page GP, Kanias T, Guo YJ, et al. Multiple-ancestry genome-wide association study identifies 27 loci associated with measures of hemolysis following blood storage. J Clin Invest. 2021;131(13):e146077. 38. Sullivan KD, Galbraith MD, Kinning KT, et al. The COVIDome Explorer researcher portal. Cell Rep. 2021;36(7):109527. 39. Ribeiro PR, Teixeira RdS, Souza AR, et al. Blood plasma metabolomics of children and adolescents with sickle cell anaemia treated with hydroxycarbamide: a new tool for uncovering biochemical alterations. Br J Haematol. 2021;192(5):922-931. 40. Culp-Hill R, Srinivasan AJ, Gehrke S, et al. Effects of red blood cell (RBC) transfusion on sickle cell disease recipient plasma and RBC metabolism. Transfusion. 2018;58(12):2797-2806. 41. Darghouth D, Koehl B, Madalinski G, et al. Pathophysiology of sickle cell disease is mirrored by the red blood cell metabolome. Blood. 2011;117(6):e57-e66. 42. Ilboudo Y, Garrett ME, Bartolucci P, et al. Potential causal role of l-glutamine in sickle cell disease painful crises: A Mendelian randomization analysis. Blood Cells Mol Dis. 2021;86:102504. 43. Selak MA, Armour SM, MacKenzie ED, et al. Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-α prolyl hydroxylase. Cancer Cell. 2005;7(1):77-85. 44. Tannahill GM, Curtis AM, Adamik J, et al. Succinate is an inflammatory signal that induces IL-1β through HIF-1α. Nature. 2013;496(7444):238-242. 45. Ridlon JM, Kang DJ, Hylemon PB, Bajaj JS. Bile acids and the gut microbiome. Curr Opin Gastroenterol. 2014;30(3):332-338. 46. Soderborg TK, Clark SE, Mulligan CE, et al. The gut microbiota in infants of obese mothers increases inflammation and susceptibility to NAFLD. Nat Commun. 2018;9(1):4462. 47. El Kasmi KC, Vue PM, Anderson AL, et al. Macrophage-derived IL-1beta/NF-kappaB signaling mediates parenteral nutrition-

Haematologica | 108 December 2023

3431

ARTICLE - Plasma metabolome in SCD

A. D’Alessandro, et al.

associated cholestasis. Nat Commun. 2018;9(1):1393. 48. Wiener G, Moore HB, Moore EE, et al. Shock releases bile acid inducing platelet inhibition and fibrinolysis. J Surg Res. 2015;195(2):390-395. 49. Roager HM, Licht TR. Microbial tryptophan catabolites in health and disease. Nat Commun. 2018;9(1):3294. 50. Su X, Gao Y, Yang R. Gut microbiota-derived tryptophan metabolites maintain gut and systemic homeostasis. Cells. 2022;11(15):2296. 51. Clement CC, D'Alessandro A, Thangaswamy S, et al. 3-hydroxyL-kynurenamine is an immunomodulatory biogenic amine. Nat Commun. 2021;12(1):4447. 52. Dvořák Z, Poulíková K, Mani S. Indole scaffolds as a promising class of the aryl hydrocarbon receptor ligands. Eur J Med Chem. 2021;215:113231. 53. Powers RK, Culp-Hill R, Ludwig MP, et al. Trisomy 21 activates the kynurenine pathway via increased dosage of interferon receptors. Nat Commun. 2019;10(1):4766. 54. D'Alessandro A, Thomas T, Akpan IJ, et al. Biological and clinical factors contributing to the metabolic heterogeneity of hospitalized patients with and without COVID-19. Cells. 2021;10(9):2293. 55. Thomas T, Stefanoni D, Reisz JA, et al. COVID-19 infection alters kynurenine and fatty acid metabolism, correlating with IL-6 levels and renal status. JCI Insight. 2020;5(14):e140327. 56. Domizio JD, Gulen MF, Saidoune F, et al. The cGAS–STING pathway drives type I IFN immunopathology in COVID-19. Nature. 2022;603(7899):145-151. 57. Huang LS, Hong Z, Wu W, et al. mtDNA activates cGAS signaling and suppresses the YAP-mediated endothelial cell proliferation program to promote inflammatory injury. Immunity. 2020;52(3):475-486. 58. Jagadeeswaran R, Lenny H, Vazquez B, et al. The abnormal presence of mitochondria in circulating red blood cells cause an increased oxygen consumption rate, ROS generation and hemolysis in patients with sickle cell disease. Blood. 2017;130(Suppl1):S2237. 59. Moriconi C, Dzieciatkowska M, Roy M, et al. Retention of functional mitochondria in mature red blood cells from patients with sickle cell disease. Br J Haematol. 2022;198(3):574-586. 60. Hanby HA, Bao J, Noh J-Y, et al. Platelet dense granules begin to selectively accumulate mepacrine during proplatelet formation. Blood Adv. 2017;1(19):1478-1490. 61. Dembélé KC, Mintz T, Veyrat-Durebex C, et al. Metabolomic profiling of plasma and erythrocytes in sickle mice points to altered nociceptive pathways. Cells. 2020;9(6):1334. 62. Hazegh K, Fang F, Bravo MD, et al. Blood donor obesity is associated with changes in red blood cell metabolism and susceptibility to hemolysis in cold storage and in response to

osmotic and oxidative stress. Transfusion. 2021;61(2):435-448. 63. Qu H-Q, Glessner J, Qu J, et al. Metabolomic profiling for dyslipidemia in pediatric patients with sickle cell disease, on behalf of the IHCC consortium. Metabolomics. 2022;18(12):101. 64. Wu H, Bogdanov M, Zhang Y, et al. Hypoxia-mediated impaired erythrocyte Lands’ Cycle is pathogenic for sickle cell disease. Sci Rep. 2016;6(1):29637. 65. Nemkov T, Skinner S, Diaw M, et al Plasma levels of acylcarnitines and carboxylic acids correlate with cardiovascular and kidney function in subjects with sickle cell trait. Front Physiol. 2022;13:916197. 66. Yoshida T, Prudent M, D'Alessandro A. Red blood cell storage lesion: causes and potential clinical consequences. Blood Transfus. 2019;17(1):27-52. 67. Nemkov T, Skinner SC, Nader E, et al. Acute cycling exercise induces changes in red blood cell deformability and membrane lipid remodeling. Int J Mol Sci. 2021;22(2):896. 68. Bissinger R, Nemkov T, D'Alessandro A, et al. Proteinuric chronic kidney disease is associated with altered red blood cell lifespan, deformability and metabolism. Kidney Int. 2021;100(6):1227-1239. 69. Xu P, Chen C, Zhang Y, et al. Erythrocyte transglutaminase-2 combats hypoxia and chronic kidney disease by promoting oxygen delivery and carnitine homeostasis. Cell Metab. 2022;34(2):299-316. 70. El-Beshlawy A, Abd El Raouf E, et al. Diastolic dysfunction and pulmonary hypertension in sickle cell anemia: is there a role for L-carnitine treatment? Acta Haematol. 2006;115(1-2):91-96. 71. A. Khaled S, M. Ashry I. Drug therapy in patients with severe forms of sickle cell anemia: a nonrandomized clinical trial of combining l-carnitine with hydroxycarbamide therapy. J Appl Hematol. 2022;13(4):237-248. 72. Hay AM, Nemkov T, Gamboni F, et al. S1P has a negative effect on RBC storage quality. Blood Adv. 2023;7(8):1379-1393. 73. Sun K, Zhang Y, D'Alessandro A, et al. Sphingosine-1-phosphate promotes erythrocyte glycolysis and oxygen release for adaptation to high-altitude hypoxia. Nat Commun. 2016;7:12086. 74. Lewis IA, Campanella ME, Markley JL, Low PS. Role of band 3 in regulating metabolic flux of red blood cells. Proc Natl Acad Sci U S A. 2009;106(44):18515-18520. 75. Issaian A, Hay A, Dzieciatkowska M, et al. The interactome of the N-terminus of band 3 regulates red blood cell metabolism and storage quality. Haematologica. 2021;106(11):2971-2985. 76. Vu TM, Ishizu AN, Foo JC, et al. Mfsd2b is essential for the sphingosine-1-phosphate export in erythrocytes and platelets. Nature. 2017;550(7677):524-528. 77. Nemkov T, Stefanoni D, Bordbar A, et al. Blood donor exposome and impact of common drugs on red blood cell metabolism. JCI Insight. 2021;6(3):e146175.

Haematologica | 108 December 2023

3432

LETTER TO THE EDITOR

Prognostic value of early positron emission tomography in patients with large B-cell lymphoma treated with antiCD19 chimeric antigen receptor T-cell therapy The prognostic implications of early (at 1 month) metabolic response on long-term outcome after anti-CD19 autologous chimeric antigen receptor (CAR) T-cell therapy remain poorly defined. We conducted a multicenter, retrospective study of 329 patients with large B-cell lymphoma who received commercial CAR T-cell therapy at six academic medical centers. We found that outcomes were excellent in patients with an early complete metabolic response, and did not vary by Deauville score (DS). Patients with a partial metabolic response or stable disease, with a DS 4 or 5, had inferior responses. Among early responders, univariable and multivariable analyses demonstrated that elevated baseline lactate dehydrogenase, grade 3 or higher cytokine release syndrome, and DS 4 and 5 were associated with inferior progression-free survival (PFS). Based on these findings, a risk score was developed that could stratify patients into four groups at significantly different risks of progression or death. The 24-month PFS in the four groups was 67%, 49%, 38% and 8% (P<0.001). This score has the potential to predict outcome more accurately based on early response and guide response-adapted treatment approaches after CAR T-cell infusion. While response rates to CAR T-cell therapy are high, the majority of patients will ultimately relapse. One potential strategy to improve outcomes is through consolidation therapy for high-risk patients. While prior studies have demonstrated that patients with a partial metabolic response or stable disease at an early assessment of response have inferior outcomes and thus could be potential candidates for this type of approach, the majority were single-institution studies or included relatively small numbers of patients.1-4 We therefore analyzed long-term outcomes based on early positron emission tomography (PET) and other clinical variables in patients with large B-cell lymphoma treated with commercial CAR T-cell therapy across six academic medical centers. To our knowledge this is the largest real-world experience to evaluate predictors of response among early responders. We performed a retrospective, multicenter study of adult patients with large B-cell lymphoma treated with commercial anti-CD19 CAR T-cell therapy between July 2018 and May 2021. This analysis was approved by the Institutional Review Boards at participating centers. The selection of patients and products, supportive care,

toxicity management, and response assessment followed institutional practice. The use of bridging therapy and the timing of imaging were according to the prescribers’ discretion. Bulky disease (defined as a mass >10 cm), performance status, International Prognostic Index, lactate dehydrogenase level, and C-reactive protein concentration were evaluated prior to lymphodepletion. All patients included in this analysis had a PET scan 30 (±15) days following CAR T-cell infusion (range, 15-45 days). Cytokine release syndrome was graded according to the modified Lee criteria and the American Society for Transplantation and Cellular Therapy (ASTCT) criteria once available.5,6 Neurotoxicity was graded according to the Common Terminology Criteria for Adverse Events version 4 and ASTCT criteria once available. Response was determined by PET scan using the 5-point DS based on local readings (not centrally reviewed) applying the Lugano criteria.7 Kaplan-Meier analysis was used to estimate PFS and overall survival and Cox proportional hazards models were applied to identify predictors of PFS. A risk score was developed based on the hazard ratios for PFS determined in the multivariable models. The baseline characteristics of the 329 patients are shown in Table 1. The median duration of follow-up after CAR T-cell administration was 24 months. The median PFS for the entire cohort was 10 months. The 12- and 24-month PFS was 48% (95% confidence interval [95% CI]: 43-54) and 42% (95% CI: 37-49), respectively (Figure 1A). There was no significant difference in PFS between patients who received axicabtagene ciloleucel or tisagenlecleucel (Figure 1B). We evaluated PFS based on the metabolic response at the time of the 1-month response assessment. Among patients with a complete metabolic response at 1 month (n=169), the 24-month PFS was 61% (95% CI: 53-70) (Figure 1C). This did not vary by DS, as patients with DS 1/2 (n=108) had a 24-month PFS of 65% (95% CI: 56-76) as compared to 59% (95% CI: 47-75) in patients with a DS 3 (n=55), with a hazard ratio (HR) of 1.23 (95% CI: 0.682.23; P=0.49) (Figure 1D). Patients with a patial metabolic response (n=92) or stable disease (n=7) had a 24-month PFS of 37% (95% CI: 28-49) and 14% (95% CI: 28-49), respectively (Figure 1C). Responses varied by DS, with a 24-month PFS of 46% (95% CI: 34-61) in patients with a DS 4 (n=62) as compared to 11% (95% CI: 3-39) in pa-

Haematologica | 108 December 2023

3433

LETTER TO THE EDITOR tients with a DS 5 (n=20) (Figure 1D). The DS was missing from the PET report in 4% of patients with a complete response and 11% of patients with a partial response or stable disease. PFS was inferior among patients with a DS 4 (HR=1.87 [95% CI: 1.09-3.19], P=0.23) and DS 5 (HR=4.65 [95% CI: 2.14-10.1], P<0.001) as compared to patients with DS 1/2/3. Overall survival was also inferior among patients with partial metabolic response or stable disease at the time of early imaging (Online Supplementary Figure S1). We also evaluated the timing of progression. The median PFS in patients who subsequently relapsed with complete metabolic response (n=45), partial metabolic response (n=49), and stable disease (n=3) at the 1-month response was 5.3 (95% CI: 4.2-6.8), 3.0 (95% CI: 2.6-3.5), and 1.6 (95% CI: 1.6-not reached) months, respectively. Twenty-seven percent of patients with an initial partial metabolic response or stable disease subsequently converted to a complete response without further therapy. The median time to conversion to a complete response was 111 days. We next examined potential predictors of relapse using a variety of variables previously evaluated in prior CAR T-cell studies.8 A univariable Cox regression analysis showed that elevated baseline lactate dehydrogenase, development of grade 3 or higher cytokine release syndrome, and DS 4 or 5 on the 1-month PET were associated with increased risk of progression (Online Supplementary Figure S2A). Patients with a DS 5 had inferior outcomes (HR=4.65 [95% CI: 2.1410.1], P<0.001); within this group, there was limited impact of other prognostic variables. Given this, a decision tree was used to first separate patients with DS less than or equal to 4 and DS 5. In a multivariable Cox regression model restricted to patients with DS 4 or less, elevated baseline lactate dehydrogenase (HR=1.59 [95% CI: 1-2.51], P=0.049), grade 3 or higher cytokine release syndrome (HR=2.39 [95% CI: 1.05-5.41], P=0.037), and DS 4 (HR=2.02 [95% CI: 1.28-3.18], P=0.002) remained independently associated with risk of progression (Online Supplementary Figure S2B). Based on the hazard ratios within the latter model, we developed a risk score with one point each assigned for each of these variables. DS 5 was designated high risk and assigned 3 points. The score separated patients into a low-risk group (score=0; 41% of patients; 24month PFS 67% [95% CI: 58-77]); low-intermediate-risk group (score=1; 30% of patients; 24-month PFS 49% [95% CI: 39-62]); high-intermediate-risk group (score=2; 21% of patients; 24-month PFS 38% [95% CI: 24-59]); and highrisk group (score=3+; 8% of patients; 24-month PFS 8% [95% CI: 2-30]) (Figure 2A). The risk score was also associated with overall survival (Figure 2B). This study is to date the largest real-world experience to evaluate the prognostic role of early metabolic response following CAR T-cell therapy in patients with relapsed/re-

fractory large B-cell lymphoma. As expected, we found that patients with a complete metabolic response, with a DS 1/2/3, had improved outcomes as compared to those with a partial metabolic response or stable disease, with a DS 4

Table 1. Baseline characteristics of the 329 patients studied. Characteristic Product, N (%) Axicabtagene ciloleucel Tisagenlecleucel

297 (90) 32 (10)

Gender, N (%) Female Male

111 (34) 218 (66)

Histology, N (%) DLBCL HGBCL PMBCL Richter syndrome THRLBCL Transformed lymphoma

221 (67) 11 (3) 13 (4) 4 (1) 4 (1) 76 (23)

Age in years, median (range)

61 (19-83)

N of prior therapies, median (range)

3 (1-9)

IPI, N (%) 0-1 2 3-5 Unavailable

88 (27) 84 (25) 151 (46) 6 (2)

Bulky disease, N (%) Yes No Unavailable

32 (10) 231 (70) 66 (20)

Elevated LDH, N (%) Yes No

145 (44) 184 (56)

Elevated CRP, N (%) Yes No Unavailable

57 (17) 132 (40) 140 (43)

Bridging therapy, N (%) Chemotherapy, % Steroids alone, % Radiation, % Targeted therapy, % Anti-CD20 therapy, %

160 (49) 66 15 12 6 1

Grade 3+ CRS, N (%) Yes No Unavailable

17 (5) 303 (92) 9 (3)

Grade 3+ neurotoxicity, N (%) Yes No Unavailable

61 (19) 214 (65) 54 (16)

DLBCL: diffuse large B-cell lymphoma; HGBCL: high-grade B-cell lymphoma; PMBCL: primary mediastinal B-cell lymphoma; THRLBCL: Tcell histiocyte-rich large B-cell lymphoma; IPI: International Prognostic Index; LDH: lactate dehydrogenase; CRP: C-reactive protein; CRS: cytokine release syndrome.

Haematologica | 108 December 2023

3434

LETTER TO THE EDITOR

Figure 1. Progression-free survival following chimeric antigen receptor T-cell therapy. (A) Progression-free survival (PFS) among the entire cohort. (B) PFS among patients who received axicabtagene ciloleucel versus tisagenlecleucel. (C) PFS among patients with a complete metabolic response, partial metabolic response or stable disease at the 1-month assessment. (D) PFS among patients with a Deauville score of 1/2, 3, 4, or 5. CAR: chimeric antigen receptor; axi-cel: axicabtagene ciloleucel; tisa-cel: tisagenlecleucel; CR: complete response; PR: partial response; SD: stable disease.

or 5. Furthermore, within this group, patients with DS 5 had dismal outcomes, even if they did not have progressive disease at the 1-month timepoint. We also sought to identify readily available clinical features that could discriminate patients at greatest risk of subsequent progression, particularly among patients with a partial response at the time of the 1-month PET. In our cohort, elevated baseline lactate dehydrogenase, grade 3 or higher cytokine release syndrome, and DS 4 or 5 were all associated with inferior PFS. We therefore used these variables to develop a risk score that could stratify patients into four groups with significantly different PFS and

overall survival. We hope that this risk score, especially if validated in other datasets, may be used for prognostication for individual patients, for the selection of patients for clinical trials evaluating consolidative strategies after CAR T-cell therapy, and as a tool to define the null hypothesis in such trials. Of note, our data also suggest that maintenance or other response-adapted strategies, if employed, should be initiated soon after the 1-month response assessment, as the majority of patients who relapsed did so soon after the initial response. This study has limitations based on its retrospective nature and the lack of centralized PET review. Additionally,

Haematologica | 108 December 2023

3435

LETTER TO THE EDITOR

Figure 2. Progression-free survival and overall survival based on risk score. (A) Progression-free survival among patients with a risk score of 0-3. (B) Overall survival among patients with a risk score of 0-3. NR: not reached.

we did not have uniform data on tumor volume or SUVmax, metrics which have demonstrated prognostic value after CAR T-cell therapy.2,9 However, this study, which is the largest retrospective study performed to date and of multicenter nature, should allow robust estimates of long-term outcomes for patients who reach the 1-month restaging timepoint. In the future, we also expect that other novel biomarkers, such as minimal residual disease, may further aid in prognostication in the CAR T-cell setting.10,11 Further research into the application of this risk score as well as novel biomarkers in treatment strategies are warranted in patients with large B-cell lymphoma receiving CAR T-cell therapy.

University of South Carolina, Charleston, SC, USA *JLC and CAJ contributed equally as first authors Correspondence: JENNIFER CROMBIE - Jennifer_Crombie@dfci.harvard.edu https://doi.org/10.3324/haematol.2022.282345 Received: November 11, 2022. Accepted: May 23, 2023. Early view: June 1, 2023. ©2023 Ferrata Storti Foundation Published under a CC BY-NC license

Authors

Disclosures JLC reports consultancy for Kite/Gilead, Karyopharm, MorphoSys,

Jennifer L. Crombie, Caron A. Jacobson, Robert Redd, Geoffrey

Incyte, ADC Therapeutics, and Genmab; research funding from

Shouse,2 Alex F. Herrera,2 Victor A. Chow,3,4 Jordan Gauthier,3,4 Erin

Bayer, Abbvie, Merck, and Genentech/Roche; and research

Mullane,3,4 Kirk Cahill,5 Justin Kline,5 Jason Romancik,6 Jonathon

support from K12 CA087723. CAJ reports consultancy for

B. Cohen, Anna Saucier, Roch Houot, Philippe Armand and Brian

Kite/Gilead, Novartis, BMS/Celgene, Bluebird Bio, Epizyme, Instil

Hess

Bio, Abintus Bio, ImmPACT Bio, Miltenyi, Caribou Bio, AztraZeneca,

Morphosys, and Ipsen. AH reports consultancy for Bristol Myers Dana-Farber Cancer Institute, Boston, MA, USA; City of Hope

Squibb, Genentech, Merck, Seattle Genetics, AstraZeneca,

National Medical Center, Duarte, CA, USA; Fred Hutchinson

Karyopharm, ADC Therapeutics, Takeda, Tubulis, Regeneron,

Cancer Research Center, Seattle, WA, USA; University of

Genmab, Pfizer, Caribou, Addict Bio, and Abbvie; and research

Washington, Seattle, WA, USA; 5University of Chicago, Chicago, IL,

funding from Bristol Myers Squibb, Genentech, Merck, Seattle

USA; 6Emory University, Atlanta, GA, USA; 7University Hospital of

Genetics, Kite Pharma, Gilead Sciences, AstraZeneca, and ADC

Rennes, Rennes, France and Hollings Cancer Center, Medical

Therapeutics. VAC reports research funding from AstraZeneca. RH

Haematologica | 108 December 2023

3436

LETTER TO THE EDITOR reports honoraria from Kite/Gilead, Novartis, Incyte, Janssen,

Contributions

MSD, Takeda and Roche; and consultancy for Kite/Gilead, Novartis,

JLC, CAJ, PA, and BH developed the research idea, performed

Bristol-Myers Squibb/Celgene, ADC Therapeutics, Incyte, and

research, and wrote the manuscript. RR performed the statistical

Miltenyi. PA reports consultancy for Merck, BMS, Pfizer, Affimed,

analysis and reviewed the manuscript. GS, AH, VAC, JG, EM, KC,

Adaptive, Infinity, ADC Therapeutics, Celgene, Morphosys, Daiichi

JK, JR, JC, and AS performed research and reviewed the

Sankyo, Miltenyi, Tessa, GenMab, C4, Enterome, Regeneron,

manuscript. RH developed the research idea and reviewed the

Epizyme, AstraZeneca, Genentech, and Xencor; research funding

manuscript.

from Kite; institutional research funding from Merck, BMS, Affimed, Adaptive, Tensha, Otsuka, Sigma Tau, Genentech/Roche,

Data-sharing statement

and IGM; and honoraria from Merck and BMS. BH reports

The data that support the findings of this study are available on

consultancy for BMS and ADC Therapeutics. RR, GS, JG, EM, KC,

request from the corresponding author.

JK, JR, JC, and AS have no conflicts of interest to disclose.

References 1. Cohen D, Luttwak E, Beyar-Katz O, et al. [(18)F]FDG PET-CT in patients with DLBCL treated with CAR-T cell therapy: a practical approach of reporting pre- and post-treatment studies. Eur J Nucl Med Mol Imaging. 2022;49(3):953-962. 2. Al Zaki A, Feng L, Watson G, et al. Day 30 SUVmax predicts progression in patients with lymphoma achieving PR/SD after CAR T-cell therapy. Blood Adv. 2022;6(9):2867-2871. 3. Hong R, Tan Su Yin E, Wang L, et al. Tumor burden measured by 18F-FDG PET/CT in predicting efficacy and adverse effects of chimeric antigen receptor T-cell therapy in non-Hodgkin lymphoma. Front Oncol. 2021;11:713577. 4. Kuhnl A, Roddie C, Kirkwood AA, et al. Early FDG-PET response predicts CAR-T failure in large B-cell lymphoma. Blood Adv. 2022;6(1):321-326. 5. Lee DW, Gardner R, Porter DL, et al. Current concepts in the diagnosis and management of cytokine release syndrome. Blood. 2014;124(2):188-195. 6. Lee DW, Santomasso BD, Locke FL, et al. ASTCT consensus grading for cytokine release syndrome and neurologic toxicity associated with immune effector cells. Biol Blood Marrow Transplant. 2019;25(4):625-638.

7. Cheson BD, Fisher RI, Barrington SF, et al. Recommendations for initial evaluation, staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: the Lugano classification. J Clin Oncol. 2014;32(27):3059-3068. 8. Jacobson CA, Hunter BD, Redd R, et al. Axicabtagene ciloleucel in the non-trial setting: outcomes and correlates of response, resistance, and toxicity. J Clin Oncol. 2020;38(27):3095-3106. 9. Breen WG, Hathcock MA, Young JR, et al. Metabolic characteristics and prognostic differentiation of aggressive lymphoma using one-month post-CAR-T FDG PET/CT. J Hematol Oncol. 2022;15(1):36. 10. Frank MJ, Hossain NM, Bukhari A, et al. Monitoring of circulating tumor DNA improves early relapse detection after axicabtagene ciloleucel infusion in large B-cell lymphoma: results of a prospective multi-institutional trial. J Clin Oncol. 2021;39(27):3034-3043. 11. Pulsipher MA, Han X, Maude SL, et al. Next-generation sequencing of minimal residual disease for predicting relapse after tisagenlecleucel in children and young adults with acute lymphoblastic leukemia. Blood Cancer Discov. 2022;3(1):66-81.

Haematologica | 108 December 2023

3437

LETTER TO THE EDITOR

Hematological abnormalities in Jacobsen syndrome: cytopenia of varying severities and morphological abnormalities in peripheral blood and bone marrow Jacobsen syndrome is caused by partial chromosomal deletion in the 11q23 region between sub-band 11q23.3 and telomeres, ranging from approximately 7 Mb to 20 Mb. Reportedly, various physical malformations and thrombocytopenia are considered common hematological comorbidities in Jacobsen syndrome, with 88.5% patients presenting with thrombocytopenia.1,2 We retrospectively analyzed four patients who were diagnosed with Jacobsen syndrome in the Japanese cohort of 1,311 pediatric patients with bone marrow failure (BMF) and found thrombocytopenia alone in only one of the patients. The other three patients had multilineage cytopenias (pancytopenia in 1 and bicytopenia in 2). Careful evaluation of the literature revealed that, in nine of 31 patients, Jacobsen syndrome was complicated by multilineage cytopenias. These findings suggest that the hematological abnormalities of Jacobsen syndrome are not limited to the platelet system but may also involve multilineage blood cells. Between February 2009 and December 2021, four patients (2 boys and 2 girls) in the Japanese BMF registry were diagnosed with Jacobsen syndrome, and they were enrolled in this study. Written informed consent was obtained from all the patients or their parents. The study was approved by the ethics committee of Nagoya University Graduate School of Medicine (2011–1352, 2019–0134). These patients had the following hematological abnormalities: thrombocytopenia (n=1), bicytopenia (thrombocytopenia + anemia, n=2), and pancytopenia (n=1). Their median age at cytopenia appearance was 60.5 months (range, 0-191 months) (Table 1). Although two patients required blood transfusion, none required hematopoietic stem cell transplantation. Other complications were listed in the Online Supplementary Table S1. Case 1 A 10-year-old girl with pulmonary hypertension and ventricular septal defect (VSD) at birth was known to have Jacobsen syndrome diagnosed via chromosome analysis of 46,XX,del(11)(q23). A blood test performed during follow-up in the clinic revealed thrombocytopenia with a platelet count of 70×109/L, although her white and red blood cell counts were within normal range. Further, it was observed that her hemoglobin level decreased gradually and that she developed cytopenia in two lineages. She did not require transfusion and is currently being followed up without treatment.

Case 2 A girl was diagnosed with Jacobsen syndrome early after birth, but she had no prominent signs of cytopenia at the time of birth. A routine medical examination which she underwent at the age of 15 years revealed anemia and thrombocytopenia. Her chromosome test result was 46,XX,add(11)(q23.3). She had a white blood cell count of 3.1×109/L, a hemoglobin level of 8.4 g/dL, and a platelet count of 107×109/L. She did not require transfusion and is currently being followed up without treatment. Case 3 A boy was born at 36 weeks of gestation with a low platelet count of 54×109/L, which further decreased gradually to 20×109/L. He also developed neutropenia (neutrophil count =0.7×109/L) and anemia (hemoglobin level =8.4 g/dL). His chromosome test result was 46,XY,del(11)(q23.3). He had cardiac complications, such as VSD and aortic stenosis, for which he underwent surgery. Red blood cell and platelet transfusions were administered as needed during the perinatal period and before and after cardiac surgery. Currently, his blood cell counts are improving without treatment. He is under observation and does not require blood transfusion. Case 4 A boy was born at 36 weeks and 5 days of gestation with a low platelet count of 46×109/L. He also had a complex congenital heart malformation (with VSD, hypoplastic left heart syndrome, mitral stenosis, truncus arteriosus, an interrupted aortic arch, and truncal valve regurgitation) and a horseshoe kidney. His chromosome test result was 46,XY,del(11)(q23.3). Currently, his platelet counts are improving without treatment; he is under observation and does not require blood transfusion. Using mononuclear cell-derived DNA from peripheral blood samples, we performed whole genome sequencing (WGS) to identify pathogenic single-nucleotide variants and determine the extent of deletions in chromosome 11q. Genomic DNA was extracted from peripheral blood mononuclear cells using the QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany). We detected germline variants using an established in-house pipeline.3 In brief, we used a Burrows-Wheeler Aligner4 to align to the hg19 reference genome. We evaluated the variants and WGS copy number

Haematologica | 108 December 2023

3438

LETTER TO THE EDITOR variations. Based on the variant interpretation guidelines of the American College of Medical Genetics, we functionally classified the identified variants into pathogenicity groups as follows: pathogenic, likely pathogenic, uncertain significance, likely benign, and benign.5 Next, we used the Genomon 2.6.3 pipeline6 and WGS data to evaluate structural variants (SV), including deletions, insertions, inversions, and translocations, in four patients diagnosed with

Jacobsen syndrome. An SV analytic pipeline was used to identify deletions of 10.0–15.5 Mb between 11q23.3 and 11q25 in each of the four patients. The deletions were then validated using a comparative genomic hybridization array (SurePrint G3 Human CGH 2 × 400k Microarray [Agilent, Inc., Santa Clara, CA]) (Figure 1). However, we found no other pathogenic single-nucleotide variant that could affect the clinical phenotypes of the patients.

Table 1. Patient characteristics of our cohort and literature review. Case

Deletion

Karyotype

Age in months

Sex

Cytopenia

Giant platelet

Reference

Site

Size, Mb

11q24.2–11q25

10.0

46,XX,del(11)(q23.3)

121

A, T

This study

11q24.1–11q25

11.5

46,XX,add(11)(q23.3)

191

A, T

This study

11q24.1–11q25

11.7

46,XY,del(11)(q23.3)

N, A, T

This study

11q23.3–11q25

15.5

46,XY,del(11)(q23.3)

This study

11q24.1–11q25

13.0

46,XY,del(11)(q23.3>qter)

11q24.1–11q25

11.3

46,XX,del(11)(q23.3)

11q24.1–11q25

12.1

46,XX,del(11)(q24.1)

11q23.3–11q25

16.0

46,XX,del(11)(q23.3)

11q23.3–11q24

16.3

46,XX,del(11)(q23.3–qter)

ND*

11q24.1–11q25

12.5

A, T

11q24.1–11q25

N, A, T

11q24.2–11q25

9.3–9.5

N, A, T

11q24.3–11q25

7.08

N, A, T

11q23.3–11q25

46,XY,del(11)(q23.3)

11q23.3–11q25

46,XY,del(11)(q23.3)

11q23.3–11q25

46,XY,del(11)(q23.3)

11q23.3–11q25

46,XY,del(11)(q23.3)

46,XX,del(11)(q23.3)

N, A, T

S10

46,XY,del(11)(q23.3)

S10

46,XX,del(11)(q23)

S10

46,XY,del(11)(q23)

S10

46,XY,del(11)(q23)

S10

46,XY,del(11)(q24.2)

N, A, T

S10

46,XX,del(11)(q24.1)

N, A, T

S10

46,XX,del(11)(q24.1)

N, A, T

S10

46,XX,del(11)(q24.2)

180

S10

46,XX,del(11)(q25)

S10

46,XY,del(11)(q23.2)

N, A, T

S11

46,XX,del(11)(qter)

S12

S13

S14

S15

S16

S17

*Pubertal age. ND: not determined; F: female; M: male; N: neutropenia; A: anemia; T: thrombocytopenia; Y: yes; S1–S17: Supplementary References (Online Supplementary Table S2).

Haematologica | 108 December 2023

3439

LETTER TO THE EDITOR Aside from giant platelets, few studies have evaluated the morphological features of the cells of Jacobsen syndrome patients.7 Smear samples were taken from the peripheral blood and bone marrow of the Jacobsen syndrome patients in our cohort. These were evaluated by physicians (AH, MI, and HI) with expertise in pediatric hematological morphology. Peripheral blood morphology could be evaluated for all four patients, whereas bone marrow morphology could be evaluated only for the three patients with pancytopenia or bicytopenia. Consistent with the findings of a previous study,2 we observed giant platelets in the peripheral blood. We also found abnormal neutrophil segmentation in the peripheral blood and dysmorphic megakaryocytes in the bone marrow. These may be common

hematological phenotypes in patients with Jacobsen syndrome (Figure 2). However, as these features have not been previously reported, future verification with larger patient samples is required. We conducted a literature search on August 31, 2022. The PubMed database (https://pubmed.ncbi.nlm.nih.gov) was searched for articles published between 1977 and August 2022. The search terms included Jacobsen syndrome, thrombocytopenia, anemia, neutropenia, and pancytopenia. The literature search extracted 14 articles written in English, and a manual literature search identified five additional relevant articles. Thus, a total of 19 articles were found to be eligible for literature review. After careful examination of the abstracts and main texts of the ar-

Figure 1. Scheme of chromosomal deletions of 11q23.3-qter in Jacobsen syndrome. Black lines indicate patients with pancytopenia, blue lines indicate patients with anemia and thrombocytopenia, and red lines indicate patients with thrombocytopenia. The deletion regions were identified via whole genome sequencing (WGS) in the 4 patients in our cohort (cases 1–4) and via comparative genomic hybridization array (*) in 5 patients from our literature review (cases 6, 7, 10, 11, and 13). Genes within chromosomal deletion regions are shown on a screenshot of the UCSC genome browser (http://genome.ucsc.edu; accessed on May 4, 2023). Chr: chromosome; tRNA: transfer RNA. Haematologica | 108 December 2023

3440

LETTER TO THE EDITOR ticles, we identified 31 Jacobsen syndrome patients with hematological abnormalities. These were thrombocytopenia (n=22), pancytopenia (n=8), and bicytopenia (n=1) (Table 1; Online Supplementary Figure S1). The median age of cytopenia onset was 0 months (range, 0–191 months) in all 35 patients, including four patients in our cohort (Table 1; Online Supplementary Table S2). Of the 31 patients with Jacobsen syndrome who showed hematological abnormalities, two patients also showed immunological abnormalities (low IgG in 1 and low IgM in the other). Data were available on the gestational week of

12 of the 31 patients. Of these 12, four were born prematurely; three with thrombocytopenia alone, and one with pancytopenia. Patients with Jacobsen syndrome are at risk of multilineage cytopenias and should be followed up closely, with attention to bleeding symptoms and infections. During the follow-up period, it is important to consider red blood cell and platelet transfusions before even minor surgical interventions. Patients should be instructed to strictly adhere to the vaccination schedule. It was reported that haploinsufficiency of friend leukemia

Figure 2. Morphological abnormalities in peripheral blood and bone marrow. May-Giemsa staining of peripheral blood and bone marrow smears. The black bar represents the scale and is equivalent to 10 μm. (A) Giant platelets in the peripheral blood of all 4 patients (case 1-4). (B) Hypersegmented neutrophils in peripheral blood. (C) Abnormal morphology of bone marrow megakaryocytes. Haematologica | 108 December 2023

3441

LETTER TO THE EDITOR integration 1 transcription factor (FLI1) plays a crucial role in the thrombocytopenia observed in Jacobsen syndrome.8 FLI1 was included in the extent of deletion of the long arm of chromosome 11 observed in all four newly diagnosed patients with Jacobsen syndrome in our cohort, and this finding is consistent with the results of previous studies.8–9 After combining the five patients with a clear extent of deletion who were identified in the literature review with the four patients in our cohort to obtain a total of nine patients, we evaluated the association between chromosome deletion site and severity of cytopenia in the nine patients; we did not find a clear association between deletion size or deletion of a specific gene and the severity of cytopenia (Figure 1). Although FLI1 may be involved in anemia and/or neutropenia as well as thrombocytopenia, we believe that the accumulation of future cases might reveal regions of deletion associated with the severity of hematological phenotypes. The current study has some limitations. We included a small number of patients, and the observation period was limited to childhood. Further, we observed fluctuations in the severity of cytopenia in some patients in our cohort. Future larger longitudinal studies must clarify the clinical picture of multilineage cytopenia and its molecular pathogenesis in patients with Jacobsen syndrome. Preterm delivery and other pregnancy-related risk factors induce epigenetic changes including DNA methylation.10 This can influence the degree of cytopenia observed in patients with Jacobsen syndrome. However, information on the gestational week was only available for a subset of cases in our cohort. Therefore, further investigation is warranted. In conclusion, multilineage cytopenia might be a recurrent finding in Jacobsen syndrome, and giant platelets, abnormal segmentation of neutrophils, and dysmorphic megakaryocytes may be considered blood cell morphological abnormalities associated with this syndrome.

Medicine, Nagoya; 2Department of Virology, Nagoya City University Graduate School of Medical Sciences, Nagoya; 3Department of Pediatrics, Hiroshima Red Cross Hospital & Atomic-Bomb Survivors Hospital, Hiroshima; 4Department of Hematology/Oncology, Saitama Children’s Medical Center, Saitama; 5Department of Pediatrics, Graduate School of Medicine, Kyoto University, Kyoto; 6Department of Pediatric Cardiology, Aichi Children’s Health and Medical Center, Obu; 7Department of Pathology, Shizuoka Children’s Hospital, Shizuoka; 8Department of Pathology, Japanese Red Cross Aichi Medical Center Nagoya First Hospital, Nagoya and 9Department of Hematology and Oncology, Children’s Medical Center, Japanese Red Cross Aichi Medical Center Nagoya First Hospital, Nagoya, Japan Correspondence: H. MURAMATSU - hideki-muramatsu@med.nagoya-u.ac.jp Y. TAKAHASHI - ytakaha@med.nagoya-u.ac.jp https://doi.org/10.3324/haematol.2022.282513 Received: January 4, 2023. Accepted: June 7, 2023. Early view: June 15, 2023. ©2023 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures No conflicts of interest to disclose. Contributions DY, HM, AN, MW, and Y Tsumura performed laboratory work, gathered clinical information, designed and conducted the study, analyzed data, and wrote the paper. AH, M Ito, HI, and YO performed laboratory work, gathered clinical information, and analyzed data. SK and Y Takahashi directed the study and analyzed the data. NN, RT, SK, KN, M Imaya, AY, RM, and DS performed laboratory work and gathered clinical information. NF, KK, KU, and EM gathered clinical information. All authors reviewed and approved the final version of the manuscript.

