Haematologica, Volume 106, Issue 7 by Haematologica

haematologica Journal of the Ferrata Storti Foundation

Editor-in-Chief Jacob M. Rowe (Haifa)

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

Managing Director Antonio Majocchi (Pavia)

Associate Editors Hélène Cavé (Paris), Monika Engelhardt (Freiburg), Steve Lane (Brisbane), Pier Mannuccio Mannucci (Milan), Pavan Reddy (Ann Arbor), David C. Rees (London), Francesco Rodeghiero (Vicenza), Gilles Salles (New York), Kerry Savage (Vancouver), Aaron Schimmer (Toronto), Richard F. Schlenk (Heidelberg), Sonali Smith (Chicago)

Assistant Editors Britta Dorst (English Editor), Catherine Klersy (Statistical Consultant), Rachel Stenner (English Editor)

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), Mary Eapen (Milwaukee), Francesca Gay (Torino), Ajay Gopal (Seattle), Alex Herrera (Duarte), Shai Izraeli (Ramat Gan), 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), Neha Mehta-Shah (St. Louis), Alison Moskowitz (New York), Yishai Ofran (Haifa), Farhad Ravandi (Houston), John W. Semple (Lund), Liran Shlush (Toronto), Sara Tasian (Philadelphia), Pieter van Vlieberghe (Ghent), Ofir Wolach (Haifa), Loic Ysebaert (Toulouse)

Editorial Office Simona Giri (Production & Marketing Manager), Lorella Ripari (Peer Review Manager), Paola Cariati (Senior Graphic Designer), Igor Ebuli Poletti (Senior Graphic Designer), Marta Fossati (Peer Review), Diana Serena Ravera (Peer Review)

Affiliated Scientific Societies SIE (Italian Society of Hematology, www.siematologia.it) SIES (Italian Society of Experimental Hematology, www.siesonline.it)

haematologica Journal of the Ferrata Storti Foundation

Information for readers, authors and subscribers 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, research articles, review articles, guideline articles and letters. 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 online (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. Detailed information about subscriptions is available online at www.haematologica.org. Haematologica is an open access journal. Access to the online journal is free. Use of the Haematologica App (available on the App Store and on Google Play) 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 International edition for the year 2021 are as following: Print edition

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 Journal of the Ferrata Storti Foundation

Table of Contents Volume 106, Issue 7: July 2021 Obituary 1779

Edoardo Ascari, a man to whom Haematologica owes much Carlo Balduini https://doi.org/10.3324/haematol.2021.00000

About the Cover 1780

Images from the Haematologica Atlas of Hematologic Cytology: MYH9-related disease Carlo L. Balduini and Alessandro Pecci https://doi.org/10.3324/haematol.2021.279022

Editorials 1781

A chemotherapy-free regimen for Philadelphia chromosome-positive acute lymphoblastic leukemia: are we there yet? Yishai Ofran https://doi.org/10.3324/haematol.2020.278077

1783

JAK out of the box: myeloproliferative neoplasms--associated JAK2 V617F mutations contribute to aortic aneurysms Shannon E. Elf https://doi.org/10.3324/haematol.2020.277111

1785

Be mindful of the central nervous system in Burkitt lymphoma Mark Roschewski https://doi.org/10.3324/haematol.2020.278181

Perspective Article 1787

Interrogating the molecular genetics of chronic myeloproliferative malignancies for personalized management in 2021 Tariq I. Mughal et al. https://doi.org/10.3324/haematol.2020.267252

Review Articles 1794

Dose intensity for conditioning in allogeneic hematopoietic cell transplantation: can we recommend “when and for whom” in 2021? Nico Gagelmann and Nicolaus Kröger https://doi.org/10.3324/haematol.2020.268839

1805

Red cell transfusion and alloimmunization in sickle cell disease Grace E. Linder and Stella T. Chou https://doi.org/10.3324/haematol.2020.270546

Articles Acute Lymphoblastic Leukemia

1816

WEE1 inhibition induces glutamine addiction in T-cell acute lymphoblastic leukemia Juncheng Hu et al. https://doi.org/10.3324/haematol.2019.231126

1828

A multicenter total therapy strategy for de novo adult Philadelphia chromosome positive acute lymphoblastic leukemia patients: final results of the GIMEMA LAL1509 protocol Sabina Chiaretti et al. https://doi.org/10.3324/haematol.2020.260935

Acute Myeloid Leukemia

1839

Comparison of total body irradiation versus non-total body irradiation containing regimens for de novo acute myeloid leukemia in children Christopher E Dandoy et al. https://doi.org/10.3324/haematol.2020.249458

Haematologica 2021; vol. 106 no. 7 - July 2021 http://www.haematologica.org/

haematologica Journal of the Ferrata Storti Foundation Cell Therapy & Immunotherapy

1846

Anti-RhD antibody therapy modulates human natural killer cell function Shlomo Elias et al. https://doi.org/10.3324/haematol.2019.238097

1857

Fc-engineering significantly improves the recruitment of immune effector cells by anti-ICAM-1 antibody MSH-TP15 for myeloma therapy Katja Klausz et al. https://doi.org/10.3324/haematol.2020.251371

Chronic Lymphocytic Leukemia

1867

Efficacy of minimal residual disease driven immune-intervention after allogeneic hematopoietic stem cell transplantation for high-risk chronic lymphocytic leukemia: results of a prospective multicenter trial Olivier Tournilhac et al. https://doi.org/10.3324/haematol.2019.239566

Chronic Myeloid Leukemia

1876

Fatigue in chronic myeloid leukemia patients on tyrosine kinase inhibitor therapy: predictors and the relationship with physical activity Lando Janssen et al. https://doi.org/10.3324/haematol.2020.247767

Hematopoiesis

1883

Thrombopoietin maintains cell numbers of hematopoietic stem and progenitor cells with megakaryopoietic potential Aled O’Neill et al. https://doi.org/10.3324/haematol.2019.241406

Hemophagocytosis

1892

A pilot study of ruxolitinib as a front-line therapy for 12 children with secondary hemophagocytic lymphohistiocytosis Qing Zhang et al. https://doi.org/10.3324/haematol.2020.253781

Hemostasis

1902

Factor VIII activity and bleeding risk during prophylaxis for severe hemophilia A: a population pharmacokinetic model Andreas Tiede et al. https://doi.org/10.3324/haematol.2019.241554

Myeloproliferative Disorders

1910

Crucial role of hematopoietic JAK2 V617F in the development of aortic aneurysms Tetsuro Yokokawa et al. https://doi.org/10.3324/haematol.2020.264085

Non-Hodgkin Lymphoma

1923

Impact on survival through consolidation radiotherapy for diffuse large B-cell lymphoma: a comprehensive meta-analysis Martin D. Berger et al. https://doi.org/10.3324/haematol.2020.249680

1932

Outcomes of Burkitt lymphoma with central nervous system involvement: evidence from a large multicenter cohort study Adam S. Zayac et al. https://doi.org/10.3324/haematol.2020.270876

Plasma Cell Disorders

1943

1957

Haematologica 2021; vol. 106 no. 7 - July 2021 http://www.haematologica.org/

haematologica Journal of the Ferrata Storti Foundation Platelet Biology & its Disorders

1968

Antiplatelet properties of Pim kinase inhibition are mediated through disruption of thromboxane A2 receptor signaling Amanda J. Unsworth et al. https://doi.org/10.3324/haematol.2019.223529

Red Cell Biology & its Disorders

1979

In vitro and in vivo induction of fetal hemoglobin with a reversible and selective DNMT1 inhibitor Aidan G. Gilmartin et al. https://doi.org/10.3324/haematol.2020.248658

Letters to the Editor 1988

Mixed myeloid chimerism and relapse of myelofibrosis after allogeneic stem cell transplantation Samer A. Srour, et al. https://doi.org/10.3324/haematol.2019.223503

1991

Symptom burden in transplant-ineligible patients with newly diagnosed multiple myeloma: a population-based cohort study Hira S. Mian et al. https://doi.org/10.3324/haematol.2020.267757

1995

Mixed-lineage leukemia protein modulates the loading of let-7a onto AGO1 by recruiting RAN Shouhai Zhu et al. https://doi.org/10.3324/haematol.2020.268474

2000

Glenzocimab does not impact glycoprotein VI-dependent inflammatory hemostasis Soumaya Jadoui et al. https://doi.org/10.3324/haematol.2020.270439

2004

Comparison of CD38 antibodies in vitro and ex vivo mechanisms of action in multiple myeloma Michelle Kinder et al. https://doi.org/10.3324/haematol.2020.268656

2009

Immunophenotypic changes of leukemic blasts in children with relapsed/refractory B-cell precursor acute lymphoblastic leukemia who have been treated with blinatumomab Ekaterina Mikhailova et al. https://doi.org/10.3324/haematol.2019.241596

2013

Whole genome CRISPR screening identifies TOP2B as a potential target for IMiD sensitization in multiple myeloma Matteo Costacurta et al. https://doi.org/10.3324/haematol.2020.265611

2018

Myocardial injury and coronary microvascular disease in sickle cell disease Kiranveer Kaur et al. https://doi.org/10.3324/haematol.2020.271254

2022

Entospletinib and obinutuzumab in patients with relapsed/refractory chronic lymphocytic leukemia and B-cell malignancies Adam S. Kittai et al. https://doi.org/10.3324/haematol.2020.270298

2026

An international retrospective study for tolerability of 6-mercaptopurine on NUDT15 bi-allelic variants in children with acute lymphoblastic leukemia Yoichi Tanaka et al. https://doi.org/10.3324/haematol.2020.266320

Case Reports 2030

A mutation in the iron-responsive element of ALAS2 is a modifier of disease severity in a patient suffering from CLPX associated erythropoietic protoporphyria Sarah Ducamp et al. https://doi.org/10.3324/haematol.2020.272450

2034

Light chain proteinuria revealing mu-heavy chain disease: an atypical presentation of Waldenström macroglobulinemia in two cases Hélène Vergneault et al. https://doi.org/10.3324/haematol.2020.277137

Haematologica 2021; vol. 106 no. 7 - July 2021 http://www.haematologica.org/

haematologica Journal of the Ferrata Storti Foundation

The origin of a name that reflects Europe’s cultural roots.

Ancient Greek

Scientific Latin

Modern English

aÂma [haima] = blood a·matow [haimatos] = of blood lÒgow [logos]= reasoning haematologicus (adjective) = related to blood haematologica (adjective, plural and neuter, used as a noun) = hematological subjects The oldest hematology journal, publishing the newest research results. 2019 JCR impact factor = 7.116

OBITUARY Edoardo Ascari, a man to whom Haematologica owes much Carlo Balduini President of the Ferrata-Storti Foundation, Pavia, Italy E-mail: CARLO BALDUINI - carlo.balduini@unipv.it. doi:10.3324/haematol.2021.279273

Edoardo Ascari 1932-2021.

doardo Ascari, former Editor-in-Chief of Haematologica and President of the Ferrata-Storti Foundation, passed away peacefully in the morning of May 14 this year after a long and fruitful life. Ascari was born in Modena, Italy, in 1932 and in this city he met the two most important people in his life. The first was the woman destined to become his wife and to give him two wonderful children. The second was Edoardo Storti, who at the time headed the Department of Internal Medicine of Modena University. Once he graduated in Medicine, Ascari became a pupil of Storti and began to take his first steps in the field of Hematology, focusing mainly on defects of blood coagulation. These were the times when infusions of fresh plasma and cryoprecipitate were being used to improve hemostasis of hemophiliacs and it is thanks to their prophylactic use that Storti and his collaborators, first and foremost Ascari, were, for the first time, able to practice synovectomy successfully to prevent relapsing hemarthrosis. When, in 1969, Storti became Director of the Medical Clinic of the University of Pavia, Ascari followed him and there organized a hemostasis laboratory that was very advanced for those times. In the 1980s, having become full Professor of Internal Medicine, Ascari took over the direction of Clinical Medicine in Pavia and led numerous collaborators destined to make important contributions in various fields of Hematology. Under his direction, the Pavia Medical Clinic continued to be a reference center for hemorrhagic and thrombotic diseases, as well as for many other neoplastic and non-neoplastic blood disorders.

haematologica | 2021; 106(7)

Ascari played an important role in Hematology not only as a researcher at the beginning of his career and then as the mentor of an important school of Hematology, but also as Editor-in-Chief from 1990 to 2002 of Haematologica, the journal founded by Adolfo Ferrata in 1920, and as President of the Ferrata-Storti Foundation, the institution founded by Storti mainly to manage and support the journal, from 2002 to 2019. Directly or indirectly, he therefore took care of Haematologica for nearly 30 years. When Ascari took over as Editor-in-Chief of Haematologica, the journal was in a critical condition. The number of submitted manuscripts was small, the selection of the articles to be published could not therefore be too strict and the result of this was that the citations received by the journal were few. This was a vicious circle from which it was difficult to escape. Ascari, among the very first to do so in the field of Hematology, took the courageous decision to make the journal open access, knowing that this choice could have led to a further decline in subscriptions. It was, however, a winning move: the number of citations increased which meant that the number of authors who wanted to publish in the journal also increased. The selection of the papers to be published could become ever more discerning, and a virtuous circle was triggered, which progressively enhanced the prestige of the journal and led it to now being one of the best reputed in the field of Hematology. The Ferrata-Storti Foundation and all the people who work today to make the publication of Haematologica possible remember Edoardo Ascari with great affection and gratitude.

1779

ABOUT THE COVER Images from the Haematologica Atlas of Hematologic Cytology: MYH9-related disease Carlo L. Balduini1 and Alessandro Pecci2 1

Ferrata-Storti Foundation, Pavia, Italy and 2University of Pavia-IRCCS Policlinico San Matteo, Pavia, Italy

E-mail: CARLO L. BALDUINI - carlo.balduini@unipv.it doi:10.3324/haematol.2021.279022

characteristic of MYH9-related disease (MYH9-RD), the most frequent inherited thrombocytopenia, is the presence of aggregated MYH9 protein in the cytoplasm of neutrophils. Immunofluorescence staining of normal neutrophils shows the homogeneous distribution of MYH9 protein (A). In MYH9-RD MYH9 protein aggregates due to mutations in the N- or C-terminus. C-terminal mutations usually result in one or a few large aggregates which potentially associate with small aggregates (B), while N-terminal mutations (mutations in the motor domain) of MYH9 lead to the formation of many small aggregates only (C). Both small and large aggregates, named Döhle-like inclusion bodies, are detectable by immunofluorescence analysis in all neutrophils of all patients with a detection specificity and sensitivity close to 100%. However, in May-Grünwald-Giemsa stained blood films, these aggregates are more difficult to identify and are only detected in a small percentage of neutrophils in around 50% of patients. Due to their characteristic round or spindle shape large aggregates are more easily identified as light blue corpuscles usually located at the cell periphery (D to F), whereas small aggregates (G to I, arrows) are more difficult to identify.1 Disclosures No conflicts of interest to disclose. Contributions Both authors contributed.

Reference 1. Balduini CL, Pecci A. Inherited thrombocytopenias. Haematologica. 2020;105(Suppl 1):S237-247.

1780

haematologica | 2021; 106(7)

EDITORIALS A chemotherapy-free regimen for Philadelphia chromosome-positive acute lymphoblastic leukemia: are we there yet? Yishai Ofran1,2 1

Department of Hematology and Bone Marrow Transplantation, Rambam Health Care Campus, and 2The Ruth and Bruce Rappaport Faculty of Medicine, Technion, Haifa, Israel E-mail: YISHAI OFRAN - y_ofran@rambam.health.gov.il doi:10.3324/haematol.2020.278077

ovel therapies are revolutionizing the treatment strategies for Philadelphia chromosome-positive (Ph+) acute lymphoblastic leukemia (ALL). Ten years ago, the GIMEMA (Gruppo Italiano Malattie EMatologiche dell’Adulto) ALL Working Party pioneered a chemotherapyfree induction regimen using the dasatinib-steroid combination,1 which brought about complete hematological remission (CHR) in all 53 evaluable patients2 and complete molecular response (CMR) in ten of them (18.8%). In this issue, final results of a subsequent LAL1509GIMEMA prospective single-arm trial are reported.3 The treatment protocol included 1-month induction with the aforementioned combination, followed by dasatinib extension until day 85. The post-remission regimen was assigned based on the minimal residual disease (MRD) status. Of the 60 enrolled patients (median age 41.9 years [range, 18-60]), those who achieved CMR were subject to dasatinib maintenance with no further intensification. The majority of patients (47 of 60; 78%) achieved CHR, while testing MRDpositive post-induction. These patients were assigned to allogeneic stem cell transplantation (allo-SCT) with or without consolidation chemotherapy. Patients ineligible for transplantation were consolidated with chemotherapy only. No induction deaths were reported and the CHR rate by day 85 was 97%, with CMR achieved in 11 of 60 (18.3%) patients. At a median follow-up of almost 5 years, overall survival (OS) and disease-free survival (DFS) were 56.3% and 47.2%, respectively. Are these impressive data sufficient to set the stage for a new standard of care in Ph+ALL? One of the critical achievements of these two studies, that should be taken into consideration, is the absence of induction deaths among the 113

patients treated. To that end, future Ph+ALL treatment protocols should be designed with the aim to maintain such a minimal induction-related mortality rate. However, this attractive low-intensity induction regimen may be unsuitable for higher-risk Ph+ALL patients. In the LAL1509GIMEMA protocol, the post-induction MRD status has been the sole factor used to stratify patients for intensive consolidation followed by allo-SCT versus dasatinib maintenance only. Yet, the clinical significance of MRD results depends on a variety of patient- and treatment-related parameters. Based on data from previous ALL studies, the Food and Drug Admisnsitration has accepted an MRD level of less than 0.01% as a surrogate efficacy endpoint for new drugs in ALL.4 Surprisingly and disappointingly, in the current trial, four of 11 (36%) patients who achieved CMR with the dasatinib-steroid induction eventually relapsed. Three of these relapses were diagnosed early at a molecular level and therefore never fulfilled the former criteria of relapse. This must raise a red flag and a message regarding the complexity of MRD clinical interpretation should be played out loud. MRD negativity is not synonymous to cure but it is rather its biomarker. No matter how sensitive the available tests are, there is still room for residual disease presence at a level below the threshold of detection. Thus, ultimate degrees of disease eradication for patients who achieve MRD negativity following intensive and less intensive induction may be different. A negative MRD result following intensive induction reflects an even deeper response than the sensitivity cutoff of the test used. This should not be extrapolated to the outcome prediction following less intensive protocols when the actual level of response below the MRD negativity cutoff may be lower (Figure 1). Not only the specific induction protocol but also

Figure 1. Response after intensive and non-intensive induction in a population of Philadelphia chromosome-positive acute lymphoblastic leukemia patients. (A) Distribution of response levels after intensive induction. (B) Distribution of response levels after non-intensive induction. The distribution of the actual depth of response across patients who achieved minimal residual disease negativity following intensive and non-intensive induction differs and so does the risk of relapse.

haematologica | 2021; 106(7)

1781

Editorials

the characteristics of the patient population should be taken into account. For instance, in Ph-negative ALL, achievement of identical laboratory major molecular response levels following treatment with intensive chemotherapy and tyrosine kinase inhibitors, was shown to predict considerably different relapse-free survival rates for newly diagnosed versus relapsed patients (26.1 vs. 12 months, respectively).5,6 Notably, progression-free survival for relapsing patients who achieved MRD negativity with inotuzumab ozogamicin treatment was as short as 8.6 months.7 Thus, in different clinical settings, identical laboratory results may be associated with completely different predicted outcomes. The LAL1509 trial, while demonstrating feasibility of a chemotherapy-free regimen for some patients, has also highlighted that biological differences within the Ph+ALL patient population go far beyond the presence or absence of BCR/ABL mutations such as T315I. In this trial, patients presenting with genetic aberrations in IKZF1 plus CDKN2A/B and/or PAX5 (IKZF1+) demonstrated a relapse-free survival rate of 0%. This combination of genetic abnormalities is known to portend poor prognosis even in patients undergoing allo-SCT. Yet, the relapsefree survival rates reported in previous studies using intensive induction regimens have been better.8 Moreover, the GIMEMA group has lately reported results of using a still more attractive combination of steroids, dasatinib and blinatumomab,9 that has led to considerable improvement in the survival of all patients, including those presenting with IKZF1+ aberrations. What have we learned from the current trial? First, induction death in Ph+ALL is preventable and for some patients even non-intensive induction should be considered. Second, the way MRD negativity is achieved influences its clinical implications. Third, routine screening for IKZF1+abnormalities is advised and may be considered when the intensity of an induction regimen is selected. And last but definitely not least, a combination of steroids, tyrosine kinase inhibitors and blinatumomab, given as first-line therapy, is an attractive option for

1782

Ph+ALL patients and is currently being studied as part of a large intergroup prospective phase III trial, led by ECOG-ACRIN, EA9181 (clinicaltrials gov. Identifier: NCT04530565).10 Disclosures No conflicts of interest to disclose.

References 1. Vignetti M, Fazi P, Cimino G, et al. Imatinib plus steroids induces complete remissions and prolonged survival in elderly Philadelphia chromosome-positive patients with acute lymphoblastic leukemia without additional chemotherapy: results of the Gruppo Italiano Malattie Ematologiche dell'Adulto (GIMEMA) LAL0201-B protocol. Blood. 2007;109(9):3676-3678. 2. Foa R, Vitale A, Vignetti M, et al. Dasatinib as first-line treatment for adult patients with Philadelphia chromosome-positive acute lymphoblastic leukemia. Blood. 2011;118(25):6521-6528. 3. Chiaretti S, Ansuinelli M, Vitale A, et al. A multicenter total therapy strategy for de novo adult Philadelphia chromosome positive acute lymphoblastic leukemia patients. Final results of the GIMEMA LAL 1509 protocol. Haematologica. 2021;106(7):1828-1838. 4. FDA Guidance Document: Hematologic Malignancies: Regulatory Considerations for use of minimal residual disease in development of drug and biological products for treatment. Available at https://www.fda.gov/media/134605/download. Accessed in December, 2020. 5. Short NJ, Jabbour E, Sasaki K, et al. Impact of complete molecular response on survival in patients with Philadelphia chromosome-positive acute lymphoblastic leukemia. Blood. 2016;128(4):504-507. 6. Abou Dalle I, Kantarjian HM, Short NJ, et al. Philadelphia chromosome-positive acute lymphoblastic leukemia at first relapse in the era of tyrosine kinase inhibitors. Am J Hematol. 2019;94(12):13881395. 7. Jabbour E, Gokbuget N, Advani A, et al. Impact of minimal residual disease status in patients with relapsed/refractory acute lymphoblastic leukemia treated with inotuzumab ozogamicin in the phase III INO-VATE trial. Leuk Res. 2020;88:106283. 8. Pfeifer H, Raum K, Markovic S, et al. Genomic CDKN2A/2B deletions in adult Ph(+) ALL are adverse despite allogeneic stem cell transplantation. Blood. 2018;131(13):1464-1475. 9. Foa R, Bassan R, Vitale A, et al. Dasatinib-Blinatumomab for Ph-positive acute lymphoblastic leukemia in adults. N Engl J Med. 2020; 383(17):1613-1623. 10. Testing the use of steroids and tyrosine kinase inhibitors with blinatumomab or chemotherapy for newly diagnosed BCR-ABL-positive acute lymphoblastic leukemia in adults. Available at https://clinicaltrials.gov/ct2/show/NCT04530565. Accessed in December, 2020.

haematologica | 2021; 106(7)

Editorials

JAK out of the box: myeloproliferative neoplasms--associated JAK2 V617F mutations contribute to aortic aneurysms Shannon E. Elf The Ben May Department for Cancer Research, The University of Chicago, Chicago, IL, USA E-mail: SHANNON ELF - shannonelf@uchicago.edu doi:10.3324/haematol.2020.277111

yeloproliferative neoplasms (MPN) are clonal disorders of hematopoiesis arising in the hematopoietic stem cell (HSC) compartment and characterized by the excess production of mature myeloid cells.1 BCR-ABL-negative MPN include polycythemia vera (PV), essential thrombocythemia (ET), and myelofibrosis (MF). PV is characterized by uncontrolled red blood cell production; ET, by megakaryocytic hyperplasia and elevated platelet counts; and MF, by megakaryocytic hyperplasia and bone marrow fibrosis. The molecular basis of MPN remained unknown until 2005, when four different groups described a point mutation in the pseudokinase domain of Janus kinase-2 (JAK2), a non-receptor tyrosine kinase, in the majority of MPN patients. The resulting JAK2 V617F mutant protein was found to be constitutively active, leading to hyperactive JAK-STAT signaling downstream of multiple hematopoietic cytokine receptors.2-5 Although MPN, particularly PV and ET, are commonly characterized as “indolent” diseases, patients have significantly decreased life expectancies compared to the general population. JAK2 V617F mutations are associated with increased vascular complications, which to date primarily include venous and arterial thrombosis and advanced atherosclerosis.6,7 Indeed, fatal thrombotic events represent the leading cause of death in JAK2

V617F-positive MPN patients, though the underlying mechanisms remain elusive. Mouse models of Jak2 V617F-driven MPN faithfully recapitulate this phenotype, with lethality primarily attributed to thrombosis.8 Interestingly, the prevalence of the JAK2 V617F mutation has also been found to be significantly increased in non-MPN patients with coronary artery disease and peripheral artery disease,9,10 suggesting that the JAK2 V617F mutation may play a role in additional vascular diseases. This observation is supported by the phenomenon of clonal hematopoiesis, in which clonal expansion of hematopoietic cells carrying is associated with significantly increased risk of vascular disorders.11 However, the mechanisms underlying JAK2 V617F-mediated vascular disease remain unclear, and the association of JAK2 V617F with vascular diseases outside of thrombosis, atherosclerosis, coronary artery disease and peripheral arterial disease has yet to be studied. In this issue of Haematologica, Yokokawa et al.12 investigate the contribution of bone marrow (BM)-derived JAK2 V617F to the development of aortic aneurysms (AA), a vascular disease not yet studied in MPN patients. AA often progress asymptomatically and can lead to sudden death, so understanding the mechanisms underlying their onset and associated risk factors is critically impor-

Figure 1. Hematopoietic JAK2 V617F mutations lead to the development of abdominal aortic aneurysms (AAA). Somatic JAK2 V617F mutations are acquired in the bone marrow at the hematopoietic stem cell (HSC) level. The resulting bone marrow-derived JAK2 V617F-mutated macrophages infiltrate the abdominal aorta, where they demonstrate increased levels of two genes critical for aortic aneurysm formation, matrix metalloproteins 2 and 9 (MMP-2, MMP-9). This expression and the resulting AAA are decreased upon treatment with a JAK inhibitor (JAK2i), ruxolitinib, confirming that the development of AAA is mediated by JAK2 V617F.

haematologica | 2021; 106(7)

1783

Editorials

tant. Here, the authors perform a prevalence study in 39 JAK2 V617F-positive MPN patients, and found that 23% displayed signs of AA. Intriguingly, they find that JAK2 V617-positive circulating leukocytes demonstrated upregulation of genes associated with AA, including matrix metalloproteinase 9 (MMP-9), which plays a well-established role in AA. This finding provided the authors’ first insight into the potential mechanism underlying JAK2 V617F-mediated AA. In order to further understand the role that hematopoietic-derived JAK2 V617F plays in the development of AA, the authors turn to a well characterized bone marrow transplantation (BMT) model of JAK2 V617F-driven MPN, in which JAK2 V617F expression is restricted to donor BM cells. Here, the authors utilize mice deficient in apolipoprotein E (ApoE-/-) as recipients, and subject them to continuous infusion of angiotensin II (AngII), a model that has been shown to promote development and expansion of AA. In their endpoint analysis, the authors compare animals receiving wild-type (WT) versus JAK2 V617F-expressing BM cells, and find that, in addition to the expected MPN-like phenotype, JAK2 V617F BMT mice exhibit significantly increased abdominal aorta diameter, indicative of abdominal AA (AAA). Moreover, they find that JAK2 V617F expression accelerated the AAA hallmark of arterial extracellular matrix proteolysis as measured by aortic elastic lamina degradation, and led to activation of MMP-9 as well as MMP-2 in the abdominal aorta, both of which are required to produce AAA. Together, these results suggest that BM-derived JAK2 V617F promotes the development of AAA. Digging deeper into the molecular mechanism underlying how JAK2 V617F leads to the development of AAA, the authors find infiltration of JAK2 V617F-mutant inflammatory cells, including CD68+ macrophages and Ly6B.2+ neutrophils, as well as increased phosphorylation of JAK2 V617F target STAT3, in the abdominal aortas of JAK2 V617F BMT mice receiving AngII. They go on to show that the inflammatory cells are strictly BM-derived, confirming that it is indeed hematopoietic JAK2 V617F causing the development of AAA in these animals. Finally and most intriguingly, they show that JAK2 V617F BM-derived CD68+ macrophages exhibit significantly increased mRNA expression levels of Mmp2 and Mmp9, both of which could be decreased upon treatment with the JAK2 inhibitor ruxolitinib. Additionally, AngII-treated JAK2

1784

V617F BMT mice treated with ruxolitinib experienced decreased incidence of AAA. These data suggest that JAK2 V617F promotes activation of MMP in BM-derived inflammatory cells, which leads to AAA development (Figure 1). Taken together, the results from this study identify a novel vascular disorder associated with JAK2 V617F mutations, provide a direct link between JAK2 V617F and the pathogenesis of AAA, and offer an additional therapeutic application for Food and Drug Admionstrationapproved JAK2 inhibitors in the prevention of AAA development in JAK2 V617F-positive MPN patients. Disclosures No conflicts of interest to disclose.

References 1. Campbell PJ, Green AR. The myeloproliferative disorders. N Engl J Med. 2006;355(23):2452-2466. 2. Baxter EJ, Scott LM, Campbell PJ, East C, Fourouclas N, Swanton S, et al. Acquired mutation of the tyrosine kinase JAK2 in human myeloproliferative disorders. Lancet. 2005;365(9464):1054-1061. 3. James C, Ugo V, Le Couedic JP, Staerk J, Delhommeau F, Lacout C, et al. A unique clonal JAK2 mutation leading to constitutive signalling causes polycythaemia vera. Nature. 2005;434(7037):1144-1148. 4. Kralovics R, Passamonti F, Buser AS, et al. A gain-of-function mutation of JAK2 in myeloproliferative disorders. N Engl J Med. 2005;352(17):1779-1790. 5. Levine RL, Wadleigh M, Cools J, et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis. Cancer Cell. 2005;7(4):387-397. 6. Elliott MA, Tefferi A. Thrombosis and haemorrhage in polycythaemia vera and essential thrombocythaemia. Br J Haematol. 2005;128(3):275-290. 7. Wang W, Liu W, Fidler T, et al. Macrophage inflammation, erythrophagocytosis, and accelerated atherosclerosis in Jak2 (V617F) Mice. Circ Res. 2018;123(11):e35-e47. 8. Mullally A, Lane SW, Ball B, et al. Physiological Jak2V617F expression causes a lethal myeloproliferative neoplasm with differential effects on hematopoietic stem and progenitor cells. Cancer Cell. 2010;17(6):584-596. 9. Muendlein A, Gasser K, Kinz E, et al. Evaluation of the prevalence and prospective clinical impact of the JAK2 V617F mutation in coronary patients. Am J Hematol. 2014;89(3):295-301. 10. Muendlein A, Kinz E, Gasser Ket al. Occurrence of the JAK2 V617F mutation in patients with peripheral arterial disease. Am J Hematol. 2015;90(1):E17-21. 11. Jaiswal S, Natarajan P, Silver AJ, et al. Clonal hematopoiesis and risk of atherosclerosis cardiovascular disease. N Engl J Med. 2017;377(2): 111-121. 12. Yokokawa T, Misaka T, Kimishima Y, et al. Crucial role of hematopoietic JAK2V617F in the development of aortic aneurysms. Haematologica. 2021;106(7):1910-1922.

haematologica | 2021; 106(7)

Editorials

Be mindful of the central nervous system in Burkitt lymphoma Mark Roschewski Lymphoid Malignancies Branch, Center for Cancer Research, National Cancer Institute, Bethesda, MD, USA E-mail: MARK ROSCHEWSKI - mark.roschewski@nih.gov doi:10.3324/haematol.2020.278181

n this issue of Haematologica, Zayac et al. illustrate the adverse prognostic impact of central nervous system (CNS) involvement for adult patients with Burkitt lymphoma (BL).1 The authors reported data from a retrospective series of 641 patients treated at 30 institutions in the USA and identified 120 (19%) patients with baseline CNS involvement. In the majority of cases (81%), the leptomeninges were involved, 17% had brain parenchymal lesions, and the site was unspecified in 2%. The risk was not uniformly distributed as patients with concomitant human immunodeficiency virus (HIV) infection had CNS involvement in 30% of cases whereas extranodal disease (particularly testicular involvement), poor performance status, and markedly elevated lactate dehydrogenase level were strongly correlated with CNS spread in HIV-negative patients. Most patients (89%) were treated with one of three standard frontline BL regimens: CODOX-M/IVAC, HyperCVAD/MA, and DA-EPOCH-R, presumably all with curative intent.2-4 Notably, active CNS involvement was evenly distributed across the three regimens, although this was not the case for other risk factors. The data clearly demonstrated that baseline CNS involvement

identified patients who are difficult to cure with standard frontline approaches as they had lower rates of complete response (59% vs. 77%; P<0.001) and a significantly lower 3-year overall survival (49% vs. 74%; P<0.001) compared to patients without baseline CNS involvement. Importantly, the negative prognostic impact of baseline CNS involvement was significant for all three regimens. A second focus of the study was predictors of CNS recurrence in 570 patients treated with the three standard regimens. The overall rate of disease recurrence was 26% and the recurrence involved the CNS in 23% of cases. Most of these recurrences (82%) were isolated to the CNS without concomitant systemic recurrence and 87% occurred within the first year of treatment. These data strongly suggest that current strategies typically fail to eradicate active baseline CNS disease rather than the CNS being a sanctuary site for late relapses. Indeed, patients with baseline CNS involvement had the highest CNS recurrence rate of 18%. Considered by regimen, patients who received DA-EPOCH-R had a 35% rate of CNS recurrence among those with baseline CNS disease. These data illustrate that baseline CNS involvement is a

Figure 1. Multiple barriers prevent current strategies from effectively treating and preventing central nervous system disease in adults with Burkitt lymphoma. Routine cytology is insufficiently sensitive for the detection of occult lymphoma cells in the cerebrospinal fluid (CSF) and widespread adoption of more sensitive techniques such as CSF flow cytometry or the development of assays that detect cell-free tumor DNA in the CSF are potential solutions. Therapy intolerance is a barrier unique to adults since pediatric regimens that employ high intensity chemotherapy and are highly effective in children and young adults with central nervous system (CNS) disease are often not tolerated by older patients or those with co-morbid conditions. Treating “possible Burkitt lymphoma (BL)” as a medical emergency with rapid employment of aggressive supportive care can improve performance status to allow for proper CNS-directed interventions and reduce the risk of early toxic death. The blood-brain barrier limits the use of many chemotherapy agents, but novel pathway inhibitors and immunotherapy agents with activity in BL are in clinical development, are often more tolerable, effectively cross the blood-brain barrier, and may overcome chemotherapy resistance. A more complete understanding of the genetic basis for mechanisms of drug resistance in BL may allow the development of additional novel agents that penetrate the CNS.

haematologica | 2021; 106(7)

1785

Editorials

greater problem in real-world practice than observed in clinical trials, and that isolated CNS recurrence is a common cause of treatment failure in adults with BL. The contributing factors are multiple and one should not conclude that the problem is readily addressed by choice of frontline regimen. In reality, the absence of highly effective CNS-penetrating agents is a major barrier, and new strategies should focus on this limitation (Figure 1). In the meantime, optimal management mandates being ever mindful of CNS detection, prophylaxis, and treatment. The first important question is whether high-dose methotrexate-containing regimens such as CODOXM/IVAC or HyperCVAD/MA are superior to DA-EPOCHR in patients with CNS disease. The inherent selection bias of this retrospective study design precludes a definitive answer since the patients who received DA-EPOCHR were older and were more likely to have a poor performance status.5 It is unclear whether these patients would have even been candidates for highly dose-intensive regimens since toxicities rise sharply with age.6,7 In a multicenter study of DA-EPOCH-R in 113 adults with BL, we reported that baseline CNS involvement was associated with an inferior progression-free survival.4 Importantly, these patients had equal risk of disease progression as well as early toxic death due to sepsis compared to patients without CNS disease. Intensification of chemotherapy may improve control of CNS disease, but will increase the risk of early toxic death. In reality, only a prospective randomized study, such as the HOVON 127 trial (EudraCT number: 2013-004394-27), can adequately address the question of optimal therapy selection for adults with BL across various subsets. Careful review of the data in the report from Zayac et al. and in the published literature suggests that additional barriers contribute to this complex problem (Figure 1). An important component of proper use of DA-EPOCH-R for BL is a risk-adapted approach that includes a careful analysis of the cerebrospinal fluid (CSF) with flow cytometry to detect occult disease. Using this approach, patients who are positive by flow cytometry are treated with an intensive intrathecal chemotherapy schedule while patients who are negative by flow cytometry receive less intrathecal treatment as CNS prophylaxis. Given the nature of the retrospective study by Zayac et al., the proper use of CSF flow cytometry to identify and adjust treatment with DA-EPOCH-R is unknown as is the use of brain imaging and CSF cytology. The absence of proper CNS evaluation in this series confounds any interpretation that DA-EPOCH-R is less effective than other highdose regimens. Furthermore, only 45% of the patients received the schedule of intrathecal therapy as published. Lastly, intrathecal chemotherapy alone would not be adequate therapy for brain parenchymal lesions and these patients have been excluded from clinical trials of DAEPOCH-R; yet four patients in the study discussed had brain parenchymal lesions and received DA-EPOCH-R, which contributed to the high rate of CNS recurrence.

1786

The authors adequately address the limitations of their data and should be commended for performing a sobering analysis of outcomes in adults with BL which we commonly consider “highly curable.” In reality, these data illustrate that we successfully cure fewer adult BL patients than suggested by clinical trials, although the proper implementation of the treatment regimens in general practice likely contributes to this disparity. Advances in the detection of occult CSF disease, including assays for cellfree tumor DNA, may be important in future studies. Perhaps most importantly, these data highlight that no matter which chemotherapy regimen is selected, effective treatment of CNS disease is an unmet need in BL. Novel targeted agents or immunotherapy that can overcome chemotherapy resistance, penetrate the blood-brain barrier, and are tolerated by adult patients of all ages offer the most promise. Sadly, the patients with the highest risk of treatment failure, such as those with CNS involvement or HIV, are often excluded from clinical trials testing these novel agents.8-10 Until we collectively change our minds about the wisdom of this practice, we should expect to be disappointed with ‘real-world’ outcomes. Disclosures No conflicts of interest to disclose.

References 1. Zayac AS, Evens A, Olszewski AJ. Outcomes of Burkitt lymphoma with central nervous system involvement: evidence from a large multi-center cohort study. Haematologica. 2021;106(7):1932-1942. 2. Mead GM, Sydes MR, Walewski J, et al. An international evaluation of CODOX-M and CODOX-M alternating with IVAC in adult Burkitt's lymphoma: results of United Kingdom Lymphoma Group LY06 study. Ann Oncol. 2002;13(8):1264-1274. 3. Thomas DA, Faderl S, O'Brien S, et al. Chemoimmunotherapy with hyper-CVAD plus rituximab for the treatment of adult Burkitt and Burkitt-type lymphoma or acute lymphoblastic leukemia. Cancer. 2006;106(7):1569-1580. 4. Roschewski M, Dunleavy K, Abramson JS, et al. Multicenter study of risk-adapted therapy with dose-adjusted EPOCH-R in adults with untreated Burkitt lymphoma. J Clin Oncol. 2020;38(22):2519-2529. 5. Evens AM, Danilov AV, Jagadeesh D, et al. Burkitt lymphoma in the modern era: real world outcomes and prognostication across 30 US cancer centers. Blood. 2021;137(3):374-386. 6. Costa LJ, Xavier AC, Wahlquist AE, Hill EG. Trends in survival of patients with Burkitt lymphoma/leukemia in the USA: an analysis of 3691 cases. Blood. 2013;121(24):4861-4866. 7. Kelly JL, Toothaker SR, Ciminello L, et al. Outcomes of patients with Burkitt lymphoma older than age 40 treated with intensive chemotherapeutic regimens. Clin Lymphoma Myeloma. 2009;9(4):307-310. 8. Uldrick TS, Ison G, Rudek MA, et al. Modernizing clinical trial eligibility criteria: recommendations of the American Society of Clinical Oncology-Friends of Cancer Research HIV Working Group. J Clin Oncol. 2017;35(33):3774-3780. 9. Lin NU, Prowell T, Tan AR, et al. Modernizing clinical trial eligibility criteria: recommendations of the American Society of Clinical Oncology-Friends of Cancer Research Brain Metastases Working Group. J Clin Oncol. 2017;35(33):3760-3773. 10. Lichtman SM, Harvey RD, Damiette Smit MA, et al. Modernizing clinical trial eligibility criteria: recommendations of the American Society of Clinical Oncology-Friends of Cancer Research Organ Dysfunction, Prior or Concurrent Malignancy, and Comorbidities Working Group. J Clin Oncol. 2017;35(33):3753-3759.

haematologica | 2021; 106(7)

PERSPECTIVE ARTICLE

Interrogating the molecular genetics of chronic myeloproliferative malignancies for personalized management in 2021

Ferrata Storti Foundation

Tariq I. Mughal,1,2 Bethan Psaila,3 Daniel J. DeAngelo,4 Giuseppe Saglio,5 Richard A. Van Etten6 and Jerald P. Radich7

Tufts University Medical Center, Boston, MA, USA; 2University of Buckingham Medical School, Buckingham, UK; 3MRC Weatherall Institute of Molecular Medicine, University of Oxford, Oxford, UK; 4Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA, USA; 5Orbassano University Hospital, Turin, Italy; 6University of California Irvine, Irvine, CA, USA and 7Frederick Hutchinson Cancer Research Center, Seattle, WA, USA 1

Haematologica 2021 Volume 106(7):1787-1793

Introduction Seminal papers over the past four decades have described phenotypic driver mutations, such as BCR-ABL1, JAK2, MPL, CALR, KIT and CSF3R, in subsets of myeloproliferative neoplasms (MPN). These mutations affect cytokine signaling or regulation, and result in malignant hematopoiesis.1-3 Such discoveries and the accompanying biological insights have resulted in successful therapeutic approaches for many people diagnosed with MPN, in particular chronic myeloid leukemia (CML), myelofibrosis (MF) and systemic mastocytosis (SM).4-6 The greatest advance has been in CML, with a significant proportion of patients being able to achieve a major molecular response (MMR or MR3; BCR-ABL1 ≤0.1% on the International Scale) following treatment with an ABL1-tyrosine kinase inhibitor (TKI), leading to these patients having lifespans indistinguishable from those of the general population, although the time by which this response milestone should be reached remains controversial.7 Qualified success with significant symptomatic benefit and modest gains in survival have also been achieved in people with MF and SM, following treatment with JAK inhibitors and KIT inhibitors, respectively.6 Some of these achievements have been facilitated by the rational integration of next-generation sequencing (NGS) assays, high-sensitive polymerase chain reaction (PCR) assays on DNA or RNA (sensitivity 0.01%0.1%), and single-cell analyses, in efforts to improve personalized treatment approaches.8 Such efforts have opened a new era of precision medicine for diverse malignancies in which relatively non-specific and often toxic drugs are being replaced by safer and better tolerated agents whose mechanism of action is precisely defined, and for which the treatment algorithm is guided by individual genetic information. Here we examine how molecular testing in MPN can shape diagnosis, monitoring, and treatment algorithms and enable more precise early identification of targeted therapy resistance, particularly in CML (Figure 1). We also discuss the potential impact of persistent or new clonal hematopoiesis on the molecular testing of individuals with MPN and the potential impact on measurable residual disease. This manuscript describes some of the current highlights and challenges related to genetic testing in MPN in 2021.

Correspondence: TARIQ MUGHAL tariq.mughal@tufts.edu Received: September 2, 2020. Accepted: January 13, 2021. Pre-published: March 4, 2021.

Clinically validated tests for detecting and monitoring BCR-ABL1 and KIT mutations

https://doi.org/10.3324/haematol.2020.267252

For patients with CML, a qualitative reverse transcriptase PCR, conducted at diagnosis on peripheral blood cells, enables precise identification of BCR-ABL1 transcripts. Once TKI therapy has commenced, quantitative reverse transcriptase PCR, which adheres to the well-established International Scale, is used for sensitive and accurate monitoring of the levels of BCR-ABL1. However, this has inherent limitations with regard to its lower limit of detection and quantification of BCR-ABL1 transcripts, problems which became important as TKI that were more powerful, compared to imatinib, were developed.9 This, in turn, affected the definition and identification of deep molecular response (DMR), defined as a greater than 4 (MR4), 4.5 (MR4.5) or 5 (MR5) log reduction in BCR-ABL1 transcripts on the International Scale, below the standardized baseline (Figure 2). DMR is now recognized to be of considerable clinical importance to prospectively identify patients likely to remain in remission after discontinuing TKI therapy. Indeed, several current CML guidelines advo-

haematologica | 2021; 106(7)

Material published in Haematologica is covered by copyright. All rights are reserved to the Ferrata Storti Foundation. Use of published material is allowed under the following terms and conditions: https://creativecommons.org/licenses/by-nc/4.0/legalcode. Copies of published material are allowed for personal or internal use. Sharing published material for non-commercial purposes is subject to the following conditions: https://creativecommons.org/licenses/by-nc/4.0/legalcode, sect. 3. Reproducing and sharing published material for commercial purposes is not allowed without permission in writing from the publisher.

1787

T.I. Mughal et al. Figure 1. Treatment response and potential uses of emerging technologies for diagnostics, monitoring and mutation testing in chronic myeloproliferative malignancies. At diagnosis, methods can be used to quickly identify breakpoints useful for designing monitoring assays, as well as other mutations that might influence the initial response to treatment with a tyrosine kinase inhibitor. RNAsequencing can be used to assay specific genes and pathways associated with early response. Single-cell genetics can be used to identify potentially troublesome complex heterogeneity and populations with a resistance signature. During therapy, digital and DNA polymerase chain reaction analyses can be used as more sensitive determinants of deep molecular response, and next-generation sequencing can be used to identify mutations in patients without a deep response who appear to be relapsing. RNA-seq: RNA-sequencing; NGS: nextgeneration sequencing; MRD: measurable residual disease; ITH: intratumoral heterogeneity; PCR: polymerase chain reaction. (Adapted with permission from Radich JP et al.7)

cate the use of a sustained DMR, for at least 2 years, to select patients for consideration of TKI cessation.10,11 At present, there are several tests in the clinic for the detection and monitoring of BCR-ABL1 transcripts, some of which have not been validated robustly; furthermore, they show considerable interlaboratory variations and variable levels of sensitivity. With the current focus on treatmentfree remission (TFR), the importance of using analytically and clinically well-validated tests, preferably approved by regulatory bodies, is being recognized. As an illustration, three recently US Food and Drug Administration (FDA)approved tests appear to perform better than the ‘standard’ quantitative reverse transcriptase PCR tests and may have greater appeal for monitoring very low levels of BCR-ABL1 transcripts. Indeed, it is of interest that over a decade ago, learning from the earlier lessons following the introduction of DNA PCR for BCR-ABL1 and the harmonizing the International Scale BCR-ABL1 transcript measurements, Goldman and colleagues in London, instigated a patientspecific DNA-based method of detection and quantification of an individual patient’s CML clone.12,13 This method involved the rapid identification of BCR-ABL1 fusion junctions by targeted NGS, coupled with the use of a dPCR platform, in patients with very low-level molecular residual disease.. The first FDA-approved test (Asuragen Inc., Austin, TX, USA) is performed with a manufactured kit that can be used on several thermal cyclers, although the FDA approval specifies a specific machine.14 The second approved test is the Cepheid cartridge technique (Cepheid, Sunnyvale, CA, USA), which is attractive given its simplicity and short turnaround time.15 More recently, a water-oil emulsion droplet technology, developed by Bio-Rad (Bio-Rad Laboratories Inc., Hercules, CA, USA), known as digital droplet PCR, was approved by the FDA.16 The digital droplet PCR assay has recently been tested in studies evaluating the feasibility of discontinuing TKI therapy safely in patients with CML who had been in DMR for >2 years.17 The studies documented that, compared with quantitative reverse transcriptase PCR, digital droplet PCR was better at forecasting the 1788

success of TKI discontinuation. Digital droplet PCR technology has several advantages over quantitative reverse transcriptase PCR tests that rely on exponential amplification to estimate the target amount. The digital droplet PCR assay relies on a binary endpoint (yes or no), which is much more lenient to poor RNA quality and inhibitors. The BCRABL1 transcript level is estimated using Poisson distribution based on the number of positive droplets. The Cepheid test is particularly suitable for analysis of smaller sample batches analyzed frequently; for larger batches of samples tested infrequently, the Asuragen or Bio-Rad platforms might be more cost-effective. The use of DNA for CML monitoring is difficult, specifically because the breakpoint on chromosome 9 is vast compared with the ‘small’ breakpoint cluster region on chromosome 22. There are techniques to find the breakpoint using a series of BCR and ABL primers. Once an amplification product has been generated, it can be sequenced, with patient-specific primers generated for subsequent PCR amplification of the BCR-ABL1 breakpoint. However, there may be new technologies that will make breakpoint detection much faster. Breakthroughs in ‘realtime’ sequencing, such as Pacific Bioscience and Nanopore technology, which can read exceeding long sequences at breathtaking speeds, potentially allow BCR-ABL1 breakpoint detection to be performed with a single sequencing run.8 In CML, there are two variants of the BCR-ABL1 transcript, depending on whether the break in BCR occurs in the intron between exons e13 and e14, or in the intron between exons e14 and e15.6 A break in the former intron yields an e13a2 mRNA junction and a break in the latter intron yields an e14a2 junction. Most patients have transcripts with features of either e13a2 or e14a2, but occasional patients have both transcripts present in their leukemia cells. The prognostic significance of the precise type of BCR-ABL1 transcript is now being increasingly recognized in efforts to achieve DMR and potential TFR following successful TKI treatment. Earlier studies in patients treated with imatinib had suggested that patients with e14a2 tranhaematologica | 2021; 106(7)

Genetic-based personalized treatment of MPN

Figure 2. The International Scale for quantitative reverse transcriptase polymerase chain reaction analysis of BCR-ABL1 transcripts. IRIS: International Randomized Study of Interferon and STI571; MMR: major molecular response; MR4, MR4.5 and MR5: 4-, 4.5and 5-log reductions, respectively, from the IRIS standardized baseline; Ph: Philadelphia chromosome; RT-qPCR: quantitative reverse transcriptase polymerase chain reaction.

scripts had deeper and faster responses compared to those with e13a2 transcripts.18 Subsequently, Baccarani and colleagues observed lower response rates and inferior outcomes following nilotinib treatment in patients with e13a2 transcripts.19 These investigators also noted a relationship of transcript type with age and gender. More recently, a French study, led by Genthon and colleagues, also documented deeper and faster responses, in terms of achieving MR3 and MR4, in patients expressing e14a2 compared with those expressing e13a2 transcripts.20 The differences in clinical outcomes based on transcript subtype has so far only been investigated in small studies and this issue needs to be assessed prospectively in larger cohorts receiving any of the five currently licensed TKI treatments for CML. A study of 20 patients who were in TFR for longer than 1 year, having achieved sustained MR4.5 following treatment with TKI, suggested the potential importance of the lineage of measurable residual disease as a potential predictive biomarker of TFR outcome.21 The investigators, Pagani and colleagues, used fluorescence-activated cell sorting of CML cells known to express BCR-ABL1 mRNA, granulocytes, monocytes, B cells T cells and natural killer cells; BCR-ABL1 DNA PCR was used to investigate the lineage of these residual CML cells. The observation of BCR-ABL1 DNA being present only in B- and T-lymphocytes begs the question of the CML cell lineage contributing to the success of TFR. The study could not, however, establish the absence of CML stem cells (CD34+38–26+) in a cohort of patients in TFR with undetectable BCR-ABL1 mRNA transcripts, which are considered to account for the loss of TFR in such patients.22 Regardless, larger studies assessing the lineage of measurable residual disease in TFR patients are warranted. In patients with SM, the identification and quantification of the KIT D816V mutation in hematopoietic stem and progenitor cells, as well as mature mast cells, by highly sensitive (sensitivity <0.01%) and specific assays, such as allelespecific oligonucleotide quantitative PCR and digital droplet PCR, allows an accurate diagnosis.23-26 This information is also useful for risk stratification, complementing conventional biomarkers such as serum tryptase level and percentage of bone marrow infiltration by mast cells, and may be used for monitoring patients receiving treatment. More recently, the use of targeted NGS panels for the characterihaematologica | 2021; 106(7)

zation of associated gene mutations, such as SRSF2, ASXL1 and RUNX1, has improved prognostication and has led to the development of novel prognostic scoring systems to optimize clinical management.27

Testing for BCR ABL kinase domain mutations Mutation testing for BCR-ABL1 kinase domain mutations and the co-existence of subclones in general can be assessed by a variety of technologies. In general, the more sensitive the test, the more complex and expensive it is. Sanger sequencing has a low error rate but a poor sensitivity of only 10-20%. In contrast, NGS has a better sensitivity of roughly 1% and is useful for the identification of compound mutations.28 However, the error rate associated with the library amplification and preparation can be up to one in a 1,000 base pairs, particularly when sequencing from mRNA, which relies on the error-prone reverse transcriptase. Barcode correction techniques may improve this, but the best technique may be so-called duplex sequencing, which is novel in that it sequences both DNA strands, dramatically reducing the error rate since mutations are only called if the complementary base change is seen on the other strand. Indeed, recent reports suggest that low frequency mutations that are detected by NGS but have unpredictable clinical courses (e.g., disappear spontaneously in some patients) may represent artifacts from the error rate inherent in NGS.29 Soverini and colleagues recently conducted the first prospective, ‘real-world’ assessment of NGS-based BCR-ABL1 knockdown mutation testing compared with Sanger sequencing in a large cohort of consecutively studied CML patients in whom TKI had failed or who were in the ‘warning’ category described by the European Leukemia Net (ELN) guidelines.30 The researchers demonstrated the importance of low-level mutations, defined as mutations with a variant allele frequency of 320%, and their clinical relevance (Figure 3). These observations are the basis for guiding genomic-based personalized therapies for CML further, in particular through the identification of high-risk individuals prior to the ELN ‘warning’ stage. These tests should also improve the efficiency and safety of clinical trials designed to reduce the risk of blast transformation in patients who respond suboptimally to TKI. Important challenges now are to improve sensitivity 1789

T.I. Mughal et al.

further, reduce the turnaround time and lower the costs, which remain high.

Intratumoral heterogeneity Studies of hematologic malignancies as well as diverse solid tumors have revealed a surprising amount of intratumor heterogeneity, i.e., the finding of multiple, related clones rather than one uni-clonal monolith. Thus, the kinetics of disease progression, response and relapse follow the rules of Darwinian selection. Many neoplasms, including CML, MF and advanced SM, have been found to exhibit greater clonal complexity than previously thought, as new myeloid mutations have been found in these diseases.31 For example, Jawaher and colleagues identified the emergence of the KIT D816V mutation as a distinct and late event in patients with multi-mutated advanced SM.32 Trying to map intratumor heterogeneity by the sequencing of bulk populations is limited by the simple fact that one is sequencing the average mutation frequency of all the

cells from various clones. New technologies allow for the sequencing of RNA (e.g., 10x Genomics) or DNA (e.g., Mission Bio) from single cells. Major advantages of singlecell technologies include the higher resolution offered to understand the types of cells present, to detect rarer cell populations and to study their function (inferred by RNA expression) or clonal structure (inferred by mutation pattern). Disadvantages, other than the financial costs, are that each cell is a ‘one and done’ experiment. Furthermore, it can be difficult to determine real signal versus experimental noise, which is especially problematic with RNA, for which simple factors such as time from sample acquisition to experiment can influence gene expression. Recent work using single-cell RNA-sequencing has garnered considerable novel insight into normal and aberrant hematopoiesis, cell-cell interactions, characterization of bone marrow and immune (non-clonal) cells as well as tissue stroma and leukemia-initiating cells.33-36 This technology enables detection and characterization of intratumor heterogeneity and provides much needed granularity to key issues, including acquisition of individual or specific combi-

1790

Figure 3. Comparison between Sanger sequencing and next-generation sequencing – the NEXT-in-CML study. (A) Percentage of patients positive for mutations, as determined by Sanger sequencing (SS) and by next-generation sequencing (NGS). Among patients positive for mutations by NGS, 31 (13.1%) had high-level mutations only (≥20%; detectable by SS too); 29 (12.3%) had both ≥1 high-level mutations and ≥1 low-level mutations (≤20%; detectable by NGS only); 51 (21.6%) had only lowlevel mutations. A low-level T315I mutation was detected in ten patients; 59 additional patients had ≥1 low-level mutations known to be associated with resistance to imatinib or second-generation tyrosine kinase inhibitors other than the T315I mutation (Y253H; E255K/V; V299L; F317L/V/I/C; F359V/I/C). The remaining ten patients had only low-level mutations with an unknown resistance profile and/or not listed in the COSMIC database. (B) Patients positive for one or multiple mutations as assessed by SS versus NGS. CML: chronic myeloid leukemia; pts: patients; IMA: imatinib; DAS: dasatinib; NIL: nilotinib; BOS: bosutinib. (Adapted, with permission, from Soverini S et al.30).

haematologica | 2021; 106(7)

Genetic-based personalized treatment of MPN

nations of somatic mutations or proteins expressed by cellular subtypes and mechanisms of relapse or treatment resistance. As an illustration, the use of single-cell-level sequencing in a study of MF, conducted by Psaila and colleagues, revealed megakaryocyte-biased hematopoiesis with megakaryocyte progenitors demonstrating distinct inflammatory and metabolic signatures, and increased expression of the surface antigen G6B (MPIG6B) on MF stem cells and progenitors (Figure 4).36 These findings raise the possibility of new therapies for MF, which could target

the MF clone and MF-associated fibrosis. In CML, insights into the intratumoral heterogeneity of CML stem cells has revealed subgroups with distinct molecular signatures that are resistant to TKI.34 Furthermore novel molecular pathways related metabolism and the bone marrow microenvironment are being deciphered. Patients with advanced SM, in particular those with an associated hematologic malignancy, often have multilineage involvement by KIT and multimutated clones, which are associated with a poorer prognosis.27 Additionally, through single-cell analysis,

Figure 4. Single-cell -omics demonstrate a trajectory for megakaryocyte-biased hematopoiesis in myelofibrosis. (A–D) Force-directed graphs (FDG) for aggregates of all control + myelofibrosis (MF) cells (A), MF only (B), control only (C), and control + down-sampled MF dataset (D). In (D), the left graph shows the lineage signature gene score and in the right graph cells are colored according to the donor type (healthy donors, blue; MF patient, red). Gene expression trajectories are visualized by superimposing the expression scores of lineage signature gene sets on the FDG. Gray cells represent uncommitted hematopoietic stem and progenitor cells or cells with expression of more than one lineage signature. (Published, with permission, from Psalia B, et al.36)

haematologica | 2021; 106(7)

1791

T.I. Mughal et al.