Authors

Acknowledgments The authors acknowledge all clinicians, patients, and their families and thank Ms. Yoshie Miura, Ms. Hiroko Ono, and Ms. Chie Amahori

Daiki Yamashita, Hideki Muramatsu, Atsushi Narita, Manabu 1

for their valuable assistance.

Wakamatsu, Yusuke Tsumura, Daichi Sajiki, Ryo Maemura, Ayako 1

Yamamori, Masayuki Imaya, Kotaro Narita, Shinsuke Kataoka, Rieko 1

Taniguchi,1 Nobuhiro Nishio,1 Yusuke Okuno,2 Naoto Fujita,3

Funding

Katsuyoshi Koh, Katsutsugu Umeda, Eiji Morihana, Hideto

This study was supported by the Nagoya Pediatric Cancer Fund.

Iwafuchi, Masafumi Ito, Seiji Kojima, Asahito Hama and Yoshiyuki 7

Data-sharing statement

Takahashi1

The data used in this study will be provided to qualified researchers 1

Department of Pediatrics, Nagoya University Graduate School of

on reasonable request.

Haematologica | 108 December 2023

3442

LETTER TO THE EDITOR

References 1. Linares Chávez EP, Toral López J, Valdés Miranda JM, et al. Jacobsen syndrome: surgical complications due to unsuspected diagnosis, the importance of molecular studies in patients with craniosynostosis. Mol Syndromol. 2016;6(5):229-235. 2. Mattina T, Perrotta CS, Grossfeld P. Jacobsen syndrome. Orphanet J Rare Dis. 2009;4:9. 3. Muramatsu H, Okuno Y, Yoshida K, et al. Clinical utility of nextgeneration sequencing for inherited bone marrow failure syndromes. Genet Med. 2017;19(7):796-802. 4. Jo H. Multi-threading the generation of Burrows-Wheeler Alignment. Genet Mol Res. 2016;15(2). 5. Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405-424. 6. Yoshida K, Sanada M, Shiraishi Y, et al. Frequent pathway

mutations of splicing machinery in myelodysplasia. Nature. 2011;478(7367):64-69. 7. Ichimiya Y, Wada Y, Kunishima S, et al. 11q23 deletion syndrome (Jacobsen syndrome) with severe bleeding: a case report. J Med Case Rep. 2018;12(1):3. 8. Raslova H, Komura E, Le Couédic JP, et al. FLI1 monoallelic expression combined with its hemizygous loss underlies ParisTrousseau/Jacobsen thrombopenia. J Clin Invest. 2004;114(1):77-84. 9. Stevenson WS, Rabbolini DJ, Beutler L, et al. Paris-Trousseau thrombocytopenia is phenocopied by the autosomal recessive inheritance of a DNA-binding domain mutation in FLI1. Blood. 2015;126(17):2027-2030. 10. Serra G, Memo L, Antona V, et al. Jacobsen syndrome and neonatal bleeding: report on two unrelated patients. Ital J Pediatr. 2021;47(1):147.

Haematologica | 108 December 2023

3443

LETTER TO THE EDITOR

Specific blood monocyte distribution in histiocytoses correlates with vascular involvement and disease activity Histiocytoses are rare hematological disorders characterized by the proliferation and accumulation of CD68+ histiocytes in tissues.1 Previously considered as inflammatory conditions, several adult’s histiocytoses (Erdheim-Chester disease [ECD], Langerhans cell histiocytosis [LCH], RosaiDorfman disease [RDD]) are now classified as myeloid neoplasms.2 Indeed, these patients display recurrent molecular features like somatic mutations in the mitogen activating-kinase (MAP-kinase) pathway genes,1 myeloid neoplasms (chronic myelomonocytic leukemia [CMML] and essential thrombocytopenia [ET])3 or clonal hematopoiesis (CH),4 which gave rise to this paradigm shift. Despite these significant advances, the cellular origin of histiocytes is still unknown. Nevertheless, monocytes are essential to the ontogeny of histiocytic disorders.5 Circulating monocytes are divided into three subsets (“classical”, “intermediate” and “non-classical”) with a distribution influenced by the innate immune system and environmental signals.6 During histiocytoses, circulating monocytes arising from bone marrow progenitors carry most MAP-kinase gene mutations, but only “classical” monocytes can differentiate into tissue histiocytes.5 Increase in “classical” monocytes was reported in CMML7 and inflammatory states,8 while a decrease in the “nonclassical” subset was described in vascular disorders.9 However, little is known about the distribution of the circulating monocyte subsets in histiocytoses10 and their differences from other myeloid neoplasms. We, therefore, sought to evaluate the distribution of the circulating monocyte subsets in patients with histiocytoses compared to patients with myeloid neoplasms (CMML and ET) and healthy donors (HD). Peripheral blood cells were obtained from patients diagnosed at Dijon University Hospital between 2020 and 2021, with histiocytoses (n=17), CMML (n=7) mainly subtype CMML-0 (n=6), ET (n=7), and HD (n=21), after informed consent. All blood samples were obtained at a steady state, after ruling out infection and worsening of the hemopathy. Among histiocytoses, eight patients had ECD (62.5% BRAF V600E mutated), five LCH (40% BRAF V600E mutated), and 4 RDD (50% MAP2K1 mutated). Samples were obtained at the time of diagnosis for two patients (#4, #15) and four (#12, #15, #16, #17) were treatmentnaïve. Three patients had concomitant myeloid neoplasms (2 CMML and 1 ET in ECD patients), and six had concomitant CH (4 ECD, 1 LCH, 1 RDD). Patient characteristics are reported in the Online Supplementary Table S1. Blood cells were stained for CD7, CD11b, CD13, CD14, CD15, CD16, CD33

and CD45. Data were acquired on a Navios cytometer and analyzed with Kaluza software (Beckman-Coulter®). Monocytes were separated on a CD14/CD16 scatter into CD14++/CD16− (classical), CD14++/CD16+ (intermediate), and CD14+/CD16++ (non-classical) as previously reported.11 The gating procedure is described in Figure 1. Diagnosis of histiocytoses was performed according to current guidelines, expert analysis and biopsies samples were centrally reviewed by an expert pathologist (JFE) for molecular analysis. Organ involvement and disease activity were assessed by imaging including 18fluorodeoxyglucose positron emission tomography.1 Diagnosis of myeloid neoplasms and clonal hematopoiesis (CH) was assessed according to biological tests completed by bone marrow analysis.2 CH was defined by at least one myeloid gene mutation in the bone marrow with a variant allele frequency over the threshold of 2% and no morphological evidence for hematological malignancies. The myeloid gene panel analyzed by next-generation sequencing (NGS) included: ASXL1, ASXL2, ATM, BCOR, CALR, CBL, CEBPA, DDX41, DNMT3A, EZH2, FLT3, GATA2, HRAS, IDH1, IDH2, IKZF1, JAK2, KIT, KRAS, MPL, NPM1, NRAS, PHF6, PTPN11, RAD21, RUNX1, SETBP1, SF3B1, SH2B3, SMC1A, SMC3, SRSF2, STAG2, TET2, TP53, U2AF1, WT1, ZRSR2, BRAF, ETV6, BCORL1, CCND3, EP300, EPOR, ETNK1, NFE2, PPM1D, STAG1, TERC, TERT, TET3, THPO, U2AF2, UBA1, EIF6, SRP72, SRP68, ANKRD26, RBBP6, GATA1, RPL23, PRPF8 and CSNK1A1. Quantitative data are expressed by median with interquartile range (IQR) and were compared between groups using Mann-Whitney tests. Qualitative data are expressed by number (percentage) and were compared using Χ2 or Fisher tests, as appropriate. For multiple group comparisons (>2), we used an analysis of variance (ANOVA) test with the Kruskal-Wallis procedure followed by Dunn’s multiple comparison with histiocytoses as the reference group. Multiple linear regression analysis was performed with all variables with a P value <0.2 in simple linear regression analysis for all monocyte subsets. Statistical significance was set at P<0.05 (two-sided). Analyses were performed with GraphPad Prism software V.9 (GraphPad, San Diego, California, USA). The study was approved by the Ethics Committee of Dijon University Hospital and is in accordance with the principles of the Declaration of Helsinki. During histiocytoses, an increase in “classical” monocytes was observed, compared to ET (P=0.01) while “intermediate” (vs. ET; P=0.02) and “non-classical” monocytes (vs.

Haematologica | 108 December 2023

3444

LETTER TO THE EDITOR

Figure 1. Procedure for gating monocytes by flow cytometry. Monocytes were identified from peripheral blood mononucelar cells with CD45 and CD33 expression after successive exclusion of T lymphocytes, natural killer (NK) cells, residual lymphocytes and basophils. Then monocytes were analyzed regarding CD14/CD16 expression for classical, intermediate and non-classical monocytes. KO: knockout; SSC: side scatter.

HD; P=0.04) were decreased (Figure 2A; Online Supplementary Table S2). Monocyte repartition did not differ between the different types of histiocytoses. Nevertheless, ECD patients had increased “classical” monocytes compared to ET patients (93%; IQR, 83-96 vs. 76%; IQR, 71-84; P=0.032) and decreased “non-classical” monocytes compared to HD (1.5% IQR, 1-4 vs. 6.1%; IQR, 4-10; P=0.029), echoing the study published by Papo et al.10 The distribution of monocyte subsets in histiocytoses was close to that in CMML (Figure 2A), suggesting some common pathogenesis pathways between those two disorders. Two arguments may partly explain this close distribution. First, CMML is the most frequent myelodysplastic/myeloproliferative neoplasms associated with histiocytoses.3 Secondly, the involvement of the MAP-kinase pathway activation has been reported in both pathologies (mostly by BRAF or MAP2K1 gene mutations in histiocytoses and KRAS, NRAS followed by BRAF in CMML).1,12 Furthermore, we have evaluated the impact of bone marrow mutation, i.e., clonal hematopoiesis, in monocyte distribution in patients with histiocytoses with or without CH compared to CMML patients. The presence of CH in histiocytoses patients induced significant changes in monocyte distribution compared to CMML patients with a

decrease in “classical” monocytes (92%; IQR, 76-93 vs. 97%; IQR, 96-98; P=0.002), and an increase in intermediate and “non-classical” monocytes (5%; IQR, 4.0-6.5 vs. 2.3%; IQR, 1.0-2.5; P=0.03; and 4%; IQR, 2.5-7.5 vs. 0.6%; IQR, 0.4-1.5; P=0.01) while no significant difference was observed in CH histiocytoses patients compared to CMML patients. These results suggest an impact of clonal hematopoiesis in the distribution of monocytes during histiocytoses (Figure 2B). We investigated whether the distribution of the monocyte subsets was associated with specific organ damage. Seven (6 ECD including 5 BRAF V600E, 1 RDD) patients had vascular involvement. They presented an increase in classical monocytes (96.00%; IQR, 92.0-96.0 vs. 86.00%; IQR, 82.5-92.0; P=0.008) and a decrease in “non-classical” monocytes (1.00%; IQR, 1.0-2.0 vs. 5.00%; IQR, 3.50-9.50; P=0.007) (Figure 3A). The correlation between “non-classical” monocytes <4% and vascular involvement was confirmed by Pearson model (0.648; 95% confidence interval [CI]: 0.25-0.86; P=0.005) (Online Supplementary Figure S1). Our results are in line with the fact that “non-classical” monocytes are associated with vascular disorders,9 their decrease being correlated with the progression of coronary disease in atherosclerotic patients.13 Thus, our results

Haematologica | 108 December 2023

3445

LETTER TO THE EDITOR

Figure 2. Distribution of monocyte subsets in histiocytoses and myeloid neoplasms and according to molecular status. (A) Distribution of monocyte subsets in patients with histiocytoses (N=17), chronic myelomonocytic leukemia (CMML) (N=7), essential thrombocythemia (ET) (N=7) and heathy donors (HD) (N=21). (B) Distribution of monocyte subsets in histiocytoses with clonal hematopoiesis (CH+) (N=6), without clonal hematopoiesis (CH-) (N=9) and CMML patients (N=7). Percentages are given among total monocytes. Horizontal bars show the median and error bars the interquartile range. P value is the result of the ANOVA test. *P<0.05; **P<0.01; ns: non-significant.

in histiocytoses suggest that “non-classical” monocytes may be a specific marker of vascular damage, irrespective of the mechanism. We then investigated whether the distribution of monocyte subsets correlated with disease activity. Histiocytoses patients achieving a metabolic response had a lower percentage of “intermediate” monocytes (3.5%; IQR, 2.0-5.0 vs. 7.0%; IQR, 4.0-13.0; P=0.04) and lower CRP levels (3.0 mg/L; IQR, 1.1-8.7 vs. 33.65 mg/L; IQR, 5.3-59.5 P=0.04) (Figure 3B), which is in line with the fact that “intermediate” monocytes produce higher levels of IL-12, TNF-α, IL-1β, and IL-6 than “classical” monocytes.14 Thus patients achieving a metabolic response have the lowest “inflammatory” state, as supported by low CRP levels and low “intermediate” monocyte frequency. We then assessed whether intrinsic or extrinsic factors could influence the distribution of monocyte subsets in patients with histiocytoses. Multiple linear regression showed a relationship between clonal hematopoiesis and the percentage of “classical” monocytes (β coefficient: -10.78; 95%

CI: -16.83 to -4.737; P=0.002). Monocyte regulation depends on the innate immune system and environment.6 Recently, the concept of trained immunity in myeloid cell homeostasis has emerged. This elaborated program is driven by epigenetic machinery providing memory immunity in the monocyte/macrophage system.15 As the trained immunity influences hematopoiesis and inflammation leads to clonal hematopoiesis, our data question the role of trained immunity in all histiocytoses, not only in BRAF-mutated ECD.16 Our study has some limitations. First we focused on circulating cells, whereas monocytes bearing the mutations represent a marginal proportion of mutated bone marrow-derived cells. However, in order to determine biomarkers useful to the management of patients, circulating cells remain the most convenient. Secondly, the analyses were performed at different times in the disease course, and drugs may have interfered with the distribution of monocyte subsets. Nevertheless, it is currently the best way to assess the modification of monocyte subset distribution related to disease activity. In addition, comple-

Haematologica | 108 December 2023

3446

LETTER TO THE EDITOR

Figure 3. Distribution of monocyte subsets in histiocytoses according to vascular injury and disease activity. Comparison of the distribution of monocyte subsets according to (A) the presence of vascular injury (N=7) or not (N=10), and (B) to the achievement of a metabolic response (N=10) or not (N=7). Disease activity was established using the last metabolic evaluation with 18 fluorodeoxyglucose positron emission tomography-computed tomography according to PERCIST criteria. Complete metabolic response was defined by normalization of all lesions to at or below standardized uptake value (SUV) of liver background. Partial metabolic response was defined by a ≥50% decrease in the sum of all target lesion baseline SUV. Progressive metabolic disease was defined by a ≥50% increase in the nadir sum of all target or new evaluable lesion SUV. Stable metabolic disease was defined as condition that did not meet previous criteria. Patients with "complete metabolic response" and "partial metabolic response" were considered as responders and patients with "stable metabolic disease" and "progressive metabolic disease" considered as non-responders.

mentary molecular analysis in all the tissues available revealed a higher frequency of myeloid gene mutations in the bone marrow rather than in the MAP kinase pathway which may be explained by the combined effect of drugs17 and aging.18 Finally, we did not perform Grubb’s test and outlier exclusion because of the difficulty of obtaining samples in these rare diseases. However, our work highlights for the first time the difference in monocyte distribution in histiocytoses compared to myeloid malignancies and HD. It also showed a correlation between “non-classical” monocytes and vascular involvement, which can be helpful for both initial staging and follow-up. As the decrease in “intermediate” monocytes was associated with lower CRP levels and metabolic response, its accuracy in response assessment will be prospectively explore to determine which of "intermediate" monocytes or CRP variation is more effective in predicting relapse. In summary, circulating monocytes may be the precursors of pathogenic histiocytes in tissues. Their subset distribu-

tion is singular in histiocytoses compared to other myeloid neoplasms and is influenced by clonal hematopoiesis. The decrease in the “non-classical” subset could represent a surrogate marker of vascular involvement, while the decrease of the intermediate fraction is associated with a metabolic response.

Authors Jerome Razanamahery,1 Maxime Samson,1 Julien Guy,2 Jessica Racine,2 Celine Row,2 Hélène Greigert,1 Barbara Nicolas,1 Stephanie Francois,3 Jean-François Emile,4 Fleur Cohen-Aubart,5 Sylvain Audia,1 Julien Haroche5 and Bernard Bonnotte1 Department of Internal Medicine and Clinical Immunology, Francois

Mitterrand Hospital, Dijon University Hospital, Dijon; 2Hematology Laboratory, Dijon University Hospital, Dijon; 3Immunology

Haematologica | 108 December 2023

3447

LETTER TO THE EDITOR Laboratory, Dijon University Hospital, Dijon; 4Department of

Disclosures

Pathology, Ambroise-Paré Hospital, Assistance-Publique Hopitaux

FC-A and JH are investigators of an academic study on the efficacy

de Paris and Sorbonne Université, Assistance Publique Hôpitaux de

of cobimetinib for treating histiocytoses. The other authors have no

Paris, Pitié-Salpêtrière Hospital, Internal Medicine Department 2,

conflicts of interest to disclose.

National Reference Center for Histiocytosis, Paris, France Contributions Correspondence:

JR, MS and BB designed the study. JG, JR, CR performed the flow

J. RAZANAMAHERY - razanamahery.jerome@hotmail.fr

cytometry analysis. JFE carried out the pathology review and molecular status determination for histiocytoses. JR, MS and HG

https://doi.org/10.3324/haematol.2023.282739

performed statistical analysis. JH and FCA provided clinical expertise. All authors edited the manuscript and approved the final

Received: January 18, 2023.

version of the manuscript for submission.

Accepted: June 1, 2023. Early view: June 15, 2023.

Data-sharing statement Complete data including flow cytometry analysis can be requested

from the corresponding author.

Published under a CC BY-NC license

References 1. Emile JF, Cohen-Aubart F, Collin M, et al. Histiocytosis. Lancet. 2021;398(10295):157-170. 2. Khoury JD, Solary E, Abla O, et al. The 5th edition of the World Health Organization classification of haematolymphoid tumours: myeloid and histiocytic/dendritic neoplasms. Leukemia. 2022;36(7):1703-1719. 3. Papo M, Diamond EL, Cohen-Aubart F, et al. High prevalence of myeloid neoplasms in adults with non-Langerhans cell histiocytosis. Blood. 2017;130(8):1007-1013. 4. Cohen Aubart F, Roos-Weil D, Armand M, et al. High frequency of clonal hematopoiesis in Erdheim-Chester disease. Blood. 2021;137(4):485-492. 5. Durham BH, Roos-Weil D, Baillou C, et al. Functional evidence for derivation of systemic histiocytic neoplasms from hematopoietic stem/progenitor cells. Blood. 2017;130(2):176-180. 6. Geissmann F, Manz MG, Jung S, et al. Development of monocytes, macrophages, and dendritic cells. Science. 2010;327(5966):656-661. 7. Selimoglu-Buet D, Wagner-Ballon O, Saada V, et al. Characteristic repartition of monocyte subsets as a diagnostic signature of chronic myelomonocytic leukemia. Blood. 2015;125(23):3618-3626. 8. Fingerle G, Pforte A, Passlick B, et al. The novel subset of CD14+/CD16+ blood monocytes is expanded in sepsis patients. Blood. 1993;82(10):3170-3176. 9. Chimen M, Yates CM, McGettrick HM, et al. Monocyte subsets co regulate inflammatory responses by integrated signaling through TNF and IL-6 at the endothelial cell Interface. J Immunol. 2017;198(7):2834-2843. 10. Papo M, Corneau A, Cohen-Aubart F, et al. Immune phenotyping of

Erdheim-Chester disease through mass cytometry highlights decreased proportion of non-classical monocytes and increased proportion of Th17 cells. Ann Rheum Dis. 2020;79(11):1522-1524. 11. Ziegler-Heitbrock L, Ancuta P, Crowe S, et al. Nomenclature of monocytes and dendritic cells in blood. Blood. 2010;116(16):e74-80. 12. Zhang L, Singh RR, Patel KP, et al. BRAF kinase domain mutations are present in a subset of chronic myelomonocytic leukemia with wild-type RAS. Am J Hematol. 2014;89(5):499-504. 13. Zhuang J, Han Y, Xu D, et al. Comparison of circulating dendritic cell and monocyte subsets at different stages of atherosclerosis: insights from optical coherence tomography. BMC Cardiovasc Disord. 2017;17(1):270. 14. Passlick B, Flieger D, Ziegler-Heitbrock HW. Identification and characterization of a novel monocyte subpopulation in human peripheral blood. Blood. 1989;74(7):2527-2534. 15. Netea MG, Domínguez-Andrés J, Barreiro LB, et al. Defining trained immunity and its role in health and disease. Nat Rev Immunol. 2020;20(6):375-388. 16. Molteni R, Biavasco R, Stefanoni D, et al. Oncogene-induced maladaptive activation of trained immunity in the pathogenesis and treatment of Erdheim-Chester disease. Blood. 2021;138(17):1554-1569. 17. Coombs CC, Zehir A, Devlin SM, et al. Therapy-related clonal hematopoiesis in patients with non-hematologic cancers is common and associated with adverse clinical outcomes. Cell Stem Cell. 2017;21(3):374-382. 18. Shlush L. Age-related clonal hematopoiesis. Blood. 2018;131(5):496-504.

Haematologica | 108 December 2023

3448

LETTER TO THE EDITOR

Mechanisms of endothelial injury and transplant-associated thrombotic microangiopathy in tandem autologous hematopoietic stem cell transplant for neuroblastoma Tandem autologous hematopoietic stem cell transplant (auto-HSCT) with cyclophosphamide/thiotepa (Cy/Thio) prior to the first auto-HSCT and carboplatin/etoposide/melphalan (CEM) prior to the second auto-HSCT is now standard therapy for high-risk neuroblastoma patients.1 This regimen is associated with an increased risk of transplantassociated thrombotic microangiopathy (TA-TMA), reported as 25% in a prospective multicenter TA-TMA screening study.2 It is unclear why neuroblastoma patients are more prone to this complication than other auto-HSCT recipients who infrequently develop TA-TMA. We hypothesized that increased endothelial injury occurs after the second autoHSCT and that novel endothelial injury pathways contribute to TA-TMA in this unique patient population. In order to test this hypothesis, we prospectively measured circulating endothelial cells (CEC) in auto-HSCT recipients with neuro-

blastoma. Moreover, we performed an RNA-sequencing (RNA-seq) analysis using an in vitro model of TA-TMA. Neuroblastoma patients who consented to our HSCT repository and underwent tandem auto-HSCT between July of 2019 and July of 2020 were included in the CEC study. TA-TMA was diagnosed and risk-stratified prospectively using Jodele criteria.3 Previously published methods for immunomagnetic separation and identification of CEC were used.4 For our in vitro TA-TMA model studies, we used stored serum samples from neuroblastoma patients who consented to our HSCT repository and underwent tandem auto-HSCT between 2017 and 2022. Primary kidney glomerular endothelial cells (CellBiologics, H-6014G) were used in these experiments based on the predominance of kidney involvement in TA-TMA.5 Glomerular endothelial cells were cultured in two conditions: 20%

Figure 1. Circulating endothelial cells change from baseline during tandem autologous autologous hematopoietic stem cell transplant for neuroblastoma. (A) Circulating endothelial cells (CEC) isolated from autologous hematopoietic stem cell transplant (HSCT) patients and stained with acridine orange to confirm nuclear material show numerous immunomagnetic beads coupled to a CD146 antibody. (B) CEC kinetics were prospectively measured in 4 pediatric patients with neuroblastoma who underwent tandem auto-HSCT with cyclophosphamide/thiotepa (auto-HSCT 1) and carboplatin/ etoposide/ melphalan (auto-HSCT 2) conditioning. Patients 1-3 suffered from transplant-associated thrombotic microangiopathy TA-TMA and/or veno-occlusive disease (VOD) after their second auto-HSCT and all 3 of these patients more than doubled their baseline CEC count. Patient 4 had no major complications and did not double her baseline CEC count after auto-HSCT. Haematologica | 108 December 2023

3449

LETTER TO THE EDITOR serum from auto-HSCT recipients with neuroblastoma who developed TA-TMA and 20% serum from auto-HSCT recipients with neuroblastoma who did not develop TATMA. Endothelial cells were harvested after 24 and 72 hours of culture and bulk RNA-seq was performed by the Genomics, Epigenomics and Sequencing Core at the University of Cincinnati. RNA-seq data were deposited at Gene Expression Omnibus (GEO) and are publicly accessible (GSE227176). Analysis for differentially expressed genes (DEG) was performed using Basespace Illumina (RNA-Seq Differential Expression version 1.0.1). Enrichment analyses of DEG were performed using MSigDB and Enrichr databases.6 Adjusted P values were calculated using Benjamini-Hochberg correction for false discovery. Adjusted P values for DEG and pathway comparisons were considered statistically significant at a cutoff of <0.05. Descriptive statistics were performed and expressed as medians with range or interquartile range (IQR). Differences in sC5b-9 levels and fold changes were analyzed using two-side t tests and a P value cutoff of <0.05 for statistical significance. Four neuroblastoma patients were prospectively enrolled in the CEC study, and three were diagnosed with high-risk TA-TMA after their second auto-HSCT (Online Supplemen-

tary Table S1). TA-TMA diagnosis occurred at a median of 9 days (range, 6-10 days) after CEM conditioning for the second auto-HSCT. Two of three patients with high-risk TA-TMA received therapy with the terminal complement blocker, eculizumab (patients 1 and 3). The remaining patient with high-risk TA-TMA (patient 2) was simultaneously diagnosed with hepatic veno-occlusive disease (VOD) and responded to VOD-directed therapy (defibrotide and methylprednisolone), threfore, did not require eculizumab. CEC were measured in 68 blood samples from the four enrolled patients (median, 16 samples/patient; range, 1422 samples/patient). All patients diagnosed with TA-TMA more than doubled their CEC count from baseline, and TA-TMA uniformly occurred after CEM conditioning for the second auto-HSCT (Figure 1B). The maximum change in CEC (∆CEC) scores from baseline for TA-TMA patients similarly occurred during (day -6, patient 1; day -1, patient 2) or after (day 14, patient 3) CEM conditioning for the second HSCT. Patient 4 did not have any significant complications, and the ∆CEC score remained <2 for the duration of transplant (maximum ΔCEC score=1.8-fold on day 12 from Cy/Thio). In our in vitro TA-TMA studies, TA-TMA serum samples (n=5) were obtained prior to eculizumab initiation and

Figure 2. RNA sequencing analysis of glomerular endothelial cells exposed to serum from patients with transplantassociated thrombotic microangiopathy (TA-TMA) versus no TA-TMA. Heatmaps showing differentially expressed genes (DEG) in a comparison of cells exposed to serum from patients with TA-TMA and without TA-TMA at (A) 24 hours (N=28 DEG) and (B) 72 hours (N=2,051 DEG).

Haematologica | 108 December 2023

3450

LETTER TO THE EDITOR time point matched samples were selected as controls from patients who did not develop TA-TMA (n=4; Online Supplementary Table S2). Soluble C5b-9 levels were significantly higher (P=0.008) in TA-TMA patient serum (median, 324 ng/mL; IQR, 295.5-403 ng/mL) compared to serum from patients without TA-TMA (median, 185 ng/mL; IQR, 158.3-261.3 ng/mL; Online Supplementary Figure S1). RNA-seq from cells cultured for 24 hours with TA-TMA patient serum compared to control patient serum showed 28 DEG (27,914 potential genes annotated and 11,453 genes assessed for statistical significance; Figure 2A). Pathway analysis using MsigDB and Enrichr software identified four differentially expressed pathways at this time point: hypoxia, KRAS signaling, interferon-γ response, and TGF-β signaling (Table 1A). A comparison of cells cultured for 72 hours with TA-TMA patient serum versus control patient serum identified 2,051 DEG (27,914 potential genes annotated and 13,338 genes assessed for statistical significance; Figure 2B). Pathway analysis of the top 500 differentially expressed genes using MsigDB and Enrichr software identified 18 differentially expressed pathways at this time point (Table 1B). The most significant pathway differences are related to cell cycle regulation (E2F targets; P=1E-12 and G2-M checkpoint; P=1E-10)). Complement (P=7E-6) and coagulation (P=8E-7) pathways were also significantly different. KRAS signaling and hypoxia pathways remained significant at 72 hours, while interferon-γ and TGF-β pathways were no longer significantly different. This perhaps suggests an early contribution of interferon-γ and TGF-β to TA-TMA initiation events. This study measured CEC in a cohort of pediatric neuroblastoma patients undergoing tandem auto-HSCT, a population at increased risk of TA-TMA. CEC doubled from baseline in all TA-TMA patients and elevations were observed early after the second auto-HSCT, during conditioning chemotherapy. This supports our hypothesis that peak endothelial injury occurs after the second autoHSCT for neuroblastoma leading to TA-TMA. Dvorak et al. describe a “three-hit hypothesis” of TA-TMA initiation where endothelial injury from conditioning chemotherapy is the second hit.7 In tandem auto-HSCT for neuroblastoma this third hit could reasonably be CEM conditioning for the second transplant. TA-TMA onset following CEM is also earlier than typical TA-TMA onset after allogeneic HSCT, which further supports the hypothesis that CEM conditioning itself, or release of toxic intracellular molecules as a consequence of lysis of hematopoietic cells by conditioning, acts as a third endothelial hit for these patients.3,8 TA-TMA risk factors are well-described but the mechanism of TA-TMA initiation and maintenance remain elusive. Complement and coagulation pathways have a known role in TA-TMA and we observed differential expression of

these pathways in our in vitro model comparing endothelial cells cultured with TA-TMA serum with endothelial cells cultured with serum from auto-HSCT recipients without TA-TMA. These complement and coagulation pathway observations support the relevance of this in vitro model and suggest valuable mechanistic information can be derived from these experiments. Differential expression of several novel pathways was observed in our in vitro TATMA model and likely contributes to both TA-TMA initi-

Table 1. Pathway analyses using Enrichr and MsigDB Hallmark 2020 show differentially expressed pathways in cells exposed to serum from patients with transplant-associated thrombotic microangiopathy (TA-TMA) versus no TA-TMA. A MsigDB hallmark pathway analysis (24 hours, TA-TMA vs. no TA-TMA) Term

Adjusted P

Hypoxia

9.33E-04

KRAS signaling up

9.33E-04

Interferon-γ response

9.33E-04

TGF-β signaling

0.01098898 B

MsigDB hallmark pathway analysis (72 hours, TA-TMA vs. no TA-TMA) Term

Adjusted P

E2F targets

1.07E-12

G2-M checkpoint

1.05E-10

Epithelial mesenchymal transition

1.05E-10

Coagulation

8.22E-07

Complement

7.08E-06

Apoptosis

1.1E-04

Cholesterol homeostasis

0.0033116

Myc targets V1

0.0033116

KRAS signaling Up

0.00903304

UV response down

0.01445485

Mitotic spindle

0.01445485

Hypoxia

0.01445485

Estrogen response late

0.01445485

Apical junction

0.01445485

Inflammatory response

0.01445485

Xenobiotic metabolism

0.01445485

Myogenesis

0.0357266

Myc targets V2

0.04109434

Differentially expressed pathways at (A) 24 hours (N=4 pathways) and (B) 72 hours (N=18 pathways) are shown. Adjusted P value calculated using Benjamini-Hochberg correction for multiple hypotheses testing. UV: ultraviolet.

Haematologica | 108 December 2023

3451

LETTER TO THE EDITOR ation and persistence. Hypoxia pathways were differentially expressed in cells exposed to TA-TMA serum compared to non-TA-TMA serum, consistent with recent studies that concluded hypoxia-inducible factor-1α (HIF1α) is a key contributor to complement activation. Hypoxia pathways may be a valuable therapeutic target in TATMA.9,10 Future studies can identify whether the differential expression of hypoxia pathways is related to localized or systemic hypoxia, or both. Hypoxia pathways may also be related to mesenchymal transition which occurs when stressed endothelial cells lose their endothelial cell phenotype and acquire mesenchymal cell morphology and function.11 This is driven primarily through TGF-β signaling11 and we found that TGF-β signaling was differentially expressed in cells exposed to TATMA serum after 24 hours incubation. This early time point observation suggests TGF-β signaling occurs early in TATMA-mediated endothelial injury and is, therefore, an attractive therapeutic target for TA-TMA prevention and treatment. Mauro et al. showed that TGF-β production in microvascular cells exposed to TMA patient sera varied significantly based on the microvascular organ system of origin.12 Pulmonary microvascular cells produced significantly more TGF-β compared to other organ systems, which may be clinically relevant to organ injury patterns in various TMA.12 Future studies should, therefore, incorporate endothelial cells from other organ systems in this in vitro model to investigate tissue-specific differences in endothelial injury mechanisms from TA-TMA. We previously reported that upregulation of interferon-responsive complement genes occurs at the onset of TATMA and returns to baseline after resolution of TA-TMA.13 In the current study of our in vitro TA-TMA model we observed significant interferon pathway activation after 24 hours of culture with serum from patients with TA-TMA. Interestingly, interferon pathway activation preceded complement and coagulation pathway activation in our model. It is possible that genetic predispositions (e.g., interferon pathway polymorphisms) contributed to the increased interferon activation in this study, however, if if this timeline occurs in vivo in TA-TMA patients, early intervention with interferon blockade may prevent pathologic complement activation and the development of TA-TMA. In conclusion, the early timing of CEC peaks and early TA-TMA diagnosis after CEM conditioning supports the hypothesis that CEM conditioning might serve as the “third hit” for TA-TMA initiation in these patients. Differences in hypoxia, interferon-γ response, TGF-β signaling and mesenchymal transition pathways may further contribute to the high incidence of TA-TMA in neuroblastoma patients who received tandem auto-HSCT. This in vitro human model will be useful for studying other aspects of TA-TMA biology, including novel mechanisms of TA-

TMA initiation and strategies for TA-TMA treatment and prevention.

Authors Anthony Sabulski,1,2 Sheyar Abdullah,1 Nathan Luebbering,1 Benjamin Aunins,1,2 Caitlin Castillo,1 Kelly Lake,1 Alexandra Duell,1 Lauren Strecker,1 Lucille Langenberg,1 William Broomhead,1 Scott DiMeo,1,2 Elizabeth A. Odegard,3 Jason T. Blackard,3 Assem G. Ziady,1,2 Alix E. Seif,4 Christopher E. Dandoy,1,2 Benjamin L. Laskin,5 Sonata Jodele1,2 and Stella M. Davies1,2 Division of Bone Marrow Transplantation and Immune Deficiency,

Cincinnati Children’s Hospital Medical Center, Cincinnati, OH; Department of Pediatrics, University of Cincinnati College of

Medicine, Cincinnati, OH; 3Division of Digestive Diseases, University of Cincinnati College of Medicine, Cincinnati, OH; 4Division of Oncology, the Children’s Hospital of Philadelphia, Philadelphia, PA and 5Division of Nephrology, the Children’s Hospital of Philadelphia, Philadelphia, PA, USA Correspondence: A. SABULSKI - Anthony.Sabulski@cchmc.org https://doi.org/10.3324/haematol.2023.283351 Received: April 13, 2023. Accepted: June 1, 2023. Early view: June 15, 2023. ©2023 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures AS has consulted for SOBI and received honorarium from Medscape. SJ holds US Patent US 10815,296 B2, has received research support from Alexion Pharmaceuticals, and consultancies from Omeros, SOBI and Alexion. SMD has received research support from Alexion Pharmaceuticals and consultancies with Novartis, Rocket Pharma, CIRM and neurogene. SJ and BL are co-inventors on US Patent PCT/US2014/055922 compositions and methods for treatment of HSCT-associated thrombotic microangiopathy. The remaining authors have no conflicts of interest to disclose. Contributions AS wrote the manuscript, designed the study, performed the experiments, analyzed the data, performed statistical analyses and performed chart reviews. SA, NL, BA and CC collected specimens, performed the experiments and analyzed the data. KL, AD, LS, WB and LL collected, processed and stored patient samples. SD performed statistical analyses and chart reviews. CED and AES analyzed data and reviewed and edited the manuscript. EAO, JTB

Haematologica | 108 December 2023

3452

LETTER TO THE EDITOR and AZ contributed to study design, reviewed and edited the

Funding

manuscript and provided transcriptomic expertise. BLL analyzed

Research and investigators reported in this publication are

data, performed statistical analyses, and reviewed and edited the

supported by the Eunice Kennedy Shriver National Institute of Child

manuscript. SJ analyzed the data, contributed to study design,

Health and Human Development of the National Institute of Health

performed chart reviews and reviewed and edited the manuscript.

(NIH) under award number R01HD093773 (SJ and SMD), the

SMD designed and supervised the study, wrote and edited the

Thrasher Research Fund Early Career Award 15712 (AS), Cincinnati

manuscript, and analyzed the data.

Children’s Hospital Medical Center Trustee Award (AS) and NIDDK grant DK125418 (BL and JTB).

Acknowledgments We would like to acknowledge Xiang Zhang from the Genomics,

Data-sharing statement

Epigenomics and Sequencing Core in the Department of

All data presented in this manuscript will be shared upon email

Environmental Health at the University of Cincinnati for his

request. RNA-sequencing data were deposited at Gene Expression

assistance with RNA-sequencing processing and interpretation.

Omnibus (GEO) and are publicly accessible (GSE227176).

References 1. Park JR, Kreissman SG, London WB, et al. Effect of tandem autologous stem cell transplant vs single transplant on eventfree survival in patients with high-risk neuroblastoma: a randomized clinical trial. JAMA. 2019;322(8):746-755. 2. Dandoy CE, Rotz S, Alonso PB, et al. A pragmatic multiinstitutional approach to understanding transplant-associated thrombotic microangiopathy after stem cell transplant. Blood Adv. 2021;5(1):1-11. 3. Jodele S, Davies SM, Lane A, et al. Diagnostic and risk criteria for HSCT-associated thrombotic microangiopathy: a study in children and young adults. Blood. 2014;124(4):645-653. 4. Sabulski A, Abdullah S, Luebbering N, et al. Circulating endothelial cells and the study of vascular injury in children undergoing hematopoietic stem cell transplant. Haematologica. 2022;107(12):2950-2954. 5. Laskin BL, Goebel J, Davies SM, Jodele S. Small vessels, big trouble in the kidneys and beyond: hematopoietic stem cell transplantation-associated thrombotic microangiopathy. Blood. 2011;118(6):1452-1462. 6. Chen EY, Tan CM, Kou Y, et al. Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinformatics. 2013;14:128. 7. Dvorak CC, Higham C, Shimano KA. Transplant-associated

thrombotic microangiopathy in pediatric hematopoietic cell transplant recipients: a practical approach to diagnosis and management. Front Pediatr. 2019;7:133. 8. Jodele S, Dandoy CE, Lane A, et al. Complement blockade for TA-TMA: lessons learned from a large pediatric cohort treated with eculizumab. Blood. 2020;135(13):1049-1057. 9. Qi J, Pan T, You T, et al. Upregulation of HIF-1α contributes to complement activation in transplantation-associated thrombotic microangiopathy. Br J Haematol. 2022;199(4):603-615. 10. Sabulski A, Jodele S. What complements complement in transplant-associated thrombotic microangiopathy? Br J Haematol. 2022;199(4):477-479. 11. Piera-Velazquez S, Jimenez SA. Endothelial to mesenchymal transition: role in physiology and in the pathogenesis of human diseases. Physiol Rev. 2019;99(2):1281-1324. 12. Mauro M, Kim J, Costello C, Laurence J. Role of transforming growth factor beta1 in microvascular endothelial cell apoptosis associated with thrombotic thrombocytopenic purpura and hemolytic-uremic syndrome. Am J Hematol. 2001;66(1):12-22. 13. Jodele S, Medvedovic M, Luebbering N, et al. Interferoncomplement loop in transplant-associated thrombotic microangiopathy. Blood Adv. 2020;4(6):1166-1177.

Haematologica | 108 December 2023

3453

LETTER TO THE EDITOR

Cross-intolerance with bosutinib after prior tyrosine kinase inhibitors for Philadelphia chromosome-positive leukemia: long-term analysis of a phase I/II study The approval of tyrosine kinase inhibitors (TKI) has significantly improved patient outcomes versus previous standard of care for Philadelphia chromosome–positive (Ph+) chronic myeloid leukemia (CML) and reduced the frequency of progression from chronic phase (CP) to accelerated phase (AP) and blast phase (BP). Although most patients have a favorable outcome with front-line TKI, approximately half may require switching to an alternative TKI due to resistance or intolerance to one or more prior TKI.1 Long-term (≥4 years) follow-up data from a phase I/II study (clinicaltrials.gov identifier: NCT00261846) was analyzed to assess the safety profile of bosutinib, an orally active, dual Src/Abl TKI, in patients with prior TKI intolerance, and to identify cross-intolerance between bosutinib and prior TKI across patients with Ph+ CP CML and advanced (ADV) leukemias (AP or BP CML or Ph+ acute lymphoblastic leukemia [ALL]). The results confirm that cross-intolerance with bosutinib in patients with prior TKI intolerance was <20%. Bosutinib was approved as a second-line (2L) or third-/ fourth-line (3L/4L) therapy for Ph+ CP, AP, or BP CML following resistance or intolerance to prior TKI therapy based on data from a phase I/II study (clinicaltrials gov identifier: NCT00261846).2-4 Bosutinib is also approved as front-line therapy based on results from the phase III BFORE trial.5 Since TKI share a common mechanism of action, cross-intolerance, i.e., discontinuation of a TKI due to the same adverse event (AE) responsible for discontinuation of a prior TKI, is a frequent concern. Previous reports have suggested this to be modest to minimal for dasatinib or nilotinib after intolerance to prior imatinib therapy.6,7 Study details were published previously.2,3 This analysis included all patients who had developed intolerance to prior imatinib, dasatinib, and/or nilotinib, defined as an inability to take the drug(s) due to treatment-related grade 4 hematologic toxicity lasting >7 days, treatment-related grade ≥3 non-hematologic toxicity, persistent grade 2 toxicity not responding to dose reductions and/or medical management, or loss of previously attained response on a lower TKI dose and an inability to receive a higher dose due to treatmentrelated toxicity. Patients should have recovered to grade ≤1 or to baseline from any toxicities of prior anti-cancer treatment prior to enrollment. Patients received a starting dose of 500 mg bosutinib once daily. Dose increases were permitted for lack of efficacy (in the absence of treatment-related grade ≥3 AE) and dose decreases/interruptions for treatment-related toxicities.

Patients continued treatment with bosutinib until disease progression, death, unacceptable toxicity, or withdrawal of consent. Safety was assessed throughout the study and AE graded according to National Cancer Institute Common Terminology Criteria for Adverse Events, version 3.0. A treatmentemergent AE (TEAE) was defined as any AE that first occurred or worsened in severity after the first dose of bosutinib through 30 days after the last dose. Here we assessed the incidence of AE recurrence, bosutinib dose delay (temporary halt due to an AE) or reduction (a decrease in dose from the initial dose or from an escalated dose, due to an AE), and cross-intolerance across AE and AE clusters (which included a set of Medical Dictionary for Regulatory Activities AE terms). Data from ≥4 years of follow-up from a locked database were used for this analysis. The study was conducted in accordance with the Declaration of Helsinki, the International Conference on Harmonization Good Practice Guidelines, and local regulatory requirements. The protocol was approved by the Institutional Review Board at each study center. All patients provided written informed consent. This analysis included 148 imatinib-intolerant, 75 dasatinib-intolerant (Table 1) and seven nilotinib-intolerant patients. Given the small nilotinib-intolerant cohort, only AE recurrence, bosutinib dose modifications, and cross-intolerance data are presented. Twenty-seven imatinib-intolerant patients were also intolerant to dasatinib (21 in the CP CML 3L cohort and 6 in the ADV cohort) and were included in both imatinib-intolerant and dasatinib-intolerant groups. At study completion, 81.8% and 89.3% of patients had discontinued treatment in the imatinib-intolerant and dasatinib-intolerant cohort respectively, 37.8% and 36.0%, respectively, due to AE. Median duration of bosutinib treatment for imatinib-intolerant and dasatinib-intolerant patients was 11.1 and 8.6 months, respectively; median dose intensity was 417.5 and 428.6 mg/day, respectively (Online Supplementary Figure S1A). Sixty (40.5%) and 27 (36.0%) imatinib- and dasatinib-intolerant patients received bosutinib for >2 years, and 46 (31.1%) and 20 (26.7%) for >4 years, respectively. At all time points, 500 mg/day was the most commonly-utilized dosage and ≥68% of patients with prior imatinib and/or dasatinib intolerance were receiving ≥400 mg/day (Online Supplementary Figure S1B, C). In the imatinib-intolerant group, 147 (99.3%) patients had at least one TEAE of any grade with bosutinib, and 122 (82.4%) had a grade ≥3 TEAE (Online Supplementary Table

Haematologica | 108 December 2023

3454

LETTER TO THE EDITOR S1). In the dasatinib-intolerant group, all patients experienced at least one TEAE while receiving bosutinib; 65 (86.7%) had a grade ≥3 TEAE (Online Supplementary Table S1). In both groups, the most frequent TEAE of any grade were diarrhea, nausea, vomiting, thrombocytopenia, and rash. Overall, of the 142 patients in which the AE responsible for imatinib intolerance was reported, 23 (16.2%) patients were cross-intolerant to bosutinib. The incidence of grade 3/4 recurrence with bosutinib of hematologic AE leading to imatinib discontinuation was highest for thrombocytopenia (n=23 [67.6%]), followed by pancytopenia (n=7 [63.6%]) (Table 2). Thrombocytopenia was also the AE that most frequently led to bosutinib dose delay (n=22 [64.7%]) and dose reduction (n=15 [44.1%]), and led to cross-intolerance in 11 (32.4%) patients. Pancytopenia led to dose delay, dose reduction, or cross-intolerance in six (54.5%), four (36.4%), and four (36.4%) patients, respectively. In addition, three (12.5%) patients were cross-intolerant to bosutinib due to neutropenia. Non-hematologic AE that resulted in bosutinib cross-intolerance were rash (4.5%), diarrhea (7.7%), vomit-

ing (14.3%), pleural effusion (33.3%), renal impairment (100%), pulmonary fibrosis (100%), and glomerular filtration rate decreased (100%) (n=1 each; Table 2). Overall, of the 74 patients in which the AE responsible for dasatinib intolerance was reported, 13 (17.6%) patients were cross-intolerant to bosutinib. Thrombocytopenia and neutropenia recurred at grade 3/4 with bosutinib in all patients with prior dasatinib intolerance due to these AE; thrombocytopenia led to bosutinib dose delay or reduction in 11 (84.6%) and ten (76.9%) patients, respectively, and was the hematologic AE that most frequently led to bosutinib cross-intolerance (n=5 [38.5%]). One (25.0%) and two (16.7%) patients were cross-intolerant to bosutinib due to neutropenia and pancytopenia, respectively (Table 3). Pleural effusion was the most frequent reason for prior dasatinib intolerance (n=23) and recurred with bosutinib at any grade and grade 3/4 in 12 (52.2%) and five (21.7%) patients, respectively. Bosutinib cross-intolerance due to recurrence of pleural effusion occurred in two (8.7%) patients. Other non-hematologic AE resulting in bosutinib cross-intolerance were edema

Table 1. Demographics and disease characteristics of patients with prior imatinib and/or dasatinib intolerance. CP2L N=89 Patient demographics Median age in years (range)

Imatinib-intolerant* CP3L ADV N=36 N=23

Total N=148

Dasatinib-intolerant CP3L ADV Total N=52 N=23 N=75

55 (23-91)

59.5 (23-77)

52 (31-83)

55 (23-91)

57 (25-79)

54 (27-83)

57 (25-83)

Female

53 (59.6)

22 (61.1)

15 (65.2)

90 (60.8)

32 (61.5)

12 (52.2)

44 (58.7)

Male

36 (40.4)

14 (38.9)

8 (34.8)

58 (39.2)

20 (38.5)

11 (47.8)

31 (41.3)

White

55 (61.8)

29 (80.6)

11 (47.8)

95 (64.2)

39 (75.0)

16 (69.6)

55 (73.3)

Asian

21 (23.6)

5 (13.9)

5 (21.7)

31 (20.9)

9 (17.3)

2 (8.7)

11 (14.7)

Black

5 (5.6)

1 (2.8)

5 (21.7)

11 (7.4)

2 (3.8)

4 (17.4)

6 (8.0)

Other

8 (9.0)

1 (2.8)

2 (8.7)

11 (7.4)

2 (3.8)

1 (4.3)

3 (4.0)

2.8 (0.2-15.4)

3.2 (0.1-18.3)

6.1 (0.6-18.3)

6.6 (1.5-20.0)

6.4 (0.6-20.0)

Sex, N (%)

Race, N (%)

Duration of CML and baseline cytogenetic response Median time in years since first 2.7 4.7 diagnosis of Ph+ leukemia (0.1-13.6) (0.6-18.3) (range) Baseline MCyR rate# (%)

24 (30.0) N=80

10 (31.3) N=32

6 (30.0) N=20

40 (30.3) N=132

17 (36.2) N=47

7 (36.8) N=19

24 (36.4) N=66

Baseline CCyR rate# (%)

12 (15.0) N=80

3 (9.4) N=32

4 (20.0) N=20

19 (14.4) N=132

7 (14.9) N=47

4 (21.1) N=19

11 (16.7) N=66

Median duration in months of prior imatinib (range)†

18.1 (<0.1-90.1)

19.7 (1.1-79.2)

13.0 (0.5-88.3)

18.1 (<0.1-90.1)

40.7 (1.1-79.4)

29.3 (4.1-63.0)

34.9 (1.1-79.4)

Median duration in months of prior dasatinib (range)

11.7 (0.9-39.0)

6.2 (0.2-34.6)

8.3 (0.2-39.0)

14.4 (0.9-35.7)

8.8 (0.1-32.7)

11.9 (0.1-35.7)

Previous history of TKI therapy

ADV: advanced Ph+ leukemia cohort; CML: chronic myeloid leukemia; CP2L: chronic phase chronic myeloid leukemia second-line cohort; CP3L: chronic phase chronic myeloid leukemia third-/fourth-line cohort; NA: not applicable; Ph+: Philadelphia chromosome-positive; TKI: tyrosine kinase inhibitor. *All patients in this study were treated with imatinib first; 38 (28 CP3L and 10 ADV) were subsequently treated with dasatinib. † Imatinib-intolerant: data missing for 1 patient in the CP2L cohort; dasatinib-intolerant: data missing for 1 patient in the CP3L cohort. #Analysis included patients who had a valid baseline molecular or cytogenetic assessment. Haematologica | 108 December 2023

3455

LETTER TO THE EDITOR Table 2. Adverse event recurrence, bosutinib dose modifications, and cross-intolerance in patients with prior imatinib intolerance.* Imatinib-intolerant patients, N=148†

Patients who discontinued imatinib due to intolerance, N

Same AE with bosutinib N (%)

142

88 (62.0)

Same grade 3/4 AE with bosutinib N (%) 47 (33.1)

Thrombocytopenia

27 (79.4)

23 (67.6)

22 (64.7)

15 (44.1)

11 (32.4)

Neutropenia

11 (45.8)

9 (37.5)

7 (29.2)

3 (12.5)

Anemia

9 (52.9)

8 (47.1)

3 (17.6)

1 (5.9)

Pancytopenia

10 (90.9)

7 (63.6)

6 (54.5)

4 (36.4)

Leukopenia

4 (66.7)

1 (16.7)

2 (33.3)

Rash

11 (50.0)

2 (9.1)

5 (22.7)

1 (4.5)

Edema

4 (25.0)

Diarrhea

11 (84.6)

5 (38.5)

4 (30.8)

1 (7.7)

Asthenia

2 (25.0)

1 (12.5)

Hepatotoxicity

2 (25.0)

Vomiting

4 (57.1)

1 (14.3)

3 (42.9)

1 (14.3)

Nausea

4 (66.7)

1 (16.7)

Myalgia

3 (75.0)

1 (25.0)

Pleural effusion

1 (33.3)

Abdominal pain

1 (50.0)

Pyrexia

1 (50.0)

Bone pain

1 (50.0)

Renal impairment

1 (100.0)

Angioedema

1 (100.0)

Pulmonary fibrosis

1 (100.0)

Weight decreased

1 (100.0)

Arthralgia

1 (100.0)

Glomerular filtration rate decreased

1 (100.0)

Cause of imatinib intolerance‡

Any AE

Bosutinib dose Bosutinib dose Bosutinib delay due to reduction due to discontinuation same AE same AE due to same AE N (%) N (%) N (%) 48 (33.8) 32 (22.5) 23 (16.2)

Hematologic AE

Non-hematologic AE

AE: adverse event. *All patients who received at least 1 dose of bosutinib were included in the safety analysis. †The AE responsible for imatinib intolerance was not reported for 6 patients. ‡In at least 5 patients in the total population of imatinib-intolerant and/or those resulting in recurrent grade 3/4 AE on bosutinib, or bosutinib dose delay, reduction, or discontinuation. Treatment-emergent AE clusters based on Medical Dictionary for Regulatory Activities (MedDRA; v18) preferred terms (PT): clusters for AE leading to discontinuation of previous tyrosine kinase inhibitor (TKI): Abdominal pain: PT abdominal discomfort, abdominal pain, abdominal pain lower, abdominal pain upper, abdominal tenderness, gastrointestinal (GI) pain, abdominal distention, abdominal symptom. Anemia: PT anemia, hemoglobin decreased. Asthenia: PT asthenia, fatigue. Cardiac failure: PT cardiac failure, cardiac failure congestive, cardiac failure acute, cardiac failure chronic, left ventricular (LV) failure, acute LV failure, chronic LV failure, ventricular failure, ventricular dysfunction, ejection fraction decreased, cardiogenic shock, LV dysfunction, systolic dysfunction. Edema: PT generalized edema, edema, edema peripheral, fluid retention, peripheral swelling, fluid overload. GI toxicity: PT GI toxicity, GI disorder. Glomerular filtration rate (GFR) decreased: PT GFR decreased, creatinine clearance decreased, blood creatinine increased. Hepatotoxicity: PT hepatotoxicity, liver disorder, drug-induced liver injury, cholestatic liver injury, hepatitis cholestatic, hepatitis toxic, hepatocellular injury, liver injury. Leukopenia: PT leukopenia, white blood cell count decreased. Neutropenia: PT neutropenia, neutrophil count decreased. Pancytopenia: PT pancytopenia, bone marrow failure, cytopenia, myelosuppression, hematotoxicity. Rash: PT rash, rash generalized, rash macular, rash maculo-papular, rash popular, eczema, erythema, rash erythematous, exfoliative rash, drug eruption, urticaria. Renal impairment: PT renal impairment, acute kidney injury, renal failure. Thrombocytopenia: PT thrombocytopenia, platelet count decreased. Clusters for AE occurring under bosutinib treatment: Abdominal pain: PT abdominal discomfort, abdominal pain, abdominal pain lower, abdominal pain upper, abdominal tenderness, GI pain, abdominal distention. Anemia: PT anemia, hemoglobin decreased. Asthenia: PT asthenia, fatigue. Cardiac failure: PT cardiac failure, cardiac failure congestive, cardiac failure acute, cardiac failure chronic, LV failure, acute LV failure, chronic LV failure, ventricular failure, ventricular dysfunction, ejection fraction decreased, cardiogenic shock, LV dysfunction, systolic dysfunction. Edema, periorbital edema: PT generalized edema, edema, edema peripheral, fluid retention, peripheral swelling, fluid overload, eyelid edema, periorbital edema, swelling of eyelid. GI toxicity: PT GI toxicity, GI disorder. Hepatotoxicity, hyperbilirubinemia, hypertransaminasemia: PT liver function test increased, transaminases increased, aspartate aminotransferase increased, alanine aminotransferase increased, hepatic enzyme increased, hypertransaminasemia, hyperbilirubinemia, liver function test abnormal, alkaline phosphatase increased, blood bilirubin increased, liver disorder, gamma-glutamyltransferase increased, hepatic function abnormal, hepatotoxicity, drug-induced liver injury, cholestatic liver injury, hepatitis cholestatic, hepatitis toxic, hepatocellular injury, liver injury. Leukopenia: PT leukopenia, white blood cell count decreased. Continued on following page.

Haematologica | 108 December 2023

3456

LETTER TO THE EDITOR Neutropenia: PT neutropenia, neutrophil count decreased. Pancytopenia: PT pancytopenia, bone marrow failure, cytopenia, myelosuppression, hematotoxicity OR all 3 any time on-treatment anemia/hemoglobin decreased AND thrombocytopenia/platelet count decreased AND any of neutropenia/neutrophil count decreased OR lymphopenia/lymphocyte count decreased OR leukopenia/white blood cell count decreased. Pericardial effusion: PT pericardial effusion, cardiac tamponade. Pleuropericarditis: PT pleuropericarditis, pleurisy, pericarditis. Rash: PT rash, rash generalized, rash macular, rash maculo-papular, rash popular, eczema, erythema, rash erythematous, exfoliative rash, drug eruption, urticaria. Renal impairment, GFR decrease: PT acute kidney injury, renal failure, blood creatinine increased, GFR decreased, creatinine renal clearance decreased, renal impairment. Thrombocytopenia: PT thrombocytopenia, platelet count decreased.

Table 3. Adverse event recurrence, bosutinib dose modifications, and cross-intolerance in patients with prior dasatinib intolerance.* Dasatinib-intolerant patients, N=75† Cause of dasatinib intolerance‡

Any AE

Patients Same grade 3/4 Bosutinib dose Bosutinib dose Bosutinib discontinued Same AE with AE with delay due to reduction due discontinuation dasatinib due to bosutinib bosutinib same AE to same AE due to same AE intolerance, N N (%) N (%) N (%) N (%) N (%) 74

48 (64.9)

30 (40.5)

32 (43.2)

21 (28.4)

13 (17.6)

Thrombocytopenia

13 (100.0)

11 (84.6)

10 (76.9)

5 (38.5)

Pancytopenia

7 (58.3)

5 (41.7)

4 (33.3)

2 (16.7)

Neutropenia

4 (100.0)

1 (25.0)

Leukopenia

1 (50.0)

Pleural effusion

12 (52.2)

5 (21.7)

7 (30.4)

3 (13.0)

2 (8.7)

Gastrointestinal toxicity

2 (66.7)

1 (33.3)

Headache

2 (66.7)

1 (33.3)

Rash

1 (33.3)

Edema

1 (50.0)

Cardiac failure

1 (50.0)

Pericardial effusion

1 (50.0)

Gastritis

1 (100.0)

Pleuropericarditis

1 (100.0)

Pyrexia

1 (100.0)

Hematologic AE

Non-hematologic AE

†

AE: adverse event. *All patients who received at least 1 dose of bosutinib were included in the safety analysis. The AE responsible for dasatinib intolerance was not reported or classified for 1 patient. ‡In at least 5 patients in the total population of dasatinib-intolerant patients and/or those resulting in recurrent grade 3/4 AE on bosutinib, or bosutinib dose delay, reduction, or discontinuation. Treatment-emergent AE clusters based on Medical Dictionary for Regulatory Activities (MedDRA; v18) preferred terms (PT): see definition of clusters in Table 2 footnotes.

(50%), cardiac failure (50%), and pericardial effusion (50%) (n=1 each; Table 3). Seven patients were nilotinib-intolerant. One (14.3%) patient with nilotinib intolerance due to pleural effusion was cross-intolerant to bosutinib. Three patients who discontinued nilotinib due to a hematologic AE experienced grade 3/4 AE recurrence (thrombocytopenia: n=2; neutropenia: n=1); none led to bosutinib cross-intolerance (Online Supplementary Table S2). No deaths due to bosutinib cross-intolerance occurred in prior imatinib-, dasatinib-, or nilotinib-intolerant patients. Cross-intolerance with bosutinib was low in prior imatinib-, dasatinib-, and nilotinib-intolerant patients, consistent with a previous study in patients with CP CML receiving bosutinib as a 4L treatment.8 Most incidences of cross-intolerance occurred in those who discontinued prior TKI treatment due to a hematologic AE, a finding consistent with previous cross-intolerance analyses of dasatinib and nilotinib.6,7 AE that recurred during treatment with bosutinib

were manageable through bosutinib dose modifications and, in the majority of cases, did not lead to bosutinib discontinuation. The incidence of cross-intolerance due to hematologic AE was similar in imatinib- and/or dasatinib-intolerant patients. Overall, thrombocytopenia was a common recurring hematologic AE leading to bosutinib discontinuation in prior imatinib- and dasatinib-intolerant patients. This finding supports previous studies that suggest thrombocytopenia to be the most likely cause of crossintolerance among TKI.6,7 The most common non-hematologic AE leading to discontinuation of prior TKI were rash, edema, and diarrhea in imatinib-intolerant patients and pleural effusion in dasatinib-intolerant patients. For these AE that recurred with bosutinib, discontinuations were rare. Indeed, despite the high incidence of recurrence of diarrhea (84.6%) in patients with prior imatinib intolerance due to this AE, previous diarrhea is not a contraindication for bosutinib,

Haematologica | 108 December 2023

3457

LETTER TO THE EDITOR as <40% of patients experienced recurrence at grade 3/4, and <10% (1 patient) discontinued bosutinib due to this AE. Similarly, despite recurrence of pleural effusion in 52.2% of patients with previous dasatinib intolerance due to this AE, only 8.7% (2 patients) discontinued bosutinib due to pleural effusion. In conclusion, results showed that cross-intolerance with bosutinib in patients with prior TKI intolerance was <20%. Results also confirmed that cross-intolerance was limited for the most common AE associated with other TKI. Patients were pooled across cohorts (CP and ADV) to have a larger patient number for the purposes of these analyses. The authors recognize the limitation of including patients with advanced leukemias, as these patients may experience a higher incidence of specific types of AE, e.g., hematologic AE, compared with CP CML cohorts. Additionally, another limitation of this analysis is the small number of patients in some subgroups; therefore, an analysis of AE recurrence and cross-intolerance due to infrequent but important AE, such as pericardial effusion or renal impairment, is limited. In line with recommendations for management of AE,9 recurrent AE with bosutinib were generally manageable with dose modifications and/or standard medical therapy. Although 500 mg is the recommended starting dose for patients previously treated with ≥1 TKI, other reports have shown that patients with prior TKI intolerance receive a lower dose intensity compared with patients resistant to previous TKI.10 This suggests a higher dose intensity may be necessary for patients resistant to previous TKI, while a lower dose intensity might be sufficient to maintain response and improve tolerability in patients intolerant to prior TKI.11 Some studies have evaluated and/or are currently evaluating run-in dosing regimens or alternative schedules starting at lower bosutinib doses and escalating based on tolerability and response.12-15 This dosing strategy has also been suggested for the management of AE during bosutinib treatment.9 Bosutinib has demonstrated durable efficacy at 2- and 5-year follow-ups in patients receiving the drug as a 2L therapy16,17 and at ≥4-year follow-up when used as a 3L/4L therapy18 in patients with prior TKI intolerance. Taken together, efficacy and safety data support use of bosutinib as an effective treatment option for Ph+ CML patients with prior TKI intolerance.

“Seràgnoli”, IRCCS Azienda Ospedaliero-Universitaria di Bologna, Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna, Bologna, Italy; 4McGill University, Montreal, Quebec, Canada; 5Department of Hematology, Oncology, Hemostaseology and Stem Cell Transplantation, Faculty of Medicine, RWTH Aachen University Hospital, Aachen, Germany; 6

Center for Integrated Oncology Aachen Bonn Cologne Düsseldorf

(CIO ABCD), Aachen Bonn Cologne Düsseldorf, Germany; 7Pfizer Inc, Cambridge, MA, USA; 8Pfizer SLU, Madrid, Spain and 9University of Milano-Bicocca, Monza, Italy Correspondence: J.E. CORTES - jorge.cortes@augusta.edu https://doi.org/10.3324/haematol.2022.281944 Received: August 15, 2022. Accepted: June 30, 2023. Early view: July 13, 2023. ©2023 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures JEC received research support to his institution from Bristol Myers Squibb, Novartis, Pfizer, Sun Pharmaceutical Industries, and Takeda; served as a consultant for Bristol Myers Squibb, Novartis, Pfizer, Takeda, and Fusion Pharmaceuticals. JHL received lecture fees and research support to his institution from Pfizer. VK received honorarium for advisory board participation from Ariad Pharmaceuticals, Incyte, Novartis, Pfizer, and Xcenda, and received research support to his institution from Pfizer. FC received research support to his institution from Pfizer and has served as a speaker or consultant for Bristol Myers Squibb, Incyte, Novartis, and Pfizer. SA received research support to his institution from Roche/Genentech and Pfizer and has served as a speaker or consultant for Roche Canada, Pfizer, Bristol Myers Squibb, Palladin, and Lundbeck. THB served as a consultant for Novartis, Pfizer, Janssen, Merck, Takeda, and received research funding from Novartis and Pfizer. AV and EL are employees of Pfizer. CG-P provides consultancy to Bristol Myers Squibb and received honoraria and research support from Pfizer. Contributions JEC, THB, and CG-P were involved in the study conception/design.

Authors

All authors were involved in the acquisition, analysis, or interpretation of data. All authors contributed to the drafting of the manuscript and approved the final version. 1

Jorge E. Cortes, Jeff H. Lipton, Vamsi Kota, Fausto Castagnetti,

Sarit Assouline,4 Tim H. Brümmendorf,5,6 Eric Leip,7 Andrea Viqueira8 and Carlo Gambacorti-Passerini

Funding This study (clinicaltrials gov. identifier: NCT00261846) was sponsored by Pfizer. Medical writing support was provided by Katy

Georgia Cancer Center, Augusta, GA, USA; Princess Margaret 3

Cancer Center, Toronto, Ontario, Canada; Istituto di Ematologia

Beck, PhD, of Engage Scientific Solutions (Horsham, UK) and funded by Pfizer.

Haematologica | 108 December 2023

3458

LETTER TO THE EDITOR Acknowledgments

that support the findings of this study. Subject to certain criteria,

The authors thank Dr Simon Durrant for his contributions to the study.

conditions and exceptions, Pfizer may also provide access to the related individual de-identified participant data. See

Data-sharing statement

https://www.pfizer.com/science/clinical-trials/trial-data-and-results

Upon request, and subject to review, Pfizer will provide the data

for more information.

References 1. Garcia-Gutierrez V, Hernandez-Boluda JC. Tyrosine kinase inhibitors available for chronic myeloid leukemia: efficacy and safety. Front Oncol. 2019;9:603. 2. Cortes JE, Kantarjian HM, Brummendorf TH, et al. Safety and efficacy of bosutinib (SKI-606) in chronic phase Philadelphia chromosome-positive chronic myeloid leukemia patients with resistance or intolerance to imatinib. Blood. 2011;118(17):4567-4576. 3. Khoury HJ, Cortes JE, Kantarjian HM, et al. Bosutinib is active in chronic phase chronic myeloid leukemia after imatinib and dasatinib and/or nilotinib therapy failure. Blood. 2012;119(15):3403-3412. 4. Gambacorti-Passerini C, Kantarjian HM, Kim DW, et al. Longterm efficacy and safety of bosutinib in patients with advanced leukemia following resistance/intolerance to imatinib and other tyrosine kinase inhibitors. Am J Hematol. 2015;90(9):755-768. 5. Cortes JE, Gambacorti-Passerini C, Deininger MW, et al. Bosutinib versus imatinib for newly diagnosed chronic myeloid leukemia: results from the randomized BFORE trial. J Clin Oncol. 2018;36(3):231-237. 6. Cortes JE, Hochhaus A, le Coutre PD, et al. Minimal crossintolerance with nilotinib in patients with chronic myeloid leukemia in chronic or accelerated phase who are intolerant to imatinib. Blood. 2011;117(21):5600-5606. 7. Khoury HJ, Goldberg SL, Mauro MJ, et al. Cross-intolerance with dasatinib among imatinib-intolerant patients with chronic phase chronic myeloid leukemia. Clin Lymphoma Myeloma Leuk. 2016;16(6):341-349. 8. Garcia-Gutierrez V, Martinez-Trillos A, Lopez Lorenzo JL, et al. Bosutinib shows low cross intolerance, in chronic myeloid leukemia patients treated in fourth line. Results of the Spanish compassionate use program. Am J Hematol. 2015;90(5):429-433. 9. Cortes JE, Apperley JF, DeAngelo DJ, et al. Management of adverse events associated with bosutinib treatment of chronicphase chronic myeloid leukemia: Expert Panel review. J Hematol Oncol. 2018;11(1):143. 10. Hochhaus A, Gambacorti-Passerini C, Abboud C, et al. Bosutinib

for pretreated patients with chronic phase chronic myeloid leukemia: primary results of the phase 4 BYOND study. Leukemia. 2020;34(8):2125-2137. 11. Gambacorti-Passerini C, Brümmendorf TH, Abruzzese E, et al. Efficacy and safety of bosutinib in previously treated patients with chronic myeloid leukemia: final results from the Byond Trial. Blood. 2021;138(Suppl 1):S1475. 12. Mita A, Abumiya M, Miura M, et al. Correlation of plasma concentration and adverse effects of bosutinib: standard dose or dose-escalation regimens of bosutinib treatment for patients with chronic myeloid leukemia. Exp Hematol Oncol. 2018;7:9. 13. Isfort S, Wolf D, Teichmann LL, et al. Step-in dosing in the Bosutinib Dose Optimization Study (BODO) failed to reduce gastrointestinal (GI) toxicity in patients failing second generation TKI (2G-TKI) in chronic phase chronic myeloid leukemia (CML) but suggests promising molecular response. Blood. 2021;138(Suppl 1):S3608. 14. Castagnetti F, Gugliotta G, Bocchia M, et al. Dose optimization in elderly CML patients treated with bosutinib after intolerance or failure of first-line tyrosine kinase inhibitors. Blood. 2019;134(Suppl 1):S496. 15. Latagliata R, Attolico I, Trawinska MM, et al. Bosutinib in the real-life treatment of chronic myeloid leukemia patients aged >65 years resistant/intolerant to previous tyrosine-kinase inhibitors. Hematol Oncol. 2021;39(3):401-408. 16. Gambacorti-Passerini C, Brümmendorf TH, Kim DW, et al. Bosutinib efficacy and safety in chronic phase chronic myeloid leukemia after imatinib resistance or intolerance: minimum 24-month follow-up. Am J Hematol. 2014;89(7):732-742. 17. Gambacorti-Passerini C, Cortes JE, Lipton JH, et al. Safety and efficacy of second-line bosutinib for chronic phase chronic myeloid leukemia over a five-year period: final results of a phase I/II study. Haematologica. 2018;103(8):1298-1307. 18. Cortes JE, Khoury HJ, Kantarjian HM, et al. Long-term bosutinib for chronic phase chronic myeloid leukemia after failure of imatinib plus dasatinib and/or nilotinib. Am J Hematol. 2016;91(12):1206-1214.

Haematologica | 108 December 2023

3459

LETTER TO THE EDITOR

Efficacy and toxicity of midostaurin with idarubicin and cytarabine induction in FLT3-mutated acute myeloid leukemia Among patients able to tolerate intensive induction, standard acute myeloid leukemia (AML) induction therapy is a regimen commonly referred to as “7+3,” consisting of 7 days of intravenous cytarabine and 3 days of daunorubicin (DNR) or idarubicin (IDA).1 Depending on patients’ genetic markers, targeted therapies may be added to the 7+3 backbone; in the pivotal trial establishing midostaurin as part of the standard of care for induction in fms-like tyrosine kinase 3 (FLT3)-mutated AML (RATIFY trial2), DNR 60 mg/m2 was the anthracycline used. Although dose equivalence of IDA compared to DNR has not been definitively established, IDA 12 mg/m2 is considered equivalent to DNR 90 mg/m2. DNR 60 mg/m2 was the anthracycline dose selected for the RATIFY trial, despite data supporting benefit of DNR 90 mg/m2 over 60 mg/m2 in FLT3-internal tandem duplication (ITD)-mutated AML.3-5 Few published data exist assessing efficacy or toxicity of midostaurin in combination with an IDA-based 7+3 AML induction regimen. This multicenter, retrospective study assessed efficacy and toxicity associated with midostaurin and IDA-based 7+3 induction. Complete remission (CR) was attained in 83 of 114 patients (73%) receiving midostaurin and IDA-based 7+3 induction. Event-free survival (EFS) was 11.7 months, and no unexpected toxicity signals were noted. Data regarding the use of midostaurin with IDA-based 7+3 induction are limited. Two small, retrospective studies examining the use of IDA-based 7+3 with midostaurin reported outcomes similar to those seen with DNR-based induction. A midostaurin expanded access program of 32 patients given IDA-based 7+3 induction reported higher rates of diarrhea and hypophosphatemia with IDA and midostaurin compared to DNR and midostaurin, but rates of other non-hematologic adverse events were similar, and rates of hematologic adverse events by induction anthracycline were not reported.6 A single-center retrospective review of 20 patients with FLT3-mutated AML given midostaurin in addition to IDA-based 7+3 induction found a higher incidence of rash compared with the RATIFY results, with higher CR rate and shorter duration of neutropenia; however, all patients in this small study received granulocyte colony-stimulating factor 14 days after starting induction, which likely contributed to the shorter duration of neutropenia.7 In order to assess the midostaurin/IDA-based 7+3 induction regimen, we performed a multicenter, retrospective study of patients who received midostaurin and 7+3 with IDA as the anthracycline. The primary objective was CR

rate after induction with midostaurin and IDA-based 7+3. CR was attained if patients met all of the following criteria: less than 5% blasts in the marrow, absence of extramedullary leukemia, an absolute neutrophil count of ≥1,000/μL, a platelet count of ≥100,000/μL, and the absence of blasts in the peripheral blood. The secondary objective was EFS, defined as time to relapse, mortality, or failure to achieve CR by day 60. Secondary objectives included assessment of hematologic and non-hematologic toxicities during induction. Adverse events were graded per CTCAE version 4.03. We performed post hoc analyses assessing efficacy and toxicity in those patients 60 years of age and older as those patients would have been excluded from the RATIFY trial, and sub-group analyses of EFS stratified by FLT3 and NPM1 mutation status. At each study institution, a report was generated through the electronic medical record identifying patients who initiated IDA-based 7+3 with midostaurin induction therapy for AML between April 1, 2017 and December 31, 2020. Patients under the age of 18 years, incarcerated patients, pregnant patients, patients who received midostaurin as part of a clinical trial, and patients receiving midostaurin for systemic mastocytosis were excluded. Data collected from eligible patients’ charts were recorded in a REDCap database hosted by the University of Utah. At each institution, research was exempt by the respective Institutional Review Boards and Ethics Committees; written informed consent was waived due to the retrospective nature of the study. In all, 114 patients were included across five academic centers in the US. Baseline characteristics (Table 1) show a median age at induction of 56 years (range, 19-74), with 35 patients (31%) 60 years or older. The median duration of follow-up was 18.1 months (range, 0.5-51.3). Thirty-two patients (28%) carried FLT3-tyrosine kinase domain, and 87 (76%) carried FLT3-ITD mutations. Eighty-eight patients (78%) completed a 14-day midostaurin course. Eleven of 114 (9%) received midostaurin dose reduction due to drugdrug interactions and 2 (1.8%) received dose reduction due to toxicity. Twenty-four (21%) received growth factor during induction after day 14. Seventy-nine (69%) of patients proceeded to consolidation chemotherapy, and midostaurin was given during consolidation to 74 patients (64.3%), with 14 (12.2%) continuing maintenance midostaurin (Online Supplementary Figure S1). Seventy-four (64.3%) proceeded to allogeneic stem cell transplant, and four patients received maintenance midostaurin post-transplant.

Haematologica | 108 December 2023

3460

LETTER TO THE EDITOR Of the 114 patients who underwent induction with midostaurin and IDA-based 7+3, 83 (72.8%) achieved CR, with a relatively higher CR rate in the age under 60 years subgroup (Table 2). Median time to CR was 34 days. Median duration of neutropenia was 25 days, and median duration of thrombocytopenia was 27 days (Table 2). Median duration of follow-up was 18.1 months. Median EFS was 11.7 months (95% confidence interval [CI]: 9.8-24) (Figure 1) with median EFS of 13.4 months (95% CI: 10.3-not reached [NR]) in the age under 60 years subgroup and 10.8 months (95% CI: 5.1-NR) in the age over 60 years subgroup (Online Supplementary Figure S2). Censoring for transplant, median EFS was 9.8 months (95% CI: 7.9-NR). Among the FLT3-ITD- mutated population, there was a trend toward improved EFS with mutated NPM1 compared to wild-type, with median EFS not reached in the NPM1 mutant population and 6.9 months in the wild-type population (Online Supplementary Figure S2). Table 2 shows rates of grade 3 or higher adverse events during induction. In both the under the age of 60 years and over the age 60 years subgroups, IDA-based 7+3 induction with midostaurin appeared to have a similar toxicity profile to DNR-based 7+3 induction reported in the RATIFY trial. This multicenter, retrospective study found an overall CR rate of 72.8% with use of IDA in 7+3 induction with midostaurin in patients with FLT3-mutant AML, including patients 60 years of age or older. The CR rate was 77.2% and 62.9%, in patients younger and older than 60, respectively. Although a direct comparison to the RATIFY study is not possible, these rates stand favorably in comparison to the RATIFY trial, in which a 58.9% CR rate was achieved in pa-

tients younger than 60 years. The high CR rate observed in this work could in part be due to the IDA dose of 12 mg/m2; whereas the RATIFY study used DNR 60 mg/m2. Previous studies in patients Table 1. Baseline characteristics and patient disposition. Baseline characteristics

All patients, N=114

Median age at induction in years (range)

56 (19-74)

Female sex, N (%)

53 (46.5)

ECOG performance status, N (%) 0 1 2 3 Not reported

40 (35.1) 50 (43.8) 9 (7.9) 2 (1.8) 13 (11.4)

FLT3 mutation subtype, N (%) TKD ITD-higha ITD-low ITD allelic ratio not quantified

32 (28.0) 19 (16.7) 29 (25.4) 39 (34.2)

2017 ELN classification, N (%) Favorable Intermediate Adverse Unknown

17 (14.9) 57 (50.0) 38 (33.3) 2 (1.8)

Median WBC on day 1, x103/μL (range)

21.3 (0.2-301)

Median platelets on day 1, x103/μL (range)

39 (8-247)

Median ANC on day 1, x103/μL (range)

1.1 (0-22.2)

Defined as FLT3 ITD ratio greater than 0.7. ECOG: Eastern Cooperative Oncology Goup; TKD: tyrosine kinase domain; ITD: internal tandem dublication; ELN: European Leukemia Net; WBC: white blood cell count; ANC: absolute neutrophil count. a

Table 2. Complete remission summary and summary of grade 3 or higher adverse events during induction. All patients N=114

Age <60 years N=79

Age ≥60 years N=35

Protocol-specified CR, N (%)

83 (72.8)

61 (77.2)

22 (62.9)

Did not attain protocol-specified CR, N (%)

28 (24.6)

16 (20.2)

12 (34.2)

Remission assessment data missing, N (%)

3 (2.6)

2 (2.5)

1 (2.8)

Median time to CR in days (range)

34 (23-56)

34 (23-66)

35 (24-56)

Median duration of ANC <500x103/μL in days (IQR)

25 (21-30)

25 (22-31)

25 (21-30)

Median duration of platelets <100x103/μL in days (IQR)

27 (24-35)

27 (23-35)

28 (26-50)

Febrile neutropenia

101 (89)

73 (92)

28 (80)

Other infection

32 (28)

20 (25)

12 (34)

Mucositis

15 (13)

9 (11)

6 (17)

Nausea

5 (4)

4 (5)

1 (3)

Diarrhea

19 (17)

9 (11)

10 (29)

Pneumonitis

1 (1)

0 (0)

1 (3)

Adverse events, grade 3 or higher, N (%)

CR: complete remission; ANC: absolute neutrophil count; IQR: interquartile range. Haematologica | 108 December 2023

3461

LETTER TO THE EDITOR

Figure 1. Kaplan-Meier curves of event-free survival. (A) Event-free survival in the overall study population and (B) overall study population censoring at time of transplant. NR: not reached.

with the FLT3-ITD mutation found an overall survival benefit using high-dose DNR (90 mg/m2) rather than lower DNR doses (60 or 45 mg/m2)3-5 or IDA 12 mg/m2.8 For patients with FLT3-mutated AML, high-dose DNR equivalence to IDA in combination with midostaurin remains an open question. Overall survival was not assessed in this study, due to the short median follow-up of 18.1 months. However, in a triallevel analysis of eight randomized studies of intensive chemotherapy in newly-diagnosed AML, an association was noted between EFS and overall survival, suggesting EFS may serve as a valid surrogate endpoint for initial AML treatment.9 A favorable EFS of 11.7 months was observed across our study population, with a high proportion of patients who proceeded to transplantation. Between the younger than 60 years and 60 years and older subgroups, EFS was comparable, with EFS of 13.4 months and 10.8 months respectively. Subgroup analysis of patients with FLT3-ITD mutations showed improved EFS with mutated NPM1 compared to unmutated NPM1. The toxicity profile of IDA-based 7+3 with midostaurin was comparable to data reported with DNR-based 7+3 induction. We observed a low rate of grade 3 or higher infection, likely due to high rates of posaconazole or voriconazole use as antifungal prophylaxis during induction (87% of patients), with only two patients (1.8%) contracting a fungal infection. Other toxicities noted, such as those observed with mucositis, diarrhea, and pneumonitis were limited by small numbers and the nature of retrospective chart review. In conclusion, this study describes the largest series of FLT3-mutated AML patients treated with midostaurin and IDA in 7+3 induction. Midostaurin and IDA in 7+3 induction demonstrated excellent efficacy and a safety profile similar to that of midostaurin with DNR as the induction anthracycline, in both the age below 60 years and the age of 60 years and older subgroups. Limitations of this work

include the retrospective nature and short follow-up, but these results suggest IDA 12 mg/m2 is a suitable alternative anthracycline to DNR 60 mg/m2 in 7+3 with midostaurin in newly-diagnosed FLT3-mutated AML.