Grootens and colleagues identified KIT 816V in early SM stem cells and progenitors, suggesting that this biomarker may not be restricted to the mast cell lineage.37 In the same study it was observed that the mutation frequency was 100% in mature CD45RA+ mast cells.

Screening for clonal hematopoiesis of indeterminate potential NGS-based genetic analysis of large unselected populations assessing acquisition of somatic mutations has provided evidence of age-related clonal expansion of recurrent mutations in known oncogenes in hematopoietic stem and progenitor cells in the absence of overt hematologic malignancies, a condition termed clonal hematopoiesis.38 The somatic mutations that define clonal hematopoiesis tend to be similar to those found in hematologic malignancies, such as ASXL1, DNMT3A, RUNX1, JAK2, TET2, SF3B1 and others, with a variant allele frequency >2% in subjects without cytopenias or a history of a myeloid neoplasm. The mutations are typically present in circulating granulocytes, monocytes, and natural killer cells; they can also affect B cells and, rarely, T cells. Clonal hematopoiesis is considered to represent the early steps of leukemogenesis and is associated with an increase in the risk of myeloid and lymphoid malignancies of 0.5% to 1.0% per year.39 Evidence from retrospective case-control studies supports the role of clonal hematopoiesis in therapy-related myeloid neoplasms, too.40 The mutations can also be associated with acquired drug resistance. Somewhat surprisingly, clonal hematopoiesis is also associated with a pro-inflammatory state and an increased risk of various non-hematologic diseases, in particular cardiovascular disease, attributed to genes that are involved in regulating inflammation and accelerating atherosclerosis.41,42 Interestingly, Hameisterr and colleagues recently investigated whether clonal hematopoiesis might affect the course of COVID-19 in hospitalized older patients who had tested positive for SARS-CoV-2 infection and found no significant association.43 Although it is recognized that people with clonal hematopoiesis develop MPN at a higher rate than those without mutations, the precise impact of clonal hematopoiesis on the prognosis and treatment of MPN is an enigma.44 An important challenge is, therefore, to assess the clinical relevance of clonal hematopoiesis at the time of diagnosis of MPN and assess its potential prognostic value, in particular in the transformation to acute leukemia and in treatment resistance. The presence of clonal hematopoiesis may also affect the ‘real-world’ situation in people with CML, MF and SM who have been treated ‘successfully’, including those who have undergone allogeneic stem cell transplantation, when such mutations could be donorderived and influence the assessment of measurable residual disease.45 Genomic studies, in particular those involving single-cell sequencing, are being increasingly integrated into the investigation of MPN at diagnosis and transformation, which also raises the question of which genes should ideally be included in the NGS panel. Furthermore, we are beginning to fathom the complexity of the cancer tissue ecosystem and how this is affected by different features, such as the cells’ metabolism and how it co-opts normal genes, stromal and immune cells within the microenvironment, among other variables. In this regards the recent work of Van Etten, Krause and others on the specific, tar1792

getable interactions with the microenvironment in people with imatinib-resistant CML is important.46 Despite the therapeutic advances in CML, MF and SM, the outlook for people whose disease transforms remains bleak and highlights the need for suitable prognostic scores to identify those at high risk of progression who may benefit from more intensive initial treatments, including allogeneic stem cell transplantation.

Future prospects We have clearly made significant progress by examining MPN through genetics and physiology, by the unprecedented application of ‘-omics’ technology, ultrasensitive sequencing technology and single-cell genomics. Such approaches have arguably enhanced our understanding of chronic myeloproliferative malignancies, and the characterization of the underlying intratumor heterogeneity and the ability of the neoplastic clone to evolve and adapt has been recognized as a principal challenge for targeted treatments and immunotherapies.47 The different genetic tests in MPN clinics have undoubtedly improved our ability to monitor patients effectively and have refined diagnostic risk stratification.48 For CML patients, they also enhance the probability of TKI cessation and achieving TFR; for patients with MF and SM, they are complementary to the World Health Organization 2016 diagnostic criteria and help in navigating treatment decisions. These tests also allow for a better selection of targeted agents to be tested in subgroups with variant somatic mutations. Nevertheless, much work remains. For example, we have little understanding of the cell-intrinsic and -extrinsic mechanisms underpinning the transformation of MPN into acute leukemia, or the mechanisms of resistance to the newer inhibitors, to mention a few.49 The emerging picture is complex but has created a platform upon which to build novel therapeutic approaches. Disclosures DJDA has provided consultancy/scientific advisory board services for Amgen, Autolos, Agios, Blueprint, Forty-Seven, Incyte, Jazz, Novartis, Pfizer, Shire, and Takeda and has received research funding from Abbvie, Glycomimetics, Novartis and Blueprint Pharmaceuticals. TIM has stocks or employment connections with Stemline; and receives royalties from Oxford University Press and Informa. BP has provided consultancy services for Novartis and Constellation. JPR has been part of scientific advisory boards for Novartis, Amgen, BMS, and Jazz Pharmaceuticals; and performed laboratory contract work for Novartis. GS has no relevant disclosures to make. RAVE has provided consultancy services for Novartis. Contributions TIM, BP and JPR wrote the draft versions of the manuscript, without any writing assistance provided by a third party. All authors edited and approved the final version of the manuscript. Acknowledgments The authors would like to thank Dr Alpa Parmar and participants of the XIV Post ASH MPN workshop for their assistance. Funding The authors would like to thank Alpine Oncology Foundation for funding. haematologica | 2021; 106(7)

Genetic-based personalized treatment of MPN

References 1. Vainchenker W, Kralovics R. Genetic basic and molecular pathophysiology of classical myeloproliferative neoplasms. Blood. 2017; 129(6):667-679. 2. Grinfeld J, Nangalia J, Baxtyer EJ, et al. Classification and personalized prognosis in myeloproliferative neoplasms. N Engl J Med. 2018;379(15):1416-1430. 3. Mughal TI, Gotlib J, Mesa R, et al. Recent advances in the genomics and therapy of BCR/ABL1-positive and -negative chronic myeloproliferative neoplasms. Leuk Res. 2018;67:67-74. 4. Verstovsek S, Mesa RA, Gotlib J, et al. A double-blind, placebo-controlled trial of ruxolitinib for myelofibrosis. N Engl J Med. 2012;366(9):799-807. 5. Arock M, Sotlar K, Gotlib J, et al. New developments in the field of mastocytosis and mast cell activation syndromes: a summary of the Annual Meeting of the European Competence Network on Mastocytosis (ECNM) 2019. Leuk Lymphoma. 2020;61(5): 1075-1083. 6. Mughal TI, Radich JP, Deininger MW, et al. Chronic myeloid leukemia: reminiscences and dreams. Haematologica. 2016;101(5): 541-558. 7. Smith G, Apperley J, Milojkovic D, et al. A British Society for Haematology Guideline on the diagnosis and management of chronic myeloid leukaemia. Br J Haematol. 2020;191(2):171-193. 8. Radich J, Yeung C, Wu D. New approaches to molecular monitoring in CML (and other diseases). Blood. 2019;134(19):1578-1584. 9. Mughal TI, Pemmaraju N, Radich JP, et al. Emerging translational science discoveries, clonal approaches, and treatment trends in chronic myeloproliferative neoplasms. Hematol Oncol. 2019;37(3):240-252. 10. Hochhaus A, Baccarani M, Silver RT, et al. European Leukemia Net 2020 recommendations for treating chronic myeloid leukemia. Leukemia. 2020;34(4):1495-1502. 11. Deininger MW, Shah NP, Altman JK, et al. Chronic myeloid leukemia - NCCN guidelines -version 2.2021. J Natl Compr Canc Netw. 2020;18(10):1385-1415. 12. Alikian M, Ellery P, Forbes M, et al. Next generation sequencing-assisted DNA-based digital PCR for a personalized approach to the detection and quantification of residual disease in chronic myeloid patients. J Mol Diagn. 2016;18(2):176-189. 13. Hughes T, Deininger M, Hochhaus A, et al. Monitoring CML patients responding to treatment with tyrosine kinase inhibitors: review and recommendations for harmonizing current methodology for detecting BCRABL transcripts and kinase domain mutations and for expressing results. Blood. 2006;108(1):28-37. 14. Brown JT, Laosinchai-Wolf W, Hedges JB, et al. Establishment of a standardized multiplex assay with the analytical performance required for qualitative measurement of BCR-ABL1 on the international reporting scale. Blood Cancer J. 2011;1(3):e13. 15. Radich JR. Chronic myeloid leukemia: Global impact from a local laboratory. Cancer. 2017;123(14):2594-2596. 16. Yan D, Pomicter AD, O’Hare T, Deininger

haematologica | 2021; 106(7)

MW. ddeeper than deep: can ddPCR predict successful imatinib cessation? Clin Cancer Res. 2019;25(22):6561-6563. 17. Hochhaus A, Breccia M, Saglio G, et al. Expert opinion- management of chronic myeloid after resistance to second-generation tyrosine kinase inhibitors. Leukemia. 2020;34(6):1495-1502. 18. Ercaliskan A, Eskazan AE. The impact of BCR-ABL1 transcript type on tyrosine kinase inhibitor responses and outcomes in patients with chronic myeloid leukemia. Cancer. 2018;124(19):3806-3818. 19. Baccarani M, Castagnetti F, Gugliotta G, et al. The proportion of different BCR-ABL1 transcript types in chronic myeloid leukemia. An international overview. Leukemia. 2019;33(5):1173-1183. 20. Genthon A, Nicolini FE, Huguet F, et al. Infleunce of major BCR-ABL1 transcript subtype on outcome in patients with chronic myeloid leukemia in chronic phase treated frontline with nilotinib. Oncotarget. 2020;11(26):2560-2570. 21. Pagani IS, Dang P, Saunders V, et al. Lineage of measureable residual disease in patients with chronic myeloid leukemia in treatment-free remission. Leukemia. 2020;34(4): 1052-1061. 22. Bocchia M, Sicuranza A, Abruzzese E, et al. Residual peripheral blood CD26+ leukemic stem cells in chronic myeloid leukemia patients during TKI therapy and during treatment-free remission. Front Oncol. 2018;8:194. 23. Arock M, Sotlar M, Akin C, et al. KIT mutation analysis in mast cell neoplasms: recommendations of the European Competence Network on Mastocytosis. Leukemia. 2015;29(6):1223-1232. 24. Greiner G, Gurbisz M, Ratzinger F, et al. Digital PCR : a sensitive and precise method for KIT D816V quantification in mastocytosis. Clin Chem. 2018;64(3):547-555. 25. Reiter A, George TI, Gotlib J. New developments in diagnosis, prognostication, and treatment of advanced systemic mastocytosis. Blood. 2020;135(16):1365-1376. 26. Martelli M, Monaldi C, De Santis S, et al. Recent advances in the molecular biology of systtemic mastocytosis: implications for diagnosis, prognosis, and therapy. Int J Mol Sci. 2020;21(11):3987. 27. Jawhar M, Schwaab J, Hausmann D, et al. Splenomegaly, elevated alkaline phosphatase and mutations in SRSF2/ASXL1/ RUNXL1 gene panel are strong adverse prognostic markers in patients with systemic mastocytosis. Leukemia. 2016;30(12): 2342-2350. 28. Wu D, Sherwood A, Fromm JR, et al. Highthroughput sequencing detects minimal residual disease in acute T lymphoblastic leukemia. Sci Transl Med. 2012;4(134): 134ra63. 29. Sasine JP, Schiller GJ. Acute myeloid leukemia: how do we measure success? Curr Hematol Malig Rep. 2016;11(6):528-536. 30. Soverini S, Bavaro L, De Benedittis C, et al. Prospective assessment of NGS-detectable mutations in CML patients with nonoptimal response: the NEXT-in-CML study. Blood. 2020;135(8):534-541. 31. Lee J, Godfrey AL, Nangalia J. Genomic heterogeneity in myeloproliferative neoplasms

and applications to clinical practice. Blood Rev. 2020;42:100708. 32. Jawhar M, Schwaab J, Schnittger S, et al. Molecular profiling of myeloid progenitor cells in multi-mutated advanced systemic mastocytosis identifies KIT D816V as a distinct and late event. Leukemia. 2015;29 (5):1115-1122. 33. Tang F, Barbacioru C, Wang Y, et al. mRNASeq whole-transcriptome analysis of a single cell. Nat Methods. 2009;6(5):377-382. 34. Psaila B, Mead AJ. Single-cell approaches reveal novel cellular pathways for megakaryocyte and erythroid differentiation. Blood. 2019;133(13):1427-1435. 35. Morris V, Marion W, Hughes T, et al. Singlecell analysis reveals mechanisms of plasticity of leukemia initiating cells. bioRxiv. 2020; April 30. [Epub ahead of print] 36. Psaila B, Wang G, Rodriguez-Meira A, et al. Single-cell analyses reveal megakaryocytebiased hematopoiesis in myelofibrosis and identify mutant clone-specific targets. Mol Cell. 2020;78(3):477-492. 37. Grootens J, Ungerstedt JS, Ekoff M, et al. Single-cell analysis reveals the KIT D816V mutation in haematopoietic stem and progenitor cells in systemic mastocytosis. EBioMedicine. 2019;43:150-158. 38. Mead A. Single cell genomics in chronic myeloid leukemia. Hemasphere. 2018;2(S2): 54-55. 39. 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. 40. Hasserjian RH, Steensma DP, Graubert TA, Ebert BL. Clonal hematopoiesis and measureable disease assessment in acute myeloid leukemia. Blood. 2020;135(20): 1729-1738. 41. Warren JT, Link DC. Clonal hematopoiesis and risk for hematologic malignancy. Blood. 2020;136(14):1599-1605. 42. Jaiswal S, Natarajan P, Ebert BL. Clonal hematopoiesis and atherosclerosis. N Engl J Med. 2017;377(14):1401-1402. 43. Hameister E, Stolz SM, Fuhrer Y, et al. Clonal hematopoiesis in hospitalized elderly patients with COVID-19. Hemasphere. 2020;4(4):e453. 44. Jaiswal S. Clonal hematopoiesis and nonhematologic disorders. Blood. 2020;136(14): 1606-1614. 45. Boettcher S, Wilk CM, Singer J, et al. Clonal hematopoiesis in donors and long-term survivors of related allogeneic hematopoietic stem cell transplantation. Blood. 2020;135 (18):1548-1559. 46. Kumar R, Pereira RS, Zanetti C, et al. Specific, targetable interactions with the microenvironment influence imatinib-resistant chronic myeloid leukemia. Leukemia. 2020;34(8):2087-2101. 47. Nangalia J, Mitchell E, Green AR. Clonal approaches to understand the impact of mutations on hematologic disease development. Blood. 2019;133(13):1436-1455. 48. Branford S, Kim DDH, Apperley JA, et al. Laying the foundation for genomicallybased risk assessment in chronic myeloid leukemia. Leukemia. 2019;33(8):1835-1850. 49. Dunbar A, Rampal R, Levine R. Leukemia secondary to myeloproliferative neoplasms. Blood. 2020;136(1):61-70.

1793

REVIEW ARTICLE Ferrata Storti Foundation

Dose intensity for conditioning in allogeneic hematopoietic cell transplantation: can we recommend “when and for whom” in 2021? Nico Gagelmann and Nicolaus Kröger

Department of Stem Cell Transplantation, University Medical Center HamburgEppendorf, Hamburg, Germany

Haematologica 2021 Volume 106(7):1794-1804

ABSTRACT

Correspondence:

llogeneic hematopoietic stem-cell transplantation is a potentially curative therapy for various hematologic diseases. An essential component of this procedure is the pre-transplant conditioning regimen, which should facilitate engraftment and reduce or eliminate tumor cells. The recognition of the substantial association of a graft-versus-tumor effect and the high toxicity of the commonly used conditioning regimen led to the introduction of more differentiated intensity strategies, with the aim of making hematopoietic stem-cell transplantation less toxic and safer, and thus more applicable to broader populations such as older or unfit patients. In general, prospective and retrospective studies suggest a correlation between increasing intensity and nonrelapse mortality and an inverse correlation with relapse incidence. In this review, we will summarize traditional and updated definitions for conditioning intensity strategies and the landscape of comparative prospective and retrospective studies, which may help to find the balance between the risk of non-relapse mortality and relapse. We will try to underscore the caveats regarding these definitions and analyses, by missing complex differences between intensity and toxicity as well as the broad influences of other factors in the transplantation procedure. We will summarize evidence regarding several confounders which may influence decisions when selecting the intensity of the conditioning regimen for any given patient, according to the individual risk of relapse and non-relapse mortality.

NICOLAUS KRÖGER nkroeger@uke.uni-hamburg.de

Categorized traditional and updated definitions Received: December 1, 2020. Accepted: February 2, 2021. Pre-published: March 18, 2021. https://doi.org/10.3324/haematol.2020.268839

©2021 Ferrata Storti Foundation Material published in Haematologica is covered by copyright. All rights are reserved to the Ferrata Storti Foundation. Use of published material is allowed under the following terms and conditions: https://creativecommons.org/licenses/by-nc/4.0/legalcode. Copies of published material are allowed for personal or internal use. Sharing published material for non-commercial purposes is subject to the following conditions: https://creativecommons.org/licenses/by-nc/4.0/legalcode, sect. 3. Reproducing and sharing published material for commercial purposes is not allowed without permission in writing from the publisher.

1794

Although full consensus has not been reached within the hematopoietic stemcell transplantation (HSCT) community,1 conditioning regimens have usually been classified as high-dose (MAC), reduced-intensity (RIC), and non-myeloablative (NMA).2 Based on these criteria,3 myeloablative, or “high-dose” regimens, consisting of alkylating agents with or without total body irradiation (TBI), are expected to ablate marrow hematopoiesis, not allowing autologous hematologic recovery. In contrast, NMA regimens, although causing minimal cytopenia, do not require stem cell support.4 Regimens that do not fit the definition of MAC or NMA are classified as RIC regimens: they result in potentially prolonged cytopenia, and they require hematopoietic stem cell support. What differentiates RIC regimens from myeloablative regimens is that the dose of alkylating agents or TBI is generally reduced by ≥30%. Notably, “intensity” was defined here on the basis of grade of reversible and irreversible myelotoxicity rather than of non-hematologic toxicity. Despite this imprecise definition of intensity and toxicity and the lack of their universality (agreement for these criteria was found in <75%), this classification has served as a clinical tool and enabled some, although still limited, comparability with registries such as that of the European Society for Blood and Marrow Transplantation (EBMT).5–7 It is important to recognize, however, that regimens classified as MAC or RIC can vary substantially, regarding intensity and toxicity, which is also reflected by so-called “sequential” strategies;8,9 and the rigidity of this scheme further impeded haematologica | 2021; 106(7)

Conditioning intensity in allogeneic HSCT

the inclusion of new regimens showing reduced nonhematologic toxicity and different safety profiles in general.10 Moreover, this classification ignores the additional intensity of purine analogs used for immune-ablation11 or of disease-specific drugs used to achieve reduction in relapse risk. Thus, in literature less well-defined terms, such as “reduced toxicity”,12 “hyper-intensive”13 or “augmented reduced intensity” are now more commonly used.14 To address this increasing variability, the EBMT has recently proposed an updated refined classification, assigning intensity weight scores for frequently used conditioning regimen components in relation to their prognostic value for non-relapse mortality (NRM); and using their sum to generate a transplant conditioning intensity score. This also categorized classification showed only slightly improved discrimination for the outcome of NRM, while discrimination and thus prognostic utility with respect to the outcome of relapse were comparable to the previous dichotomized RIC/MAC classification.15

Prospective randomized studies with different dose intensities Prospective randomized studies comparing different condition regimen intensities exist only for acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) (including very few patients with chronic myeloid leukemia; see Table 1), while no prospective randomized studies exist for other hematologic malignancies includ-

ing non-Hodgkin lymphoma, Hodgkin lymphoma and multiple myeloma.

Toxicity-reduced myeloablative conditioning versus myeloablative conditioning The first trial of toxicity-reduced MAC versus MAC was a randomized phase III trial by Bornhäuser et al. in 195 patients (median age 44 years) with intermediate- or high-risk AML in first complete remission.16 Treatment in the lower intensity arm consisted of a lower dose of TBI (800 cGy) plus fludarabine which was compared to a standard MAC approach of cyclophosphamide-TBI (1200 cGy). Although the study was concluded early because of slow accrual of patients, outcomes were not significantly different with regards to NRM, cumulative incidence of relapse, and survival, with these findings being confirmed in a long-term follow-up report.17 Interestingly, severe mucositis and in-hospital mortality were less frequent in the RIC group, leading to the conclusion that RIC regimens lessened the toxic effects of transplantation, and the short-term mortality at 1 year was lower in the RIC group. An age limit of 60 years may impede the interpretation of the findings. In contrast, these results suggest that perhaps RIC regimens should be used preferentially in patients <60 years old with AML in first complete remission. Aside from this, the RIC regimen of TBI 800 cGy is, according to the current definition, still a MAC regimen . Another attempt to reduce intensity and toxicity without losing myeloablative intensity of the conditioning regimen was made by replacing cyclophosphamide with

Table 1. Characteristics and results of prospective randomized trials comparing different intensities and toxicity of conditioning regimens.

Trial

Population

Regimen

Toxicity reduced MAC vs. MAC or RIC Rambaldi et al.20 AML BuFlu (MAC) BuCy (MAC) Age >40 y Bornhäuser et al.16 AML CR1 8 GyTBIFlu (MAC) 12 GyTBI/Cy (MAC) Age 18-60 y IR/HR cytogenetics Beelen et al.24 AML/MDS TreoFlu (MAC) BuFlu (RIC) Age ≥50 y and/or CI >2/KPS >60% RIC vs. NMA Blaise et al.2 Hematologic BuFlu (RIC) FluTBI (NMA) malignancies RIC vs. MAC Ringdén et al.90 AML/CML BuFlu (RIC) BuCy(MAC) Age ≤60 y incl n=4 CML (NMA) Scott et al.26 AML/MDS in CR BuFlu; FluMel BuFlu; BuCy; Age 18-65 y (RIC) TBICy (MAC) Kröger et al.25 MDS/sAML BuFlu (RIC) BuCy (MAC) Age 18-60 y UD Age 18-65 RD RIC vs. sequential RIC Craddock et al.14 AML /MDS FLAMSA-Bu Bu/Flu or Age 18-75 y (seq RIC) Mel/Flu (RIC)

RFS % (P)

Relapse % (P)

NRM % (P)

OS % (P)

40 vs. 47 (ns) 58 vs. 56 (ns)

24 vs. 21 (ns) 28 vs. 26 (ns)

8 vs. 18 (0.03) 13 vs. 18 (ns)

27 vs. 35 (ns) 61 vs. 58 (ns)

64 vs. 50 (0.001)

25 vs. 23 (ns)

11 vs. 23 (0.05)

71 vs. 56 (0.01)

35 vs. 23 (ns)

27 vs. 54 (<0.01)

38 vs. 22 (0.03)

41 vs. 41 (ns)

NR 47 vs. 68 (<0.01) 62 vs. 58 (ns)

12 vs. 35 (ns) 48 vs. 14 (<0.001) 17 vs. 15 (ns)

11 vs. 11 (ns) 4 vs. 16 (<0.01) 17 vs. 25 (ns)

76 vs. 62 (ns) 78 vs. 68 (0.07) 76 vs. 63 (0.08)

54 vs. 49 (ns)

27 vs. 30 (ns)

21 vs. 17 (ns)

61 vs. 59 (ns)

RIC: reduced intensity conditioning; MAC: myeloablative conditioning; NMA: nonmyeloablative; RFS: relapse-free survival; NRM: non-relapse mortality; OS: overall survival; (s)AML: (secondary) acute myeloid leukemia; CML: chronic myeloid leukemia; CR: complete remission; Cy: cyclophosphamide; Treo: treosulfan; Flu: fludarabine; TBI: total body irradiation; Bu: busulfan; IR: intermediate-risk; HR: high-risk; Mel: melphalan; MDS: myelodysplastic syndrome; UD: unrelated donor; RD: related donor; ns: not significant; y: years; CI: comorbidity index; KPS: Karnofsky performance status; NR: not reported; seq: sequential.

haematologica | 2021; 106(7)

1795

N. Gagelmann and N. Kröger

a less toxic immunosuppressive agent, fludarabine.18,19 The important prospective randomized study by Rambaldi et al. comparing busulfan-fludarabine conditioning versus busulfan-cyclophosphamide in patients >40 years of age with AML showed significantly reduced 1year transplant-related mortality (8% vs. 17%).20 Importantly, the comparison showed reduced NRM specifically for patients with a higher comorbidity index and busulfan-fludarabine conditioning, as well as in patients in first complete remission. Adverse event rates were similar in the two groups, except for organ-failure, which was more frequent in the group treated with cyclophosphamide. Another option to maintain myeloablation, according to the given definition, and immunosuppression but also to reduce non-hematologic toxicity was investigated by replacing busulfan by the alkylator treosulfan,21 which exhibits low inter- and intra-patient variability without the need for dose adjustments.22,23 A recent prospective randomized trial by Beelen et al. in older (≥50 years) and/or comorbid AML/MDS patients randomly assigned patients to receive either intravenous 3x10 g/m2 treosulfan or reduced intensity busulfan. The initial treosulfan dose of 14 g/m2 daily was changed due to safety concerns.24 Both groups received 30 mg/m² intravenous fludarabine. Overall, the 2-year event-free survival rate was 64% in the treosulfan group and 50% in the busulfan group, but differences were most pronounced in the subgroup of patients ≥50 years receiving matched unrelated HSCT, whereas there was not a significant difference among patients with a comorbidity index of ≥2. Notably, despite higher intensity, the survival benefit of treosulfan was caused by a higher NRM in the busulfan group which did, however, appear somewhat higher than previously reported.25,26

Reduced intensity conditioning versus myeloablative conditioning The BMT-CTN study reported by Scott et al.26 prospectively compared RIC versus MAC approaches in AML and MDS. In order to have more flexibility, transplant physicians had some choice of preparative approaches. The study design allowed higher-dose busulfan (12.8 mg/kg intravenously) with fludarabine or busulfan with cyclophosphamide along with cyclophosphamide-TBI in the MAC arm and lower-dose busulfan (8 mg/kg intravenously) with fludarabine or fludarabine with melphalan in the RIC arm. Enrollment was more rapid and accured 272 patients with AML/MDS patients aged 18-65 years with a comorbidity index <5 and <5% marrow myeloblasts prior to matched-related or unrelated donor HSCT. The study was closed early because of the finding of superior relapse-free survival in the MAC arm. This study clearly demonstrated that myeloablative busulfan regimens resulted in a significantly improved relapse-free survival (despite a higher NRM rate) and in a significantly lower relapse incidence than the lower-dose busulfan/fludarabine RIC arm. The main cause of death in the MAC arm was graft-versus-host disease (GvHD) (50%), as compared to relapse (86%) in the RIC arm. The ability to perform subgroup analyses was limited by the early closure. Nevertheless, these analyses showed survival benefit for MAC for AML patients, high-risk patients and patients with a comorbidity index of 0. The EBMT RICMAC trial25 prospectively addressed this 1796

issue in patients with MDS or secondary AML. The patients included in this trial had to be 18-60 years for those with unrelated donors and 18-65 years for those with related donors. Eighty-five percent of chemotherapies before transplantation were administered in advanced MDS (chronic myelomonocytic leukemia, refractory anemia with excess of blasts, and secondary AML) to reduce the number of blasts. Regimens were busulfan (16 mg/kg orally or 12.8 mg/kg intravenously) and cyclophosphamide (120 mg/kg) for MAC and busulfan (8 mg/kg orally or 6.4 mg/kg intravenously) and fludarabine (150 mg/m2) for RIC. The trial also accrued slowly, assigning 129 patients, and was closed early after calculations suggested enough power to address the primary aim of determining differences in NRM. The trial showed similar 2-year incidences of relapse, relapse-free survival, and overall survival. Short-term NRM, at 1 year, was also similar but was much lower than predicted. In the multivariable model of NRM, an interaction was found between conditioning intensity and cytogenetics, which led to a subgroup analysis stratified by cytogenetic risk group. In the low-risk cytogenetic group, lower performance status was associated with higher NRM, while the comparison of conditioning intensities showed lower NRM after RIC in this risk group. In the intermediate- and high-risk cytogenetic groups, RIC resulted in a higher NRM rate.

Reduced intensity conditioning versus sequential reduced intensity conditioning More recently, Craddock et al. compared, in a prospective, randomized fashion, a sequential transplant regimen with fludarabine-amsacrine-cytarabine followed by busulfan “augmented” RIC to a fludarabine-based RIC in high-risk AML and MDS and did not find any statistically significant difference in therapy-related mortality, relapse or overall survival between the two groups.14

Reduced intensity conditioning versus non-myeloablative conditioning A further reduction of intensity and toxicity was introduced by a non-myeloablative regimen with only 2 Gy TBI and fludarabine. A prospective randomized study reported by Blaise et al. compared a 2 Gy TBI-based NMA regimen with a busulfan-fludarabine-based RIC regimen.27 The incidence of grade 2-4 acute GvHD was 47% in the RIC group versus 27% in the group given NMA conditioning, with no difference in chronic GvHD. The RIC group showed a lower relapse rate (27% vs. 54%), while the NRM rate was higher (38% vs. 22%). At 5 years, the overall survival rates were identical (41%).

Evidence summary In summary, prospective studies in AML/MDS show a challenging landscape of evidence. No superiority for any arm regarding relapse or NRM was found for AML patients with intermediate-/high-risk cytogenetics and ≤60 years or MDS/secondary AML patients,16,25 with a trend towards better overall survival after RIC in the trial by Kröger et al.25 In contrast, Scott et al.26 showed a clearly reduced risk of relapse and better overall survival for MAC, despite higher NRM in AML/MDS patients. Intensifying RIC by administering sequential RIC did not improve outcome compared to conventional RIC in high-risk AML/MDS.14 Meta-analyses summarizing findings from the aforehaematologica | 2021; 106(7)

Conditioning intensity in allogeneic HSCT

mentioned randomized trials have been published but are limited by the heterogeneous characteristics across the trials.28,29 In conclusion, for AML and MDS, the overall quality of evidence for the optimal conditioning intensity is low. For higher-risk patients, MAC appeared to provide some benefit.29 However, in the current setting, only data synthesis on the level of individual patient data may be of additional value.

Retrospective studies in acute myeloid leukemia/myelodysplastic syndromes The wide basis for current clinical consensus statements with regards to conditioning intensity in AML/MDS draws on data from retrospective comparisons (Table 2). The first study to compare outcomes of conventional MAC versus RIC was from an EMBT registry study that looked at 722 patients with AML >50 years.30 Four hundred seven patients received MAC consisting of TBI doses >10 Gy or busulfan doses >8 mg/kg plus other drugs, and 315 patients underwent RIC including fludarabine in combination with low-dose TBI (<2 Gy) or busulfan doses <8 mg/kg. The results showed that NRM was higher after MAC, while RIC transplants were associated with a higher relapse risk, even after adjustment for various factors. There was, however, no difference in 2-year relapse-free and overall survival. The incidences of grades 2-4 acute GvHD and chronic GvHD were also lower after RIC. (Table 2). Another retrospective EBMT registry study by Martino et al. in MDS patients compared RIC (mainly consisting of fludarabine-busulfan regimens) with standard MAC (mainly TBI-cyclophosphamide or busulfan-cyclophosphamide).31 In multivariate analysis, the 3-year incidence of relapse was significantly higher in the RIC group, with a risk increased by 64%. In contrast, the risk of NRM was significantly decreased by 39% compared with that in the MAC arm. Acute GvHD was seen more frequently in MAC, while rates of chronic GvHD were comparable. The 3-year estimated progression-free and overall survival rates were similar in both groups. Of note, these findings were also confirmed after long-term follow-up.32 Similar conclusions were reached from comparative analyses of unrelated donor transplants.33 More stratified comparisons of NMA, RIC, conventional, and so-called hyper-intensive MAC in patients with AML/MDS with <10% blasts13 showed significant and interesting differences in NRM over time: while the NRM at day 100 was highest for hyper-intensive MAC (22%) followed by MAC (11%) and RIC (4%) and NMA (0%), the landmark NRM

after day 100 showed the highest NRM rate for NMA (32%) followed by RIC (17%), MAC (14%) and hyperintensive conditioning (11%). Another large Center for International Blood & Marrow Transplant Research (CIBMTR) analysis considered 3,731 MAC and 1,448 RIC/NMA procedures performed between 1997 and 2004. The relapse rates were significantly higher in the RIC and NMA groups than in the MAC group, but there was no difference in NRM rates. Adjusted overall survival rates at 5 years were 34%, 33% and 26% for MAC, RIC and NMA transplants, respectively. NMA conditioning resulted in inferior disease-free survival and overall survival, but there was no difference in these survival outcomes between RIC and MAC regimens (Table 2).34 In summary, retrospective studies in AML/MDS depict a homogeneous landscape of evidence, suggesting increased risk of relapse after RIC and a higher NRM after MAC, while overall survival appeared to be similar comparing both intensities.

Retrospective studies in other diseases Other myeloid malignancies In chronic phase chronic myeloid leukemia, using CIBMTR information 1,395 allogeneic HSCT recipients aged 18-60 years were evaluated in the era of tyrosine kinase inhibition (2007-2014).35 No significant differences between conditioning intensities were detected in multivariable analysis with respect to leukemia-free survival and NRM. Regarding relapse, the RIC group showed a higher risk of early relapse, and the incidence of chronic GvHD was lower with RIC than with MAC. In myelofibrosis, an EBMT analysis using data from 2000-2014 showed comparable incidences of NRM between the intensity groups and slightly increased relapse incidence for RIC, with rates at 1 and 5 years of, respectively, 14% and 23% for RIC, and 11% and 20% for MAC.36 No significant difference in 5-year overall survival was seen between the two arms.

Lymphoid malignancies In acute lymphoblastic leukemia, an analysis evaluated the outcomes of 576 adult patients ≥45 years, undergoing HSCT from an HLA-identical sibling in complete remission.37 With a median follow-up of 16 months, the 2-year NRM rate was higher in the MAC group and relapse was increased following RIC. In multivariate analysis, the type of conditioning regimen was not significantly associated with survival.

Table 2. Selected retrospective registry comparisons of conditioning intensity in acute myeloid leukemia/myelodysplastic syndromes.

Trial

Registry

Population

LFS/RFS

Relapse NRM %, RIC vs. MAC (P)

Aoudjhane et al.30

EBMT

722

Martino et al.13

EBMT

AML Age >50 y AML/MDS Blasts <10% AML/MDS Age 18-69 y

40 vs. 47 (ns) 48 vs. 54 (ns) 30 vs 33 (ns)

41 vs. 24 (<0.01) 34 vs. 24 (0.01) 40 vs 32* (<0.01)

Luger et al.34

CIBMTR

878 5179

18 vs. 32 (<0.01) 18 vs. 22 (ns) 29 vs 29 (ns)

OS 44 vs. 46 (ns) 53 vs. 56 (ns) 33 vs 34 (ns)

RIC: reduced intensity conditioning; EBMT: European Society for Blood and Marrow Transplantation; CIBMTR: Center for International Blood & Marrow Transplant Research; MAC: myeloablative conditioning; AML: acute myeloid leukemia; CML: chronic myeloid leukemia; DLBCL: diffuse large B-cell lymphoma; ALL: acute lymphoblastic leukemia; MM: multiple myeloma; MDS: myelodysplastic syndromes; N: number; LFS: leukemia-free survival; RFS: relapse-free survival; NRM: non-relapse mortality; OS: overall survival; ns: not significant; y: years. *in the RIC arm of the CIBMTR study non-myeloablative conditioning regimens are also included.

haematologica | 2021; 106(7)

1797

N. Gagelmann and N. Kröger

Regarding diffuse large B-cell lymphoma, the CIBMTR analyzed 396 patients, of whom 165 received MAC, 143 RIC, and 88 NMA.38 NRM was higher after MAC, while relapse incidence was higher after RIC. Survival and GvHD rates did not differ between groups. In patients with relapsed/refractory Hodgkin lymphoma,39 NRM rates were similar and relapse risk was slightly reduced for MAC compared with RIC. Subsequently, survival appeared to be better for recipients of MAC. In multiple myeloma patients, an analysis from the EBMT (1991-2012) evaluated patients 40-60 years old at the time of HSCT.40 At a median follow-up of 54 months, MAC was associated with a higher risk of death than RIC.41 Notably, results after 2002 were comparable. The main results of the retrospective analyses for hematologic malignancies other than AML/MDS comparing RIC versus MAC are summarized in Table 3.

Factors which may be helpful in the decision process: balance between the risk of relapse and non-relapse mortality Since, in general, compared to a lower intensity conditioning regimen, a higher intensity one is associated with more NRM but less relapse, selecting the optimal intensity of the conditioning regimen requires an appropriate

balance between the risk of relapse and NRM (Figure 1). Thus, other disease-, patient-, and transplant-specific risk factors that affect the risk of relapse and NRM should be taken into account. Next we summarize specific factors that may complicate clinical decision-making (Table 4).

Measurable disease status Multiple studies, mainly in acute lymphoblastic leukemia and more recently in AML, have investigated the association between the presence of measurable residual disease (MRD) prior to allogeneic HSCT,42 showing an increased risk of relapse and death among MRDpositive patients. However, significant between-study heterogeneity was found, underscoring site-specific methodological differences. A recent European LeukemiaNet consensus document identified key clinical and scientific issues in the measurement and application of MRD in AML, providing guidelines for the current and future use of MRD in clinical practice.43 With regards to the conditioning intensity, Walter et al.44 showed that MRD status had strong predictive value both in the MAC and NMA settings, with MRD-defined depth of response prior to HSCT being the most important predictor of outcome. Conversion from MRD-positivity before HSCT to MRD-negativity after MAC was shown to not substantially improve the incidence of relapse or survival rate.45

Table 3. Selected retrospective registry comparisons of conditioning intensity in other hematologic malignancies.

Trial McLornan et al.36

Registry

Population

LFS/RFS

Relapse NRM %, RIC vs MAC (P)

EBMT

MF Age 18-74 y CML Age18-60 y DLBCL Age 18-69 y rrHL Age 25-40 y ALL Age ≥45 y MM Age 29-66 y

2224

26 vs. 32 (0.001)* 43 vs. 44 (ns) 15 vs. 18 (ns) 36 vs. 48 (0.07) 32 vs. 38 (0.07) 19 vs. 34 (<0.01)

20 vs. 23 (0.08) 25 vs. 26 (ns) 38 vs. 26 (0.03) 60 vs. 50 (ns) 47 vs. 31 (<0.01) 54 vs. 27 (<0.01)

Chhabra et al.35

CIBMTR

Bacher et al.38

CIBMTR

Genadieva-Stravrik et al.39

EBMT

Mohty et al.37

EBMT

Crawley et al.41

EBMT

1395 396 312 576 516

34 vs. 34 (ns) 29 vs. 32 (ns) 47 vs. 56 (0.01) 12 vs. 13 (ns) 21 vs. 29 (0.03) 24 vs. 37 (<0.01)

OS 51 vs. 53 (ns) 53 vs. 53 (ns) 20 vs. 18 (ns) 62 vs. 73 (ns) 48 vs. 45 (ns) 39 vs. 51 (ns)

RIC: reduced intensity conditioning; EBMT: European Society for Blood and Marrow Transplantation; CIBMTR: Center for International Blood & Marrow Transplant Research; MAC: myeloablative conditioning; AML: acute myeloid leukemia; CML: chronic myeloid leukemia; DLBCL: diffuse large B-cell lymphoma; ALL: acute lymphoblastic leukemia; MM: multiple myeloma; MDS: myelodysplastic syndromes; N: number; LFS: leukemia-free survival; RFS: relapse-free survival; NRM: non-relapse mortality; OS: overall survival; ns: not significant; y: years. *Unadjusted graft-versus-host disease/relapse-free survival at 5 years.

Figure 1. The balance between risk for non-relapse mortality and risk for relapse when choosing conditioning intensity.

1798

haematologica | 2021; 106(7)

Conditioning intensity in allogeneic HSCT

In a comparison of conditioning regimen intensities, a correlative analysis was performed of ultra-deep DNA sequencing of blood samples from patients treated in the BMT-CTN trial.26,46 Of the 218 AML patients, 190 patients had blood samples collected prior to HSCT, which were analyzed using a 13-gene, targeted error-corrected, next-generation sequencing panel for the presence of genomic MRD. Among these patients, 63% randomized to RIC and 68% randomized to MAC had evidence of MRD by detection of one or more of the tested genes. It should, however, be pointed out that current European LeukemiaNet recommendations do not support mutations such as FLT3-ITD, NRAS, DNMT3A, or ASXL1, and expression levels of EVI1 as single MRD markers.43 Instead, these markers are suggested to be useful when used in combination with a second MRD marker. Among MRD-positive patients in this analysis,46 outcomes were dismal for those given RIC, with a 3-year incidence of relapse of 67% versus 19% in the MAC group. After adjusting for disease risk and donor group, RIC was associated with a significantly increased risk of relapse and decreased survival in MRD-positive patients, when compared to MAC. However, the 3-year NRM rate was higher in those who underwent MAC (27%) than in those who received RIC (9%), and this difference was not affected by MRD status. Furthermore, overall survival was comparable in the two groups among patients who were MRD-negative. In line with these results, an EBMT analysis of 2,292 AML patients in first complete remission showed that less intensive conditioning was only inferior to MAC for patients <50 years old who were MRD-positive, showing higher relapse and lower survival rates.47 Irrespective of age, conditioning intensities were associated with similar outcomes in patients who were MRD-negative. This analysis also revealed the better caption of the “real world” using our proposed balanced conditioning approach, because RIC/MAC groups shared the same heterogeneous spectrum of regimens used, with only different distributions of certain regimens. A significant caveat regarding this study is that the MRD methodology and allocation were determined by individual participating centers, utilizing molecular and/or immunophenotyping criteria. An earlier CIBMTR analysis of 197 patients with acute lymphoblastic leukemia showed that MRDpositive patients had a higher risk of relapse with RIC,48 but MRD-negative patients who also received tyrosine kinase inhibitor therapy before HSCT had superior survival after RIC compared to a similar population after MAC. After multivariable adjustment, RIC reduced NRM but increased relapse risk. In light of these results that suggest benefits of more intensive conditioning primarily for some or all AML patients who were MRD-positive prior to HSCT, a recent analysis suggested that MAC should also be considered for MRD-negative AML patients if tolerated.49 In an analysis of 287 patients with MDS,50 of whom one quarter had >5% marrow blasts and more than a half were MRD-positive at HSCT, as determined by multiparameter flow cytometry and cytogenetics on marrow aspirates, it was found that the risk of overall mortality was higher with lower intensity retimens than with higher intensity regimens among the MRD-positive patients. On the other hand, MRD-negative patients had similar risks of mortalihaematologica | 2021; 106(7)

Table 4. Risk factors influencing treatment failure (relapse or NRM) after allogeneic HSCT. Disease-specific factors Advanced disease status relapse > NRM Unfavorable cytogenetics/molecular genetics relapse > NRM Susceptibility to GVL-effect relapse > NRM Patient-specific risk factors Age NRM > relapse Performance status NRM > relapse Comorbidities NRM > relapse Transplant-specific risk factors MRD positivity relapse > NRM HLA disparity NRM > relapse CMV incompatibility NRM > relapse Center effect (JACIE accredited) NRM > relapse NRM, non-relapse mortality; HSCT, hematopoietic stem cell transplantation; GVL, graft-versusleukemia effect; MRD, measurable residual disease; CMV, cytomegalovirus; JACIE, Joint Accreditation Committee ISCT-Europe & EBMT.

ty. The main reason for mortality after lower intensity conditioning in MRD-positive patients was relapse. Evidence from comparison of conditioning regimens of different intensities according to MRD status in AML is still based on retrospective or post-hoc analyses including heterogeneous use of regimens for each group. Current signals point to a benefit for more intensive approaches in MRD-positive AML patients but the increased risk of NRM needs to be considered. In addition, other factors such as donor selection or graft type may further affect MRD-related outcomes. For example, one retrospective analysis suggested that pre-HSCT MRD-positive patients receiving cord blood as the source of stem cells had a reduced incidence of relapse. in comparison with those receiving transplants from matched unrelated donors and even better survival than those given grafts from mismatched unrelated donors.51 Another study reported better outcome after haploidentical HSCT in comparison to HLAidentical sibling transplants in MRD-positive AML patients.52

Disease risk The evidence regarding the impact of conditioning intensity on different disease risk index (DRI) groups53,54 is still limited to some retrospective analyses. One analysis evaluated 380 AML/MDS patients with either high/very high or low/intermediate DRI.55 Among patients with high/very high DRI, there was no difference in outcome between the RIC and MAC groups. For low/intermediate risk DRI, recipients of MAC showed better 3-year overall survival (69% vs. 57%), disease-free survival (65% vs. 51%), and a decreased incidence of relapse (17% vs. 32%) but similar, slightly increased NRM (19% vs. 17%). Except for overall survival, which was not significantly different in multivariable analysis, results for the remaining outcomes were confirmed after multivariable adjustment. In a larger, very recent CIBMTR analysis, MAC resulted in an improved survival in comparison to RIC in AML/MDS patients 40 to 65 years old with low/intermediate risk DRI, but similar clinical benefit to RIC despite higher risk of relapse in patients with high/very high risk DRI.56 In conclusion, compared with RIC, MAC may be associated with improved outcomes among patients with low/intermediate DRI, while no benefit has been noted for any intensity among those in high/very high DRI groups. 1799

N. Gagelmann and N. Kröger

Transplant risk score Several studies have been published to further aid physicians in balancing the risks and benefits of HSCT for patients. The EBMT transplantation risk score,57 initially established for chronic myeloid leukemia, has now been expanded and validated to assess post-transplant risk for multiple hematologic disorders that can be treated with a transplant. The risk score includes: the patient’s age class (<20, 20-40, or >40 years), disease stage (early, intermediate, or advanced), donor type (HLA-identical sibling or unrelated donor), and donor-recipient sex match/mismatch (specifically, an increased score for a male recipient with female donor). The score held for all acquired hematologic disorders (score 0-5) and was independent of the HSCT procedure itself. This risk scheme does not specifically consider comorbidities (see below), and notably places all patients >40 years in a high-risk category. Importantly, the EBMT risk score was independently valid, irrespective of the intensity of the conditioning. In line with these findings, an EBMT mega-file analysis showed that patients' pre-HSCT risk factors determine survival,58 independently of conditioning intensities, suggesting that outcomes may be improved more effectively through better identifying patients with their individual pre-HSCT characteristics. In summary, the recent increased efforts will shape future research to establish more individualized risk models for predicting outcome after HSCT in several diseases, irrespective of conditioning intensity.59–62

Genetic risk The Japanese Society of Hematopoietic Stem Cell Transplantation reported a significant survival benefit from MAC versus RIC in 840 AML patients with poor cytogenetics, irrespective of subgroup analysis for age <60 years or high comorbidity index.63 In contrast, EBMT studies failed to show any survival benefit of either type of conditioning regimen in patients with cytogenetically poor-risk AML64,65 One such study with AML patients 40-60 years old in first complete remssion, stratified according to cytogenetic risk, found better survival with RIC in low-risk patients but not in the intermediate- or poor-risk groups.65 In the latter groups, relapse incidence was lower with MAC, but NRM was higher with MAC in all cytogenetic risk groups. The analysis concluded that in patients 40-60 years old, MAC had no significant advantage. Bornhäuser et al. also reported no difference between MAC and RIC for intermediateand high-risk AML patients in a prospective randomized trial.16 Another very recent study in MDS showed that the adverse impact of shorter telomeres on NRM was more frequently observed in patients receiving more intensive conditioning and was associated with the development of GvHD.66 Thus, strategies in MDS patients with shorter telomere length may focus on minimizing toxicity and reducing conditioning intensity.

Performance status, comorbidities and age There is conflicting evidence regarding the association of conditioning intensity and comorbidities. As described above, the study by Rambaldi et al. showed significantly reduced NRM after busulfan-fludarabine conditioning compared with busulfan-cyclophosphamide conditioning in patients with a comorbidity index >2.20 A recent analysis in patients ≥50 years old with Philadelphia-positive acute 1800

lymphoblastic leukemia who received tyrosine kinase inhibitor therapy before HSCT and who achieved MRDnegativity showed similar outcomes after RIC or MAC,67 with subgroup analyses suggesting better outcomes for RIC in patients with a poor performance status or a high HCT-Comorbidity Index.68,69 In contrast, the above mentioned Japanese study in patients with cytogenetically poor-risk AML and first complete remission showed an association with better outcomes for MAC,63 irrespective of subgroup analysis for patients <60 years old or high HCT-Comorbidity Index. A single-center experience including 875 adults highlights both the pitfalls of arbitrary dichotomization of conditioning intensity and the value of patient-specific balanced evaluations.70 The following were classified as RIC: fludarabine 150 mg/m2 with busulfan 6.4 mg/kg; fludarabine 150 mg/m2 with treosulfan 30 g/m2; and fludarabine 150 mg/m2 with melphalan 100-140 mg/m2. With respect to specific comorbidities in the overall population, which varied widely across regimens, renal dysfunction, hypoalbuminemia, and severe hepatic disease were associated with worse NRM. Notably, the risk was not associated with intensity as classified. Instead, outcome was associated with regimen-specific profiles, showing increased NRM for fludarabine-busulfan in patients with cardiac disease, and for fludarabine-melphalan and fludarabine-treosulfan in patients with severe pulmonary disease and a pre-existing infection. The HCT-Comorbidity Index was only associated with worse outcome in patients receiving fludarabine-melphalan conditioning but not in those given other regimens. With respect to RIC versus MAC, several analyses showed similar outcomes in patients ≥50 years old.47,71 Of note, generally, patients receiving RIC are older by a median of 10 years.72 For the evaluation of less intensive conditioning, a retrospective analysis compared the efficacy of RIC in MDS patients >50 years, analyzing patients <65 or ≥65 years at HSCT separately. Subsequently, in patients <65 years, NMA conditioning was associated with higher NRM and shorter survival, while the cumulative incidence of relapse was similar in both the RIC and NMA groups. The EBMT recently analyzed the outcome of AML patients with reduced performance status according to the Karnofsky performance status of ≤80%. Patients with a Karnofsky performance status of 80% benefited more from MAC, while patients with a performance status <80% benefited more from RIC.73 In conclusion, simple recommendation of RIC/NMA for older unfit patients or MAC for young and fit ones is not reflected by current evidence. Neither age nor comorbidities are associated with significantly different outcomes for these categorizations.74,75 Ultimately, only patients with exceptionally limiting comorbidities/performance may experience different outcomes.

Infections and late effects Presumably, less intensive conditioning may lead to reduced rates of infection. However, published evidence on this issue remains limited. Early investigations suggested that RIC may decrease the risk of dying from an opportunistic infection, reducing the frequency of cytomegalovirus infection/disease.76 In a recent CIBMTR analysis of 1,755 AML patients ≥40 years old, although absolute numbers of patients with ≥1 infection were not different in the RIC/NMA (58%) and MAC (61%) groups, haematologica | 2021; 106(7)

Conditioning intensity in allogeneic HSCT

the median time to initial infection after MAC occurred earlier, at a median of 15 versus 21 days.77 Patients receiving MAC were more likely to experience ≥1 bacterial infection, whereas ≥1 viral infection was more prevalent among those receiving RIC/NMA. Another recent smaller analysis regarding mucositis found infections were less frequent after less intensive conditioning, with the rate being lowest after fludarabine-treosulfan conditioning.78 Evaluations regarding long-term effects of different intensity conditioning regimens are scarce. One study showed similar long-term leukemia-free survival and GvHD-free/relapse-free survival at 10 years for RIC and MAC, with most events occurring within the first 2 years after allogeneic HSCT.79 Relapse was the major cause of late death in both groups; while NRM and especially chronic GvHD as well as second cancers were more frequent causes of late death after MAC.

maintenance therapy with tyrosine kinase inhibitors in AML, specifically for patients with FLT3-ITD mutations. Two prospective studies showed that sorafenib maintenance was associated with significantly better relapse and relapse-free survival outcomes.88,89 In the Chinese study all patients were given MAC,88 while 58% of patients in the German study received sorafenib and underwent RIC HSCT;89 no stratified comparisons according to conditioning intensity were conducted within this latter study. Limited indirect comparison of the two studies suggest at least comparable outcomes in both, with 2-year survival rates in the sorafenib arm of 82% in the Chinese study and 90% in the German study. Whether the impressive effect of maintenance on relapse, with a 60% risk reduction, may broaden the utility of less intensive conditioning, is yet to be determined.

Graft-versus-host disease

Summary

Prospective and retrospective comparisons show that relapse is the major cause of death in RIC patients while most MAC patients die from GvHD. This led to the concept of influencing risk of relapse and GvHD to improve outcomes for any given conditioning intensity.80 Regarding the latter, the use of post-HSCT cyclophosphamide significantly reduced the risk of GvHD and was associated with better outcomes when used in haploidentical HSCT, at least compared with transplants from mismatched unrelated donors.81 In view of this consistent effect of post-HSCT cyclophosphamide, no significant differences according to conditioning intensity in haploidentical transplants were found after meta-regression analyses81 and in two large registry analyses.82,83 So far, it has not been possible to confirm the hypothesis of an association between reduced GvHD after post-HSCT cyclophosphamide and better survival in patients treated with MAC regimens, but comparisons still lack stringent control and assessments in specific donor settings. Here, it is important to underscore the complex interplay of GvHD and risk of relapse with the type of disease or tumor burden, including phenomena that are not yet fully clarified.84 In acute lymphoblastic leukemia and BCR-ABL-negative myeloproliferative neoplasms there are similar and obvious correlations between the occurrence of GvHD and risk of relapse when compared with chronic myeloid leukemia, and in MDS and lymphoproliferative disorders there are intermediate correlations between GvHD and relapse risk. Only in AML and plasma cell disorders is GvHD associated with only modest reductions in relapse risk.

The answer to “when and f\or whom” with respect to HSCT conditioning intensity is complex, individualized, and constantly evolving. Apart from factors such as pretreatment, disease risk, donor source, GvHD prophylaxis, and maintenance strategies, the conditioning regimen is only one factor affecting the risk of treatment failure through relapse or NRM after allogeneic HSCT. As of 2021, the traditional, simplified section of a conditioning regimen between RIC and MAC is no longer appropriate because it significantly underestimates the complexity of currently used regimens with respect to toxicity. Updated categorizations from the EBMT15 are one step in the right direction but still provide only modest improvements in facilitating decision-making, considering all outcomes. A critical individual balance between the risk of NRM and the risk of relapse must be inclluded in a personalized medicine approach. These individual and continuous considerations may include diverse factors such as disease burden, MRD status and other disease-specific, patient-specific, and transplant-specific risk factors. Furthermore, recent advances in the incorporation of toxicity-reduced conditioning regimens (e.g., treosulfan) and improvements in relapse reduction by including maintenance strategies, as well as immunotherapy approaches after allogeneic HSCT may further refine considerations regarding conditioning intensity, steering towards the use of less intensive and toxic regimens in the future. Discosures No conflicts of interest to disclose.

Maintenance therapy A more personalized approach might incorporate posttransplant strategies to prevent relapse.85-87 There is accumulating evidence regarding the efficacy of post-HSCT

References 1. Gyurkocza B, Sandmaier BM. Conditioning regimens for hematopoietic cell transplantation: one size does not fit all. Blood. 2014;124(3):344-353. 2. Giralt S, Ballen K, Rizzo D, et al. Reducedintensity conditioning regimen workshop: defining the dose spectrum. Report of a workshop convened by the Center for International Blood and Marrow Transplant

haematologica | 2021; 106(7)

Contributions NG and NK drafted the manuscript and approved the final version.

Research. Biol Blood Marrow Transplant. 2009;15(3):367-369. 3. Champlin R, Khouri I, Kornblau S, Molldrem J, Giralt S. Reinventing bone marrow transplantation: reducing toxicity using nonmyeloablative, preparative regimens and induction of graft-versus-malignancy. Curr Opin Oncol. 1999;11(2):87-95. 4. Bacigalupo A, Ballen K, Rizzo D, et al. Defining the intensity of conditioning regimens: working definitions. Biol Blood

Marrow Transplant. 2009;15(12):1628-1633. 5. Bacigalupo A. Third EBMT/AMGEN Workshop on reduced-intensity conditioning allogeneic haemopoietic stem cell transplants (RIC-HSCT), and panel consensus. Bone Marrow Transplant. 2004;33(7):691696. 6. Bacigalupo A. Second EBMT Workshop on reduced intensity allogeneic hemopoietic stem cell transplants (RI-HSCT). Bone Marrow Transplant. 2002;29(3):191-195.