Authors Julia S. Lee,1 Charlotte B. Wagner,1 Stacy Prelewicz,1 Heena P. Kurish,2 Robert Walchack,2 Danielle A. Cenin,2 Seema Patel,2 Mimi Lo,3 Danielle Schlafer,4 Belinda K. T. Li,4 R. Donald Harvey III,4 Bestis Wasef,5 Jian Ying6 and Tibor Kovacsovics7 Department of Pharmacy, Huntsman Cancer Institute at the

University of Utah, Salt Lake City, UT; 2Department of Pharmacy, Cleveland Clinic Foundation, Cleveland, OH; 3Department of Pharmaceutical Services, University of California San Francisco, San Francisco, CA; 4Department of Hematology and Medical Oncology, Winship Cancer Institute/Emory Healthcare, Atlanta, GA; Department of Pharmacy, Oregon Health and Science University

Hospital, Portland, OR; 6Department of Internal Medicine, University of Utah School of Medicine, Salt Lake City, UT and 7Division of Hematology, Huntsman Cancer Institute at the University of Utah, Salt Lake City, UT, USA Correspondence: J. LEE - julialee@stanfordhealthcare.org https://doi.org/10.3324/haematol.2022.281967 Received: December 15, 2022. Accepted: June 15, 2023. Early view: June 22, 2023. ©2023 Ferrata Storti Foundation Published under a CC BY-NC license

Haematologica | 108 December 2023

3462

LETTER TO THE EDITOR Disclosures

data collection form, cleaned and analyzed the data, and drafted

RDH reports financial relationships with Abbisko, AbbVie, Actuate,

the letter. JSL, CW, SPr, DC, HK, SPa, RW, ML, DS, BL and BW

Amgen, AstraZeneca, Bayer, Bristol-Myers Squibb, Boston

collected data. JY assisted with statistical analysis. All authors

Biomedical, Genmab, GlaxoSmithKline, Infinity, InhibRx, Janssen,

reviewed and provided input on the manuscript.

Merck, Mersana, Meryx, Nektar, Novartis, Pfizer, Regeneron, Sanofi, Sutro, Takeda, Turning Point Therapeutics and, Xencor. HK reports

Funding

honoraria from Market Access Transformation. ML reports

Use of REDCap was supported by NIH/NCATS grant UL1TR002538.

consultancy with Oncopeptides and EUSA Pharma. TK reports research funding from Abbvie, Caelum, Gilead, Glycomimetics,

Data-sharing statement

Janssen, Jazz Pharmaceuticals and Syndax; honoraria from Novartis

The REDcap study database is not available per the protocols

and Kite. All other authors have no conflicts of interest to disclose.

approved by the participating institutions’ Institutional Review Boards. The data collection form will be made available upon request to the corresponding author.

Contributions CW, SPr and TK conceived the study question. JSL designed the

References 1. National Comprehensive Cancer Network. Acute myeloid leukemia (Version 3.2023). https://www.nccn.org/professionals/physician_gls/pdf/aml.pdf Accessed June 6, 2023. 2. Stone RM, Mandrekar SJ, Sanford BL, et al. Midostaurin plus chemotherapy for acute myeloid leukemia with a FLT3 mutation. N Engl J Med. 2017;377(5):454-464. 3. Burnett AK, Russell NH, Hills RK, United Kingdom National Cancer Research Institute Acute Myeloid Leukemia Study Group. Higher daunorubicin exposure benefits FLT3 mutated acute myeloid leukemia. Blood. 2016;128(3):449-452. 4. Choi EJ, Lee JH, Lee JH, et al. Comparison of anthracyclines used for induction chemotherapy in patients with FLT3-ITDmutated acute myeloid leukemia. Leuk Res. 2018;68:51-56. 5. Luskin MR, Lee JW, Fernandez HF, et al. Benefit of high-dose daunorubicin in AML induction extends across cytogenetic and molecular groups. Blood. 2016;127(12):1551-1558.

6. Roboz GJ, Strickland SA, Litzow MR, et al. Updated safety of midostaurin plus chemotherapy in newly diagnosed FLT3 mutation-positive acute myeloid leukemia: the RADIUS-X expanded access program. Leuk Lymphoma. 2020;61(13):3146-3153. 7. Azzi J, Cirrone F, Abdul Hay M, et al. Midostaurin in combination with idarubicin and cytarabine (3+7) induction for FLT3 positive AML - very high complete response rates and transition to allogeneic transplantation. Blood. 2018;132(Suppl 1):S5216. 8. Lee JH, Kim H, Joo YD, et al. Prospective randomized comparison of idarubicin and high-dose daunorubicin in induction chemotherapy for newly diagnosed acute myeloid leukemia. J Clin Oncol. 2017;35(24):2754-2763. 9. Norsworthy KJ, Gao X, Ko CW, et al. Response rate, event-free survival, and overall survival in newly diagnosed acute myeloid leukemia: US Food and Drug Administration trial-level and patient-level analyses. J Clin Oncol. 2022;40(8):847-854.

Haematologica | 108 December 2023

3463

LETTER TO THE EDITOR

Latexin deletion protects against radiation-induced hematopoietic damages via selective activation of Bcl-2 prosurvival pathway Exposure to ionizing radiation (IR) causes dysfunction of multiple organs of which the hematopoietic system is the most sensitive tissue.1 Radiation damage to the hematopoietic system induces acute myelosuppression that increases the risk of infection and bleeding.2 It also causes long-term bone marrow (BM) injury, which underlies the development of BM failure or hematological malignancies.3 Therefore, IR-induced acute and long-term BM injuries are the most significant consequence of accidental or intentional exposure to IR and also represent a serious side effect of radiation therapy. To date, few effective medical countermeasures have been developed to protect and mitigate radiation-induced hematological toxicity. Recently, a significant progress has been made toward identifying novel radioprotectants, such as toll-like receptor 5 agonists4 and CDK4/6 inhibitors5, inhibitors of the protein C (aPC) pathway,6 and cell-based therapy such as infusion of endothelial or progenitor cells.7 However, further novel approaches, based on mechanistic data, are warranted since the translation of these findings into the clinic remains a significant challenge. Elucidating cellular and molecular pathways that govern the high sensitivity of hematopoietic cells, particularly hematopoietic stem cells (HSC), to IR can provide a better strategy to rationally develop medical countermeasures against radiation-induced hematopoietic toxicity. We previously discovered that lack of the protein latexin (Lxn) in HSC enhances HSC survival and regeneration in vivo, while Lxn overexpression sensitizes myeloid cell line to radiation.8,9 In this study, we examined the role of Lxn in IR-induced BM injury and HSC damage using the Lxn knock-out mouse model (Lxn-/-). Lxn-/- mice have a significantly increased survival advantage compared to wild-type (WT) mice after lethal doses (8 Gy and 9 Gy) of total body irradiation (TBI) (Figure 1A). IR-induced acute myelosuppression is a primary cause of lethality. Therefore, we monitored the dynamic changes of blood and BM cells at different time points for 2 months post a sub-lethal dose (6.5 Gy) of TBI, and found that blood leukocytes and platelets and BM hematopoietic stem/progenitor cell (HSPC)enriched lineage-Sca1+c-Kit+ (LSK) cells recovered significantly faster in Lxn-/- mice than WT mice during the first month post-IR (Figure 1B). The rapid recovery was not due to the increased proliferation of LSK cells (Online Supplementary Figure S1A). We next determined the role of Lxn deletion in protecting HSC from radiation-induced long-

term damage. We irradiated Lxn-/- and WT mice with 6.5 Gy TBI, collected BM cells at 56 days post-IR at which time blood cell and HSPC counts returned to the normal level (Figure 1B), and performed various functional assays to evaluate HSC regeneration and self-renewal functions, including cobblestone area forming cell (CAFC) assay, in vivo limiting dilution competitive repopulation unit (CRU) assay, and serial transplantation. The results showed that Lxn-/BM had a significantly elevated number of CAFC day 35 cells compared to the WT mice after IR (Figure 1C). The CRU assay showed that Lxn-/- mice indeed preserved a significantly higher frequency of long-term repopulating HSC in the BM (Figure 1D). In competitive repopulation and serial transplantation, BM cells (CD45.2) from irradiated Lxn-/- and WT mice were transplanted into lethally irradiated primary recipient mice (CD45.1) along with an identical number of competitor cells (CD45.1). At 16 weeks post-transplantation, PB and BM chimerism was analyzed for CD45.2-derived cells, and CD45.2 BM cells were sorted and transplanted into the secondary recipients. The same regimen was repeated in the tertiary transplantation. The result showed that Lxn-/- HSC had a higher capacity to regenerate PB and BM LSK cells in the secondary and tertiary recipients than WT HSC (Figure 1E), demonstrating enhanced HSC self-renewal activity.10 Mice or humans exposed to radiation, especially fractionated low-dose radiation regimen, exhibit residual HSC functional defects even months after hematopoiesis has recovered from the exposure.11 We exposed Lxn-/- and WT mice with clinically relevant fractionated low-dose radiation (2 Gy daily for 5 days), and examined the long-term effect at a 16-20month period post-IR along with age-matched non-IR WT and Lxn-/- mice. We found that numbers of BM LSK cells, long-term (LT-), short-term HSC (ST-HSC), and multipotent progenitors (MPP), identified by flow cytometry, were much better preserved in Lxn-/- mice compared to WT mice after radiation although radiation reduced the numbers of these cell populations in both strains (Figure 1F). Radiation induces the accumulation of reactive oxygen species (ROS) and senescence, and DNA damage. We didn’t find any changes in ROS level and senescence in irradiated Lxn-/LSK cells compared to WT cells (Online Supplementary Figure S1B, C). By using γ-H2A.X staining and the comet assay, we found that Lxn-/- LSK cells had fewer γ-H2A.X foci and a shorter length of comet tail post-IR compared to WT cells (Online Supplementary Figure S1D, E), suggest-

Haematologica | 108 December 2023

3464

Continued on following page.

LETTER TO THE EDITOR

Haematologica | 108 December 2023

3465

LETTER TO THE EDITOR Figure 1. Latexin deletion protects radiation-induced acute myelosuppression and long-term hematopoietic stem cells functional impairment. (A) Significant more latexin knock-out (Lxn-/-) mice survived after exposure to 8 Gy (left) or 9 Gy (right) of total body irradiation (TBI) in 137Cesium from JL Shepard Mark I irradiator. The mean survival time for Lxn-/- mice was 16 days compared to 13 days of wild-type (WT) mice in 8 Gy TBI. The mean survival time for Lxn-/- mice was 21 days compared to 9 days in 9 Gy TBI. Prism software was used to determine the mean survival time (N=5/group). (B) Lxn-/- mice had accelerated recovery in the numbers of blood leukocytes, platelets, and hematopoietic stem/progenitor cells (HSPC) post sublethal 6.5 Gy TBI. HSPC was identified as Lin-, Sca1+, c-kit+ cells (LSK cells). Data were collected on days 0, 1, 3, 7, 14, 28, and 56 days post-TBI (N=5/group at each time point). (C-E) WT and Lxn-/- mice were subject to 6.5 Gy TBI. BM cells were isolated from these mice at 56 days postTBI and evaluated for the HSC function by cobblestone area forming cell (CAFC) assay, limiting dilution competitive repopulation unit (CRU) assay, and competitive repopulation and serial transplantation. (C) More CAFC day 35 bone marrow (BM) cells were preserved in Lxn-/- mice after IR. BM cells were pooled from 5 mice in each group (N=5). Two independent CAFC assays were performed. (D) CRU frequency (left panel) and absolute number per femur (right panel) were significantly higher in Lxn-/- mice compared to WT mice. 2x104, 6x104, 2x105 or 6x105 donor BM cells were collected from Lxn-/- or WT (CD45.2) mice at day 56 after 6.5 Gy TBI, and mixed with 2x105 competitor BM cells (CD45.1) and retro-orbitally injected into lethally irradiated (9 Gy) recipient mice (CD45.1). Percentages of donor (CD45.2+)-derived peripheral blood (PB) cells were determined at 16 weeks post transplantation (N=10 recipient mice in each dilution). CRU frequency was calculated with LDA software. (E) In serial transplantation, Lxn-/HSC had a higher capacity to regenerate PB and BM LSK cells in secondary and tertiary recipients. 1x106 donor cells from Lxn-/or WT BM (CD45.2) mice were mixed with an equal number of competitor BM cells (CD45.1) and retro-orbitally injected into lethally irradiated (9 Gy) first recipient mice (CD45.1). At 16 weeks post-transplant, Lxn-/- or WT BM cells (CD45.2) were sorted from the primary recipients and transplanted into secondary recipients. An identical regimen was repeated one more time, culminating in the engrafted tertiary hosts. Percentages of CD45.2-derived PB and BM LSK cells were evaluated in each round of transplantation recipients (N=10 recipient mice at each time point). (F) Age-matched (8 weeks old) WT and Lxn-/- mice with or without the fractionated dose TBI (2 Gy daily for 5 days) were aged naturally for 16-20 months, and HSPC populations were analyzed for long-term effect. Representative flow cytometry analysis of HSPC subsets. Gating strategy for HSPC-enriched Lin-, Sca1+, c-kit+ (LSK) cells, long-term HSC (LT-HSC) (LSK, CD34-, Flt3-), short-term HSC (ST-HSC) (LSK, CD34+, Flt3-), and multipotent progenitor (MPP) (LSK, CD34+, Flt3+) is shown. Percentages of LSK cells, LT-HSC, ST-HSC, and MPP cells were significantly higher in Lxn-/- mice than in WT mice (N=5/ group). The statistical analysis was consulted with the Markey Cancer Center Biostatistics & Bioinformatics Shared Resource Facility. Data were examined for homogeneity of variances (F test), then analyzed by a two-tailed, unpaired Student’s t test. The log-rank (Mantel-Cox) test was used to assess survival curves. All statistical analyses were performed using GraphPad Prism Software version 7.0. Differences were considered significant at P<0.05. Results shown represent mean ± standard deviation; *P≤ 0.05, **P<0.01, ***P<0.001 and ****P<0.0001. Female mice were used at the age 7-8 weeks old. Mice were housed and handled at the University of Kentucky animal facilities following National Institutes of Healthmandated guidelines for animal welfare and with Institutional Animal Care and Use Committee approval number 2021-3881.

ing that survived Lxn-/- LSK cells maintained the genomic integrity. This may explain the absence of hematologic malignancies in Lxn-/- mice even 2 years after radiation (data not shown). Overall, all these short-term and long-term studies strongly suggest that Lxn deletion protects against both IR-induced acute myelosuppression and long-term HSC damage. We determined the signaling pathways involved in Lxn deletion-mediated radiation protection. We previously identified ribosome protein subunit 3 (Rps3) as a novel Lxn-binding protein in a myeloid cell line.9 Rps3 was reported to interact with the NF-kB p65 subunit and direct the complex to the promoter of some specific pro-survival genes upon IR, thus providing regulatory specificity.12,13 We performed immunofluorescence staining in LSK cells and co-immunoprecipitation (Co-IP) in Lin- cells, and confirmed the binding of Lxn and Rps3, and the interaction between Rps3 and p65 in primary HSPC (Figure 2A). Lxn deletion didn’t change Rps3 mRNA level before and after radiation (Online Supplementary Figure S2A). Rps3 itself is involved in ribosome assembly and protein synthesis. We asked whether Lxn deletion could affect protein synthesis in HSPC. We performed in vivo O-propargyl-puromycin (OP-Puro) incorporation assay and found similar OP-puro incorporation between different subsets of Lxn-/- and WT cells except for the CMP, indicating that Lxn deletion did

not affect overall protein synthesis in hematopoietic cells (Online Supplementary Figure S2B).14 We thus hypothesized that Lxn deletion releases Rps3 protein, which promotes the nuclear translocation of the NF-kB complex and stimulates prosurvival pathways upon radiation, enhancing HSC survival. We used immunofluorescence-conjugated Rps3 and p65 antibodies to detect their signal intensity in the nucleus of single LSK cells and found there were more Rps3 and NF-kB p65 detected in the nucleus of Lxn-/- LSK cells compared with WT LSK cells post-IR (Figure 2B), suggesting that Lxn deletion does enhance nuclear translocation of the Rps3-NF-kB complex. This result was further confirmed by western blot in less primitive Lin- cells (Online Supplementary Figure S2C). We further identified Bcl2 as one of the downstream target survival genes that were upregulated in Lxn-/- LSK cells (Figure 2C; Online Supplementary Figure S2D). Chromatin immunoprecipitation quantitative polymerase chain reaction (ChIP-qPCR) and electrophoretic mobility shift assay (EMSA) confirmed that Bcl-2 was the direct target of NF-kB p65 in Lxn-/- cells, and there was more binding in Lxn-/- cells compared with WT cells (Figure 2D, E). Consistently, we found that Lxn-/LSK cells were less apoptotic than WT cells at different time points after radiation (Figure 2F). All these data confirm that Lxn deletion promotes nuclear translocation of the Rps3-NF-kB complex upon IR and activates Bcl-2 tran-

Haematologica | 108 December 2023

3466

Continued on following page.

LETTER TO THE EDITOR

Haematologica | 108 December 2023

3467

LETTER TO THE EDITOR Figure 2. Latexin deletion upregulates RPS3-NF-kB-mediated transcriptional activation of Bcl-2 upon radiation and increases hematopoietic stem/progenitor cell survival. (A) Immunofluorescence staining shows the co-localization of LXN and RPS3, as well as RPS3 and NFKB p65 in a single LSK cell (top panel). Binding of LXN and RPS3, as well as RPS3 and NFKB p65 in Lin- cells was confirmed by reciprocal co-immunoprecipitation (CO-IP, bottom panel). (B) Lxn-/- LSK cells had increased nuclear translocation of RPS3 and NF-KB p65 after ionizing radiation (IR). Representative microimages of RPS3 and p65 staining in single LSK cell from Lxn-/- and wild-type (WT) mice with (6.5 Gy) or without IR (top panel). RPS3 (bottom left panel) and p65 (bottom right panel) staining intensity in the nuclear of LSK cells from Lxn-/- and WT mice without or with IR (6.5 Gy) were quantified and presented (N=50-70 LSK cells evaluated per group). (C) Increased Bcl-2 mRNA expression of LSK cells from Lxn-/- mice compared with WT mice at day 0, 7 and 28 after 6.5 Gy total body irradiation (TBI). Quantitative real-time poymerase chain reaction (PCR) was performed. mRNA was extracted from a pool of 5 mice/group. Two independent real-time PCR were performed and each experiment has 3 replicates. The data were pooled from 2 experiments. (D) Bcl-2 was a direct target of p65 and more p65 was enriched in the Bcl-2 promoter in Lxn-/- Lin- cells compared to WT cells. Chromatin immunoprecipitation assay was performed by p65 antibody and Bcl-2 promoter sequence was determined by real-time PCR. The gel image (top panel) and quantification (bottom panel) of real-time PCR were shown. (E) Electrophoresis mobility shift assay shows the direct binding of p65 to Bcl-2 promoter sequence, and more binding was present in Lxn-/- Lin- cells compared to WT cells at day 28 days post 6.5 Gy TBI. The quantification of the super shift band is labeled. (F) Percentage of apoptotic (Annexin V+ 7-AAD-) LSK cells was significantly lower in Lxn-/- mice than WT mice at day 0, 7, 14, and 28 days post 6.5 Gy TBI (top panel), consistent with the decreased expression of cleaved caspase 3 in LSK cells at day 28 (bottom panel) (N=5 mice/group); *P< 0.05, **P<0.01, ***P<0.001, and ****P< 0.0001.

Figure 3. Inhibition of Rps3-NFkB p65-Bcl-2 pathway sensitizes Lxn-/- LSK cells to radiation-induced apoptosis. (A) Knock-down of Rps3 increases apoptosis of wild-type (WT) and Lxn-/- LSK cells and diminishes their survival difference without or with ionizing radiation (IR) (left panel). Consistently, knockdown of Rps3 decreased p65, phosphorylated p65 (p-p65), and Bcl-2 expression (right panel). Flow cytometry-sorted LSK cells from WT or Lxn-/- mice were stimulated with cytokines including 100 ng/mL FMSlike tyrosine kinase-3 ligand, 50 ng/mL mouse stem cell factor, 20 ng/mL interleukin-3 (IL-3), and 20 ng/mL TPO in StemSpan SFEM (STEMCELL Technologies). After 24 hours (hrs), the cells were transduced with lentiviral particles encoding Rps3 small hairpin RNA (shRNA) (catalog no: MSH030789-LVRU6GP, GeneCopoeia) or its related scramble control vector, at a MOI of 100 along with 8 µg/mL polybrene for 6 hrs at 37°C. After 48 hrs, the GFP-positive cells were sorted for apoptosis assay and western blot. (B) NF-kB p65 specific inhibitor, JSH23, leads to the increased apoptosis of WT and Lxn-/- LSK cells and diminishes their apoptosis difference without or with IR (left panel). JSH23 also results in decreased Bcl-2 expression in both WT and Lxn-/- LSK Continued on following page. Haematologica | 108 December 2023

3468

LETTER TO THE EDITOR cells (right panel). The fold change of each protein expression was marked underneath. (C) Bcl-2 specific inhibitor, ABT-199, leads to the increased apoptosis of WT and Lxn-/- LSK cells and diminishes their apoptosis difference without or with IR (left panel). It also inhibits Bcl-2 expression in Lxn-/- LSK cells (right panel). The fold change of each protein expression was marked underneath. In (B and C), sorted WT or Lxn-/- LSK cells were cultured in StemSpan SFEM (STEMCELL Technologies) with cytokines including 100 ng/mL FMS-like tyrosine kinase-3 ligand, 50 ng/mL mouse stem cell factor, 20 ng/mL interleukin-3 (IL-3), and 20 ng/mL TPO. Immediately after 6.5 Gy irradiation, P65 inhibitor II, JSH23 (final concentration 7.1 µm, Millipore Sigma, catalog# 481408), or Bcl-2 inhibitor, ABT-199 (final concentration 100 µm, CHEMIETEK, catalog# CT-A199) or control (dimethyl sulfoxide [DMSO]), was added into the cells. After 6 hrs, cells were collected for apoptosis analysis or western blot. Apoptosis was measured by Annexin V+ 7AAD. Protein expression was measured by western blot. Two independent experiments were performed and each experiment has 3 replicates; **P<0.01, ***P<0.001, ****P<0.0001. (D) In vivo treatment of Lxn-/- mice by ABT-199 suppresses the survival advantage of radiation. New cohorts of WT and two groups of Lxn-/- mice were subject to 8 Gy TBI. One group of Lxn-/- mice was intraperitoneally injected with ABT-199 the day after total body irradiation (TBI) with a dose 100 mg/kg daily for 1 week. Mice survival was monitored and the mean survival time was calculated by using Prism software. The mean survival times for WT group and Lxn-/- group with ABT-199 post TBI were 23 days and 22 days respectively, whereas Lxn-/- group all survived during the 30 days of observation. The mice used in this experiment were 16-18 weeks old females (N=5/group); **P<0.01. (E) Model of the molecular mechanism for Lxn deletion-mediated radiation protection. Lxn binds to Rps3. Rps3 is an integral component of the NF-kB complex (p65/p50 dimer) by its specific interaction with the p65 subunit. In the absence of Lxn, Rps3 enhances the nuclear translocation of the NF-kB complex upon radiation and guides NF-kB complex specifically binding to the promoter of the prosurvival gene, Bcl-2, and increases its transcription and expression. This leads to the increased survival of Lxn-/- hematopoietic stem/progenitor cells upon radiation.

scription, thus enhancing the survival of Lxn-/- HSPC. We next genetically or pharmaceutically inhibited each key component of the Rps3- NF-kB-Bcl-2 pathway in Lxn-/LSK cells and determined whether it could blunt the survival advantage upon radiation. Annexin V+ apoptotic cells and Bcl-2 expression were used for functional and molecular evaluation. Rps3 knock-down increased apoptosis in both WT and Lxn-/- cells, abolished the survival advantage of Lxn-/- cells, and suppressed both NF-kB p65 nuclear translocation and Bcl-2 expression (Figure 3A). We next treated cells with the NF-kB specific inhibitor JSH23 and found that NF-kB inhibition also attenuated radiation protection and Bcl-2 activation in Lxn-/- cells (Figure 3B). Similar effects were observed with the Bcl-2 specific inhibitor ABT-199 (Figure 3C). In order to further confirm the relationship between Lxn and Bcl2 in vivo, we treated Lxn-/- mice with ABT-199 in vivo after lethal IR (8 Gy), and found that ABT-199 diminished the survival advantage of Lxn-/- mice after IR in comparison to the irradiated WT and Lxn-/- mice without treatment (Figure 3D). Altogether, these data suggest that in the absence of Lxn, HSC are protected from radiation-induced apoptosis via activation of a novel Rps3-NF-KB-Bcl-2 prosurvival pathway (Figure 3E). Lxn has thus a unique dual function in that it provides protection against both acute and long-term hematopoietic damages upon radiation. Lxn is a novel regulator of NF-kB signaling pathway via the specific interaction to Rps3 protein. Such Rps3-dependent activation of specific NF-kB target genes has been proposed as a novel strategy to selectively, rather than globally, manipulate NF-kB activity to reduce off-target side effects.12,13,15 In the future, an investigation of how Lxn is involved in this pathway will be warranted. Lxn might be a new target for developing novel radiomitigation compounds to minimize radiation-induced injury. Moreover, we have reported that Lxn-/- mice had better hematopoietic recovery from 5-FU-induced myelo-

suppression.8 Pharmacological inhibition of Lxn could be of clinical importance in improving outcomes for patients with radiation and chemotherapy.

Authors Cuiping Zhang,1 Xiaojing Cui,1 Yi Liu,2 Fang Wang,1 Robert Signer,3 Kapana Nattamai,4 Daohong Zhou,5 Yi Zheng,4 Hartmut Geiger,6 Fengyi Wan7 and Ying Liang1° Department of Toxicology and Cancer Biology University of Kentucky,

Lexington, KY, USA; 2Department of Physiology, University of Kentucky, Lexington, KY, USA; 3Division of Regenerative Medicine, Department of Medicine, Moores Cancer Center, University of California San Diego, La Jolla, CA, USA; 4Experimental Hematology and Cancer Biology, Cincinnati Children Hospital Medical Center, University of Cincinnati, OH, USA; 5Department of Biochemistry & Structural Biology, University of Texas Health Science Center, San Antonio, TX, USA; 6Institute of Molecular Medicine, Ulm University, Ulm, Germany and 7Department of Biochemistry and Molecular Biology, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD, USA °current address: New York Blood Center, New York, NY, USA Correspondence: Y. LIANG - yliang@nybc.org https://doi.org/10.3324/haematol.2022.282028 Received: September 8, 2022. Accepted: June 15, 2023. Early view: June 22, 2023. ©2023 Ferrata Storti Foundation

Haematologica | 108 December 2023

3469

LETTER TO THE EDITOR Office for editing and graphics support. We thank Dr. Sean Morrison

Published under a CC BY-NC license

for the critical insights into studies and comments on manuscripts. Disclosures Funding

The work is related to a patent US 10,604,756.

The authors are supported by the National Institutes of Health under awards R01HL124015(YL), R21HL140213 (to YL), R43AI145726 (to YL),

Contributions CZ performed the majority of experiments and wrote the manuscript. XC was involved in Lxn

-/-

mice maintenance. YLiu helped with blood

and R01CA211963 (to DZ) and the Markey Cancer Center’s Flow Cytometry and Immune Monitoring Core and Biostatistics and

and bone marrow cell monitoring. FW did the ABT-199 survival

Bioinformatics Shared Resource Facilities (P30CA177558).

experiment. RS helped with protein synthesis and review of the manuscript. DZ, YZ, KN, FW, and HG helped with the radiation strategy

Data-sharing statement

and phenotype characterization, and manuscript revision. YLiang

All data generated or analyzed during this study are included in this

guided the overall project, designed the experiments, and wrote the

published article (and its Online Supplementary Appendix). Data will be

manuscript.

available from the corresponding author upon reasonable request.

Acknowledgments We thank the Markey Cancer Center's Research Communications

References 1. Mauch P, Constine L, Greenberger J, et al. Hematopoietic stem cell compartment: acute and late effects of radiation therapy and chemotherapy. Int J Radiat Oncol Biol Phys. 1995;31(5):1319-1339. 2. Shao L, Luo Y, Zhou D. Hematopoietic stem cell injury induced by ionizing radiation. Antioxid Redox Signal. 2014;20(9):1447-1462. 3. Harbron RW, Feltbower RG, Glaser A, Lilley J, Pearce MS. Secondary malignant neoplasms following radiotherapy for primary cancer in children and young adults. Pediatr Hematol Oncol. 2014;31(3):259-267. 4. Burdelya LG, Krivokrysenko VI, Tallant TC, et al. An agonist of toll-like receptor 5 has radioprotective activity in mouse and primate models. Science. 2008;320(5873):226-230. 5. Johnson SM, Torrice CD, Bell JF, et al. Mitigation of hematologic radiation toxicity in mice through pharmacological quiescence induced by CDK4/6 inhibition. J Clin Invest. 2010;120(7):2528-2536. 6. Geiger H, Pawar SA, Kerschen EJ, et al. Pharmacological targeting of the thrombomodulin-activated protein C pathway mitigates radiation toxicity. Nat Med. 2012;18(7):1123-1129. 7. Chute JP, Muramoto GG, Salter AB, et al. Transplantation of vascular endothelial cells mediates the hematopoietic recovery and survival of lethally irradiated mice. Blood. 2007;109(6):2365-2372. 8. Liu Y, Zhang C, Li Z, et al. Latexin Inactivation Enhances survival and long-term engraftment of hematopoietic stem cells and expands the entire hematopoietic system in mice. Stem Cell

Rep. 2017;8(4):991-1004. 9. You Y, Wen R, Pathak R, et al. Latexin sensitizes leukemogenic cells to gamma-irradiation-induced cell-cycle arrest and cell death through Rps3 pathway. Cell Death Dis. 2014;5(10):e1493. 10. Liang Y, Jansen M, Aronow B, Geiger H, Van Zant G. The quantitative trait gene latexin influences the size of the hematopoietic stem cell population in mice. Nat Genet. 2007;39(2):178-188. 11. Chua HL, Plett PA, Fisher A, et al. Lifelong residual bone marrow damage in murine survivors of the hematopoietic acute radiation syndrome (H-ARS): a compilation of studies comprising the Indiana University Experience. Health Phys. 2019;116(4):546-557. 12. Wan F, Anderson DE, Barnitz RA, et al. Ribosomal protein S3: a KH domain subunit in NF-kappaB complexes that mediates selective gene regulation. Cell. 2007;131(5):927-939. 13. Wan F, Weaver A, Gao X, et al. IKKbeta phosphorylation regulates RPS3 nuclear translocation and NF-kappaB function during infection with Escherichia coli strain O157:H7. Nat Immunol. 2011;12(4):335-343. 14. Signer RA, Magee JA, Salic A, Morrison SJ. Haematopoietic stem cells require a highly regulated protein synthesis rate. Nature. 2014;509(7498):49-54. 15. Wier EM, Neighoff J, Sun X, Fu K, Wan F. Identification of an Nterminal truncation of the NF-kappaB p65 subunit that specifically modulates ribosomal protein S3-dependent NFkappaB gene expression. J Biol Chem. 2012;287(51):43019-43029.

Haematologica | 108 December 2023

3470

LETTER TO THE EDITOR

ETV6 fusions from insertions of exons 3-5 in pediatric hematologic malignancies Current risk classification and treatment of patients with B-cell lymphoblastic leukemia (B-ALL) relies on rapid genomic testing for the identification of subtype-defining, prognostically significant, or potentially targetable alterations, which often occur as fusion events.1,2 Karyotype and fluorescence in situ hybridization (FISH) are the traditional methodologies used for detecting fusions; however, there is increasing recognition of false positive and false negative results with these tests depending on the structure of the rearrangement. Although FISH has greater resolution than karyotype, it is still limited to targeted aberrations of approximately 100 kb or larger, thus clinically important fusions with false negative FISH results

due to relatively small insertions have been observed. ETV6::RUNX1 fusions define one of the most common pediatric B-ALL subtypes comprising 20-25% of cases.3 Although not universally favorable, its status is used in some treatment protocols to classify patients as provisionally low risk.4,5 These fusions typically result from chromosomal translocations t(12;21)(p13;q22) that are cryptic to karyotyping due to banding pattern similarities of the 12p and 21q chromosomal arms.5 As a result, FISH or reverse transcription polymerase chain reaction (RTPCR) is usually performed for diagnosis. Here we report two pediatric B-ALL cases that were negative for ETV6::RUNX1 rearrangements by FISH, but were deduced

Table 1. Patient characteristics and molecular findings. Case 1

Case 2

Case 3

Age in years

Sex

Diagnosis

B-ALL

AML

BMA blast %

CNS involvement

46,XY[20]

Failed

47,XX,del(2)(p16p22), t(3;19)(p21;p13.3),del(9)(q21q31), del(12)(q24),+21[19] / 46,XX[1]

Clinical FISH for ETV6::RUNX1

Negative 1 copy ETV6 in 68.5%

Negative

Fusions (RNA)

RUNX1::ETV6 (e2 to e3) ETV6::RUNX1 (e5 to e3)

EP300::ETV6 (e1 to e3) ETV6::EP300 (e5 to e2)

1. (chr21:36421139-, chr12:11992074+) 2. (chr12:12022903+, chr21:36265260-)

1. (chr22:41489102+, chr12:11992074+) 2. (chr12:12022903+, chr22:41513191+)

Isoforms (RNA)

ETV6 (e2 to e6) IKZF1 (e1 to e8)

ETV6 (e2 to e6)

ETV6 del (DNA)

1 copy del, whole gene

IKZF1 del (DNA)

1 copy del, exons 2-7

NFE2 p.Y172H (VUS)

1 copy del SETD2 (3p) 1 copy del CUX1 (7q) Gain RUNX1, ERG, U2AF1 (21q)

Karyotype

Breakpoints (RNA)

Other molecular findings (DNA)

KRAS p.F156L KRAS p.L19F NRAS p.G12D SF3B1 p.E595K

Details of case 4 (Figure 1C) can be found in the Supplementary Table of the total RNA sequencing study (sample ID: 36).8 The case was from a 4 year-old male with B-ALL with 89% blasts, a karyotype of 45,XY,-9,der(12)t(9;12)(q21.11;p11.22),der(18)t(17;18)(q21.31;q21.2)[8], positive FISH for ETV6::RUNX1 with an unusual pattern of signals (described in the main text), RNA sequencing fusions of RUNX1::ETV6 (e2 to e3) and ETV6::RUNX1 (e5 to e3) (by re-analysis of the raw data), outlier expression of the isoform ETV6 e2 to e6 (by isoform analysis of the raw data), and 1 copy whole-gene deletions of ETV6, KRAS, JAK2, and PAX5 as well as a VUS in IKZF2 p.N35S (by review of the report from targeted DNA next generation sequencing testing, also performed at Brigham and Women’s Hospital). AML: acute myeloid leukemia; B-ALL: B-lymphoblastic leukemia; BMA: bone marrow aspirate; CNS: central nervous system; e: exon; F: female; M: male; del: deletion; VUS: variant of unknown significance; FISH: fluorescence in situ hybridization. Haematologica | 108 December 2023

3471

LETTER TO THE EDITOR through targeted RNA next generation sequencing (NGS) to harbor ETV6::RUNX1 fusions characterized by focal insertions of ETV6 exons 3-5 into RUNX1 with concomitant intragenic deletions of the same exons from ETV6. We further report one pediatric AML case with an ETV6::EP300 rearrangement similarly involving insertion and deletion of ETV6 exons 3-5. Finally, by searching public whole-transcriptome sequencing data, we identify another ETV6::RUNX1-positive B-ALL with the same pattern and characterize the underlying insertion and deletion breakpoints on the DNA level.

Cases 1 and 2 were identified within 2 weeks of each other during routine review at Boston Children’s Hospital of a clinically validated targeted RNA NGS assay for detecting fusions based on ArcherDx FusionPlex Heme v2 (FPHeme; IDT, Coralville, IA, USA). Cytogenetics and FISH were performed at Integrated Oncology or partner institutions. Targeted DNA NGS was performed at Brigham and Women’s Hospital. Isoform analysis of RNA sequencing data was performed using the software isoformSR (https://github.com/ht50/isoformSR), as previously described.6 Clinical features and molecular findings are summarized in Table 1.

Continued on following page. Haematologica | 108 December 2023

3472

LETTER TO THE EDITOR

Figure 1. RNA sequencing results. (A) ETV6 exon skipping isoform and ETV6 fusion transcripts detected on targeted RNA sequencing (FPHeme) over historical cohort (N=474) consisting of cases 1-3 (black square, circle, and triangle) and others (gray x’s). Cases 1-3 had outlier expression of the ETV6 isoform junction e2e6 connecting exon 2 to exon 6, consistent with an underlying intragenic deletion of the skipped exons 3-5, as well as fusion breakpoints consistent with insertion of exons 3-5 between exons 2 and 3 of the partner gene RUNX1 (cases 1 and 2) or between exons 1 and 2 of the partner gene EP300 (case 3). Case 1 was also associated with single copy whole gene deletion of ETV6 and accordingly had the highest expressed variant allele frequency (VAF) of e2e6. (B) Clustering by t-distributed stochastic neighbor embedding (t-SNE) of the targeted transcriptional profiles (100 genes) of cases 1 and 2 among all B-cell lymphoblastic leukemia (B-ALL) cases (N=209) colored by subtype including ETV6::RUNX1 fusions (green dots). (C) Screening of public total RNA sequencing data identified one more pediatric ETV6::RUNX1-positive B-ALL case (sample ID: 36; SRR15301257) with ETV6 e2e6 at high VAF (95.3%; not shown), fusion breakpoints consistent with insertion of ETV6 exons 3-5 between RUNX1 exons 2 and 3, and single copy whole gene deletion of ETV6 by targeted DNA next generation sequencing. Split-read analysis of intronic alignments identified 3 sets of expressed DNA breakpoints within split-reads: (i) chr12:11952219+ (ETV6 intron 2) to chr12:12032918+ (ETV6 intron 5) illustrated in the example paired split-read SR1 (light blue), (ii) chr21:36296008-(RUNX1 intron 2) to chr12:11951870+ (ETV6 intron 2) illustrated in the example paired split-read SR2 (yellow), and (iii) chr12:12032916+ (ETV6 intron 5) to chr21:36296069- (RUNX1 intron 2) illustrated in the example paired split-read SR3 (gray). The breakpoints of SR1 implied a deletion of the intervening 80.7 kb segment between the ETV6 intron 2 and intron 5 breakpoints. The breakpoints of SR2 and SR3 implied a focal insertion of the near-exact 81 kb segment of ETV6 (chr12:11951870+ to chr12:12032916+) into RUNX1 intron 2 at breakpoints separated by 62 bp (chr21:36296008- and chr21:36296069-), thus also associated with a duplication of this short 62 bp segment. The inclusion of an additional ~350 bp from intron 2 in the inserted segment compared to the deleted segment suggested some amount of repair/replication within intron 2 during formation of the deletion. In theory, the breakpoints of SR2 and SR3 alternatively could represent a reciprocal translocation with duplication of ETV6 exons 3-5, however single copy loss of the other ETV6 allele made it impossible for the remaining allele to harbor both a reciprocal translocation and a deletion of the same region. The intronic splitreads harboring DNA deletion and rearrangement breakpoints were presumably derived from pre-mRNA, as described in other studies.6 Notation - VAF: expressed variant allele fraction. 5’ and 3’: refers to the 5’ and 3’ ends of the split-read alignments. +/refers to the DNA strand of the alignment; note that ETV6 is transcribed from the plus strand (+) of chromosome 12 whereas RUNX1 is transcribed from the minus strand (-) of chromosome 21.

Case 1 was from a 4-year-old male with newly-diagnosed B-ALL with unrevealing genetic analysis, including normal karyotype and negative FISH for ETV6::RUNX1 and other subtype-defining rearrangements. FPHeme subsequently revealed an ETV6::RUNX1 fusion with standard breakpoints connecting ETV6 exon 5 (NM_001987.5) to RUNX1 exon 3 (NM_001754.5), together with an atypical reciprocal RUNX1::ETV6 fusion with rare breakpoints connecting RUNX1 exon 2 to ETV6 exon 3 and abnormally high expression (quantified by split-reads sequencing the junction) of an in-frame ETV6 exon-skipping isoform junction connecting exon 2 to exon 6 (i.e., skipping exons 3-5), satisfying previously established criteria for “outlier expression” relative to other FPHeme cases (Figure 1A). We have previously

shown outlier expression to be a sensitive and specific marker for underlying intragenic deletions in other genes, thus this case was most consistent with DNA deletion of ETV6 exons 3-5.6 Despite its unusual aspects, the targeted gene expression profile clustered with B-ALL cases harboring typical FISH-positive ETV6::RUNX1 fusions (Figure 1B). Theoretically, the observed fusion breakpoints could represent either a reciprocal translocation or a focal insertion; however, FISH analysis showed loss of one ETV6 signal in 68.5% of cells, similar to the 66% blast estimate by flow cytometry and confirmed to represent a single-copy, whole-gene ETV6 deletion by targeted DNA sequencing, making it impossible for the single remaining ETV6 allele to harbor both deletion of ETV6 exons 3-5 and a reciprocal

Haematologica | 108 December 2023

3473

LETTER TO THE EDITOR translocation to RUNX1. The overall findings, therefore, implied an intragenic deletion of ETV6 exons 3-5 with associated insertion between RUNX1 exons 2 and 3 (Figure 2). Case 2 was from a 6-year-old female with newly diagnosed B-ALL and similar findings of a normal karyotype and negative FISH for ETV6::RUNX1 and other fusions. FPHeme again revealed ETV6::RUNX1 and RUNX1::ETV6 fusions with the same exon structure as case 1, outlier expression of the same ETV6 exon-skipping isoform, and a gene expression profile clustering with typical cases of ETV6::RUNX1 B-ALL. Given this recurrent pattern of insertions and deletions of intragenic exons, we retrospectively searched historical FPHeme cases (n=474) for outlier expression of any ETV6 exon-skipping isoform as a potential marker for cryptic insertions. We identified one additional case with outlier expression involving skipping of ETV6 exons 3-5 (case 3), but did not identify outlier expression involving skipping of other exons. Case 3 was from a 5-year-old female with relapsed acute myeloid leukemia (AML) with detected in-frame reciprocal fusions ETV6::EP300 connecting exon 5 of ETV6 to exon 2 of EP300 (NM_001429.4) and EP300::ETV6 connecting exon 1 of EP300 to exon 3 of ETV6 (Figure 1A). The predicted chimeric protein contained the helix-loop-helix (HLH) domain of ETV6 inserted between the nuclear localization se-

quence and transactivation domains of EP300 and retained the EP300 chromatin modification region (Figure 2). Outlier expression of the ETV6 isoform skipping exons 3-5 was again observed, consistent with single-copy intragenic deletion (Figure 1A). Per report, interphase FISH analysis showed two strong ETV6 signals and one weak ETV6 signal. Metaphase FISH analysis showed that the weak ETV6 signal was located on a small G-size chromosome. EP300 FISH was not performed. Based on these data, we hypothesize that the weak ETV6 signal represented partial binding of the ETV6 probe to ETV6 exons 3-5 inserted into the EP300 gene on chromosome 22. Another case of ETV6::EP300, with unspecified structure, has been reported in the literature.7 Finally, we analyzed ETV6 exon-skipping isoforms in public total RNA sequencing data from a pediatric ALL cohort, revealing one ETV6::RUNX1 case (sample ID: 36) with outlier expression of the isoform skipping exons 3-5, comprising 3.4% (1/29) of ETV6::RUNX1-positive B-ALL in the cohort, where it again co-occurred with the only instance of a reciprocal fusion connecting RUNX1 exon 2 to ETV6 exon 3.6,8 Targeted DNA NGS showed single copy loss of ETV6, thus the exon-skipping isoform was expressed at a high variant allele fraction (95.3%; data not shown), similar to case 1 (Figure 1A). Intronic split-read analysis identified three sets of expressed DNA breakpoints derived presumably from

Figure 2. Schematics of ETV6 insertions. Structure of the rearrangements detected in cases 1 and 2 (left side) and case 3 (right side) with the predicted chimeric protein products. Haematologica | 108 December 2023

3474

LETTER TO THE EDITOR pre-mRNA, compatible with deletion of 80.7 kb between ETV6 introns 2 and 5 (chr12:11,952,220 to chr12:12,032,917) and insertion of a near-exact 81.0 kb segment (chr12:11,951,870 to chr12:12,032,916) into RUNX1 intron 2 at breakpoints separated by 62 bp (chr21:36,296,008 and chr21:36,296,069) (Figure 1C). Per report, FISH analysis described a non-classical ETV6::RUNX1 rearrangement, where the translocated 5’ part of ETV6 to the derived 21 demonstrated a smaller green signal than usual, together with an absence of the small extra red signal that ordinarily represents the 5’ part of RUNX1 translocated to the derived 12. The FISH positivity of this case, in contrast to cases 1-2, might be related to differences in FISH probes or ETV6 insertion sizes, which theoretically could be as small as ~30 kb versus ~81 kb in this case. An ETV6::RUNX1 fusion cryptic to FISH but detected by RT-PCR, which was designed to amplify any fusion connecting ETV6 exon 5 to RUNX1 exon 3 or exon 4 (including the insertion fusions described here), was previously reported in a pediatric B-ALL, although underlying genomic structure was not determined.9 Importantly, in the absence of RNA sequencing or RT-PCR, such FISH-negative cases would likely remain unclassified and, in some clinical protocols, may lead to unintended higher risk stratification and more intensive treatment regimens.3,4 Similarly, although uncommon in pediatric AML, ETV6 rearrangements are important to identify given their association with adverse risk regardless of fusion partner; chromosome 12p abnormalities/ETV6 rearrangements are accordingly an indication in pediatric AML for allogeneic hematopoietic stem cell transplantation in first remission in the current Children’s Oncology Group AAML1831 clinical trial (clinicaltrials gov. Identifier: NCT04293562).7,10 The mechanistic consequences of insertion events will require elucidation. The gene expression profiles of cases 1-2 were similar to typical cases of ETV6::RUNX1 positive B-ALL in our historical cohort, suggesting functional similarity. An ETV6::RUNX1like gene expression profile, which may also confer a favorable prognosis, has been described as a provisional entity in otherwise unclassified B-ALL and has been associated with inactivating ETV6 variants and shown to be mediated by microsatellite enhancers ordinarily bound by ETV6.2,11 Detection and screening for exon-skipping isoforms may be an effective way of recognizing fusions resulting from concurrent deletion and insertion of intragenic genomic material, which otherwise may be mistaken for typical balanced reciprocal translocations in standard short read sequencing data. Indeed, targeted RNA sequencing of case 3 at a partner institution reported the ETV6::EP300 fusion as connecting exons 1-5 of ETV6 to exons 2-31 of EP300. The deletion-insertion pattern has also been described in the context of YAP1::KMT2A rearranged sarcomas although screening

our cohort did not reveal outlier expression of any KMT2A exon-skipping isoforms in either KMT2A-rearranged or non-rearranged hematologic cases.12 In summary, clinically relevant ETV6 fusions from focal insertions of ETV6 may be more common in childhood leukemia than previously recognized, particularly in FISHnegative cases. Larger studies are necessary to determine their true frequency, assess clinical significance, and inform the use of NGS fusion testing in B-ALL.

Authors Sarah B. Mueller,1 Yana Pikman,2,3 Sarah K. Tasian,4,5 Lewis B. Silverman,2,3 Marian H. Harris1 and Harrison K. Tsai1,6 Department of Pathology, Boston Children’s Hospital, Harvard

Medical School, Boston, MA; 2Department of Pediatric Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA; Division of Hematology/Oncology, Boston Children’s Hospital,

Harvard Medical School, Boston, MA; 4Division of Oncology and Center for Childhood Cancer Research, Children’s Hospital of Philadelphia, Philadelphia, PA; 5Department of Pediatrics and Abramson Cancer Center, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA and 6Department of Pathology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA Correspondence: H. K. TSAI - Harrison.Tsai@childrens.harvard.edu https://doi.org/10.3324/haematol.2022.282498 Received: January 19, 2023. Accepted: June 19, 2023. Early view: June 29, 2023. ©2023 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures No conflicts of interest to disclose. Contributions SM and HKT designed the study, performed data analysis, and wrote the manuscript. SKT, YP, LBS, and MHH contributed data. All authors edited the manuscript. Acknowledgments SKT is a scholar of the Leukemia and Lymphoma Society and holds the Joshua Kahan Endowed Chair in Pediatric Leukemia Research. This work includes two cases from clinical trials (clinicaltrails gov. Identifier: NCT02670525 and NCT03020030).

Haematologica | 108 December 2023

3475

LETTER TO THE EDITOR Funding

Data-sharing statement

This work was supported by St. Baldrick’s Foundation Consortium

Select data is available upon request to the corresponding author.

and the Charles H. Hood Foundation grants (YP).

References 1. Alaggio R, Amador C, Anagnostopoulos I, et al. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: Lymphoid Neoplasms. Leukemia. 2022;36(7):1720-1748. 2. Arber DA, Orazi A, Hasserjian RP, et al. International Consensus Classification of Myeloid Neoplasms and Acute Leukemias: integrating morphologic, clinical, and genomic data. Blood. 2022;140(11):1200-1228. 3. Pui CH, Campana D, Pei D, et al. Treating childhood acute lymphoblastic leukemia without cranial irradiation. N Engl J Med. 2009;360(26):2730-2741. 4. Mattano LA Jr, Devidas M, Maloney KW, et al. Favorable trisomies and ETV6-RUNX1 predict cure in low-risk B-cell acute lymphoblastic leukemia: results from Children's Oncology Group trial AALL0331. J Clin Oncol. 2021;39(14):1540-1552. 5. Shurtleff SA, Buijs A, Behm FG, et al. TEL/AML1 fusion resulting from a cryptic t(12;21) is the most common genetic lesion in pediatric ALL and defines a subgroup of patients with an excellent prognosis. Leukemia. 1995;9(12):1985-1989. 6. Tsai HK, Gogakos T, Lip V, et al. Outlier expression of isoforms by targeted or total RNA sequencing as clinical markers of genomic variants in B lymphoblastic leukemia and other tumor types. medRxiv. 2022 July 29.

doi:https://doi.org/10.1101/2022.07.29.22278149 [preprint, not peer-reviewed]. 7. Smith JL, Ries RE, Wang Y-C, et al. ETS family transcription factor fusions in childhood AML: distinct expression networks and clinical mplications. Blood. 2021;138(Suppl 1):S2356. 8. Tran TH, Langlois S, Meloche C, et al. Whole-transcriptome analysis in acute lymphoblastic leukemia: a report from the DFCI ALL Consortium Protocol 16-001. Blood Adv. 2022;6(4):1329-1341. 9. Hahm C, Han SH, Mun YC, et al. ETV6/RUNX1 rearrangement identified by RT-PCR without evidence on FISH. Acta Haematol. 2014;132(1):122–124. 10. Lamble AJ, Tasian SK. Opportunities for immunotherapy in childhood acute myeloid leukemia. Blood Adv. 2019;3(22):3750-3758. 11. Kodgule R, Goldman JW, Monovich AC, et al. ETV6 Deficiency unlocks ERG-dependent microsatellite enhancers to drive aberrant gene activation in B-lymphoblastic leukemia. Blood Cancer Discov. 2023;4(1):34-53. 12. Massoth LR, Hung YP, Nardi V, et al. Pan-sarcoma genomic analysis of KMT2A rearrangements reveals distinct subtypes defined by YAP1-KMT2A-YAP1 and VIM-KMT2A fusions. Mod Pathol. 2020;33(11):2307-2317.

Haematologica | 108 December 2023

3476

LETTER TO THE EDITOR

Modified carfilzomib dosing is associated with improved treatment responses and longer time on treatment in patients with multiple myeloma Carfilzomib is a second-generation proteasome inhibitor (PI) approved for the treatment of patients with multiple myeloma (MM). Compared to bortezomib and ixazomib, carfilzomib is associated with higher rates of cardiovascular (CV) adverse events (AE).1 For example, compared to bortezomib-lenalidomide-dexamethasone (VRd) in the newly diagnosed setting, carfilzomib-lenalidomide-dexamethasone (KRd) was associated with three-times more frequent grade ≥3 cardiac disorders (2.1% vs. 6.3%, respectively).2 Similarly, in the relapsed/refractory setting, grade ≥3 cardiac failure rates for patients treated with bortezomib-dexamethasone (Vd) were 1.8% compared to 4.8% in the Kd arm.3 One proposed mechanism of carfilzomib-induced cardiotoxicity appears to be through the inhibition of the AMPKα/mTORC1 pathways via increased PP2A activity, with downstream effects of autophagy downregulation in the myocardium.4 Higher doses of carfilzomib may result in decreased phosphorylation and activation of the PI3K/AKT/eNOS pathway. Multi-omics analyses have also revealed the importance of the glutamate-dependent acid resistance and pyruvate oxidation pathways in patients that experience CV AE.5 Finally, clinicians inexperienced with carfilzomib may overhydrate patients with normal saline during drug administration that may contribute to CV toxicity.6 We conducted a retrospective analysis to assess incidence of CV AE and outcomes of MM patients treated with carfilzomib using a step-up titration dosing (TD) schedule (e.g., step-wise C1D1 20 mg/m2, C1D8 27 mg/m2, C1D15 36 mg/m2, for goal C2D1 70 mg/m2 onwards) compared to standard dosing (SD) schedule (e.g., C1D1 20mg/m2 and C1D8 70mg/m2 onwards) (other examples of dosing schedules are given in the Online Supplementary Table S1). We included adult MM patients treated with more than one cycle of carfilzomib at the Ohio State University (OSU) from January 1, 2013 to September 1, 2019. Patients were excluded if they were not carfilzomib-naïve when receiving treatment institutionally, received a single dose of carfilzomib while inpatient with no intent to continue treatment outpatient, or had a treatment plan entered but never received it. Patient demographics, disease and carfilzomib-related characteristics, CV AE, and follow-up information were collected. Of the 166 patient charts analyzed, 36 were treated using a TD method and 130 were treated using a SD method.

Baseline demographics and disease characteristics are described in the Online Supplementary Table S2. The median age at diagnosis of myeloma was 60 years old and the majority of patients were Caucasian males. There were no differences in International Staging System (ISS) and Revised-ISS scores between the two groups. There was similar use of autologous hematopoietic cell transplantation, immunomodulatory drugs (IMiD), anti-CD38 monoclonal antibodies, and cyclophosphamide prior to the use of carfilzomib in both groups. There was a slight difference in baseline renal function (median creatinine clearance [CrCl]: SD 76 mL/min vs. TD 94mL/min; P=0.01) and prior PI use (SD 95.4% vs. TD 86.1%; P=0.048) between the two groups. Pre-existing CV risk factors were similar between the two groups; there was similar incidence of pre-existing hypertension (HTN; n=104; SD 61.5% vs. TD 66.7%), congestive heart failure (CHF; n=11; SD 6.9% vs. TD 5.6%), ischemic heart disease (IHD; n=17; SD 11.5% vs. TD 5.6%), arrhythmia (n=27; SD 16.2% vs. 16.7%), prior anthracycline exposure (n=16; SD 10.0% vs. TD 8.3%), and prior chest radiation (n=18; SD 10.0% vs. TD 13.9%). The majority of patients were treated with carfilzomib in the relapsed/refractory setting (SD 98.5% vs. TD 91.7%) compared to the newly-diagnosed setting (SD 1.5% vs. TD 8.3%). Most patients were treated with carfilzomib using the twice-weekly dosing schedule (SD 83.1% vs. TD 75.0%) compared to once-weekly dosing schedule (SD 16.9% vs. TD 25.0%). Carfilzomib was most often combined with an IMiD (n=84, 50.6%; SD 70% vs. TD 30%), with cyclophosphamide in 18.1% (n=30; SD 97% vs. TD 3%), and with another agent in 1.8% of patients (n=3; SD 67% vs. TD 33%). Carfilzomib and dexamethasone doublet was used in 29.5% of patients (n=49; SD 82% vs. TD 18%). Additional carfilzomib therapy details can be found in the Online Supplementary Table S3. The provider treatment selection reasoning was not explicitly documented in the charts reviewed. Furthermore, information regarding fluid volumes infused during carfilzomib treatment were not able to be collected. There were no differences in incidence of HTN (n=137; SD 83.1% vs. TD 80.6%), CHF (n=17; SD 10.8% vs. TD 8.3%), IHD (n=20; SD 13.1% vs. 8.3%), arrythmias (n=28; SD 16.9% vs. 16.7%), or pulmonary HTN (n=5; SD 3.1% vs. TD 2.8%) between the different dosing schedules. Approximately 80% of patients had documentation of HTN while on carfilzomib, however, the onset of HTN was delayed in the TD

Haematologica | 108 December 2023

3477

LETTER TO THE EDITOR group compared to the SD group. In the SD group, HTN developed earlier within the first five cycles (SD 70.3% vs. TD 46.7%) compared to sixth cycle and beyond in the TD group (SD 1.8% vs. TD 6.7%; P=0.02). Consistent with published reports, patients with pre-existing CV risk factors were more likely to develop CV toxicities associated with carfilzomib use (HTN, P=0.005; CHF, P=0.04; IHD, P=0.004; arrhythmia, P<0.001). Furthermore, there was a lower incidence of dyspnea (not included in the definition of CV AE) in the TD group compared to SD group (SD 36.2% vs. TD 16.7%; P=0.03). There was no statistical difference in CV AE being the cause for treatment discontinuation between the two groups. The median number of cycles administered to titrated patients was seven (range, 2-56) compared to three (range, 1-44) for standard patients (P<0.001) (Online Supplementary Table S3). The cumulative carfilzomib dose adminis-

tered to patients with TD was 1,669 mg compared to 593 mg in the SD group (P<0.001). There was a trend towards higher objective response rate in TD group compared to the SD group (SD 42.8% vs. TD 63.9%; P=0.09). Five-year OS was improved in patients treated with TD than with SD (SD 17.5% vs. TD 43.8%; P<0.001; Figure 1). In the univariable analysis on association between patient treatment characteristics and overall survival (OS), TD (hazard ratio [HR] =0.43; 95% confidence interval [CI]: 0.27-0.70; P=0.001) and carfilzomib being part of triplet therapy (HR=0.64; 95% CI: 0.44-0.93; P=0.019) were associated with improved OS (Table 1). In multivariable analysis controlling for age, disease status, and pre-existing CV risk, TD (HR=0.46; 95% CI: 0.28-0.74; P=0.002) and carfilzomib being part triplet therapy (HR=0.59; 95% CI: 0.41-0.86; P=0.007) retained significance for association with improved OS.

Figure 1. 5-year overall survival for multiple myeloma patients treated with carfilzomib. OS: overall survival; TD: titration dosing; SD: standard dosing.

Table 1. Univariable and multivariable analyses of factors associated with overall survival. Factor

Univariable analysis HR

95% CI

Multivariable analysis

Factor

95% CI

Titration vs. standard

0.43

0.27

0.70

0.001

Titration vs. standard

0.46

0.28

0.74

0.002

Age at diagnosis

1.03

1.01

1.05

0.007

Age at diagnosis

1.03

1.01

1.05

0.008

CrCl at start of therapy

0.99

0.112

Multiple vs. doublet

0.64

0.44

0.93

0.019

Multiple vs. doublet

0.59

0.41

0.86

0.007

Relapsed vs. frontline

5.75

0.8

41.2

0.081

Relapsed vs. frontline

4.3

0.59

31.25

0.15

Prior PI

1.47

0.69

3.16

0.321

Pre-existing CV risk

1.42

0.94

2.15

0.093

1.3

0.86

1.97

0.22

Pre-existing CV risk

Factors with P<0.10 in the univariable model were included in the multivariable model. HR: hazard ratio; CI: confidence interval; CrCl: creatinin clearance; PI: proteasome inhibitor; CV: cardiovascular risk. Haematologica | 108 December 2023

3478

LETTER TO THE EDITOR Overall, a simple modification of carfilzomib titration dosing resulted in significantly reduced perception of dyspnea with carfilzomib-based therapy and did not compromise efficacy. The incidence of HTN while on carfilzomib therapy was similar between the two groups although the patients treated with TD developed HTN later. However, an improvement in OS was observed in the TD group due to prolonged duration of therapy, including both the cumulative carfilzomib dose administered and cycles of therapy. As there were no differences in CV AE being the cause of treatment discontinuation between the two dosing schedules, future pharmacokinetic studies may identify the mechanism by which carfilzomib TD results in improved clinical outcomes compared to SD. Limitations of the study include its retrospective design, different carfilzomib target doses in the various treatment combinations, and choice of dosing schedule potentially being influenced by the aggressiveness of disease. In a large analysis of carfilzomib treated patients, titration dosing allows more time on therapy, similar efficacy, and improved 5-year OS.

Division of Hematology, Department of Internal Medicine, The Ohio State University, Columbus, OH, USA Correspondence: A. KHAN - Abdullah.Khan@osumc.edu https://doi.org/10.3324/haematol.2022.282521 Received: December 2, 2022. Accepted: June 30, 2023. Early view: July 13, 2023. ©2023 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures AMK and NB are part of Amgen Speakers Bureau. All other authors have no conflicts of interest to disclose. Contributions KD and AMK collected data. QZ and AMK performed data analysis. AMK, QZ, and AR drafted the manuscript. All authors reviewed the manuscript.

Authors

Data-sharing statement Data are available from the corresponding author upon reasonable request.

Abdullah M. Khan, Kiatlyn Dvorak, Qiuhong Zhao, Naresh Bumma, Francesca Cottini, Srinivas Devarakonda, Elvira Umyarova, Nidhi Sharma, Don Benson and Ashley Rosko

References 1. Das A, Dasgupta S, Gong Y, et al. Cardiotoxicity as an adverse effect of immunomodulatory drugs and proteasome inhibitors in multiple myeloma: a network meta-analysis of randomized clinical trials. Hematol Oncol. 2022;40(2):233-242. 2. Kumar SK, Jacobus SJ, Cohen AD, et al. Carfilzomib or bortezomib in combination with lenalidomide and dexamethasone for patients with newly diagnosed multiple myeloma without intention for immediate autologous stem-cell transplantation (ENDURANCE): a multicentre, open-label, phase 3, randomised, controlled trial. Lancet Oncol. 2020;21(10):1317-1330. 3. Dimopoulos MA, Moreau P, Palumbo A, et al. Carfilzomib and dexamethasone versus bortezomib and dexamethasone for patients with relapsed or refractory multiple myeloma

(ENDEAVOR): a randomised, phase 3, open-label, multicentre study. Lancet Oncol. 2016;17(1):27-38. 4. Efentakis P, Kremastiotis G, Varela A, et al. Molecular mechanisms of carfilzomib-induced cardiotoxicity in mice and the emerging cardioprotective role of metformin. Blood. 2019;133(7):710-723. 5. Tantawy M, Chekka LM, Huang Y. et al. Lactate dehydrogenase B and pyruvate oxidation pathway associated with carfilzomibrelated cardiotoxicity in multiple myeloma patients: result of a multi-omics integrative analysis. Front Cardiovasc Med. 2021;8:645122. 6. Landgren O, Siegel D, Kazandjian D, et al. Treatments for newly diagnosed multiple myeloma: when endurance is interrupted. Lancet Oncol. 2020;21(12):e540.

Haematologica | 108 December 2023

3479

LETTER TO THE EDITOR

Impact of pinworm infection on the development of murine B-cell leukemia/lymphoma in the presence and absence of ETV6::RUNX1 B-cell acute lymphoblastic leukemia (B-ALL) is the most common malignant disease in childhood with a peak age at 2-6 years old. A two-step model has been proposed for the development of childhood B-ALL.1 The first step is an early somatic genetic rearrangement, such as the ETV6::RUNX1 fusion gene, followed by a broad range of secondary mutational events driven by environmental stimuli (including infection and abnormal cytokine release from immunologically untrained cells).1 The involvement of at least two discrete steps suggests that B-ALL may be preventable in infants with a genetically initiated step, who could be protected from harmful postnatal environmental stimuli. In the past decades, infections have been regarded as the environmental stimuli with the most impact in the etiology of childhood B-ALL. Common pathogens may drive secondary mutations in genetically predisposed subjects.2 Experimental models of infection can be leveraged in xenograft and animal models that closely resemble the pathophysiology of childhood B-ALL. For example, two animal studies demonstrated that transgenic mice (with the ETV6::RUNX1 fusion or with Pax5+/- heterozygosity) only developed B-ALL when they were exposed to common infections, although with incomplete penetrance.3,4 These past studies indicate that infections can act as important promoters of B-ALL development in the context of genetic predispositions. However, exposure to pathogens early in life via childhood contacts (daycare, microbiome) may modulate immune reactivity and decrease risk.2 Through a serendipitous observation, we found that the impact of pinworm infection on leukemogenesis was markedly different depending on the presence or absence of a common human somatic genetic change. Pinworms are a commonly found intestinal helminth in laboratory animals and the control of these pathogens in animal holdings is quite difficult.3 We performed a retrospective analysis of the latency and incidence of leukemia/lymphoma of two strains of mice during and after a pinworm outbreak in a specific-pathogen-free (SPF) facility (Figure 1A). Cdkn2a-/- and ETV6::RUNX1+ Cdkn2a-/- (referred to as E6R1+Cdkn2a-/- in the current work and Cre+ TA+ Cdkn2a-/- in the past work4) mice were maintained on the FVB/N strain background and were age- and sex-matched for each survival experiment. In line with the details of our animal use protocol, body condition scoring, clinical signs, and a diagnosis of neoplasia were used, in consultation

with veterinarians, to identify animals that had reached our predefined study endpoints. A gross necropsy was performed to identify potential sources of illness. Selected tissues were preserved through formalin-fixed paraffin embedding and stained with hematoxylin & eosin. Diagnoses were based upon gross necropsy and histopathology. If gross necropsy findings suggested a hematopoietic neoplasm, single-cell suspensions of involved tissues were cryopreserved in medium containing 10% dimethylsulfoxide. Additional diagnostic information was obtained by immunophenotyping when necessary. All experiments were performed following institutional review and approval by the University of California San Francisco Institutional Animal Care and Use Committee. We previously reported that ETV6::RUNX1 expression cooperates with Cdkn2a deletion to promote the development of B-ALL in mice (Figure 1B). After 2013, the leukemogenic effect of E6R1 expression was no longer observed, as demonstrated by two independent experiments showing overlap (Figure 1C) or minimal separation (Figure 1D) between the Cdkn2a-/- and E6R1+ Cdkn2a-/- survival curves. Review of the infection records in the SPF facility revealed that an outbreak of the pinworm Aspicularis was detected by fecal floatation testing of sentinels during the timeframe when decreased latency for leukemia/lymphoma had been observed in E6R1+ Cdkn2a-/- mice in comparison with Cdkn2a-/- mice. Following the outbreak of Aspicularis, all mice in the room were treated with the broad spectrum antihelminthic fenbendazole. The diminished effect of E6R1 on promoting leukemia/lymphoma development was observed after the eradication of pinworm. We then prospectively investigated the impact of intentional pinworm exposure on leukemogenesis. To determine whether pinworm infection could restore the leukemogenic effect of E6R1 in the Cdkn2a-/- model, 4week-old Cdkn2a-/- and E6R1+ Cdkn2a-/- mice were transferred from an SPF facility to an Aspicularis-infected conventional facility (also detected by fecal floatation testing of sentinels), where they were followed for survival (Figure 1A). In the context of pinworm infection, E6R1+ Cdkn2a-/- mice developed leukemia/lymphoma earlier and with a higher incidence than Cdkn2a-/- mice (Figure 1E). Together, these survival studies indicate a different impact of pinworm infection on the development of leukemia/lymphoma in Cdkn2a-/- and E6R1 Cdkn2a-/- mice. Given the well-established role of the E6R1 mutation in co-

Haematologica | 108 December 2023

3480

LETTER TO THE EDITOR

Figure 1. Pinworm exposure drives differences in leukemia/lymphoma development between Cdkn2a-/- and E6R1+ Cdkn2a-/- mice. (A) Timeline of individual survival studies following Cdkn2a-/- and E6R1+ Cdkn2a-/- mice in specific pathogen-free (SPF) and conventional facilities relative to the 2013 fenbendazole treatment. (B) Survival curves of leukemia/lymphoma development in Cdkn2a-/(N=34) and E6R1+ Cdkn2a-/- (N=22) mice housed in an SPF facility during a pinworm outbreak prior to fenbendazole treatment. Arrows indicate the chronological order of the survival studies. The year in brackets corresponds to the date of euthanasia of the last mouse to develop illness in each cohort. (C, D) Survival curves from two independent experiments of mice housed in an SPF facility after pinworm was eradicated with fenbendazole treatment. 2016-2017: Cdkn2a-/- (N=58) and E6R1+ Cdkn2a-/- (N=40); 2017: Cdkn2a-/- (N=18) and E6R1+ Cdkn2a-/- (N=15). (E) Survival curve from one experiment in which Cdkn2a-/- (N=20) and E6R1+ Cdkn2a-/(N=22) mice were housed in an SPF facility for 4 weeks, then transferred to a conventional facility for exposure to pinworm bedding. The log-rank (Mantel-Cox) test was applied to the survival curves.

Haematologica | 108 December 2023

3481

LETTER TO THE EDITOR operating with radiation, chemicals, and infectious exposures to promote the development of lymphoid malignancies,3,4,6,7 we had hypothesized that pinworm infection would promote leukemogenesis and have a stronger effect in E6R1+ Cdkn2a-/- mice than in Cdkn2a-/- mice. We examined this hypothesis by aggregating the survival cohorts from the studies shown in Figure 1. Mice In a pinworm-free facility mice (cohorts shown in Figure 1C, D) were aggregated to serve as pinworm-free controls (Figure 2A). For comparison, survival cohorts of pinworm-infected mice (Figure 1B, E) were also aggregated. Consistent with individual experiments, only pinworm-infected mice demonstrated a statistically significant difference in the development of leukemia/lymphoma

between E6R1+ Cdkn2a-/- mice and Cdkn2a-/- animals (Figure 2B). In contrast to our hypothesis, leukemia/lymphoma free survival curves revealed that pinworm infection elicited a protective effect in Cdkn2a-/- mice (Figure 2C) that was not observed in E6R1+ Cdkn2a-/- mice (Figure 2D). In pinworm-infected Cdkn2a-/- mice, the median latency of leukemia/lymphoma development increased from 305 days to 365 days (Table 1). In contrast, pinworm infection was associated with a decrease in the median latency of leukemia/lymphoma in E6R1+ Cdkn2a-/- mice from 253 days to 231 days (Table 1). Although there was an effect on latency, the incidence of leukemia/lymphoma was not affected by pinworm infection (Online Supplementary

Figure 2. Impact of pinworm exposure on leukemia/lymphoma development in the absence and presence of E6R1. Cumulative survival curves showing combined data from Figure 1 of four independent experiments of mice housed in a pinworm-free (SPF) facility or pinworm-infected facility. Leukemia/lymphoma-free survival for (A) SPF-housed (open symbols) and (B) pinworm-exposed (filled symbols) Cdkn2a-/- mice (black triangles) and E6R1+ Cdkn2a-/- mice (orange squares). Leukemia/lymphoma-free survival for genotype-matched (C) Cdkn2a-/- mice and (D) E6R1+ Cdkn2a-/- mice. SPF Cdkn2a-/- mice (N=76), SPF-housed E6R1+ Cdkn2a-/- mice (N=55), pinworm-exposed Cdkn2a-/- mice (N=54), and pinworm-exposed E6R1+ Cdkn2a-/- mice (N=44). The logrank (Mantel-Cox) test was applied to the survival curves. Haematologica | 108 December 2023

3482

LETTER TO THE EDITOR Table 1. Median latency of development of malignancy in Cdkn2a-/- and E6R1+ Cdkn2a-/- mice housed in pinworm-free or pinworm-infected facilities. Median latency, days Malignancy

SPF E6R1+Cdkn2a-/N=55

Pinworm E6R1+Cdkn2a-/N=44

SPF Cdkn2a-/N=76

Pinworm Cdkn2a-/N=54

Cancer

248

202

218

259

Leukemia/lymphoma

253

231

305

365

SPF: specific pathogen-free facility (i.e., free of pinworm).

Table S1). Of note, pinworm exposure had a modest impact on leukemia/lymphoma-free survival curves of Cdkn2a-/- mice (Figure 2C) and E6R1+ Cdkn2a-/- mice (Figure 2D), but the cumulative bidirectional effect of pinworm resulted in a significant difference between these two mouse strains (Figure 2B). Because Cdkn2a-/- mice are susceptible to other cancers in addition to leukemia/lymphoma, we examined cancer-free survival curves: there was no difference between Cdkn2a-/- and E6R1+ Cdkn2a-/mice in SPF conditions (Online Supplementary Figure S1A), but there was a statistically significant difference in the presence of pinworm (Online Supplementary Figure S1B). Interestingly, pinworm was associated with divergent responses: protection in Cdkn2a-/- mice (Online Supplementary Figure S1C) and a promotion in E6R1+ Cdkn2a-/- nice (Online Supplementary Figure S1D). Neither E6R1 (Online Supplementary Figure S2A, B) nor pinworm exposure (Online Supplementary Figure S2C, D) had a statistically significant impact on the development of solid tumors. Taken together, this work demonstrates a protective effect of an infection on leukemia/lymphoma development of Cdkn2a null mice in the absence of the E6R1 mutation, but that this protection is reversed in the presence of this common prenatally acquired genetic change. The increased latency of leukemia/lymphoma in pinwormexposed Cdkn2a-/- mice is a novel in vivo demonstration of a protective role of pathogenic infection in a mouse model of B-ALL. These results build upon past findings, in which immune stimulation with CpG, a TLR9 agonist, protected against B-ALL in Eu-ret mice.5 Exposure to microbes that are capable of priming the early immune system to respond appropriately to infections has long been associated with reduced risk of childhood B-ALL.1,2 Interestingly, pinworms have immunomodulatory properties, and their decades-long decline in western societies is inversely correlated with the rising incidence of B-ALL and other childhood immune disorders.6 While the current study does not characterize the immune response to pinworm exposure in Cdkn2a-/- mice, pinworm infections are well-described to stimulate anti-inflammatory immune responses that are characterized by the production of the cytokine interleukin-10 (IL-10) and Tregulatory cells.7 As IL-10 is a protective factor for B-cell leukemia/lymphoma in humans8 and mice,9 it is possible

that pinworm-induced IL-10 and/or pinworm-induced microbial diversity provide protection against B-ALL in specific genetic settings. In addition to parasites, viral and bacterial pathogens are also capable of supporting the gut microbiome and inducing production of IL-10 and T-regulatory cells.10 It will therefore be interesting to determine whether the protection garnered from parasitic infection can be acquired through other infectious pathogens that stimulate similar immune pathways. The gut microbiome is well known for shaping immune responses through structural components or metabolites of its constituent bacteria11 and has recently been identified as a player in the development of B-ALL.12 Considering the impact of helminth infections in shaping the composition of the gut microbiome,13,14 it is likely that pinworm exerts a protective effect in Cdkn2a-/- mice by enhancing microbial diversity. Although the timing of pinworm introduction varied (from birth [Figure 1B] or from weaning age [Figure 1E]), we did not find evidence for a role of timing differences in leukemia/lymphoma development. We instead observed a strong genetic-environmental interaction, in which E6R1 expression completely inhibited the protective effect of pinworm exposure in Cdkn2a-/- mice. This result may be explained by the recently described ability of E6R1 to induce a state of microbial dysbiosis12 or the well-established role of E6R1 in converting B-cell precursors into B-ALL.2 Pinworm and probiotic interventions that are currently being investigated for the prevention of childhood autoimmune disorders may also have the capacity to prevent childhood B-ALL.2 Future studies should be aimed at understanding how genetics and infectious exposures interact to affect the gut-immune axis. This will be an important step toward leveraging the full potential of preventative strategies in children who are genetically predisposed to B-ALL.

Authors Briana A. Fitch,1 Jamilla Situ,2 Joseph L. Wiemels,3 Scott C. Kogan2,4 and Mi Zhou2 Department of Pathology, Keck School of Medicine, University of

Haematologica | 108 December 2023

3483

LETTER TO THE EDITOR Southern California, Los Angeles; 2Department of Laboratory

Contributions

Medicine, University of California San Francisco, San Francisco;

BAF and SCK conceived the study. BAF, MZ, and SCK designed the

Center for Genetic Epidemiology, Department of Population and

study. BAF, MZ, and JS performed experiments and analyzed data.