1801

N. Gagelmann and N. Kröger 7. Bacigalupo A. Hematopoietic stem cell transplants after reduced intensity conditioning regimen (RI-HSCT): report of a workshop of the European Group for Blood and Marrow Transplantation (EBMT). Bone Marrow Transplant. 2000;25(8):803-805. 8. Kolb H-J, Schmid C. The FLAMSA conceptpast and future. Ann Hematol. 2020;99(9): 1979-1988. 9. Schmid C, Schleuning M, Ledderose G, Tischer J, Kolb H-J. Sequential regimen of chemotherapy, reduced-intensity conditioning for allogeneic stem-cell transplantation, and prophylactic donor lymphocyte transfusion in high-risk acute myeloid leukemia and myelodysplastic syndrome. J Clin Oncol. 2005;23(24):5675-5687. 10. Bacigalupo A, Raiola AM, Lamparelli T, et al. Thiotepa-based reduced intensity conditioning regimen: a 10 year follow up. Bone Marrow Transplant. 2007;40(11):1091-1093. 11. Langenhorst JB, van Kesteren C, van Maarseveen EM, et al. Fludarabine exposure in the conditioning prior to allogeneic hematopoietic cell transplantation predicts outcomes. Blood Adv. 2019;3(14):21792187. 12. Marks R, Potthoff K, Hahn J, et al. Reducedtoxicity conditioning with fludarabine, BCNU, and melphalan in allogeneic hematopoietic cell transplantation: particular activity against advanced hematologic malignancies. Blood. 2008;112(2):415-425. 13. Martino R, Wreede L de, Fiocco M, et al. Comparison of conditioning regimens of various intensities for allogeneic hematopoietic SCT using HLA-identical sibling donors in AML and MDS with <10% BM blasts: a report from EBMT. Bone Marrow Transplant. 2013;48(6):761-770. 14. Craddock C, Jackson A, Loke J, et al. Augmented reduced-intensity regimen does not improve postallogeneic transplant outcomes in acute myeloid leukemia. J Clin Oncol. 2021;39(7):768-778. 15. Spyridonidis A, Labopin M, Savani BN, et al. Redefining and measuring transplant conditioning intensity in current era: a study in acute myeloid leukemia patients. Bone Marrow Transplant. 2020;55(6):1114-1125. 16. Bornhäuser M, Kienast J, Trenschel R, et al. Reduced-intensity conditioning versus standard conditioning before allogeneic haemopoietic cell transplantation in patients with acute myeloid leukaemia in first complete remission: a prospective, open-label randomised phase 3 trial. Lancet. Oncol 2012;13(10):1035-1044. 17. Fasslrinner F, Schetelig J, Burchert A, et al. Long-term efficacy of reduced-intensity versus myeloablative conditioning before allogeneic haemopoietic cell transplantation in patients with acute myeloid leukaemia in first complete remission: retrospective follow-up of an open-label, randomised phase 3 trial. Lancet Haematol. 2018;5(4):e161e169. 18. Martino R, Badell I, Brunet S, et al. Highdose busulfan and melphalan before bone marrow transplantation for acute nonlymphoblastic leukemia. Bone Marrow Transplant. 1995;16(2):209-212. 19. Giralt S, Thall PF, Khouri I, et al. Melphalan and purine analog-containing preparative regimens: reduced-intensity conditioning for patients with hematologic malignancies undergoing allogeneic progenitor cell transplantation. Blood. 2001;97(3):631-637. 20. Rambaldi A, Grassi A, Masciulli A, et al. Busulfan plus cyclophosphamide versus busulfan plus fludarabine as a preparative

1802

regimen for allogeneic haemopoietic stemcell transplantation in patients with acute myeloid leukaemia: an open-label, multicentre, randomised, phase 3 trial. Lancet Oncol. 2015;16(15):1525-1536. 21. Peccatori J, Mastaglio S, Giglio F, et al. Clofarabine and treosulfan as conditioning for matched related and unrelated hematopoietic stem cell transplantation: results from the Clo3o phase II trial. Biol Blood Marrow Transplant. 2020;26(2):316322. 22. Shimoni A, Hardan I, Shem-Tov N, Rand A, Yerushalmi R, Nagler A. Fludarabine and treosulfan: a novel modified myeloablative regimen for allogeneic hematopoietic stem-cell transplantation with effective antileukemia activity in patients with acute myeloid leukemia and myelodysplastic syndromes. Leuk Lymphoma. 2007;48(12):2352-2359. 23. Kröger N, Shimoni A, Zabelina T, et al. Reduced-toxicity conditioning with treosulfan, fludarabine and ATG as preparative regimen for allogeneic stem cell transplantation (alloSCT) in elderly patients with secondary acute myeloid leukemia (sAML) or myelodysplastic syndrome (MDS). Bone Marrow Transplant. 2006;37:339-344. 24. Beelen DW, Trenschel R, Stelljes M, et al. Treosulfan or busulfan plus fludarabine as conditioning treatment before allogeneic haemopoietic stem cell transplantation for older patients with acute myeloid leukaemia or myelodysplastic syndrome (MCFludT.14/L): a randomised, non-inferiority, phase 3 trial. Lancet Haematol. 2020;7(1): e28-e39. 25. Kröger N, Iacobelli S, Franke G-N, et al. Dose-reduced versus standard conditioning followed by allogeneic stem-cell transplantation for patients with myelodysplastic syndrome: a prospective randomized phase III study of the EBMT (RICMAC trial). J Clin Oncol. 2017;35(19):2157-2164. 26. Scott BL, Pasquini MC, Logan BR, et al. Myeloablative versus reduced-Intensity hematopoietic cell transplantation for acute myeloid leukemia and myelodysplastic syndromes. J Clin Oncol. 2017;35(11):11541161. 27. Blaise D, Tabrizi R, Boher J-M, et al. Randomized study of 2 reduced-intensity conditioning strategies for human leukocyte antigen-matched, related allogeneic peripheral blood stem cell transplantation: prospective clinical and socioeconomic evaluation. Cancer. 2013;119(3):602-611. 28. Rashidi A, Meybodi MA, Cao W, et al. Myeloablative versus reduced-intensity hematopoietic cell transplantation in myelodysplastic syndromes: systematic review and meta-analysis. Biol Blood Marrow Transplant. 2020;26(6):e138-e141. 29. Ma S, Shi W, Li Z, et al. Reduced-intensity versus myeloablative conditioning regimens for younger adults with acute myeloid leukemia and myelodysplastic syndrome: a systematic review and meta-analysis. J Cancer. 2020;11(17):5223-5235. 30. Aoudjhane M, Labopin M, Gorin NC, et al. Comparative outcome of reduced intensity and myeloablative conditioning regimen in HLA identical sibling allogeneic haematopoietic stem cell transplantation for patients older than 50 years of age with acute myeloblastic leukaemia: a retrospective survey from the Acute Leukemia Working Party (ALWP) of the European group for Blood and Marrow Transplantation (EBMT). Leukemia. 2005;19 (12):2304-2312.

31. Martino R, Iacobelli S, Brand R, et al. Retrospective comparison of reduced-intensity conditioning and conventional highdose conditioning for allogeneic hematopoietic stem cell transplantation using HLAidentical sibling donors in myelodysplastic syndromes. Blood. 2006;108(3):836-846. 32. Martino R, Henseler A, van Lint M, et al. Long-term follow-up of a retrospective comparison of reduced-intensity conditioning and conventional high-dose conditioning for allogeneic transplantation from matched related donors in myelodysplastic syndromes. Bone Marrow Transplant. 2017; 52(8):1107-1112. 33. Ringdén O, Labopin M, Ehninger G, et al. Reduced intensity conditioning compared with myeloablative conditioning using unrelated donor transplants in patients with acute myeloid leukemia. J Clin Oncol. 2009;27(27):4570-4577. 34. Luger SM, Ringdén O, Zhang M-J, et al. Similar outcomes using myeloablative vs reduced-intensity allogeneic transplant preparative regimens for AML or MDS. Bone Marrow Transplant. 2012;47(4):203211. 35. Chhabra S, Ahn KW, Hu Z-H, et al. Myeloablative vs reduced-intensity conditioning allogeneic hematopoietic cell transplantation for chronic myeloid leukemia. Blood Adv. 2018;2(21):2922-2936. 36. 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. 37. Mohty M, Labopin M, Volin L, et al. Reduced-intensity versus conventional myeloablative conditioning allogeneic stem cell transplantation for patients with acute lymphoblastic leukemia: a retrospective study from the European Group for Blood and Marrow Transplantation. Blood. 2010;116 (22):4439-4443. 38. Bacher U, Klyuchnikov E, Le-Rademacher J, et al. Conditioning regimens for allotransplants for diffuse large B-cell lymphoma: myeloablative or reduced intensity? Blood. 2012;120(20):4256-4262. 39. Genadieva-Stavrik S, Boumendil A, Dreger P, et al. Myeloablative versus reduced intensity allogeneic stem cell transplantation for relapsed/refractory Hodgkin's lymphoma in recent years: a retrospective analysis of the Lymphoma Working Party of the European Group for Blood and Marrow Transplantation. Ann Oncol. 2016;27 (12):2251-2257. 40. Hayden PJ, Iacobelli S, Pérez-Simón JA, et al. Conditioning-based outcomes after allogeneic transplantation for myeloma following a prior autologous transplant (19912012) on behalf of EBMT CMWP. Eur J Haematol. 2020;104(3):181-189. 41. Crawley C, Iacobelli S, Björkstrand B, Apperley JF, Niederwieser D, Gahrton G. Reduced-intensity conditioning for myeloma: lower nonrelapse mortality but higher relapse rates compared with myeloablative conditioning. Blood. 2007;109(8):3588-3594. 42. Buckley SA, Wood BL, Othus M, et al. Minimal residual disease prior to allogeneic hematopoietic cell transplantation in acute myeloid leukemia: a meta-analysis. Haematologica. 2017;102(5):865-873. 43. Schuurhuis GJ, Heuser M, Freeman S, et al.

haematologica | 2021; 106(7)

Conditioning intensity in allogeneic HSCT Minimal/measurable residual disease in AML: a consensus document from the European LeukemiaNet MRD Working Party. Blood. 2018;131(12):1275-1291. 44. Walter RB, Othus M, Burnett AK, et al. Resistance prediction in AML: analysis of 4601 patients from MRC/NCRI, HOVON/SAKK, SWOG and MD Anderson Cancer Center. Leukemia. 2015;29(2):312320. 45. Zhou Y, Othus M, Araki D, et al. Pre- and post-transplant quantification of measurable ('minimal') residual disease via multiparameter flow cytometry in adult acute myeloid leukemia. Leukemia. 2016;30(7): 1456-1464. 46. Hourigan CS, Dillon LW, Gui G, et al. Impact of conditioning intensity of allogeneic transplantation for acute myeloid leukemia with genomic evidence of residual disease. J Clin Oncol. 2020;38(12):12731283. 47. Gilleece MH, Labopin M, Yakoub-Agha I, et al. Measurable residual disease, conditioning regimen intensity, and age predict outcome of allogeneic hematopoietic cell transplantation for acute myeloid leukemia in first remission: a registry analysis of 2292 patients by the Acute Leukemia Working Party European Society of Blood and Marrow Transplantation. Am J Hematol. 2018;93(9):1142-1152. 48. Bachanova V, Marks DI, Zhang M-J, et al. Ph+ ALL patients in first complete remission have similar survival after reduced intensity and myeloablative allogeneic transplantation: impact of tyrosine kinase inhibitor and minimal residual disease. Leukemia. 2014;28(3):658-665. 49. Morsink LM, Sandmaier BM, Othus M, et al. Conditioning intensity, pre-transplant flow cytometric measurable residual disease, and outcome in adults with acute myeloid leukemia undergoing allogeneic hematopoietic cell transplantation. Cancers. 2020;12(9):2339. 50. Festuccia M, Deeg HJ, Gooley TA, et al. Minimal identifiable disease and the role of conditioning intensity in hematopoietic cell transplantation for myelodysplastic syndrome and acute myelogenous leukemia evolving from myelodysplastic syndrome. Biol Blood Marrow Transplant 2016;22(7): 1227-1233. 51. Milano F, Gooley T, Wood B, et al. Cordblood transplantation in patients with minimal residual disease. N Engl J Med. 2016;375(10):944-953. 52. Chang Y-J, Wang Y, Liu Y-R, et al. Haploidentical allograft is superior to matched sibling donor allograft in eradicating pre-transplantation minimal residual disease of AML patients as determined by multiparameter flow cytometry: a retrospective and prospective analysis. J Hematol Oncol. 2017;10(1):134. 53. Armand P, Gibson CJ, Cutler C, et al. A disease risk index for patients undergoing allogeneic stem cell transplantation. Blood. 2012;120(4):905-913. 54. Armand P, Kim HT, Logan BR, et al. Validation and refinement of the Disease Risk Index for allogeneic stem cell transplantation. Blood. 2014;123(23):3664-3671. 55. Solh MM, Solomon SR, Morris LE, Zhang X, Holland HK, Bashey A. The dilemma of conditioning intensity: when does myeloablative conditioning improve outcomes for allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2019;25 (3):606-612.

haematologica | 2021; 106(7)

56. Bejanyan N, Zhang M, Bo-Subait K. Myeloablative conditioning for allogeneic transplantation results in superior diseasefree survival for acute myeloid leukemia and myelodysplastic syndromes with low/intermediate, but not high disease risk Index: a CIBMTR study. Transplant Cell Ther. 2021;27(1):68.e1-68.e9. 57. Gratwohl A. The EBMT risk score. Bone Marrow Transplant. 2012;47(6):749-756. 58. Gratwohl A, Duarte R, Snowden JA, et al. Pre-transplantation risks and transplanttechniques in haematopoietic stem cell transplantation for acute leukaemia. EClinicalMedicine. 2019;15:33-41. 59. Gagelmann N, Ditschkowski M, Bogdanov R, et al. Comprehensive clinical-molecular transplant scoring system for myelofibrosis undergoing stem cell transplantation. Blood. 2019;133(20):2233-2242. 60. Gagelmann N, Eikema D-J, Stelljes M, et al. Optimized EBMT transplant-specific risk score in myelodysplastic syndromes after allogeneic stem-cell transplantation. Haematologica. 2019;104(5):929-936. 61. Shaffer BC, Ahn KW, Hu Z-H, et al. Scoring system prognostic of outcome in patients undergoing allogeneic hematopoietic cell transplantation for myelodysplastic syndrome. J Clin Oncol. 2016;34(16):1864-1871. 62. Versluis J, Cornelissen JJ. Risks and benefits in a personalized application of allogeneic transplantation in patients with AML in first CR. Semin Hematol. 2019;56(2):164-170. 63. Konuma T, Kondo T, Mizuno S. Conditioning intensity for allogeneic hematopoietic cell transplantation in acute myeloid leukemia patients with poor-prognosis cytogenetics in first complete remission. Biol Blood Marrow Transplant. 2020;26(3):463-471. 64. Versluis J, Labopin M, Ruggeri A, et al. Alternative donors for allogeneic hematopoietic stem cell transplantation in poor-risk AML in CR1. Blood Adv. 2017;1 (7):477-485. 65. Passweg JR, Labopin M, Cornelissen J, et al. Conditioning intensity in middle-aged patients with AML in first CR: no advantage for myeloablative regimens irrespective of the risk group-an observational analysis by the Acute Leukemia Working Party of the EBMT. Bone Marrow Transplant. 2015;50 (8):1063-1068. 66. Myllymäki M, Redd R, Reilly CR, et al. Short telomere length predicts nonrelapse mortality after stem cell transplantation for myelodysplastic syndrome. Blood. 2020;136 (26):3070-3081. 67. Akahoshi Y, Nishiwaki S, Arai Y, et al. Reduced-intensity conditioning is a reasonable alternative for Philadelphia chromosome-positive acute lymphoblastic leukemia among elderly patients who have achieved negative minimal residual disease: a report from the Adult Acute Lymphoblastic Leukemia Working Group of the JSHCT. Bone Marrow Transplant. 2020;55(7):1317-1325. 68. Sorror ML, Maris MB, Storb R, et al. Hematopoietic cell transplantation (HCT)specific comorbidity index: a new tool for risk assessment before allogeneic HCT. Blood. 2005;106(8):2912-2919. 69. Mor V, Laliberte L, Morris JN, Wiemann M. The Karnofsky Performance Status Scale. An examination of its reliability and validity in a research setting. Cancer. 1984;53(9):20022007. 70. Fein JA, Shimoni A, Labopin M, et al. The impact of individual comorbidities on non-

relapse mortality following allogeneic hematopoietic stem cell transplantation. Leukemia. 2018;32(8):1787-1794. 71. Gilleece MH, Labopin M, Savani BN, et al. Allogeneic haemopoietic transplantation for acute myeloid leukaemia in second complete remission: a registry report by the Acute Leukaemia Working Party of the EBMT. Leukemia. 2020;34(1):87-99. 72. Sengsayadeth S, Gatwood KS, Boumendil A, et al. Conditioning intensity in secondary AML with prior myelodysplastic syndrome/myeloproliferative disorders: an EBMT ALWP study. Blood Adv. 2018;2(16): 2127-2135. 73. Saraceni F, Labopin M, Forcade E, et al. Allogeneic stem cell transplant in patients with acute myeloid leukemia and karnofsky performance status score less than or equal to 80%: a study from the Acute Leukemia Working Party of the European Society for Blood and Marrow Transplantation (EBMT). Cancer Med. 2021;10(1):23-33. 74. Sorror ML, Estey E. Allogeneic hematopoietic cell transplantation for acute myeloid leukemia in older adults. Hematology Am Soc Hematol Educ Program. 2014;2014:2133. 75. Castagna L, Fürst S, Marchetti N, et al. Retrospective analysis of common scoring systems and outcome in patients older than 60 years treated with reduced-intensity conditioning regimen and alloSCT. Bone Marrow Transplant. 2011;46(7): 1000-1005. 76. Martino R, Caballero MD, Canals C, et al. Reduced-intensity conditioning reduces the risk of severe infections after allogeneic peripheral blood stem cell transplantation. Bone Marrow Transplant. 2001;28(4):341347. 77. Ustun C, Kim S, Chen M, et al. Increased overall and bacterial infections following myeloablative allogeneic HCT for patients with AML in CR1. Blood Adv. 2019;3(17): 2525-2536. 78. Shouval R, Kouniavski E, Fein J, et al. Risk factors and implications of oral mucositis in recipients of allogeneic hematopoietic stem cell transplantation. Eur J Haematol. 2019;103(4):402-409. 79. Shimoni A, Labopin M, Savani B, et al. Long-term survival and late events after allogeneic stem cell transplantation from HLAmatched siblings for acute myeloid leukemia with myeloablative compared to reduced-intensity conditioning: a report on behalf of the acute leukemia working party of European group for blood and marrow transplantation. J Hematol Oncol. 2016;9 (1):118. 80. Storb R, Gyurkocza B, Storer BE, et al. Graftversus-host disease and graft-versus-tumor effects after allogeneic hematopoietic cell transplantation. J Clin Oncol. 2013;31(12): 1530-1538. 81. Gagelmann N, Bacigalupo A, Rambaldi A, et al. Haploidentical stem cell transplantation with posttransplant cyclophosphamide therapy vs other donor transplantations in adults with hematologic cancers: a systematic review and meta-analysis. JAMA Oncol. 2019;5(12):1739-1748. 82. Brissot E, Labopin M, Ehninger G, et al. Haploidentical versus unrelated allogeneic stem cell transplantation for relapsed/refractory acute myeloid leukemia: a report on 1578 patients from the Acute Leukemia Working Party of the EBMT. Haematologica. 2019;104(3):524-532. 83. Ciurea SO, Zhang M-J, Bacigalupo AA, et al.

1803

N. Gagelmann and N. Kröger Haploidentical transplant with posttransplant cyclophosphamide vs matched unrelated donor transplant for acute myeloid leukemia. Blood. 2015;126(8):1033-1040. 84. Stern M, Wreede LC de, Brand R, et al. Sensitivity of hematological malignancies to graft-versus-host effects: an EBMT megafile analysis. Leukemia. 2014;28(11):2235-2240. 85. Gökbuget N, Canaani J, Nagler A, Bishop M, Kröger N, Avigan D. Prevention and treatment of relapse after stem cell transplantation with immunotherapy. Bone Marrow Transplant. 2018;53(6):664-672. 86. Falkenburg F, Ruggiero E, Bonini C, et al. Prevention and treatment of relapse after

1804

stem cell transplantation by cellular therapies. Bone Marrow Transplant. 2019;54(1): 26-34. 87. Zeiser R, Beelen DW, Bethge W, et al. Biology-driven approaches to prevent and treat relapse of myeloid neoplasia after allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2019;25(4):e128-e140. 88. Xuan L, Wang Y, Huang F, et al. Sorafenib maintenance in patients with FLT3-ITD acute myeloid leukaemia undergoing allogeneic haematopoietic stem-cell transplantation: an open-label, multicentre, randomised phase 3 trial. Lancet Oncol. 2020;21(9):1201-

1212. 89. Burchert A, Bug G, Fritz LV, et al. Sorafenib maintenance after allogeneic hematopoietic stem cell transplantation for acute myeloid leukemia with FLT3-internal tandem duplication mutation (SORMAIN). J Clin Oncol. 2020;38(26):2993-3002. 90. Ringdén O, Erkers T, Aschan J, et al. A prospective randomized toxicity study to compare reduced-intensity and myeloablative conditioning in patients with myeloid leukaemia undergoing allogeneic haematopoietic stem cell transplantation. J Intern Med. 2013;274(2):153-162.

haematologica | 2021; 106(7)

REVIEW ARTICLE

Red cell transfusion and alloimmunization in sickle cell disease

Ferrata Storti Foundation

Grace E. Linder1 and Stella T. Chou2

Department of Pathology and Lab Medicine, Children’s Hospital of Philadelphia, and Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, PA, USA

1 2

ABSTRACT

ed cell transfusion remains a critical component of care for acute and chronic complications of sickle cell disease. Randomized clinical trials demonstrated the benefits of transfusion therapy for prevention of primary and secondary strokes and postoperative acute chest syndrome. Transfusion for splenic sequestration, acute chest syndrome, and acute stroke are guided by expert consensus recommendations. Despite overall improvements in blood inventory safety, adverse effects of transfusion are prevalent among patients with sickle cell disease and include alloimmunization, acute and delayed hemolytic transfusion reactions, and iron overload. Judicious use of red cell transfusions, optimization of red cell antigen matching, and the use of erythrocytapheresis and iron chelation can minimize adverse effects. Early recognition and management of hemolytic transfusion reactions can avert poor clinical outcomes. In this review, we discuss transfusion methods, indications, and complications in sickle cell disease with an emphasis on alloimmunization.

Haematologica 2021 Volume 106(7):1805-1815

Introduction Transfusion remains a central intervention for sickle cell disease (SCD), with most patients receiving one or more transfusions by adulthood.1 Prospective, randomized clinical trials support transfusion for primary and secondary stroke prevention, but for many other indications, treatment is based on expert consensus. Guidelines on transfusion management for SCD are limited by availability of welldesigned studies. Thus, many recommendations are based on low or moderate quality evidence or expert consensus, as well as the balance of benefits and harm for any given intervention.2-5 While transfusion therapy reduces SCD-associated morbidity and mortality, attention to prevention and management of alloimmunization, hemolytic transfusion reactions, and iron overload is critical.

Correspondence: STELLA T. CHOU chous@chop.edu Received: January 4, 2021. Accepted: March 7, 2021. Pre-published: April 1, 2021.

Goals of transfusion Red cell transfusion improves oxygen-carrying capacity and symptoms of anemia. For SCD, it may be used to increase a patient’s hematocrit and/or to reduce endogenous production of red cells containing hemoglobin S (HbS). Episodic transfusions are used for preoperative preparation or treatment of acute complications. Chronic transfusion therapy is utilized when the goal is to sustain a lower HbS level, such as for primary or secondary stroke prevention.2,6,7 A standard goal is to maintain HbS levels ≤30% or to raise the hemoglobin to 10-12 g/dL depending on the transfusion indication.1-3 Raising the hemoglobin to levels greater than 10-12 g/dL is generally avoided to limit the risk of hyperviscosity.3

Transfusion method Red cell transfusions can be provided by simple or exchange transfusion. In pediatric patients, simple transfusions are dosed by volume (i.e., 10-15 mL/kg), while in adults simple transfusions are provided in units (i.e., 1-2 units). Simple haematologica | 2021; 106(7)

https://doi.org/10.3324/haematol.2020.270546

1805

G.E. Linder and S.T. Chou

transfusion is convenient, requires one point of peripheral venous access, and utilizes fewer red cell units. Additionally, simple transfusion does not require specialized personnel or devices. Drawbacks of simple transfusion include risks of volume overload and hyperviscosity. Simple transfusion invariably leads to iron overload over time, necessitating treatment with iron chelation or alteration in transfusion modality. Red cell exchange (RCE) procedures involve removal of the patient’s red cells and replacement with cells from the donor. RCE can be provided via automated (erythrocytapheresis) or manual methods. Manual RCE is performed using a series of repeated phlebotomies and transfusions, is time-consuming, and provides less consistent control of fluid balance during the procedure.8 Erythrocytapheresis requires apheresis machines and operators with technical expertise and may only be available at specialty centers. Since RCE typically replaces one or two times the patient’s total red cell volume, a higher volume of replacement cells is required. Despite increased exposure to donors with RCE compared to simple transfusion, several studies have shown no increase in alloimmunization rates associated with RCE.9,10 Limited studies indicate that RCE is cost-effective and may decrease hospitalization rates.8,11 Red cell replacement volume and hematocrit can be tightly controlled with RCE, allowing for significant reductions in HbS levels, while minimizing or preventing iron loading.12 Erythrocytapheresis requires draw and return lines. Venous access must allow a steady flow of blood and withstand the high negative pressures of the draw line. Most adult patients have an adequate peripheral venous access to support RCE, but smaller pediatric patients often require central venous access. Indwelling catheters incur additional risks of infection and thromboembolic events.13 RCE can be further modified to include isovolemic hemodilution, a process that includes initial removal of the patient’s red cells and replacement with normal saline or albumin followed by RCE. RCE with isovolemic hemodilution is not recommended in patients with recent or severe cerebrovascular or cardiopulmonary disease. Potential benefits of isovolemic hemodilution include improved efficiency of RCE, reduced number of red cell units per exchange, and decreased procedure frequency, however a recent meta-analysis found little evidence to support the use of RCE with isovolemic hemodilution over RCE without isovolemic hemodilution.2 RCE is recommended over simple transfusion for acute ischemic stroke, severe acute chest syndrome, for patients with high baseline hematocrits requiring transfusion, and for chronically transfused patients with significant iron overload. Guidelines published by the American Society of Hematology (ASH) suggest using automated RCE in all patients with SCD receiving chronic transfusion therapy; however, individualized decisions for patients should consider availability of compatible red cell units and venous access.2

General transfusion considerations Prior to transfusion, an extended red cell antigen profile, including typing for C/c, E/e, K/k, Fya/Fyb, Jka/Jkb, M/N, and S/s, should be obtained for all patients with 1806

SCD.2 An antigen profile performed by genotyping is preferred, as it provides increased accuracy for C and Fyb antigen expression in this population.2 Serological phenotyping may be inaccurate if the patient has been transfused in the preceding 3 months. Extended red cell antigen profiles guide antigen matching and evaluation of positive antibody screens. It is critical to obtain a patient’s antibody history from all hospitals that provided prior transfusions. The majority of antibodies are not detectable 6 to 12 months after initial identification.14,15 Knowledge of antibody history is necessary to avoid re-exposure to implicated antigens and reduce risk of hemolytic transfusion reactions. Leukocyte reduction decreases the transmission of cytomegalovirus as well as the occurrence of HLA alloimmunization and febrile non-hemolytic transfusion reactions and is standard practice at most transfusion services treating patients with SCD.16 Irradiation prevents transfusion-associated graft-versus-host disease and is required for patients undergoing hematopoietic stem cell transplantation.16 Patients with SCD should receive transfusions negative for sickle cell trait. This aids accurate monitoring of post-transfusion HbS levels, a parameter utilized in chronic exchange programs and when assessing possible delayed hemolytic transfusion reactions (DHTR).1

Indications for transfusion in sickle cell disease Transfusions are a key component of managing SCDassociated complications (Table 1). Patients can experience acute exacerbations of anemia due to parvovirusinduced red cell aplasia, splenic and hepatic sequestration, and vaso-occlusive episodes. Transfusion therapy should be based on symptomatic anemia and hemodynamic compromise rather than hemoglobin value.1 Transfusion is also utilized to decrease the HbS level rapidly in patients experiencing stroke, acute chest syndrome (ACS), and multiorgan failure. The benefit of transfusion has not been well studied for pulmonary hypertension, priapism, and leg ulcers. Transfusion is not indicated for uncomplicated vaso-occlusive episodes.

Neurological complications Cerebrovascular accidents are a significant source of morbidity and mortality in patients with SCD. Prior to implementation of routine screening, between 4-11% of patients experienced a stroke within the first two decades of life, and, without further therapy, two-thirds of patients developed recurrent stroke within 36 months.17,18 The landmark Stroke Prevention Trial in Sickle Cell Anemia (STOP trial) identified children at high risk of stroke using transcranial Doppler to detect elevated internal carotid or middle cerebral artery blood flow velocity.6 Among a randomized cohort of 130 patients, those receiving chronic transfusion therapy had a 92% lower risk of stroke than those in the standard-ofcare arm. In the STOP II study, children whose transcranial Doppler findings had normalized after receiving transfusion therapy for 30 months were randomized to continue or stop chronic transfusion therapy.7 The study was terminated early after a significant proportion of the children who stopped receiving transfusions developed haematologica | 2021; 106(7)

Red cell transfusion and alloimmunization in SCD Table 1. Summary of indications for transfusion therapy in patients with sickle cell disease.

Indication Transient aplastic crisis Acute multisystem organ failure Acute hepatic sequestration Acute splenic sequestration Acute splenic sequestration, recurrent Acute ischemic stroke Primary stroke prevention Secondary stroke prevention Moderate acute chest Severe acute chest Acute chest, recurrent Preoperative with > 1 hour with general anesthesia Pregnancy with complications Pregnancy, uncomplicated Prior to hematopoietic stem cell transplant Uncomplicated vaso-occlusive episode Priapism Leg ulcers Avascular necrosis

Transfusion method

Level of support

Simple Simple or exchange* Simple or exchange Simple via small volume aliquots of 3-5 mL/kg Simple as a bridge to splenectomy Exchange > simple Simple or exchange Simple or exchange Simple or exchange Exchange Simple or exchange Simple or exchange Simple or exchange Simple or exchange Simple or exchange ---------

Expert consensus Expert consensus Expert consensus Expert consensus Expert consensus Observational studies; expert consensus Randomized clinical trial Randomized clinical trial Expert consensus Expert consensus Hydroxyurea preferable Randomized clinical trial Expert consensus Under investigation Under investigation Transfusion not recommended Transfusion not recommended Transfusion not recommended Transfusion not recommended

*The British Committee for Standards in Haematology recommend exchange transfusion for acute multisystem organ failure. Red cell exchange may be preferred for severe, acute multisystem organ failure.

high-risk transcranial Doppler findings or overt stroke in contrast to none of the children in the continued-transfusion arm. Discontinuing transfusions in the STOP II trial was also associated with higher occurrence of silent cerebral infarcts.19 More recently, the Transcranial Doppler with Transfusions Changing to Hydroxyurea (TWiTCH) trial explored transitioning patients with abnormal transcranial Doppler findings but no severe vasculopathy and at least 1 year of chronic transfusions to hydroxyurea versus continuation of chronic transfusions.20 Neither treatment group developed new stroke or evidence of new cerebral infarcts on magnetic resonance imaging, and hydroxyurea therapy at maximum tolerated dose was determined to be non-inferior to standard transfusions. Hydroxyurea can be considered as an alternative therapy for selected patients on chronic transfusion for primary stroke prevention. The Stroke With Transfusions Changing to Hydroxyurea (SWiTCH) study was a multicenter randomized trial comparing transfusions plus chelation to hydroxyurea and phlebotomy in pediatric patients with a history of stroke and iron overload.21 The primary composite endpoint of the study included quantitative liver iron content and stroke recurrence rate. The study closed early after interim analysis indicated that liver iron content was not significantly different between groups. Importantly, although within the range of the study’s non-inferiority margin, there was an imbalance of seven strokes in the hydroxyurea arm compared to no strokes in the subjects receiving transfusions. Silent cerebral infarcts are common in children with SCD and are associated with cognitive deficits and poor educational attainment.22 A history of silent cerebral infarcts predicts an increased risk of recurrent infarct, in the form of both other silent cerebral infarct and overt stroke.23,24 The haematologica | 2021; 106(7)

Silent Infarct Transfusion trial showed that chronic transfusion therapy reduced the incidence of recurrent cerebral infarction in children with SCD.25 However, this study did not compare the efficacy of hydroxyurea to chronic transfusion, so implementation of chronic transfusion for patients with silent cerebral infarcts has not been robust given the availability of hydroxyurea and the burdens of chronic transfusion therapy. Erythrocytapheresis is the preferred transfusion modality for acute stroke given its ability to decrease HbS levels rapidly while limiting effects on serum viscosity. In settings of both acute cerebral ischemia and stroke prevention, maintenance of HbS level ≤30% has been the standard of care.26

Acute chest syndrome Acute chest syndrome (ACS) is one of the most common complications of SCD and is a leading cause of hospitalization and death.27 ACS is defined as a new pulmonary infiltrate on chest radiograph in the presence of respiratory symptoms, hypoxia, chest pain, or fever. Episodes can be triggered by infection, fat embolism, atelectasis, and infarction.27 The clinical course and spectrum of the disease are variable. While studies defining standardized criteria to assess ACS severity are lacking, patients with significant hypoxia or rapidly declining hemoglobin are considered to have severe disease.2 Simple transfusion provided early in the course of moderate ACS often prevents progression of the disease and the need for RCE.28-31 The ASH 2020 guidelines recognized the paucity of large-scale studies but suggest RCE over simple transfusion for patients with severe ACS, rapidly progressive ACS, or ACS in patients with high baseline hemoglobin.2 Recurrent episodes of ACS can lead to chronic lung disease including pulmonary hypertension and fibrosis. 1807

G.E. Linder and S.T. Chou

Hydroxyurea is the primary treatment for prevention of recurrent ACS. Analysis of patients in the STOP trial showed a significantly decreased incidence of ACS in the chronic transfusion arm, and small studies suggest it may reduce frequency of ACS recurrence but not severity.32,33 Chronic transfusion therapy can be considered in patients with recurrent ACS when hydroxyurea is not well tolerated or when hydroxyurea is insufficient to prevent severe, recurrent ACS.

Preoperative transfusion support Operations, including cholecystectomy, splenectomy, and hip surgery, are common in patients with SCD and carry a significant risk of morbidity and mortality.34 The observational Cooperative Study of Sickle Cell Disease demonstrated that patients with SCD undergoing surgery had high rates of pain and ACS, as well as complications such as fever, bleeding, and death.34,35 The Transfusion Alternatives Preoperatively in Sickle Cell Disease (TAPS) study randomized patients with hemoglobin SS and Sβ0 thalassemia SCD requiring low- or medium-risk operations to transfusion or no transfusion preoperatively.36 The rate of postoperative ACS was markedly reduced among patients in the transfusion arm. The majority (85%) of study patients underwent medium-risk operations and, therefore, the relevance of these findings to low-risk surgeries is uncertain. The effect of preoperative transfusion on postoperative pain crises was less clear. Multiple studies have found no significant reduction in postoperative pain in preoperatively transfused patients.2,36,37

The Preoperative Transfusion in Sickle Cell Disease Study Group conducted a multicenter study comparing perioperative complication rates among patients randomized to a conservative preoperative transfusion regimen intended to increase hemoglobin concentration to 10 g/dL or an aggressive transfusion regimen to decrease HbS below 30%.38 There was no difference in rates of ACS, vaso-occlusive episodes, or other serious complications between the two study arms. As such, preoperative transfusion to achieve a hemoglobin of 9-11 g/dL, rather than a goal HbS level, is suggested. Recent guidelines recommend preoperative transfusion for patients undergoing surgery with general anesthesia that is expected to last longer than 1 hour.2 For patients with a high baseline hemoglobin that precludes simple transfusion, preoperative RCE should be performed. RCE should also be considered for patients undergoing high risk cardiovascular or neurosurgical procedures.2

Transfusion support for transplantation and curative therapies There are limited studies examining transfusion considerations in patients with SCD undergoing allogeneic or autologous hematopoietic stem cell transplantation. Blood transfusion prior to transplantation may reduce SCD-associated bone marrow changes and inflammation, possibly improving transplant outcomes. Reducing HbS levels to ≤30% prior to transplantation may also minimize SCD-related complications in the peri-transplant period. Recent studies and ongoing clinical trials utilize

Figure 1. Factors that contribute to alloimmunization in sickle cell disease. The prevalence of alloimmunization in sickle cell disease (SCD) is high compared to that in the general population. One of the key factors behind alloimmunization is recipient-donor mismatch of Rh and K antigens. The majority of Blacks lack C, E, and K antigens. The frequencies of C, E, and K are higher in blood donor populations, leading to increased risk of alloantigen exposure with red cell transfusion. Transfusing red cells matched for Rh and K decreases the rate of alloimmunization; however, RH genetic diversity contributes to a persistent risk of Rh antibody development. Most patients with SCD have one or more RH variants. Common RH alleles in patients with SCD are depicted in panel 2. Red boxes represent RHD exons, and blue boxes represent RHCE exons. The dashed line indicates gene deletion. Vertical lines reflect the positions of amino acid substitutions. Patients with RH variants can form antibodies against the Rh epitopes they lack. SCD is a chronic inflammatory state. Hemolysis leads to elevated levels of circulating hemoglobin and free heme, which activate macrophages and neutrophils, and leads to the secretion of pro-inflammatory cytokines. Patients with high levels of inflammation are at increased risk of alloimmunization. While the immunological pathways contributing to alloimmunization are complex, it is becoming increasingly clear that immune system dysregulation influences antibody formation. IL-1: interleukin -; IL-6: interleukin-6; IFNγ: interferon gamma. Created with Biorender.com.

1808

haematologica | 2021; 106(7)

Red cell transfusion and alloimmunization in SCD

transfusion parameters of hemoglobin <10-11 g/dL and HbS percentage <30% prior to autologous stem cell mobilization and transplantation.39 The optimal transfusion modality and timing peri-transplant are important areas of further study.

Alloimmunization in sickle cell disease Alloantigen specificity and donor/recipient antigen discrepancies Alloimmunization, or formation of antibodies to nonself antigens, is a major adverse effect of transfusion. Alloimmunization increases the risk of hemolytic transfusion reactions and leads to delays in identification of compatible red cell units. While multicenter, registrybased studies have identified a prevalence of red cell alloimmunization of 2-5% in the general population, the prevalence in patients with SCD ranges from 5-75%.40-42 The pathophysiology of alloimmunization in SCD is complex and is associated with level of antigen matching, Rh blood group system diversity, and immune factors (Figure 1). Antibodies to Rh system antigens C and E and to the Kell system antigen K historically accounted for up to two-thirds of alloantibodies in SCD.43,44 Rh and Kell system antigens are among the most immunogenic. Thus, a primary driver behind alloimmunization is recipientdonor mismatch of Rh and K antigens. The majority of Blacks lack C and E antigens (73% and 78%, respectively), and only 2% express the K antigen.45 In the predominantly white blood donor populations in the USA and Europe, the frequencies of C, E, and K are higher, leading to recipient-donor mismatch.43,45 Other clinically relevant antigens, including Jkb in the Kidd system, Fya in the Duffy system, and S in the MNS system are also more common in individuals of European descent. Several studies report a lower prevalence of alloimmunization when the blood donor and SCD patient populations share greater antigenic similarity, but an important caveat is the low transfusion burden of the patients in these reports.46-48 Differences in red cell antigens among patient and donor populations have led to efforts to recruit Black donors to support transfusion of SCD populations.49,50

Red cell antigen matching The British Society for Haematology guidelines, the ASH 2020 guidelines, and the National Institutes of Health Expert Panel recommend prophylactic matching for Rh (C, E or C/c, E/e), and K in addition to ABO and D in patients with SCD.2-4,51 Transfusion of red cells matched for Rh and K decreases the rate of alloimmunization from 1.7-3.9 antibodies per 100 units transfused to 0.26-0.50 antibodies per 100 units transfused.41 Additional antigen matching extended for Fya/Fyb, Jka/Jkb, M/N, S/s, Lea/Leb, and P antigens further reduces the rate of alloantibody formation to 0.1-0.3 per 100 units transfused, but finding sufficient compatible units becomes significantly more challenging.52-54 An extended red cell antigen profile should be obtained in all patients with SCD at the earliest opportunity.2 Extended antigen identification has traditionally been performed by manual serological phenotyping methods. As most blood group antigens are due to single nucleotide polymorphisms, high throughput genotyping systems haematologica | 2021; 106(7)

can identify Rh, Kell, Kidd, Duffy, MNS, Lutheran, Diego, Dombrock, and Colton blood group system antigens.55 DNA-based red cell typing is more accurate than serological phenotyping and provides additional information, such as whether a patient has a GATA mutation in the ACKR1 gene and is, therefore, not at risk of forming antiFyb antibodies.56 Genotyping methods also provide increased accuracy for C antigen expression in this population, predict antigen expression when no antisera is available, and should be used over serological phenotyping if the patient has been transfused in the preceding 3 months. Several studies have utilized molecular genotyping to support extended antigen matching between blood donors and patients with SCD.57,58

RH diversity and the role of RH genotyping Despite serological matching for D and C, E or C/c, E/e antigens, Rh alloimmunization persists due to RH genetic diversity in individuals of African descent.49,59 The RHD and RHCE genes are located on chromosome 1, arose through gene duplication, and encode D and C, c, E, e antigens, respectively.60 The two loci are highly homologous, leading to many gene recombination events resulting in variant RHD and RHCE alleles that encode altered antigens.45,60 Rh variant antigens are difficult to distinguish serologically and require RH genotyping for identification. While RH variants can result in weak (decreased antigen density) or partial (missing epitopes) antigen expression, Blacks typically carry alleles in the latter category and are at risk of alloantibody formation when exposed to the epitopes they lack via transfusion, pregnancy, or transplantation. High suspicion must be maintained when apparent autoantibodies with Rh specificity or unexplained Rh antibodies are detected in patients with SCD, and further investigation with RH genotyping should be pursued. Most patients with SCD have one or more RH allele variants.49,54,59 Two common variants in Blacks, RHD*DAU0 and RHCE*ce48C, have not been shown to encode Rh proteins lacking epitopes and are considered “altered” antigens.54 RHD*D 4.0, *DIVa, *DAU3, and *DIIIa are frequently detected variants in patients with SCD and result in partial D antigen expression.61,62 The hybrid RHD*DIIIa-CE(4-7)-D allele results from RHCE exons 4 through 7 replacing the corresponding exons of RHD. This allele encodes a partial C antigen and no D antigen. Individuals with RHD*DIIIa-CE(4-7)-D who lack conventional RHCE*Ce or *CE alleles serologically type as C positive but are at risk of developing alloanti-C if exposed to conventional C antigen.63 Variant RhCE antigens resulting in partial c and e antigens are particularly common in Blacks. Individuals with homozygous ce variants often make allo-anti-e antibodies and may also lack the high frequency hrB and hrS antigens.64 Formation of alloantibodies against hrB and hrS, present on the red cells of 98% of individuals, can pose a challenge to identification of compatible donor units.45 Anti-hrB and -hrS may initially appear as having an anti-e specificity. Knowledge of the patient’s RH genotype, which identifies those who are hrB and hrS negative, can facilitate proper antibody evaluation and distinguish these antibodies from anti-e. E-e+ patients with partial e antigens who form allo-anti-e are at risk of anti-E if transfused with E+e- red cells. While each clinical scenario requires individual decision-making, if there was no associated DHTR 1809

G.E. Linder and S.T. Chou

with the anti-e, we have cautiously transfused patients with e+ blood without evidence of a DHTR or anti-e reappearance. In the future, RH genotype matched red cells would be the ideal choice. The role of RH genotyping in blood donors and patients with SCD and genotype matching to prevent alloimmunization is currently under investigation. While systematic RH genotyping of patients with SCD may aid in blood product selection and reduce the risk of alloimmunization in patients with variant RH alleles, a large pool of genotyped Black donors would be needed as well. Universal RH genotyping is currently cost-prohibitive in most settings, but one study has shown that prophylactic RH matching based on genotype for patients with SCD is achievable but would require recruitment of double the number of Black blood donors as compared to those for serological matching.54

Inflammation and immune system regulation in alloimmunization A subset of patients with SCD do not form alloantibodies despite repeated transfusions. Genetic modifiers and differences in immune regulation likely contribute to an individual’s risk. HLA molecules present foreign red cell antigens to T cells. Activated CD4+ T cells stimulate B-cell responses and differentiation into plasma cells. The class II HLA alleles HLA-DQ2, HLA-DQ3, and HLA-DQ5 are associated with lower risk of red cell alloimmunization, while HLA-DQ7, HLA-DRB1*04, and HLA-DRB1*15 may be associated with increased risk.65,66 Efforts to identify genetic markers for alloimmunization in patients with SCD have demonstrated moderate associations but no strong predictors.67,68 Patients with high baseline levels of inflammation, such as those with autoimmune disease, have higher rates of alloimmunization.40 SCD is a chronic inflammatory state in which hemolysis results in elevated levels of circulating hemoglobin and free heme, activating neutrophils and macrophages and causing secretion of pro-inflammatory cytokines. Accordingly, patients with SCD have higher levels of pro-inflammatory cytokines, including interleukin-1, interleukin-6, and interferon-γ, as compared to the levels in healthy controls.69 Fasano and colleagues demonstrated that patients with SCD who received transfusions during inflammatory events such as ACS and vaso-occlusive episodes had an increased rate of alloimmunization.70 Chronic inflammation can lead to immune system dysregulation. Several studies have shown that regulatory T cells, which control T-cell responses, display higher levels of inhibitory markers such as CTLA-4 and are dysfunctional in patients with SCD.71 Furthermore, regulatory B cells from alloimmunized patients with SCD have a decreased ability to suppress monocyte activation.72 Pal et al.73 demonstrated that hemolysis and cell-free heme typically suppress B cells and plasma cell differentiation, but alloimmunized patients with SCD had altered B-cell inhibition. Further mechanistic studies are required to elucidate the complex immunological pathways contributing to alloimmunization in SCD and to determine whether targeted reversal of immune dysregulation can reduce antibody formation.

Clinical impact of alloimmunization Red cell alloimmunization in patients with SCD significantly increases the risk of hemolytic transfusion reac1810

tions, including hyperhemolytic reactions. Identifying compatible blood for patients with multiple alloantibodies or antibodies to high prevalence antigens can be challenging or even impossible, potentially leading to transfusion delays and poor outcomes.

Additional complications of packed red blood cell transfusion Delayed hemolytic transfusion reactions and hyperhemolysis DHTR is a feared adverse outcome of transfusion in SCD.74 DHTR classically occurs after re-exposure to a red cell antigen that the patient had previously been immunized against. As many as 80% of alloantibodies in patients with SCD become undetectable.75 In patients with SCD, 3040% of DHTR are associated with no identifiable antibodies, and in one-third of cases, autoantibodies or antibodies of unclear specificity are the only detectable finding.76,77 The mechanisms of red cell destruction in antibody-negative DHTR have not been elucidated, however hypotheses include hyperactivated macrophages and red cells with increased membrane exposure of phosphatidylserine.78,79 Alternatively, the antibody may simply be difficult to detect. The most severe complication is hyperhemolysis, in which hemolysis of bystander autologous cells occurs, leading to a hemoglobin level lower than pre-transfusion levels and often life-threatening anemia. The reported incidence of DHTR in SCD is 4.8-7.7%.80,81 These rates may be underestimates, as many DHTR are misdiagnosed as vaso-occlusive episodes or go undetected. In one of the largest cohorts to date, the most common clinical manifestations of DHTR were hemoglobinuria, pain, and fever.76 Only 44% of patients had overt signs of anemia. Signs and symptoms of DHTR vary among individuals. The recent ASH guidelines define DHTR as a significant drop in hemoglobin within 21 days after transfusion in the presence of hemoglobinuria, newly detected alloantibodies, accelerated increase in HbS, significant change in reticulocyte percentage, or increase in lactate dehydrogenase level above baseline.2 Rapid decline in HbA concentration relative to an early post-transfusion measurement is highly predictive of DHTR. Risk factors include a history of alloimmunization, prior DHTR, and transfusion for acute complications.76,80-82 High suspicion must be maintained when patients with SCD present with pain, fever, or worsening anemia in the days to weeks after transfusion. Habibi et al. reported that 92% of patients with DHTR were not immediately diagnosed.76 Review of the transfusion history is required, and an antibody screen, direct antiglobulin test (DAT) with elution, and hemoglobin electrophoresis should be performed if the patient has been recently transfused. Between 2560% of DHTR are associated with newly detected red cell antibodies, and approximately 80% of DHTR are DAT positive.76,81,82 Time to DAT positivity is variable, and it is recommended that the DAT and antibody screen are repeated 1-2 weeks after presentation in cases of DHTR that are initially antibody-negative. Additional transfusions may exacerbate hemolysis, particularly when the antibody is not identified. Upon recognition of a DHTR, further transfusion should be avoided if possible. If transfusion is necessary and no antibody specificity has been identified, extended antigen matching for haematologica | 2021; 106(7)

Red cell transfusion and alloimmunization in SCD

C/c, E/e, K, Jka/Jkb, Fya/Fyb, and S/s is recommended.2 Many patients improve with hydration, oxygen support, and pain management alone, but others develop severe complications such as ACS and multiorgan failure.76 Erythropoietin with or without intravenous iron is an additional supportive measure. High-dose steroids and intravenous immunoglobulins are suggested first-line therapy for patients with severe DHTR or ongoing hemolysis, although caution must be maintained, as high-dose steroids have been associated with rebound symptoms and worsening vaso-occlusive episodes.2,82,83 Prophylactic treatment with steroids and intravenous immunoglobulins prior to transfusion should be considered for patients with a history of multiple or life-threatening DHTR or for those for whom compatible blood is not available. Rituximab, an anti-CD20 monoclonal antibody, can be used to reduce the risk of further alloimmunization when future transfusion is likely and may be employed as a prophylactic therapy prior to transfusion for patients with a history of multiple or severe DHTR.83 There is growing evidence that the complement pathway plays a pivotal role in the pathogenesis of DHTR. Alloantibody-antigen complexes activate the classical complement pathway. Free heme and hemoglobin trigger the alternative complement pathway, leading to endothelial damage and organ injury.84,85 Eculizumab, a monoclonal anti-C5 antibody targeting terminal complement activation, has been used as salvage therapy in cases of severe DHTR and hyperhemolysis.85-92 Case reports have also described treatment of hyperhemolysis with tocilizumab, a monoclonal antibody against the interleukin-6 receptor.93-95 These cases showed marked improvement after targeted antiinterleukin-6 receptor therapy, suggesting that blockade of macrophage activation may be an effective treatment strategy. Table 2 summarizes published reports describing use of eculizumab and tocilizumab in patients with SCD.

DHTR are associated with high mortality, underscoring the importance of early recognition and treatment, but prevention is key.80 Red cell exposure should be minimized by transfusing only for evidence-based indications.2 All patients should be prophylactically matched for Rh (C, E or C/c, E/e) and K antigens. Patients at high risk of DHTR should be identified and transfusions avoided or delivered with immunomodulatory agents as possible.2,80 Incomplete transfusion and alloantibody histories contribute to the incidence of DHTR. Most countries lack national transfusion databases, so patients should be encouraged to make new providers aware of their transfusion history and limit transfusions to one institution, if possible.

Iron overload Each milliliter of transfused red cells contains 0.8-1 mg of iron. Transfusion of 3-5 units of packed red cells delivers 1 g of iron, a significant burden considering the total body iron of an average adult is 4-5 g. The human body has no mechanisms for excreting excess iron. While small amounts of iron are lost through the gastrointestinal tract and skin, iron homeostasis is primarily regulated by hepcidin, a protein synthesized by the liver in response to iron overload and inflammation.96 Hepcidin inhibits dietary iron absorption and blocks iron recycling through the reticuloendothelial system. Transfusional iron is delivered outside of these normal regulatory mechanisms, and there are no means of eliminating large amounts of iron from transfusion. In SCD, iron accumulation is most prominent in the liver. Compared to thalassemia patients with equivalent transfusion volumes, patients with SCD are less vulnerable to iron overload-induced endocrinopathies and heart failure; however, iron cardiomyopathy is detectable in 2.5% of SCD patients receiving chronic transfusion therapy.97,98 Iron toxicity is estimated to contribute to 7-11% of deaths in patients with SCD.99,100

Table 2. Reports of studies investigating eculizumab and tocilizumab for the treatment of delayed hemolytic transfusion reactions and hyperhemolysis in patients with sickle cell disease.

Study

Drug investigated

Dose

Number of patients

Adverse events

Boonyasampant et al. 2015

Eculizumab

None reported

Dumas et al. 2016

Eculizumab

1200 mg weekly x 4 weeks followed by every 2 weeks for 2 more doses 900 mg x 2 dosed 1 week apart

Chonat et al. 2018 Vlachaki et al. 2018 Unnikrishnan et al. 2019 Chonat et al. 2020 Floch et al. 2020

Eculizumab Eculizumab Eculizumab Eculizumab Eculizumab

600 mg x 2 900 mg x 1 900 mg x 1 600 mg weekly x 4 weeks 1-3 doses

1 1 1 1 18

Mpinganzima et al. 2020 Sivapalaratnam et al. 2019 Lee et al. 2020

Eculizumab Tocilizumab Tocilizumab

900 mg x 2 dosed 6 days apart 8 mg/kg daily x 2 days 8 mg/kg daily x 4 days

1 1 1

Eculizumab and Tocilizumab

900 mg x 3 8 mg/kg x 1

1 death secondary to severe pulmonary infection None reported None reported None reported None reported 3 patient deaths (2 from complications of encapsulated bacterial infection) None reported None reported Seizure (in the setting of methemoglobinemia secondary to hemoglobin-based oxygen carrier) None reported

Hair et al. 2021

haematologica | 2021; 106(7)

1811

G.E. Linder and S.T. Chou

Regular assessment of iron overload is recommended for patients with SCD receiving chronic transfusion therapy.2 While serum ferritin levels are widely available, relatively inexpensive, and can be easily serially monitored, ferritin is an acute phase reactant and its levels do not always correlate with total body iron stores. Magnetic resonance imaging is currently the recommended technique for quantifying liver iron.101,102 Both R2 and R2* magnetic resonance imaging data show strong correlation with iron levels on liver biopsy but are not interchangeable, so the same method should be used to monitor a patient longitudinally.102 Regular assessment of liver iron concentration by liver magnetic resonance imaging every 1-2 years is recommended for chronically transfused patients with SCD or those with sustained serum ferritin levels ≥1000 ng/mL.2 Given the rarity of cardiac iron overload in SCD, routine screening for cardiac iron levels by T2* magnetic resonance imaging is recommended only for patients with evidence of cardiac dysfunction or a severe iron overload (liver iron content >15-20 mg/g).2 Iron chelation is recommended for patients on chronic transfusion therapy who have sustained serum ferritin levels >1000 ng/mL or liver iron content >3-7 mg/g liver dry weight (normal range 0.8-1.5 mg/g liver dry weight).102 There are currently three iron chelators licensed and approved for use in Europe and the USA, all of which have been shown to be effective in mitigating iron overload in patients with hemoglobinopathies (Table 3).102-104 Successful chelation therapy is dependent upon the patients’ adherence and the tolerability and toxicities of the drugs used. Chronic transfusion modality and transfusion parameters can be modulated to reduce iron loading. In erythrocytapheresis, post-procedure hematocrit can be targeted to a value equal to or lower than the pre-procedure hematocrit to maintain a neutral or net negative iron balance. For patients who transition from simple transfusion to RCE due to iron overload, targeting a slightly lower hematocrit at the end of the procedure than the pre-transfusion hematocrit will reduce total body iron stores over time. Chelation therapy can be combined with RCE for greater reduction in liver iron content for iron-overloaded patients requiring chronic transfusions.105

Global challenges in transfusion support for sickle cell disease The worldwide incidence of SCD is highest in subSaharan Africa, accounting for approximately 75% of the global burden of SCD.106,107 Although red cell transfusion significantly reduces morbidity and mortality associated

with SCD, transfusion support in sub-Saharan Africa is limited by the availability and safety of blood products. While 13% of the global population resides in subSaharan Africa, only 4% of blood donations occur in this region.108 Blood donations in Africa have increased over the past decade, but widespread blood shortages remain.108 The high cost of blood products in these regions poses a further challenge.109 Red cell and whole blood transfusions in patients with SCD in sub-Saharan Africa are often restricted to patients with acute complications. Transcranial Doppler screening is not widely available, and chronic transfusion therapy is often unattainable. Several studies have supported higher rates of transfusion reactions in regions of Africa compared to those in higher-resource regions, attributable to factors including limited implementation of leukoreduction, challenges in maintaining temperature control during storage of blood products, and need for effective quality and transfusion education systems.109,110 While most countries in Africa routinely screen blood products for human immunodeficiency virus, hepatitis B, and hepatitis C, the residual risk of transfusion-transmitted viral infection is relatively high, particularly in countries with higher percentages of paid or family/replacement blood donors.108,111 Transfusion-transmitted malaria and emerging infectious diseases pose additional burdens. Pre-transfusion testing in sub-Saharan Africa typically comprises ABO and D typing and saline crossmatching. Antiglobulin reagents are in limited supply, and antibody screening is not consistently performed in most settings.109,112 Although pre-transfusion antigen typing and prophylactic antigen matching are not routinely available in low-income countries, rates of alloimmunization among patients with SCD in sub-Saharan Africa may be equivalent to or lower than those in higher or middle income countires.112 It is hypothesized that reduced transfusion rates and greater antigenic similarity between donor and recipient populations contribute to these findings, although further studies are required. Efforts to improve transfusion support for patients with SCD living in lower-resource countries are paramount and, along with other measures, such as increasing availability of hydroxyurea, would undoubtedly improve care and quality of life for the majority of patients with SCD.

Future directions Transfusion therapy is a cornerstone of treatment in SCD. Clinical trials have proven that transfusions are

Table 3. Characteristics of iron chelators. Dose Route of administration

Route of excretion Toxicities

1812

Deferasirox (Exjade, Jadenu)

Deferoxamine (Desferal)

Deferiprone (Ferriprox)

14-28 mg/kg/day (Jadenu) 20-40 mg/kg/day (Exjade) Oral [Jadenu: coated tablet or sprinkles] [Exjade: dispersible tablet] Fecal Gastrointestinal upset, proteinuria, renal dysfunction, raised transaminases, gastrointestinal bleeding (rare)

40-50 mg/kg/day

75-100 mg/kg/day

Parenteral (intravenous or subcutaneous)

Oral

Urine and fecal Injection site reactions, anaphylaxis (rare), infection, renal and auditory impairment (rare)

Urine Gastrointestinal upset, raised transaminases, arthropathy, rash, neutropenia, agranulocytosis haematologica | 2021; 106(7)

Red cell transfusion and alloimmunization in SCD

effective means of preventing stroke, decreasing postoperative ACS, and reducing morbidity and mortality in SCD. Given the risks of iron overload, alloimmunization, and DHTR, transfusion should be used judiciously for evidence-based indications or those defined by expert consensus. Despite increased understanding of the pathophysiology of alloimmunization in SCD and improved execution of Rh and K antigen matching, high rates of alloimmunization persist. Future work is necessary to determine whether extended antigen matching or prophylactic RH genotype matching can reduce alloimmunization in a cost-effective manner. Recruitment of diverse donors and broad implementation of donor genotyping will increase compatible donors for patients with

References 1. Smith-Whitley K, Thompson AA. Indications and complications of transfusions in sickle cell disease. Pediatr Blood Cancer. 2012;59(2):358-364. 2. Chou ST, Alsawas M, Fasano RM, et al. American Society of Hematology 2020 guidelines for sickle cell disease: transfusion support. Blood Adv. 2020;4(2):327-355. 3. Davis BA, Allard S, Qureshi A, et al. Guidelines on red cell transfusion in sickle cell disease. Part I: principles and laboratory aspects. Br J Haematol. 2017;176(2):179-191. 4. National Institutes of Health. Evidencebased management of sickle cell disease: expert panel report. National Institutes of Health, Bethesda, MD: https://www nhlbi nih gov/health-pro/guidelines/sickle-celldisease-guidelines. 2014. 5. Davis BA, Allard S, Qureshi A, et al. Guidelines on red cell transfusion in sickle cell disease Part II: indications for transfusion. Br J Haematol. 2017;176(2):192-209. 6. Adams RJ, McKie VC, Hsu L, et al. Prevention of a first stroke by transfusions in children with sickle cell anemia and abnormal results on transcranial Doppler ultrasonography. N Engl J Med. 1998;339(1):5-11. 7. Adams RJ, Brambilla D. Discontinuing prophylactic transfusions used to prevent stroke in sickle cell disease. N Engl J Med. 2005;353(26):2769-2778. 8. Dedeken L, Lê PQ, Rozen L, et al. Automated RBC exchange compared to manual exchange transfusion for children with sickle cell disease is cost-effective and reduces iron overload. Transfusion. 2018;58(6):1356-1362. 9. Venkateswaran L, Teruya J, Bustillos C, Mahoney D Jr., Mueller BU. Red cell exchange does not appear to increase the rate of allo- and auto-immunization in chronically transfused children with sickle cell disease. Pediatr Blood Cancer. 2011;57(2):294-296. 10. Wahl SK, Garcia A, Hagar W, Gildengorin G, Quirolo K, Vichinsky E. Lower alloimmunization rates in pediatric sickle cell patients on chronic erythrocytapheresis compared to chronic simple transfusions. Transfusion. 2012;52(12):2671-2676. 11. Tsitsikas DA, Ekong A, Berg L, et al. A 5year cost analysis of automated red cell exchange transfusion for the management of recurrent painful crises in adult patients with sickle cell disease. Transfus Apher Sci. 2017;56(3):466-469. 12. Kim HC, Dugan NP, Silber JH, et al. Erythrocytapheresis therapy to reduce iron

haematologica | 2021; 106(7)

SCD. The need to improve safety and ensure access to reliable transfusion therapy for patients with SCD worldwide remains. Disclosures No conflicts of interest to disclose. Contributions GEL and STC wrote the manuscript. Acknowledgements We acknowledge support from the National Institutes of Health/National Heart Lung Blood Institute through grants HL134696 and HL147879-01 (to STC).

overload in chronically transfused patients with sickle cell disease. Blood. 1994;83(4):1136-1142. 13. Jeng MR, Feusner J, Skibola C, Vichinsky E. Central venous catheter complications in sickle cell disease. Am J Hematol. 2002;69 (2):103-108. 14. Tormey CA, Stack G. The persistence and evanescence of blood group alloantibodies in men. Transfusion. 2009;49(3):505-512. 15. Coleman S, Westhoff CM, Friedman DF, Chou ST. Alloimmunization in patients with sickle cell disease and underrecognition of accompanying delayed hemolytic transfusion reactions. Transfusion. 2019;59(7): 2282-2291. 16. Gehrie EA, Dunbar NM. Modifications to blood components: when to use them and what is the evidence? Hematol Oncol Clin North Am. 2016;30(3):653-663. 17. Ohene-Frempong K, Weiner SJ, Sleeper LA, et al. Cerebrovascular accidents in sickle cell disease: rates and risk factors. Blood. 1998;91(1):288-294. 18. Powars D, Wilson B, Imbus C, Pegelow C, Allen J. The natural history of stroke in sickle cell disease. Am J Med. 1978;65(3):461471. 19. Abboud MR, Yim E, Musallam KM, Adams RJ. Discontinuing prophylactic transfusions increases the risk of silent brain infarction in children with sickle cell disease: data from STOP II. Blood. 2011;118(4):894-898. 20. Ware RE, Davis BR, Schultz WH, et al. Hydroxycarbamide versus chronic transfusion for maintenance of transcranial doppler flow velocities in children with sickle cell anaemia-TCD With Transfusions Changing to Hydroxyurea (TWiTCH): a multicentre, open-label, phase 3, non-inferiority trial. Lancet. 2016;387(10019):661-670. 21. Ware RE, Helms RW. Stroke With Transfusions Changing to Hydroxyurea (SWiTCH). Blood. 2012;119(17):3925-3932. 22. Schatz J, Brown RT, Pascual JM, Hsu L, DeBaun MR. Poor school and cognitive functioning with silent cerebral infarcts and sickle cell disease. Neurology. 2001;56(8): 1109-1111. 23. Jordan LC, Kassim AA, Donahue MJ, et al. Silent infarct is a risk factor for infarct recurrence in adults with sickle cell anemia. Neurology. 2018;91(8):e781-e784. 24. Miller ST, Macklin EA, Pegelow CH, et al. Silent infarction as a risk factor for overt stroke in children with sickle cell anemia: a report from the Cooperative Study of Sickle Cell Disease. J Pediatr. 2001;139(3):385-390. 25. DeBaun MR, Gordon M, McKinstry RC, et al. Controlled trial of transfusions for silent

cerebral infarcts in sickle cell anemia. N Engl J Med. 2014;371(8):699-710. 26. DeBaun MR, Jordan LC, King AA, et al. American Society of Hematology 2020 guidelines for sickle cell disease: prevention, diagnosis, and treatment of cerebrovascular disease in children and adults. Blood Adv. 2020;4(8):1554-1588. 27. Vichinsky EP, Neumayr LD, Earles AN, et al. Causes and outcomes of the acute chest syndrome in sickle cell disease. National Acute Chest Syndrome Study Group. N Engl J Med. 2000;342(25):1855-1865. 28. Emre U, Miller ST, Gutierez M, Steiner P, Rao SP, Rao M. Effect of transfusion in acute chest syndrome of sickle cell disease. J Pediatr. 1995;127(6):901-904. 29. Turner JM, Kaplan JB, Cohen HW, Billett HH. Exchange versus simple transfusion for acute chest syndrome in sickle cell anemia adults. Transfusion. 2009;49(5):863-868. 30. Saylors RL, Watkins B, Saccente S, Tang X. Comparison of automated red cell exchange transfusion and simple transfusion for the treatment of children with sickle cell disease acute chest syndrome. Pediatr Blood Cancer. 2013;60(12):1952-1956. 31. Miller ST, Rao SP. Acute chest syndrome, transfusion, and neurologic events in children with sickle cell disease. Blood. 2003;102(4):1556; author reply 1556-1557. 32. Hankins J, Jeng M, Harris S, Li CS, Liu T, Wang W. Chronic transfusion therapy for children with sickle cell disease and recurrent acute chest syndrome. J Pediatr Hematol Oncol. 2005;27(3):158-161. 33. Miller ST, Wright E, Abboud M, et al. Impact of chronic transfusion on incidence of pain and acute chest syndrome during the Stroke Prevention Trial (STOP) in sickle-cell anemia. J Pediatr. 2001;139(6):785-789. 34. Koshy M, Weiner SJ, Miller ST, et al. Surgery and anesthesia in sickle cell disease. Cooperative Study of Sickle Cell Diseases. Blood. 1995;86(10):3676-3684. 35. Haberkern CM, Neumayr LD, Orringer EP, et al. Cholecystectomy in sickle cell anemia patients: perioperative outcome of 364 cases from the National Preoperative Transfusion Study. Preoperative Transfusion in Sickle Cell Disease Study Group. Blood. 1997;89 (5):1533-1542. 36. Howard J, Malfroy M, Llewelyn C, et al. The Transfusion Alternatives Preoperatively in Sickle Cell Disease (TAPS) study: a randomised, controlled, multicentre clinical trial. Lancet. 2013;381(9870):930-938. 37. Al-Jaouni SK, Al-Muhayawi SM, Qari MH, Nawas MA, Al-Mazrooa A. Randomized clinical trial to evaluate the safety of avoid-

1813

G.E. Linder and S.T. Chou ing pre-operative transfusion in sickle cell anaemia. Bahrain Med Bull. 2006;28(4):164167. 38. Vichinsky EP, Haberkern CM, Neumayr L, et al. A comparison of conservative and aggressive transfusion regimens in the perioperative management of sickle cell disease. The Preoperative Transfusion in Sickle Cell Disease Study Group. N Engl J Med. 1995;333(4):206-213. 39. Lagresle-Peyrou C, Lefrère F, Magrin E, et al. Plerixafor enables safe, rapid, efficient mobilization of hematopoietic stem cells in sickle cell disease patients after exchange transfusion. Haematologica. 2018;103(5):778-786. 40. Karafin MS, Westlake M, Hauser RG, et al. Risk factors for red blood cell alloimmunization in the Recipient Epidemiology and Donor Evaluation Study (REDS-III) database. Br J Haematol. 2018;181(5):672-681. 41. Fasano RM, Meyer EK, Branscomb J, White MS, Gibson RW, Eckman JR. Impact of red blood cell antigen matching on alloimmunization and transfusion complications in patients with sickle cell disease: a systematic review. Transfus Med Rev. 2019;33(1):12-23. 42. Chou ST, Liem RI, Thompson AA. Challenges of alloimmunization in patients with haemoglobinopathies. Br J Haematol. 2012;159(4):394-404. 43. Vichinsky EP, Earles A, Johnson RA, Hoag MS, Williams A, Lubin B. Alloimmunization in sickle cell anemia and transfusion of racially unmatched blood. N Engl J Med. 1990;322(23):1617-1621. 44. Rosse WF, Gallagher D, Kinney TR, et al. Transfusion and alloimmunization in sickle cell disease. The Cooperative Study of Sickle Cell Disease. Blood. 1990;76(7):1431-1437. 45. Reid ME, Lomas-Francis C. The Blood Group Antigen FactsBook. 2004. Academic Press. 46. Olujohungbe A, Hambleton I, Stephens L, Serjeant B, Serjeant G. Red cell antibodies in patients with homozygous sickle cell disease: a comparison of patients in Jamaica and the United Kingdom. Br J Haematol. 2001;113(3):661-665. 47. Natukunda B, Schonewille H, Ndugwa C, Brand A. Red blood cell alloimmunization in sickle cell disease patients in Uganda. Transfusion. 2010;50(1):20-25. 48. Boateng LA, Campbell AD, Davenport RD, et al. Red blood cell alloimmunization and minor red blood cell antigen phenotypes in transfused Ghanaian patients with sickle cell disease. Transfusion. 2019;59(6):2016-2022. 49. Chou ST, Jackson T, Vege S, Smith-Whitley K, Friedman DF, Westhoff CM. High prevalence of red blood cell alloimmunization in sickle cell disease despite transfusion from Rh-matched minority donors. Blood. 2013;122(6):1062-1071. 50. Sesok-Pizzini DA, Friedman DF, SmithWhitley K, Nance SJ. Transfusion support of patients with sickle cell disease at the Children's Hospital of Philadelphia. Immunohematology. 2006;22(3):121-125. 51. Yawn BP, Buchanan GR, Afenyi-Annan AN, et al. Management of sickle cell disease: summary of the 2014 evidence-based report by expert panel members. JAMA. 2014;312 (10):1033-1048. 52. Ambruso DR, Githens JH, Alcorn R, et al. Experience with donors matched for minor blood group antigens in patients with sickle cell anemia who are receiving chronic transfusion therapy. Transfusion. 1987;27(1):9498. 53. Lasalle-Williams M, Nuss R, Le T, et al. Extended red blood cell antigen matching

1814

for transfusions in sickle cell disease: a review of a 14-year experience from a single center (CME). Transfusion. 2011;51(8):17321739. 54. Chou ST, Evans P, Vege S, et al. RH genotype matching for transfusion support in sickle cell disease. Blood. 2018;132(11):1198-1207. 55. Veldhuisen B, van der Schoot CE, de Haas M. Blood group genotyping: from patient to high-throughput donor screening. Vox Sang. 2009;97(3):198-206. 56. Casas J, Friedman DF, Jackson T, Vege S, Westhoff CM, Chou ST. Changing practice: red blood cell typing by molecular methods for patients with sickle cell disease. Transfusion. 2015;55(6 Pt 2):1388-1393. 57. Wilkinson K, Harris S, Gaur P, et al. Molecular blood typing augments serologic testing and allows for enhanced matching of red blood cells for transfusion in patients with sickle cell disease. Transfusion. 2012;52(2):381-388. 58. Ribeiro KR, Guarnieri MH, da Costa DC, Costa FF, Pellegrino J Jr, Castilho L. DNA array analysis for red blood cell antigens facilitates the transfusion support with antigen-matched blood in patients with sickle cell disease. Vox Sang. 2009;97(2):147-152. 59. Sippert E, Fujita CR, Machado D, et al. Variant RH alleles and Rh immunisation in patients with sickle cell disease. Blood Transfus. 2015;13(1):72-77. 60. Cohn CS, Delaney M, Johnson ST, Katz LM. Technical Manual, 20th edition. Bethesda, MD: AABB Press, 2020. 61. Chou ST, Flanagan JM, Vege S, et al. Wholeexome sequencing for RH genotyping and alloimmunization risk in children with sickle cell anemia. Blood Adv. 2017;1(18):14141422. 62. Srivastava K, Polin H, Sheldon SL, et al. The DAU cluster: a comparative analysis of 18 RHD alleles, some forming partial D antigens. Transfusion. 2016;56(10):2520-2531. 63. Tournamille C, Meunier-Costes N, Costes B, et al. Partial C antigen in sickle cell disease patients: clinical relevance and prevention of alloimmunization. Transfusion. 2010;50 (1):13-19. 64. Noizat-Pirenne F, Lee K, Pennec PY, et al. Rare RHCE phenotypes in black individuals of Afro-Caribbean origin: identification and transfusion safety. Blood. 2002;100(12): 4223-4231. 65. Tatari-Calderone Z, Gordish-Dressman H, Fasano R, et al. Protective effect of HLADQB1 alleles against alloimmunization in patients with sickle cell disease. Hum Immunol. 2016;77(1):35-40. 66. Hoppe C, Klitz W, Vichinsky E, Styles L. HLA type and risk of alloimmunization in sickle cell disease. Am J Hematol. 2009;84(7): 462-464. 67. Meinderts SM, Gerritsma JJ, Sins JWR, et al. Identification of genetic biomarkers for alloimmunization in sickle cell disease. Br J Haematol. 2019;186(6):887-899. 68. Williams LM, Qi Z, Batai K, et al. A locus on chromosome 5 shows African ancestry-limited association with alloimmunization in sickle cell disease. Blood Adv. 2018;2(24): 3637-3647. 69. Jison ML, Munson PJ, Barb JJ, et al. Blood mononuclear cell gene expression profiles characterize the oxidant, hemolytic, and inflammatory stress of sickle cell disease. Blood. 2004;104(1):270-280. 70. Fasano RM, Booth GS, Miles M, et al. Red blood cell alloimmunization is influenced by recipient inflammatory state at time of transfusion in patients with sickle cell dis-

ease. Br J Haematol. 2015;168(2):291-300. 71. Vingert B, Tamagne M, Desmarets M, et al. Partial dysfunction of Treg activation in sickle cell disease. Am J Hematol. 2014;89(3): 261-266. 72. Bao W, Zhong H, Manwani D, et al. Regulatory B-cell compartment in transfused alloimmunized and non-alloimmunized patients with sickle cell disease. Am J Hematol. 2013;88(9):736-740. 73. Pal M, Bao W, Wang R, et al. Hemolysis inhibits humoral B cell responses and modulates alloimmunization risk in patients with sickle cell disease. Blood. 2021;137(2):269280. 74. Thein SL, Pirenne F, Fasano RM, et al. Hemolytic transfusion reactions in sickle cell disease: underappreciated and potentially fatal. Haematologica. 2020;105(3):539-544. 75. Harm SK, Yazer MH, Monis GF, Triulzi DJ, Aubuchon JP, Delaney M. A centralized recipient database enhances the serologic safety of RBC transfusions for patients with sickle cell disease. Am J Clin Pathol. 2014;141(2):256-261. 76. Habibi A, Mekontso-Dessap A, Guillaud C, et al. Delayed hemolytic transfusion reaction in adult sickle-cell disease: presentations, outcomes, and treatments of 99 referral center episodes. Am J Hematol. 2016;91 (10):989-994. 77. Chadebech P, Habibi A, Nzouakou R, et al. Delayed hemolytic transfusion reaction in sickle cell disease patients: evidence of an emerging syndrome with suicidal red blood cell death. Transfusion. 2009;49(9):17851792. 78. Win N, Doughty H, Telfer P, Wild BJ, Pearson TC. Hyperhemolytic transfusion reaction in sickle cell disease. Transfusion. 2001;41(3):323-328. 79. Yasin Z, Witting S, Palascak MB, Joiner CH, Rucknagel DL, Franco RS. Phosphatidylserine externalization in sickle red blood cells: associations with cell age, density, and hemoglobin F. Blood. 2003;102(1):365-370. 80. Narbey D, Habibi A, Chadebech P, et al. Incidence and predictive score for delayed hemolytic transfusion reaction in adult patients with sickle cell disease. Am J Hematol. 2017;92(12):1340-1348. 81. Vidler JB, Gardner K, Amenyah K, Mijovic A, Thein SL. Delayed haemolytic transfusion reaction in adults with sickle cell disease: a 5-year experience. Br J Haematol. 2015;169(5):746-753. 82. de Montalembert M, Dumont MD, Heilbronner C, et al. Delayed hemolytic transfusion reaction in children with sickle cell disease. Haematologica. 2011;96(6):801807. 83. Pirenne F, Yazdanbakhsh K. How I safely transfuse patients with sickle-cell disease and manage delayed hemolytic transfusion reactions. Blood. 2018;131(25):2773-2781. 84. Merle NS, Grunenwald A, Rajaratnam H, et al. Intravascular hemolysis activates complement via cell-free heme and heme-loaded microvesicles. JCI Insight. 2018;3(12): e96910. 85. Chonat S, Quarmyne MO, Bennett CM, et al. Contribution of alternative complement pathway to delayed hemolytic transfusion reaction in sickle cell disease. Haematologica. 2018;103(10):e483-e485. 86. Floch A, Morel A, Zanchetta-Balint F, et al. Anti-C5 antibody treatment for delayed hemolytic transfusion reactions in sickle cell disease. Haematologica. 2020;105(11):26942697.

haematologica | 2021; 106(7)

Red cell transfusion and alloimmunization in SCD 87. Dumas G, Habibi A, Onimus T, et al. Eculizumab salvage therapy for delayed hemolysis transfusion reaction in sickle cell disease patients. Blood. 2016;127(8):10621064. 88. Chonat S, Graciaa S, Shin HS, et al. Eculizumab for complement mediated thrombotic microangiopathy in sickle cell disease. Haematologica. 2020;105(12):28872891. 89. Unnikrishnan A, Pelletier JPR, Bari S, et al. Anti-N and anti-Do(a) immunoglobulin G alloantibody-mediated delayed hemolytic transfusion reaction with hyperhemolysis in sickle cell disease treated with eculizumab and HBOC-201: case report and review of the literature. Transfusion. 2019;59(6):19071910. 90. Boonyasampant M, Weitz IC, Kay B, Boonchalermvichian C, Liebman HA, Shulman IA. Life-threatening delayed hyperhemolytic transfusion reaction in a patient with sickle cell disease: effective treatment with eculizumab followed by rituximab. Transfusion. 2015;55(10):2398-2403. 91. Mpinganzima C, Haaland A, Holm AGV, Thein SL, Tjønnfjord GE, Iversen PO. Two consecutive episodes of severe delayed hemolytic transfusion reaction in a sickle cell disease patient. Case Rep Hematol. 2020;2020:2765012. 92. Vlachaki E, Gavriilaki E, Kafantari K, et al. Successful outcome of hyperhemolysis in sickle cell disease following multiple lines of treatment: the role of complement inhibition. Hemoglobin. 2018;42(5-6):339-341. 93. Sivapalaratnam S, Linpower L, Sirigireddy B, et al. Treatment of post-transfusion hyperhaemolysis syndrome in sickle cell disease with the anti-IL6R humanised monoclonal antibody tocilizumab. Br J Haematol. 2019;186(6):e212-e214. 94. Lee LE, Beeler BW, Graham BC, Cap AP, Win N, Chen F. Posttransfusion hyperhemolysis is arrested by targeting macrophage

haematologica | 2021; 106(7)

activation with novel use of tocilizumab. Transfusion. 2020;60(1):30-35. 95. Hair PS, Heck TP, Carr DT, et al. Delayed hemolytic transfusion reaction in a patient with sickle cell disease and the role of the classical complement pathway: a case report. J Hematol. 2021;10(1):18-21. 96. Wang CY, Babitt JL. Liver iron sensing and body iron homeostasis. Blood. 2019;133(1): 18-29. 97. Vichinsky E, Butensky E, Fung E, et al. Comparison of organ dysfunction in transfused patients with SCD or beta thalassemia. Am J Hematol. 2005;80(1):70-74. 98. Meloni A, Puliyel M, Pepe A, Berdoukas V, Coates TD, Wood JC. Cardiac iron overload in sickle-cell disease. Am J Hematol. 2014;89 (7):678-683. 99. Perronne V, Roberts-Harewood M, Bachir D, et al. Patterns of mortality in sickle cell disease in adults in France and England. Hematol J. 2002;3(1):56-60. 100. Darbari DS, Kple-Faget P, Kwagyan J, Rana S, Gordeuk VR, Castro O. Circumstances of death in adult sickle cell disease patients. Am J Hematol. 2006;81(11):858-863. 101. Wood JC, Zhang P, Rienhoff H, Abi-Saab W, Neufeld EJ. Liver MRI is more precise than liver biopsy for assessing total body iron balance: a comparison of MRI relaxometry with simulated liver biopsy results. Magn Reson Imaging. 2015;33(6):761-767. 102. Coates TD, Wood JC. How we manage iron overload in sickle cell patients. Br J Haematol. 2017;177(5):703-716. 103. Maggio A, Kattamis A, Felisi M, et al. Evaluation of the efficacy and safety of deferiprone compared with deferasirox in paediatric patients with transfusion-dependent haemoglobinopathies (DEEP-2): a multicentre, randomised, open-label, non-inferiority, phase 3 trial. Lancet Haematol. 2020;7(6):e469-e478. 104. Sridharan K, Sivaramakrishnan G. Efficacy and safety of iron chelators in thalassemia

and sickle cell disease: a multiple treatment comparison network meta-analysis and trial sequential analysis. Expert Rev Clin Pharmacol. 2018;11(6):641-650. 105. Fasano RM, Leong T, Kaushal M, Sagiv E, Luban NL, Meier ER. Effectiveness of red blood cell exchange, partial manual exchange, and simple transfusion concurrently with iron chelation therapy in reducing iron overload in chronically transfused sickle cell anemia patients. Transfusion. 2016;56(7):1707-1715. 106. Piel FB, Patil AP, Howes RE, et al. Global epidemiology of sickle haemoglobin in neonates: a contemporary geostatistical model-based map and population estimates. Lancet. 2013;381(9861):142-151. 107. 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. 108. World Health Organization. Global status report on blood safety and availability 2016. Geneva: World Health Organization. Accessed in April, 2017. 109. Dzik WS, Kyeyune D, Otekat G, et al. Transfusion medicine in sub-Saharan Africa: conference summary. Transfus Med Rev. 2015;29(3):195-204. 110. Waiswa MK, Moses A, Seremba E, Ddungu H, Hume HA. Acute transfusion reactions at a national referral hospital in Uganda: a prospective study. Transfusion. 2014;54(11): 2804-2810. 111. Jayaraman S, Chalabi Z, Perel P, Guerriero C, Roberts I. The risk of transfusion-transmitted infections in sub-Saharan Africa. Transfusion. 2010;50(2):433-442. 112. Boateng LA, Ngoma AM, Bates I, Schonewille H. Red blood cell alloimmunization in transfused patients with sickle cell disease in sub-Saharan Africa; a systematic review and meta-analysis. Transfus Med Rev. 2019;33(3):162-169.

1815

ARTICLE Ferrata Storti Foundation

Acute Lymphoblastic Leukemia

WEE1 inhibition induces glutamine addiction in T-cell acute lymphoblastic leukemia Juncheng Hu,1,2,* Tianci Wang,1,2,* Jin Xu,2 Sanyun Wu,1 Liyuan Wang,2 Hexiu Su,2 Jue Jiang,2 Ming Yue,3 Jingchao Wang,2 Donghai Wang,2 Peng Li,4 Fuling Zhou,1 Yu Liu,5 Guoliang Qing2 and Hudan Liu1,2

Department of Hematology, Zhongnan Hospital, Wuhan University, Wuhan; 2Frontier Science Center for Immunology and Metabolism, Medical Research Institute, Wuhan University, Wuhan; 3Department of Pharmacy, The Central Hospital of Wuhan, Tongji Medical College, Huazhong University of Science and Technology, Wuhan; 4South China Institute for Stem Cell Biology and Regenerative Medicine, Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences, Guangzhou and 5Pediatric Translational Medicine Institute, Shanghai Children’s Medical Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China 1

Haematologica 2021 Volume 106(7):1816-1827

*JH and TW contributed equally as co-first authors.

ABSTRACT

Correspondence: HUDAN LIU hudanliu@whu.edu.cn Received: July 2, 2019. Accepted: January 2, 2020. Pre-published: January 9, 2020. https://doi.org/10.3324/haematol.2019.231126

1816

cell acute lymphoblastic leukemias (T-ALL) are aggressive and heterogeneous hematologic tumors resulting from the malignant transformation of T-cell progenitors. The major challenges in the treatment of T-ALL are dose-limiting toxicities of chemotherapeutics and drug resistance. Despite important progress in deciphering the genomic landscape of T-ALL, translation of these findings into effective targeted therapies remains largely unsuccessful. New targeted agents with significant antileukemic efficacy and less toxicity are urgently needed. Here we report that the expression of WEE1, a nuclear tyrosine kinase involved in cell cycle G2-M checkpoint signaling, is significantly elevated in T-ALL. Mechanistically, oncogenic MYC directly binds to the WEE1 promoter and activates its transcription. T-ALL cells particularly rely on the elevated WEE1 for cell viability. Pharmacological inhibition of WEE1 elicits global metabolic reprogramming which results in a marked suppression of aerobic glycolysis in T-ALL cells, leading to an increased dependency on glutaminolysis for cell survival. As such, dual targeting of WEE1 and glutaminase (GLS1) induces synergistic lethality in multiple TALL cell lines and shows great efficacy in T-ALL patient-derived xenografts. These findings provide mechanistic insights into the regulation of WEE1 kinase in T-ALL and suggest an additional vulnerability during WEE1 inhibitor treatments. We also highlight a promising combination strategy of dual inhibition of cell cycle kinase and metabolic enzymes for T-ALL therapeutics.

Introduction T-cell acute lymphoblastic leukemias (T-ALL) are highly proliferative hematologic tumors,1 which represent 10-15% of pediatric and 25% of adult acute lymphoblastic leukemia cases.2 Introduction of intensified chemotherapy protocols in T-ALL treatment significantly improves the overall survival in pediatric patients.3 Despite this progress, chemotherapeutic treatments come with significant shortterm and long-term side effects4 and the prognosis of patients with resistant or relapsed diseases remains dismal.5 Moreover, the remarkable success of pediatric treatment has not been achieved in adult patients as they do not always tolerate intensive pediatric regimens.6 Identification of activating mutations in NOTCH1 in over 50% of T-ALL cases has stimulated much interest in the development of antiNOTCH1 therapies. However, the clinical development of γ-secretase inhibitors (GSI), which block a critical proteolytic step required for NOTCH1 activation, has been hampered by limited efficacies in human patients and significant gastrointestinal toxicity.7 Facing these clinical challenges, new tarhaematologica | 2021; 106(7)

Synergistic targeting of WEE1 and GLS1 in T-ALL

geted therapies are needed to improve the outcomes of those patients with a poor prognosis and reduce the side effects associated with chemotherapies. Uncontrolled proliferation is one of the hallmarks of cancer. Many cancer cells possess a deficient G1 cell cycle checkpoint, for example due to p53 loss, and this impairs the ability of cells to halt the cell cycle and repair DNA damage before replication (S-phase).8 It provides cancer cells with a means to accumulate mutations and propagate irregularities that are favorable for proliferation. Meanwhile, tumor cells become more reliant on the G2-M cell cycle checkpoint to prevent mitotic entry with excessive DNA damage, which may lead to apoptosis due to mitotic catastrophe. WEE1 is a tyrosine kinase that plays a crucial role as the gatekeeper of the G2-M checkpoint.9 When DNA damage occurs, it leads to activation of WEE1 which phosphorylates CDK1 and maintains the CDK1cyclin B complex in an inactive form, preventing entry into mitosis.10 To limit excessive genomic instability in tumor cells, it is not surprising that WEE1 is highly expressed in a variety of cancer types.11-13 Moreover, high WEE1 expression has been associated with poor rates of disease-free survival.11,14,15 Despite considerable studies on the role of WEE1 in cell cycle checkpoints, it remains unclear how expression of WEE1 is increased and how increased WEE1 expression promotes neoplastic phenotypes. Cellular metabolism is at the foundation of all biological activities, and altered tumor cell metabolism is now firmly established as another hallmark of human cancer. Normal cells primarily rely on aerobic respiration/oxidative phosphorylation to meet their energy requirements, yet fastgrowing, poorly differentiated tumor cells typically exhibit increased aerobic glycolysis by converting a majority of glucose-derived pyruvate to lactate.16 Because of this, tumor cells depend on glutamine anaplerosis to replenish tricarboxylic acid (TCA) cycle intermediates (e.g., a-ketoglutarate) to sustain the metabolic integrity and produce nicotinammide adenina dinucleotide (NADH).17 Reprogramming of glucose and glutamine metabolism not only provides tumor cells with building blocks for macromolecule biosynthesis but also rescues them from a stressed cellular microenvironment by maintaining proper redox homeostasis.16 In T-ALL, both glycolysis and glutaminolysis play crucial roles in mediating leukemia cell proliferation, survival and drug resistance.18,19 In this study, we show that the elevated expression of WEE1 kinase in T-ALL results from oncogenic MYC-mediated transcriptional activation, and this WEE1 upregulation significantly contributes to efficient aerobic glycolysis. Pharmacological inhibition of WEE1 leads to a marked decrease in glycolytic flux, rendering T-ALL cells particularly vulnerable to glutamine deficiency. Based on these findings, dual targeting of WEE1 and glutaminase (GLS1), the key rate-limiting enzyme in the glutaminolysis pathway,20 shows great promise in anti-T-ALL targeted therapies.

informed consent from Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences and Zhongnan Hospital, Wuhan University, China. Polymerase chain reaction (PCR) primer sequences and antibodies used in this study are listed in Online Supplementary Tables S1 and S2, respectively.

Metabolomic analysis HPB-ALL cells were treated with mock (DMSO) or selective WEE1 inhibitor MK177523 (200 nM) for 20 hours (h). Ten million cells in each treatment were collected and quenched in liquid nitrogen. Metabolite samples were prepared for analysis using standard solvent extraction methods and then subjected to the gas chromatograph system (Agilent Technologies, Santa Clara, CA, USA) coupled with a Pegasus HT gas chromatography timeof-flight mass spectrometer (GC-TOF-MS; LECO Corporation, St Joseph, MI, USA).24,25 Identification of chemical entities was based on comparison to Fiehn metabolomics library. Chroma TOF 4.3x software and the LECO-Fiehn Rtx5 database were used for raw peak identification and integration of the peak area. Both mass spectrum match and retention index match were taken into consideration.26 Normalized data were uploaded using the SIMCA software package (V14.1, Sartorius Stedim Data Analytics AB, Umea, Sweden) for principal component analysis (PCA) and orthogonal projections for latent structuresdiscriminant analysis (OPLS-DA). Differential metabolites between experimental groups were determined by variable importance in the projection (VIP) values (VIP>1) and Student ttest. The metabolic pathway enrichment analysis was performed using Kyoto Encyclopedia of Genes and Genomes (KEGG, http://www.genome.jp/kegg/) and MetaboAnalyst 4.0 (http://www.metaboanalyst.ca).

Mouse studies HPB-ALL xenografts were carried out as previously described.21,27 NOD-Prkdcscid IL2Rγnull NPG mice (4-6 weeks old, Beijing Vitalstar Biotechnology Co., Beijing, China) were injected with five million cells infected with lentiviruses expressing the green fluorescent protein (GFP) and luciferase (pWPXLdLuciferase-GFP). Mice were subjected to bioimaging at day 6 post engraftment with IVIS Lumina II (Waltham, MA, USA) to ensure equivalent tumor onset in vivo. These animals were then randomly divided into four groups undergoing treatments in a three days on and three days off mode for four cycles. MK1775 (20 mg/kg) was administered twice daily by oral gavage and BPTES (25 mg/kg) was intraperitoneally injected once daily. Disease progression and therapy response were evaluated by bioimaging. For drug synergy studies in the patient-derived xenograft (from primary T-ALL sample #1, Online Supplementary Table S3), T-ALL cells were injected into irradiated 4-6 week old NPG mice (2 Gray), which were subjected to treatment at day 6 post engraftment. Control, MK1775 (20 mg/kg), CB-839 (200 mg/kg) or both were administrated by oral gavage every other day for two consecutive weeks. Leukemia burden was assessed by flow cytometry analysis of human CD45+ cells. All animal experiments were performed under animal ethical regulations and the study protocol was approved by the Institutional Animal Care and Use Committee of Wuhan University.

Statistical analysis Methods Cell culture and reagents T-ALL cells were maintained in RPMI-1640 (Hyclone, Logan, UT, USA) supplemented with 10% FBS (Hyclone) as described.21,22 Human primary specimens were obtained with haematologica | 2021; 106(7)

Spearman rank correlation test was used to analyze the WEE1 and MYC expression in primary T-ALL samples (Figure 2 F-H). Log-rank analysis was used to evaluate differences in KaplanMeier survival curves. Student t-test or one-way ANOVA was used in other statistical analysis. P<0.05 was considered statistically significant. 1817

J. Hu et al.

Results WEE1 is highly expressed in T-cell acute lymphoblastic leukemia To determine the expression of WEE1 in T-ALL, we analyzed gene expression profiles from three independent T-ALL patient cohorts.28-30 WEE1 expression in primary TALL was significantly elevated as compared to normal bone marrow (BM) cells (Figure 1A). Analysis of the Cancer Cell Line Encyclopedia (CCLE)31 demonstrated that WEE1 is the third highest expressed in T-ALL among 1,429 human cancer cell lines derived from 40 tumor origins (Online Supplementary Figure S1), suggesting the particular-

ly important role of WEE1 in T-ALL. Quantitative PCR analysis verified these findings in ten T-ALL patient samples and eight T-ALL cell lines that WEE1 mRNAs were significantly increased in T-ALL compared with normal BM (Figure 1B and Online Supplementary Table S3). To further correlate the level of WEE1 transcript with its protein expression, we next compared the WEE1 protein expression between normal and transformed thymocytes. Again, WEE1 was generally more abundant in T-ALL cells than normal thymocytes from healthy donors (Figure 1C). Related to our finding, increased WEE1 expression had previously been found in 58 adult T/B-ALL samples as compared to normal mononuclear cells.13 We next assessed

Figure 1. High WEE1 expression in T-cell acute lymphoblastic leukemia (T-ALL). (A) WEE1 mRNA expression was analyzed in primary T-ALL and normal bone marrow (BM) from microarray datasets (left, 6 normal BM and 11 T-ALL samples in GSE7186; middle, 4 normal BM and 46 T-ALL samples in GSE28497; right, 7 normal BM and 117 T-ALL samples in GSE26713). The distributions of WEE1 mRNA expression are presented as Log median-entered intensity and shown in Box-and-Whisker plots with the median value (line), the interquartile range (box), and up to 1.5x the interquartile range (bars). (B) Relative WEE1 mRNA expression in normal BM, primary T-ALL cells and T-ALL cell lines as indicated. (C) Immunoblots of WEE1 in normal human thymus, primary T-ALL cells and T-ALL cell lines as indicated. ACTIN serves as a loading control. (D) Analysis of WEE1 expression in 264 primary T-ALL samples that are categorized into eight different subtypes. y-axis denotes WEE1 FPKM values in Log scale from the RNA-sequencing (Seq) dataset. 2

1818

haematologica | 2021; 106(7)

Synergistic targeting of WEE1 and GLS1 in T-ALL

WEE1 expression in different T-ALL subtypes from 264 primary T-ALL samples, based on unique gene expression signatures reflecting distinct stages of arrest during T-cell development.32 Interestingly, WEE1 mRNA levels varied with the highest expression in the TLX1 subtype (Figure 1D). Regardless of these variations, we identified a global upregulation of WEE1 in T-ALL.

MYC directly activates WEE1 transcription in T-cell acute lymphoblastic leukemia To understand the molecular mechanism underlying WEE1 upregulation in T-ALL, we conducted in silico analysis in the UCSC genome browser gateway to identify potential transcription factor regulating WEE1 expression. MYC, TAL1 and GATA3 were predicted to activate the WEE1 promoter (Online Supplementary Figure S2A). We

individually knocked down each of these transcription factors, using two individual shRNA (sh#1 or sh#2) in human T-ALL cells, and found that only depletion of MYC decreased WEE1 mRNA and protein levels (Figure 2A and Online Supplementary Figure S2B). Consistently, bromodomain 4 (Brd4) inhibitor JQ1, which represses MYC transcription,33 decreased WEE1 mRNA and protein levels concomitant with downregulation of MYC expression in multiple T-ALL cells (Figure 2B and Online Supplementary Figure S2C). In contrast, JQ1 treatment, inhibiting N-MYC as well,34 caused minimal effect on the WEE1 steady-state level in T-ALL LOUCY cells which predominantly express N-MYC (Online Supplementary Figure S2D). In addition, inactivation of MYC similarly down-regulated WEE1 (expression in Burkitts lymphoma P493 cells (Online Supplementary Figure S2E). These data suggest that MYC,

Figure 2. MYC directly activates WEE1 transcription. (A) Jurkat and CUTLL1 cells were infected with lentiviruses expressing control (shCtrl) or MYC shRNA (shMYC#1 or shMYC#2). WEE1 mRNA and protein levels were determined by real-time-quantitative polymerase chain reaction (RT-qPCR) and immunoblots. ACTIN served as a loading control. (B) CCRF-CEM and KOPTK1 cells were treated with JQ1 as indicated, and WEE1 protein levels were determined by immunoblots. (C) Schematic presentation of MYC binding site (E-box, -39 ~ -34) on the WEE1 promoter. The potential MYC responsive element (wild-type, WT) and its mutant (MUT) are shown as indicated. (D) Binding of MYC to the WEE1 promoter was analyzed by chromatin immunoprecipitation (ChIP) in CUTLL1 cells. Averages of fold enrichment between MYC and isotype IgG are shown. NCL was analyzed as a positive control. (E) Luciferase reporter activities of the WEE1 promoter (-210−342bp) containing MYC RE-WT (and MUT) were detected in the presence of ectopically expressed MYC in 293T cells. 3xMYC E-box (3Ebox) sequences were used as a positive control. Reporter activities relative to empty pGL3-Basic vector (Vector) are shown. (F) Correlation of WEE1 expression with MYC in 117 primary T-cell acute lymphoblastic leukemia (T-ALL) samples (GSE26713). (G) Correlation of WEE1 expression with MYC in 264 primary T-ALL samples. (H) Correlation of WEE1 expression with MYC in NOTCH1 mutant or WT T-ALL cases shown in (G). Gene expression levels from primary T-ALL samples are presented in Log scale. Data of RT-qPCR and luciferase reporter analysis shown represent the means (± standard deviation) of biological triplicates. **P<0.01. 2

haematologica | 2021; 106(7)

1819

J. Hu et al. A

Figure 3. Figure legend on next page

1820

haematologica | 2021; 106(7)

Synergistic targeting of WEE1 and GLS1 in T-ALL

Figure 3 (from previous page). WEE1 inhibition compromises T-cell acute lymphoblastic leukemia (T-ALL) cell viability and aerobic glycolysis. (A) Cell viability analysis in multiple T-ALL cell lines subjected to MK1775 treatment at various concentrations (50, 100, 200, 400, 800 and 1600 nM) for 24 hours (h). Cell viability was determined using the CCK8 Cell Proliferation Assay Kit. Percentages of viable cells are shown relative to the untreated control. (B) CCRF-CEM and KOPTK1 cells were infected with lentiviruses expressing control (shCtrl) or WEE1 (sh#1 or sh#2). Live cells were counted at the indicated time points and cell growth was plotted as shown. (C) HPB-ALL cells were treated with MK1775, or infected with lentiviruses expressing WEE1 shRNA (sh#1 or sh#2) as indicated. Phosphorylated CDK1 (pCDK1), WEE1 and MYC protein levels were analyzed by immunoblots. As a WEE1 substrate, CDK1 phosphorylation was used to reflect WEE1 activity. ACTIN serves as a loading control. (D) Heatmap of down-regulated metabolites in MK1775 (200 nM)-treated HPB-ALL cells. Intracellular metabolites were prepared and analyzed by gas chromatography time-of-flight mass spectrometer (n=5). Identification of significantly different metabolites between experimental groups were determined by variable importance in projection (VIP) values (VIP > 1) and Student t-test (P<0.05). Colors indicate relative metabolite abundance. (E) Metabolite set enrichment analysis of significantly down-regulated metabolites (P<0.01) in MK1775-treated HPB-ALL cells. (F) CUTLL1 cells were treated with MK1775 (200 nM) or GSI Compound E (1 mM) for 24 h. Glucose consumption and lactate secretion were analyzed and normalized to the same live cell number. (G) HPB-ALL cells were treated with MK1775 for 24 h as indicated. Six representative genes involved in glycolysis pathway were analyzed by real-time quantitative polymerase chain reaction (RTqPCR) (left). (Right) Immunoblots of indicated proteins. (H) Primary cells from a T-ALL patient were injected into NPG mice which underwent control (Ctrl) or MK1775 (20 mg/kg) treatments as described in the Methods section. Human CD45+ cells from the spleens of control or MK1775-treated mice were purified for RT-qPCR and immnoblotting to assess glycolytic gene expression. Data shown represent the means (± standard deviation) of biological triplicates. *P<0.05, **P<0.01. Editor's note. This figure is slightly different from the one pre-published as early view because the authors noticed an error in panel C and asked to replace it with the one shown above.

but not N-MYC, activates WEE1 expression in T-ALL, and most likely in other tumor contexts as well. Previous chromatin immunoprecipitation sequencing (ChIP)-Seq analysis manifested a strong binding signal of MYC, but not TAL1 or GATA3, in the WEE1 locus (Online Supplementary Figure S2F),35,36 further supporting the concept that MYC directly activates WEE1 transcription. To validate these data, we performed conventional ChIP assays and revealed a significant increase in MYC recruitment to the potential binding site on WEE1, similar to a well-characterized MYC target NCL (Figure 2C and D). We next constructed the WEE1 promoter containing the MYC responsive element (RE, E-box) into a luciferase reporter vector. Similar to the triple MYC RE (3 E-box) reporter as a positive control, the WT WEE1 reporter was strongly activated by MYC whereas the mutant RE with disrupted E-box was only slightly induced (Figure 2C and E). The weak activation on the mutant RE was probably due to additional unrevealed MYC binding sites or other transcription factors contributing to WEE1 luciferase reporter expression. In further support of this, gene expression profiling of 117 primary T-ALL30 revealed a strong and significant correlation between WEE1 and MYC mRNA levels (R=0.335, P<0.001) (Figure 2F). MYC showed a significant correlation with WEE1 (R=0.187, P=0.002) in another expression profile dataset with 264 primary T-ALL (Figure 2G).32 Interestingly, MYC was more strongly and significantly correlated with WEE1 in NOTCH1 mutant samples (R=0.224, P=0.001) than NOTCH1 WT specimens (R=0.038, P=0.754) (Figure 2H). These data suggest that MYC preferentially regulates WEE1 in NOTCH1 mutant cases, and MYC probably activates differential gene expression in NOTCH1 WT and mutant T-ALL. Consistent with this, MYC depletion barely changed WEE1 expression in NOTCH1 WT SUP-T1 cells (Online Supplementary Figure S3). Our data thus provide strong evidence demonstrating that oncogenic MYC specifically binds to WEE1, particularly in NOTCH1 mutant T-ALL cells, for direct transcriptional activation.

Inhibition of WEE1 impairs aerobic glycolysis Elevated WEE1 expression driven by MYC poses the possibility that T-ALL cells may be particularly dependent on WEE1 for cell proliferation and survival. Indeed, a selective WEE1 inhibitor MK1775 as a single agent reduced cell viability in a dose-dependent manner in seven T-ALL cell lines, whereas the effect on normal BM cells was minimal (Figure 3A). Similarly, WEE1 depletion haematologica | 2021; 106(7)

significantly suppressed KOPTK1 and CCRF-CEM cell growth (Figure 3B). In support of our findings, MK1775 was shown to elicit ALL cell apoptosis primarily due to disruption of the G2-M cell cycle checkpoint and increased DNA damage,13 which were also detected in CUTLL1 cells (Online Supplementary Figure S4). Of particular interest, we found that WEE1 depletion or inhibition reduced MYC steady-state levels (Figure 3C and Online Supplementary Figure S5A and B). Given that MYC is a master regulator in controlling cancer cell metabolism in the majority of tumor contexts,37 we reasoned that WEE1 inhibition would lead to metabolic change in T-ALL cells through MYC downregulation. To test this hypothesis, HPB-ALL cells treated with or without MK1775 were subjected to metabolomic analysis. Relative to the mock treatment, MK1775 induced global metabolic changes. Notably, WEE1 inhibition caused a decreased production of fructose-6-phosphate, 3-phosphoglycerate and lactic acid, which are crucial metabolic intermediates or products involved in aerobic glycolysis (Figure 3D). We further conducted metabolite set enrichment analysis of all down-regulated molecules and revealed that the Warburg effect was the top hit affected by WEE1 inhibition (Figure 3E). Consistent with these data, MK1775 significantly inhibited glucose uptake concomitant with a decrease in lactic acid secretion in multiple T-ALL cells, similar to GSI compound E which was previously shown to repress glycolysis in T-ALL (Figure 3F and Online Supplementary Figure S5C).18 WEE1 depletion yielded similar results to the MK1775 treatment (Online Supplementary Figure S5D). Importantly, addition of membrane soluble pyruvate, commonly encountered as one of the end products of glycolysis, significantly rescued MK1775-induced cell death (Online Supplementary Figure S5E), suggesting that the metabolic effect is one of the primary mechanisms underlying MK1775-mediated antileukemic activity. This metabolic change is presumably due to altered MYC expression resulting from WEE1 inhibition, as enforced MYC expression significantly rescued glycolysis defects induced by MK1775 (Online Supplementary Figure S6A). Most likely, MYC acts as an important downstream player mediating the role of WEE1 in regulation of glycolysis. Consistent with previous reports that MYC activates a panel of glycolytic gene expression,37 WEE1 inhibition led to downregulation of these genes in T-ALL cells (Figure 3G and Online Supplementary Figure S6B). We also assessed the in vivo effect of MK1775 on glycolysis in a T-ALL patient1821

J. Hu et al.

derived xenograft (PDX) derived from primary T-ALL sample #1 (Online Supplementary Table S3). Human CD45+ leukemia cells from the spleen were analyzed for glycolysis-related gene expression. Again, MK1775 suppressed the expression of several key enzymes involved in the glycolysis pathway (Figure 3H). Taken together, we identify a crucial role of WEE1 in regulation of glucose metabolism, in addition to its well-defined function in DNA damage response and cell cycle checkpoint.10 In

this regard, multiple mechanisms account for the antitumor efficacy of WEE1 inhibitors.

WEE1 inactivation sensitizes T-cell acute lymphoblastic leukemia cells to glutaminase inhibitors Glucose and glutamine are two primary nutrients utilized by cancer cells for their proliferation and growth. We then surmised that a decrease in glycolysis may render TALL cells more addicted to glutaminolysis and glutamine

Figure 4. WEE1 inhibition sensitizes T-cell acute lymphoblastic leukemia cells to glutaminolysis inhibition. (A) T-ALL cells were treated with MK1775 (100 nM) in the presence or absence of Glutamine (Gln, 2 mM) for 48 hours (h). Apoptotic cell death was analyzed by Annexin V/PI staining and flow cytometry analysis. (Left) Representative flow cytometry graphs of HPB-ALL cells. (Right) Quantifications of cell death from four T-ALL lines are presented on the right. (B) Immunofluorescence images of cleaved Caspase-3 (c-caspase 3, red) and DAPI (blue) in HPB-ALL cells undergoing DMSO, MK1775(100 nM), CB-839 (100 nM) or dual treatments for 48 h. Scale bar, 50 μm. (Right) Quantifications of fluorescence signals. (C) Apoptotic cell death was analyzed by Annexin V/PI staining in normal bone marrow (BM) and T-ALL cell lines after treatments as in (B). (D) Apoptotic cell death was analyzed in T-ALL cells after treatments with DMSO, MK1775 (100 nM), BPTES (10 mM) and combination for 48 h. Data shown represent the means (± standard deviation) of biological triplicates. **P<0.01.

1822

haematologica | 2021; 106(7)

Synergistic targeting of WEE1 and GLS1 in T-ALL

deficiency would synergize with WEE1 inhibition. To test this hypothesis, we treated several T-ALL cells with MK1775 in the presence or absence of glutamine in the culture medium. Glutamine deficiency induced robust cell death in combination with MK1775 treatment in multiple T-ALL cells (Figure 4A), suggesting that inhibition of glutamine uptake or metabolism exacerbates cell death induced by WEE1 inhibition. Glutaminolysis is a stepwise process by which imported glutamine is converted to glutamate (dependent on glutaminase activity) and subse-

quently transformed to a-ketoglutarate, a TCA cycle intermediate, for further catabolism and production of NADH.20 GLS1 is the glutaminase converting glutamine to glutamate, which is the first and rate-limiting step in the glutaminolysis pathway. We predicted that co-inhibition of WEE1 and GLS1 would disrupt the integrity of the TCA cycle, leading to metabolic catastrophe and subsequent cell death. Given that the selective GLS1 inhibitor CB-839 has been evaluated in clinical trials for anti-tumor activity,38,39 we therefore determined whether CB-839 elicited a

Figure 5. MK1775 and BPTES treatments demonstrate synergistic anti-leukemia efficacy in HPB-acute lymphoblastic leukemia (ALL) xenografts. (A) Graphical illustration of HPB-ALL luciferase T-cell ALL xenografts and treatment strategy (see Methods section). Single or dual treatments started at day 6 post engraftment and underwent a “three days on and one day off” schedule for four cycles as illustrated. When control mice became moribund around day 30 post engraftment, all mice were euthanized to assess leukemogenesis in vivo and therapeutic responses. (B) Representative images (left) and quantification (right) of tumor burden as assessed by luciferase luminescence signals in HPB-ALL xenografts (n=5 per group). Combo: combination. (C) GFP+ cells from the spleen and bone marrow were analyzed by flow cytometry (left). (Right) Data from five individual mice. (D) Representative spleen and bone images are shown (top) with spleen weights (bottom). **P<0.01.

haematologica | 2021; 106(7)

1823

J. Hu et al.

synergistic effect with MK1775. Immunofluorescence analysis and immunoblots of cleaved Caspase-3 in HPBALL and CUTLL1 cells manifested robust apoptosis upon combined treatments as compared to single treatment (Figure 4B and Online Supplementary Figure S7). Annexin V staining confirmed strong synergism by dual treatments in six T-ALL cell lines, while sparing normal BM cells (Figure 4C). Consistently, another GLS1 inhibitor, bis-2-(5-phenylacetamido-1,3,4-thiadazol-2-yl) ethyl sulfide (BPTES), elicited synergistic cell death in combination with MK1775 (Figure 4D). Therefore, these findings indicate the strong anti-T-ALL efficacy by co-inhibition of WEE1 and GLS1.

Targeting of WEE1 and glutaminase suppresses HPB-ALL xenografts To evaluate the in vivo efficacy of the combination strategy, we established a human xenograft using HPB-ALL cells. In order to visualize leukemia cell expansion in vivo,

lentiviruses expressing firefly luciferase were infected into HPB-ALL cells and five million luciferase-expressing cells were intravenously injected into each NPG mouse. The intensities of luciferase luminescence signals were equivalent in all mice at day 6 post engraftment, and these animals were then randomly divided into four groups subjected to control, MK1775, BPTES or combination treatment (Figure 5A and B). Whereas single administration showed a moderate inhibition of leukemia progression, MK1775/BPTES combination elicited a much stronger anti-leukemia effect, as evidenced by bioimaging signals at day 13 and day 20 post xenograft (Figure 5B). When the control cohort became moribund, all the mice were sacrificed to assess the leukemia burden in vivo. Administration of either MK1775 or BPTES alone moderately inhibited engraftment, and MK1775/BPTES combination induced a synergistic and remarkable suppression of leukemogenesis (Figure 5C). As a result, dual treatment ameliorated spleen enlargement and resulted in more reddish bones (Figure

Figure 6. Synergistic anti-leukemia effects of MK1775 and CB-839 on a patient-derived T-cell acute lymphoblastic leukemia (T-ALL) xenograft. Human primary T-ALL cells were injected into irradiated NPG mice. (A) Human CD45+ cells from bone marrow (BM) and spleen were analyzed by flow cytometry when the controltreated mice became moribund around day 25 post xenograft. (Right) Quantifications of CD45+ percentages. Combo: combination. (B) Representative immunohistological images of PCNA and cleaved Caspase-3 (c-Caspase 3) in the spleen sections from mice receiving different treatments. Scale bar, 50 mm. (A and B) Quantifications shown represent the means (± standard deviation) of biological triplicates. (C) Representative spleen and bone images of mice subjected to different treatments (left) with spleen weights (right). (D) Kaplan-Meier survival curves of T-ALL PDX treated with MK1775 and/or CB-839 (n=5 in each group). Dotted lines define the treatment time frame. **P<0.01.

1824

haematologica | 2021; 106(7)

Synergistic targeting of WEE1 and GLS1 in T-ALL

5D). These findings corroborate the anti-T-ALL efficacy of MK1775/BPTES combination in vivo. We also evaluated the toxicity of the drug combination in healthy C57/BL6 mice using the same treatment strategy as shown in Figure 5A. Body weights of these mice were comparable to the control cohort (Online Supplementary Figure S8A). Analysis of blood parameters at the end of the treatment revealed a mild decrease in red and white cell counts (Online Supplementary Figure S8B), and the decline in blood cell counts vanished two weeks after treatment ended (Online Supplementary Figure S8C). Moreover, hematoxylin & eosin (H&E) staining revealed undetectable damage in various organs (Online Supplementary Figure S8D). Consistently, healthy human BM cells were not sensitive to dual treatment either (Figure 4C), arguing of manageable toxicity of this combination strategy. This is probably due to lower WEE1 and MYC expression in the majority of normal cells. Therefore, therapeutic targeting of WEE1 may not induce significant inhibition of MYC and glycolysis as much as it does in tumor cells.

Combined WEE1/GLS1 inhibition prolongs overall survival in a T-cell acute lymphoblastic leukemia patient-derived xenograft Given CB-839 is currently being evaluated in clinical trials,39 we next explored the translational potential of MK1775 in combination with CB-839 in a T-ALL PDX. Human CD45+ leukemia cell distributions in vivo were analyzed to assess tumor burden and therapeutic responses. Again, the combination treatment induced synergistic tumor growth inhibition compared with the control or monotherapy, as evidenced by much lower human CD45+ tumor cell percentages (Figure 6A). Immunohistological (IHC) analysis of the spleen sections further confirmed reduced cell proliferation and massive intratumoral apoptosis due to administration of both compounds, as quantified by attenuated proliferating cell nuclear antigen (PCNA) and increased cleaved Caspase-3 staining, respectively (Figure 6B). Injection of MK1775 and CB-839 resulted in much smaller spleen size and more reddish bones (Figure 6C). More importantly, dual treatments significantly prolonged the lifespans of the T-ALL PDX as compared to single treatment (Figure 6D). Together with the preclinical studies in HPB-ALL xenografts, these results strongly suggest the clinical potential of WEE1/GLS1 inhibitors as a promising T-ALL targeted therapeutics.