Public Health Sciences, University of Southern California, Keck School

BAF wrote the manuscript with contributions from SCK and MZ. All

of Medicine, Los Angeles and Helen Diller Family Comprehensive

authors interpreted data, reviewed the work critically, and revised

Cancer Center, University of California San Francisco, San Francisco,

the manuscript.

CA, USA Acknowledgments We thank Todd Whitehead and Kamir Hiam for useful discussions.

Correspondence: MI ZHOU - mi.zhou@ucsf.edu

Funding This work was supported by National Institutes of Health/National

https://doi.org/10.3324/haematol.2022.282591

Cancer Institute grants R01 CA185058 (to SCK and JLW) and F31 CA221157 (to BF).

Received: December 14, 2022. Accepted: June 15, 2023.

Data-sharing statement

Early view: June 22, 2023.

The original data and protocols are available to other investigators without restriction.

References 1. Hauer J, Fischer U, Borkhardt A. Toward prevention of childhood ALL by early-life immune training. Blood. 2021;138(16):1412-1428. 2. Greaves M. A causal mechanism for childhood acute lymphoblastic leukaemia. Nat Rev Cancer. 2018;18(8):471-484. 3. Taffs LF. Pinworm infections in laboratory rodents: a review. Lab Anim. 1976;10(1):1-13. 4. Li M, Jones L, Gaillard C, et al. Initially disadvantaged, TEL-AML1 cells expand and initiate leukemia in response to irradiation and cooperating mutations. Leukemia. 2013;27(7):1570-1573. 5. Seif AE, Barrett DM, Milone M, Brown VI, Grupp SA, Reid GSD. Long-term protection from syngeneic acute lymphoblastic leukemia by CpG ODN-mediated stimulation of innate and adaptive immune responses. Blood. 2009;114(12):2459-2466. 6. Gale EAM. A missing link in the hygiene hypothesis? Diabetologia. 2002;45(4):588-594. 7. Maizels RM, McSorley HJ. Regulation of the host immune system by helminth parasites. J Allergy Clin Immunol. 2016;138(3):666-675. 8. Chang JS, Zhou M, Buffler PA, Chokkalingam AP, Metayer C, Wiemels JL. Profound deficit of IL10 at birth in children who

develop childhood acute lymphoblastic leukemia. Cancer Epidemiol Biomarkers Prev. 2011;20(8):1736-1740. 9. Fitch BA, Zhou M, Situ J, et al. Decreased IL-10 accelerates Bcell leukemia/lymphoma in a mouse model of pediatric lymphoid leukemia. Blood Adv. 2022;6(3):854-865. 10. Couper KN, Blount DG, Riley EM. IL-10: the master regulator of immunity to infection. J Immunol. 2008;180(9):5771-5777. 11. Schluter J, Peled JU, Taylor BP, et al. The gut microbiota is associated with immune cell dynamics in humans. Nature. 2020;588(7837):303-307. 12. Vicente-Dueñas C, Janssen S, Oldenburg M, et al. An intact gut microbiome protects genetically predisposed mice against leukemia. Blood. 2020;136(18):2003-2017. 13. Ramanan D, Bowcutt R, Lee SC, et al. Helminth infection promotes colonization resistance via type 2 immunity. Science. 2016;352(6285):608-612. 14. Lee SC, Tang MS, Lim YAL, et al. Helminth colonization is associated with increased diversity of the gut microbiota. PLoS Negl Trop Dis. 2014;8(5):e2880.

Haematologica | 108 December 2023

3484

LETTER TO THE EDITOR

CAR T-cell therapy for central nervous system lymphomas: blood and cerebrospinal fluid biology, and outcomes Therapeutic options are limited for relapsed/refractory (R/R) central nervous system lymphomas (CNSL), including primary CNSL (PCNSL) and secondary CNSL (SCNSL).1-3 Chimeric antigen receptor (CAR) T cells has been one of the most promising novel cancer therapies. Given that almost all CNSL express the CD19 antigen and that T cells are known to pass the blood brain barrier, there is a strong biological rationale for treating them with CAR T cells but they have been excluded from most clinical trials. Recent data have shown that CAR T cells could be efficient in R/R PCNSL and we recently reported an overall response rate (ORR) of 67% in nine patients.4–7 However, the dynamics of CAR T cells and their trafficking and persistence in cerebrospinal fluid (CSF) have been rarely described. Identification of biomarkers associated with increased disease control could enhance understanding of the biological basis for efficiency and enables more effective treatment interventions. We investigated the clinical outcomes and the CAR T-cell expansion and phenotype in the peripheral blood and CSF within 21 patients with CD19+-R/R CNSL. Patients with isolated R/R CD19+-PCNSL or SCNSL and treated with tisagenlecleucel (n=19) or axicabtagene ciloleucel (n=2) from January 2020 to January 2022 at PitiéSalpêtrière Hospital (France) were retrospectively selected. The disease response was assessed at day 28 (D28) by brain magnetic resonance imaging (MRI) and then every 2 months. Clinical results of eight PCNSL patients were previously reported with shorter follow-up and without any biological data.4 All patients gave written informed consent, the study was performed in accordance with the Declaration of Helsinki and was approved by national (CNIL 913611) and local (CPP Ile-De-France PP 13-022) Ethics Committees. Fresh blood samples were collected for realtime quantitative polymerase chain reaction (RT-qPCR) every 3 days the first 2 weeks, every week the following 2 weeks and then monthly. PMBC were collected and used for mass and flow cytometry at the peak of expansion. For detection of integrated CAR-expressing vectors, DNA was extracted from blood using a QIAamp DNA Blood Mini Kit. RT-qPCR was performed on 100 ng of extracted DNA using the 2X TaqMan Universal PCR Master Mix. The specific primers and TaqMan probes detected the CD28-CD3ζ (Yescarta) and the 41BB-CD3ζ (Kymriah) junctions of the CAR T transgene (Online Supplementary Figure S1). For CAR T immune profiling by mass cytometry, thawed PBMC were stained with the CD19 CAR Detection Reagent, Biotin (Mil-

tenyi), washed, and stained with an anti-biotin-106-116Cd antibody for 20 minutes as previously described.8 They were then incubated with the MDIPA panel (Fluidigm), plus five antibodies (anti-PD-1-175Lu, anti-TIGIT-209Bi, anti-TIM-3169 Tm, anti-CD69-162Dy and anti-CXCR4-165Ho). Samples were acquired on a Helios machine and analyzed with Maxpar Pathsetter, FlowJo, and OMIQ. For CAR T functionality assessment by flow cytometry, 1x106 thawed PBMC were stimulated 5 hours with a CD19+ B-lymphocyte cell line immortalized by Epstein-Barr virus (Raji cell line), then incubated with biotin-labeled CD19 for 30 minutes and stained with Live/Dead, anti-biotin-PE, CD107a-APC-R700, CD3APC780, CD8-BV605, CCR5-BV650, CCR6-BV700, CD69-PECF594, interleukin (IL) 17-BV421, interferon-γ (IFN-γ)-FITC, tumor necrosis factor-α (TNF-α)-PE-Cy7, and IL22-AF647. Fresh CSF samples were collected every 2 weeks during the first month and then monthly, and tested after incubation with biotin-labeled CD19 for 30 minutes, with antiCD3-APC-H7, CD4-PercPCy5.5, CD8-APC-Alexa700 and anti-biotin-PE. IL-6 levels were measured by Cytometric Bead Array technique in freshly thawed CSF samples, on a FACSCanto II cytometer. Twenty-one R/R patients (13 PCNSL, 8 SCNSL) were selected, all but one patients with brain parenchymal or meningeal involvement and nine with CSF involvement (Online Supplementary Table S1). The median age was 67 years and the median number of prior therapies three (range, 2–5), including ASCT in 16 patients. At the time of CAR T-cell infusion, 62% had progressive disease (PD), 29% partial response (PR), 5% stable disease and 5% complete response (CR). The median follow-up was 12 months (range, 1–29). At D28, ORR was observed in 67% cases, including 29% CR and 38% PR. At month 3 (M3), ORR and CR without new treatment were observed in nine (43%) (6/13 PCNSL, and 3/8 SCNSL) and six (29%) (5/13 PCNSL, and 1/8 SCNSL) patients, respectively. Among the nine patients with response at M3, the median response duration was 19 (range, 8–29) months. On final follow-up, eight (38%) patients had persistent response: six CR (4/13 PCNSL, and 2/8 SCNSL) and two PR (1 PCNSL and 1 SCNSL), and all the 13 remaining patients died because of R/R disease. The median overall survival was 15 months and tended to be higher in PCNSL than in SCNSL (20 vs. 12 months; P=0.63) (Figure 1), and the median progression-free survival was 3 months. For subsequent biological analysis, patients with response

Haematologica | 108 December 2023

3485

LETTER TO THE EDITOR ≥6 months without new systemic treatment were defined as responders (R, n=8), and the others as non-responders (NR, n=13). Cytokine release syndrome and immune effector cell-associated neurotoxicity syndrome occurred in 16 (1 grade 3) and seven (2 grades≥ 3) patients respectively, all having been resolutive using tociluzumab and corticosteroids for 14 and five patients. We first assessed the expansion of CAR T cells in the CSF for 16 patients. All tested patients demonstrated CSF positivity for CAR T cells regardless of clinical outcome with an initial phase of rapid expansion followed by a slow decrease with long-term persistence (Figure 2). On D30, the median number of CAR-T cells in the CSF was 0.17/mm3, reflecting 19% (range, 4–45) of CD3+ T cells, and the median CD4/CD8 CAR T-cells ratio was 4.2. The later CSF analysis assessed 17 months after the infusion in one R patient still demonstrated the presence of CAR T cells (41% of CD3). We demonstrated, focusing on the first month, that the expansion peak in the CSF was significantly higher in R than in NR patients (0.50/mm3 vs. 0.19/mm3 (P=0.01) (Figure 2). Finally, a transient increase in IL-6 dosage was detectable during the first month for 76% cases regardless of their outcomes. We next addressed the phenotypic characterization of

blood CAR T cells at the expansion peak with mass cytometry for 20 patients. We first demonstrated that R patients had a higher frequency of CD8 T cells with a terminal effector phenotype (CD8 TE, mean=24% vs. 11% of cells; P<0.05) and a lower frequency of regulatory T cells (Tregs, 0.5% vs. 1.8%; P< 0.05) compared to NR. We next performed a datadriven non-supervised analysis using the Phenograph clustering algorithm which identified 68 distinct cells subtypes (Figure 3A). Among them, PG-4 which presented a phenotype similar to T cells CD8 TE, confirming our previous observation, and PG-34 corresponding to CD4+ CAR T cells expressing high levels of IL7R and IL3 were particularly overrepresented in the R group (Figure 3B). Conversely, three clusters were less present among the responders: PG-31, phenotypically comparable to TIGIT-expressing Tregs, which refines the above findings; PG-8, CD4+ T cells positive for interleukin 7 receptor (IL7-R) and PG-13 which fits the description of natural killer (NK) cells. We further applied Phenograph specifically to CAR T cells and we confirmed the over-representation of CD4+ IL7R+ CAR T cells (PG-CAR13) among the responders (Figure 3C, D). In addition, PD1 showed a weaker expression among responders, and this was particularly striking regarding the CAR T cells (Figure 3E). This reduction of PD1 was paralleled by a CAR T-cell-

Figure 1. Overall survival and progression-free survival. (A) Overall survival (OS) and progression-free survival (PFS) in (A) all the patients, (B) in primary central nervous system lymphomas (PCNSL) patients (N=13) and (C) in secondary CNSL (SCNSL) patients (N=8).

Haematologica | 108 December 2023

3486

LETTER TO THE EDITOR specific reduction of CXCR4 expression, especially among CD4+ CAR. Finally, we analyzed the CAR T cells functionality and the expression of chemokine receptors at the peak of expansion by using thawed PBMC (n=16) after 5 hours of stimulation with tumor B cells in order to replicate the in vivo interaction with the CAR T-cell target. We showed that the IFNγ expression by CD4+ CAR T cells was higher in R than in NR patients (P=0.03), and that the expression of CD107a and IFN-γ by CD8+ CAR T cells tended to be higher in R patients (P=0.08) (Online Supplementary Figure S2). In addition, the expression of CCR5 and CCR6 on CD8+ CAR T cells was higher (54% and 47%, respectively) than that on the CD8+ non-CAR T cells (6% and 4%, respectively; P<0.0001) after the stimulation with tumor B cells. This upregulation on CAR T cells seemed to be the result of the specific costimulation since their expression on the CD8+ CAR T cells cultured without target B cells (negative control) was significantly lower (22% for CCR5 and 14% for CCR6; P=0.0096). We first reported persistent response in 38% cases with highly R/R disease, and it is very encouraging in view of

the very poor prognosis of CNSL. CNS CAR T-cell trafficking was reported in all of our patients, confirming their ability to migrate into the CNS and persist there despite the low level of target antigen in the blood. Of note, the R patients demonstrated higher CAR T-cell peaks in the CSF, reinforcing the recently published data6 but these results should be interpreted with caution due to far apart time points. We next suggested that a strong cytotoxic TE CD8+ T-cell response, and a diminished suppressive mechanism mediated by Tregs may be crucial for the efficacy of CAR T cells reinforcing that non-CAR T cells are also critical for the responses.9 We further demonstrated that a subset of CD4+IL3R+IL7R+ CAR T cells are associated with a better clinical response. Indeed, IL7 prolongs the survival time of tumor-specific T cells and the effector pool generation, and CAR T cells engineered for IL7-R constitutive activation had a higher anti-tumor activity and persistence in preclinical models.10–14 On the other hand, the expression of PD-1 and CXCR4 paralleled with a poorer outcome, probably reflecting an impaired capacity to migrate to the target tissue and then to kill the tumor cells. Our results

C Figure 2. Cerebrospinal fluid CAR T-cell expansion. (A) Individual profiles of chimeric antigen receptor (CAR) T cells expansion (% of CD3) in cerebrospinal fluid (CSF) for each patient. Red color denotes responder (R) patients, circle plots denote the patients with primary central nervous system lymphomas (PCNSL) and triangle plots the patients with secondary CNSL (SCNSL). (B) Number of CART cell/mm3 in CSF on the best expansion peak during the first month. (C) Interleukin 6 (IL-6) dosage (pg/mL) in CSF. The timing scale has not been respected for the first 30 days in order to better see all the dots. Red color denotes R patients, circle plots denote the patients with PNCSL and triangle plots the patients with SCNSL Mann-Whitney test.

Haematologica | 108 December 2023

3487

LETTER TO THE EDITOR

Figure 3. Immune profiling of peripheral blood mononuclear cells in central nervous system lymphoma patients treated with CAR T cells. (A) t-distributed stochastic neighbor embedding (t-SNE) representation of single-cell cytometry by time-of-flight (CyTOF) data for all patients (N= 20), depicting 68 cell clusters after Phenograph unsupervised clustering based on the expression of each of the 35 markers. Numbers within colored boxes correspond to the respective cluster number. (B) Abundances of selected clusters identified in and comparison of their frequencies between responders (R, red) and non-responders (NR, black) patients. (C) t-SNE representation depicting specific 34 CAR T-cell clusters after Phenograph unsupervised clustering based on the expression of each of the 35 markers. Numbers within colored boxes correspond to the respective cluster number. (D) Abundances of selected clusters identified in (C) and comparison of their frequencies between R (red) and NR (black) patients. (E) Overlayed histograms comparing expression of PD-1 (left) or CXCR4 (right) of gated total CAR T cells, and specific CD8+ or CD4+ CAR between R (red) and NR (black) patients. Haematologica | 108 December 2023

3488

LETTER TO THE EDITOR strongly support that engineered CAR T cells might increase the level of response. Finally, we showed for the first time that the production of IFN-γ was associated with clinical outcomes, suggesting that the peripheral immunological invigoration of CAR T reflects the activity at the tumor site. In conclusion, the use of CAR T-cell therapy in CNSL patients answers an unmet medical need, and we suggest that the CSF expansion as well as the functionality and phenotype of CAR T cells are implicated in the clinical outcome, paving the way for the development of novel CAR T cells with higher anti-tumor activity.

Correspondence: M. BARON - marine.baron@aphp.fr https://doi.org/10.3324/haematol.2023.282875 Received: February 23, 2023. Accepted: June 14, 2023. Early view: June 22, 2023. ©2023 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures No conflicts of interest to disclose.

Authors

Contributions AG, BA, VV, FN, DR-W and MB designed the study. CM collected and

Claire Lacan,1,2* Jonathan Caron,3* Nadine Tarantino,1 Baptiste

preserved CAR T-cell products. MC performed the cell banking. BF

Fouquet, Mustapha Cherai, Christophe Parizot, Véronique Morel, 4

performed the PCR analysis. CL, NT, MB, EL, AV, G.G, MM and JC

Laetitia Souchet, Madalina Uzunov, Guy Gorochov, Stéphanie 2

performed the immunological analysis. ES and MLG-T performed

Nguyen-Quoc, Elise Sourdeau, Vincent Vieillard, Makoto Miyara,

the cytokine dosage in the CSF. CP monitored the flow analysis. DR-

Angélique Vinit, Silvia Solorzano, Carole Soussain, Caroline

W, SC, FN, VM, LS, MU, SN-Q, CL, MB and CS took care of patients.

Houillier, Carole Metz, Brigitte Autran, Elena Litvinova, Magali Le

CL and MB collected clinical data. MB wrote the manuscript. All the

Garff-Tavernier,

authors reviewed the manuscript.

1,2

3,5

Françoise Norol, Damien Roos-Weil, Sylvain 2

Choquet,2 Amélie Guihot1 and Marine Baron1,2 Acknowledgments Sorbonne Université, INSERM U1135, CNRS EMR 8255, CIMI-Paris;

We acknowledge all the patients included in the study. We

Sorbonne Université, Department of Clinical Hematology, AP-HP,

acknowledge the French national expert network for oculo-cerebral

Pitié- Salpêtrière Hospital; Centre de Recherche des Cordeliers,

lymphomas (LOC) for the recruitment of the patients and the

INSERM, Cell Death and Drug Resistance in Lymphoproliferative

treatment decisions.

Disorders Team, Sorbonne Université, Université Sorbonne Paris Cité, Université Paris Descartes, Université Paris Diderot; 4Sorbonne

Funding

Université, Department of Immunology, AP-HP, Pitié- Salpêtrière

This study was supported by grants from INCA-DGOS-Inserm_12560

Hospital; 5Sorbonne Université, Department of Biological

(SiRIC CURAMUS is financially supported by the French National

Hematology, AP-HP, Pitié- Salpêtrière Hospital; Sorbonne

Cancer Institute, the French Ministry of Solidarity and Health and

Université, UMS37-PASS, Plateforme de Cytométrie CyPS, Pitié-

INSERM with financial support from ITMO Cancer AVIESAN), by

Salpêtrière Hospital; Hematology Unit, Institut Curie, Site de Saint-

Association Capucine and SFGM-TC, prix Force Hémato 2020 and by

Cloud et Center for Cancer Immunotherapy, Institut Curie, PSL

the Fondation ARC pour la recherche sur le cancer (to DRW). JC

Research University, INSERM U932; Sorbonne Université, INSERM,

was financially supported by SiRIC CURAMUS.

CNRS, UMR S 1127, Institut du Cerveau, ICM, Department of Neurology 2-Mazarin, AP-HP, Pitié Salpêtrière Hospital and

Data-sharing statement

Sorbonne Université, Unité REQPHARM, Pharmacie à Usage

All biological data were linked and shared at the Department of

Intérieur, AP-HP, Pitié- Salpêtrière Hospital, Paris France

Immunology, and of Hematology, Pitié-Salpêtrière Hospital, Paris, France. External users with a formal analysis plan can request access to these data.

CL and JC contributed equally as first authors.

References 1. Kridel R, Dietrich P-Y. Prevention of CNS relapse in diffuse large B-cell lymphoma. Lancet Oncol. 2011;12(13):1258-1266. 2. Bachanova V, Perales M-A, Abramson JS. Modern management of relapsed and refractory aggressive B-cell lymphoma: a perspective on the current treatment landscape and patient selection for CAR T-cell therapy. Blood Rev. 2020;40:100640.

3. Chapuy B, Roemer MGM, Stewart C, et al. Targetable genetic features of primary testicular and primary central nervous system lymphomas. Blood. 2016;127(7):869-881. 4. Alcantara M, Houillier C, Blonski M, et al. CAR T-cell therapy in primary central nervous system lymphoma: the clinical experience of the French LOC network. Blood. 2022;139(5):792-796.

Haematologica | 108 December 2023

3489

LETTER TO THE EDITOR 5. Siddiqi T, Wang X, Blanchard MS, et al. CD19-directed CAR T-cell therapy for treatment of primary CNS lymphoma. Blood Adv. 2021;5(20):4059-4063. 6. Frigault MJ, Dietrich J, Gallagher K, et al. Safety and efficacy of tisagenlecleucel in primary CNS lymphoma: a phase 1/2 clinical trial. Blood. 2022;139(15):2306-2315. 7. Wu J, Meng F, Cao Y, et al. Sequential CD19/22 CAR T-cell immunotherapy following autologous stem cell transplantation for central nervous system lymphoma. Blood Cancer J. 2021;11(7):131. 8. Corneau A, Cosma A, Even S, et al. Comprehensive mass cytometry analysis of cell cycle, activation, and coinhibitory receptors expression in CD4 T cells from healthy and HIVinfected individuals. Cytometry B Clin Cytom. 2017;92(1):21-32. 9. Scholler N, Perbost R, Locke FL, et al. Tumor immune contexture is a determinant of anti-CD19 CAR T cell efficacy in large B cell lymphoma. Nat Med. 2022;28(9):1872-1882.

10. Melchionda F, Fry TJ, Milliron MJ, McKirdy MA, Tagaya Y, Mackall CL. Adjuvant IL-7 or IL-15 overcomes immunodominance and improves survival of the CD8+ memory cell pool. J Clin Invest. 2005;115(5):1177-1187. 11. Shum T, Omer B, Tashiro H, et al. Constitutive signaling from an engineered IL7 receptor promotes durable tumor elimination by tumor-redirected T cells. Cancer Discov. 2017;7(11):1238-1247. 12. Zhao Z, Li Y, Liu W, Li X. Engineered IL-7 receptor enhances the therapeutic effect of AXL-CAR-T cells on triple-negative breast cancer. Biomed Res Int. 2020;2020:4795171. 13. Kim MY, Jayasinghe R, Devenport JM, et al. A long-acting interleukin-7, rhIL-7-hyFc, enhances CAR T cell expansion, persistence, and anti-tumor activity. Nat Commun. 2022;13(1):3296. 14. Li L, Li Q, Yan Z-X, et al. Transgenic expression of IL-7 regulates CAR-T cell metabolism and enhances in vivo persistence against tumor cells. Sci Rep. 2022;12(1):12506.

Haematologica | 108 December 2023

3490

LETTER TO THE EDITOR

CD47 expression in acute myeloid leukemia varies according to genotype CD47 is a “don’t eat me” signal to phagocytes, that is overexpressed in solid and blood tumors and represents a key mechanism of immune evasion in cancer.1 Engagement of signal-regulating protein α (SIRPα) on phagocytic cells by CD47 prevents phagocytosis of tumor cells.1 In acute myeloid leukemia (AML), CD47 is known to be upregulated on leukemia stem cells (LSC) to avoid phagocytosis2,3 and increased expression of CD47 commonly predicts worse overall survival.3 CD47-directed blocking monoclonal antibodies are effective against AML engraftment in preclinical models, by inducing phagocytosis of LSC,3 and several clinical trials are currently evaluating the efficacy of the Hu5F9-G4 monoclonal antibody magrolimab, alone or in combination with other anti-leukemic compounds in adult AML patients. Early results from two phase Ib trials have shown some clinical efficacy of anti-CD47/magrolimab in AML,4,5 but solid data on the leukemic patients who are more likely to benefit from this immunotherapeutic approach, according to the underlying genotype, is still lacking. Here, we provide a detailed immunohistochemical characterization of CD47 protein expression on leukemic cells from formalin- or B5-fixed paraffin-embedded bone marrow sections across AML genomic spectrum, according to 2022 European-LeukemiaNET (ELN) risk stratification, and complement it with bulk transcriptome analysis of leukemic cells from AML patients included in the Beat-AML dataset.6 We investigated a cohort of adult AML patients (n=53) from University College of London Cancer Center, London, UK, and the Institute of Hematology, University Hospital of Perugia, Italy. Risk stratification was as follows: 19 of 53 (36%) had low-risk, 19 of 53 (36%) intermediate-risk, and 15 of 53 (28%) adverse-risk AML. Low-risk AML included three (16%) cases with RUNX1/RUNX1T1, eight (42%) with CBFβ/MYH11 and eight (42%) patients with NPM1mutFLT3wt AML. The intermediate-risk category included seven (37%) patients with NPM1mutFLT3-internal tandem dublication (ITD), four (21%) with NPM1wtFLT3-ITD and eight (42%) with KMT2A/MLLT3rearranged AML. The adverse-risk category included five (33%) patients carrying KMT2A-rearranged AML (other than KMT2A/MLLT3 fusion), three (20%) having chromosome 3 aberrations, four (27%) with monosomy 7, and three (20%) with a complex karyotype. The clinicopathologic features of the study cohort are summarized in Table 1. Immunostaining for CD47 of bone marrow (BM) biopsy specimens from AML patients was performed using the recombinant rabbit monoclonal anti-CD47 antibody, EPR21794 clone (Abcam, ab218810); notably, the anti-CD47 antibody (EPR21794 clone) used in this study, as well as the Hu5F9-G4 monoclonal

antibody magrolimab,7 are both raised against the extracellular domain of the human CD47 molecule. The immunostaining procedure was carried out as previously described.8 In details, the anti-CD47 antibody was tested with similar results in both the BOND-III AutoStainer (Leica Microsystems, Newcastle-upon-Tyne, UK) and the Benchmark ULTRA (VENTANA/ROCHE Diagnostics, Tucson, AZ, USA). The antigen retrieval method was the same for both platforms, as was the EDTA pH-9 based procedure. The HRP-labeled polymers were used as per suppliers’ specification and DAB was applied to detect the antigen-antibody reaction. The BOND-III AutoStainer platform was then used to perform the whole series of staining. The CD47 immunohistochemical expression and antigen intensity was scored from 0 to 3 (0: absent membranous staining; 1: weak; 2: moderate: 3: strong) and independently assessed by three investigators (AM, BF, and TM). Fixation methods did not influence CD47 antigen detection or intensity; as such, results from our analysis appeared fixation-independent across all genomic subtypes of AML. In the low-risk category, 16 of 19 cases had strong CD47 expression, while, among the remaining three cases, two of three (67%) with RUNX1/RUNXT1 and one of eight (12.5%) with NPM1mut AML showed weak expression (intensity score: 1). In particular, all CBFβ/MYH11 AML samples (n=8) showed a diffuse and strong (intensity score: 3) expression of CD47 on leukemic blasts (Figure 1A, B; Table 1). This was in keeping with the levels of CD47 mRNA in leukemic cells from patients with CBFβ/MYH11 AML, included in the BeatAML project dataset.6 Indeed, CBFβ/MYH11 AML consistently showed an elevated expression of the CD47 gene transcript compared to other genomic subtypes, such as RUNX1/RUNX1T1 or KMT2A/MLLT3 (Figure 2). Notably, CBFβ/MYH11 AML cases bearing KIT-D816V mutations included in the Beat-AML dataset, all clustered within the CD47-high subgroup (n=4/9, Fisher-exact test; P=0.08), that has been stratified according to the median normalized expression of the CD47 gene transcript. However, in our study cohort, only one patient (1/8) with CBFβ/MYH11 AML was KITD816V-mutated, thereby precluding validation by immunohistochemistry of the transcriptomic analysis performed with the Beat-AML dataset.6 Among AML cases carrying an NPM1 mutation (n=15), 13 of 15 were positive for CD47, but two of 13 (15%) had weak expression (intensity score: 1) (Table 1). Our results were in keeping with a previous tissue microarray study,9 pointing to a significant association between an increased CD47 expression and the presence of NPM1 mutation, but not of FLT3-ITD mutation. Accordingly,

Haematologica | 108 December 2023

3491

LETTER TO THE EDITOR Table 1. Demographic, diagnostic features and immunohistochemical analysis of CD47 antigen expression in the acute myeloid leukemia patients’ cohort. Age Blast ID in Sex percentage yrs

Blast phenotype

CD47 expression

CD47 intensity score

Low-risk category RUNX1/RUNX1T1

1 44 2 25 3 56

M M M

90-95 90 95

CD34+, MPO+/dot-like, CD15-/+ focal, CD68-, CD117+ CD34-, MPO+, CD68 weak CD34-/+, MPO+, CD68-/weak

Positive Weak Weak

3 1 1

CBFβ/MYH11

4 45 5 79 6 54 7 41 8 62 9 71 10 50 11 28

M F M M M M M F

Diffuse 90 Diffuse Diffuse Diffuse 30-40 Diffuse 5-10

CD34+/-, MPO-/+, CD68+/CD34+ (10-15%), MPO+/-, CD68+/CD34-/+, MPO+, CD68+ CD34+ (15%), MPO+/-, CD68+/CD34-/+, MPO-/+, CD68-/+ CD34+ (10-15%),MPO+/-, CD68-/+ CD34-/+, MPO+/-, CD68-/+ + CD34 (5-10%), CD117+ (15%), CD38-/weak

Positive Positive Positive Positive Positive Positive Positive Positive

3 3 3 3 3 3 3 3

NPM1mut FLT3wt

12 74 13 66 14 56 15 43 16 66 17 53 18 47 19 65

F M F F M M M M

80 90 95 Diffuse 60-70 Diffuse >90 Diffuse

CD34-, MPO+/-, CD68-/+ CD34-, MPO-/+, CD68+/CD34-, MPO+, CD68CD34-, MPO+, CD68-/+ CD34-, MPO-/+, CD68+/CD34-, MPO+/-, CD68-/+ CD34-, CD33+, CD117+ (10-15%), MPO+ (10-15%) CD34-, CD117 very weak, MPO focal

Negative/weak Positive Positive Positive Positive Positive Positive Positive

1 3 3 3 3 3 3 3

NPM1mut FLT3-ITD

20 62 21 27 22 48 23 52 24 79 25 70 26 54

F F M M F F M

Diffuse Diffuse Diffuse 90 Diffuse 90-95 95

CD34-, MPO-/+, CD68-/+ CD34-, MPO+, CD68CD34-, MPO-/+, CD68-/+ CD34-, MPO-, CD68-/+ CD34-, MPO+, CD68CD34-, MPO+/focal, CD68+ CD33+, CD34-, CD117-

Weak Positive Positive Positive Positive Negative Negative

1 3 3 3 3 0 0

NPM1wt FLT3-ITD

27 73 28 73 29 84 30 76

M F M F

40 Diffuse 90-95 Diffuse

CD34+ (4-5%), MPO+/- weak, CD68+/CD34+ (5%), MPO+/-, CD68+/CD34-, MPO+/weak, CD68+ CD34+, MPO+/- weak, CD68-

Weak Focal positive (20-30%) Negative Negative

1 1 0 0

KMT2A/MLLT3

31 57 32 17 33 16 34 32 35 65 36 69 37 47 38 33

M M F F F F M M

90-95 100 100 80 >90 40 100 95

CD34-, MPO+/-, PGM1+/CD34-, CD117-, CD33+, MPO-, TdTCD34-, CD117-, MPO+ CD34 , CD38-, CD117+, CD11c+, MPO+ CD34+, CD117+/weak, MPO focal weak CD34-, CD117-, MPO-, CD11c+, CD68+ CD34+, CD117+ (6%) CD34-, MPO weak, CD68+, CD117+ focal, CD33+, CD14 weak

Negative Negative Negative Negative Weak Negative Weak Weak

0 0 0 0 1 0 1 1

39 24 40 26 KMT2A-rearranged* 41 57 42 59 43 34

M M F M F

Diffuse 25-30 Diffuse Diffuse 90-95

CD34+ (40-50%), CD117+ (40-50%) CD34+, CD117+ CD34-, MPO-/+, CD68-/+ CD34-, MPO-/+, CD68+ CD34+ (20-25%), MPO-/+, CD68-/+

Negative Negative Negative/weak Negative Weak

0 0 0/1 0 1

44 74 3q rearrangements 45 62 46 57

M M M

70-80 80 90

CD34+ (40%), CD117 weak, MPO focal weak, CD68CD34+, MPO+/-, CD68 focal weak CD34+/-, MPO focal weak, CD68+ focal

Positive Weak Positive

3 1 3

47 64 48 64 49 57 50 57

F F F F

40 95 Diffuse 90

CD34+/-, MPO+, CD68CD34+, MPO+/-, CD68+/CD34-, MPO+, CD68+ focal CD34+, MPO+, CD68-

Weak Negative Positive Positive

1 0 3 3

51 62 Complex karyotype 52 62 53 67

M M F

Diffuse 20-25 20

CD34+, MPO+, CD68CD34+ CD34+

Focal weak Positive Positive

1 2 2

Intermediate-risk category

Adverse-risk category

Monosomy 7

Continued on following page. Haematologica | 108 December 2023

3492

LETTER TO THE EDITOR Detailed are the demographic characteristics of the study cohort, including patient’s sex and age at acute myeloid leukemia (AML) diagnosis. Blast percentage and phenotype were evaluated on the diagnostic bone marrow biopsy specimen. CD47 expression was then assessed and scored according to differential grades of antigen immunoreactivity (score 0=negative; score 1=weak positive; score 2=moderate positive; score 3=strong positive) from 3 investigators (AM, BF and TM). *KMT2A-rearranged AML cases had genetic abnormalities other than KMT2A/MLLT3 fusion. yrs: years; mut: mutated; wt: wild-type; ITD: internal tandem duplication. +/-: in case of >50% positivity of AML cells; -/+: in case of <50% positivity of AML cells.

in our cohort, two of four NPM1wtFLT3-ITD mutated AML patients had negative CD47 expression on leukemic cells, and the other two cases had weak (ID 27) or focal positive (ID 28) expression of CD47 antigen (Table 1). Interestingly, all AML patients carrying KMT2A/MLLT3 (n=8) and KMT2Ar (n=5), respectively included in the intermediate- and adverse-risk ELN genomic categories, showed no/low expression of CD47 on leukemic blasts (Table 1), with four of 13 cases (30%) having low expression (intensity score: 1; Figure 1C, D) while the other nine cases displayed complete absence of CD47 antigen expression (intensity score: 0; Figure 1E, F). Notably, the immunohistochemical findings in the KMT2A/MLLT3 subgroup of AML patients were in keeping with bulk transcriptome analysis performed

with the Beat-AML dataset6 (Figure 2) and were supported by previous studies with KMT2A/MLLT3 rearranged AML cell lines showing low protein levels of CD47 by western blot.2 In the adverse-risk category, which included patients with complex karyotype (n=3), monosomy 7 (n=4), KMT2A (n=5) or 3q rearrangements (n=3), 11 of 15 (73%) had positive CD47 expression, including five of 11 cases with weak (intensity score: 1), two of 11 with moderate (intensity score: 2) and four of 11 with strong (intensity score: 3) immunohistochemical expression (Table 1). Unfortunately, the limited number of TP53-mutated AML cases (patients ID 52 and 53; Table 1), both of whom demonstrated moderate expression of CD47 (intensity score: 2), precluded any further insights into this very high-risk disease setting, so it

Figure 1. Immunohistochemical expression of CD47 immune antigen across genomic subtypes of acute myeloid leukemia. (A and B) Strong positive (intensity score: 3) and diffuse expression of CD47 by leukemic cells in a patient with CBFβ/MYH11-rearranged acute myeloid leukemia (AML, original magnification 4X (A) and 40X (B)). (C and D) CD47-weak positive expression (intensity score: 1) by leukemic cells in a patient with KMT2A/MLLT3-rearranged AML. Focal strong staining (black arrows) is seen in background dysplastic megakaryocytes and scattered small cells with round regular nuclei probably of lymphoid origin (asterisk) (original magnification 40X). (E and F) CD47-negative expression (intensity score: 0) by leukemic cells (E and F) in a patient with KMT2A-rearranged AML. Scattered background positive small/medium-sized rounded cells, most likely of lymphoid origin, show membranous CD47 expression (F, black arrows) (original magnification 40X). Immunostaining for CD47 was performed using recombinant anti-CD47 antibody, EPR21794 clone, Abcam. Haematologica | 108 December 2023

3493

LETTER TO THE EDITOR is not clear if such cases could benefit from anti-CD47/magrolimab immunotherapy as suggested by preliminary results from clinicaltrials gov. Identifier: NCT03248479 trial.5 In this study, we have reported that adult AML shows differential patterns of expression of the CD47 immune molecule on leukemic cells. Importantly, KMT2A/MLLT3 and KMT2A-rearranged AML cells were mostly negative for CD47, supporting the use of drugs forcing CD47 antigen expression, such as the hypomethylating agent azacitidine,10 in combination with anti-CD47 immunotherapy.1,4,5 Conversely, CD47 was strongly and consistently expressed in CBFβ/MYH11 AML as well as in the majority of NPM1-mutated AML cases (in the absence of FLT3-ITD mutation), which suggest these patients may respond better to antiCD47/magrolimab immunotherapy, urging clinical trials to address this issue. To date, no study has previously evaluated the importance of measuring CD47 antigen density on AML blasts by immunohistochemistry in whole BM section, as well as its relative impact on predicting response to anti-CD47 immunotherapy. However, our data strongly encourage to assess CD47 antigen density in each AML patient undergoing a magrolimab-based treatment.

Emerging results from clinical trials are suggesting that magrolimab holds promise to be effective against highrisk genetic AML category,4,5 which, in our study cohort, was found to have a lower expression of CD47 molecule compared to other categories. Although we cannot exclude that even a small amount of CD47 antigen could mediate response to magrolimab, it should be noted that in the mentioned clinical trials,4,5 anti-CD47 immunotherapy was combined to azacitidine, an hypomethylating agent being reported to i) increase CD47 expression,10 ii) upregulate eat-me signals, as calreticulin10 and iii) enhance phagocytosis of leukemic cells, thereby acting synergistically with 5F9 (magrolimab).11,12 Thus, such therapeutic approach may prove effective in treating even AML cases with low levels of CD47 antigen expression on leukemic cells, as those herein more frequently found within the high-risk genetic category. In conclusion, this is the first report quantitatively assessing the immunohistochemical expression of CD47 immune antigen on leukemic cells across distinct genomic subtypes of AML, thus providing a potentially useful guide to immunotherapeutic approaches targeting CD47/SIRPα axis in AML.

Figure 2. CD47 mRNA expression across genomic subtypes of acute myeloid leukemia. Transcriptomic data were retrieved from the “Beat-AML” project (http://www.vizome.org/). Bioinformatic and statistics analyses were performed on RNA-sequencing data from bone marrow (BM) samples. Fragments per kilobase of transcript per million mapped reads (FPKM) values for each gene/sample were calculated from RNA-sequencing raw counts, using rpkm function of “edgeR” (R package). According to European LeukemiaNet genetic-risk categories, 10 distinct genomic subgroups of acute myeloid (AML) were identified: the low-risk category includes RUNX1/RUNX1T1, CBFβ/MYH11 and NPM1mutFLT3wt genotypes; the intermediate-risk category includes NPM1mutFLT3-internal tandem duplication (ITD), NPM1wtFLT3-ITD and KMT2A/MLLT3 genotypes; the high-risk category includes AML cases with the following genotypes, as KMT2A-rearrangement (other than KMT2A/MLLT3), GATA2/MECOM, monosomy 7, and complex karyotype. Haematologica | 108 December 2023

3494

LETTER TO THE EDITOR

Authors

Andrea Marra,1,2,3,4 Ayse U. Akarca,5 Giovanni Martino,1,6 Alan Ramsay,5

Disclosures

Stefano Ascani, Vincenzo Maria Perriello, Jenny O’Nions, Andrew J.