Discussion Intensified T-ALL chemotherapies face the challenges of significant side effects, frequent relapses, and drug resistance. To improve the treatment of T-ALL and reduce associated toxicity, introduction of new targeted agents is desperately needed. We here show that WEE1 kinase is a promising therapeutic target. Reduced cell viability upon WEE1 inhibitor MK1775 treatment is partly attributable to significant suppression of aerobic glycolysis, leading to TALL cells more addicted to glutaminolysis. Administration of WEE1 and GLS1 inhibitors induces synergistic lethality in T-ALL cells and leukemia xenografts. These results highlight a promising therapeutic strategy of dual targeting of cell cycle kinase and metabolic enzymes in T-ALL treatments. haematologica | 2021; 106(7)

Loss of cell cycle control plays a prominent role in the pathogenesis of T-ALL. The tumor suppressors p16INK4A and p14ARF encoded by the CDKN2A locus are frequently lost in T-ALL due to chromosomal deletions.40 p16INK4A and p14ARF inactivates the cyclin D1-CDK4/6 and cyclin ECDK2 complexes, respectively, leading to G1-S arrest for DNA repair.9 As such, loss of CDKN2A results in overactivation of these CDK complexes, enabling T-ALL cells to enter S phase for replication despite DNA damage. T-ALL cells, with a deficient G1-S checkpoint, are therefore more reliant on the G2-M cell cycle checkpoint to prevent excessive DNA damage that may lead to mitotic catastrophe and cell death. In support of this notion, we and others have shown that WEE1 expression is significantly increased in a variety of T-ALL cell lines and patientderived primary cells.13 We further delineate the molecular mechanism underlying WEE1 expression in T-ALL by identifying MYC as a prominent regulator directly activating the WEE1 transcription. MYC-mediated WEE1 upregulation is also found in Burkitt’s lymphoma cells, suggesting the MYC-WEE1 axis as a general regulatory mode governing WEE1 expression in human cancers. Inhibition of WEE1 in cancer cells circumvents cell cycle arrest during the G2 phase and enables cell division despite accumulation of DNA damage. T-ALL cells crucially depend on the G2 checkpoint in the presence of DNAdamage inducing drugs. WEE1 inhibitor manifested synergistic anti-leukemic activity in combination with cytarabine or olaparib.13,41,42 We here identify an additional metabolic vulnerability of T-ALL cells in response to WEE1 inhibition after which they become particularly addicted to glutaminolysis for cell survival. The underlying mechanism involves MK1775-mediated glycolytic suppression at least in part via downregulation of MYC, a master regulator in controlling glucose metabolism in the majority of tumor contexts.37 Indeed, we demonstrate that overexpression of MYC significantly rescued glycolysis defect due to MK1775 treatment, suggesting that MYC acts as an important downstream player mediating the role of WEE1 in the regulation of glycolysis. It is interesting to note that although a previous report suggests that MYC promotes glutaminolysis as well by activating GLS1 expression in B lymphoma and prostate cancer cells,43 neither MYC downregulation nor WEE1 inhibition affected GLS1 expression in the context of T-ALL (data not shown). Moreover, WEE1 inhibition switches T-ALL cells to a more glutamine-dependent state such that simultaneous suppression of glycolysis and glutaminolysis by MK1775 and GLS1 inhibitor respectively induced potent synergistic anti-T-ALL effects (Online Supplementary Figure S9). Our findings reveal a molecular link between cell cycle and cancer metabolism by demonstrating the contribution of WEE1 to glycolysis. WEE1 inhibition, which unleashes the G2-M checkpoint and accelerates cell cycle, reprograms the cellular metabolism such that tumor cells with decreased glycolysis become more addicted to glutaminolysis. These results are reminiscent of a recent finding that activation of cyclin D1-CDK4/6 by dysregulation of Fbxo4-cyclin D1 axis leads to cellular dependency on glutamine metabolism and sensitizes tumor cells to CB-839.44 As such, increased glutamine dependency could be a consequence of over-activated CDK complex, which provides a potential therapeutic opportunity in fast-dividing tumor cells. Loss of cell cycle checkpoints due to genetic mutations and/or utilizing checkpoint inhibitors in cancer treat1825

J. Hu et al.

ment may sensitize tumor cells to GLS1 inhibitors. Taken together, our findings support further investigation of MK1775 and CB-839 combination in clinical settings for T-ALL treatments, given that the respective monotherapy has been evaluated in multiple clinical trials and shows tolerable toxicity.39,45,46 Disclosures No conflicts of interest to disclose. Contributions HL conceived and designed the study; HL supervised the study. HL, JH and TW wrote the manuscript; JH and TW performed the majority of experiments; JX, LW, HS, JJ, MY, JW and DW provided technical support; SW, PL and FZ provided primary T-ALL samples; YL helped with data analysis of primary T-ALL samples; HL and GQ analyzed and interpreted the data.

References 1. Belver L, Ferrando A. The genetics and mechanisms of T cell acute lymphoblastic leukaemia. Nat Rev Cancer. 2016;16(8):494507. 2. Van Vlierberghe P, Ferrando A. The molecular basis of T cell acute lymphoblastic leukemia. J Clin Invest. 2012;122(10):33983406. 3. Hunger SP, Lu X, Devidas M, et al. Improved survival for children and adolescents with acute lymphoblastic leukemia between 1990 and 2005: a report from the Children's Oncology Group. J Clin Oncol. 2012; 30(14):1663-1669. 4. Inaba H, Greaves M, Mullighan CG. Acute lymphoblastic leukaemia. Lancet. 2013; 381(9881):1943-1955. 5. Bhojwani D, Pui C-H. Relapsed childhood acute lymphoblastic leukaemia. Lancet Oncol. 2013;14(6):e205-e217. 6. Litzow MR, Ferrando AA. How I treat T-cell acute lymphoblastic leukemia in adults. Blood. 2015;126(7):833-841. 7. Palomero T, Ferrando A. Therapeutic targeting of NOTCH1 signaling in T-cell acute lymphoblastic leukemia. Clin Lymphoma Myeloma. 2009;9(Suppl 3):S205-S210. 8. Chen T, Stephens PA, Middleton FK, Curtin NJ. Targeting the S and G2 checkpoint to treat cancer. Drug Discov Today. 2012;17(56):194-202. 9. Otto T, Sicinski P. Cell cycle proteins as promising targets in cancer therapy. Nat Rev Cancer. 2017;17(2):93-115. 10. Matheson CJ, Backos DS, Reigan P. Targeting WEE1 kinase in cancer. Trends Pharmacol Sci. 2016;37(10):872-881. 11. Magnussen GI, Holm R, Emilsen E, Rosnes AK, Slipicevic A, Florenes VA. High expression of Wee1 is associated with poor disease-free survival in malignant melanoma: potential for targeted therapy. PLoS One. 2012;7(6):e38254. 12. Mir SE, De Witt Hamer PC, Krawczyk PM, et al. In silico analysis of kinase expression identifies WEE1 as a gatekeeper against mitotic catastrophe in glioblastoma. Cancer Cell. 2010;18(3):244-257. 13. Ghelli Luserna Di Rora A, Beeharry N, Imbrogno E, et al. Targeting WEE1 to enhance conventional therapies for acute lymphoblastic leukemia. J Hematol Oncol. 2018;11(1):99. 14. Slipicevic A, Holth A, Hellesylt E, Trope CG,

1826

Acknowledgments The authors would like to thank Liu Lab members for technical support and critical reading of the manuscript, the Core Facility of Medical Research Institute at Wuhan University for immunofluorescence, flow cytometry and histological analysis. Funding This research was supported by grants from National Key R&D Program of China (2017YFA0505600 to GQ), National Natural Science Foundation of China (81770177, 81970152 to HL, 81803003 to MY), National Science Fund for Distinguished Young Scholars (81725013 to GQ), Hubei Provincial Natural Science Fund for Distinguished Young Scholars (2017CFA072 to HL), Hubei Provincial Natural Science Fund for Creative Research Groups (2018CFA018 to GQ), and Innovative Research Grants from Wuhan University (2042019kf0338 to HL and 2042017kf0282 to GQ).

Davidson B, Florenes VA. Wee1 is a novel independent prognostic marker of poor survival in post-chemotherapy ovarian carcinoma effusions. Gynecol Oncol. 2014; 135(1):118-124. 15. Magnussen GI, Hellesylt E, Nesland JM, Trope CG, Florenes VA, Holm R. High expression of wee1 is associated with malignancy in vulvar squamous cell carcinoma patients. BMC Cancer. 2013;13:288. 16. Pavlova NN, Thompson CB. The emerging hallmarks of cancer metabolism. Cell Metab. 2016;23(1):27-47. 17. Zhang J, Pavlova NN, Thompson CB. Cancer cell metabolism: the essential role of the nonessential amino acid, glutamine. EMBO J. 2017;36(10):1302-1315. 18. Palomero T, Sulis ML, Cortina M, et al. Mutational loss of PTEN induces resistance to NOTCH1 inhibition in T-cell leukemia. Nat Med. 2007;13(10):1203-1210. 19. Herranz D, Ambesi-Impiombato A, Sudderth J, et al. Metabolic reprogramming induces resistance to anti-NOTCH1 therapies in T cell acute lymphoblastic leukemia. Nat Med. 2015;21(10):1182-1189. 20. Hensley CT, Wasti AT, DeBerardinis RJ. Glutamine and cancer: cell biology, physiology, and clinical opportunities. J Clin Invest. 2013;123(9):3678-3684. 21. Wang Z, Hu Y, Xiao D, et al. Stabilization of Notch1 by the Hsp90 chaperone is crucial for T-cell leukemogenesis. Clin Cancer Res. 2017;23(14):3834-3846. 22. Su H, Hu J, Huang L, et al. SHQ1 regulation of RNA splicing is required for T-lymphoblastic leukemia cell survival. Nat Commun. 2018;9(1):4281. 23. Hirai H, Iwasawa Y, Okada M, et al. Smallmolecule inhibition of Wee1 kinase by MK1775 selectively sensitizes p53-deficient tumor cells to DNA-damaging agents. Mol Cancer Ther. 2009;8(11):2992-3000. 24. Yang C, Hao R, Du X, Wang Q, Deng Y, Sun R. Response to different dietary carbohydrate and protein levels of pearl oysters (Pinctada fucata martensii) as revealed by GC-TOF/MS-based metabolomics. Sci Total Environ. 2019;650(Pt 2):2614-2623. 25. Yue M, Jiang J, Gao P, Liu H, Qing G. Oncogenic MYC activates a feedforward regulatory loop promoting essential amino acid metabolism and tumorigenesis. Cell Rep. 2017;21(13):3819-3832. 26. Kind T, Wohlgemuth G, Lee DY, et al. FiehnLib: mass spectral and retention index libraries for metabolomics based on quadru-

ple and time-of-flight gas chromatography/mass spectrometry. Anal Chem. 2009; 81(24):10038-10048. 27. Hu Y, Su H, Liu C, et al. DEPTOR is a direct NOTCH1 target that promotes cell proliferation and survival in T-cell leukemia. Oncogene. 2017;36(8):1038-1047. 28. Andersson A, Ritz C, Lindgren D, et al. Microarray-based classification of a consecutive series of 121 childhood acute leukemias: prediction of leukemic and genetic subtype as well as of minimal residual disease status. Leukemia. 2007; 21(6):1198-1203. 29. Coustan-Smith E, Song G, Clark C, et al. New markers for minimal residual disease detection in acute lymphoblastic leukemia. Blood. 2011;117(23):6267-6276. 30. Homminga I, Pieters R, Langerak AW, et al. Integrated transcript and genome analyses reveal NKX2-1 and MEF2C as potential oncogenes in T cell acute lymphoblastic leukemia. Cancer Cell. 2011;19(4):484-497. 31. Ghandi M, Huang FW, Jane-Valbuena J, et al. Next-generation characterization of the Cancer Cell Line Encyclopedia. Nature. 2019;569(7757):503-508. 32. Liu Y, Easton J, Shao Y, et al. The genomic landscape of pediatric and young adult T-lineage acute lymphoblastic leukemia. Nat Genet. 2017;49(8):1211-1218. 33. Delmore JE, Issa GC, Lemieux ME, et al. BET bromodomain inhibition as a therapeutic strategy to target c-Myc. Cell. 2011; 146(6):904-917. 34. Puissant A, Frumm SM, Alexe G, et al. Targeting MYCN in neuroblastoma by BET bromodomain inhibition. Cancer Discov. 2013;3(3):308-323. 35. Kourtis N, Lazaris C, Hockemeyer K, et al. Oncogenic hijacking of the stress response machinery in T cell acute lymphoblastic leukemia. Nat Med. 2018;24(8):1157-1166. 36. Sanda T, Lawton LN, Barrasa MI, et al. Core transcriptional regulatory circuit controlled by the TAL1 complex in human T cell acute lymphoblastic leukemia. Cancer Cell. 2012; 22(2):209-221. 37. Hsieh AL, Walton ZE, Altman BJ, Stine ZE, Dang CV. MYC and metabolism on the path to cancer. Semin Cell Dev Biol. 2015;43:1121. 38. Song M, Kim SH, Im CY, Hwang HJ. Recent development of small molecule glutaminase inhibitors. Curr Top Med Chem. 2018;18(6): 432-443. 39. DeMichele A, Harding JJ, Telli ML, et al.

haematologica | 2021; 106(7)

Synergistic targeting of WEE1 and GLS1 in T-ALL

Phase 1 study of CB-839, a small molecule inhibitor of glutaminase (GLS) in combination with paclitaxel (Pac) in patients (pts) with triple negative breast cancer (TNBC). J Clin Oncol. 2016;34(Suppl 15):1011-1011. 40. Girardi T, Vicente C, Cools J, De Keersmaecker K. The genetics and molecular biology of T-ALL. Blood. 2017;129(9): 1113-1123. 41. Ford JB, Baturin D, Burleson TM, Van Linden AA, Kim YM, Porter CC. AZD1775 sensitizes T cell acute lymphoblastic leukemia cells to cytarabine by promoting apoptosis over DNA repair. Oncotarget. 2015;6(29):28001-28010.

haematologica | 2021; 106(7)

42. Garcia TB, Snedeker JC, Baturin D, et al. A small-molecule inhibitor of WEE1, AZD1775, synergizes with olaparib by impairing homologous recombination and enhancing DNA damage and apoptosis in acute leukemia. Mol Cancer Ther. 2017; 16(10):2058-2068. 43. Gao P, Tchernyshyov I, Chang TC, et al. cMyc suppression of miR-23a/b enhances mitochondrial glutaminase expression and glutamine metabolism. Nature. 2009; 458(7239):762-765. 44. Qie S, Yoshida A, Parnham S, et al. Targeting glutamine-addiction and overcoming CDK4/6 inhibitor resistance in human

esophageal squamous cell carcinoma. Nat Commun. 2019;10(1):1296. 45. Leijen S, van Geel RM, Pavlick AC, et al. Phase I study evaluating WEE1 inhibitor AZD1775 as monotherapy and in combination with gemcitabine, cisplatin, or carboplatin in patients with advanced solid tumors. J Clin Oncol. 2016;34(36):43714380. 46. Do K, Wilsker D, Ji J, et al. Phase I study of single-agent AZD1775 (MK-1775), a Wee1 kinase inhibitor, in patients with refractory solid tumors. J Clin Oncol. 2015;33(30): 3409-3415.

1827

ARTICLE Ferrata Storti Foundation

Haematologica 2021 Volume 106(7):1828-1838

Acute Lymphoblastic Leukemia

A multicenter total therapy strategy for de novo adult Philadelphia chromosome positive acute lymphoblastic leukemia patients: final results of the GIMEMA LAL1509 protocol

Sabina Chiaretti,1 Michela Ansuinelli,1 Antonella Vitale,1 Loredana Elia,1 Mabel Matarazzo,1 Alfonso Piciocchi,2 Paola Fazi,2 Francesco Di Raimondo,3 Lidia Santoro,4 Francesco Fabbiano,5 Catello Califano,6 Giovanni Martinelli,7 Francesca Ronco,8 Felicetto Ferrara,9 Nicola Cascavilla,10 Catia Bigazzi,11 Alessandra Tedeschi,12 Simona Sica,13,14 Nicola Di Renzo,15 Angela Melpignano,16 Germana Beltrami,17 Marco Vignetti2 and Robin Foà1

Section of Hematology, Department of Translational and Precision Medicine, Sapienza University, Rome; 2GIMEMA Data Center, Fondazione GIMEMA, Rome; 3Section of Hematology, Department of General Surgery and Medical-Surgical Specialties, University of Catania, Catania; 4Struttura Complessa di Ematologia e Trapianto Emopoietico-A.O. S.G.Moscati, Avellino; 5Division of Hematology and Bone Marrow Transplantation, Ospedali Riuniti Villa Sofia-Cervello, Palermo; 6Onco-Ematologia Ospedale Pagani, Salerno; 7Seragnoli Institute of Hematology, Bologna University School of Medicine, Bologna; 8Operative Unit of Hematology, Grande Ospedale Metropolitano "Bianchi-Melacrino-Morelli", Reggio Calabria; 9Division of Hematology and Stem Cell Transplantation Program, AORN Cardarelli Hospital, Naples; 10Casa Sollievo della Sofferenza IRCCS, San Giovanni Rotondo; 11Department of Hematology and Stem Cell Transplantation Unit, C.G. Mazzoni Hospital, Ascoli Piceno; 12ASST Grande Ospedale Metropolitano Niguarda, Milan; 13Fondazione Policlinico Universitario A. Gemelli, Rome; 14 Università Cattolica del Sacro Cuore, Rome; 15Department of Hematology and Stem Cell Transplant, Presidio Ospedaliero Vito Fazzi, Lecce; 16A. Perrino Hospital, Brindisi and 17 Policlinico San Martino, Genoa, Italy 1

ABSTRACT

Correspondence: SABINA CHIARETTI chiaretti@bce.uniroma1.it ROBIN FOÀ rfoa@bce.uniroma1.it Received: May 29, 2020. Accepted: December 24, 2020. Pre-published: February 4, 2021. https://doi.org/10.3324/haematol.2020.260935

1828

he GIMEMA LAL1509 protocol, designed for adult (≥18-60 years) de novo Philadelphia chromosome positive (Ph+) acute lymphoblastic leukemia (ALL) patients, was based on dasatinib plus steroids induction - with central nervous system prophylaxis - followed by dasatinib alone in patients in complete molecular response or by chemotherapy and/or allogeneic transplant in patients not reaching complete molecular response. Sixty patients (median age 41.9 years) were enrolled: 33 were p190+, 18 p210+ and nine p190/p210+. At the end of induction (day +85), 58 patients (97%) achieved complete hematologic remission. No deaths in induction were recorded. Eleven patients (18.3%) obtained complete molecular response. Among the incomplete molecular responders (n=47), 22 underwent an allogeneic transplant. Seventeen hematologic relapses occurred (median 7 months; range, 340.1): 13 during consolidation and four post-transplant. ABL1 mutations (five T315I, three V299L, one E281K and one G254E) were found in ten of 13 relapsed cases. With a median follow-up of 57.4 months (range, 4.2-75.6), overall survival and disease-free survival were 56.3% and 47.2%. A better disease-free survival was observed in patients who obtained a molecular response at day +85 compared to cases who did not. The presence of additional copy number aberrations - IKZF1 plus CDKN2A/B and/or PAX5 deletions - was the most important unfavorable prognostic factor on overall and disease-free survival (P=0.005 and P=0.0008). This study shows that in adult Ph+ ALL long-term survivals can be achieved with a total-therapy strategy based on a chemotherapyfree induction and, in complete molecular responders, also without further systemic chemotherapy. Finally, the screening of additional copy number aberrations should be included in the diagnostic work-up (clinicatrial gov. Identifier: EudraCT 2010-019119-39). haematologica | 2021; 106(7)

Final results of the GIMEMA LAL1509 for Ph+ ALL

Introduction Philadelphia positive (Ph+) acute lymphoblastic leukemia (ALL) was historically recognized as the ALL subset with the most unfavorable outcome. The Ph chromosome designates the shortened chromosome 22 which encodes the BCR-ABL fusion gene/protein kinase. It arises from a translocation termed t(9;22)(q34;q11) and leads to BCRABL1 rearrangement.1 The incidence of the Ph+ ALL increases with age, is detected in more than 50% of elderly B-lineage ALL patients, and represents the most common genetic abnormality in adult ALL.2-4 The advent of tyrosine kinase inhibitors (TKI) has dramatically changed the management and the prognosis of this high-risk group of patients.5-9 In fact, in the pre-TKI era, treatment with standard chemotherapy rarely produced sustained complete remissions and less than 20% of Ph+ ALL patients were long-term survivors. Allogeneic stem cell transplant (allo-SCT), when feasible according to age and comorbidities, represented the only possibility of cure, if complete hematologic remission (CHR) was achieved.10-12 Nowadays, TKI with5,9,13-15 or without7,8,16,17 systemic chemotherapy in induction represent the gold standard approach. TKI-based schemes have led to a significant improvement in terms of CHR, reached in virtually all patients, disease-free survival (DFS) and overall survival (OS) rates. Nevertheless, the combination of TKI with conventional chemotherapy is aggravated by toxicities and treatment-related mortalities and a 2-7% death rate during induction has been reported in different studies.9,13,18 Deintensified chemotherapy regimens have been utilized to limit toxicity associated with the combined use of a TKI and standard chemotherapy.19,20 In order to overcome the treatment-related toxicity of the combined TKI-chemotherapy approach, the GIMEMA (Gruppo Italiano Malattie EMatolologiche dell’Adulto) group has applied - over the past 15 years - a chemotehrapy-free induction strategy using first, second or third generation TKI in combination with steroids and central nervous system (CNS) prophylaxis. The first trial with imatinib (clinicaltrials gov. Identifier: LAL0201) was designed for elderly patients (>60 years) and represented the proof of principle that CHR could be achieved without the use of systemic chemotherapy in Ph+ ALL.7 In the following study (clinicaltrials gov. Identifier: LAL0904) the same induction backbone was applied to adults up to the age of 60 years of age and patients received chemotherapy as consolidation treatment; if feasible by donor availability and clinical fitness, patients then underwent allo-SCT. The results of this study showed an OS and DFS at 60 months of 48.8% and 45.8%, respectively.17 The subsequent GIMEMA trial (clinicaltrials gov. Identifier: LAL1205) used the second generation TKI dasatinib for 12 weeks as firstline induction treatment for all Ph+ ALL over 18 years of age and with no upper age limit. All 53 patients obtained CHR at the end of the induction with eight patients (15.1%) being in complete molecular response (CMR) and no deaths or progressions were recorded during induction; consolidation was left to each treating center. At 20 months, the OS was 69.2% and the DFS was 51.1%.16 In the subsequent GIMEMA LAL1509 protocol the same induction strategy was utilized - dasatinib in combination with steroids together with CNS prophylaxis - followed by a consolidation strategy modulated according to the molecular response upon induction: patients in CMR continued haematologica | 2021; 106(7)

dasatinib until relapse/progression, while patients who did not reach CMR status continued with systemic chemotherapy and/or allo-SCT. We herein report the final results of the study.

Methods Study design, therapy and endpoint The GIMEMA LAL1509 trial (clinicaltrials gov. Identifier: EudraCT 2010-019119-39) was designed for adult Ph+ ALL patients (18-60 years). The identification of the BCR-ABL1 transcript was performed centrally within the steroid pre-phase (see below). Induction consisted of 12 weeks of dasatinib administration (140 mg/day), preceded by a 7-day steroid pre-phase (escalating doses from 20 up to 60 mg/m2): steroids were administered for 24 days, then tapered until day 31. Patients continued treatment with dasatinib until day 84 after induction, cases in CMR continued dasatinib until progression or minimal residual disease (MRD) increase or occurrence of toxicity (grade ≥3). MRD-positive patients were stratified according to transplant eligibility: i) those with a promptly available human leukocyte antigen-compatible donor proceeded to an allo-SCT (by protocol guidelines, a haploidentical donor was not permitted); ii) those with a compatible, but not readily available donor, received one chemotherapy consolidation cycle prior to allo-SCT; iii) those non-eligible for alloSCT received two chemotherapy consolidation cycles. The first consolidation cycle consisted of clofarabine (40 mg/m2, days 1-5) and cyclophosphamide (400 mg/m2, days 1-5); the second consolidation cycle consisted of 3 days of the same combination; dasatinib was interrupted during the 5 and 3 days of chemotherapy consolidation to avoid excessive toxicity and was restarted as soon as possible, at hematologic recovery. Upon consolidation, patients received dasatinib, until intolerance or relapse. Figure 1 summarizes the treatment scheme and patients’ disposition. CNS prophylaxis is detailed in the Online Supplementary Materials and Methods. The primary endpoint of the study was to evaluate the rate of patients alive in CHR. The study was approved by the Ethics Committees of all participating centers; patients gave their written informed consent in accordance with the Declaration of Helsinki.

Hematologic response assessment CHR was defined as bone marrow (BM) containing less than 5% blasts, absence of blasts in the peripheral blood (PB) and no extramedullary involvement. Hematologic relapse was defined as the re-appearance of blasts in the BM and/or in extramedullary sites.

Molecular diagnosis, definition of molecular response and MRD monitoring Molecular analyses were centrally performed at the Hematology Center of the Sapienza University of Rome. Molecular testing was carried out to identify the presence of BCR-ABL1 gene in BM samples at baseline and for MRD monitoring. MRD monitoring was performed by quantitative real-time polymerase chain reacxtion21,22 during induction at days +22, +45, +57 and +85 (end of induction) and at fixed time points according to the post-induction therapy (Online Supplementary Materials and Methods). CMR was defined as a BCR-ABL1/ABL1=0, with a confirmatory molecular BM aspirate after 15 days. Molecular relapse was defined as an at least 2-log increase of the BCR-ABL1/ABL1 gene expression confirmed in an additional control, which was solicited as soon as a molecular increase was documented. 1829

Figure 1. Protocol scheme and patients’ disposition. Dashed lines represent protocol deviation. n= number; CHR: complete hematologic remission; CMR: complete molecular response; allo-SCT: allogeneic stem cell transplant; MRD: minimal residual disease; CTX: cyclophosphamide.

S. Chiaretti et al.

1830

haematologica | 2021; 106(7)

Final results of the GIMEMA LAL1509 for Ph+ ALL

Mutational screening was performed in relapsed cases depending on material availability by Sanger sequencing (Online Supplementary Materials and Methods). Copy number aberrations analysis was performed on diagnostic samples by Cytoscan arrays, as previously described.23

Statistical analysis Survival was estimated using the Kaplan-Meier Product Limit. Differences in OS and DFS were evaluated by means of the LogRank test in univariate analysis and by the Cox regression model in multivariate analysis, after assessment of proportionality of hazards. The role of allo-SCT on DFS was evaluated in the Cox regression model by means of a time-dependent covariate. All analyses were performed using the SAS system software (version 9.4). All tests were two-sided, accepting P≤0.05 as statistically significant. Study data were collected and managed using REDCap24,25 electronic data capture tools hosted at the GIMEMA Foundation.

Results Patients From July 2011 to October 2013, 60 patients with de novo Ph+ ALL were enrolled in the GIMEMA LAL1509 trial. Thirty-four patients were males and 26 females. Median age at presentation was 41.9 years (range, 18.7-59.1). The median white blood cell count (WBC) at diagnosis was 12.4x109/L (range, 1.4-178). A BCR-ABL1 p190 fusion transcript was detected in 33 patients (55%), a BCR-ABL1 p210 fusion transcript in 18 (30%), while 9 patients (15%) were positive for both transcripts; p210 and p190/p210 were grouped together for all statistical analyses. The median follow-up was 57.4 months (range, 4.2-75.6) (Table 1).

Hematologic response during and after induction After the steroid pre-phase, 38 of 60 patients (63%) showed a PB blast reduction of ≥75%. All patients achieved CHR by day +57, 51 patients (85%) by day +22. By the end of the induction (day +85), all 60 patients were evaluable for response: 58 patients (97%) were in CHR, while two patients showed a disease progression between day +57 and day +85, both of these were p210-positive.

Minimal residual disease monitoring during and at the end of induction Molecular analysis showed that transcript levels rapidly and constantly decreased during the induction phase. The rate of transcript reduction was highly significant (P<0.0001) between the onset of the disease and day +22, between days +22 and +45, and between days +57 and +85. Importantly, a significantly greater clearance was observed in BCR-ABL1 p190 (n=33) versus p190/210 and p210 (n=27) cases at every time point of MRD evaluation (P=0.0032, P=0.0031, P=0.0016, P=0.007 at days +22, +45, +57 and +85, respectively) (Figure 2A). At the end of the induction, CMR was obtained in 11 patients (18.3%). In these patients, the rate of BCR-ABL1 transcript clearance was significantly higher at every point of MRD evaluation since day +22 (Figure 2B).

Post-remission treatment and response Figure 1 summarizes the patients’ disposition. Of the 11 patients who were in CMR at the end of induction, seven continued with dasatinib: four remain in CHR and CMR after 64, 54, 55 and 53 months (one patient was allografted haematologica | 2021; 106(7)

Table 1. Patients’ characteristics

Patients’ characteristics Median age at diagnosis (range) Sex M F Median WBC count at diagnosis (range) Type of fusion transcript p190 p210* p190/p210* Median FU (range)

N=60 41.9 years (18.7-59.1) 34 (56.6%) 26 (43.3%) 12.4x109/L (1.4-178) 33 (55%) 18 (30%) 9 (16%) 57.4 months (4.2-75.6)

*Considered together for subsequent analyses: M: male; F: female; WBC: white blood cell count; FU: follow-up.

after 6 months of dasatinib administration), one experienced a hematologic relapse, one experienced a molecular recurrence and one patient died of transplant-related complications, performed after 6 months of dasatinib maintenance while in CHR and CMR. The remaining four patients stopped dasatinib maintenance: two due to a molecular relapse (BCR-ABL1 T315I mutation was detected in one) and were both allografted, and one patients for a protocol violation due to a medical decision (i.e., transplant allocation within 6 months); all three patients are in CHR after 57, 58 and 60 months. Finally, one patient went off-study for non-compliance. The 47 patients who were in CHR but not in CMR at the end of induction were evaluated for allo-SCT eligibility. According to protocol guidelines, 25 of 47 patients were considered eligible for the procedure. Seventeen patients were transplanted, 11 after a consolidation cycle with clofarabine plus cyclophosphamide and of these, six patients obtained CMR. Eight patients did not undergo an allo-SCT due to: i) one case of patient’s refusal, ii) a documented hematologic or extramedullary relapse in four cases, iii) an unforeseen unavailability of the donor in one case and iv) a medical decision in two cases (one major protocol violation). The remaining 22 of 47 patients were considered ineligible for transplant: 20 patients underwent at least one cycle of clofarabine-cyclophosphamide and 13 received both cycles, with six patients obtaining CMR, while two patients went off protocol prior to consolidation treatment due to a major violation. After the second cycle of clofarabine-cyclophosphamide, 11 of 13 patients started maintenance with dasatinib. Five patients, initially deemed ineligible for transplant, could undergo the procedure after one cycle of consolidation chemotherapy: in two cases for a medical decision, in one case because of the presence of the BCR-ABL1 T315I mutation and in two cases for dasatinib interruption (hematologic toxicity in one and patient’s decision in the other). Overall, 22 patients underwent an alloSCT and after the procedure 12 patients restarted dasatinib as maintenance treatment. After allo-SCT, five patients were persistently MRD-positive and two of them went off treatment due to the emergence of the BCR-ABL1 T315I mutation and for a medical decision, respectively, whereas the others restarted treatment with dasatinib. Thirteen patients were MRD-negative after the transplant and in nine patients dasatinib was restarted: i) in three patients dasatinib was administered for the planned 6 months; ii) in one patient for 5 months; iii) in one patient for 3 months; iv) in two patients for 2 and 1 months each. 1831

S. Chiaretti et al.

Figure 2. Minimal residual disease clearance. Patients were stratified according to fusion protein (A) and molecular status at the end of induction (B). CMR: complete molecular response; d= day.

Relapse and cumulative incidence of relapse Seventeen relapses occurred at a median time of 7 months from the end of induction (range, 3-40.1), of which two were extramedullary (one CNS and one psoas muscle): 13 patients relapsed during consolidation and four patients relapsed after allo-SCT. At a median follow up of 57.4 months, the cumulative incidence of hematologic relapse (CIR) was 29.8% (95% Confidence Interval [CI]: 18.4-42.1) (Figure 3). The CIR, stratified according to the molecular response at day +85, was 10% in CMR patients as opposed to 35.6% in nonCMR patients; this difference did not reach statistical significance, probably due to the small number of relapses.

Toxicity and deaths Overall, 188 adverse events (AE) (grade >2) were recorded in 41 patients (Online Supplementary Table S1), the most frequent were hematologic (124 of 177, 66%), gastrointestinal (12 of 188, 6%) and infectious (12 of 188, 6%). As expected, hematologic toxicity was mainly recorded during induction and was most likely sustained by the disease itself. The most frequent AE reported with dasatinib was one registered case of pleural effusion and one registered case of peripheral edema, while no cases of pulmonary hypertension were recorded. Eighteen severe adverse events (SAE) were recorded in ten patients (Table 2), and six SAE were due to treatment: 1832

infections and gastrointestinal disorders were the most frequent (27% and 22%, respectively). While no deaths occurred during the induction phase, 24 deaths were recorded in the post-induction phase: 12 for disease progression, 11 in CHR for toxicity and one for both. Causes of death in CHR patients were represented by sepsis in four patients, multi-organ failure (kidney and lung) in two patients and transplant-related mortality in five patients (one for graft versus host disease, two for viral pneumonitis, two for an encephalitis and two for hemolytic anemia).

BCR-ABL1 mutations and copy number evolution analysis BCR-ABL1 mutational screening was performed in cases with biological material available at either MRD increase or overt hematologic relapse. A total of 13 cases were screened (ten with a hematologic relapse and three with MRD increase) and the following BCR-ABL1 mutations were detected in ten cases: five cases of T315I, three cases of V299L (one of which with concomitant F317I and F317L mutations), one case of E281K and one case of G254E. No difference in the incidence of BCR-ABL1 mutations was found between p190+ and p210+ patients. Notably, in the three cases with a MRD increase a mutation was detected (E281K, T315I and V299L), but it did not translate into a subsequent hematologic relapse. A total of nine mutated cases were retrospectively evaluated on the diagnostic material and no mutations were detected. haematologica | 2021; 106(7)

Final results of the GIMEMA LAL1509 for Ph+ ALL

Figure 3. Cumulative incidence of relapse. CI: Confidence Interval; CHR: complete hematologic remission; yrs: years.

Single nucleotide polymorphism array analysis was carried out at presentation in 39 cases; the most frequent aberrations were deletions of IKZF1 (84.6%), PAX5 (38.5%), CDKN2A/B (33.3%), MLLT3 (33.3%), JAK2 (28.2%) and RB1 (28.2%). Furthermore, in a small fraction of patients (four of 37; 10.8%) a deletion involving MEF2C was identified.

Disease-free survival and overall survival At a median follow up of 57.4 months, the DFS was 47.2% (95% CI: 35.6.0-72.6) (Figure 4A), with a median DFS of 30.4 months, and the OS was 56.3% (95% CI: 44.5-71.3), with a median OS not reached (Figure 4B). A better, though not significant, DFS was observed in patients who obtained a CMR at day +85, compared to cases with a MRD positivity (70% vs. 42%, P=0.08) (Figure 5A). The observed DFS prolongation in CMR patients did not translate into a statistically significant advantage in OS, even though better survival was observed (70% vs. 52.9%, P=0.27) (Figure 5B). This is probably related to the number of patients and because of the efficacy of consolidation treatments in inducing MRD negativity. DFS was also evaluated on the basis of the fusion protein and the presence of additional copy number evolution (CNA). There were no significant differences in DFS between BCR-ABL1 p190+ patients compared to p210+ and/or p190/p210+ cases (49.7% vs. 41.7%, P=0.4). Likewise, OS was superimposable among the two subsets (56% vs. 44.2%, P=0.9, not shown). At variance, the presence of additional CNA cases with the co-occurrence of IKZF1 plus CDKN2A/B and/or PAX5 (IKZF1 plus) at presentation had a strong impact on DFS and OS had a significantly worse DFS and OS than cases with no alterations and/or with IKZF1 deletions only: 0% versus 60% (P=0.0008) and 20% versus 69.5% (P=0.0068) at 60 months (Figure 6). Importantly, the presence of IKZF1 deletions alone did not affect DFS. Unfortunately, as also reported by Pfeifer and colleagues,26 the negative prognostic impact was also not abrogated when adjusted by transplant haematologica | 2021; 106(7)

Table 2. Number and types of serious adverse events occurred during the protocol.

Serious adverse events

Total number of events (N=8)

Events related to dasatinib or chemotherapy

Gastrointestinal system Nervous system Serous effusions Infectious diseases Cardiologic toxicity General disorders Graft versus host disease

4 2 2 5 1 3 1

2 1 2 0 0 1 0

allocation. In addition, patients with IKZF1 plus CDKN2A/B or PAX5 deletions showed a significantly increased CIR (P=0.0031) i.e., all patients harboring these alterations relapsed within 24 months from induction therapy. Patients carrying a MEF2C deletion appeared to have a better DFS as opposed to patients without it (80% vs. 39.6%, P=0.12).23 Finally, allo-SCT, evaluated in a Cox model by a time-dependent covariate, did not impact on OS and DFS survival.

Univariate and multivariate survival analyses Several factors (age, sex, WBC count, type of fusion protein, additional genetic lesions, etc.) were investigated in univariate analysis to establish their impact on DFS: patients who achieved a CMR after induction, showed a better, though non-significant, DFS, also in multivariate analysis (P=0.07). The co-occurrence of IKZF1 deletions plus CDKN2A/B or PAX5 deletions was the only significant factor associated with a worse DFS (P=0.0008) in our cohort of patients. Along the same line, IKZF1 deletions plus CDKN2A/B or PAX5 were the only significant factors impacting on OS in both univariate (P=0.01) and multivariate analysis (P=0.005). 1833

S. Chiaretti et al. A

Figure 4. Survival estimates. (A) Disease-free survival (DFS) and (B) overall survival (OS). CI: Confidence Interval; CHR: complete hematologic remission; yrs: years.

Discussion We herein reported the final results of the GIMEMA LAL1509 trial for adult Ph+ ALL patients. This protocol was based on a total therapy strategy that comprised a chemotherapy-free induction based on dasatinib, plus steroids and CNS prophylaxis, followed by a post-remission treatment tailored according the molecular response. Patients in sustained CMR continued with dasatinib, while patients in CHR who did not achieve CMR underwent chemotherapy consolidation with clofarabinecyclophosphamide plus an allo-SCT for eligible patients. The protocol was designed to evaluate the efficacy of a post-induction treatment modulated on the basis of the depth of molecular response and represents, so far, the first protocol to investigate the possibility of chemotherapy-free consolidation in adult patients (18-60 years) in CMR. This study confirms the effectiveness of chemotherapy-free induction with dasatinib plus steroids in Ph+ ALL adult patients in inducing a rapid clearance of a neoplastic clone16 and also confirms that this strategy is less toxic than combinational strategies,19,27-30 since no deaths during 1834

induction were recorded, as already reported in previous GIMEMA trials for Ph+ ALL.7,16,17 Of the 60 eligible patients, 97% reached CHR at the end of the induction with two patients in CHR at day +57, both harboring the p210 fusion protein, who relapsed between day +57 and +85 of the induction phase. At the end of the induction (day +85), CMR - confirmed by a second BM aspirate after 15 days - was achieved in 11 cases (18.3%). Seven of these 11 patients continued treatment with dasatinib: four patients are still alive and disease-free (one underwent allo-SCT while in CMR), one patient underwent alloSCT and died of transplant-related complications, one patient experienced a hematologic relapse and one patient suffered molecular recurrence; the latter two cases both had an IKZF1-plus phenotype. The remaining four CMR patients were rapidly switched to different therapeutic options, two due to a molecular increase and two due to protocol violations (i.e., transplant allocation and non-compliance). With regards to the two cases with molecular recurrence, a strict MRD monitoring prevented an overt hematologic relapse (resulting in a benefit for patients) and may thus represent a bias of the study; nevertheless, it further highlights the importance of molecular monitoring, haematologica | 2021; 106(7)

Final results of the GIMEMA LAL1509 for Ph+ ALL

Figure 5. Survival stratified according to molecular response at the end of induction. (A) Disease-free survival (DFS) and (B) overall survival (OS). CI: Confidence Interval; CHR: complete hematologic remission; yrs: years.

mostly in an era in which a plethora of novel compounds are available. Even if the numbers are small and protocol violations represent a major limit, the DFS of the few cases in persistent CHR and CMR seems to be better than that of patients who did not achieve CMR. This suggest that some patients, regardless of their age, may be spared systemic chemotherapy and transplant procedure. However, the applicability of this strategy must be carefully evaluated and possibly refined by additional genetic information. Indeed, this chemo/transplant-sparing strategy is addressed in an ancillary study (clinicaltrials gov. Identifier: NCT03318770) to our recently published paper on adult Ph+ ALL patients of all ages treated in induction with dasatinib and consolidated with the bispecific monoclonal antibody blinatumomab.31 Regarding patients in CHR who did not achieve CMR, 35 received chemotherapy consolidation with clofarabinecyclophosphamide: overall, this treatment was capable of haematologica | 2021; 106(7)

inducing CMR in 12 cases (34.3%), thus indicating its effectiveness and in some cases a valuable bridge to transplant. Furthermore, this strategy was safe, since only 18 SAE were recorded in ten patients. Similarly, allo-SCT proved effective in inducing MRD negativity in 32% of the patients. Seventeen hematologic relapses (comprising two extramedullary) and 11 deaths in CHR were recorded. We observed a 5-year DFS of 47.2% and a 5-year OS 56.3%, making the outcome of Ph+ ALL patients almost comparable to that of Ph- ALL cases. These results represent a slight improvement compared to our previous trial - GIMEMA LAL090417 - based on a chemotherapy-free imatinib administration in induction, followed by consolidation and alloSCT, in which the 5-year DFS and OS were 45.8% and 48.8%, respectively. The finding that DFS was not significantly superior might be related to the well-established fact that dasatinib, even though more effective in rapidly eradicating a leukemic clone, is not capable of overcoming the occurrence of TKI-resistant mutations, particularly the 1835

S. Chiaretti et al. A

Figure 6. Impact of genetic lesions on survival. (A) Disease-free survival (DFS) and (B) overall survival (OS) of IKZF1 deletion with or without other abnormalities. CI: Confidence Interval; CHR: complete hematologic remission; yrs: years.

T315I mutation. Although mutations were not systematically screened in this trial, ten patients in hematologic relapse were evaluated and seven indeed harbored a mutation, with T315I being the most frequent (n=4). These results are in line with our previous experience16 and with Rousselot and colleagues who reported a high incidence of T315I mutations (18 of 24 relapsing patients, 75%).32 Importantly, three cases were screened and proved positive for ABL1 mutations at MRD increase and none of them relapsed, thus corroborating the notion that an earlier detection can better drive therapeutic decisions. A 7.5% increase in OS was observed in the current study compared to the results of the LAL0904 protocol, from 48.8% to 56.3%. Our results compare favorably also with other studies based on a combination of dasatinib plus chemotherapy, which were also aggravated by toxicities and treatment discontinuations.27,32 The European Working Group on Adult ALL used dasatinib in combination with low-dose chemotherapy in elderly patients with Ph+ ALL: 61 patients 1836

were enrolled (median age 69 years), 96% of patients achieved CR after induction, but three patients died. Overall, the 5-year OS was 36%, suggesting that a proportion of patients may experience long-term survival without intensive therapies.32 Ravandi et al.27 published the final results of their phase II study with dasatinib plus HyperCVAD in 72 adult Ph+ ALL patients: they reported CHR of 96% with 83% of complete cytogenetic responses after induction with three early deaths. Overall, the 5-years DFS and OS were 44% and 46%, respectively. The US Intergroup reported the results of a multicenter trial for Ph+ ALL patients aged 18-60 years treated with dasatinib plus HyperCVAD and allo-SCT for eligible patients.33 The overall complete remission/complete remission with incomplete hematologic recovery (CR-CRi) rate was 88% and the 3-year OS and DFS were 69% and 62%, respectively. The shorter follow-up does not allow a direct comparison with our study. In addition, a sizable proportion of patients (n=34) had received one course of therapy prior to enrollhaematologica | 2021; 106(7)

Final results of the GIMEMA LAL1509 for Ph+ ALL

ment in the study, including 16 who had already achieved a CR or CRi. In the pre-TKI era, allo-SCT in first remission was considered the standard of care for all eligible patients. After the introduction of TKI, the use of allo-SCT in first CHR has been questioned.19,34-38 Although our study was not powered to define its role, allo-SCT did not seem to have an impact on survival. This might be due to the small sample size of patients who performed the procedure and may also be influenced by protocol violations, which represent a major limit of the study: indeed, five patients in CMR after dasatinib induction underwent allo-SCT. Finally, allo-SCT was aggravated by mortality in five patients. In the report by Ravandi et al.27 only a small proportion of patients underwent allo-SCT in first remission without any improvement in survival. A recent report from Chang and colleagues36 also reported no benefit in OS and DFS at 3 years between transplanted and nontransplanted patients in a retrospective study (76% vs. 71.3%, P=0.56; 70.5% vs. 80.1%, P=0.94, respectively) in Ph+ ALL patients treated with dasatinib plus chemotherapy followed or not by allo-SCT. In contrast, in the report by the US Intergroup study, the landmark analysis at 175 days showed a significant advantage in both OS and relapse-free survival (P=0.038 and P=0.0037, respectively) in transplanted patients.33 In our cohort of transplanted patients, for patients relapsed after allo-SCT, all of them were MRD-positive before the procedure. This finding is in line with a recent publication by Candoni and colleagues in which the 5-year CIR was higher in Ph+ ALL patients that were MRD+ prior to the allograft.37 With regard to the specific biologic features investigated, we observed, during induction, a significantly better clearance of the BCR-ABL1 transcript in p190 patients as opposed to p210 patients; interestingly, this, however, did not translate into an improved DFS and OS, thus suggesting that the kinetics of MRD clearance is different between the p190 and p210 cases. Further investigations are ongoing on this specific issue. Similarly, DFS was not statically better for patients achieving CMR after induction, mostly because of the small sample size and also indicating that patients who failed to achieve CMR can be rescued by an effective post-induction treatment. At variance, patients with IKZF1 deletions plus CDKN2A/B and/or PAX5 deletions had a significantly worse DFS and OS, all relapsing within 24 months: this finding held true also in cases that achieved CMR. Also, the IKZF1-plus signature was the only prognostic factor to have an impact on DFS and OS in both univariate and multivariate analyses. Therefore, screening for these genetic lesions

References 1. Nowell PC, Hungerford DA. Chromosome studies on normal and leukemic human leukocytes. J Natl Cancer Inst. 1960;25:85109. 2. Mancini M, Scappaticci D, Cimino G, et al. A comprehensive genetic classification of adult acute lymphoblastic leukemia (ALL): analysis of the GIMEMA 0496 protocol. Blood. 2005;105(9):3434-3441. 3. Chiaretti S, Vitale A, Cazzaniga G, et al. Clinico-biological features of 5202 patients with acute lymphoblastic leukemia enrolled in the Italian AIEOP and GIMEMA

haematologica | 2021; 106(7)

should be performed in all Ph+ ALL patients at the time of diagnosis for a more refined prognostic stratification, and to further optimize treatment.23,26,31 Taken together, the results of this study document that durable remissions and long-term survival can be achieved with a total therapy strategy based on a chemotherapy-free induction, with an OS rate of 56.3% at a median follow-up of 57.4 months. Even in the TKI era, allo-SCT still remains an effective treatment for eligible adult patients; further efforts need to be made to identify which patients may be spared the procedure and its related toxicity, through the design of trials with a larger sample size that allow patients’ stratification according to their molecular features. Whether the use of more powerful BCR-ABL1 inhibitors such as ponatinib, may improve OS in Ph+ ALL remains to be determined.38,39 The cardiovascular toxicity of ponatinib may counteract the potential benefit of this drug in patients with serious comorbidities. Finally, the upfront integration of TKI with immunoconjugates or BiTE antibodies - such as blinatumomab - appears capable of inducing a high percentage of molecular response in adult Ph+ ALL patients and could eventually lead to higher rates of long-term responses.31 Disclosures RF has served on advisory boards and at speaker bureaus of Janssen, AbbVie, Novartis, Amgen, Incyte, Pfizer and Servier; SC has served on advisory boards of Amgen, Pfizer, Incyte and Shire; the other authors declare no conflicts of interest. Contributions SC provided samples and clinical data, analyzed data and wrote the manuscript; MA analyzed data and wrote the manuscript; AV provided clinical data; LE and MM performed experiments; AP performed statistical analyses; PF contributed to protocol management; FDR, LS, FF, CC, GM, FR, FF, NC, CB, AT, SS, NDR, AM and GB provided clinical data; MV contributed to protocol management; RF designed the research and the trial, and critically revised the manuscript. Acknowledgements The authors wish to thank Sanofi and Bristol Myers Squibb, which provided clofarabine and dasatinib, respectively, free of charge. Funding The authors wish to thank the Associazione Italiana per la Ricerca sul Cancro (AIRC) 5x1000, Special Program Metastases (21198), Milan (Italy) to RF; Finanziamento Medi Progetti Universitari 2015 to SC (Sapienza University of Rome) and PRIN 2017 (2017PPS2X4_002) to SC.

protocols and stratified in age cohorts. Haematologica. 2013;98(11):1702-1710. 4. Burmeister T, Schwartz S, Bartram CR, et al. Patients’ age and BCR–ABL frequency in adult B- precursor ALL: a retrospective analysis from the GMALL study group. Blood. 2008;112(3):918-919. 5. Yanada M, Takeuchi J, Sugiura I, et al. High complete remission rate and promising outcome by combination of imatinib and chemotherapy for newly diagnosed BCRABL-positive acute lymphoblastic leukemia: a phase II study by the Japan Adult Leukemia Study Group. J Clin Oncol. 2006;24(3):460-466.

6. Talpaz M, Shah NP, Kantarjian H, et al. Dasatinib in imatinib-resistant Philadelphia chromosome-positive leukemias. N Engl J Med. 2006;354(24):2531-2541. 7. Vignetti M, Fazi P, Cimino G, et al. Imatinib plus steroids induces complete remissions and prolonged survival in elderly Philadelphia chromosome-positive patients with acute lymphoblastic leukemia without additional chemotherapy: results of the Gruppo Italiano Malattie Ematologiche dell’Adulto (GIMEMA) LAL0201-B protocol. Blood. 2007;109(9):3676-3678. 8. Ottmann OG, Wassmann B, Pfeifer H, et al. Imatinib compared with chemotherapy as

1837

S. Chiaretti et al. front-line treatment of elderly patients with Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph+ALL). Cancer. 2007;109(10):2068-2076. 9. Bassan R, Rossi G, Pogliani EM, et al. Chemotherapy-phased imatinib pulses improve long- term outcome of adult patients with Philadelphia chromosomepositive acute lymphoblastic leukemia: Northern Italy Leukemia Group protocol 09/00. J Clin Oncol. 2010;28(22):3644-3652. 10. Dombret H, Gabert J, Boiron JM, et al. Outcome of treatment in adults Philadelphia chromosome-positive acute lymphoblastic leukemia-results of the prospective multi-center LALA-94 trial. Blood. 2002;100(7):2357-2366. 11. Laport GG, Alvarnas JC, Palmer JM. Longterm remission of Philadelphia chromosome-positive acute lymphoblastic leukemia after allogeneic hematopoietic cell transplantation from matched sibling donors: a 20-year experience with the fractionated total body irradiation-etoposide regimen. Blood. 2008; 112(3):903-909. 12. Fielding AK, Rowe JM, Richards SM. Prospective outcome data on 267 unselected adult patients with Philadelphia chromosome–positive acute lymphoblastic leukemia confirms superiority of allogeneic transplantation over chemotherapy in the pre-imatinib era: results from the International ALL Trial MRC UKALLXII/ECOG2993. Blood. 2009; 113(19):4489-4496. 13. Wassmann B, Pfeifer H, Goekbuget N, et al. Alternating versus concurrent schedules of imatinib and chemotherapy as front-line therapy for Philadelphia-positive acute lymphoblastic leukemia (Ph+ALL). Blood. 2006;108(5):1469-1477. 14. de Labarthe A, Rousselot P, Huguet-Rigal F, et al. Group for Research on Adult Acute Lymphoblastic Leukemia (GRAALL). Imatinib combined with induction or consolidation chemotherapy in patients with de novo Philadelphia chromosome-positive acute lymphoblastic leukemia: results of the GRAAPH-2003 study. Blood. 2007;109 (4):1408-1413. 15. Ribera JM, Oriol A, Gonzalez M, et al. Concurrent intensive chemotherapy and imatinib before and after stem cell transplantation in newly diagnosed Philadelphia chromosome-positive acute lymphoblastic leukemia. Final results of the CSTIBES02 trial. Haematologica. 2010;95(1):87-95. 16. Foà R, Vitale A, Vignetti M, et al. Dasatinib as first-line treatment for adult patients with Philadelphia chromosome-positive acute lymphoblastic leukemia. Blood. 2011;118(25):6521-6528. 17. Chiaretti S, Vitale A, Vignetti M, et al. A sequential approach with imatinib, chemotherapy and transplant for adult Ph+ acute lymphoblastic leukemia: final results of the GIMEMA LAL0904 study. Haematologica. 2016;101(12):1544-1552. 18. Fielding AK, Rowe JM, Buck G, et al.

1838

UKALLXII/ECOG2993: addition of imatinib to a standard treatment regimen enhances long-term outcomes in Philadelphia positive acute lymphoblastic leukemia. Blood. 2014;123(6):843-850. 19. Ribera JM, García O, Montesinos P, et al. Treatment of young patients with Philadelphia chromosome-positive acute lymphoblastic leukaemia using increased dose of imatinib and deintensified chemotherapy before allogeneic stem cell transplantation. Br J Haematol. 2012;159 (1):78-81. 20. Chalandon Y, Thomas X, Hayette S, et al; Group for Research on Adult Acute Lymphoblastic Leukemia (GRAALL). Randomized study of reduced-intensity chemotherapy combined with imatinib in adults with Ph-positive acute lymphoblastic leukemia. Blood. 2015;125(24):37113719. 21. Gabert J, Beillard E, van der Velden VH, et al. Standardization and quality control studies of ‘real-time’ quantitative reverse transcriptase polymerase chain reaction of fusion gene transcripts for residual disease detection in leukemia - a Europe Against Cancer program. Leukemia. 2003;17(12): 2318-2357. 22. Elia L, Mancini M, Moleti L, et al. A multiplex reverse transcriptase polymerase chain reaction strategy for the diagnostic molecular screening of chimeric genes: a clinical evaluation on 170 patients with acute lymphoblastic leukemia. Haematologica. 2003;88(3):275-279. 23. Fedullo AL, Messina M, Elia L, et al. Prognostic implications of additional genomic lesions in adult Philadelphia chromosomepositive acute lymphoblastic leukemia. Haematologica. 2019;104(2):312-318. 24. Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap) – A metadata- driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377-381. 25. Harris PA, Taylor R, Minor BL, et al. The REDCap Consortium: building an international community of software partners. J Biomed Inform.2019;95:103208. 26. Pfeifer H, Raum K, Markovic S, et al. Genomic CDKN2A/2B deletions in adult Ph+ ALL are adverse despite allogeneic stem cell transplantation. Blood. 2018;131 (13):1464-1475. 27. Ravandi F, O'Brien SM, Cortes JE, et al. Long-term follow-up of a phase 2 study of chemotherapy plus dasatinib for the initial treatment of patients with Philadelphia chromosome-positive acute lymphoblastic leukemia. Cancer. 2015;121(23):4158-4164. 28. Daver N, Thomas D, Ravandi F, et al. Final report of a phase II study of imatinib mesylate with hyper-CVAD for the front-line treatment of adult patients with Philadelphia chromosome-positive acute lymphoblastic leukemia. Haematologica. 2015;100(5):653-661.

29. Lim SN, Joo YD, Lee KH, et al. Long-term follow-up of imatinib plus combination chemotherapy in patients with newly diagnosed Philadelphia chromosome-positive acute lymphoblastic leukemia. Am J Hematol. 2015;90(11):1013-1020. 30. Hatta Y, Mizuta S, Matsuo K, et al. Final analysis of the JALSG Ph+ALL202 study: tyrosine kinase inhibitor-combined chemotherapy for Ph+ ALL. Ann Hematol. 2018;97(9):1535-1545. 31. Foà R, Bassan R, Vitale A, et al. Dasatinibblinatumomab for Ph-positive acute lymphoblastic leukemia in adults. N Engl J Med. 2020;383(17):1613-1623. 32. Rousselot P, Coudé MM, Gokbuget N, et al. Dasatinib and low-intensity chemotherapy in elderly patients with Philadelphia chromosome-positive ALL. Blood. 2016;128(6): 774-782. 33. Ravandi F, Othus M, O'Brien SM, et al. US Intergroup study of chemotherapy plus dasatinib and allogeneic stem cell transplant in Philadelphia chromosome positive ALL. Blood Adv. 2016;1(3):250-259. 34. Kim DY, Joo YD, Lim SN, et al. Nilotinib combined with multiagent chemotherapy for newly diagnosed Philadelphia-positive acute lymphoblastic leukemia. Blood. 2015;126(6):746-756. 35. Short N, Kantarjian HM, Ravandi F, et al. Long-term safety and efficacy of HyperCVAD plus ponatinib as frontline therapy for adults with Philadelphia chromosomepositive acute lymphoblastic leukemia. Blood. 2019;134(Suppl. 1):S283. 36. Chang J, Douer D, Aldoss I, et al. Combination chemotherapy plus dasatinib leads to comparable overall survival and relapse-free survival rates as allogeneic hematopoietic stem cell transplantation in Philadelphia positive acute lymphoblastic leukemia. Cancer Med. 2019;8(6):28322839. 37. Candoni A, Rambaldi A, Fanin R, et al. Outcome of allogeneic hematopoietic stem cell transplantation in adult patients with Philadelphia chromosome-positive acute lymphoblastic leukemia in the era of tyrosine kinase inhibitors: a registry-based study of the Italian Blood and Marrow Transplantation Society (GITMO). Biol Blood Marrow Transplant. 2019;25(12): 2388-2397. 38. Jabbour E, Short NJ, Ravandi F, et al. Combination of hyper-CVAD with ponatinib as first-line therapy for patients with Philadelphia chromosome-positive acute lymphoblastic leukaemia: long-term follow-up of a single-centre, phase 2 study. Lancet Haematol. 2018;5(12):e618-e627. 39. Martinelli G, Piciocchi A, Papayannidis C, et al. First report of the GIMEMA LAL1811 phase II prospective study of the combination of steroids with ponatinib as frontline therapy of elderly or unfit patients with Philadelphia chromosome-positive acute lymphoblastic leukemia. Blood. 2017;130 (Suppl.1):S99.

haematologica | 2021; 106(7)

ARTICLE

Acute Myeloid Leukemia

Comparison of total body irradiation versus non-total body irradiation containing regimens for de novo acute myeloid leukemia in children

Christopher E. Dandoy,1,2 Stella M. Davies,1,2 Kwang Woo Ahn,3 Yizeng He,3 Anders E. Kolb,4 John Levine,5 Stephanie Bo-Subait,6 Hisham Abdel-Azim,7 Neel Bhatt,8 Joseph Chewing,9 Shahinaz Gadalla,10 Nicholas Gloude,11 Robert Hayashi,12 Nahal R. Lalefar,13 Jason Law,14 Margaret MacMillan,15 Tracy O’Brien,16 Timothy Prestidge,17 Akshay Sharma,18 Peter Shaw,19 Lena Winestone20 and Mary Eapen21

Division of Hematology-Oncology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH, USA; 2Department of Pediatrics, University of Cincinnati, Cincinnati, OH, USA; 3Division of Biostatics, Institute for Heath and Equity, Medical College of Wisconsin, Milwaukee, WI, USA; 4Division of Hematology-Oncology, Alfred I. duPont Hospital for Children, Wilmington, DE, USA; 5Blood and Marrow Transplant Program, Icahn School of Medicine at Mount Sinai, New York, NY, USA; 6Center for International Blood and Marrow Transplant Research, National Marrow Donor Program/Be The Match, Minneapolis, MN, USA; 7Division of Hematology, Oncology and Blood & Marrow Transplantation, Children's Hospital Los Angeles, University of Southern California Keck School of Medicine, Los Angeles, CA, USA; 8Division of Clinical Research, Department of Data Abstraction, Fred Hutchinson Cancer Research Center, Seattle, WA, USA; 9Division of Hematology-Oncology, University of Alabama at Birmingham, Birmingham, AL, USA; 10Division of Cancer Epidemiology & Genetics, NIH-NCI Clinical Genetics Branch, Rockville, MD, USA 11Division of Hematology-Oncology, Rady Children’s Hospital San Diego, San Diego, CA, USA; 12 Division of Pediatric Hematology/Oncology, Department of Pediatrics, Washington University School of Medicine in St. Louis, St. Louis, MO, USA; 13Division of HematologyOncology, Children’s Hospital and Research Center Oakland, Oakland, CA, USA; 14Division of Pediatric Hematology-Oncology, Tufts Medical Center, Boston, MA, USA; 15Blood and Marrow Transplant Program, Department of Pediatrics, University of Minnesota, Minneapolis, MN, USA; 16Blood & Marrow Transplant Program, Kids Cancer Center, Sydney Children's Hospital, Sydney, New South Wales, Australia; 17Blood and Cancer Center, Starship Children’s Hospital, Auckland, New Zealand; 18Division of Bone and Marrow Transplantation and Cellular Therapy, St. Jude Children’s Research Hospital, Memphis, TN, USA; 19The Children’s Hospital at Westmead, Westmead, New South Wales, Australia; 20Division of Hematology-Oncology, Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, PA, USA and 21Division of Hematology-Oncology, Department of Medicine, Medical College of Wisconsin, Milwaukee, WI, USA

Ferrata Storti Foundation

Haematologica 2021 Volume 106(7):1839-1845

ABSTRACT

ith limited data comparing hematopoietic cell transplant outcomes between myeloablative total body irradiation (TBI) containing and non-TBI regimens in children with de novo acute myeloid leukemia, the aim of this study was to compare transplant-outcomes between these regimens. Cox regression models were used to compare transplant-outcomes after TBI and non-TBI regimens in 624 children transplanted between 2008 and 2016. Thirty two percent (n=199) received TBI regimens whereas 68% (n=425) received non-TBI regimens. Five-year non-relapse mortality was higher with TBI regimens (22% vs. 11%, P<0.0001) but relapse was lower (23% vs. 37%, P<0.0001) compared to non-TBI regimens. Consequently, overall (62% vs. 60%, P=1.00) and leukemia-free survival (55% vs. 52%, P=0.42) did not differ between treatment groups. Grade 2-3 acute graft versus host disease was higher with TBI regimens (56% vs. 27%, P<0.0001) but not chronic graft versus host disease. The 3-year incidence of gonadal or growth hormone deficiency was higher with TBI regimens (24% vs. 8%, P<0.001) but there were no differences in late pulmonary, cardiac or renal impairment. In the absence of a survival advantage, the choice of TBI or non-TBI regimen merits careful consideration with the data favoring non-TBI regimens to limit the burden of morbidity associated with endocrine dysfunction. haematologica | 2021; 106(7)

Correspondence: MARY EAPEN meapen@mcw.edu Received: February 6, 2020. Accepted: June 12, 2020. Pre-published: June 18, 2020. https://doi.org/10.3324/haematol.2020.249458

1839

C.E. Dandoy et al.

Introduction Hematopoietic cell transplant (HCT) is accepted as the standard of care for children and adolescents with high risk or relapsed/refractory acute myeloid leukemia (AML).1-3 Total body irradiation (TBI) or busulfan (Bu) containing myeloablative conditioning regimens are commonly used to treat these patients.4,5 In 1992, a randomized trial in adults showed improved 2-year leukemia-free survival using conditioning with TBI-cyclophosphamide (Cy) compared to Bu-Cy using oral busulfan formulation, which was available at that time,6 however, there are limited data in children and adolescents. Intravenous Bu has subsequently been developed, which provides more consistent pharmacokinetics and reliable dosing.7 Further, pharmacokinetic targeting of intravenous Bu dosing reduces treatment-related toxicity.8 In the modern era, in adults, intravenous Buconditioning has been shown to be associated with improved non-relapse mortality, and overall and diseasefree survival in comparison to TBI-containing regimens.9,10 However, acute graft versus host disease (GvHD),11 acute liver injury,12 pulmonary injury (e.g., diffuse alveolar hemorrhage), and bloodstream infections1,13,14 are reported to be higher with TBI-containing compared to non-TBI regimens. Late complications, including secondary malignancies, endocrine, metabolic, renal, ocular, and neurocognitive complications are also higher with TBI-containing compared to non-TBI myeloablative regimens.15-17 Although TBI-containing regimens are associated with significant toxicity in children,18 many physicians continue to use TBIcontaining regimens for transplantation for de novo AML in children and adolescents. In the absence of a randomized trial comparing TBI-containing versus non-TBI regimens in children, we utilized data on HCT reported to an observational registry, the Center for International Blood and Marrow Transplant Research (CIBMTR) to compare outcomes between the two treatment groups.

cause was considered an event. Leukemia-free survival was defined as being alive in continuous remission. Neutrophil recovery was defined as achieving a count of ≥0.5x109/L for 3 consecutive days. Platelet recovery was defined as achieving a count of ≥20x109/L without transfusions for 7 consecutive days. The day100 incidence of veno-occlusive disease, systemic bacterial, viral and fungal infection were compared between the two treatment groups. The 5-year incidence of post-transplant interstitial pneumonitis, congestive heart failure, gonadal dysfunction, growth hormone deficiency and renal failure severe enough to warrant dialysis were compared between the two treatment groups.

Statistical methods Patient-related, disease-related, and transplant-related outcomes were compared between treatment groups using MannWhitney tests (continuous variables) and Fisher’s exact/Chisquare test (categorical variables). A P-value of <0.05 was considered statistically significant. Cox regression models were built for acute and chronic GvHD, non-relapse mortality, relapse, overall and leukemia-free survival.21 The main effect (TBI-containing vs. non-TBI regimens) was forced in all models, and other covariates were retained in the final model if they met a significance level of less than 0.05. Forward stepwise selection was used to identify significant covariates. The interaction between the main effect and significant covariates was examined. Assessment of the proportional-hazards assumption was done by examining the coefficient of the logarithm of time from transplant to the last followup for each covariate. The coefficients for the covariates which violated the proportional hazards assumption were added as time-varying effects. The adjusted survival or cumulative incidence probabilities were calculated based on the final Cox models.22,23 Center effects were tested for non-relapse mortality, relapse, overall and leukemia-free survival using the score test.24 All analyses were performed using SAS 9.4 (SAS Institute Inc, Cary, NC).

Results Methods Patients Data were reported prospectively to the CIBMTR, a voluntary working group of more than 450 transplant centers worldwide that contribute detailed data on allogeneic and autologous HCT. Participating centers report consecutive transplants and compliance is monitored by on-site audits. All patients are followed longitudinally until death or lost to follow-up. Eligible patients were aged ≤21 years undergoing first allogeneic transplantation with myeloablative conditioning for de novo AML in first or second complete remission between 2008 and 2016 and consented for research. Excluded were patients with an antecedent hematologic disorder or secondary AML, mismatched related donor transplant and non-calcineurin inhibitor GvHD prophylaxis regimens. Patients were broadly grouped into TBI-containing (TBICy), TBI-Cy-fludarabine (Flu) and non-TBI (Bu-Cy and Bu-Flu) regimens. The study was approved by the Institutional Review Board of the National Marrow Donor Program.

Endpoint Grades 2-3 acute GvHD, grade 3-4 acute GvHD, and chronic GvHD were defined using standard definitions.19,20 Relapse was defined as the recurrence of AML (morphologic, cytogenetic or molecular) and non-relapse mortality was defined as death in remission. Overall survival was defined where death from any 1840

Patient, disease and transplant characteristics Six hundred and twenty-four patients transplanted at 124 transplant centers were eligible and their characteristics are shown in Table 1. TBI-containing regimens included TBI-Cy (38%, 76 of 199) and TBI-Cy-Flu (62%, 123 of 199). Non-TBI regimens included Bu-Cy (76%, 322 of 425) and Bu-Flu (24%, 103 of 425). Bu pharmacokinetics with dose adjustments were performed for 80% (338 of 425) of non-TBI transplantations. Patient and disease characteristics differed by treatment group. TBI-containing regimens were less likely to be used for children aged 3 years and younger, for transplants in first complete remission and more likely with umbilical cord blood (67%). In very young children (age ≤3 years, n=170), only 19% (33 of 170) received a TBI regimen. Bone marrow was the predominant graft for non-TBI regimen transplants (48%). In vivo Tcell depletion with anti-thymocyte globulin was common with non-TBI regimens accounting for 52% of transplantations compared to only 11% with TBI-containing regimens. The predominant GvHD prophylaxis with TBI-containing regimens was cyclosporine with mycophenolate and for non-TBI regimen, tacrolimus or cyclosporine with methotrexate. There were no differences between treatment groups regarding performance score, hematopoietic co-morbidity index, sites at diagnosis and cytogenetic risk. Most transplant centers used both TBI-containing and nonhaematologica | 2021; 106(7)

TBI versus non-TBI regimens for de novo AML

TBI regimens (n=57) or non-TBI regimens alone (n=53). Only 14 centers used TBI-containing regimens alone. Further, between 2012 and 2016, only a third of transplants used TBI-containing regimens. Non-TBI regimens were equally likely to be used between 2008 and 2011 and between 2012 and 2016. The median follow-up of patients who received TBI-containing regimens was 63 months (range, 3-122 months) and for those who received non-TBI regimens, 50 months (range, 3–122 months). Table 1. Patient, disease and transplant characteristics.

Variable Sex Male/female

TBI- regimens

Non-TBI regimens

101 (51%) / 98 (49%)

217 (51%) / 208 (49%)

P-value 0.94

Age ≤ 3 years 33 (17%) 4 – 10 years 49 (25%) 11 – 21 years 117 (59%) Performance score 90 - 100 168 (84%) ≤ 80 26 (13%) Not reported 5 ( 3%) HCT co-morbidity index ≤2 172 (86%) ≥3 25 (13%) Not reported 2 ( 1%) Site(s) at diagnosis Bone marrow only 132 (66%) Bone marrow + central 52 (26%) nervous system Bone marrow + other sites 12 ( 6%) Not reported 3 ( 2%) Cytogenetic risk Favorable 29 (15%) Intermediate 123 (62%) Poor 41 (21%) Not reported 6 ( 3%) Disease status at transplant 1st complete remission 107 (54%) 2nd complete remission 92 (46%) Donor HLA-matched sibling 17 ( 9%) HLA-matched unrelated donor 34 (17%) HLA-mismatched unrelated 14 ( 7%) donor HLA-matched unrelated 21 (11%) cord blood HLA-mismatched unrelated 101 (51%) cord blood Unrelated cord blood 12 ( 6%) (not reported) Graft versus host disease prophylaxis Tacrolimus + mycophenolate 7 ( 4%) Tacrolimus + methotrexate 40 (20%) Tacrolimus alone` 4 ( 2%) Cyclosporine + mycophenolate 119 (60%) Cyclosporine + methotrexate 20 (10%) Cyclosporine alone` 9 ( 5%) Anti-thymocyte globulin 22 (11%) Transplant period 2008 – 2011 133 (67%) 2012 – 2016 66 (33%) TBI: total body irradiation; HCT: hematopoietic cell transplant.

haematologica | 2021; 106(7)

<0.001 137 (32%) 93 (22%) 195 (46%) 0.06 376 (88%) 47 (11%) 2 (<1%) 0.55 375 (88%) 43 (10%) 7 ( 2%) 0.14 295 (69%) 84 (20%) 29 ( 7%) 17 ( 4%) 0.14 36 ( 8%) 285 (67%) 90 (21%) 14 ( 3%) 277 (65%) 148 (35%) 123 (29%) 109 (26%) 32 ( 8%) 29 ( 7%) 99 (23%) 33 ( 8%) <0.001

192 (45%) 233 (55%)

The median time to neutrophil recovery was 20 days and 17 days after TBI-containing and non-TBI regimens, respectively (P=0.04). The corresponding time to platelet recovery were 38 days and 30 days, P=0.002. Consequently, the day28 incidence of neutrophil recovery was lower after TBIcontaining (79%, 95% Confidence Interval [CI]: 73–84) compared to non-TBI regimens (85%, 95% CI: 82–88), P=0.04. Similarly, day-100 incidence of platelet recovery was also lower after TBI-containing (81%, 95% CI: 75–86) compared to non-TBI regimens (87%, 95% CI: 84–90), P=0.002.

Acute and chronic graft versus host disease Grade 2-3 acute GvHD risk was higher with TBI-containing compared to non-TBI regimens (Table 2). The day-100 incidence of grade 2-3 acute GvHD were 56% (95% CI: 49– 63) and 27% (95% CI: 22–30), respectively, P<0.0001. Compared to HLA-matched sibling donors, risks were higher with HLA-matched unrelated (hazard ratio [HR] 3.03, 95% CI: 1.75–5.25, P<0.0001), HLA-mismatched unrelated (HR 4.12, 95% CI: 2.18–7.77, P<0.0001), HLAmatched cord blood (HR 3.02, 95% CI: 1.44 –6.34, P=0.0035) and HLA-mismatched cord blood (HR 2.95, 95% CI: 1.57–5.56, P=0.0008). Grade 3-4 acute GvHD risk did not differ between the treatment groups (Table 2). Compared to bone marrow grafts, risk of acute GvHD was higher with peripheral blood (HR 3.22, 95% CI: 1.72–6.03, P=0.003) and cord blood (HR 2.24, 95% CI: 1.29–3.88, P=0.0041). Chronic GvHD risk also did not differ between treatment groups (Table 2). The 5-year incidence of chronic GvHD was 37% (95% CI: 30– 44) and 30% (95% CI: 26– 35) after TBI-containing and non-TBI regimens. Chronic GvHD risks were higher in patients aged 11–21 years compared to those aged ≤3 years (HR 1.78, 95% CI: 1,13–2.81, Table 2. Results of multivariate analysis

0.006

42 (10%) 139 (33%) 8 ( 2%) 77 (18%) 115 (27%) 44 (10% 221 (52%)

Hematopoietic recovery

<0.001 <0.001

Outcome Grade 2-4 acute GvHD* TBI-containing regimen Non-TBI regimen Grade 3-4 acute GvHD TBI-containing regimen Non-TBI regimen Chronic GvHD* TBI-containing regimen Non-TBI regimen Non-relapse mortality║ TBI-containing regimen Non-TBI regimen Relapse# TBI-containing regimen Non-TBI regimen Leukemia-free survival** TBI-containing regimen Non-TBI regimen Overall survival║ TBI-containing regimen Non-TBI regimen

P-value

Events/ Number

Hazard Ratio (95% CI)

109/196 112/420

1.00 0.44 (0.33 – 0.58)

<0.0001

34/196 53/420

1.00 0.69 (0.44 – 1.08)

0.10

70/198 116/422

1.00 0.82 (0.59 – 1.13)

0.23

42/199 46/425

1.00 0.53 (0.35 – 0.81)

0.003

43/199 149/425

1.00 1.46 (1.04 – 2.07)

0.03

85/199 195/425

1.00 1.01 (0.78 – 1.31)

0.95

73/199 159/425

1.00 0.98 (0.74 – 1.30)

0.91

GvHD: graft versus host disease; TBI. total body irradiation: CI: Confidence Interval; TBI: total body irradiation; *adjusted for age, donor type and GvHD prophylaxis; ║adjusted for age; ♯ adjusted for age and site(s) at diagnosis and **adjusted for age and cytogenetic risk.

1841

C.E. Dandoy et al.

P=0.01) and 4–10 years (HR 2.29, 95% CI: 1.57–3.33, P<0.0001). Chronic GvHD was higher with HLA-mismatched cord blood compared to HLA-matched sibling donors (HR 2.04, 95% CI: 1.10–3.67, P=0.02). Chronic GvHD risks did not differ between other donor groups (data not shown).

Non-relapse mortality and relapse Non-relapse mortality was higher with TBI-containing regimens (Table 2, Figure 1A). The 1- and 5-year incidence of non-relapse mortality with TBI-containing regimens were 17% (95% CI: 12–22) and 22% (95% CI: 16–28). The corresponding incidence with non-TBI regimens were 8% (95% CI: 6–11) and 11% (95% CI: 8–15). Compared to patients aged 4–10 years, non-relapse mortality was higher in patients aged 11–21 years (HR 1.99, 95% CI: 1.15–3.46, P=0.01) but not in those aged ≤3 years (HR 1.71, 95% CI: 0.85–3.44, P=0.13). Infections were higher with TBI-containing compared to non-TBI regimens (Table 3A). Venoocclusive disease was lower with TBI-containing regimens (Table 3A). Patients who survived at least 1 year after transplantation in remission were evaluable for organ dysfunction (Table 3B). Endocrine dysfunction (thyroid or gonadal) was higher with TBI-containing regimens. Pulmonary, cardiac and renal complications did not differ between treatment groups. Relapse risks were lower in TBI-containing regimens (Table 2, Figure 1B). The 1- and 5-year incidence of relapse with TBI-containing regimens were 15% (95% CI: 11–22) and 23% (95% CI: 17– 29). The corresponding relapse incidence with non-TBI regimens were 26% (95% CI: 22–31) and 37% (95% CI: 32–42), P<0.0001. Relapse risks did not differ between patients aged 4-10 and 11-21 years (HR 1.17, 95% CI: 0.82–1.69, P=0.39). Relapse was higher in patients aged ≤3 years compared to those aged 4-10 years (HR 2.49, 95% CI: 1.68–3.69, P<0.0001) and 11–21 years (HR 2.12, 95% CI: 1.51–2.98, P<0.0001). Compared to bone marrow and central nervous system involvement at diagnosis,

relapse risks were higher in patients with bone marrow involvement alone (HR 1.93, 95% CI: 1.27–2.93, P=0.002) and bone marrow with extramedullary site(s) excluding central nervous system involvement (HR 1.88, 95% CI: 1.01–3.50, P=0.04). Acute grade 2-4 GvHD was associated with lower relapse risk (HR 0.63, 95% CI: 0.44–0.89, P=0.008) but this was independent of conditioning regimen. The effect of acute grade 3-4 (HR 0.65, 95% CI: 0.39– 1.09, P=0.10) and chronic GvHD (HR 0.74, 95% CI: 0.48– 1.15, P=0.19) on relapse did not meet the level of significance that was set a priori.

Overall and leukemia-free survival There were no differences in overall or leukemia-free survival by treatment groups (Table 2, Figure 2A and B). Age was associated with both overall and leukemia-free survival and cytogenetic risk with leukemia-free survival. Compared to patients aged 4-10 years, survival was lower for those aged 11– 21 years (HR 1.82, 95% CI: 1.28–2.59, P<0.0001) and ≤3 years (HR 2.79, 95% CI: 1.90–4.10, P<0.0001). Survival was also lower in patients aged ≤3 years compared to those aged 11–21 years (HR 1.52, 95% CI: 1.14–2.08, P=0.005). Compared to favorable cytogenetics, leukemia-free survival was lower with intermediate risk (HR 1.85, 95% CI: 1.10–3.09, P=0.0198) and poor risk (HR 2.46, 95% CI: 1.42–4.27, P=0.0013). The 5-year overall survival was 61% (95% CI: 54–68) and 61% (95% CI: 56–66) after TBI-containing and non-TBI containing regimens. The corresponding leukemia-free survival was 53% (95% CI: 46–60) and 53% (95% CI: 48–58).

Transplant period As the current analysis included patients transplanted between 2008 and 2016, we tested for an effect of transplant period (2012-2016 vs. 2008-2011) on non-relapse mortality (HR 0.75, 95% CI 0.48 - 21.17, P=0.21), relapse (HR 1.04, 95% CI: 0.78–1.41, P=0.78), overall (HR 0.93, 95% CI: 0.71– 1.22, P=0.62) and leukemia-free survival (HR 0.96,

Figure 1. Non-relapse mortality and relapse. (A) Cumulative incidence of non-relapse mortality with total body irradiation (TBI)-containing and non-TBI regimens, (B) cumulative incidence of relapse with TBI-containing and non-TBI regimens.

1842

haematologica | 2021; 106(7)

TBI versus non-TBI regimens for de novo AML

95% CI: 0.75–1.23, P=0.15) and found none. We also examined for differences in infection rates by transplant period and observed lower rate of fungal infection between 2012 and 2016 compared to the earlier period in patients who received TBI-containing (2% vs. 8%) and non-TBI (1% vs. 4%) regimens.

0.50–1.12, P=0.16), non-relapse mortality (HR 1.09, 95% CI: 0.60–1.99, P=0.78), relapse (HR 1.07, 95% CI: 0.77–1.50, P=0.68), overall (HR 1.24, 95% CI: 0.90–1.72, P=0.19) and leukemia-free survival (HR 1.05, 95% CI: 0.78–1.40, P=0.75) and found none.

Subset analysis

Discussion

In a subset analysis limited to patients who received BuCy and Bu-Flu, we examined for an effect of anti-thymocyte globulin on grade 2-4 acute GvHD (HR 0.68, 95% CI: 0.44–1.03, P=0.07), grade 3-4 acute GvHD (HR 0.93, 95% CI: 0.51–1.70, P=0.82), chronic GvHD (HR 0.75, 95% CI:

To our knowledge this is the largest study to compare TBI and non-TBI intravenous Bu containing regimens in children and adolescents with de novo AML. Non-relapse mortality was higher with TBI regimens, and relapse was

Table 3A. Day-100 incidence of veno-occlusive disease and systemic infection.

Outcome VOD Bacterial bloodstream infection Viral bloodstream infection Fungal bloodstream infection

N eval 197 199 199 199

TBI-containing regimen Probability (95% CI)

N eval

8% (4-12) 47% (40-54) 43% (36-50) 6% (3-9)

422 425 425 425

Non-TBI regimen Probability (95% CI) 15% (11-18) 30% (26-35) 30% (26-34) 2% (1-4)

P-value 0.03 <0.001 0.001 0.04

TBI: total body irradiation; VOD: veno-occlusive disease; CI: Confidence Interval; eval: evaluated.

Table 3B. 3-year incidence of organ dysfunction in patients who were alive and in remission for at least 1-year post-transplant

Outcome Interstitial pneumonitis / Idiopathic pneumonia syndrome Endocrine dysfunction (gonadal or growth hormone) Cardiac failure, renal failure requiring dialysis

N eval

TBI-containing regimen Probability (95% CI)

N eval

Non-TBI regimen Probability (95% CI)

P-value

129

5% (2-9)

267

7% (4-11)

0.36

125

24% (17-32)

265

8% (5-12)

<0.001

129

5% (2-9)

267

3% (1-5)

0.26

Number of events: gonadal dysfunction TBI n=28 of 125; non-TBI n=29 of 265; growth hormone deficiency TBI n=14 of 129, non-TBI n=7 of 265; cardiac failure TBI n=3 of 129, non-TBI n=1 of 267; renal failure TBI n=4 of 129, non-TBI N=7 of 267. TBI: total body irradiation; CI: Confidence Interval; n: number; eval: evaluated.

Figure 2. Leukemia-free survival and overall survival. (A) Probability of overall survival with total body irradiation (TBI)-containing and non-TBI regimens, (B) probability of leukemia-free survival with TBI-containing and non-TBI regimens.

haematologica | 2021; 106(7)

1843

C.E. Dandoy et al.

higher with non-TBI regimens, negating an advantage for overall or leukemia-free survival. The net contribution of non-relapse mortality or relapse for either treatment group was not sufficient to lead towards an overall or leukemia-free survival advantage. Our findings are in keeping with another pediatric study from Japan that also failed to show differences in overall and leukemia-free survival between TBI-containing and non-TBI Bu regimens.1 We hypothesize there are several factors that influenced relapse risks including acute grade 2-4 GvHD. TBI regimens were largely used with cord blood transplants and TBI-Cy-Flu was the predominant regimen. Others have reported lower relapse with TBI-Cy-Flu regimen compared to other TBI- and non-TBI containing regimens for cord blood transplant.25 The higher incidence of bacterial, viral and fungal infections with the TBI-containing regimens within the first 3 months after transplantation likely contributed to early transplant-related mortality. Whether this is an effect of the conditioning regimen or the type of donor is challenging to differentiate as TBI regimens were predominantly used for cord blood transplants. In the subset, limited to transplants between 2012 and 2016, the incidence of bacterial and viral infections was also higher with TBI-regimens and consistent with the main analysis. However, the incidence of invasive fungal infection decreased to 2% with TBI regimens and 1% with non-TBI regimens for transplants between 2012 and 2016 (P=0.89) although this had negligible effect on non-relapse mortality (HR 1.18, P=0.72). A higher 5-year overall survival recorded with TBI-Cy-Flu compared to non-TBI regimens may be acceptable for some considering cord blood transplant even though growth hormone and gonadal deficiency is higher with TBI-Cy-Flu regimen.25 For transplantations with HLA-matched sibling or adult unrelated donors intravenous Bu-Cy or Bu-Flu is preferred.26,27 A recent study from the European Society for Blood and Marrow Transplant (EBMT) observed lower relapse and higher leukemia-free survival for AML in first complete remission with Bu-Cy-melphalan compared to Bu-Cy and TBI-Cy.28 Our study did not include the Bu-Cy-melphalan regimen. Hematopoietic recovery was lower with TBI-containing regimens. We hypothesize the lower recovery rates are in part explained by the predominant use of umbilical cord blood graft with TBI-containing regimens and in part, by use of intravenous Bu for all patients and pharmacokinetic data available for 80% of patients in the non-TBI group. Higher neutrophil but not platelet recovery with intravenous Bu containing regimens compared to TBI-containing regimens has been reported in adults with acute leukemia.10 Consistent with other reports, TBI-containing regimens were associated with higher incidence of thyroid and growth hormone deficiency compared to non-TBI regimens.21,22 Although not studied in the current analysis, others have recorded higher risk of cataracts, neuropsychological and cognitive abnormalities with TBI-containing regimens.22-24 The 5-year incidence of cardiac failure and renal failure were modest (<10%) but did not differ between treatment groups. Although not the focus of the current study, two recent publications studied the association between myeloablative conditioning regimens and second neoplasm. Those reports did not record a higher risk with TBI-containing compared to non-TBI regimens.27,29

1844

There are limitations to studying the effect of transplant conditioning regimen in a retrospective cohort. First, we do not know the factors that influenced choice of conditioning regimen other than in the youngest age group (≤3 years), approximately 70% of those who received TBI regimen were in second complete remission. Although we performed a carefully controlled analyses there may be unknown or unmeasured factors that may have influenced the outcomes recorded. Second, over the course of the study, effective molecular flow cytometric measures of detectable disease in patients in complete remission at transplantation may have helped refine prognosis after transplantation30 although it can be argued that the effect of minimal residual disease (MRD) would be consistent across both treatment groups. Among 166 patients for whom MRD status was available, 6 of 33 (18%) patients who received TBI and 17 of 133 (13%) patients who received non-TBI regimens were MRD negative at transplantation. Third, we know most patients who received Bu had pharmacokinetic dose adjustments, but we do not have data on dose adjustments to examine whether an increase or decrease to the prescribed Bu dose was associated with outcomes. Our study spanned a 9year period, a strength considering our sample size, but leukemia-free and overall survival may be influenced by transplant period. A careful analysis failed to find an effect of transplant period on outcomes other than a lower incidence of invasive fungal infection with TBI regimens. Our findings are relevant regarding a discussion on the choice of TBI-containing or non-TBI regimen when considering allogeneic transplantation for children and adolescents with de novo AML. In the absence of a survival advantage with either regimen group, the non-TBI regimens, Bu-Cy or Flu-Bu, are preferred although when considering umbilical cord blood transplantation TBI-Cy-Flu may be preferred.25 Disclosures No conflicts of interest to disclose. Contributions CED, SMD, KWA, AEK, JL and ME designed the study; SBS prepared the study file, KWA and YH analyzed the data; CED, SMD, KWA, YH and ME interpreted the results; CED drafted the manuscript; SMD, KWA, YH, AEK, JL, SB-S, HA-A, NB, JC, SG, NG, RH, NRL, JL, MM, TO, TP, AS, PS, LW and ME critically reviewed and edited the manuscript; all authors approved the final version. Funding The CIBMTR is supported primarily by Public Health Service Grant/Cooperative Agreement 5U24-CA076518 from the National Cancer Institute (NCI), the National Heart, Lung and Blood Institute (NHLBI) and the National Institute of Allergy and Infectious Diseases (NIAID); 5U10HL069294 from NHLBI and NCI; a contract HHSH250201200016C with Health Resources and Services Administration (HRSA/DHHS); grants N0001415-1-0848 and N00014-16-1-2020 from the Office of Naval Research. The views expressed in this article do not reflect the official policy or position of the National Institute of Health, the Department of the Navy, the Department of Defense, Health Resources and Services Administration (HRSA) or any other agency of the U.S. Government.

haematologica | 2021; 106(7)

TBI versus non-TBI regimens for de novo AML

References 1. Ishida H, Kato M, Kudo K, et al. Comparison of outcomes for pediatric patients with acute myeloid leukemia in remission and undergoing allogeneic hematopoietic cell transplantation with myeloablative conditioning regimens based on either intravenous busulfan or total body irradiation: a report from the Japanese Society for Hematopoietic Cell Transplantation. Biol Blood Marrow Transplant. 2015;21(12):2141-2147. 2. Kinnunen U, Koistinen P, Ohtonen P, et al. Influence of chemotherapy courses on the rate of bloodstream infections during neutropenia in adult acute myeloid leukaemia. Scand J Infect Dis. 2008;40(8):642-647. 3. Rubnitz JE. How I treat pediatric acute myeloid leukemia. Blood. 2012;119(25): 5980-5988. 4. Socié G, Clift RA, Blaise D, et al. Busulfan plus cyclophosphamide compared with total-body irradiation plus cyclophosphamide before marrow transplantation for myeloid leukemia: long-term follow-up of 4 randomized studies. Blood. 2001;98(13): 3569-3574. 5. Thomas ED, Buckner CD, Banaji M, et al. One hundred patients with acute leukemia treated by chemotherapy, total body irradiation, and allogeneic marrow transplantation. Blood. 1977;49(4):511-533. 6. Blaise D, Maraninchi D, Archimbaud E, et al. Allogeneic bone marrow transplantation for acute myeloid leukemia in first remission: a randomized trial of a busulfanCytoxan versus Cytoxan-total body irradiation as preparative regimen: a report from the Group d'Etudes de la Greffe de Moelle Osseuse. Blood. 1992;79(10):2578-2582. 7. Andersson BS, Kashyap A, Gian V, et al. Conditioning therapy with intravenous busulfan and cyclophosphamide (IV BuCy2) for hematologic malignancies prior to allogeneic stem cell transplantation: a phase II study. Biol Blood Marrow Transplant. 2002;8(3):145-154. 8. Pidala J, Kim J, Anasetti C, et al. Pharmacokinetic targeting of intravenous busulfan reduces conditioning regimen related toxicity following allogeneic hematopoietic cell transplantation for acute myelogenous leukemia. J Hematol Oncol. 2010;3:36. 9. Copelan EA, Hamilton BK, Avalos B, et al. Better leukemia-free and overall survival in AML in first remission following cyclophosphamide in combination with busulfan

haematologica | 2021; 106(7)

compared with TBI. Blood. 2013;122(24): 3863-3870. 10. Bredeson C, LeRademacher J, Kato K, et al. Prospective cohort study comparing intravenous busulfan to total body irradiation in hematopoietic cell transplantation. Blood. 2013;122(24):3871-3878. 11. Nakasone H, Fukuda T, Kanda J, et al. Impact of conditioning intensity and TBI on acute GVHD after hematopoietic cell transplantation. Bone Marrow Transplant. 2015;50(4):559-565. 12. Radhakrishnan K, Bishop J, Jin Z, et al. Risk factors associated with liver injury and impact of liver injury on transplantationrelated mortality in pediatric recipients of allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2013;19(6):912-917. 13. Ustun C, Young JH, Papanicolaou GA, et al. Bacterial blood stream infections (BSIs), particularly post-engraftment BSIs, are associated with increased mortality after allogeneic hematopoietic cell transplantation. Bone Marrow Transplant. 2019;54:1254-1265. 14. Dandoy CE, Ardura MI, Papanicolaou GA, et al. Bacterial bloodstream infections in the allogeneic hematopoietic cell transplant patient: new considerations for a persistent nemesis. Bone Marrow Transplant. 2017;52(8):1091-1106. 15. Freycon F, Casagranda L, Trombert-Paviot B. The impact of severe late-effects after 12 Gy fractionated total body irradiation and allogeneic stem cell transplantation for childhood leukemia (1988-2010). Pediatr Hematol Oncol. 2019;36(16):86-102. 16. Bresters D, Lawitschka A, Cugno C, et al. Incidence and severity of crucial late effects after allogeneic HSCT for malignancy under the age of 3 years: TBI is what really matters. Bone Marrow Transplant. 2016;51(11): 1482-1489. 17. Faraci M, Barra S, Cohen A, et al. Very late nonfatal consequences of fractionated TBI in children undergoing bone marrow transplant. Int J Radiat Oncol Biol Phys. 2005;63 (5):1568-1575. 18. Bhatia S, Davies SM, Scott Baker K, et al. NCI, NHLBI first international consensus conference on late effects after pediatric hematopoietic cell transplantation: etiology and pathogenesis of late effects after HCT performed in childhood--methodologic challenges. Biol Blood Marrow Transplant. 2011;17(10):1428-1435. 19. Przepiorka D, Weisdorf D, Martin P, et al. 1994 Consensus Conference on Acute GVHD Grading. Bone Marrow Transplant. 1995;15(6):825-828.

20. Lee SJ, Klein JP, Barrett AJ, et al. Severity of chronic graft-versus-host disease: association with treatment-related mortality and relapse. Blood. 2002;100(2):406-414. 21. Klein KP and Moeschberger ML. Survival analysis. techniques for censored and truncated data. Springer Science & Business Media. 2006. 22. Zhang X, Zhang MJ. SAS macros for estimation of direct adjusted cumulative Incidence curves under proportional subdistribution hazards models. Comput Methods Programs Biomed. 2011;101(1):87-93. 23. Zhang X, Loberiza FR, Klein JP, Zhang MJ. A SAS macro for estimation of direct adjusted survival curves based on a stratified Cox regression model. Comput Methods Programs Biomed . 2007;88(2):95-101. 24. Commenges D, Andersen PK. Score test of homogeneity for survival data. Lifetime Data Anal. 1995;1(2):145-156. 25. Eapen M, Kurtzberg J, Zhang MJ, et al. Umbilical cord blood transplantation in children with acute leukemia: impact of conditioning on transplantation outcomes. Biol Blood Marrow Transplant. 2017;23(10): 1714-1721. 26. Wilhelmsson M, Vatanen A, Borgström B, et al. Adverse health events and late mortality after pediatric allogeneic hematopoietic SCT-two decades of longitudinal follow-up. Bone Marrow Transplant. 2015;50(6):850857. 27. Vrooman LM, Millard HR, Brazauskas R, et al. Survival and late effects after allogeneic hematopoietic cell transplantation for hematologic malignancy at less than three years of age. Biol Blood Marrow Transplant. 2017;23(8):1327-1334. 28. Lucchini G, Labopin M, Beohou E, et al. Impact of conditioning regimen on outcomes for children with acute myeloid leukemia undergoing transplantation in first complete remission. An analysis on behalf of the Pediatric Disease Working Party of the European Group for Blood and Marrow Transplant. Biol Blood Marrow Transplant. 2017;23(3):467-474. 29. Lee CJ, Kim S, Tecca HR, et al. Late effects after ablative stem cell transplantation for adolescents and young adults with acute myeloid leukemia. Blood Adv. 2020;4(6): 983-992. 30. Araki D, Wood BL, Othus M, et al. Allogeneic hematopoietic cell transplantation for acute myeloid leukemia: time to move toward a minimal residual diseasebased definition of complete remission? J Clin Oncol. 2016;34(4):329-336.

1845

ARTICLE Ferrata Storti Foundation

Cell Therapy & Immunotherapy

Anti-RhD antibody therapy modulates human natural killer cell function

Shlomo Elias,1,2* Inbal Kol,1* Shira Kahlon,1 Rajaa Amore,3 Mariam Zeibak,3 Dror Mevorach,3 Uriel Elchalal,4 Orly Zelig2 and Ofer Mandelboim1

Haematologica 2021 Volume 106(7):1846-1856

1 The Concern Foundation Laboratories at the Lautenberg Center for Immunology and Cancer Research, Institute for Medical Research Israel Canada (IMRIC); 2Department of Hematology; 3Rheumatology Research Center, Department of Medicine and 4 Department of Obstetrics and Gynecology, Hadassah – Hebrew University Medical Center, Jerusalem, Israel.

* SE and IK contributed equally as co-first authors.

ABSTRACT

nti-RhD antibodies are widely used in clinical practice to prevent immunization against RhD, principally in hemolytic disease of the fetus and newborn. Intriguingly, this disease is induced by production of the very same antibodies when an RhD negative woman is pregnant with an RhD positive fetus. Despite over five decades of use, the mechanism of this treatment is, surprisingly, still unclear. Here we show that anti-RhD antibodies induce human natural killer (NK) cell degranulation. Mechanistically, we demonstrate that NK cell degranulation is mediated by binding of the Fc segment of anti-RhD antibodies to CD16, the main Fcγ receptor expressed on NK cells. We found that this CD16 activation is dependent upon glycosylation of the anti-RhD antibodies. Furthermore, we show that anti-RhD antibodies induce NK cell degranulation in vivo in patients who receive this treatment prophylactically. Finally, we demonstrate that the anti-RhD drug KamRho enhances the killing of dendritic cells. We suggest that this killing leads to reduced activation of adaptive immunity and may therefore affect the production of anti-RhD antibodies

Correspondence: OFER MANDELBOIM oferm@ekmd.huji.ac.il Received: September 13, 2019. Accepted: May 27, 2020. Pre-published: May 28, 2020. https://doi.org/10.3324/haematol.2019.238097

1846

Introduction Anti-RhD immunoglobulins are polyclonal antibodies which are commonly used in clinical practice. These antibodies are produced from human sera, and are mainly administered to prevent endogenous production of anti-RhD antibodies in case of exposure to the RhD antigen.1 This prophylactic treatment is commonly used in RhDwomen pregnant with an RhD+ fetus. As such, they are at risk of developing antiRhD antibodies which can cause hemolytic disease of the fetus and newborn (HDFN). Paradoxically, anti-RhD antibodies are also used as prophylactic treatment for this condition. Anti-RhD antibodies are administered not only as a preventive therapy, but can also be used for treating immune thrombocytopenic purpura,2 an autoimmune disease characterized by peripheral destruction of platelets. Despite over five decades of use, the mechanism behind anti-RhD’s effect remains unclear.3 Several hypotheses have been raised to explain the clinical impact of these antibodies, but none has been definetely proven.1-5 One of the leading notions is that anti-RhD antibodies can cause antibody-mediated immune suppression (AMIS), even though a mechanism to explain this has not been established yet. One of the possibilities to explain AMIS is that anti-RhD antibodies lead to macrophage-mediated destruction of RhD+ erythrocytes.6 A second theory hypothesizes that the antiRhD antibodies mask the RhD antigen on erythrocytes. Such masking could prevent the RhD antigen from being recognized by the immune system. It is estimated, however, that most of the RhD antigen sites remain unbound by the anti-RhD antibodies, and therefore should generate an immune response.4 Additional reported effects of these preparations include an increase of the cytokines transforming growth factor-β (TGF-β) and prostaglandin E2.7 It has also been suggested that anti-RhD antibodies might cross-link the B-cell receptor and the inhibitory fragment crystallizable receptor haematologica | 2021; 106(7)

anti-RhD antibodies modulate NK cell activity

(FcR) CD32B (FcγRIIb).4 However, Fcγ receptors have not been shown to play a role in AMIS.8 Since the clinical effect of anti-RhD antibodies implies that they convey some immune suppression, we wondered whether these preparations affect immune cells other than B cells. Here we concentrate on natural killer (NK) cells, which are innate lymphoid cells that play a significant role in eliminating virus-infected and malignant cells.9 NK-cell activity is governed by a balance of signals from a vast array of activating (e.g., CD16a, FcγRIIIa) and inhibitory receptors that are activated by self or foreign ligands. NK cells can also interact with dendritic cells (DC) and are able to kill DC in peripheral tissues, but this cytotoxic effect mostly affects immature DC (iDC).10 Notably, the killing of iDC by NK cells might attenuate adaptive immunity. 10 NK cells are also able to kill cells coated with antibody, a phenomenon known as antibody-dependent cellular cytotoxicity (ADCC). ADCC is mediated by the Fc fragment of antibodies which bind to CD16, the main FcR expressed on NK cells.11 Freshly isolated primary NK cells are mainly composed of a large, CD56dimCD16+ subpopulation (which expresses CD56 at intermediate levels and CD16 at high levels), and a much smaller, CD56brightCD16- subpopulation (which expresses CD56 at high levels and does not express CD16).12 Here we show that anti-RhD antibodies activate NK cells via binding of their Fc segment to CD16 in a glycosylationdependent manner. We show that this activation occurs not only in vitro, but also in vivo, in patients who receive this treatment. We further show that the anti-RhD drug KamRho enhances killing of iDC by NK cells and discuss how this might lead to immune suppression.

Methods Erythrocyte extraction and staining We used commercial erythrocytes of defined phenotype (cat. 004310, Bio-Rad) or erythrocytes extracted from whole blood samples from healthy volunteers with a known RhD expression profile. For extraction of erythrocytes from whole blood samples, the samples were centrifuged and the supernatant was discarded. Erythrocytes were then washed three times with phosphate buffered saline (PBSx1). For flow cytometry analysis, erythrocytes were incubated at 37°C for 15 minutes with antibodies, washed and then incubated with a secondary antibody at room temperature for 30 minutes. Details of other procedures related to erythrocytes are described in the Online Supplementary Appendix.

Natural killer cell purification and CD107a degranulation assay Primary NK cells were isolated from the peripheral blood of healthy human volunteers, and tested for purity as previously described.13 Unless stated otherwise, in all experiments we used NK cells which were activated as previously described.13 For the CD107a degranulation assay, primary NK cells were incubated with different polyclonal antibodies (described in the Online Supplementary Appendix) together with anti-CD107a and anti-CD56 antibodies. The NK and target cells were incubated at 37°C and 5% CO for 2 hours and analyzed by flow cytometry. In each well we used 50,000 NK cells and 0.5 µg of the relevant antibody in a final volume of 100 µL. Degranulation of NK cells was assessed by calculating the percent of CD107a+ NK cells out of the total NK cells in a given well. The data was normalized to the 2

haematologica | 2021; 106(7)

basal percent of CD107a+ NK cells (without addition of antibodies). The assay was performed in triplicate. For the mIgG2a blocking experiment, NK cells were pre-incubated for 30 minutes on ice with 5 µg per well of the mouse IgG2a isotype control.

Ethics The collection of patient samples was approved by the Institutional Review Board of The Hadassah Medical Center and informed consent was obtained from all participants (HMO-0091-18).

Human samples Two peripheral blood samples were collected from RhDwomen who received prophylactic anti-RhD treatment while hospitalized. After erythrocyte lysis, NK cell degranulation was quantified using flow cytometry as the percent of CD107a+ CD56+ CD3- cells out of CD56+CD3- cells. The assay was performed in triplicate.

Dendritic cell purification Monocyte-derived DC were generated from CD14+ peripheral blood mononuclear cell (PBMC), which were isolated as described previously.14 Briefly, anti-CD14 magnetic beads were used to isolate monocytes from PBMC according to the manufacturer's instructions (Miltenyi Biotech, Auburn, CA, USA). Monocytes were placed in wells at a concentration of 1.25×106 cells/1.5 mL culture medium.

Cytotoxicity assays The in vitro cytotoxic activity of NK cells was assessed in 5-hour S-release assays as previously described.13 DC were incubated for 48 hours with 20 ml of 35S- methionine prior to incubation with NK cells. K562, 721.221 or primary B cells were incubated for 24 hours with 4 ml of 35S- methionine, in methionine-free medium prior to incubation with NK cells. NK cells were pre-incubated for 2 hours at 37°C with different polyclonal antibodies (described in the Online Supplementary Appendix), at a concentration of 0.5 µg per well, diluted in RPMI medium. For the blocking experiment, NK cells were pre-incubated for 30 minutes on ice with 5 µg mIgG2a per well.

Results Anti-RhD preparations induce natural killer cell degranulation In order to investigate the effect of anti-RhD antibodies on NK cells, we used two common commercial preparations of anti-RhD antibodies: Rhophylac and KamRho. Both are produced from human sera and used clinically. In order to verify that these preparations specifically bind the RhD antigen, we first stained erythrocytes with known RhD expression. As expected, RhD+ erythrocytes were recognized by both anti-RhD products, in contrast to RhDerythrocytes (Figure 1A). Next, we tested whether the antiRhD preparations affect NK cell activation. We incubated activated human NK cells (which only contain CD56dimCD16+ cells, Online Supplementary Figure S1A) with anti-RhD antibodies and used flow cytometry to assess the level of CD107a, a marker of NK cell degranulation.15 Since anti-RhD preparations also contain non-specific human antibodies, we used intravenous immunoglobulin (IVIG) at the same concentration as the total antibody concentration of these preparations as a control. Neither of the IVIG products (Intratect and Kiovig) significantly affected the degranulation level of NK cells (Figure 1B). Both of the anti-RhD 1847

S. Elias et al.

preparations, however, caused significant activation of NK cells, demonstrated by a substantial increase in CD107a levels (Figure 1B). Primary NK cells obtained from eight different healthy donors displayed similar effects (Figure 1C). In order to corroborate these findings, we examined the degranulation level of primary activated NK cells incubated with increasing doses of the anti-RhD preparation KamRho, and observed a dose-dependent response (Figure

1D). Similar results were obtained using the anti-RhD preparation Rhophylac (Online Supplementary Figure S1B).

Natural killer cell degranulation is mediated specifically by anti-RhD antibodies Since both anti-RhD antibody mixtures contain mainly non-specific antibodies, we sought to investigate whether the degranulation effect is mediated specifically by anti-

Figure 1. Anti-RhD antibodies induce degranulation of natural killer cells. (A) Staining of erythrocytes (red blood cells [RBC]) with a pre-defined RhD expression profile using two commercial anti-RhD preparations: Rhophylac (left two panels) and KamRho (right two panels). Gray filled histograms represent the background staining with the secondary antibody. One representative experiment is shown out of three performed. (B) Degranulation of natural killer (NK) cells incubated with various commercial human polyclonal antibodies. NK-cell degranulation was assessed by calculating the percentage of CD107a+ NK cells out of the total NK cells (analyzed by flow cytometry). The percent of degranulated NK cells is indicated in each scatter plot. (C) The average NK-cell degranulation of eight different NK-cell donors after incubation with different polyclonal antibodies. The NK-cell degranulation level was calculated as in (B) and then normalized to the basal percent of CD107a+ NK cells without antibody (no Ab). ***P<0.001; ANOVA with Tukey’s HSD post-hoc test. Error bars represent standard error. (D) Degranulation of NK cells following incubation with increasing doses of KamRho. Error bars represent standard deviation of triplicates. One representative experiment is shown out of three performed.

1848

haematologica | 2021; 106(7)

anti-RhD antibodies modulate NK cell activity

haematologica | 2021; 106(7)

Figure 2. Specific anti-RhD antibodies mediate the degranulation of natural killer cells. (A to D) Preparation of the "unbound" antibody fractions and their effect on natural killer (NK)-cell degranulation. (A) Illustration of the production of unbound antiRhD antibodies. Anti-RhD antibodies were incubated with RhD(B) or RhD+ (C) erythrocytes and then the supernatant was collected and used for staining RhD- (left) and RhD+ (right) erythrocytes. (D) Degranulation of NK cells incubated with the unbound antibody fractions, prepared as illustrated in (A). The NK-cell degranulation level was normalized to the basal percent of CD107a+ NK cells. (E to H) Preparation of the "bound" antibody fractions and their effect on NK-cell degranulation. (E) Illustration of the production of bound anti-RhD antibodies. AntiRhD antibodies were incubated with RhD- (F) or RhD+ (G) erythrocytes. The eluates containing the bound antibody fractions were then extracted and used for staining RhD- (left) and RhD+ (right) erythrocytes. (H) Degranulation of NK cells incubated with the bound antibody fractions, prepared as illustrated in (E), and normalized as in (D). For all experiments, one representative experiment is shown out of at least two performed. ***P<0.001; Student's t-test. Error bars represent standard deviation of triplicates.

1849

S. Elias et al. A

Figure 3. Anti-RhD antibodies induce natural killer cell degranulation by binding to CD16. (A to B) Staining of transfected BW cells which express the extracellular domain of specific NK-cell receptors. The receptor expressed by each BW cell line is indicated above each histogram. (A) Staining with different antibodies as indicated below each histogram, in order to verify receptor expression; gray filled histograms represent staining with isotype control antibodies. (B) Staining with KamRho; gray filled histograms represent staining with secondary antibody and gray histograms represent staining of the parental BW cells. (C) Staining of BW-CD16 cells with intravenous immunoglobulin (IVIG) (gray filled histograms) or KamRho (black histograms), without (left panel) or with (right panel) blocking of CD16 by mouse IgG2a. (D) Staining of freshly isolated NK cells with the bound RhD+ fraction. Gating was on the CD56+ cell population (gray box). The y-axis presents the CD56 expression level; cells above the black line are CD56bright (CD16-) and those below it are CD56dim (CD16+). (E) Staining of bulk activated NK cells with the bound RhD+ fraction, with or without blocking of CD16 by mouse IgG2a. Gating was on all living NK cells. The figure shows the average of three experiments, performed on NK cells from three different donors. For each donor, the percent of stained NK cells in both conditions was normalized to the percent of stained NK cells without blocking. *P<0.05; paired student’s t-test. Error bars represent standard deviation. (F) Degranulation of NK cells incubated with KamRho or IVIG, with or without blocking of CD16 by mouse IgG2a. The NK-cell degranulation level was normalized to the basal percentage of CD107a+ NK cells without antibody (no Ab.). **P<0.01; Student's t-test. Error bars represent standard deviation of triplicates. (G) Primary NK cells were isolated from donors who express the low- (158F) or high-affinity (158V) variants of CD16. These NK cells were incubated with increasing doses of KamRho and then the NK-cell degranulation was assessed. The NK-cell degranulation level was normalized to the basal percent of CD107a+ NK cells. One representative experiment is shown out of two performed. ***P<0.001; Student's t-test. Error bars represent standard deviation of triplicates.

1850

haematologica | 2021; 106(7)

anti-RhD antibodies modulate NK cell activity

RhD antibodies. In order to address this, we incubated KamRho with RhD+ erythrocytes and collected the supernatant, which consists of antibodies that did not bind to erythrocytes ("unbound" fraction, Figure 2A). As a control, we performed the same adsorption process with RhD- erythrocytes, which are not expected to adsorb RhD-specific antibodies (Figure 1A). We first confirmed the efficiency of the process by staining RhD- and RhD+ erythrocytes with the two unbound antibody fractions. As expected, the unbound fraction from RhD- erythrocytes did not stain RhD- erythrocytes (Figure 2B, left panel), but did stain RhD+ erythrocytes (Figure 2B, right panel). In contrast, the unbound fraction from RhD+ erythrocytes no longer stained erythrocytes regardless of their RhD status (Figure 2C), thus demonstrating the efficiency of the adsorption process. We next examined the effect of the unbound fraction on NK-cell degranulation. We found that the unbound fraction from RhD- erythrocytes highly activated NK cells, in contrast to the unbound fraction from RhD+ (Figure 2D). In order to further corroborate the potency of specific antiRhD antibodies in inducing NK-cell degranulation, we eluted the specific anti-RhD antibodies bound to RhD+ erythrocytes ("bound" fraction, Figure 2E). As a control, we performed the same elution process with RhD- erythrocytes. We verified the elution process by staining RhD- and RhD+ erythrocytes with these bound fractions (Figure 2 F to G). Similar to the original antibody preparation, neither bound fraction stained RhD- erythrocytes (Figure 2 F to G, left panels). While the bound fraction from RhD- erythrocytes minimally stained RhD+ erythrocytes, the bound fraction from RhD+ erythrocytes, as expected, significantly stained RhD+ erythrocytes (Figure 2 F to G, right panels). Both bound fractions were then incubated with NK cells and we determined degranulation levels as before (Figure 2H). The bound fraction from RhD+ erythrocytes induced significant NK cell degranulation as compared to the bound fraction from RhD- erythrocytes (Figure 2H). Based on these results we concluded that it is the specific anti-RhD antibodies which induce NK-cell degranulation.

Anti-RhD antibodies induce natural killer cell degranulation by binding to CD16 We next explored how anti-RhD antibodies induce NKcell degranulation. Because NK cells do not express the RhD antigen,16 we focused on CD16, the main FcR expressed by NK cells, as we suspected it binds the Fc portion of the antiRhD antibodies. In order to test this possibility, we stained several transfected BW cells which express the extracellular domain of CD16 or of a control NK-cell receptor (NTB-A, NKp44 or DNAM-1). We first verified that each of the transfected BW cells expresses the extracellular domain of the given receptor (Figure 3A). We then stained these transfected BW cells with KamRho and analyzed the cells by flow cytometry (Figure 3B). We found that KamRho binds to CD16, but not to other tested NK-cell receptors (Figure 3B). Similar results were obtained with Rhophylac (Online Supplementary Figure S1C). Pre-blocking of CD16 with mouse IgG2a isotype, known to bind CD16 with high affinity,17 abolished the binding of KamRho to BW-CD16, as compared with IVIG binding (Figure 3C). In order to further explore the possibility that anti-RhD antibodies bind CD16, we double-stained primary freshly isolated NK cells using the bound RhD+ fraction we generated (shown in Figure 2 E, G to H), and an anti-CD56 antihaematologica | 2021; 106(7)

body. In concordance with our expectations, we observed binding of the anti-RhD antibodies to CD56dim NK cells that express CD16, but not to CD56bright CD16 negative NK cells (Figure 3D). The relatively low binding to CD56dim NK cells is probably due to limitations of staining human NK cells with human antibodies, as discussed below. Furthermore, blocking of CD16 on activated NK cells significantly reduced the binding of the bound RhD+ fraction to activated NK cells (Figure 3E). In order to corroborate these results, we repeated the staining of freshly isolated NK cells with a fluorochromeconjugated KamRho which we generated. This staining clearly demonstrated a specific binding to CD56dim (CD16+) NK cells (Online Supplementary Figure S2). Taken together, these findings indicate that anti-RhD antibodies bind CD16 on NK cells. Next, in order to determine whether anti-RhD – CD16 interaction is responsible for the NK-cell degranulation, we repeated the degranulation assay with pre-blocking of CD16 on NK cells. Pre-blocking significantly reduced the degranulation levels induced by anti-RhD antibodies (KamRho in Figure 3F and Rhophylac in the Online Supplementary Figure S3A). In order to further elucidate the role of CD16 in anti-

Figure 4. Anti-RhD antibodies induce degranulation of natural killer cells through their Fc segment. (A to B) Natural killer (NK)-cell degranulation assays. (A) F(ab') fragments were produced from KamRho antibodies. NK cells were incubated with equal concentrations of the F(ab') fragments or whole KamRho and intravenous immunoglobulin (IVIG). The NK-cell degranulation level was normalized to the basal percent of CD107a+ NK cells without antibody (no Ab.). (B) KamRho and IVIG were deglycosylated with PNGase under non-denaturing conditions. NK cells were incubated with equal concentrations of the deglycosylated or the original antibodies. The NK-cell degranulation level was normalized to the basal percent of CD107a+ NK cells without antibody (no Ab.). ***P<0.001; Student's t-test. Error bars represent standard deviation of triplicates. 2

1851

S. Elias et al.

RhD–induced NK-cell degranulation, we cultured primary NK cells from donors who are homozygous for either of the low- or high-affinity alleles of CD16 (158F and 158V respectively).18 We performed degranulation assays with both NK cells in parallel, which were incubated with KamRho (Figure 3G) or Rhophylac (Online Supplementary Figure S3B). We found significant differences in the extent of NK-cell degranulation between the donors (Figure 3G; Online Supplementary Figure S3B). We therefore concluded that anti-RhD antibodies induce degranulation of NK cells by binding to CD16.