BF licensed a patent on NPM1 mutants (n. 102 004 901 256 449) and

Wilson, Rajeev Gupta,

declares honoraria from Rasna Therapeutics, Inc. for scientific

7,8

Anna Childerhouse, Ian Proctor, Manuel 5

Rodriguez-Justo, Sabine Pomplun, Maria Paola Martelli, Cristina Lo

advisor activities. BF is a member of the Neogenomics scientific

Celso,

advisory board. MPM declares honoraria/consultancy at scientific

2,9

Brunangelo Falini and Teresa Marafioti 1#

advisory board for AbbVie, Amgen, Celgene, Janssen, Novartis, Pfizer Institute of Hematology and Center for Hemato-Oncological

and Jazz Pharmaceuticals. The other authors declare no competing

research (CREO), University of Perugia and Santa Maria della

financial interests.

Misericordia Hospital, Perugia, Italy; Department of Life Sciences, 2

Imperial College London, South Kensington Campus, London, UK;

Contributions

Laboratory of Molecular Medicine and Biotechnology, Department

AM, BF and TM conceived the study and designed the research. AM,

of Medicine, University Campus Bio-Medico of Rome, Rome, Italy;

AUA, BF and TM performed most of the research and analyzed

Institute of Translational Pharmacology, National Research Council

immunohistochemical expression of CD47 in the cross-institutional

of Italy (CNR), Rome, Italy; 5Department of Pathology, University

cohort of patients. AM performed analysis of CD47 mRNA expression

College London, London, UK; 6Pathology Unit, Azienda Ospedaliera

across genomic subtypes of AML with the Beat-AML dataset. JO’N, RG

"Santa Maria" di Terni, Terni, Italy; Department of Hematology,

and AJW contributed wet clinical diagnosis from the UCL cohort of

University College London NHS Foundation Trust, London, UK; UCL

patients. MPM contributed clinical diagnosis from the Perugia cohort of

Cancer Institute, University College London, London, UK and Sir

patients. AA, GM, SP, VMP, SA, AR, AC, IP and MRJ performed

Francis Crick Institute, London, UK

pathological examination of AML bone marrow diagnostic specimens.

AM, BF and TM wrote the paper with critical input from all the authors. #

BF and TM contributed equally as senior authors.

AR, CLC and BF edited the paper. AM, BF and TM directed the study.

Correspondence:

Funding

T. MARAFIOTI - t.marafioti@ucl.ac.uk

This work was supported by the UK National Institute of Health Research University College London Hospital Biomedical Research

https://doi.org/10.3324/haematol.2023.283154

Center.

Received: March 16, 2023.

Data-sharing statement

Accepted: June 19, 2023.

Original data are available upon email request to the corresponding

Early view: June 29, 2023.

author.

References 1. Maute R, Xu J, Weissman IL. CD47-SIRPα-targeted therapeutics: status and prospects. Immunooncol Technol. 2022;13:100070. 2. Jaiswal S, Jamieson CH, Pang WW, et al. CD47 is upregulated on circulating hematopoietic stem cells and leukemia cells to avoid phagocytosis. Cell. 2009;138(2):271-285. 3. Majeti R, Chao MP, Alizadeh AA, et al. CD47 is an adverse prognostic factor and therapeutic antibody target on human acute myeloid leukemia stem cells. Cell. 2009;138(2):286-299. 4. Sallman D, Asch A, Kambhampati S, et al. AML-196: the first-inclass anti-CD47 antibody magrolimab in combination with azacitidine is well tolerated and effective in AML patients: phase 1b results. Clin Lymphoma Myeloma Leuk. 2021;21:S290 5. Daver NG, Vyas P, Kambhampati S, et al. Tolerability and efficacy of the first-in-class anti-CD47 antibody magrolimab combined with azacitidine in frontline TP53m AML patients: phase 1b results. J Clin Oncol. 2022;40(Suppl 16):S7020. 6. Tyner JW, Tognon CE, Bottomly D, et al. Functional genomic landscape of acute myeloid leukaemia. Nature. 2018;562(7728):526-531. 7. Liu J, Wang L, Zhao F, et al. Pre-clinical development of a

humanized anti-CD47 antibody with anti-cancer therapeutic potential. PLoS One. 2015;10(9):e0137345. 8. Marafioti T, Jones M, Facchetti F, et al. Phenotype and genotype of interfollicular large B cells, a subpopulation of lymphocytes often with dendritic morphology. Blood. 2003;102(8):2868-2876. 9. Galli S, Zlobec I, Schürch C, Perren A, Ochsenbein AF, Banz Y. CD47 protein expression in acute myeloid leukemia: A tissue microarray-based analysis. Leuk Res. 2015;39(7):749-756. 10. Boasman K, Chris B, Simmonds MJ, Rinaldi CR. Role of prophagocytic calreticulin and anti-phagocytic CD47 in MDS and MPN models treated with azacytidine or ruxolitinib. Haematologica. 2017;102(s2):763. 11. Feng D, Gip P, Mckenna KM, et al. Combination treatment with 5F9 and azacytidine enhances phagocytic elimination of acute myeloid leukemia. Blood. 2018;132(Suppl 1):S2729. 12. Jia Y, Zhang Q, Weng C, et al. Combined blockade of CD47-Sirpa interaction by 5F9 (magrolimab) and azacytidine/venetoclax therapy facilitates macrophage-mediated anti-leukemia efficacy in AML pre-clinical models. Blood. 2021;138(Suppl 1):S510.

Haematologica | 108 December 2023

3495

LETTER TO THE EDITOR

Co-stimulatory and immune checkpoint molecules are important in the tumor microenvironment of Hodgkinlike adult T-cell leukemia/lymphoma Classic Hodgkin lymphoma (CHL) harbors CD30+ Hodgkin and Reed–Sternberg (HRS) cells with occasional EpsteinBarr virus (EBV) infection among numerous non-neoplastic lymphocytes. CD4+ T cells, mainly composed of helper type 2 T (Th2) and regulatory T cells (Treg), surround HRS cells and form T-cell rosettes.1 CD4+ T-cell rosettes variably express immune-suppressive immune checkpoint molecules, including programmed cell death protein-1 (PD-1), cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), lymphocyte activation gene 3 (LAG-3), and T-cell immunoreceptor with immunoglobulin and ITIM domain (TIGIT).2-5 Interaction between HRS cells and CD4+ T cells via these molecules is considered important in the tumor microenvironment (TME) of CHL. Adult T-cell leukemia/lymphoma (ATLL) is a T-cell malignancy caused by human T-cell lymphotropic virus type I (HTLV-1) and typically exhibits a CD4+ Treg phenotype.6 Hodgkin-like ATLL is a rare subtype of ATLL that histologically mimics CHL. CD30+ HRS-like cells with occasional EBV infection are scattered among small to medium monoclonal CD4+ T cells infected with HTLV-1.6,7 Hodgkinlike ATLL exhibits mild to moderate atypia in CD4+ T cells and could be confused with CHL occurring in HTLV-1 carriers; however, patients with Hodgkin-like ATLL show a worse prognosis than those with CHL.7 The TME of Hodgkin-like ATLL remains unclear. Here, we aimed to elucidate the interaction between HRS-like and CD4+ T cells using digital spatial profiling (DSP). DSP involves comprehensive gene expression analysis in specific cells or areas in formalin-fixed, paraffin-embedded (FFPE) slides and provides information on the location of HRSlike cells and CD4+ cells based on their gene expression profiles. We compared gene expression profiles between CD4+ T-cell rosettes and those away from HRS-like cells using DSP in four patients with Hodgkin-like ATLL, with the aim of providing new insights into the TME of this condition and identifying novel target candidates for treatment. Biopsied lymph nodes from 11 patients with Hodgkin-like ATLL were reviewed. All patients were newly diagnosed at Kurume University between 2006 and 2020 based on the latest World Health Organization classification. The study was approved by the Research Ethics Committee of Kurume University and conducted in accordance with the guidelines of the Declaration of Helsinki. An opt-out method was applied for this retrospective study.

DSP of patients 1, 2, 8, and 9 was performed using a GeoMx digital spatial profiler (NanoString Technologies, Seattle, WA, USA). As summarized in Figure 1A, three areas of interest (AOI) were designed for each patient: CD30+ HRSlike cells, CD4+ cells located within 20 μm from HRS-like cells (CD4+ T-cell rosettes), and the remaining CD4+ cells separated from the HRS-like cells (other CD4+ T cells). Immunofluorescence for CD30, CD4, and SYTO 83 was performed on FFPE slides to select AOI (Figure 1B; Online Supplementary Table S1). A barcoded RNA probe mix (Whole Transcriptome Atlas, NanoString Technologies) was hybridized on the FFPE slides, and oligos were then separately cleaved from each AOI via ultraviolet exposure (Figure 1C–E). A sequencing library constructed from the obtained oligos was paired-end sequenced using the NovaSeq 6000 instrument (Illumina, San Diego, CA, USA). The data were analyzed using GeoMx DSP Control Center software (version 2.5.1.145; NanoString Technologies). Q3 normalization was performed according to the manufacturer’s instructions. Differential gene expression was analyzed using a linear mixed model test with Benjamini–Hochberg correction. Results with P<0.05 were considered statistically significant. Immunohistochemistry (IHC) was performed using a DAKO Link autostainer (DAKO, Glostrup, Denmark). Antibodies against CD30, CD4, paired box 5 (PAX5), major histocompatibility complex (MHC) class I and class II, CD28, inducible T-cell co-stimulator (ICOS), TIGIT, PD-1, CTLA-4, LAG-3, CD80, and CD86 were used (Online Supplementary Table S1). EBV-encoded small RNA (EBER) in situ hybridization was performed.7 The expression of markers on HRSlike cells was considered positive if the markers were expressed on ≥50% of the cells. The expression of TIGIT, CD28, ICOS, PD-1, and CTLA-4 in CD4+ T cells was evaluated using the TIGIT scoring system (Online Supplementary Figure S1).8 Cells with score 3 were considered positive. The patient characteristics are listed in Table 1. All patients tested positive for HTLV-1 antibodies and showed rearrangement of T-cell receptor γ. Infection was confirmed in the biopsied lymph nodes using in situ hybridization of HTLV-1 bZIP factor (HBZ) with/without Southern blotting of HTLV-1 in all patients. All HRS-like cells expressed CD30. EBER expression was positive in six cases (54.5%), and PAX5 expression was positive in seven cases (63.6%). MHC classes I and II were expressed in eight (72.7%) and five (45.5%) cases, respectively.

Haematologica | 108 December 2023

3496

LETTER TO THE EDITOR

Figure 1. Digital spatial profiling of Hodgkin-like adult T-cell leukemia/lymphoma. (A) A schema of digital spatial profiling (DSP) in Hodgkin-like adult T-cell leukemia/lymphoma (ATLL). CD4+ cells within 20 µm from CD30+ Hodgkin and Reed–Sternberg (HRS)like cells were defined as CD4+ T-cell rosettes (within the blue circle). The remaining CD4+ cells (other CD4+ T cells) were selected to compare with CD4+ T-cell rosettes. (B) Immunofluorescence-based detection of CD30+ HRS-like cells (red) and CD4+ cells (green). (C-E) Each area was exposed to ultraviolet radiation to cleave the oligos. (C) CD30+ HRS-like cells. (D) CD4+ T-cell rosettes. (E) Other CD4+ T cells. (F-K) Differential gene expression analyses comparing CD4+ T cells. Gene expression of (F) CD28, (G) ICOS, (H) PD-1, (I) CTLA-4, (J) LAG-3, and (K) TIGIT is shown. ICOS: inducible T-cell co-stimulator; PD-1: programmed cell death protein-1; CTLA-4: cytotoxic T lymphocyte-associated protein 4; LAG-3: lymphocyte activation gene 3; TIGIT: T-cell immunoreceptor with immunoglobulin and ITIM domain. *Indicates a significant difference. Haematologica | 108 December 2023

3497

LETTER TO THE EDITOR A total of 1,118 genes were significantly upregulated in CD4+ T-cell rosettes compared with their expression in other CD4+ cells present further away, including those encoding the co-stimulatory molecules CD28 and ICOS (P=0.00340 and P=0.04777, respectively) (Figure 1F, G; Online Supplementary Table S2). The expression of immune checkpoint molecules, including PD-1, CTLA-4, LAG-3, and TIGIT, was not significantly different between the areas (Figure 1H– K). IHC was performed on patient 1–11 (summarized in Table 1). CD30+ HRS-like cells were scattered among small to medium CD4+ T cells (Figure 2A–C). CD4+ T-cell rosettes expressed CD28 in all cases (Figure 2D). ICOS (6/11, 54.5%), TIGIT (7/11, 63.6%), PD-1 (3/11, 27.2%), and CTLA-4 (4/11, 36.3%) were variably expressed in CD4+ T-cell rosettes (Figure 2E–H). A few LAG-3+ cells were detected (data not shown). CD80 (10/11, 90.9%) and CD86 (11/11, 100%), the ligands of CD28, were expressed in HRS-like cells (Figure 2I, J). In the present study, DSP enabled the direct integration of comprehensive gene expression profiling and spatial analyses of CD4+ T cells in Hodgkin-like ATLL. We identified the co-stimulatory molecules, CD28 and ICOS, to be specifically upregulated in CD4+ T-cell rosettes. Our results in-

dicated the interaction between HRS-like cells and CD4+ T cells via distinct co-stimulatory molecules in Hodgkinlike ATLL. CD28 plays a pivotal role in the activation of T cells. ATLL frequently harbors activating alterations in CD28, as well as in genes involved in the T-cell receptor (TCR) pathway.9,10 Yoshida et al.11 reported frequent CD28 fusion genes in young patients with ATLL. They indicated that the interaction between the CD28-fusion protein and its ligands CD80 and CD86 could induce early progression of ATLL via constitutive T-cell activation.11 Sakamoto et al.12 reported various CD28 alterations in 33% of patients with ATLL, and the patients harboring CD28 alterations, especially the smoldering or chronic type, were refractory to chemotherapy and showed a worse prognosis than those without CD28 alterations.12 We previously reported that patients with Hodgkin-like ATLL showed a poor response to conventional therapies and a poor prognosis.7 The interaction between CD28 on HTLV-1-infected CD4+ T cells and CD80/CD86 on HRS-like cells may constitutively activate CD4+ T cells, which might induce disease progression and poor prognosis in patients with Hodgkin-like ATLL (Figure 2K). ICOS is another co-stimulatory molecule that is induced on various T cells upon TCR ligation and CD28 co-

Table 1. Clinicopathological characteristics of patients with Hodgkin-like adult T-cell leukemia/lymphoma. HRS-like cells

Age in years

Sex

Patients

CD4+ T-cell rosettes

MHC MHC class I class II

CD30

EBER

PAX5

CD80

CD86

CD28

ICOS

TIGIT

PD-1

CTLA-4

Patient background, immunohistochemistry for Hodgkin and Reed–Sternberg (HRS)-like cells, and immunohistochemistry for surrounding CD4+ cells are summarized. EBER: Epstein Barr virus-encoded small RNA; PAX5: paired box 5; MHC: major histocompatibility; ICOS: inducible T-cell stimulator; TIGIT: T-cell immunoglobulin and ITIM domain; PD-1: programmed cell death-1; CTLA-4: cytotoxic T lymphocyte-associated protein 4; F: female; M: male. Haematologica | 108 December 2023

3498

LETTER TO THE EDITOR

Figure 2. Co-stimulatory and immune checkpoint molecules in Hodgkin-like adult T-cell leukemia/lymphoma. (A-C) CD30+ Hodgkin and Reed–Sternberg (HRS)-like cells scattered among small to medium CD4+ lymphocytes. (A) Hematoxylin and eosin, (B) CD30, (C) CD4. (D-G) CD4+ T-cell rosettes variably express co-stimulatory and immune checkpoint molecules: (D) CD28, (E) inducible T-cell co-stimulator (ICOS), (F) programmed cell death protein-1 (PD-1), (G) T-cell immunoreceptor with immunoglobulin and ITIM domain (TIGIT), and (H) cytotoxic T lymphocyte-associated protein (CTLA-4). (I) HRS-like cells express CD80. (J) HRSlike cells expressing CD86. Microscope, Olympus BX53; original magnification, 400× for all images; scale bar, 20 μm; camera, Olympus DP53. (K) A schema of the tumor microenvironment (TME) of Hodgkin-like ATLL. HRS-like cells express CD80/CD86 and major histocompatibility complex (MHC) class II that interact with CD28 and T-cell receptor (TCR) expressed on human T-cell lymphotropic virus type I (HTLV-1)-infected CD4+ T cells. Although MHC class II expression can be lost or decreased in HRS-like cells, gene alteration in the TCR pathway may activate TCR signaling. Constitutive T-cell activation by CD28-CD80/CD86 interaction as well as TCR signaling might induce disease progression in Hodgkin-like ATLL.

stimulatory signal. ICOS could also be important for T-cell activation in Hodgkin-like ATLL. Anti-CD28 and ICOS dual antagonists developed for auto-immune diseases might inhibit constitutive T-cell activation in Hodgkin-like ATLL.13 However, blocking co-stimulatory molecules might also disrupt anti-tumor immunity and induce disease progression. Our findings also indicated that CD4+ T cells might interact with HRS-like cells via immune-suppressive immune checkpoint molecules, including TIGIT, PD-1, and CTLA-4, and contribute to an immunosuppressive TME around HRS-like cells. The TME of Hodgkin-like ATLL might be par-

tially similar to that of CHL, which frequently involves CD4+ T cells expressing co-stimulatory and immune-suppressive immune checkpoint molecules around HRS cells. PD-1 blockade was highly effective for patients with relapsed/refractory CHL.14 Thus, immune checkpoint inhibitors might be candidates for new therapies in some patients with Hodgkin-like ATLL. Two types of HRS-like cells have been reported: PAX5+ HRS-like cells derived from polyclonal B cells with occasional EBV infection and PAX5- HRS-like cells showing uncertain cell lineages.7,15 Given the limited number of relevant studies, it remains unclear whether these cell

Haematologica | 108 December 2023

3499

LETTER TO THE EDITOR types should be distinguished as different entities. In this study, the phenotype of CD4+ T cells was dependent on neither PAX5 expression nor EBV infection in HRS-like cells. Further studies are required to elucidate the pathogenesis of Hodgkin-like ATLL. This study has some limitations. First, as Hodgkin-like ATLL is a rare disease and old samples are not recommended for DSP, only four samples were available for DSP. DSP could not extract immune checkpoint molecules as significantly upregulated genes because half of the patients (patients 8 and 9) turned out to be negative for immune checkpoint molecules by IHC. Therefore, additional DSP studies with larger cohorts are necessary to confirm our findings. Second, DSP was not performed at a singlecell resolution. We could not analyze the expression pattern of co-stimulatory and immune checkpoint molecules on each CD4+ cell. In conclusion, we presented, for the first time, distinct CD4+ T cells expressing co-stimulatory and immune checkpoint molecules in the TME of Hodgkin-like ATLL, indicating the interaction between CD4+ T cells and HRS-like cells via these molecules. Our findings provide new insights into the TME of Hodgkin-like ATLL and might pave way for the development of new therapies targeting these molecules.

Correspondence: K. OHSHIMA - ohshima_kouichi@med.kurume-u.ac.jp https://doi.org/10.3324/haematol.2023.283163 Received: March 22, 2023. Accepted: June 30, 2023. Early view: July 13, 2023. ©2023 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures No conflicts of interest to disclose. Contributions MT, HM and KO developed the concept and design of the study. MT, KN, KY, TF, MM, YO, KT, TI, FA and KK acquired data. MT, HM, YS, KS and KO analyzed and interpreted data. MT, HM and KO wrote and reviewed the manuscript. All authors approved the final manuscript. Acknowledgments We thank Mayumi Miura, Kanoko Miyazaki, Akiko Sumi, and Chie Kuroki for their technical assistance. We also thank Jin Katayama, Takashi Uematsu, and Jingjing Gong (NanoString Technologies) for their assistance with the GeoMx analysis. The manuscript was edited and formatted by Editage (https://www.editage.com).

Authors

Funding This work was partially supported by Grants-in-Aid from the Japan

Mai Takeuchi, Hiroaki Miyoshi, Yuichiro Semba, Kyohei Yamada,

Society for the Promotion of Science (KAKENHI)—grants 20K07381 (to

Kazutaka Nakashima, Kensaku Sato, Takuya Furuta, Mayuko

KO), JP20K16206 (to MT), JP22K06950 (to MT), and 20K16208 (to KY).

Moritsubo,1 Yusuke Ogura,1 Ken Tanaka,1 Teppei Imamoto,1 Fumiko Data-sharing statement

Arakawa,1 Kei Kohno1 and Koichi Ohshima1

The DSP data file has been deposited in the Gene Expression Omnibus Department of Pathology, Kurume University School of Medicine and

(https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE213461). Fur-

Department of Medicine and Biosystemic Science, Kyusyu University

thermore, the data file is available on request from the corresponding

Faculty of Medicine, Fukuoka city, Fukuoka, Japan

author.

References 1. Weniger MA, Küppers R. Molecular biology of Hodgkin lymphoma. Leukemia. 2021;35(4):968-981. 2. Carey CD, Gusenleitner D, Lipschitz M, et al. Topological analysis reveals a PD-L1-associated microenvironmental niche for ReedSternberg cells in Hodgkin lymphoma. Blood. 2017;130(22):2420-2430. 3. Patel SS, Weirather JL, Lipschitz M, et al. The microenvironmental niche in classic Hodgkin lymphoma is enriched for CTLA-4-positive T cells that are PD-1-negative. Blood. 2019;134(23):2059-2069. 4. Aoki T, Chong LC, Takata K, et al. Single-cell transcriptome analysis reveals disease-defining T-cell subsets in the tumor microenvironment of classic Hodgkin lymphoma. Cancer Discov.

2020;10(3):406-421. 5. Li W, Blessin NC, Simon R, et al. Expression of the immune checkpoint receptor TIGIT in Hodgkin’s lymphoma. BMC Cancer. 2018;18(1):1209. 6. Ohshima K, Kikuchi M, Yoshida T, Masuda Y, Kimura N. Lymph nodes in incipient adult T-cell leukemia-lymphoma with Hodgkin's disease-like histologic features. Cancer. 1991;67(6):1622-1628. 7. Ohshima K, Suzumiya J, Kato A, Tashiro K, Kikuchi M. Clonal HTLV-1-infected CD4+ T-lymphocytes and non-clonal nonHTLV-1-infected giant cells in incipient ATLL with Hodgkin-like histologic features. Int J Cancer. 1997;72(4):592-598. 8. Annibali O, Bianchi A, Grifoni A, et al. A novel scoring system for

Haematologica | 108 December 2023

3500

LETTER TO THE EDITOR TIGIT expression in classic Hodgkin lymphoma. Sci Rep. 2021;11(1):7059. 9. Vallois D, Dupuy A, Lemonnier F, et al. RNA fusions involving CD28 are rare in peripheral T-cell lymphomas and concentrate mainly in those derived from follicular helper T cells. Haematologica. 2018;103(8):e360-e363. 10. Kataoka K, Nagata Y, Kitanaka A, et al. Integrated molecular analysis of adult T cell leukemia/lymphoma. Nat Genet. 2015;47(11):1304-1315. 11. Yoshida N, Shigemori K, Donaldson N, et al. Genomic landscape of young ATLL patients identifies frequent targetable CD28 fusions. Blood. 2020;135(17):1467-1471. 12. Sakamoto Y, Ishida T, Masaki A, et al. Clinical significance of

CD28 gene-related activating alterations in adult T-cell leukaemia/lymphoma. Br J Haematol. 2021;192(2):281-291. 13. Yang J, Lickliter JD, Hillson JL, et al. First-in-human study of the safety, tolerability, pharmacokinetics, and pharmacodynamics of ALPN-101, a dual CD28/ICOS antagonist, in healthy adult subjects. Clin Transl Sci. 2021;14(4):1314-1326. 14. Ansell SM, Lesokhin AM, Borrello I et al. PD-1 blockade with Nivolumab in relapsed or refractory Hodgkin’s Lymphoma. N Engl J Med. 2015;372(4):311-319. 15. Karube K, Takatori M, Sakihama S, et al. Clinicopathological features of adult T-cell leukemia/lymphoma with HTLV-1infected Hodgkin and Reed-Sternberg-like cells. Blood Adv. 2021;5(1):198-206.

Haematologica | 108 December 2023

3501

LETTER TO THE EDITOR

Higher cyclophosphamide dose grants optimal stem cell collection after daratumumab-based induction in multiple myeloma Quadruplet induction with daratumumab, bortezomib, thalidomide and dexamethasone (Dara-VTd) followed by autologous stem cell transplantation (ASCT) and Dara-VTd consolidation has become the standard treatment for transplant-eligible newly diagnosed multiple myeloma patients (NDMM) in Europe.1 Both improved response rates and progression-free survival (PFS) advantage were reported in the Dara-VTd arm of the CASSIOPEIA trial.2 Nonetheless, concerns emerged with stem cell mobilization and collection after daratumumab.3 Following cyclophosphamide 2 to 3 g/m2 and granulocyte colony-stimulating factor (G-CSF) 10 μg/kg/day, patients treated with daratumumab experienced greater use of plerixafor, longer leukapheresis and lower total number of CD34+ cells/kg collected per patient. Although the proportion of patients undergoing ASCT and obtaining hematopoietic reconstitution was similar between arms, longer intervals to platelet and neutrophil engraftment were observed in those receiving daratumumab.3 Similar results after daratumumab exposure were reported in both GRIFFIN and MASTER study with chemotherapy-free mobilization approach based on G-CSF and plerixafor as well as in recent reallife reports with cyclophosphamide and G-CSF.4–6 We report a collaborative, retrospective analysis of NDMM that consecutively underwent stem cell mobilization and collection after Dara-VTd induction at two Italian centers (Ospedale Santo Spirito, Pescara, and IRCCS Ospedale San Raffaele, Milano). Inclusion criteria were NDMM and eligibility to ASCT. After Dara-VTd induction, patients received high-dose cyclophosphamide (HD-CTX) 4 g/m2 as per institutional practice and were monitored in outpatient setting. G-CSF was administered at a dose of 5 μg/kg/day starting on day 3 to 5 after HD-CTX, with optional increase to 10 μg/kg/day at 48 hours before leukapheresis. Plerixafor 0.24 mg/kg was administered on demand on the day of planned leukapheresis either in patients with <20 CD34+ cells/μL or in those predicted to be poor mobilizer according to subsequent criteria: i) white blood cells count >10x109/L together with CD34+ count <15/µL; ii) ratio of patient body weight (kg) and CD34+/μL >2; iii) yield <25% of total CD34+ target dose on first day of leukapheresis. Preplanned total target dose was 10x106 CD34+ cells/kg to allow for multiple ASCT. Aim of the study was to evaluate safety and efficacy of mobilization with HDCTX 4 g/m2 in terms of stem cell yield and transplantation outcomes. Patients were treated according to current institutional programs upon written informed consent for

transplantation procedures and use of medical records for research. From December 1, 2021 to February 28, 2023, 64 NDMM consecutively received Dara-VTd at our institutions. At data cut-off, 47 patients completed induction and were included in this analysis whereas 17 patients where still receiving induction (Figure 1). At diagnosis median age was 62 years (range, 38-71), 13 patients (28%) were International Staging System (ISS) III, five patients (11%) were Revised ISS (R-ISS III) and 26 patients (55%) were R2-ISS III-IV. Seventeen patients (36%) had high-risk cytogenetic abnormalities as del17p13, t(4;14) and t(14;16). Overall, a median of four Dara-VTd induction cycles (range, 4-6) was administered. Notably, 36 patients (76%) required thalidomide dose reduction, 12 patients (25%) needed bortezomib dose reduction and in two patients (4%) a single daratumumab administration was omitted due to adverse events and/or intolerance. Grade 3-4 adverse events occurred in ten patients (21%) (Figure 1). After Dara-VTd induction, overall response rate was 98%, with 22 patients (47%) in very good partial remission, three patients (6%) in complete remission and 11 patients (23%) in stringent complete remission. One patient progressed soon after the fourth Dara-VTd cycle. After a median of 133 days (range, 113-232) from start of induction and 32 days (range, 16-93) from last daratumumab administration, 46 patients received HD-CTX 4 g/m2. Most frequent grade 3-4 adverse events were anemia (n=8; 17%), neutropenia (n=40; 87%), thrombocytopenia (n=21; 45%), and febrile neutropenia (n=4; 8%) (Figure 1). Notably, only two patients (4%) required platelets transfusion and two patients (4%) discontinued mobilization due to febrile neutropenia. Fourteen patients (30%) doubled G-CSF dose at 48 hours before leukapheresis. After a median of 11 days (range, 916) from HD-CTX, 43 of 46 patients (93%) underwent leukapheresis; 21 of them (49%) received plerixafor. All patients undergoing leukapheresis completed stem cell collection, harvesting a mean total amount of 10.68x106 CD34+ cells/kg (standard deviation [SD] 2.54) (range, 4.9218.8). Notably, at first day of leukapheresis the mean amount of CD34+ cells/kg collected was 6.98x106 (SD 3.60; range, 1.4-17.6) and plerixafor was given in six patients (14%). Mean collection efficiency (CE) was high in our cohort and the majority of patients (n=28/43; 65%) required 2 days of apheresis to reach the collection target of 10x106 CD34+ cells/kg. Details on stem cell mobilization and harvesting are shown in Table 1. Overall, three of 46 patients

Haematologica | 108 December 2023

3502

LETTER TO THE EDITOR (7%) who received HD-CTX 4 g/m2 and G-CSF did not undergo stem cell collection. One patient discontinued mobilization due to concomitant severe acute respiratory syndrome coronavirus 2 infection. Unfortunately, both subsequent rescue attempts with chemotherapy-free GCSF 10 μg/kg/day plus plerixafor and then CTX 2 g/m2 plus G-CSF 10 μg/kg/day plus plerixafor failed. Two patients discontinued mobilization due to febrile neutropenia but were later able to collect enough CD34+ cells/kg for single ASCT with a second procedure (a chemotherapy-free regimen and a bone marrow harvest, respectively). After a median of 195 days from start of induction (range, 164283) and 49 days (range, 33-96) from stem cell collection, 35 of 43 (81%) patients who completed leukapheresis already underwent ASCT at time of data cut-off. Mean number of infused CD34+ cells per patient was 4.84x106/kg (SD 1.20; range, 2.96-9.86). All patient obtained stable neutrophils and platelets engraftments after a median of 12 days (range, 9-14) and 16 days (range, 10-25), respectively. No unexpected toxicities of transplant procedure were reported. At last follow up, all patients are alive.

Recent years have seen the introduction of several novel agents for the treatment of NDMM, with improved response rates and prolonged survival. Nonetheless, ASCT still remains a standard of care for younger fit patients.1,7 A long-standing debate exists about the value of double ASCT in MM: the phase III STaMINA trial showed a PFS benefit in high-risk patients receiving tandem ASCT, whereas in the EMN02/HO95 trial both PFS and overall survival advantage were reported.8–10 To date both EHAESMO and NCCN guidelines still recommend double ASCT as an option for patients who do not achieve at least a very good partial reponse (VGPR) after the first ASCT and those with high-risk features (defined by cytogenetics or clinical characteristics as extramedullary disease).1,7 Notably, ongoing randomized trials still incorporate a double ASCT arm for MRD-defined high-risk patients (MIDAS study; clinicaltrials gov. Identifier: NCT04934475). A second transplantation can be also considered at the time of disease relapse in carefully selected patients.1,7 More recently, cryopreserved autologous CD34+ cells have been used as rescue of prolonged cytopenias after CAR T, further in-

Figure 1. Profile of analyzed population. The diagram shows the distribution of analyzed patients. NDMM: newly-diagnosed multiple myeloma patients; Dara-VTd: daratumumab, bortezomib, thalidomide and dexamethasone; HD-CTX: high-dose cyclophosphamide; ASCT: autologous stem cell transplantation; PNP: peripheral neuropathy.

Haematologica | 108 December 2023

3503

LETTER TO THE EDITOR Table 1. Characteristics of stem cell mobilization and harvesting. Characteristics of stem cell mobilization and harvesting Days from start of induction to HD-CTX, median (range)

133 (113-232)

Days from last daratumumab to HD-CTX, median (range)

32 (16-93)

Days from HD-CTX to first day of leukapheresis, mean [SD] (range)

11.6 [1.65] (9-16)

Total G-CSF μg/kg administered per patient, mean [SD] (range)

440 [128] (240-768)

Peripheral white blood cells/μL on first day of leukapheresis, median (range)

13,400 (1,600-67,000)

Peripheral CD34+ cells/μL on first day of leukapheresis, median (range) Number of days of leukapheresis, N/N (%) 1 2

57 (20-226)

15/43 (35) 28/43 (65)

Days of leukapheresis, mean [SD] (range)

1.7 [0.48] (1-2)

Plerixafor use, N/N (%) Indication <20 CD34+ cells/μL Other

21/43 (49) 14/21 (66) 7/21 (34)

Amount of CD34+ cells x106/kg collected per patient, mean [SD] (range) - Day 1 - Day 2 - Total

6.98 (1.4-17.6) 5.77 (3.2-12.6) 10.68 [2.54] (4.94-18.8)

Collection efficiency %, mean [SD] (range) - Day 1 - Day 2

64 [17] (18-99) 71 [19] (26-100)

Total blood volume processed in liters, mean [SD] (range)

4.58 [2] (1.6-9.6)

HD-CTX: high-dose cyclophosphamide; SD: standard deviation; GCSF: granulocyte colony-stimulating factor.

creasing the need for high collection numbers considering the expected increase in CAR T use in the next future.11,12 Although a minimum dose of 2x106 CD34+ cells/kg is sufficient for a single ASCT, the optimal dose for a timely and stable hematopoietic reconstitution usually ranges between 3x106 and 5x106 CD34+ cells/kg. Therefore, whenever feasible, the collection target should reach 8-10x106 CD34+ cells/kg to allow for multiple ASCT.13,14 Many centers are now implementing chemotherapy-free mobilization protocols, although these are associated with an increased risk of mobilization failure or low stem cell yield among patients exposed to daratumumab.5,6 Cyclophosphamide is the most common chemotherapeutic agent used for mobilization in NDMM, with dosage ranging from 1.5 g/m2 to 4 g/m2. Despite greater toxicities, higher doses are associated with improved stem cell mobilization and harvesting, lower rate of collection failure and improved post-transplantation en-

graftment.14,15 The present analysis represents, to the best of our knowledge, the first report of a mobilization strategy based on HD-CTX 4 g/m2 after daratumumab-based quadruplet induction in NDMM. In our population, HD-CTX 4 g/m2 was administered early after the end of induction and was well tolerated with limited toxicities in an outpatient setting. Notably, the majority of patients successfully underwent leukapheresis after a median of 11 days. In our cohort, we observed a high proportion of poor mobilizers after daratumumab exposure and frequent need for plerixafor to reach the preplanned target dose of 10x106 CD34+ cells/kg. Nonetheless, HD-CTX 4 g/m2 together with patient-tailored plerixafor allowed high CD34+ cells collection in the majority of patients, with a mean total amount of 10.68x106 CD34+ cells/kg. Notably, high collection numbers with limited plerixafor administration were already achieved in the first day of leukapheresis. These results compare favorably to those reported in the Dara-VTd arm of CASSIOPEIA trial, where lower CTX doses (2-3 g/m2) led to collection of a mean of 6.7x106 CD34+ cells/kg.3 We also observed a low incidence of mobilization failure in our cohort. Patients could safely undergo ASCT, with timely hematological engraftment. While limitations of this study include retrospective non-randomized nature and patients number, multicentricity limits single-center effects and bias in patient management. In conclusion, HD-CTX 4 g/m2 and G-CSF after Dara-VTd induction in NDMM proved feasible in the outpatient setting and effective in terms of stem cell mobilization. Increased dose of CTX together with on-demand administration of plerixafor allowed high collection numbers of CD34+ cells per patient, thus limiting the potential detrimental effects of daratumumab on leukapheresis procedures and ensuring sufficient stem-cells for multiple ASCT and rescue of CAR T-related cytopenia.

Authors Carmine Liberatore,1* Tommaso Perini,2,3* Cecilia Passeri,4 Valeria Ferla,2 Francesca Fioritoni,1 Virginia Girlando,2,5 Ornella Iuliani,4 Alessia Orsini,6 Guido Montanaro,1 Francesca Farina,2 Alessandro Di Nicola,7 Rosamaria Nitti,2,5,6 Stefano Pulini,1 Sara Mastaglio,2 Stella Santarone,1 Milena Coppola,6 Sarah Marktel,2 Patrizia Accorsi,4 Fabio Ciceri,2,5 Magda Marcatti2# and Mauro Di Ianni1,7# Hematology Unit, Department of Oncology and Hematology,

Ospedale Santo Spirito, Pescara; 2Hematology and Bone Marrow Transplant Unit, IRCCS Ospedale San Raffaele, Milan; 3Age Related Diseases Laboratory, Division of Genetics and Cell Biology, IRCCS Ospedale San Raffaele, Milan; 4Blood Bank Unit, Department of Oncology and Hematology, Ospedale Santo Spirito, Pescara; 5VitaSalute San Raffaele University, Milan; 6Immunohematology and

Haematologica | 108 December 2023

3504

LETTER TO THE EDITOR Transfusion Medicine Unit, IRCCS Ospedale San Raffaele, Milan and

Department of Medicine and Sciences of Aging, University of Chieti-

Published under a CC BY-NC license

Pescara, Chieti, Italy Disclosures CL and TP contributed equally as first authors.

No conflicts of interest to disclose.

MM and MDI contributed equally as senior authors.

Contributions Correspondence:

All authors contributed to patients’ clinical care. CL and TP

C. LIBERATORE - carmine.liberatore@asl.pe.it

wrote the manuscript. MM and MDI revised the manuscript. All

T. PERINI - perini.tommaso@hsr.it

authors have read and agreed to the published version of the manuscript.

https://doi.org/10.3324/haematol.2023.283452 Data-sharing statement Received: May 2, 2023.

Original data are available in anonymous form upon request by

Accepted: June 20, 2023.

contacting corresponding author.

Early view: July 13, 2023.