Fc glycosylations of anti-RhD antibodies are essential for inducing natural killer cell degranulation As CD16 is an FcR, we wished to confirm that anti-RhD antibodies mediate their effect via their Fc segments. Therefore, we performed a degranulation assay with NK cells incubated with anti-RhD antibodies or F(ab') derivatives produced from KamRho and Rhophylac. As a control we used IVIG and F(ab') fragments of IVIG. In contrast to the degranulation induced by the whole anti-RhD antibodies, the F(ab')2 segments did not induce degranulation of NK cells (Figure 4A; Online Supplementary Figure S3C). This indicates that the Fc segment of anti-RhD antibodies is necessary for CD16-induced NK-cell degranulation. In order to understand why the Fc fragments of anti-RhD antibodies, but not those of IVIG, activate NK cells, we investigated which element of the Fc is important for this effect. Since the human IgG subclasses 1, 2, 3, and 4 differ in their immune properties,19 we initially quantified the abundance of these subclasses in KamRho and Rhophylac, as well as in IVIG. Although some differences could be observed, we did not detect significant enrichment of a specific IgG subclass in the anti-RhD preparations as compared to IVIG (Online Supplementary Table S1), indicating that the functional difference is not due to a specific IgG subclass. As Fc glycosylations have a critical influence on antibody function,20,21 we next examined the role of Fc glycosylations on anti-RhD-mediated NK-cell degranulation. We treated KamRho and Rhophylac with peptide:N-glycosidase F (PNGase F), an enzyme which cleaves N-linked oligosaccharides, and repeated the degranulation assay with the treated antibodies. Removal of glycosylations from KamRho and Rhophylac abolished their ability to induce degranulation of NK cells (Figure 4B; Online Supplementary Figure S3D, respectively), indicating that anti-RhD antibodies activate NK cells via glycosylation-dependent binding of their Fc segment to CD16. 2

Treatment with anti-RhD antibodies increases natural killer cell degranulation in humans Since anti-RhD preparations are widely used in clinical settings we wished to determine whether anti-RhD antibodies affect NK-cell degranulation in humans. We therefore collected peripheral blood samples from pregnant women who received prophylactic treatment with antiRhD. We collected two samples from each woman: the first sample was obtained just before administration of the anti-RhD injection, and the second sample was obtained 3 hours post-injection. Samples were stained and analyzed for CD107a levels by flow cytometry with gating on NK cells (CD56+CD3-). Two NK-cell populations were observed and referred to as PBL1 and PBL2 (Figure 5A and B). We initially gated on the PBL2 population, which were 1852

Figure 5. Treatment with anti-RhD antibodies increases natural killer cell degranulation in humans. (A-B) Gating strategy for assessment of natural killer (NK)-cell degranulation in human patients. Human peripheral blood (PB) samples were stained and analyzed by flow cytometry. (A) Size-based gating of two populations of cells (PBL1 and PBL2). (B) NK cells were identified as CD3-CD56+ cells, after size-based gating. (C) Summary of the effect of anti-RhD treatment on NK cell degranulation in 11 women. NK-cell populations PBL2 (left panel) and PBL1 (right panel) were stained for CD107a. For each sample (before and after), NK-cell degranulation was assessed by calculating the percentage of CD107a+ NK cells out of the total NK cells. The triplicate average for each condition (before/after) of each patient was then plotted and used for the statistical test. Each black line represents a single patient. *P<0.05, Wilcoxon signed-ranks test.

larger in size and had high baseline degranulation levels, presumably representing activated NK cells. In accordance with the in vitro degranulation assays, we noted a significant increase in the degranulation levels in almost all patients (10 of 11) with an average 2.06-fold increase in NK-cell degranulation after receiving anti-RhD (Figure 5C, left panel). Similar results were observed when gating on the PBL1 NK population (Figure 5 A to C, right panels), with an average 1.34-fold increase in NK-cell degranulation after anti-RhD treatment (Figure 5C, right panel).

The anti-RhD drug KamRho induces killing of immature and mature dendritic cells by natural killer cells Finally, we explored the possibility that anti-RhD-induced activation of NK cells helps to prevent production of antiRhD antibodies by B cells, the desired clinical outcome of this treatment. As NK cells can kill iDC in peripheral tissues, haematologica | 2021; 106(7)

anti-RhD antibodies modulate NK cell activity

Figure 6. The anti-RhD drug KamRho induces killing of immature dendritic cells by natural killer cells. (A) Cytotoxicity assay. 35S-labeled immature dendritic cells (iDC) were incubated with activated natural killer (NK) cells and different polyclonal antibodies (intravenous immunoglobulin [IVIG] or KamRho). The effector to target (E:T) ratio is indicated on the x-axis. One representative experiment is shown out of five performed. **P<0.01, ***P< 0.001; ANOVA with Tukey’s HSD post-hoc test. Error bars represent standard deviation of triplicates. (B) Cytotoxicity assay was performed as in (A), with 35S-labeled mature DC (mDC) cells used as targets. One representative experiment is shown out of two performed. *P<0.05, **P<0.01; ***P<0.001; ANOVA with Tukey’s HSD post-hoc test. Error bars represent standard deviation of triplicates. (C) Cytotoxicity assay was performed as in (A), with 35Slabeled iDC cells as targets, at an E:T ratio of 50:1. The assay was performed with pre-blocking of CD16 on NK cells using mouse IgG2a. One representative experiment is shown out of two performed. ns: non-significant; Student's t-test. Error bars represent standard deviation of triplicates. (D) Cytotoxicity assay. B lymphocytes, 721.221, or K562 cells were 35S-labeled and then incubated with activated NK cells and different polyclonal antibodies (IVIG or KamRho). The E:T ratio is 50:1 for B lymphocytes and 5:1 for 721.221 and K562 cells. One representative experiment is shown out of five performed with 721.221 and K562 cells, and two performed with B lymphocytes. ns: non-significant; Student's t-test. Error bars represent the standard deviation of triplicates.

haematologica | 2021; 106(7)

1853

S. Elias et al.

regulating adaptive immunity10, we tested whether antiRhD antibodies affect this process. We performed NK cytotoxicity assays, in which we measured the killing of mature DC (mDC) activated by lipopolysaccharide (LPS) and of iDC by NK cells which were pre-incubated with KamRho, Rhophylac or IVIG. The phenotype of iDC and mDC is presented in the Online Supplementary Figure S4A. At all effector to target ratios tested, pre-incubation with KamRho increased the killing of iDC by NK cells (Figure 6A). KamRho also increased the killing of mDC by NK cells, although the killing levels were significantly lower (Figure 6B), which is in line with previous reports.22-24 Pre-blocking of CD16 on NK cells with mouse IgG2a abolished the differences between KamRho and IVIG (Figure 6C). Interestingly, pre-incubation of NK cells with Rhophylac did not increase the killing of neither iDC nor mDC (Online Supplementary Figure S4B and C, respectively), although both drugs were able to induce degranulation of NK cells (Figure 1C). Because anti-RhD antibody preparations contain mainly non-specific antibodies, we next verified that the increased killing effect is specifically due to anti-RhD antibodies. We repeated the cytotoxicity assay, this time using the bound antibody fractions we eluted from RhD- and RhD+ erythrocytes incubated with KamRho (as shown in Figure 2 E to H). As expected, the bound fraction from RhD+ erythrocytes (which contains the specific anti-RhD antibodies) increased NK-cell-mediated killing of iDC significantly while the bound fraction obtained from RhD- erythrocytes did not (Online Supplementary Figure S4D). In order to check whether any other cells, particularly B cells, could also be more effciently killed in the presence of KamRho, we repeated the NK cytotoxicity assay with primary B cells isolated from human donors, the B-cell line 721.221 or the myeloid-derived cell line K562, and observed no effect (Figure 6D).

Discussion Anti-RhD Antibodies are widely used in clinical practice, mainly to prevent immunization against the RhD antigen. Surprisingly, although prophylactic administration of antiRhD antibodies is a common, safe, and successful treatment (in terms of preventing HDFN), the mechanism of this therapy remains unclear. One of the common theories suggests that this treatment can cause AMIS, though there is no single definitive mechanism for this suppression. One of the limitations of anti-RhD antibodies is that these products are manufactured from the sera of human subjects who are immunized against RhD. In light of the success of the preventive anti-RhD therapy, the number of immunized donors has decreased significantly.4 In order to overcome this drawback, in recent years several recombinant monoclonal antibodies against RhD have been generated.25-27 Some studies have demonstrated clinical efficiency of monoclonal anti-RhD antibodies in preventing RhD immunization, and have displayed encouraging laboratory markers (e.g., clearance of erythrocytes or antibody-dependent cytotoxicity).26-28 Nonetheless, polyclonal RhD antibodies are still considered the standard of care, which emphasizes that the precise mechanism of this treatment has not yet been elucidated. Deciphering this mechanism would potentially assist in generating efficient recombinant substitutions which would replace serum-based products. 1854

Here we explored the effect of two commercial anti-RhD antibody preparations on NK-cell activity. We showed that both products induce significant degranulation of NK cells, even in the absence of target cells, and that this is mediated specifically by anti-RhD antibodies. Importantly, we also observed an increase in NK-cell degranulation following prophylactic treatment with anti-RhD antibodies in human patients. We demonstrated that this effect occurs via binding of the Fc segment of the antibodies to CD16 in a glycosylation-dependent manner. It is possible that anti-RhD antibodies have unique glycosylations which are critical for their interaction with NK cells. This is consistent with a study which demonstrated that anti-RhD antibodies isolated from the plasma of alloimmunized pregnant women are less fucosylated.29 These properties should by explored in further studies. We demonstrated a significant binding of labeled antiRhD antibodies to the CD56dim, but not to the CD56bright NK-cell population of freshly isolated NK cells. However, when we used the non-labeled anti-RhD antibodies, little staining was observed. This is probably due to limitations of staining human NK cells by unlabeled human antibodies which requires anti-human secondary antibody staining. Since NK cells are coated with human immunoglobulins,30,31 the non-specific background staining with secondary antihuman antibodies is high, seemingly masking any specific binding. We also demonstrated the role of CD16 in the anti-RhD -mediated NK-cell degranulation in other experiments (e.g., degranulation with blocking of CD16 and with low and high-affinity CD16a-expressing donors). Blocking of CD16 abolished anti-RhD-mediated degranulation significantly, yet not completely. This incomplete blocking could result from inefficiency of the blocking or might indicate that additional NK-cell receptors are involved (even though we have not observed binding of anti-RhD antibodies to several other NK-cell receptors). We next found that the anti-RhD drug KamRho increases killing of iDC. KamRho also increased killing of mDC, but to a lesser extent. This is in line with previous reports of the sensitivity of iDC as compared to mDC to killing by NK cells, which is attributed to the increased levels of major histocompatibility complex (MHC) on mDC.22-24. Surprisingly, although the drugs Rhophylac and KamRho had similar effects in most in vitro assays performed, Rhophylac did not increase NK killing of DC. One possibility is that this difference is related to the presence of a different stabilizer in each of these drugs: albumin in Rhopylac and glycine in KamRho (glycine is also used as stabilizer in the IVIG regent we used as a control). Interestingly, increased NK-cell activation mediated by KamRho did not enhance killing of other cell types, including B cells. The reasons behind this observation are not completely understood. One possible explanation is that MHC class I proteins are involved. This option seems unlikely as K562 and 721.221 cells express little or no MHC class I, while primary B cells, iDC and to a greater extent mDC express high levels of MHC class I. Alternatively, it is possible that anti-RhD antibodies not only bind NK but can also bind DC through their Fab segment, therefore inducing ADCC. This possibility seems unlikely as the RhD antigen is expressed only on erythrocytes16. A final explanation might be that CD16 binds an unknown cellular ligand expressed on DC. Such a cellular ligand was previously haematologica | 2021; 106(7)

anti-RhD antibodies modulate NK cell activity

reported to be expressed on 1106mel cells.32 Further studies are required to decipher this differential killing by NK cells. A possible mechanism arising from our work is that antiRhD antibodies increase NK-cell activation, resulting in increased killing of iDC. This may, in turn, lead to decreased antigen presentation and a decreased humoral response. In the context of an RhD- individual exposed to RhD+ erythrocytes, this effect could diminish the production of anti-RhD antibodies, which is the clinical outcome observed upon anti-RhD prophylactic treatment. Previous studies have demonstrated that impaired killing of DC by cytotoxic T lymphocytes and NK cells can lead to continuous antigen presentation and exaggerated immune activation,33,34 as seen in hemophagocytic lymphohistiocytosis.35 We can therefore speculate that in the inverse case, activation of NK by anti-RhD antibodies can contribute to AMIS via killing of DC. If administration of the anti-RhD antibodies indeed leads to the killing of DC, a general immune suppression would be expected in these women. This should be investigated in the future in a clinical study. Although the mechanism we describe here is unrelated to ADCC of erythrocytes (anti-RhD-mediated lysis of erythrocytes by CD16 on NK cells), ADCC, if significant, could potentially also contribute to the mechanism of action of anti-RhD antibodies. Several studies have shown that NK cells are capable of lysing erythrocytes via ADCC (mainly in the context of Plasmodium falciparum infected erythrocytes)36. Only a small fraction, however, have researched this effect with human allo-antibodies, let alone anti-RhD. Furthermore, these studies rarely employ direct, reliable measurements of erythrocyte lysis.37 Therefore, this important question should be comprehensively addressed by future studies.

References 1. Brinc D, Lazarus AH. Mechanisms of anti-D action in the prevention of hemolytic disease of the fetus and newborn. Hematology Am Soc Hematol Educ Program. 2009:185-191. 2. Lazarus AH, Crow AR. Mechanism of action of IVIG and anti-D in ITP. Transfus Apher Sci. 2003;28(3):249-255. 3. Kumpel BM. On the immunologic basis of Rh immune globulin (anti-D) prophylaxis. Transfusion. 2006;46(9):1652-1656. 4. Kumpel BM. On the mechanism of tolerance to the Rh D antigen mediated by passive anti-D (Rh D prophylaxis). Immunol Lett. 2002;82(1-2):67-73. 5. Garratty G. The James Blundell Award Lecture 2007: do we really understand immune red cell destruction? Transfus Med. 2008;18(6):321-334. 6. Kumpel BM, Elson CJ. Mechanism of anti-Dmediated immune suppression--a paradox awaiting resolution? Trends Immunol. 2001;22(1):26-31. 7. Branch DR, Shabani F, Lund N, Denomme GA. Antenatal administration of Rh-immune globulin causes significant increases in the immunomodulatory cytokines transforming growth factor-beta and prostaglandin E2. Transfusion. 2006;46(8):1316-1322. 8. Bernardo L, Yu H, Amash A, Zimring JC, Lazarus AH. IgG-Mediated immune suppression to erythrocytes by polyclonal antibodies can occur in the absence of activating or inhibitory Fcgamma receptors in a full mouse model. J Immunol. 2015;195(5):2224-2230.

haematologica | 2021; 106(7)

In summary, we provide new insights into the activity of polyclonal anti-RhD antibodies which are in clinical use. These findings could contribute to the clinical efficiency of these preparations or imply that these antibodies carry unique features meriting further study. Disclosures No conflicts of interest to disclose. Contributions SE and IK designed and performed experiments, analyzed results, and wrote the paper; SK performed experiments; OZ provided reagents and contributed professional advice; RA, MZ and DM assisted in the dendrtic cell experiments; UE was responsible for the human samples and OM supervised the project. Acknowledgments The authors thank Yulia Gendler and Dr. Shulamit Metsger from the Hadassah Blood Bank (Jerusalem, Israel) for their fruitful discussions and assistance with the erythrocyte experiments, Suhair Abdeen from Hadassah Medical Center (Jerusalem, Israel) for the isotype analysis and Natan Stein from The Concern Foundation Laboratories at the Lautenberg Center for Immunology and Cancer Research (Jerusalem, Israel) for language editing. Funding This work was supported by the ISF-China program and by the ISF Moked grant. Further support came from the ICRF professorship grant, by the MOST-DKFZ grant, by the GIF grant. The study was also supported by the Israel Science Foundation (grant 502/15), the Kass Medical Research Award and the Israeli Society of Hematology and Transfusion Medicine research grant (to S.E).

9. Spits H, Bernink JH, Lanier L. NK cells and type 1 innate lymphoid cells: partners in host defense. Nat Immunol. 2016;17(7):758-764. 10. Moretta A. Natural killer cells and dendritic cells: rendezvous in abused tissues. Nat Rev Immunol. 2002;2(12):957-964. 11. Bournazos S, Ravetch JV. Fcgamma receptor function and the design of vaccination strategies. Immunity. 2017;47(2):224-233. 12. Fehniger TA, Cooper MA, Nuovo GJ, et al. CD56bright natural killer cells are present in human lymph nodes and are activated by T cell-derived IL-2: a potential new link between adaptive and innate immunity. Blood. 2003;101(8):3052-3057. 13. Yamin R, Lecker LS, Weisblum Y, et al. HCMV vCXCL1 binds several chemokine receptors and preferentially attracts neutrophils over NK cells by interacting with CXCR2. Cell Rep. 2016;15(7):1542-1553. 14. Amarilyo G, Verbovetski I, Atallah M, et al. iC3b-opsonized apoptotic cells mediate a distinct anti-inflammatory response and transcriptional NF-kappaB-dependent blockade. Eur J Immunol. 2010;40(3):699-709. 15. Alter G, Malenfant JM, Altfeld M. CD107a as a functional marker for the identification of natural killer cell activity. J Immunol Methods. 2004;294(1-2):15-22. 16. Iwamoto S, Omi T, Yamasaki M, Okuda H, Kawano M, Kajii E. Identification of 5' flanking sequence of RH50 gene and the core region for erythroid-specific expression. Biochem Biophys Res Commun. 1998; 243(1):233-240. 17. Jiang N, Chen W, Jothikumar P, et al. Effects

of anchor structure and glycosylation of Fcgamma receptor III on ligand binding affinity. Mol Biol Cell. 2016;27(22):3449-3458. 18. Koene HR, Kleijer M, Algra J, Roos D, von dem Borne AE, de Haas M. Fc gammaRIIIa158V/F polymorphism influences the binding of IgG by natural killer cell Fc gammaRIIIa, independently of the Fc gammaRIIIa-48L/R/H phenotype. Blood. 1997;90(3):1109-1114. 19. Irani V, Guy AJ, Andrew D, Beeson JG, Ramsland PA, Richards JS. Molecular properties of human IgG subclasses and their implications for designing therapeutic monoclonal antibodies against infectious diseases. Mol Immunol. 2015;67(2 Pt A):171-182. 20. Jennewein MF, Alter G. The immunoregulatory roles of antibody glycosylation. Trends Immunol. 2017;38(5):358-372. 21. Seeling M, Bruckner C, Nimmerjahn F. Differential antibody glycosylation in autoimmunity: sweet biomarker or modulator of disease activity? Nat Rev Rheumatol. 2017;13(10):621-630. 22. Ferlazzo G, Semino C, Melioli G. HLA class I molecule expression is up-regulated during maturation of dendritic cells, protecting them from natural killer cell-mediated lysis. Immunol Lett. 2001;76(1):37-41. 23. Walzer T, Dalod M, Robbins SH, Zitvogel L, Vivier E. Natural-killer cells and dendritic cells: "l'union fait la force". Blood. 2005;106(7):2252-2258. 24. Ferlazzo G, Tsang ML, Moretta L, Melioli G, Steinman RM, Munz C. Human dendritic cells activate resting natural killer (NK) cells

1855

S. Elias et al. and are recognized via the NKp30 receptor by activated NK cells. J Exp Med. 2002;195(3):343-351. 25. Miescher S, Zahn-Zabal M, De Jesus M, et al. CHO expression of a novel human recombinant IgG1 anti-RhD antibody isolated by phage display. Br J Haematol. 2000;111(1):157-166. 26. Kumpel BM, Goodrick MJ, Pamphilon DH, et al. Human Rh D monoclonal antibodies (BRAD-3 and BRAD-5) cause accelerated clearance of Rh D+ red blood cells and suppression of Rh D immunization in Rh D- volunteers. Blood. 1995;86(5):1701-1709. 27. Miescher S, Spycher MO, Amstutz H, et al. A single recombinant anti-RhD IgG prevents RhD immunization: association of RhD-positive red blood cell clearance rate with polymorphisms in the FcgammaRIIA and FcgammaIIIA genes. Blood. 2004;103(11): 4028-4035. 28. Beliard R, Waegemans T, Notelet D, et al. A human anti-D monoclonal antibody selected

1856

for enhanced FcgammaRIII engagement clears RhD+ autologous red cells in human volunteers as efficiently as polyclonal anti-D antibodies. Br J Haematol. 2008;141(1):109-119. 29. Kapur R, Della Valle L, Sonneveld M, et al. Low anti-RhD IgG-Fc-fucosylation in pregnancy: a new variable predicting severity in haemolytic disease of the fetus and newborn. Br J Haematol. 2014;166(6):936-945. 30. Sulica A, Herberman RB. Cytophilic immunoglobulins revisited via natural killer cells. FASEB J. 1996;10(13):1495-1504. 31. Sulica A, Galatiuc C, Manciulea M, et al. Regulation of human natural cytotoxicity by IgG. IV. Association between binding of monomeric IgG to the Fc receptors on large granular lymphocytes and inhibition of natural killer (NK) cell activity. Cell Immunol. 1993;147(2):397-410. 32. Mandelboim O, Malik P, Davis DM, Jo CH, Boyson JE, Strominger JL. Human CD16 as a lysis receptor mediating direct natural killer cell cytotoxicity. Proc Natl Acad Sci U S A.

1999;96(10):5640-5644. 33. Terrell CE, Jordan MB. Perforin deficiency impairs a critical immunoregulatory loop involving murine CD8(+) T cells and dendritic cells. Blood. 2013;121(26):5184-5191. 34. Yang J, Huck SP, McHugh RS, Hermans IF, Ronchese F. Perforin-dependent elimination of dendritic cells regulates the expansion of antigen-specific CD8+ T cells in vivo. Proc Natl Acad Sci U S A. 2006;103(1):147-152. 35. Manciulea M, Rabinowich H, Sulica A, et al. Divergent phosphotyrosine signaling via Fc gamma RIIIA on human NK cells. Cell Immunol. 1996;167(1):63-71. 36. Arora G, Hart GT, Manzella-Lapeira J, et al. NK cells inhibit Plasmodium falciparum growth in red blood cells via antibodydependent cellular cytotoxicity. Elife. 2018;7:e36806. 37. Urbaniak SJ. ADCC (K-cell) lysis of human erythrocytes sensitized with rhesus alloantibodies. II. Investigation into the mechanism of lysis. Br J Haematol. 1979;42(2):315-325.

haematologica | 2021; 106(7)

ARTICLE

Cell Therapy & Immunotherapy

Fc-engineering significantly improves the recruitment of immune effector cells by anti-ICAM-1 antibody MSH-TP15 for myeloma therapy

Ferrata Storti Foundation

Katja Klausz,1 Michael Cieker,1 Christian Kellner,2 Thies Rösner,1 Anna Otte,1 Steffen Krohn,1 Anja Lux,3 Falk Nimmerjahn,3 Thomas Valerius,1 Martin Gramatzki1 and Matthias Peipp1

Division of Stem Cell Transplantation and Immunotherapy, Department of Internal Medicine II, University Hospital Schleswig-Holstein and Christian-Albrechts-University, Kiel; 2Department of Transfusion Medicine, Cell Therapeutics and Hemostaseology, University Hospital, LMU Munich, Munich and 3Institute of Genetics, Department of Biology, University of Erlangen-Nürnberg, Erlangen, Germany 1

Haematologica 2021 Volume 106(7):1857-1866

ABSTRACT

espite several therapeutic advances, patients with multiple myeloma (MM) require additional treatment options since no curative therapy exists yet. In search of a novel therapeutic antibody, we previously applied phage display with myeloma cell screening and developed TP15, a single-chain fragment variable targeting intercellular adhesion molecule 1 (ICAM-1/CD54). In order to more precisely evaluate the antibody’s modes of action, fully human immunoglobulin G1 antibody variants were generated bearing the wild-type (MSH-TP15) or mutated fragment crystallizable (Fc-engineered [Fc-eng.]) region to either enhance (MSH-TP15 Fc-eng.) or prevent (MSH-TP15 Fc knockout [Fc k.o.]) Fcγ receptor binding. Especially MSH-TP15 Fc-eng. induced significant antibody-dependent cell-mediated cytotoxicity against malignant plasma cells by recruiting natural killer cells and engaged macrophages for antibody-dependent cellular phagocytosis of tumor cells. Binding studies with truncated ICAM-1 demonstrated MSHTP15 binding to ICAM-1 domain 1-2. Importantly, MSH-TP15 and MSHTP15 Fc-eng. both prevented myeloma cell engraftment and significantly prolonged survival of mice in an intraperitoneal xenograft model. In the subcutaneous model MSH-TP15 Fc-eng. was superior to MSH-TP15, whereas MSH-TP15 Fc k.o. was not effective in either of the models – reflecting the importance of Fc-dependent mechanisms of action also in vivo. The efficient recruitment of immune cells and the observed anti-tumor activity of the Fcengineered MSH-TP15 antibody hold significant potential for myeloma immunotherapy.

Correspondence: MATTHIAS PEIPP m.peipp@med2.uni-kiel.de Received: March 11, 2020. Accepted: May 28, 2020. Pre-published: June 4, 2020.

Introduction https://doi.org/10.3324/haematol.2020.251371 With approval of the monoclonal antibodies (mAb) daratumumab and elotuzumab in 2015, antibody-based immunotherapy has entered clinical practice for multiple myeloma (MM) patients. While both mAb show therapeutic activity in combination with standard treatment regimens, only daratumumab has significant singleagent activity and is additionally approved as front-line therapy. Increased survival rates and good tolerability are achieved with both mAb, but still not all patients benefit and first drug resistances have been reported.1,2 Thus, there is still a high unmet medical need for myeloma patients. For homing and growth of malignant plasma cells in the bone marrow (BM), adhesion molecules are of significant importance and ‘hiding’ tumor cells in the BM are believed to induce relapse in myeloma patients.3,4 Therefore, targeting the BM microenvironment and adhesion molecules might be explicitly reasonable for MM therapy.5 For instance, intercellular adhesion molecule-1 (ICAM-1/CD54) is reported to be highly expressed on malignant plasma cells and not down-regulated under MM therapy.6 Of note, significantly higher levels of ICAM-1 are expressed on haematologica | 2021; 106(7)

1857

K. Klausz et al.

malignant plasma cells from MM patients treated with chemotherapy and especially in tumors with a multi-drug resistance phenotype.7 This is in line with the observation that adhesion molecules such as ICAM-1 are involved in macrophage-induced drug resistance and tumor escape in myeloma.8 In contrast, expression of ICAM-1 in healthy tissue, e.g., on endothelial cells and leukocytes, is constitutively low and increases only after stimulation with cytokines, i.e., interferon (IFN)-γ or interleukin (IL)-1β, which are up-regulated and released during inflammation and infection.9 Clinical trials with anti-ICAM-1 antibodies in patients with rheumatoid arthritis, renal transplants, stroke and myeloma proved safety and tolerability, but often lacked significant activity.10-13 Most therapeutic antibodies used for cancer immunotherapy are of immunoglobulin (Ig) G1 isotype and exert their anti-tumor activity via antigen-binding fragment (Fab)- and/or fragment crystallizable (Fc)-mediated effector functions directly through antigen binding and/or interactions with the immune system.14 The Fc-mediated effector functions encompass antibodydependent cell-mediated cytotoxicity (ADCC), antibodydependent cellular phagocytosis (ADCP) and complement-dependent cytotoxicity (CDC). Over the years many efforts have been made to improve Fc-mediated effector functions of mAb to increase their cytotoxic and therapeutic activity by Fc glyco- and Fc protein-engineering techniques. Fc protein-engineering approaches make use of specific amino acid substitutions in the Fcγ receptors (FcγR) and C1q binding interfaces of a therapeutic antibody to specifically improve ADCC, ADCP and/or CDC activity.15-19 Lazar and colleagues identified the S239D and I332E (DE) mutations in the CH2 domain of IgG1 antibodies, which markedly enhanced binding to the activating FcγRIIIa/CD16a and ADCC activity of a variety of therapeutic IgG1 antibodies.15 The DE mutations have been applied to generate an Fc-engineered (Fc-eng.) CD19 antibody that demonstrates promising therapeutic activity in clinical trials for relapsed/refractory B-cell non-Hodgkin lymphoma.20 Margetuximab, an Fc-optimized anti-HER2 antibody with reduced binding to FcγRIIb/CD32b and enhanced affinity for FcγRIIIa/CD16a,21 is currently evaluated in a phase III study for breast cancer and also highlights the potential of Fc-engineering as a strategy to improve antibodies for cancer immunotherapy. Here, we applied Fc-engineering to MSH-TP15, a novel, fully human anti-ICAM-1 IgG1 antibody whose variable regions are derived from TP15-Fc originally isolated by phage display and screening of myeloma cells.22 Fab- and Fc-mediated effector functions of MSH-TP15 were analyzed by comparing three antibody variants that differed in their affinity for FcγR’s on immune cells. Antigen binding, direct anti-tumor effects and immune cell recruitment for tumor cell lysis by the MSH-TP15 antibody variants were tested in vitro and in vivo against myeloma cells.

Methods Cell separation Mononuclear cells from peripheral blood (PBMC), pleural effusion (PE) and BM of myeloma patients and healthy donors were isolated as previously described.23 Samples were taken after receiving donors’ written informed consents. Experiments were in accordance with the Declaration of Helsinki and approved by the 1858

Ethics Committee of the Christian-Albrechts-University, Kiel, Germany (D442/10).

Cell lines Ramos, CHO-K1, L363 and U266 were obtained from DSMZ (Braunschweig, Germany), Lenti-X 293T were purchased from Takara Bio (Göteborg, Sweden), MM1.S, Raji CS and CHOFcγRIIa-131H were gifts from YT Tai (Boston, MA, USA), MJ Glennie (Southampton, UK), and F Nimmerjahn (Erlangen, Germany),24 respectively, and cultured as recommended. INA-6, INA-6.Tu1 and BHK-FcγRIIIa-158V were established in our laboratory and cultured as previously described.25,26

Production of MSH-TP15 antibody variants Variable light (VL) and heavy chain (VH) sequences derived from phage PIII-1522 and BI-5056,27 were cloned into modified pSecTag2/HygroC vectors harboring sequences for the constant region of the human kappa light chain (LC) or either wild-type (wt) or mutated human IgG1 heavy chain (HC).28 Antibodies were termed MSH-TP15 (wt IgG1), MSH-TP15 Fc-engineered (Fc-eng.) and MSH-TP15 Fc knockout (Fc k.o.) and produced as described in the Online Supplementary Appendix.

Flow cytometric analyses Immunofluorescence analyses to investigate mAb binding, induction of programmed cell death (PCD) and ADCP were performed on a Navios flow cytometer and analyzed with Kaluza software (Beckman Coulter). For details please refer to the Online Supplementary Appendix.

Co-culture experiments BM stromal cells (BMSC) were isolated from MM patient BM aspirates and used for co-culture experiments with INA-6 cells as previously described.29

Cytotoxicity assays ADCC activity was analyzed in 3 hour (h) 51Cr release assays using PBMC and natural killer (NK) cells at E:T ratios of 80:1 and 10:1, respectively, as previously described.23

Live cell imaging INA-6 killing was analyzed over 24 h with IncuCyte caspase3/7 reagent (250 nM) in the presence or absence of NK cells (E:T ratio 10:1) and 20 mg/mL of the indicated antibodies with IncuCyte live cell imaging (Essen Bioscience). Green cell counts per image were analyzed with Zoom2016A software.

Animal experiments All animal experiments were performed according to the guidelines of the Christian-Albrechts-University Kiel along with the German Animal Protection Law. For both models 7-8 week-old female SCID/beige mice (Charles River, Sulzfeld, Germany) and red fluorescent protein-expressing INA-6.Tu1 cells (INA6.Tu1_red; unpublished) were used. In the intraperitoneal (i.p.) model, ten mice/group were injected with 20x106 INA-6.Tu1_red cells 48 h prior start of twice weekly i.p. treatment with mAb (initial doses 10 mg/kg followed by six doses of 5 mg/kg mAb/animal) or vehicle control. In the established tumor model 5x106 INA6.Tu1_red cells were subcutaneously (s.c.) injected into the right hind flank of five mice/group. At day 10, i.p. treatment with five doses of 10, 1 or 0.1 mg/kg mAb/mouse twice weekly was started. Tumor growth was measured by caliper and volume was calculated: (length x width2)/2. Human IL-6 receptor was analyzed with CD126 enzyme-linked immunosorbent assay (ELISA) kit (Diaclone, Besançon, France). haematologica | 2021; 106(7)

Fc-engineered CD54 antibody MSH-TP15 for myeloma therapy

Figure 1. Generation and binding characteristics of MSH-TP15 antibody variants. (A) The variable heavy (VH) (dark blue) and variable light (VL) (light blue) sin-glechain fragment variable (scFv) sequences of phage antibody PIII-15 were used to generate three human immunoglobuin (Ig) G1 MSH-TP15 monoclonal antibody (mAb) variants carrying either a wild-type (middle), a protein-engineered (Fc-eng.; top) or a knockout (k.o.; bottom) fragment crystallizable (Fc)-domain. Mutations in the constant region (grey) of the heavy chain (HC) are depicted in yellow, glycosylation is shown in light grey. (B) Purity and molecular masses of the produced antibodies was analyzed by sodium dodecyl sulfate polyacrylamide-based discontinuous gel electrophoresis (SDS-PAGE) and Coomassie staining under reducing (red.) and non-reducing (n.r.) conditions. ADCC: antibody-dependent cell-mediated cytotoxicity; L: molecular mass ladder; LC: light chain. (C) Intercellular adhesion molecule-1 (ICAM-1) binding was measured by flow cytometry using L363 myeloma cells and increasing antibody concentrations. CD20 antibody rituximab served as control (ctrl IgG1). ***P<0.001 MSH-TP15 mAb vs. ctrl IgG1. (D) Binding to human Fcγ receptor (FcγR) was analyzed with CHO cells expressing human FcγRIIa-131H or (E) BHK cells expressing either the low (FcγRIIIa-158F) or the high affinity (FcγRIIIa-158V) allelic form of human CD16a. *P<0.05 MSH-TP15/MSH-TP15 Fc-eng. vs. MSH-TP15 Fc k.o.; MFI: mean fluorescence intensity. Data represent mean values ± standard error of the mean of three independent experiments.

Data processing and statistical analyses Data were statistically analyzed with GraphPad Prism Software using appropriate tests (San Diego, CA, USA). Significance was accepted with P<0.05.

Results Generation of MSH-TP15 IgG1 monoclonal antibodies with different FcγR binding properties VL and VH sequences of scFv-Fc antibody PIII-15, previously selected by phage display and panning with human myeloma cell lines,22 were used to generate three fully human IgG1κ antibody variants. Compared to the wt IgG1 MSH-TP15, the Fc-optimized MSH-TP15 Fc-eng. and the Fc k.o. variant MSH-TP15 Fc k.o. were designed to display either enhanced FcγRIIa or FcγRIIIa binding to improve effector cell activation while retaining wt CDC activity or being incapable to mediate CDC and ADCC and therefore exclusively rely on Fab-mediated effector functions (Figure 1A). All proteins were produced in LentiX 293-T cells. Antibody preparations were highly pure and LC appeared at the calculated molecular masses of 25 kDa under reducing conditions. The slightly altered haematologica | 2021; 106(7)

migration of the calculated 50 kDa HC and 150 kDa IgG1 antibodies that were observed under reducing and nonreducing conditions is most likely due to glycosylation (Figure 1B). Importantly, the three MSH-TP15 mAb showed almost identical, concentration-dependent binding to ICAM-1 with EC50 values in the low nanomolar (nM) range, but exerted significant differences in their FcγRIIa and FcγRIIIa binding (Figure 1C to E). As intended, the MSH-TP15 Fc k.o. did not bind to these FcγR on stable transfected cells, while binding of MSH-TP15 Fceng. was significantly enhanced for both FcγR compared to wt MSH-TP15 IgG1 (Figure 1D and E). Thus, these results confirmed the expected differences of the three antibody variants.

MSH-TP15 binds domain 1 of human ICAM-1 In order to proof that by converting the scFv to a Fab fragment, epitope specificity was not altered, L363 cells were pre-incubated with an excess of TP-15-Fc (scFv-Fc antibody) prior to staining with Alexa Fluor 755-labeled MSH-TP15 IgG1. Importantly, MSH-TP15 binding was completely inhibited indicating the same ICAM-1 binding epitope (Figure 2A). In order to more precisely localize the binding region of MSH-TP15, truncated ICAM-1 variants 1859

K. Klausz et al. A

Figure 2. MSH-TP15 binds human ICAM-1 domain 1-2 and partly shares an epitope with CD54 antibodies BI-505 and 84H10. (A) In order to proof that binding specificity is retained by conversion of the single-chain fragment variable (scFv) to the antigen-binding fragment (Fab) containing antibody, cross-blocking studies with TP15-Fc and MSH-TP15 antibodies were performed on L363 cells. Pre-incubation with 500 mg/mL TP15-Fc, but not with the control molecule 4D5-Fc, significantly prevented binding of Alexa Fluor 755 (AF755)-labeled MSH-TP15 (checked bar) measured by flow cytometry. Identically labeled cetuximab (CTX-AF755) served as negative control. Binding of TP15-Fc and 4D5-Fc were detected with FITC-labeled anti-human immunoglobulin (Ig) G antibody (grey bars). (B) Antibody binding to human intercellular adhesion molecule-1 (huICAM-1) or mouse ICAM-1 (muICAM-1) was analyzed by flow cytometry. Transiently transfected CHO-K1 cells expressing myc-tagged full length or truncated human ICAM-1 domains (D) D1-2 to D1-5 were stained with mouse anti-myc antibody and detected with a FITC-labeled anti-mouse secondary antibody. Histograms show binding of MSH-TP15-AF755 (grey) and CTX-AF755 (black) on ICAM-1 expressing CHO-K1 cells. Non-transfected (non-transf.) cells served as control. (C) Cross-blocking experiments with MSH-TP15 and ICAM-1 antibodies RR1/1, 84H10 and BI-505 were performed by flow cytometry on L363 cells with directly labeled antibodies or species specific anti-human or anti-mouse IgG secondary antibody as indicated in the graphs. All graphs show mean values ± standard error of the mean of a minimum of three independent experiments. MFI: mean fluorescence intensity. Controls are shown in grey, cross-blocking in patterned bars. Statistics were calculated vs. pre-incubation with control IgG1 (ctrl IgG1) or binding of antibody without blocking (black bars). ***P<0.001, **P<0.01, *P<0.05, n.s: not significant. Drawing summarizes the results and visualizes the overlapping epitopes of MSH-TP15, BI-505 and clone 84H10.

were cloned for cell surface expression on CHO-K1 cells (Online Supplementary Figure S1). Flow cytometric analyses revealed that for binding of MSH-TP15 at least ICAM-1 domain (D) 1-2 needed to be present. No binding to mouse ICAM-1 was observed (Figure 2B). For unknown reasons isolated expression of D1 was not possible. Therefore, we performed cross-blocking experiments with MSH-TP15 and the anti-human ICAM-1 mAb BI505, 84H10 and RR1/1, both known to bind human ICAM-1 D1.30 As shown in Figure 2C, addition of MSH1860

TP15, BI-505 and 84H10 cross-blocked individual binding – suggesting that they bound to neighboring, overlapping, or even identical epitopes. This was especially observed for BI-505 and MSH-TP15 (Figure 2C). In contrast, RR1/1 binding was not cross-blocked by MSH-TP15, BI-505 or 84H10. This indicates a unique binding site for RR1/1, which was described to inhibit lymphocyte function-associated antigen-1 (LFA-1) ligand binding to ICAM-1 D1.31 In line with this, MSH-TP15 had no impact on LFA-1-ICAM1 interaction (data not shown). haematologica | 2021; 106(7)

Fc-engineered CD54 antibody MSH-TP15 for myeloma therapy

Figure 3. MSH-TP15 induces apoptosis of lymphoma cells after cross-linking on cell surface and inhibits myeloma growth in the presence of bone marrow stromal cells. (A) Induction of apoptosis was tested with Ramos lymphoma cells. After 6-hour incubation with the indicated antibody and 10 mg/mL anti-human fragment crystallizable γ (Fcγ) cross-linking antibody AV-FITC and 7-AAD positive cells were detected by flow cytometry. Rituximab (RTX) served as positive, a control immunoglobulin (Ig) G1 (ctrl IgG1) as negative control. Mean percentage ± standard error of the mean of four independent experiments is shown. (B) Results of one exemplified experiment with 10 mg/mL antibodies. AV-FITC-positive cells are marked in red boxes. (C) Growth inhibition of INA-6 myeloma cells was measured in the presence of bone marrow stromal cells (+BMSC) or absence of BMSC (-BMSC). MSH-TP15 and control IgG1 (ctrl IgG1) were used at 10 mg/mL (grey bars) or 100 mg/mL (black bars). Direct inhibitory effects on BMSC were not detected (100 mg/mL monoclonal antibody [mAb]; right graph). w/o: without; ctrl: control. Experiments were performed four times in triplicates. Graphs show mean values ± standard error of the mean. ***P<0.001, ** P<0.01.

Table 1. EC50 values achieved by MSH-TP15 Fc-Fc-engineered in antibody-dependent cell-mediated cytotoxicity experiments with peripheral blood mononuclear cells

Cell line EC50 (nM)

INA-6

L363 1

0.36 (0.07-1.97)

MM1.S 1

2.59 (0.87-7.77)

U266 1

1.82 (0.63-5.26)

3.77 (0.83-17.2)1

95% CI Confidence Interval, sigmoidal dose response curve.

Direct effector functions of MSH-TP15 First, we analyzed direct effector mechanisms of MSHTP15. MSH-TP15 did not directly inhibit cell proliferation of myeloma cell lines (data not shown), but was capable of inducing apoptosis after antibody cross-linking on the cell surface. MSH-TP15 and rituximab both induced apoptosis in Ramos lymphoma cells expressing ICAM-1 and CD20, while a control IgG1 was not effective (Figure 3A). Total percentage of Annexin V (AV)-positive cells was calculated by combining early apoptotic (AV+/7-AAD–) and dead cells (AV+/7-AAD+) as shown for one exemplified experiment in Figure 3B. Next, we analyzed the impact of MSH-TP15 on growth of IL-6-dependent INA-6 myeloma cells in the presence of BMSC isolated from myeloma patients. Growth inhibition of INA-6 cells in these co-culture experiments was observed in the presence of 10 or 100 µg/mL MSHTP15, while no direct inhibitory effect on growth of BMSC was detectable with 100 µg/mL MSH-TP15 (Figure 3C).

MSH-TP15 Fc-engineered induces natural killer cell killing and macrophage-mediated phagocytosis of myeloma cells In order to investigate Fc-mediated effector mechanisms of MSH-TP15, we performed chromium release assays. None of the antibody variants induced complementdependent lysis of myeloma cells (data not shown), but sighaematologica | 2021; 106(7)

nificant ADCC of plasmocytoma cells could be achieved with MSH-TP15 Fc-eng. and healthy donor’s PBMC (Figure 4A). Only minimal lysis was observed with MSHTP15 and as expected no killing was detectable with the Fc k.o. variant and control mAb. EC50 values of MSH-TP15 Fc-eng. ranged from 0.36 nM (INA-6) to 3.77 nM (U266) and are summarized in Table 1. Of note, MSH-TP15 Fceng. also induced significant tumor cell lysis when patientderived myeloma cells, including those from relapsed/refractory MM patients, and purified natural killer (NK) cells were used (Figure 4B, for patient details refer to Table 2). Flow cytometric analyses verified specific binding of MSH-TP15 to the CD138+ tumor cells of all patients (Figure 4B). NK cell induced killing of INA-6 myeloma cells by MSH-TP15 Fc-eng. was additionally investigated with live cell imaging (Figure 4C). Caspase activation and apoptosis of INA-6 cells was present over the entire 24 h assay time period. Exemplified microscopy images from 0, 6, 12 and 24 h are shown in Figure 4C. Quantification of the green fluorescent counts (caspasedependent apoptosis) after 24 h revealed that significant killing of INA-6 myeloma cells occured only in the presence of NK cells and MSH-TP15 Fc-eng. (Figure 4C). Finally, ADCP activity of the MSH-TP15 antibodies was analyzed with Raji lymphoma cells (CD19, CD20 and ICAM-1 positive) and monocyte-derived macrophages 1861

K. Klausz et al.

Figure 4. MSH-TP15 Fc-engineered efficiently recruits natural killer cells for antibody-dependent cell-mediated cytotoxicity and engages macrophages for antibodydependent cellular phagocytosis of myeloma and lymphoma cells. (A) Antibody-dependent cell-mediated cytotoxicity (ADCC) experiments were performed as standard 3-hour chromium release assays with increasing antibody concentrations and peripheral blood mononuclear cells (PBMC) of healthy donors at an E:T ratio of 80:1. Antibody-mediated myeloma cell lysis was analyzed for MSH-TP15 (), MSH-TP15 Fc-engineered (Fc-eng.). (▲), MSH-TP15 Fc knockout (Fc k.o.). (∇) and control monoclonal antibodies (mAb) (, ) with the indicated cell lines. (B) ADCC experiments with patient-derived tumor cells from BM aspirates of multiple myeloma (MM) patients (P#1-P#4), pleural effusion of an extramedullary myeloma (EM-MM) patient (P#5) and PB of a plasma cell leukemia (PCL) patient (P#6) were performed with PBMC (80:1; black bars) or natural killer (NK) cells (10:1; grey bars) of healthy donors and 10 µg/mL of the indicated antibodies. Graph shows mean values ± standard error of the mean of six independent experiments. Histograms on the right show staining of the CD138+ malignant plasma cells of these patients (for details please refer to Table 2) with Alexa Fluor 755-labeled MSH-TP15 (grey line) or control (ctrl) IgG1 (black line). (C) Apoptosis of INA-6 myeloma cells was measured by life cell imaging over 24 hours (h). Apoptotic tumor cells became green fluorescent when intracellular caspases were activated. Right graph shows the total green counts/image measured after 24 h in the presence (black bars) or absence (grey bars) of NK cells and the indicated antibody. Pictures on the left show images taken from start of the experiment (0 h), 6, 12 and 24 h during incubation of myeloma (MM) and NK cells (NK; E:T ratio 10:1; marked with arrows in the upper left picture) with 20 mg/mL MSH-TP15 Fc-eng. (D) For antibody-dependent cellular phagocytosis (ADCP) analyses macrophages were incubated with 10 mg/ml mAb and Raji lymphoma cells at an E:T ratio of 1:3. Percentage of phagocytosed tumor cells was quantified by flow cytometry gating on CD14+, CFSE+ and CD19– cells. Rituximab (RTX) served as positive control. Experiment was repeated six times with macrophages derived from different donors. Mean values ± standard error of the mean of phagocytosed tumor cells is summarized in the graph with significant differences between ctrl IgG1 and the other mAb marked with stars. ***P<0.001, **P<0.01 and *P<0.05 of MSH-TP15 mAb/RTX vs. ctrl IgG1 for all experiments.

Table 2. Patient characteristics.

Patient #1 #2 #3 #4 #5 #6

Diagnosis

Sex

Age

Source of material

CD138+ PC

MM (IgM), initial diagnosis MM (IgAl), progressive disease MM (IgG), refractory, progress MM (IgGκ) EM-MM (IgGl), active disease PCL (IgGκ), active disease

f f m m f f

67 y 70 y 56 y 54 y 70 y 67 y

BM, freshly isolated BM, frozen BM, frozen BM, freshly isolated PE, freshly isolated PB, freshly isolated

30 % 63 % 58 % 35 % 93 % 80 %

MM: multiple myeloma; Ig: immunoglobulin; PC: plasma cell; PCL: plasma cell leukemia; EM: extramedullary; f: female; m: male; y: years; BM: bone marrow; PB: peripheral blood; PE: pleural effusion.

1862

haematologica | 2021; 106(7)

Fc-engineered CD54 antibody MSH-TP15 for myeloma therapy

Figure 5. In vivo anti-tumor efficacy of the MSH-TP15 monoclonalantibody variants tested in two myeloma xenograft models. (A) 48 hours (h) after intraperitoneal (i.p.) injection of 2x107 INA-6.Tu1_red cells in ten SCID/beige mice per group, animals were treated twice weekly either with vehicle (phosphate buffered saline control [PBS ctrl]; dotted black line), MSH-TP15 (black line), MSH-TP15 Fc-engineered (Fc-eng.) (dotted grey line) or MSH-TP15 Fc knockout (Fc k.o.) (grey line). All mice received one dose of 10 mg/kg and six doses of 5 mg/kg by i.p. injection, time points are marked with ↓. (B) Left graph shows the human interleukin 6 receptor (huIL-6R) concentration in final sera of all mice measured by enzyme-linked immunosorbent assay. Correlation between huIL-6R concentration and tumor weight of the explanted tumors of PBS and MSH-TP15 Fc k.o. treated mice is shown in the right graph. (C) Survival and (D) tumor volume of five mice per group injected subcutaneously with 5x106 INA-6.Tu1_red cells 10 days prior start of twice weekly i.p. treatment with five doses of 10 mg/kg (left graphs), 1 mg/kg (middle) or 0.1 mg/kg (right) MSH-TP15 (black line) and MSH-TP15 Fc-eng. (dotted grey line). Animals in the MSH-TP15 Fc k.o. group (grey line) received 10 mg/kg and PBS treated animals (dotted black line) served as control. ↓ time point of treatment. Tumor volume was calculated by regular caliper measurement of subcutaneous tumors.

from healthy donors, which highly expressed FcγRII/CD32 and FcγRI/CD64 as well as lower levels of FcγRIIIa/CD16a (Online Supplementary Figure S2). In order to determine ADCP, CD14-stained macrophages were incubated with CFSE-labeled Raji cells at an E:T ratio of 1:3 in the presence of the indicated mAb. CD14/CFSE double positive and CD19 negative cells were defined as phagocytosed tumor cells by flow cytometry. Significant ADCP activity was observed for MSH-TP15 Fc-eng. and the positive control rituximab, while the MSH-TP15 Fc k.o. completely lacked activity and ADCP by MSH-TP15 was moderate (Figure 4D). The significant recruitment of NK cells and macrophages for ADCC and ADCP of myeloma cells by MSH-TP15 Fc-eng. most likely reflects the improved affinity of this variant to FcγRIIIa and FcγRIIa (Figure 1D and E).

MSH-TP15 and MSH-TP15 Fc-engineered have significant in vivo anti-myeloma activity The MSH-TP15 mAb variants were tested in INA6.Tu1_red myeloma xenograft models in immunodeficient SCID/beige mice. In the first experiment, 48 h after i.p. injection of tumor cells treatment with MSH-TP15 haematologica | 2021; 106(7)

mAb or vehicle control (phosphate buffered saline [PBS]) was started. Within 65 days, control animals and MSHTP15 Fc k.o. treated mice needed to be sacrificed due to tumor size, resulting in a median survival of 39 and 32 days, respectively (Figure 5A). Human IL-6 receptor (huIL6R) secreted by the INA-6.Tu1_red cells was detectable in final sera of these mice and its concentration correlated with explanted tumor weight (Figure 5B). In contrast, animals treated with MSH-TP15 or MSH-TP15 Fc-eng. did not develop any tumors until the end of the experiment on day 90 (P<0.001 vs. control mice) and no huIL-6R was measured (Figure 5B). Thus, treatment with both, MSHTP15 and MSH-TP15 Fc-eng., completely prevented the development of myeloma in this model. Next, in order to compare the efficacy of MSH-TP15 and MSH-TP15 Fc-eng. in more detail, doses of 10, 1 or 0.1 mg/kg were administered to s.c. tumor bearing mice. MSH-TP15 Fc k.o. was given at 10 mg/kg and PBS-treated animals served as control (Figure 5C and D). Compared to the control mice that needed to be sacrificed after 22 days, MSH-TP15 Fc k.o. treated mice reached a median survival of 27 days (Figure 5C). Direct, Fab-mediated anti-myeloma effects may account for the slightly delayed tumor 1863

K. Klausz et al.

growth and the survival benefit compared to control mice, but significance was not reached (P=0.11). Of note, treatment with 0.1 mg/kg MSH-TP15 significantly retarded tumor growth and prolonged survival (27 days; P=0.03 vs. control group; Figure 5C and D). Treatment with 1 mg/kg and 10 mg/kg further improved tumor control and resulted in a median survival of 33 days and 36 days, respectively (P<0.01 for both doses vs. control group). Importantly, with the Fc-optimized MSH-TP15 Fc-eng. survival was not only significantly prolonged compared to PBS-treated animals at doses of 0.1 mg/kg (29 days, P=0.02), 1 mg/kg (41 days, P<0.01) and 10 mg/kg (50 days, P<0.01), but also when compared to 10 mg/kg MSH-TP15 treated mice (P=0.03). This tendency of being more effective than the wt IgG1 mAb was also seen at 1 mg/kg doses, although differences did not reach statistical significance (P=0.12; Figure 5C). These results point to the importance of efficient effector cell recruitment for the anti-myeloma activity of MSH-TP15 in vivo.