References 1. Dimopoulos MA, Moreau P, Terpos E, et al. Multiple myeloma: EHA-ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2021;32(3):309-322. 2. Moreau P, Attal M, Hulin C, et al. Bortezomib, thalidomide, and dexamethasone with or without daratumumab before and after autologous stem-cell transplantation for newly diagnosed multiple myeloma (CASSIOPEIA): a randomised, open-label, phase 3 study. Lancet. 2019;394(10192):29-38. 3. Hulin C, Offner F, Moreau P, et al. Stem cell yield and transplantation in transplant-eligible newly diagnosed multiple myeloma patients receiving daratumumab plus bortezomib/thalidomide/dexamethasone in the phase III CASSIOPEIA study. Haematologica. 2021;106(8):2257-2260. 4. Lemonakis K, Tätting L, Lisak M, et al. Impact of daratumumabbased induction on stem cell collection parameters in Swedish myeloma patients. Haematologica. 2023;108(2):610-614. 5. Eleutherakis Papaiakovou E, Terpos E, Kanellias N, et al. Impact of daratumumab-containing induction on stem cell mobilization and collection, engraftment and hospitalization parameters among multiple myeloma patients undergoing autologous stem cell transplantation. Blood. 2021;138(Suppl 1):S3886. 6. Chhabra S, Callander N, Watts NL, et al. Stem cell mobilization yields with daratumumab- and lenalidomide-containing quadruplet induction therapy in newly diagnosed multiple myeloma: findings from the MASTER and GRIFFIN trials. Transplant Cell Ther. 2023;29(3):174. 7. Kumar SK, Anderson LD, Baljevic M, et al. NCCN Guidelines Version 3.2023 Multiple Myeloma. 8. Hari P, Pasquini MC, Stadtmauer EA, et al. Long-term follow-up of BMT CTN 0702 (STaMINA) of post autologous hematopoietic cell transplantation strategies in the upfront treatment of multiple myeloma. J Clin Oncol. 2020;38(Suppl 15):S8506. 9. Cavo M, Gay F, Beksac M, et al. Autologous haematopoietic stemcell transplantation versus bortezomib–melphalan–prednisone,

with or without bortezomib–lenalidomide–dexamethasone consolidation therapy, and lenalidomide maintenance for newly diagnosed multiple myeloma (EMN02/HO95): a multicentre, randomised, open-label, phase 3 study. Lancet Haematol. 2020;7(6):e456-e468. 10. Gagelmann N, Eikema DJ, Koster L, et al. Tandem autologous stem cell transplantation improves outcomes in newly diagnosed multiple myeloma with extramedullary disease and high-risk cytogenetics: a study from the Chronic Malignancies Working Party of the European Society for Blood and Marrow Transplantation. Biol Blood Marrow Transplant. 2019;25(11):2134-2142. 11. Yan L, Shang J, Shi X, et al. Successful treatment of marrow failure after CARTs for myeloma by the infusion of cryopreserved stem cells. Am J Hematol. 2020;95(1):E20-E23. 12. Gödel P, Sieg N, Heger JM, et al. Hematologic rescue of CAR Tcell-mediated prolonged pancytopenia using autologous peripheral blood hematopoietic stem cells in a lymphoma patient. Hemasphere. 2021;5(3):E545. 13. Giralt S, Stadtmauer EA, Harousseau JL, et al. International myeloma working group (IMWG) consensus statement and guidelines regarding the current status of stem cell collection and high-dose therapy for multiple myeloma and the role of plerixafor (AMD 3100). Leukemia. 2009;23(10):1904-1912. 14. Giralt S, Costa L, Schriber J, et al. Optimizing autologous stem cell mobilization strategies to improve patient outcomes: consensus guidelines and recommendations. Biol Blood Marrow Transplant. 2014;20(3):295-308. 15. Hamadani M, Kochuparambil ST, Osman S, et al. Intermediatedose versus low-dose cyclophosphamide and granulocyte colony-stimulating factor for peripheral blood stem cell mobilization in patients with multiple myeloma treated with novel induction therapies. Biol Blood Marrow Transplant. 2012;18(7):1128-1135.

Haematologica | 108 December 2023

3505

CASE REPORT

Myeloid/lymphoid neoplasms associated with eosinophilia and rearrangements of PCM1::JAK2 with erythroblastic sarcoma: a case report and literature review Translocations and rearrangements involving JAK2 are less common. Among them, the PCM1::JAK2 fusion gene derived from t(8;9)(p22;p24) is the most frequent. In the 2016 World Health Organization (WHO) classification, myeloid/lymphoid neoplasms with eosinophilia (MLN-Eo) and PCM1::JAK2 fusions were defined as new provisional entities, although eosinophilia may be absent in a subset of cases. These neoplasms share common features, such as eosinophilia; large immature erythroid islands, with some lymphoid aggregates; myelofibrosis, usually in combination with splenomegaly; and predominance in males.1 Myeloid sarcoma (MS) is a rare solid tumor formed by primitive or naive myeloid cells in extramedullary sites, mostly secondary to hematologic malignancies. Erythroblastic sarcoma (EBS) is an erythroblastic form of MS, with only a handful of case reports published to date.2 Here, we report an MLN-Eo case with PCM1::JAK2 rearrangements and EBS that was challenging to diagnose and treat. Ethical review and approval were not required for this study on human participants by the local legislation and institutional requirements. Written informed consent to participate in this study was provided by the patient. A 72-year-old male presented with fatigue in April 2021. A complete blood count analysis revealed a white blood cell count of 10.8×109/L, hemoglobin (HGB) level of 90 g/L, platelet (PLT) count of 70×109/L, and eosinophil level of 2.12×109/L. However, he did not consult a hematologist. In January 2022, his fatigue worsened, with a HGB level of 75 g/L. The patient underwent a positron emission tomography/computed tomography (PET/CT) scan, which showed moderate splenomegaly with a mild increase in diffuse fluorodeoxyglucose metabolism. He experienced the first marrow aspiration but could not receive a definite diagnosis. Iron supplementation was given but was ineffective. The HGB level decreased to 59 g/L, and the PLT count decreased to 34×109/L. The patient was referred to our institution. Repeated aspiration and biopsy of the bone marrow indicated myelofibrosis with eosinophilia and leftshifted erythroid predominance (Figure 1A-D). Chromosome analysis revealed a 46,XY,t(8;9)(p22;p24) [16]/46,XY[4] karyotype (Figure 2A). Reverse transcription polymerase chain reaction (RT-PCR) (Figure 2B) and fluorescence in situ hybridization (FISH) assay (Figure 2C, D) showed PCM1::JAK2 gene fusions in the marrow. A 248 gene DNA-based myeloid-targeted next-generation sequencing (NGS) panel was performed

on the marrow with a sequencing depth of 4,544X. Runt-related transcription factor 1 (RUNX1) mutations were found with a variant allele frequency of 8%. The BCR::ABL, MPLW515L/K, CALR, PDGFRA, PDGFRB, and FGFR1 genes were analyzed by RT-PCR, and all results were negative. A biopsy of an enlarged right posterior cervical lymph node was performed. The results revealed neoplastic cells that were positive for erythroid-specific markers such as the transferrin receptor (CD71) and glycophorin 235a (Figure 1E, F) but negative for conventional pan myeloid markers including CD34 and MPO. Therefore, all findings were diagnostic of EBS arising from an MLN-Eo with PCM1::JAK2 rearrangements. The patient declined chemotherapy, involved-field radiotherapy, and JAK2 inhibitors due to the low PLT level. Considering a normal erythropoietin concentration of 7.33 mIU/mL (normal range: 2.59-18.5 mIU/mL) at diagnosis, erythropoietin (at a dose of 10,000 units) was initially administered three times per week. The HGB level gradually increased to 105 g/L on day 165 (day 0 corresponds to erythropoietin initiation) without transfusions. However, the HGB level gradually decreased to 76 g/L on day 312 for unknown reasons. Shortly after, prednisone (30 mg daily) was added in addition to the original erythropoietin treatment and tapered to a maintenance dose of 10 mg daily on day 326. At the last followup (day 370), the HGB level increased to 119 g/L. The PLT and eosinophil counts were maintained at 36×109/L and 2.3×109/L, respectively. The patient is currently under close follow-up and enjoying an excellent performance status. The treatment of erythropoietin and prednisone continues. This case is a unique presentation of an EBS, an exceptionally rare entity. To the best of our knowledge, only a handful of EBS cases have been reported in the English literature so far.2-5 EBS in pediatric patients was manifested as either isolated sarcomas without systemic involvement or with associated erythroid leukemia, while those developed in adults present with myelodysplastic syndrome (MDS), chronic myelomonocytic leukemia (CMML), primary myelofibrosis (MF), MLN with PDGFRA or JAK2 rearrangement, myelodysplastic syndrome/myeloproliferative neoplasm (MDS/MPN), and erythroid leukemia. Due to the small number of cases, it is difficult to predict survival and optimal treatment. Generally, EBS has short-term survival and poor prognosis; in seven of the previously reported cases, the patients died within 6 months of diagnosis. In terms of MLN-Eo with

Haematologica | 108 December 2023

3506

CASE REPORT t(8;9)(p22;p24)/PCM1::JAK2 rearrangement and synchronous MS, only three cases have been reported to date (Table 1).5-7 The treatment was unknown in one case and the other two cases were treated with local radiotherapy and chemotherapy. One of them then received allogenetic stem cell transplantation (ASCT), and the other was awaiting transplantation. There is currently no consensus on the management of sarcoma with MLN. Treatment can be aggressive or expectant based on patients’ characteristics, preferences, and experience of the treating physician. If

aggressive treatment is not initiated upfront, then close follow-up is needed. This case also presented uncommon PCM1::JAK2 rearrangement. PCM1::JAK2 rearrangement affects a wide variety of hematopoietic cell lines and, therefore, can cause both myeloid and lymphoid malignancies, while MPN is the most common among the diseases. These patients had spanned the entire gamut of ages with a median age of 47 years. Nearly 80% of reported cases were male. The reason for the male preponderance is not clear.8 Lierman et al.,

Figure 1. Erythroblastic sarcoma arising in a myeloid neoplasm with PCM1::JAK2 rearrangement: morphology and immunohistochemistry at diagnosis (A) Eosinophils (red arrow) and teardrop erythrocytes (black arrow) were easily seen on the peripheral blood smear (×1,000 magnification). (B) Bone marrow biopsy showed a marked increase in argentophilic reticulin via silver impregnation staining and indicated marked reticulin fibrosis (×400 magnification). (C, D) Marrow biopsy showed numerous clusters of erythroid precursors (black arrow) and predominantly eosinophils (red arrow) via hematoxylin and eosin (H&E) staining (C, ×100 magnification; D, × 400 magnification; D is the magnified image of C). (E, F) Biopsy of the patient’s right posterior cervical lymph node via immunohistochemistry showed sheets of neoplastic cells highlighted by CD235a (E, ×400 magnification) and CD71 (F, ×400 magnification) immunostaining. Haematologica | 108 December 2023

3507

CASE REPORT first reported the use of ruxolitinib, a potent JAK1/2 inhibitor, for chronic eosinophilic leukemia with PCM1::JAK2 rearrangement at an initial dose of 10 mg twice daily, followed by a cautious increase to 15-20 mg twice daily. A

complete cytogenetic response was observed after 15 months without recurrence for at least 36 months.9 Since then, there have been several positive responses to ruxolitinib treatment against such diseases, but the small

Figure 2. Erythroblastic sarcoma arising in a myeloid neoplasm with PCM1::JAK2 rearrangement: genetic features at diagnosis (A) Chromosome analysis of marrow revealed a 46,XY,t(8;9)(p22;p24) karyotype. (B) Reverse transcription polymerase chain reaction revealed the sequence of the chimeric PCM1::JAK2 gene showing in-frame fusion between exon 36 of PCM1 and exon 9 of JAK2. (C) Fluorescence in situ hybridization (FISH) analysis with a JAK2 dual colour break-apart probe showed split signal patterns (1 red, 1 green, and 1 yellow) in 74% cells. (D) FISH analysis with a PCM1 (green)/JAK2 (red) dual-colour dual fusion probe showed a signal consistent with 2 fusions (2 yellow, 1 red, and 1 green) and PCM1::JAK2 rearrangement.

Haematologica | 108 December 2023

3508

CASE REPORT number of patients reported precludes a definitive statement regarding its effect on survival. Further studies are needed to determine how best to utilize ruxolitinib. Therefore, currently, ASCT is still the only way to cure the disease, particularly in the absence of acute leukemia.8 For those patients who are unsuitable for transplantation, the prospect seems very bleak. Some patients with MPN received hydroxyurea or interferon for initial treatment and then followed a strategy of watch and wait.10 Although effective standard treatment strategies have not yet been established for elderly patients who are not candidates for transplantation, here, we illustrate the value of symptom-directed conventional drug therapies in certain settings. The patient initially responded to erythropoietin, although he experienced progressive anemia after 4 months. Low levels of serum erythropoietin at the onset of therapy may be a predictor of the efficacy of erythropoietin. In primary MF patients, prednisone was an option for MFassociated anemia.11 Low-dose prednisone was, thus given, resulting in rapid improvement in overall symptom burden and anemia. Further, enhanced CT examination of cervical lymph nodes didn’t reveal enlargement of the originally involved lesion. This case also indicates that cytogenetic and molecular testing is a crucial part of the workup for hematopoietic neoplasms. In addition to conventional karyotype analysis, thorough evaluation should utilize a targeted sequencing panel to look for mutations that may have a significant impact on the patient’s therapy and likely, prognosis. As hematological neoplasms with PCM1::JAK2 rearrangement are rare, it is difficult to conduct large-scale clinical studies on the molecular characterization. NGS assays detected relatively small RUNX1 mutation clones in the marrow of our

case. Similar to JAK2 and spliceosome genes, RUNX1 showed the highest predictive value for myeloid neoplasms. Notably, mutations in RUNX1 are usually not detected in the prediagnostic phase but are mostly identified in the context of a full-blown myelodysplastic syndrome or acute myelogenous leukemia (AML). For example, RUNX1 is more frequently detected in secondary AML post-CMML than during CMML and is associated with disease progression and poor prognosis.12 Unlike other MLN-Eo subgroups, such as PDGFRA, PDGRFB, and PCM1::JAK2, somatic mutations are common in the FGFR1-rearranged subgroup and frequently involve RUNX1.13-14 A retrospective study involving 61 MLN-Eo cases found that 23% (14/61) of patients carried at least one mutation in addition to WHO-defined genetic changes. Eighty-three percent of cases (5/6) with FGFR1 rearrangement were found with mutations, all of which involved RUNX1. However, mutations were detected in only 14% (1/7) of the PCM1::JAK2 subgroup, which were TET2.15 Of note, TET2 is also mutated in older individuals without a hematologic malignancy, which was recently described as clonal hematopoiesis of indeterminate potential. Therefore, the role of RUNX1 mutations in the occurrence, progression, and prognosis of our PCM1::JAK2 fusion case remains uncertain. In conclusion, we reported a novel case of PCM1::JAK2 fusion mutation-related MLN-Eo with EBS. It is important that cytogenetics, sometimes FISH, PCR, and/or NGS, should be completed at the time of initial diagnosis. To date, ASCT is the only potentially curative treatment. Prospective studies are required to confirm whether ruxolitinib could be a first-line treatment option for this rare disease, especially for those who are not eligible for transplantations. Given the limitations of the data, at this point in time, supportive treatment and symptomatic care may be

Table 1. Clinical course of patients with t(8;9)(p22;p24)/PCM1::JAK2 rearrangement accompanying myeloid sarcoma.

Reference

Age, yrs Sex

Type

Biopsy site

Initial diagnosis

Right ileo-psoas MF, B-ALL muscle

Treatment

ASCT

Two cycles of FLAI regimen; a cycle of high-dose methotrexate and cytarabine, ASCT also for CNS prophylaxis; after CR local radiotherapy; ASCT; ruxolitinib as post-transplant maintenance

Rizzuto G, et al. 20226

53 F

Luedke C, et al. 20205

32 M

EBS

Axillary LN

Myeloid neoplasm

Song I, et al. 20167

42 M

Right axillary LN and an inguinal LN

MPN-U

Chemotherapy with arabinoside and erythromycin

Survival (mths) (from diagnosis Response of MS to the last follow-up reported)

Awaiting Unknown

F: female; M: male; MS: myeloid sarcoma; EBS: erythroblastic sarcoma; LN: lymph node; MF: myelofibrosis; B-ALL: B-cell acute lymphoblastic leukemia; MPN-U: myeloproliferative neoplasm, unclassifiable; FLAI: Fludarabine + Cytarabine + Idarubicine; CNS: central nervous system; ASCT: allogeneic stem cell transplantation; NA: not available; CR: complete remission; yrs: years; mths: months. Haematologica | 108 December 2023

3509

CASE REPORT considered for non-transplant candidates with relatively stable diseases.

Authors

No conflicts of interest to disclose. Contributions

Lina Zhang, Xiaoqin Zhu, Weiying Qu, Yingjia Lu, Zhou Feng and

LZ, LNZ, and XQZ gathered the clinical information and drafted the

Lin Zhao2

manuscript. LZ and LNZ revised the paper. LZ approved the final

diagnosis and formulated the therapeutic strategies. WYQ, YJL, and Department of Biochemistry, Institute of Basic Medicine, Shanghai

ZF reviewed multiple drafts of the manuscript. All authors listed

University of Traditional Chinese Medicine and Department of

have made a substantial, direct, and intellectual contribution to the

Hematology, Shuguang Hospital Affiliated to Shanghai University of

work and approved it for publication.

Traditional Chinese Medicine, Shanghai, China Acknowledgements Correspondence:

This investigation was supported by the grant 2022 NO.178 to Lin

L. Z. ZHAO - sg1520@shutcm.edu.cn

Zhao from the fifth batch of national training programs for clinical talents of Traditional Chinese Medicine, China.

https://doi.org/10.3324/haematol.2022.282228 Data-sharing statement Received: October 24, 2022.

The datasets used and/or analyzed during the current study are

Accepted: April 20, 2023.

available from the corresponding author on reasonable request.

Early view: April 27, 2023.

References 1. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood. 2016;127(20):2391-2405. 2. King RL, Siaghani PJ, Wong K, et al. Novel t(1;8)(p31.3;q21.3) NFIA-RUNX1T1 translocation in an infant erythroblastic sarcoma. Am J Clin Pathol. 2021;156(1):129-138. 3. Tomlinson B, Raess PW, Beck RC, et al. Erythroblastic sarcoma with rare PDGFRA rearrangement responding to tyrosine kinase inhibitor therapy. Br J Haematol. 2020;188(6):e98-e100. 4. Nozuchi N, Nakaya Y, Ishii N, et al. Erythroblastic sarcoma coexistent with myelodysplastic syndrome. Pathol Int. 2019;69(8):505-507. 5. Luedke C, Rein L. Transformation to erythroblastic sarcoma from myeloid neoplasm with PCM1-JAK2. Blood. 2020;136(9):1113. 6. Rizzuto G, Leoncin M, Imbergamo S, et al. Sequential allogeneic transplantation and ruxolitinib maintenance for a synchronous PCM1-JAK2 positive myeloid sarcoma and acute Blymphoblastic leukemia. Clin Case Rep. 2022;10(1):e05212. 7. Song I, Lee DH, Lee JH, et al. A t(8;9)(p22;p24)/PCM1-JAK2 translocation in a patient with myeloproliferative neoplasm and myeloid sarcoma: first report in Korea. Ann Lab Med. 2016;36(1):79-81. 8. Kaplan HG, Jin R, Bifulco CB, et al. PCM1-JAK2 fusion tyrosine kinase gene-related neoplasia: a systematic review of the clinical literature. Oncologist. 2022;27(8):e661-e670. 9. Lierman E, Selleslag D, Smits S, et al. Ruxolitinib inhibits

transforming JAK2 fusion proteins In vitro and induces complete cytogenetic remission in t(8;9)(p22;p24)/PCM1-JAK2– positive chronic eosinophilic leukemia. Blood. 2012;120(7):1529-1531. 10. Tang GL, Philip JKSS, Weinberg O, et al. Hematopoietic neoplasms with 9p24/JAK2 rearrangement: a multicenter study. Mod Pathol. 2019;32(4):490-498. 11. Tefferi A. Primary myelofibrosis: 2021 update on diagnosis, riskstratification and management. Am J Hematol. 2021;96(1):145-162. 12. Kuo MC, Liang DC, Huang CF, et al. RUNX1 mutations are frequent in chronic myelomonocytic leukemia and mutations at the C-terminal region might predict acute myeloid leukemia transformation. Leukemia. 2009;23(8):1426-1431. 13. Pozdnyakova O, Orazi A, Kelemen K, et al. Myeloid/lymphoid neoplasms associated with eosinophilia and rearrangements of PDGFRA, PDGFRB, or FGFR1 or with PCM1-JAK2. Am J Clin Pathol. 2021;155(2):160-178. 14. Kumar KR, Chen W, Koduru PR, et al. Myeloid and lymphoid neoplasm with abnormalities of FGFR1 presenting with trilineage blasts and RUNX1 rearrangement: a case report and review of literature. Am J Clin Pathol. 2015;143(5):738-748. 15. Baer C, Muehlbacher V, Kern W, et al. Molecular genetic characterization of myeloid/ lymphoid neoplasms associated with eosinophilia and rearrangement of PDGFRA, PDGFRB, FGFR1 or PCM1-JAK2. Haematologica. 2018;103(8):e348-e350.

Haematologica | 108 December 2023

3510

CASE REPORT

Myeloid lineage switch following CD7-targeted chimeric antigen receptor T-cell therapy in relapsed/refractory T-cell acute lymphoblastic leukemia Myeloid lineage switch is a rare phenomenon in relapsed B-cell acute lymphoblastic leukemia B-ALL, predominantly arising in cases harboring the KMT2A rearrangement and treated with CD19-targeted immunotherapies including blinatumomab and CD19-targeted chimeric antigen receptor (CAR) T-cell therapy,1,2 with an incidence of 3% following CD19CAR T-cell therapy in one study.3 However, myeloid lineage switch has only very rarely been observed in T-cell ALL (T-ALL),4,5 including one case with KMT2A rearrangment. Nevertheless, the extension of antigen-based targeted immunotherapy to T-ALL may increase the prevalence of this phenomenon as a potential escape mechanism. There has been a growing interest in developing CD7-targeting (CD7) CAR T-cell therapy for relapsed/refractory (r/r) T-ALL as CD7 is expressed in nearly all cases, and early clinical data for CD7CAR T-cell therapy is promising.6,7 The WU-007 CAR T-cell platform is a novel allogeneic offthe-shelf product manufactured from T cells collected from healthy donors. Allogeneic T cells are genetically edited to delete the CD7 antigen and T-cell receptor α chain (TRAC) to prevent fratricide and graft-versus-host disease (GvHD), respectively, and modified cells are subsequently transduced with a CD7CAR.8 Preclinical data supporting the in vitro expansion, antitumor activity and lack of GvHD of WU-007 CAR T cells8 led to the design of an ongoing international multicenter phase I dose escalation study testing WU-007 in patients with r/r T-ALL (clinicaltrials gov. Identifier: NCT04984356), for which our institution is a participating site. Here, we report an unusual case of r/r T-ALL that responded to WU-007 CAR T cells administered at dose level 2 (300x106 CAR T cells), but subsequently relapsed as an acute myeloid leukemia (AML) (myeloid lineage switch) with no evidence of recurrent T-ALL. The patient was a 30-year-old male diagnosed with T-ALL with a normal karyotype. Online Supplementary Figure S1 depicts key time points in this case. He was treated according to the AALL0434 regimen with nelarabine consolidation and achieved complete remission (CR) with negative minimal residual disease (MRD-negative). However, he experienced early hematological relapse during interim maintenance therapy and then received a salvage regimen of nelarabine, cyclophosphamide and etoposide that led to a CR but persistent MRD. A 5-day course of nelarabine was administered as a consolidation therapy

while a fully matched unrelated donor was being activated. Unfortunately, he had evidence of frank hematological relapse with increased circulating blasts in the peripheral blood (PB) 4 weeks later. He was enrolled on the CD7CAR T-cell study (clinicaltrials gov. Identifier: NCT04984356) and received cytoreduction with dexamethasone while undergoing screening studies. Pre-CAR bone marrow (BM) biopsy showed hypercellular marrow with 68% lymphoblasts (Figure 1A). The immunophenotype of the lymphoblasts (Figures 2A) showed positivity for CD1a, CD2, cytoplasmic CD3 (cCD3), CD4 (minor subset), CD5 (minor subset), CD7, CD33 (dim), CD34, CD38, CD45 (moderate), and HLA-DR (heterogenous). Cytogenetics showed a complex karyotype. Next generation sequencing (NGS) of the BM aspirate revealed mutations in TP53, PHF6, and RUNX1 as well as high RNA expression of CDK6 and IL7R amplification (Table 1). He received lymphodepletion with cyclophosphamide and fludarabine, but white blood cell count and blasts in the PB continued to rise after completing lymphodepletion. WU-007 CAR T cells were infused intravenously. He experienced grade 1 cytokine release syndrome on day 1 following infusion that progressed to grade 2 and resolved with administration of tocilizumab and dexamethasone. Exploratory day 7 BM biopsy showed persistent T-ALL (90%) in a hypercellular marrow; T lymphoblasts expressed cCD3, CD2, CD4 (dim, small subset), CD5 (dim, subset), CD7, CD34, CD45 (dim to moderate) and CD123 (moderate, increased). Cytogenetics showed persistent complex karyotype and fluorescence in situ hybridization (FISH) revealed TP53 deletion (Table 1). On day 10 post CAR T cells, he developed pancytopenia and had increased lactate dehydrogenase (LDH), triglyceride, bilirubin, and ferritin as well as decreased fibrinogen levels. These findings were supportive of macrophage activation syndrome, which was successfully treated with dexamethasone and anakinra. Day 28 post infusion BM biopsy showed hypocellular marrow with >80% abnormal immature T-cell population. This abnormal T-cell population expressed CD2, cCD3, and CD4, but had lost expression of CD7 by flow cytometry (Figure 2B). Touch imprint showed that atypical cells were mainly blastoid with high nuclear-to-cytoplasmic ratio, multiple prominent nucleoli, and very scant blue and nongranular cytoplasm, some morphologically indistinguishable from the original T lymphoblasts (Figure 1B, C). Flow cytometry for MRD obtained from the marrow identified

Haematologica | 108 December 2023

3511

CASE REPORT

Figure 1. Cell morphology at different time points. (A) Morphology of blasts pre-chimeric antigen receptor (CAR) T cells. Large-sized blasts with high nuclear/cytpoplasmic (N/C) ratio, finely dispersed to somewhat coarse chromatin with occasional nuclei, and scant lightly basophilic agranular cytoplasm. (B) Morphology of findings post CAR T cells day 28. Pleomorphic cells with variable size double nucleated and irregular nuclear border and cytoplasmic spike. (C) Morphology of findings post CAR T cells day 28. Large and atypical cells with enlarged nuclei and irregular nuclear contours, predominantly with ample amount of cytoplasm. Cells with bizarre nuclear shapes “Reed-Sternberg-like morphology”, depicted in the picture. (D) Morphology of blasts at myeloid linage relapse. Medium-sized blasts with a high N/C ratio, round to irregular nuclear contours, fine chromatin, inconspicuous nucleoli, and scant basophilic cytoplasm.

two T-cell populations, both CD7-. The first population represented 70% of mononuclear cells and had decreased surface CD3 (sCD3), but with normal expression of CD2, cCD3, CD4, CD5, CD38, CD45 and CD48, and absence of CD7, CD16, CD34 and CD56. The second population represented 6% of mononuclear cells and had CD4 predominance (CD4:CD8 ratio of approximately 37) with normal expression of CD2, sCD3, cCD3, CD4, CD5, CD38, CD45 and CD48, and absence of CD7, CD16, CD34 and CD56. These two unusual populations likely represented a combination of CAR T cells and reactive T cells, and there was no definitive immunophenotypic evidence of residual T-cell ALL (Figure 2B). Cytogenetics showed normal karyotype in five metaphases and polymerase chain reaction for TP53 mutation was negative (Table 1). The patient was deemed to have achieved MRD-negative morphological leukemia-free state but continued to have marked pancytopenia. During the time in which the patient’s donor was activated in preparation for transplant, the patient’s LDH levels began to rise, and therefore, we repeated the BM biopsy

on day 35 post CAR T-cell infusion. The results of this biopsy showed a hypercellular bone marrow with 80% involvement by myeloid blasts (Figure 1D). Flow cytometry demonstrated an abnormal blast population expressing CD45 (moderate intensity), CD4 (subset), CD13 (weak intensity, subset), CD33 (weak intensity), CD34 (increased), CD64 (weak intensity), CD123 (moderate intensity), HLADR and MPO that comprised 82% of total cells analyzed (Figure 2C). The blast population was negative for CD1a, CD2, sCD3, cCD3, CD5, CD7, CD8, CD10, CD14, CD16, cCD22, CD24, CD117, TCRα/β, TCRγ/δ and TdT (Table 1). NGS analysis demonstrated identical identified mutations in TP53, RUNX1 and PHF6 in addition to amplification of PIK3CD and BTG1 loss (Table 1). He was treated with salvage of venetoclax and azacitidine, but unfortunately, treatment was complicated by an invasive fungal infection, and he did not respond. In this case, the r/r T-ALL initially responded to CD7CAR T-cell therapy and the patient achieved MRD-negative morphological leukemia-free state on day 28 post infu-

Haematologica | 108 December 2023

3512

CASE REPORT

Continued on following page. Haematologica | 108 December 2023

3513

CASE REPORT

Figure 2. Phenotype of blast populations over time by flow cytometry. (A) Pre-CD7 chimeric antigen receptor (pre-CD7CAR) T cells. All plots show mononuclear cells. Abnormal blasts (0.03% of the white blood cells [WBC]) are highlighted in red and express cytoplasmic CD3, CD7, and CD34 without surface CD3. Normal T cells are shown in green and natural killer cells (NK) cells are shown in blue. CD34+ myeloid blast are shown in orange. (B) Post CD7CAR T cells. All plots show mononuclear cells. No CD7+/CD34+ abnormal blast population is present. T cells are shown in green (surface CD3-, CD4:CD8 ratio of approximately 8:1) and aqua (surface CD3+, almost all CD4+) and lack express of CD7. Essentially no NK cells are present. CD34 blasts are shown in orange, comprise approximately 5% of the WBC and lack expression of all T-cell antigens. At the time of initial evaluation, these were favored to represent early regeneration. (C) Myeloid lineage relapse. The specimen consists largely of CD34+ blasts (red, approximately 75% of the WBC) which express cMPO without cCD3 or CD7. T cells (green) still include a major subset lacking surface CD3 and have a CD4:CD8 ratio of approximately 1:0.6). NK cells remain rare.

sion, but the disease relapsed shortly thereafter and manifested as AML. The shared mutational profile between the original T-ALL and the emergent AML post CD7CAR T-cell therapy supported the notion that the AML originated from the same T-ALL clone as a lineage switch. While CAR T cells effectively eliminated CD7+ T-ALL clones, relapse with lineage switch to myeloid occurred, potentially as an escape mechanism under the pressure of targeting CD7. We postulate that the lineage switch could be the result of either a pre-existing small population of myeloid clones at the time of treatment initiation or due to transformation of early leukemic clones that were less committed to the lymphoid lineage. Indeed, a recent report provides evidence of a significant incidence of myeloid mutations associated with clonal hematopoiesis of indeterminate potential (CHIP) in NGS performed on both B- and T-ALL samples.9 The allogeneic nature of WU007 CD7CAR T cells raises concern that the development

of AML in this case could originate from donor cells as donor-derived leukemia. However, the similarity of the mutational profile for both leukemias makes this scenario unlikely. While myeloid lineage switch has been reported in patients with B-ALL with KMT2A rearrangement treated with CD19targeted immunotherapies,1,2 the occurrence in T-ALL is very rare with only a small number of reported cases.4,5 Whereas the loss of CD7 antigen upon relapse following CD7CAR T-cell therapy was observed in other studies utilizing different CAR T-cell platforms, there have been no reports on myeloid lineage switch with CD7CAR T-cell therapy.6,7 As innovative immunotherapeutics are being extended to T-ALL, the phenomenon of lineage switch may become more prevalent and necessitate additional studies to comprehend which patients are at risk and to introduce novel approaches to avert these outcomes. Our case highlights the potential activity of CD7CAR T-cell

Haematologica | 108 December 2023

3514

CASE REPORT therapy with the WU-007 platform, with the study currently in progress and the dose of WU-007 being escalated per the study design. Nonetheless, our experience exemplifies some challenges and a potential morphologic pitfall in differentiating residual T-ALL from modified CAR prod-

uct and reactive T cells post treatment as we encountered on day 28 post infusion. The cumulative findings such as extended immunophenotype and molecular results supported disease response in this case before lineage switch.

Table 1. Leukemia phenotype, cytogenetics and molecular findings at various time points. Immunophenotype

Mutations and gene expression

Cytogenetics/FISH

Diagnosis

CD2+, sCD3-, cCD3+, CD4-, dimCD5+, CD7+, CD8-, CD10-, 46,XY; FISH showed tetrasomy of CD11b-, CD13-, CD14-, CD16-, chrom 6 and 21 in 8.5%, and 1-2 CD19-, CD20-, CD23-, CD30-, extra signals/copies on chrom 8, 9, CD33-, CD34+, CD38+, CD56-, 11, and 22 in 5-13.5% CD64-, CD117-, HLA-DR+ (24%), MPO-, TdT+, CD123+

1st relapse

CD1a-, CD2+ (partial dim), sCD3-, CD4-, CD5+ (predominantly), CD7+, CD8-, CD10-, CD19-, CD33+, CD34+, CD38+, CD45+ (dim)

85-90<4n>XXYY,add(6)(p23)x2, add(8)(p21)x12, der(11)t(11;14)(p13;q11.2)x2, -2,del(17)(p11.2)x12,+58mar, inc[cp4]; FISH showed 17p deletion

Pre CAR T- cell infusion

CD1a+, CD2+, sCD3-, cCD3+, CD4+ (minor subset), CD5+ (minor subset), CD7+, CD8-, CD33+ 45,XY,add(8)(p?21),add(9)(p?21), (dim), CD34+, CD38+, TP53, RUNX1 and + der(11)t(11;14)(p1?3;q11.2), CD45 (moderate), HLA-DR PHF6 mutations; CDK6 elevated (heterogenous), TdT-. (65% of total -4,del(16)(q22),del(17)(p11.2)[4]; expression and IL7R amplification analyzed cells). On concurrent FISH studies 17p deletion peripheral blood flow CD123 dim, partial expression is seen on 71% of total analyzed cells

Day 7 post CAR T-cell infusion

CD1a-, CD2+, sCD3-, cCD3+, CD4+ (dim, small subset), CD5+ (dim, subset), CD7+, CD8-, CD15-, CD34+, CD45+, CD56-, CD64-, CD123+ (95% of total analyzed cells)

75~94<4n>,XY,-X,-Y,add(1)(q44), 4,-4,-5,-5,add(6)(p23)x2,add(8) (p?21)x2,+9,add(9)(p?21)x2, der(11)t(11;14)(p1?3;q11.2)x2, -12,-12,-14,-14,+del(16)(q22), +17,del(17)(p11.2)x2, -22,+1~5mar[cp16]

Day 28 post CAR T-cell infusion

No definite abnormal T-lymphoid blast population. T cells present showed phenotype of: CD1a-, CD2+, cCD3+, CD4+, CD7-, CD8-, CD45+, TdT(86.6% of total analyzed cells)

46,XY [5]

TP53 mutation was negative by PCR

Day 35 post CAR T-cell infusion

CD1a-, CD2-, sCD3-, cCD3-, CD4+ 45,XY,add(8)(p?21),add(9)(p?21), (minor subset), CD5-, CD7-, t(11;14)(p1?3;q11.2), TdT-, CD34+, MPO+ del(17)(p11.2),-19[16] (70% of the total analyzed cells)

Day 38 post CAR T-cell infusion

CD1a-, CD2-, sCD3-, cCD3-, CD4+ (subset), CD5-, CD7-, CD8-, CD10-, CD13+, CD14-, CD16-, cCD22-, CD24-, CD33+ (weakly), CD34+, CD117-, CD123+, TdT-, HLA-DR+, MPO+

TP53, RUNX1, and PHF6 mutations; BTG1 loss and PIK3CD amplification; PCR for TP53 mutation was positive at 50%

FISH: fluorescence in situ hybridization; ND: not done; PCR: polymerase chain reaction; chrom: chromosome. Haematologica | 108 December 2023

3515

CASE REPORT

Authors

Ibrahim Aldoss,1 Parastou Tizro,2 Davsheen Bedi,3 James K. Mangan,4

Disclosures

Mary C. Clark, David Spencer, Joo Y. Song, Sindhu Cherian, Raju Pillai,

IA discloses consulting fees from Amgen and KiTE Pharma; is part

Young Kim, Nitin Mahajan, Ketevan Gendzekhadze, Mike James,

of the speakers bureau of Amgen; has received financial support for

Kenneth Jacobs, Jan Davidson-Moncada, Stephen J. Forman, Huan-

attending meetings and/or travel from Amgen and KiTE Pharma. DS

You Wang and Michelle Afkhami

discloses consulting fees from Wugen and is a stockholder of

Wugen. KJ discloses participation on data safety monitoring board Department of Hematology and Hematopoietic Cell Transplantation,

and advisory board or Advisory Board and is a stockholder of

Gehr Family Center for Leukemia Research, City of Hope, Duarte, CA;

Wugen, and has other financial or non-financial interests in Wugen.

Department of Pathology, City of Hope, Duarte, CA; Department of

JD-M discloses a leadership role at Wugen and is a stockholder of

Pathology, University of California San Diego, La Jolla, CA; Department

Wugen and MacroGenics. All other authors have no conflicts of

of Medicine, Division of Blood and Marrow Transplantation, Moores

interest to disclose.

Cancer Center, University of California San Diego, La Jolla, CA; Division 5

of Oncology, Department of Medicine, Washington University School of

Contributions

Medicine, St. Louis, MO; 6Department of Laboratory Medicine and

IA and MA designed the study, collected data, performed analyses

Pathology, University of Washington, Seattle, WA; 7Wugen, Saint Louis,

and wrote the paper. IA, PT, DB, JM, JYS, SC, RP, YK, KG, KJ, JD, SJF,

MO and HLA Laboratory, City of Hope, Duarte, CA, USA

HW and MA collected and analyzed data. PT, DS, JYS, SC, RP, YK,

NM, MJ, HW and MA reviewed bone marrow morphology, Correspondence:

cytogenetics and molecular findings. IA and JKM treated the study

I. ALDOSS - ialdoss@coh.org

patient. MCC wrote the paper. All authors reviewed the final draft of the paper.

https://doi.org/10.3324/haematol.2023.283566 Data-sharing statement Received: May 22, 2023.

For original data, please contact ialdoss@coh.org

Accepted: July 7, 2023. Early view: July 20, 2023.

References 1. Aldoss I, Song JY. Extramedullary relapse of KMT2A(MLL)rearranged acute lymphoblastic leukemia with lineage switch following blinatumomab. Blood. 2018;131(22):2507. 2. Gardner R, Wu D, Cherian S, et al. Acquisition of a CD19-negative myeloid phenotype allows immune escape of MLL-rearranged BALL from CD19 CAR-T-cell therapy. Blood. 2016;127(20):2406-2410. 3. Lamble AJ, Myers RM, Taraseviciute A, et al. Preinfusion factors impacting relapse immunophenotype following CD19 CAR T cells. Blood Adv. 2023;7(4):575-585. 4. Aujla A, Hanmantgad M, Islam H, Shakil F, Liu D, Seiter K. Lineage switch from T-cell lymphoblastic leukemia/lymphoma to acute myeloid leukemia and back to T-cell lymphoblastic leukemia/lymphoma in a patient diagnosed during pregnancy. Stem Cell Investig. 2019;6:12. 5. Ittel A, Jeandidier E, Helias C, et al. First description of the t(10;11)(q22;q23)/MLL-TET1 translocation in a T-cell lymphoblastic

lymphoma, with subsequent lineage switch to acute myelomonocytic myeloid leukemia. Haematologica. 2013;98(12):e166-168. 6. Lu P, Liu Y, Yang J, et al. Naturally selected CD7 CAR-T therapy without genetic manipulations for T-ALL/LBL: first-in-human phase 1 clinical trial. Blood. 2022;140(4):321-334. 7. Pan J, Tan Y, Wang G, et al. Donor-derived CD7 chimeric antigen receptor T cells for T-cell acute lymphoblastic leukemia: first-inhuman, phase I trial. J Clin Oncol. 2021;39(30):3340-3351. 8. Cooper ML, Choi J, Staser K, et al. An "off-the-shelf" fratricideresistant CAR-T for the treatment of T cell hematologic malignancies. Leukemia. 2018;32(9):1970-1983. 9. Saygin C, Stauber J, Aldoss I, et al. Molecular characterization of adult acute lymphoblastic leukemia identifies a subgroup with myeloid mutations and pre-existing clonal hematopoiesis. Blood. 2022;140(Suppl 1):S1050-1052.

Haematologica | 108 December 2023

3516

haematologica Journal of the Ferrata Storti Foundation

haematologica.org