Discussion In this study, we characterized the ICAM-1 binding epitope and modes of action of the novel human IgG1 antibody MSH-TP15 originally isolated as scFv antibody by phage display and myeloma cell screening.22 Detailed binding analyses revealed that the MSH-TP15 epitope is located at the N-terminal D1-2 of human ICAM-1, which is a trans-membrane glycoprotein that consists of five, heavily glycosylated extracellular Ig domains.32 The majority of the ICAM-1 ligands bind to D1, including the β2-integrin LFA-1 (CD11a/CD18) expressed on all lymphocytes.32 ICAM-1 and β2-integrin interactions play a pivotal role in lymphocyte activation and adhesion, leukocyte trafficking, and cellular immune responses. In general, ICAM-1 and its ligands are expressed on lymphocytes, leukocytes and vascular endothelium, are increased upon stimulation with inflammatory cytokines like IFN-γ and mediate adhesion of immune cells to allow migration into tissue.9 Many anti-ICAM-1 antibodies, like RR1/1 and 84H10, bind to D1 or D2 and inhibit LFA-1-mediated adhesion of lymphocytes to ICAM-1 although they recognize different epitopes.30 Our cross-blocking experiments confirmed a unique ICAM-1 epitope recognized by RR1/1 distinct from that of 84H10, MSH-TP15 and BI-505. Potent blockade and thus overlapping epitopes were observed for MSH-TP15 and BI-505, while only partly or no overlapping epitopes were seen for the two human mAb with 84H10 and RR1/1. Accordingly, MSH-TP15 was incapable of inhibiting LFA-1 and ICAM-1 interaction. Of note, MSH-TP15 and BI-505 were both identified by phage display and are derived from a fully human scFv library based on the same VH (VH-DP47), but different LC (VL-DPL3 vs. DPK9) framework.22,33 This resulted in 89% VH sequence identity between the human IgG1l mAb BI505 and the human IgG1κ mAb MSH-TP15. Interestingly, in contrast to MSH-TP15, BI-505 was found by screening for apoptosis-inducing antibodies against Ramos Burkitt’s lymphoma cells.34 Its exact epitope was not determined, but our studies indicate that BI-505 and MSH-TP15 bind close epitopes of human ICAM-1 D1-2. Despite its important role in immunological processes, ICAM-1 was found to be highly expressed on malignant plasma cells.35,36 There ICAM-1 seems to play a pivotal role 1864

in myeloma cell adhesion to BMSC, and it is, thus considered to be an important molecule in the tumor-microenvironment and to be essential for immune escape of myeloma cells. In addition, ICAM-1 is thought to be involved in macrophage-induced drug resistance and was shown to be overexpressed on chemo-resistant residual myeloma cells. Furthermore, ICAM-1 expression correlates with tumor progression and metastasis in melanoma and renal cell carcinoma patients,38,39 and was recently identified as potential cancer stem cell marker in esophageal squamous cell carcinoma.40 Targeting ICAM-1 for therapy was initially evaluated with mouse antibodies that were clinically tested in rheumatoid arthritis, acute stroke and renal transplantation.11-13 Production of human anti-mouse Ab caused serious problems after repeated treatment.41,42 To date, BI-505 is the only clinically tested human anti-ICAM-1 Ab tested. Of note, repeated administration of 10 mg/kg BI-505 already saturated ICAM-1 on patient’s BM myeloma cells and up to 43 mg/kg were well tolerated.10,43 Many ICAM-1 targeting mAb, including BI-505 and MSH-TP15, exert only little Fab-mediated anti-tumor effects. MSH-TP15 also did not directly affect myeloma cell proliferation, but was capable of inhibiting IL-6dependent INA-6 growth in the presence of BMSC from myeloma patients. This might be an interesting function regarding the supportive role of the BM microenvironment for survival and drug resistance of myeloma cells. Additionally, MSH-TP15 induced apoptosis of ICAM-1 expressing lymphoma cells after cross-linking on the cell surface – a mechanisms which was also described for BI505.6 No CDC but moderate ADCP activity and activation of human NK cells for ADCC of myeloma cells was observed for MSH-TP15. Together these data suggest engagement of NK cells and phagocytes as a predominant effector function of MSH-TP15, a fact already evident for scFv-Fc antibody TP15-Fc.22 In an attempt to improve the Fc-mediated effector functions of MSH-TP15, we applied Fc protein-engineering. As intended, the DE mutations introduced into the Fc domain of MSH-TP15 Fc-eng. substantially improved the antibody’s affinity for FcγRIIIa and FcγRIIa,15,44 resulting in enhanced ADCC and ADCP activity of MSH-TP15 Fc-eng. against myeloma cells in vitro. Of note, significant NK cell-mediated tumor cell lysis was also achieved for patient-derived myeloma cells. In vivo, wt and Fc-optimized MSH-TP15 controlled tumor growth in two myeloma xenograft models, while with the k.o. Ab variant no significant survival benefit compared to control mice was achieved. Myeloma cell growth was completely prevented or dose-dependently inhibited by MSH-TP15 and MSH-TP15 Fc-eng. and both antibody variants significantly prolonged survival of the mice. Importantly, improvement in tumor control and survival was also seen in vivo with the Fc-optimized MSHTP15 Fc-eng. compared to MSH-TP15 pointing to an important role of immune cells in mice as well. Due to the fact that in SCDI/bg mice NK cell functions are impaired, most likely mouse myeloid effector cells (monocytes/macrophages, granulocytes) account for the observed differences. This is in line with previous findings that human IgG1 antibodies efficiently bind to mouse FcγRIV, which is the orthologue of human FcγRIIIa, but which is, in contrast to the human system, restricted to mouse myeloid cells and is not expressed on mouse NK cells.45,46 Since our novel anti-ICAM-1 antibodhaematologica | 2021; 106(7)

Fc-engineered CD54 antibody MSH-TP15 for myeloma therapy

ies not only triggered phagocytosis with human macrophages but also significantly triggered ADCC by human NK cells in vitro this might suggest that although in the mouse model myeloid immune effector cells play a dominant role. In patients, together with macrophages also NK cells may constitute a powerful additional effector cell population, which can be recruited for tumor cell killing. Nevertheless, in the subcutaneous model, cure of the mice was not achieved. This leaves room for further improvement and allows testing of combinations of our novel anti-ICAM-1 antibodies with various established anti-myeloma treatment regimen, including proteasome inhibitors, immunomodulatory drugs and chemotherapeutic substances. Systematic testing in our model system may allow the identification of optimal combination partners for potential clinical application. Taken together, our data indicate that predominantly Fc-mediated effector functions account for the anti-myeloma activity of MSH-TP15 and that by Fc protein-engineering a significant improvement of the antibody’s in vitro and in vivo activity was achieved. Particularly for myeloma therapy, where in the post-transplant setting NK cells are among the first immune cells to be present, targeting resid-

References 1. Jakubowiak A, Offidani M, Pegourie B, et al. Randomized phase 2 study: elotuzumab plus bortezomib/dexamethasone vs bortezomib/dexamethasone for relapsed/refractory MM. Blood. 2016;127(23):2833-2840. 2. Nijhof IS, Casneuf T, van Velzen J, et al. CD38 expression and complement inhibitors affect response and resistance to daratumumab therapy in myeloma. Blood. 2016;128(7):959-970. 3. Markovina S, Callander NS, O'Connor SL, et al. Bone marrow stromal cells from multiple myeloma patients uniquely induce bortezomib resistant NF-kappaB activity in myeloma cells. Mol Cancer. 2010;9:176. 4. Meads MB, Gatenby RA, Dalton WS. Environment-mediated drug resistance: a major contributor to minimal residual disease. Nat Rev Cancer. 2009;9(9):665-674. 5. Kawano Y, Moschetta M, Manier S, et al. Targeting the bone marrow microenvironment in multiple myeloma. Immunol Rev. 2015;263(1):160-172. 6. Veitonmaki N, Hansson M, Zhan F, et al. A human ICAM-1 antibody isolated by a function-first approach has potent macrophage-dependent antimyeloma activity in vivo. Cancer Cell. 2013; 23(4):502-515. 7. Schmidmaier R, Morsdorf K, Baumann P, Emmerich B, Meinhardt G. Evidence for cell adhesion-mediated drug resistance of multiple myeloma cells in vivo. Int J Biol Markers. 2006;21(4):218-222. 8. Zheng Y, Yang J, Qian J, et al. PSGL1/selectin and ICAM-1/CD18 interactions are involved in macrophage-induced drug resistance in myeloma. Leukemia. 2013;27(3):702-710. 9. Dustin ML, Rothlein R, Bhan AK, Dinarello CA, Springer TA. Induction by IL 1 and interferon-gamma: tissue distribution, biochemistry, and function of a natural adherence molecule (ICAM-1). J Immunol. 1986;137(1):245-254. 10. Hansson M, Gimsing P, Badros A, et al. A

haematologica | 2021; 106(7)

ual tumor cells in the patients BM with an Fc-eng. antiICAM-1 antibody like MSH-TP15 Fc-eng. might be a promising strategy to improve clinical outcome. Disclosures No conflicts of interest to disclose. Contributions KK, MC, CK, TR, AO, SK and AL performed the research; KK and MP wrote the manuscript; CK, FN, TR, TV and MG contributed to writing of the manuscript; MG and MP supervised the study Acknowledgments The authors would like to thank Kathrin Richter, Jan Brdon, Britta von Below and Anja Muskulus for excellent technical assistance. Funding This study was supported by a research grant from the Deutsche Krebshilfe e.V. (Mildred-Scheel-Professorship program) to MP and Else Kröner-Fresenius-Stiftung (2015_A166) to KK.

Phase I dose-escalation study of antibody BI-505 in relapsed/refractory multiple myeloma. Clin Cancer Res. 2015; 21(12):2730-2736. 11. Haug CE, Colvin RB, Delmonico FL, et al. A phase I trial of immunosuppression with anti-ICAM-1 (CD54) mAb in renal allograft recipients. Transplantation. 1993;55(4):766773. 12. Kavanaugh AF, Davis LS, Jain RI, et al. A phase I/II open label study of the safety and efficacy of an anti-ICAM-1 (intercellular adhesion molecule-1; CD54) monoclonal antibody in early rheumatoid arthritis. J Rheumatol. 1996;23(8):1338-1344. 13. Schneider D, Berrouschot J, Brandt T, et al. Safety, pharmacokinetics and biological activity of enlimomab (anti-ICAM-1 antibody): an open-label, dose escalation study in patients hospitalized for acute stroke. Eur Neurol. 1998;40(2):78-83. 14. Scott AM, Wolchok JD, Old LJ. Antibody therapy of cancer. Nat Rev Cancer. 2012;12(4):278-287. 15. Lazar GA, Dang W, Karki S, et al. Engineered antibody Fc variants with enhanced effector function. Proc Natl Acad Sci U S A. 2006;103(11):4005-4010. 16. Moore GL, Chen H, Karki S, Lazar GA. Engineered Fc variant antibodies with enhanced ability to recruit complement and mediate effector functions. MAbs. 2010;2(2):181-189. 17. Richards JO, Karki S, Lazar GA, et al. Optimization of antibody binding to FcγRIIa enhances macrophage phagocytosis of tumor cells. Mol Cancer Ther. 2008; 7(8):2517-2527. 18. de Jong RN, Beurskens FJ, Verploegen S, et al. A novel platform for the potentiation of therapeutic antibodies based on antigendependent formation of IgG hexamers at the cell surface. PLoS Biol. 2016;14(1): e1002344. 19. Kellner C, Otte A, Cappuzzello E, Klausz K, Peipp M. Modulating cytotoxic effector functions by Fc engineering to improve cancer therapy. Transfus Med Hemother.

2017;44(5):327-336. 20. Jurczak W, Zinzani PL, Gaidano G, et al. Phase IIa study of the CD19 antibody MOR208 in patients with relapsed or refractory B-cell non-Hodgkin's lymphoma. Ann Oncol. 2018;29(5):12661272. 21. Bang YJ, Giaccone G, Im SA, et al. First-inhuman phase 1 study of margetuximab (MGAH22), an Fc-modified chimeric monoclonal antibody, in patients with HER2positive advanced solid tumors. Ann Oncol. 2017;28(4):855-861. 22. Klausz K, Cieker M, Kellner C, et al. A novel Fc-engineered human ICAM-1/CD54 antibody with potent anti-myeloma activity developed by cellular panning of phage display libraries. Oncotarget. 2017; 8(44):77552-77566. 23. Peipp M, Lammerts van Bueren JJ, Schneider-Merck T, et al. Antibody fucosylation differentially impacts cytotoxicity mediated by NK and PMN effector cells. Blood. 2008;112(6):2390-2399. 24. Lux A, Yu X, Scanlan CN, Nimmerjahn F. Impact of immune complex size and glycosylation on IgG binding to human FcgammaRs. J Immunol. 2013;190(8):43154323. 25. Burger R, Guenther A, Bakker F, et al. Gp130 and ras mediated signaling in human plasma cell line INA-6: a cytokineregulated tumor model for plasmacytoma. Hematol J. 2001;2(1):42-53. 26. Glorius P, Baerenwaldt A, Kellner C, et al. The novel tribody [(CD20)(2)xCD16] efficiently triggers effector cell-mediated lysis of malignant B cells. Leukemia. 2013; 27(1):190-201. 27. Hansson MFB, inventor; The use of antibodies against ICAM-1 in the treatment of patients with relapsed cancer. 2012 Intl. Cl. C07K 16/28 (2006.01). Appl. No.: PCT/EP2011/061983. 28. Kellner C, Derer S, Klausz K, et al. Fc glycoand Fc orotein-engineering: design of antibody variants with improved ADCC and CDC activity. Methods Mol Biol. 2018;

1865

K. Klausz et al. 1827:381-397. 29. Staudinger M, Glorius P, Burger R, et al. The novel immunotoxin HM1.24-ETA' induces apoptosis in multiple myeloma cells. Blood Cancer J. 2014;4:e219. 30. Makgoba MW, Sanders ME, Ginther Luce GE, et al. A cluster of antibodies (RR1/1, LB-2 and 84H10) that inhibit LFA-1-dependent lymphoid and myeloid cell adhesion bind intercellular adhesion molecule-1 (ICAM-1). In: Dupont B, editor. Immunobiology of HLA. Vol II. Berlin: Springer-Verlag New York; 1989. 31. Rothlein R, Dustin ML, Marlin SD, Springer TA. A human intercellular adhesion molecule (ICAM-1) distinct from LFA-1. J Immunol. 1986;137(4):1270-1274. 32. Staunton DE, Marlin SD, Stratowa C, Dustin ML, Springer TA. Primary structure of ICAM-1 demonstrates interaction between members of the immunoglobulin and integrin supergene families. Cell. 1988; 52(6):925-933. 33. Soderlind E, Strandberg L, Jirholt P, et al. Recombining germline-derived CDR sequences for creating diverse single-framework antibody libraries. Nat Biotechnol. 2000;18(8):852-856. 34. Fransson J, Tornberg UC, Borrebaeck CA, Carlsson R, Frendeus B. Rapid induction of apoptosis in B-cell lymphoma by functionally isolated human antibodies. Int J Cancer. 2006;119(2):349-358.

1866

35. Tatsumi T, Shimazaki C, Goto H, et al. Expression of adhesion molecules on myeloma cells. Jpn J Cancer Res. 1996;87(8):837-842. 36. Van Riet I, De Waele M, Remels L, et al. Expression of cytoadhesion molecules (CD56, CD54, CD18 and CD29) by myeloma plasma cells. Br J Haematol. 1991; 79(3):421-427. 37. Paiva B, Corchete LA, Vidriales MB, et al. Phenotypic and genomic analysis of multiple myeloma minimal residual disease tumor cells: a new model to understand chemoresistance. Blood. 2016;127(15): 1896-1906. 38. Galore-Haskel G, Baruch EN, Berg AL, et al. Histopathological expression analysis of intercellular adhesion molecule 1 (ICAM-1) along development and progression of human melanoma. Oncotarget. 2017; 8(59):99580-99586. 39. Juengel E, Krueger G, Rutz J, et al. Renal cell carcinoma alters endothelial receptor expression responsible for leukocyte adhesion. Oncotarget. 2016;7(15):20410-20424. 40. Tsai ST, Wang PJ, Liou NJ, et al. ICAM1 is a potential cancer stem cell marker of esophageal squamous cell carcinoma. PLoS One. 2015;10(11):e0142834. 41. Kavanaugh AF, Schulze-Koops H, Davis LS, Lipsky PE. Repeat treatment of rheumatoid arthritis patients with a murine anti-intercellular adhesion molecule 1 monoclonal

antibody. Arthritis Rheum. 1997;40(5):849853. 42. Furuya K, Takeda H, Azhar S, et al. Examination of several potential mechanisms for the negative outcome in a clinical stroke trial of enlimomab, a murine antihuman intercellular adhesion molecule-1 antibody: a bedside-to-bench study. Stroke. 2001;32(11):2665-2674. 43. Wichert S, Juliusson G, Johansson A, et al. A single-arm, open-label, phase 2 clinical trial evaluating disease response following treatment with BI-505, a human anti-intercellular adhesion molecule-1 monoclonal antibody, in patients with smoldering multiple myeloma. PLoS One. 2017; 12(2):e0171205. 44. Hamaguchi Y, Xiu Y, Komura K, Nimmerjahn F, Tedder TF. Antibody isotype-specific engagement of Fcγ receptors regulates B lymphocyte depletion during CD20 immunotherapy. J Exp Med. 2006; 203(3):743-753. 45. Kerntke C, Nimmerjahn F, Biburger M. There Is (Scientific) Strength in numbers: a comprehensive quantitation of Fcγ receptor numbers on human and murine peripheral Blood Leukocytes. Front Immunol. 2020; 11:118. 46. Lux A, Nimmerjahn F. Of mice and men: the need for humanized mouse models to study human IgG activity in vivo. J Clin Immunol. 2013;33(Suppl 1):S4-8.

haematologica | 2021; 106(7)

ARTICLE

Chronic Lymphocytic Leukemia

Olivier Tournilhac,1 Magali Le Garff-Tavernier,2 Stéphanie Nguyen Quoc,3 Edouard Forcade,4 Patrice Chevallier,5 Faezeh Legrand-Izadifar,6 Gandhi Laurent Damaj,7 David Michonneau,8 Cécile Tomowiak,9 Cécile Borel,10 Corentin Orvain,11 Pascal Turlure,12 Rabah Redjou,13 Gaëlle Guillerm,14 Laure Vincent,15 Celestine Simand,16 Richard Lemal,17 Claire Quiney,2 Patricia Combes,18 Bruno Pereira,19 Laure Calvet,20 Aurélie Cabrespine,1 Jacques-Olivier Bay,1 Véronique Leblond3 and Nathalie Dhédin21

Service d’Hématologie Clinique et de Thérapie Cellulaire, CHU Estaing, Université Clermont Auvergne EA 7453 CIC1405, Clermont-Ferrand; 2Service d’Hématologie Biologique, Groupe Hospitalier Pitié-Salpêtrière, Assistance Publique - Hôpitaux de Paris, Paris; 3Service d’Hématologie Clinique, Groupe Hospitalier Pitié-Salpêtrière, Assistance Publique - Hôpitaux de Paris, Paris; 4Service d’Hématologie Clinique et de Thérapie cellulaire, CHU Bordeaux, Bordeaux; 5Service d’Hématologie Clinique, CHU Nantes Hôtel Dieu, Nantes; 6Service d’Hématologie Clinique, Département de Greffe de Moelle, CHU Nice, Nice; 7Hématologie Clinique, Institut d'Hématologie de BasseNormandie, CHU Côte de Nacre, Caen; 8Service Hématologie Greffe, Hôpital Saint-Louis, Assistance Publique - Hôpitaux de Paris, Paris; Université Paris Diderot, Paris; 9Service Oncologie Hématologique et Thérapie Cellulaire, CHU Poitiers, Poitiers; 10Service d’Hématologie, Institut Universitaire du Cancer Toulouse - Oncopole, Toulouse; 11Service Maladies du Sang, CHU Angers, Angers; 12Service d’Hématologie Clinique, CHU Dupuytren, Limoges; 13Service d’Hématologie Clinique, Hôpital Henri Mondor, Assistance Publique - Hôpitaux de Paris, Créteil; 14Service d'Hématologie Clinique, Institut de Cancéro-Hématologie, Hôpital Augustin Morvan, Brest; 15Département Hématologie Clinique, Hôpital St Eloi, Montpellier; 16Service Hématologie, CHU de Strasbourg, Strasbourg; 17Service d’Histocompatibilité, CHU, Université Clermont Auvergne EA 7453 and CIC501, Clermont-Ferrand; 18Service Cytogénétique, CHU Estaing, Clermont-Ferrand; 19Unité de Biostatistiques, Direction de la Recherche Clinique (DRCI), CHU, Clermont-Ferrand; 20Service de Réanimation Médicale, Hôpital Gabriel Monpied, CHU de Clermont-Ferrand, Clermont-Ferrand and 21Unité Adolescents et Jeunes Adultes, Hôpital St Louis, Assistance Publique - Hôpitaux de Paris, Paris, France

Ferrata Storti Foundation

Haematologica 2021 Volume 106(7):1867-1875

Correspondence: OLIVIER TOURNILHAC otournilhac@chu-clermontferrand.fr

ABSTRACT

llogeneic hematopoietic stem cell transplantation (HSCT) remains a potentially curative and useful strategy in high-risk relapsing chronic lymphocytic leukemia (CLL). Minimal residual disease (MRD) assessment at 12 months (M12) post-HSCT is predictive of relapse. This phase II study aimed to achieve M12 MRD negativity (MRD ) using an MRD-driven immune-intervention (Md-PII) algorithm based on serial flow-cytometry blood MRD, involving cyclosporine tapering followed in case of failure by donor lymphocytes infusions. Patients had high-risk CLL according to the 2006 European Society for Blood and Marrow Transplantation consensus, in complete or partial response with lymphadenopathy <5 cm and comorbidity score ≤2. Donors were HLA-matched sibling or matched unrelated (10/10). Fortytwo enrolled patients with either 17p deletion (front-line, n=11; relapse n=16) or other high-risk relapse (n=15) received reduced intensity-conditioning regimen before HSCT and were submitted to Md-PII. M12MRD status was achieved in 27 of 42 patients (64%) versus 6 of 42 (14.2%) before HSCT. With a median follow-up of 36 months (range, 1953), 3-year overall survival, non-relapse mortality and cumulative incidence of relapse are 86.9% (95% Confidence Interval [CI]: 70.8-94.4),

Received: October 15 2019. Accepted: May 29, 2020. Pre-published: June 11, 2020. https://doi.org/10.3324/haematol.2019.239566

neg

haematologica | 2021; 106(7)

1867

O. Tournilhac et al.

9.5% (95% CI: 3.7-23.4) and 29.6% (95% CI: 17.3-47.7). Incidence of 2-year limited and extensive chronic graft versus host disease (cGVHD) is 38% (95% CI: 23-53) and 23% (95% CI: 10-36) including two cases post Md-PII. Fifteen patients converted to MRD either after cyclosporine A withdrawal (n=12) or after cGvHD (n=3). As a time-dependent variable, MRD achievement at any time-point correlates with reduced relapse (Hazard ratio [HR] 0.14 [range, 0.04-0.53], P=0.004) and improvement of both progression free (HR 0.18 [range, 0.06-0.6], P<0.005) and overall (HR 0.18 [range, 0.03-0.98], P=0.047) survival. These data highlight the value of MRD-driven immune-intervention to induce prompt MRD clearance in the therapy of CLL (clinicaltrials gov. Identifier: NCT01849939). neg

neg

Introduction Until recently, patients with refractory chronic lymphocytic leukemia (CLL) or who relapse early after purine analogs and rituximab-based chemoimmunotherapy or those harboring 17p deletion (del(17p)) and/or TP53 mutations were considered high-risk patients with reduced overall survival (OS). Better understanding of the molecular and genetic aspects of CLL brought novel and highly active strategies such as targeting kinases downstream of the Bcell receptor (BCR) pathway.1-3 These therapies have profondly modified the CLL therapeutic landscape, thanks to improved efficacy and better tolerability. However, the disease is still incurable and allogeneic hematopoietic stem cell transplantation (HSCT) remains a valid option in selected high-risk patients.4,5 Prospective studies have shown that allogeneic HSCT can offer long progression free survival (PFS) and even a cure in 35% to 45% of high-risk patients. Reduced intensity conditioning (RIC) HSCT can be proposed to older patients and patients with comorbidities who represent the bulk of the CLL population. However disease recurrence, recorded in 22% to 46% patients, is still a major issue.6-9 Pre-transplantation refractoriness and bulky disease is associated with higher risk of post-transplantation progression.9,10 The level of post-transplantation minimal residual disease (MRD) is widely associated with the risk of further progression. In several studies, a negative MRD (MRD ) status at 6 to 12 months translated into a progression incidence below 10%.11-14 Moreover, the MRD status may be reached by post-transplantation immunomodulation such as cyclosporine A (CsA) tapering or donor lymphocyte infusion (DLI).15 These data led us to conduct a prospective study evaluating an approach of RIC HSCT followed by a preemptive MRD-driven immune-intervention with the aim to achieve a MRD status at 12 months post-transplantation. neg

neg

Methods

comorbidity score ≤ 2. Donors were HLA-matched sibling or unrelated (10/10).17 All responsible Institutional Review Boards in accordance with the Declaration of Helsinki approved the protocol including the study-specific informed consent form. The study was declared to the French Authorities (reference ID-RCB 2011A00906-35) and registered on clinicaltrials gov. Identifier: NCT01849939.

Transplantation modalities Conditioning regimen was fludarabine, 30 mg/m2/day, from day (D) D-5 to D-1, intravenous busulfan 3.2 mg/kg/day from D4 to D-3 and ATG (thymoglobuline) 2.5 mg/kg/day from D-3 to D-2.18 Stem cell source was G-CSF mobilized peripheral blood cells. Graft-versus-host disease (GvHD) prophylaxis was based on CsA with a short course of methotrexate in case of minor donor/recipient ABO mismatch.

Response and minimal residual disease evaluation Response evaluation was performed according to 2008 iwCLL criteria including computed tomography scan (CT-scan)19 before and 3 months (M3), 6 months (M6) and 12 months (M12) after transplantation. MRD analysis was centrally performed on blood and/or bone marrow by 10-color multiparameter flow cytometry.20-22 MRD definition was <1 CLL cell detectable per 10,000 leukocytes (<1.10-4).19 MRD (MRD ) definition was ≥1 CLL cell detectable per 10,000 leukocytes. Clusters of <20 events were considered as undetectable MRD (UD). Blood MRD evaluation was planned before transplantation, then monthly until M6, at M9 and M12. Once achieved, the blood MRD status was confirmed 1 month later in both blood and bone marrow.

neg

pos

neg

Preemptive immune-intervention Preemptive immune-intervention was applied in the absence of significant GvHD, defined by either acute GvHD (aGVHD) ≥ grade 2 or extensive chronic GvHD (cGvHD). The algorithm based on response and blood MRD assessment included acceleration of CsA tapering and withdrawal followed in case of failure by escalating DLI. The algorithm also included extension of CsA treatment in case of early achievement of MRD status (Online Supplementary Appendix; Figure A)

Study design The ICLL03 RICAC-PMM (Reduced Intensity Conditioning Allogeneic Transplantation for CLL with Preemptive MDR Management), a joint FILO (French Innovative Leukemia Organization) and SFGM-TC (Société Francophone de Greffe de Moelle et de Thérapie Cellulaire) multicenter phase II trial evaluated the efficacy and safety of a preemptive immune-intervention based on MRD assessment in high-risk CLL. Eligible patients were 18 to 70 years old, with CLL (Matutes score 4 or 5) or lymphocytic lymphoma, and high-risk features according to the 2006 European Society for Blood and Marrow Transplantation (EBMT) consensus16 (see Online Supplementary Appendix). Patients had to be in complete or partial response with lymphadenopathy <5 cm and a 1868

Chimerism and graft-versus-host disease assessment Chimerism studies were performed on peripheral blood at M1, M2, M3, M6, and M12 post HSCT by multiplex fluorescent polymerase chain reaction using Short Tandem Repeat analysis. (Online Supplementary Appendix). The diagnosis of GvHD was made according to published criteria.23,24

Trial objectives and statistical analysis The primary objective was to evaluate the efficacy of a preemptive immune-intervention to achieved MRD at M12. The probabilities of OS, and PFS were calculated using the KaplanMeier estimator. The probability of non-relapse mortality neg

haematologica | 2021; 106(7)

Allograft and preemptive immuneintervention in CLL

(NRM) and relapse/progression were calculated using the Fine and Gray approach, considering death as competing risks. In order to evaluate the impact on outcomes of MRD achievement, we performed time dependent analyses considering MRD occurrence as a time-dependent event. Outcome data were estimated by the Mantel-Byar method and graphically illustrated by Simon-Makuch plots25,26 (Online Supplementary Appendix). neg

neg

Results Patients’ characteristics, donor type and protocol adherence Between September 2012 and February 2015, 43 patients fulfilling the 2006 EBMT consensus criteria were recruited in 16 French centers; due to donor comorbidities, one patient included was not eventually transplanted. The present analysis includes the remaining 42 patients (32 male and 10 female). Patients’ characteristics are depicted in the Table 1. Before HSCT, patients had received a median of two

lines of treatment (range, 1-5); the last one being alemtuzumab for 17 patients, immunochemotherapy for 21 and BCR inhibitors for four. Details of previous lines of treatment per patients are reported in the Online Supplementary Table S1. Eight patients were in CR/CRi (including six with blood MRD status) and 34 in PR pre-transplantation. Donors were HLA-identical siblings (n=16) or HLAmatched (10/10) unrelated donors (n=26). The trial profile of the immune-intervention applied in this study and the representative protocol adherence is shown in the Online Supplementary Appendix (Online Supplementary Figure 1B). One patient died before D30. Among the 41 remaining patients, seven were not treated strictly according to study protocol: four patients had an unplanned early CsA withdrawal for primary (n=1) or secondary (n=1) graft failure or mixed chimerism (n=2). One of these last two patients relapsed at 13 months, whereas the second, who later received DLI, was still in mixed chimerism without relapse at 18 months. For two patients, CsA was tapered early, despite MRD status, due to renal failure in one patient. Finally CsA was reduced at D120 instead of D90 for one patient with D90 MRD status. neg

neg

pos

Post-transplantation response and outcomes

Table 1. Patients’ characteristics

Patients (n=42) Sex Female 10 Male 32 Median age at transplant: y (range) 58.6 (40.6 - 68.6) Median time between diagnostic and HSCT: y (range) 4.5 (0.2 - 14.7) Indication for HSCT: n del(17)p and/or TP53 mutation, 1st line 11 del(17)p and/or TP53 mutation, in relapse 16 Purine analogs refractoriness without TP53 abnormality 3 Early relapse (<2 y) after fludarabine based combination or autologous transplant without TP53 abnormality 12 Median prior treatment lines: n (range) 2 (1-5) Last line before HSCT: n Alemtuzumab (+/- Dexamethasone) 17 Bendamustine based combination (B, BR, BOMP) 14 R-DHAC 6 Ibrutinib 3 Idelalisib + rituximab 1 Rituximab 1 Median time between last line and HSCT: d (range) 63 (7-179) Prior exposure to alemtuzumab: n 20 Median interval between alemtuzumab (last line) and 85 (37-179) HSCT: d (range) HSCT done ≤ 60 days post alemtuzumab: n 6 Disease status at transplantation: n CR/CRi 8 PR 34 Lymph node > 15 mm* 25 Lymph node ≤ 15 mm 17 Blood MRD at transplantation Median MRD level: % (range) 0.78 (<10-4 - 70) Negative MRD: n 6 Donor type (HLA 10/10): n HLA Matched sibling 16 HLA Matched unrelated 26 HSCT: hematopoietic stem cell transplantation; B: bendamustine; BR: bendamustine, rituximab; BOMP: bendamustine, ofatumumab, méthylprednisolone; R-DHAC: rituximab, carboplatin, cytarabine, dexamethasone; CR: complete remission; PR: partial remission; MRD: minimal residual disease. n: number, d: days, m: months, y: years. * No patient with lymph node >50 mm.

haematologica | 2021; 106(7)

Engraftment occurred in 40 of the 42 transplanted patients while two presented graft failure. In the latter patients, both in PR at transplantation, the last line were alemtuzumab plus dexamatasone, interrupted 49 days before transplant in one patient and ibrutinib interrupted 9 days before transplant in the other. Assessment of the response according to the iwCLL criteria between 3 and 6 months found CR/CRi and PR for 13 (31%) and 26 (62%) patients, respectively. Response was not evaluated for three patients because of early death (n=1) or graft rejection (n=2). Response was classified as partial when there was lymph node >15 mm persistence (n=6), spleen enlargement (n=4), both (n=3) or incomplete evaluation (n=13) (Online Supplementary Table S2). Seventeen patients developed grade 1 (n=8), 2 (n=6) and 3 (n=3) aGvHD. Limited and extensive cGvHD occurred in 15 and nine of the 39 patients who engrafted and were still alive at D100, translating into a cumulative incidence at 2 years of cGvHD of 61% (95% Confidence Interval [CI]: 5468), including limited and extensive cGvHD in 38% (95% CI: 23-53) and in 23% (95% CI: 10%-36%) patients respectively. Two cases of primary cGvHD were diagnosed following planned immune-intervention: one was a limited cGVHD after DLI administration for early progression at D35 and one was an extensive cGvHD after cessation of CsA due to D90 MRD positivity. Seven of the 42 patients died. Causes of death were extensive cGvHD (n=2), pulmonary aspergillosis plus Pneumocystis jiroveci pneumonia associated with limited cGvHD (n=1) and early cytomegalovirus infection (n=1) in a patient who received alemtuzumab in the last weeks prior to transplantation. The three remaining deaths were related to disease progression with Richter transformation. Moreover three patients presented severe complications, namely two polyradiculopathy and one Epstein–Barr virusinduced lymphoproliferative disease. With a median follow-up of survivors of 36 months (range, 19-53) the 3-year OS, PFS, and NRM were 86.9% (95% CI: 70.8-94.4), 62.9% (95% CI: 45.8-75.9) and 9.5% (95% CI: 3.7-23.4) respectively. Ten patients had progression occurring after a median of 12 months (range, 1-34). 1869

O. Tournilhac et al.

The 3-year cumulative incidence of relapse was 29.6% (95% CI: 17.3-47.7) (Figure 1). Salvage therapy was delivered in patients who relapsed after donor engraftment; eight patients received ibrutinib; six of them are still in remission at the last follow-up (32 to 52 months) while two had a transient response followed by progression with Richter transformation. (Table 2).

At M12, 27 patients achieved MRD status, including 23 patients with an undetectable MRD (MRD < limit of detection), seven patients remained MRD , eight patients were not evaluable because either early toxic death (n=4) or other reason including graft rejection (n=2), EppsteinBarr virus-induced lymproliferation (n=1) and early relapse (n=1). Thus, at M12, MRD status was achieved in 64% (27 of 42) if we consider all patients and in 77% (27of 35) if we take into consideration all 34 patients assessed at this time point and the patient who experienced a clinical relapse at 1 month (and thus not subject to systematic MRD assessment but considered as failure) versus 14.2% before transplantation. Most patients remained MRD early after transplantation and progressively translated to MRD within the first 6 months posttransplantation. (Figure 2). Nine of the 13 (69%) D90

MRD patients who had no significant GvHD but who had an early CsA withdrawal according to the protocol, managed to reach a MRD status. For the 39 patients who engrafted and were alive after M1, MRD kinetics followed four distinctive patterns. (Figure 3). The pattern A (n=6) is constituted of the pretransplantation MRD patients. Two of these patients relapsed, one at 12 and one at 19 months. The pattern B (n=11) comprised the patients who converted from pretransplantation MRD to post-transplantation MRD status within 3 months without any immune-intervention. One pattern-B patient with M12 MRD close to the positivity threshold relapsed at 13 months. The pattern C (n=15) is constituted of the patients with pre-transplantation MRD who remained MRD during the first 3 months but became MRD either after CsA tapering and withdrawal (n=12) or after cGvHD (n=3). Two pattern-C patients relapsed at 23 and 34 months. The pattern D (n=7) comprised the patients with a pre-transplantation MRD status who remained MRD despite cGvHD (n=1) or immune-intervention including CsA tapering and withdrawal (n=6) followed by DLI for five of them. Progression was observed in five pattern-D patients including three Richter transformations, each occurring within the first 13 months. The outcome of all four panels is represented in the Online Supplementary Table S3.

Minimal residual disease (MRD) status at 12 months and MRD kinetics after hematopoietic stem cell transplantation neg

pos

neg

pos

neg

pos

neg

pos

neg

pos

neg

pos

Figure 1. Post-transplant outcome of the 42 chronic lymphocytic leukemia transplanted patients. Kaplan-Meier estimates of (A) overall survival, (B) progressionfree survival. Calculated probability of (C) non relapse mortality and (D) cumulative incidence of relapse after hematopoietic stem cell transplantation.

1870

haematologica | 2021; 106(7)

Allograft and preemptive immuneintervention in CLL

Impact of minimal residual disease on outcome In Mantel-Byar analysis, evaluating MRD as a timedependent variable, achievement of the MRD status regardless of the time point, was predictive of an improved PFS, Hazard ratio (HR) 0.18 (range, 0.06-0.60), P=0.005, and OS, HR 0.18 (range, 0.03-0.98), P=0.047 along with a reduction of CIR, HR 0.14 (range, 0.040.539, P=0.004. (Figure 4).

relapse risk, HR 0.16 (range, 0.02-1.22) P=0.08 but had no impact on OS, P=0.18.

neg

Chimerism evaluation The chimerism analyzed on unselected blood cells had no impact on outcome. Conversely T-cell donor engraftment (≥95% donor T cells) tended to be associated with higher PFS, HR 0.16 (range, 0.02-1.37), P=0.09, and lower

Discussion Since the first descriptions, allogeneic HSCT has long been the only curative treatment for CLL. Its development has benefited greatly from the advent of reduced-intensity conditioning that can be proposed until the age of 70. The availability of new alternative therapies, including both BCR and BCL-2 inhibitors have in high-risk patients taken the place of allogeneic HSCT and delayed this strategy until later in the management of the disease. Consequently, the

Table 2. Treatment and follow-up of patients in relapse after hematopoietic stem cell transplantation.

Pt#

Relapse treatment

35 2 8 19 29 13 4 15 28 18

12 19 12 34 23 11* 1 12 12 9

13 34 15 34 23 19 6 12 13 10

Ibrutinib (M13-ongoing) DLI(1), failure followed by ibrutinib (M36-ongoing) DLI(3) followed by Ibrutinib (M18 - ongoing) Ibrutinib (M34 - ongoing) DLI(2) response, followed by ruxolitinib (ongoing) for GvHD RCHOP, with initial PR, RDHAP, irradiation °Ibrutinib (M15-M45) with initial PR then PD*, RDHAC °Ibrutinib (M13-M24) with initial PR then PD*, RCHOP Ibrutinib (M13-ongoing) Ibrutinib (M10-ongoing)

Last FU

Status

29+ 52+ 48+ 42+ 36+ 35 47 43 32+ 41+

PR, MRD CR, MRD CR, MRD CR, MRD CR, MRD PD* PD* PD* CR, MRD PR, MRD

pos

neg

pos

Pt#: patient number ; ASCT: allogenic hematopoietic stem cell transplantation; T1: time from ASCT to relapse (months) ; T2: time from ASCT to relapse treatment (months) ; CR: complete response ; PR: partial response ; PD: progressive disease ; *: Richter transformation ; DLI: donor lymphocytes infusion (number) ; R-CHOP: rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone ; R-DHAC: rituximab, dexamethasone, cytarabine, carboplatinum; GvHD: graft versus host disease; FU: follow-up; M13: 13 months; M15: 15 months; M24: 24 months; M34: 34 months; M45: 45 months. , ° Patients#4 and #15 had received preemptive DLI before relapse treatment as part of the immune-intervention as per study protocol.

Figure 2. Post-transplantation minimal residual disease evaluation. At 12 months (M12), 27 of 42 (64%) patients were minimal residual disease negative (MRDneg), 7 of 42 (17%) patients remained MRD positive (MRD ), 8 of 42 (19%) patients were not evaluable because either prior early toxic death (n=4) and 4 of 42 patients (9.5%) or other reasons including graft rejection (n=2), Eppstein-Barr virus lymproliferation (n=1) and early relapse (n=1). pos

haematologica | 2021; 106(7)

1871

O. Tournilhac et al. A

Figure 3. Patterns of minimal residual disease response of the 39 patients who engrafted and were alive after 1 month. Pattern A: patients with pre-transplant minimal residual disease negative (MRDneg) status (n=6), Pattern B: patients who converted to MRDneg within 3 months post-transplant without any immune-intervention (n=11). Pattern C : patients who converted to MRD upon immune-intervention (cyclosporine A [CsA] withdrawal only) or graft-versus-host disease (GvHD) (n=16) Pattern D: patients who remained MRD positive (MRD ) during follow-up despite immune-intervention (CsA withdrawal and donor lymphocyte infusion [DLI]) or GvHD (n=7). Solid blue line: negativity limit of MRD (<0.01%). UD : undetectable MRD (MRD < limit of detection [LOD]). neg

pos

number of allogeneic HSCT for CLL has considerably decreased since 2015, both in the United States27 and Europe.28 BCR and BCL-2 inhibitors allow control of relapsed CLL with a response duration exceeding those reported after immunochemotherapy.29,30 However, relapses are the rule, particularly in patients with adverse molecular31,32 and/or complex karyotype.33,34 For such patients, CAR-T cells are also a hope.35,36 However, while this new option is very promising in several hematological diseases, in the 134 highly pre-treated CLL reported to date, the complete response rate remains 20 to 30%, with a median PFS of 18% at 18 months.37 This approach is associated with significant acute toxicity, but does not present, in contrast to allogeneic HSCT, a risk of GvHD. Hence, long-term results in large cohorts of CLL patients treated by CAR-T cells are currently needed, and allogeneic HSCT is still a valid option in CLL for selected patients.38 We report the first trial evaluating prospectively an approach of post-transplantation MRD-driven immuneintervention for CLL. M12 MRD associated with a reduced risk of relapse and an improvement of disease-free survival, was chosen as the primary endpoint.10-14 We hypothesized that early CsA tapering potentially followed by DLI in case of a post-transplantation MRD status could increase the incidence of MRD status at M12 and as a consequence could reduce the risk of relapse. Conversely, for neg

pos

neg

1872

patients with a post-transplantation MRD status, CsA administration would be extended for a longer period to reduce the risk of chronic GvHD. In order to minimize severe GvHD incidence, we selected the same ATG containing conditioning regimen as previously evaluated in a large multicenter study performed in a similar age population.18 Overall, in this population with a median age close to 60 years, we observed less than 25% extensive cGvHD which appears lower than in previous series of HSCT in CLL and can be considered very acceptable in the context of allogeneic treatment of high-risk diseases.7-10 Moreover, only four deaths were related to either GvHD or infection and the 2-year NRM less than 10%, favorably compares with those varying from 17% to 27% reported in the main series of reduced intensity conditioning transplant in CLL.7-10 In this trial low NRM highly contribute to impressive 3-year OS close to 90%. We show that post-transplantation MRD-driven immune-intervention is feasible in the setting of a multicenter trial. MRD evaluation was centralized and performed by a sensitive method of high-resolution ten-color flow cytometry. Results were available within 48 hours, allowing a rapid adaptation of the immunosuppressive therapy. Immune-intervention was conducted in accordance to the protocol in 83% of the evaluable patients; in the remaining patients CsA was withdrawn earlier due to graft-failure, neg

haematologica | 2021; 106(7)

Allograft and preemptive immuneintervention in CLL

Figure 4. Impact of minimal residual disease negative (MRD ) status achievement on post transplant outcome according to the Mantel-Byar method illustrated by Simon-Makuch plots (MRD status as a time-dependent event). (A) overall survival, (B) progression-free survival and (C) cumulative incidence of relapse. neg

mixed chimerism or renal failure. The immune-intervention related toxicity was low, with only two cases of GvHD occurring after CsA withdrawal or DLI applied as per protocol. In an Intent-To-Treat analysis, the primary end-point of M12 MRD status has been achieved in 64 % of the 42 transplanted patients and in 79% of the 34 patients who actually had a M12 MRD evaluation. This result favorably compares with both prospective10 and retrospective11-14 studies reporting 48% to 71% MRD status at 6 to 12 months after HSCT.10-14 Particularely our results are in line with one large single-center retrospective analysis of 77 allografted CLL patients submitted to immune modulation based on MRD evaluation.6 In this latter study M12 MRD clearance was achieved in 56% overall and 84% of all patients evaluable for M12 MRD status and the 3-year relapse incidence was 26%. Our data argue for the benefit of an early preemptive immune-intervention based on MRD evaluation. Thus, early CsA withdrawal applied in D90 MRD patients translated into MRD status at M12 in 69% of them. Moreover, most patients with D90 MRD status and GvHD, spontaneously switched to MRD at M12, highlighting the role of allogeneic reaction in the control of the disease. Conversely, in this context of early preemptive immune-intervention we failed to show a benefit of DLI to convert MRD from positive to negative, but three of five patients were already in clinical progression at the time of infusion. Finally, considering the non-randomized nature of neg

neg

pos

neg

pos

neg

haematologica | 2021; 106(7)

the trial, our data suggest that a MRD-driven CsA withdrawal can provide durable MRD clearance, improve GvHD leukemia effect and avoid progression. In several studies, the M12 MRD status was associated with lower incidence of relapse.13-15,39 As half of the progression occurred before M12 in our series, we chose to analyze the impact of MRD using MRD as a time dependent variable. We confirm here the strong correlation between an MRD status achievement regardless of time point and both low progression and better PFS. Interestingly, we also show that MRD status achievement translates into better survival. The impressive post-transplant OS closed to 90% at 3 years in this high-risk CLL population could also be explained by the possibility opened to physician of treating post-allograft relapses with ibrutinib. It should also be noted that the three patients who died from CLL-related cause had Richter's syndrome, including two escaping therapy with ibrutinib. These data lead us to propose early additional therapy in patients who display an MRD status despite either MRDdriven CsA withdrawal or chronic GvHD, or in the rare patients who could experience disease despite MRD status achievement. Recent reports show the efficacy of ibrutinib in post-HSCT CLL relapse without limiting toxicity or GvHD, as also observed in our series.40,41 This treatment should be evaluated preemptively in patients who fail to achieve negative MRD after CsA cessation. In conclusion, this report shows the feasibility of MRDneg

neg

pos

neg

1873

O. Tournilhac et al.

driven immune-intervention following ATG-based RIC allogeneic HSCT in CLL. These data highlight the importance to repeatedly monitor post-HSCT MRD to guide early CsA discontinuation in patients with D90 MRD and without GvHD. However, as we report the results of this study in 2020, we must emphasize that the entry criteria were based on the EBMT 2007 recommendations, which no longer represent current practice. Particularly, this is the case for untreated patients with del(17p) and/or TP53 mutation who represent 26% of the study's enrollment. A presentation of post-transplant outcome excluding these 11 patients is shown in the Online Supplementary Appendix (Online Supplementary Figure S3). Allogeneic HSCT indications have evolved in 2014 under the impulse of the European Research Initiative on CLL (ERIC) and EBMT with of a new decisional algorithm according to patient biology and prior treatment with BCR and BCL-2 inhibitors,42 the feasibility of which has just been reported in a recent analysis.43 The pre-emptive immune modulation based on post-transplant MRD, as described in our study in patients who were 90% naive of BCR and/or BCL-2 treaments, should also be effective in patients pretreated with such agents, but this will have to be demonstrated. pos

Disclosures OT has received travel grant, scientific support, and honorarium

References 1. Byrd JC, Furman RR, Coutre SE, et al. Targeting BTK with ibrutinib in relapsed chronic lymphocytic leukemia. N Engl J Med. 2013;369(1):32-42. 2. Furman RR, Sharman JP, Coutre SE, et al. Idelalisib and rituximab in relapsed chronic lymphocytic leukemia. N Engl J Med. 2014;370(11):997-1007. 3. Roberts AW, Davids MS, Pagel JM. et al. Targeting BCL2 with venetoclax in relapsed chronic lymphocytic leukemia. N Engl J Med. 2016;374(4):311-322. 4. Gribben JG. How and when I do allogeneic transplant in CLL. Blood 2018; 132(1):31-39. 5. Hallek M. Chronic lymphocytic leukemia: 2017 update on diagnosis, risk stratification, and treatment. Am J Hematol. 2017;92(9):946-965. 6. Hahn M, Böttcher S, Dietrich S, et al. Allogeneic hematopoietic stem cell transplantation for poor-risk CLL: dissecting immune-modulating strategies for disease eradication and treatment of relapse. Bone Marrow Transplant. 2015;50(10):1279-1285. 7. Khouri IF, Bassett R, Poindexter N, et al. Nonmyeloablative allogeneic stem cell transplantation in relapsed/refractory chronic lymphocytic leukemia. Cancer. 2011;117(20):4679-4688. 8. Michallet M, Socié G, Mohty M, et al. Rituximab, fludarabine, and total body irradiation as conditioning regimen before allogeneic hematopoietic stem cell transplantation for advanced chronic lymphocytic leukemia: long-term prospective multicenter study. Exp Hematol. 2013;41(2):127-133. 9. Sorror ML, Storer BE, Sandmaier BM, et al. Five-year follow-up of patients with advanced chroniclLymphocytic leukemia treated with allogeneic hematopoietic cell transplantation after nonmyeloablative con-

1874

for board participation from Amgen, Roche, Jansen, Abbevie and Gilehead; MLGT has receieved honorarium for board participation from Alexion; VL has received board participation, speaker bureau, honorarium from Roche, Abbevie, Gilehead, Amgen and Janssen. Contributions OT and ND designed and performed research including patients’ care, coordinated the study and wrote the paper; MLGT and CQ performed MRD flow cytometry assessment; PC performed molecular chimerism; SNGQ, EF, PC, FLI, GLD, DM, CT, CB, CO, PT, RR, GG, LV, CS, JOB, VL performed research especially patients’ care; RL, LC, and AC worked on study design, study application and national coordination and BP performed statistical data analysis. Acknowledgments The study was supported by a National Grant from the Fondation ARC. Logistic support was provided through the Fonds de recherche clinique en hématologie Force Hémato. Les Laboratoires Pierre Fabre provided financial research support for the study, but did not participate in the conduct of the study or data and results analysis. The authors thank Dr Reza Tabrizi and Dr Oumeladi Reman for their devoted implication in patient care and the clinical research assistants of the FILO group for the onsite monitoring process performed during the study.

ditioning. J Clin Oncol. 2008;26(30):49124920. 10. Dreger P, Döhner H, Ritgen M, et al. Allogeneic stem cell transplantation provides durable disease control in poor-risk chronic lymphocytic leukemia: long-term clinical and MRD results of the German CLL Study Group CLL3X trial. Blood. 2010;116(14):2438-2447. 11. Algrin C, Golmard J-L, Michallet M, et al. Flow cytometry minimal residual disease after allogeneic transplant for chronic lymphocytic leukemia. Eur J Haematol. 2017; 98(4):363-370. 12. Farina L, Carniti C, Dodero A, et al. Qualitative and quantitative polymerase chain reaction monitoring of minimal residual disease in relapsed chronic lymphocytic leukemia: early assessment can predict longterm outcome after reduced intensity allogeneic transplantation. Haematologica. 2009;94(5):654-662. 13. Logan AC, Zhang B, Narasimhan B, et al. Minimal residual disease quantification using consensus primers and high-throughput IGH sequencing predicts post-transplant relapse in chronic lymphocytic leukemia. Leukemia. 2013;27(8):1659-1665. 14. Moreno C, Villamor N, Colomer D, et al. Clinical significance of minimal residual disease, as assessed by different techniques, after stem cell transplantation for chronic lymphocytic leukemia. Blood. 2006;107(11):4563-4569. 15. Ritgen M, Böttcher S, Stilgenbauer S, et al. Quantitative MRD monitoring identifies distinct GVL response patterns after allogeneic stem cell transplantation for chronic lymphocytic leukemia: results from the GCLLSG CLL3X trial. Leukemia. 2008;22(7):1377-1386. 16. Dreger P, Corradini P, Kimby E, et al. Indications for allogeneic stem cell trans-

plantation in chronic lymphocytic leukemia: the EBMT transplant consensus. Leukemia. 2007;21(1):12-17. 17. Sorror ML, Maris MB, Storb R, et al. Hematopoietic cell transplantation (HSCT)specific comorbidity index: a new tool for risk assessment before allogeneic HSCT. Blood. 2005;106(8):2912-2919. 18. Blaise D, Devillier R, Lecoroller-Sorriano AG, et al. Low non-relapse mortality and long-term preserved quality of life in older patients undergoing matched related donor allogeneic stem cell transplantation: a prospective multicenter phase II trial. Haematologica. 2015;100(2):269-274. 19. Hallek M, Cheson BD, Catovsky D, et al. Guidelines for the diagnosis and treatment of chronic lymphocytic leukemia: a report from the International Workshop on Chronic Lymphocytic Leukemia updating the National Cancer Institute–Working Group 1996 guidelines. Blood. 2008;111(12):5446-5456. 20. Rawstron AC, Villamor N, Ritgen M, et al. International standardized approach for flow cytometric residual disease monitoring in chronic lymphocytic leukaemia. Leukemia. 2007;21(5):956-964. 21. Rawstron AC, Böttcher S, Letestu R, et al. Improving efficiency and sensitivity: European Research Initiative in CLL (ERIC) update on the international harmonised approach for flow cytometric residual disease monitoring in CLL. Leukemia. 2013;27(1):142-149. 22. Rawstron AC, Fazi C, Agathangelidis A, et al. A complementary role of multiparameter flow cytometry and high-throughput sequencing for minimal residual disease detection in chronic lymphocytic leukemia: an European Research Initiative on CLL study. Leukemia. 2016;30(4):929-936. 23. Glucksberg H, Storb R, Fefer A, et al. Clinical

haematologica | 2021; 106(7)

Allograft and preemptive immuneintervention in CLL

manifestations of graft-versus-host disease in human recipients of marrow from HL-Amatched sibling donors. Transplantation. 1974;18(4):295-304. 24. Przepiorka D, Weisdorf D, Martin P, et al. 1994 consensus conference on acute GVHD grading. Bone Marrow Transplant. 1995;15 (6):825-828. 25. Mantel N, Byar DP. Evaluation of responsetime data involving transient states: an illustration using heart-transplant data. J Am Stat Assoc. 1974;69:81-86. 26. Simon R, Makuch RW. A non-parametric graphical representation of the relationship between survival and the occurrence of an event: application to responder versus nonresponder bias. Stat Med. 1984;3(1):35-44. 27. Kharfan-Dabaja MA, Kumar A, Hamadani M, et al. Clinical practice recommendations for use of allogeneic hematopoietic cell transplantation in chronic lymphocytic leukemia on behalf of the guidelines committee of the American Society for Blood and Marrow Transplantation. Biol Blood Marrow Transplant. 2016;22(12):2117-2125. 28. Tournilhac O, van Gelder N, Dreger P, et al. The 10-years EBMT landscape of allogeneic hematopoietic cell transplantation (alloHCT) for chronic lymphocytic leukemia. 46th Annual Meeting of the European Society for Blood and Marrow Transplantation. Madrid, 2020 29. Chanan-Khan A, Cramer P, Demirkan F, et al. Ibrutinib combined with bendamustine and rituximab compared with placebo, bendamustine, and rituximab for previously treated chronic lymphocytic leukaemia or

haematologica | 2021; 106(7)

small lymphocytic lymphoma (HELIOS): a randomised, double-blind, phase 3 study. Lancet Oncol. 2016;17(2):200-211. 30. Deng R, Gibiansky L, Lu T. Bayesian population model of the pharmacokinetics of venetoclax in combination with rituximab in patients with relapsed/refractory chronic lymphocytic leukemia: results from the Phase III MURANO Study. Clin Pharmacokinet. 2019;58(12):1621-1634. 31. O'Brien S, Furman RR, Coutre S, et al. Single-agent ibrutinib in treatment-naïve and relapsed/refractory chronic lymphocytic leukemia: a 5-year experience. Blood. 2018;131(17):1910-1919. 32. Mato AR, Thompson M, Allan JN, et al. Real-world outcomes and management strategies for venetoclax-treated chronic lymphocytic leukemia patients in the United States. Haematologica. 2018;103(9):15111517. 33. Jain P, Thompson PA, Keating M, et al. Longterm outcomes for patients with chronic lymphocytic leukemia who discontinue ibrutinib. Cancer. 2017;123(12):2268-2273. 34. Anderson M, Tam C, Lew TE, et al. Clinicopathological features and outcomes of progression of CLL on the BCL2 inhibitor venetoclax. Blood. 2017;129(25):3362-3370. 35. Porter DL, Hwang W-T, Frey NV, et al. Chimeric antigen receptor T cells persist and induce sustained remissions in relapsed refractory chronic lymphocytic leukemia. Sci Transl Med. 2015;7(303):303ra139 36. Turtle CJ, Hay KA, Hanafi LA, et al. Durable molecular remissions in chronic lymphocytic Leukemia treated with CD19-specific

chimeric antigen receptor-modified T cells after failure of ibrutinib. J Clin Oncol. 2017;35(26):3010-3020. 37. Lemal R. Tournilhac O. State-of-the-art for CAR T-cell therapy for chronic lymphoid leukemia in 2019. J Immunother Cancer. 2019;7(1):202. 38. Dreger P, Ghia P, Schetelig J et al. High-risk chronic lymphocytic leukemia in the era of pathway inhibitors: integrating molecular and cellular therapies. Blood. 2018;132(9): 892-902. 39. Krämer I, Stilgenbauer S, Dietrich S, et al. Allogeneic hematopoietic cell transplantation for high-risk CLL: 10-year follow-up of the GCLLSG CLL3X trial. Blood. 2017;130(12):1477-1480. 40. Link CS, Teipel R, Heidenreich F, et al. Durable responses to ibrutinib in patients with relapsed CLL after allogeneic stem cell transplantation. Bone Marrow Transplant. 2016;51(6):793-798. 41. Ryan CE, Sahaf B, Logan AC, et al. Ibrutinib efficacy and tolerability in patients with relapsed chronic lymphocytic leukemia following allogeneic HSCT. Blood. 2016;128(25):2899-2908. 42. Dreger P, Schetelig J, Andersen N, et al. Managing high-risk CLL during transition to a new treatment era: stem cell transplantation or novel agents? Blood. 2014;124(26): 3841-3849. 43. Hoffmann A, Dietrich S, Hain S, et al. Allogeneic transplantation in high-risk chronic lymphocytic leukemia: a single-center, intent-to-treat analysis. Haematologica. 2019;104(7):e304-e306.

1875

ARTICLE Ferrata Storti Foundation

Haematologica 2021 Volume 106(7):1876-1882

Chronic Myeloid Leukemia

Fatigue in chronic myeloid leukemia patients on tyrosine kinase inhibitor therapy: predictors and the relationship with physical activity Lando Janssen,1,2 Nicole M.A. Blijlevens,2 Meggie M.C.M. Drissen,1 Esmée A. Bakker,1,3 Malou A.H. Nuijten,1 Jeroen J.W.M. Janssen,4 Silvie Timmers1,5 and Maria T.E. Hopman1

1 Radboud Institute for Health Sciences, Department of Physiology, Radboud University Medical Center, Nijmegen, the Netherlands; 2Radboud Institute for Health Sciences, Department of Hematology, Radboud University Medical Center, Nijmegen, the Netherlands; 3Research Institute for Sport and Exercise Sciences, Liverpool John Moores University, Liverpool, UK; 4Cancer Center Amsterdam, Department of Hematology, Amsterdam University Medical Center, location VUmc, Amsterdam, the Netherlands and 5Human and Animal Physiology, Wageningen University, Wageningen, the Netherlands

ABSTRACT

Correspondence: MARIA HOPMAN: Maria.Hopman@radboudumc.nl NICOLE BLIJLEVENS Nicole.Blijlevens@radboudumc.nl Received: January 17, 2020. Accepted: June 30, 2020. Pre-published: July 2, 2020. https://doi.org/10.3324/haematol.2020.247767

1876

atigue is a common side effect of tyrosine kinase inhibitor (TKI) therapy in patients with chronic myeloid leukemia (CML). However, the prevalence of TKI-induced fatigue remains uncertain and little is known about predictors of fatigue and its relationship with physical activity. In this study, 220 CML patients receiving TKI therapy and 110 genderand age-matched controls completed an online questionnaire to assess fatigue severity and fatigue predictors (Part 1). In addition, physical activity levels were objectively assessed for 7 consecutive days in 138 severely fatigued and non-fatigued CML patients using an activity monitor (Part 2). We demonstrated that the prevalence of severe fatigue was 55.5% in CML patients and 10.9% in controls (P<0.001). We identified five predictors of fatigue in our CML population: age (odds ratio [OR] 0.96, 95% confidence interval [95% CI]: 0.93-0.99), female gender (OR 1.76, 95% CI: 0.92-3.34), Charlson Comorbidity Index (OR 1.91, 95% CI: 1.16-3.13), the use of comedication known to cause fatigue (OR 3.43, 95% CI: 1.58-7.44), and physical inactivity (OR of moderately active, vigorously active and very vigorously active compared to inactive 0.43 (95% CI: 0.12-1.52), 0.22 (95% CI: 0.06-0.74), and 0.08 (95% CI: 0.02-0.26), respectively). Objective monitoring of activity patterns confirmed that fatigued CML patients performed less physical activity of both light (P=0.017) and moderate to vigorous intensity (P=0.009). In fact, compared to the non-fatigued patients, fatigued CML patients performed 1 hour less of physical activity per day and took 2,000 fewer steps per day. Our findings facilitate the identification of patients at risk of severe fatigue and highlight the importance of setting reduction of fatigue as a treatment goal in CML care. This study was registered at The Netherlands Trial Registry, NTR7308 (Part 1) and NTR7309 (Part 2).

Introduction Survival of patients with chronic myeloid leukemia (CML) has improved significantly since the introduction of tyrosine kinase inhibitors (TKI) in 2001. This has translated into a life expectancy that is almost the same as that of the general population.1 Nevertheless, health-related quality of life (QoL) of CML patients is inferior to that of the general population, which is mainly the result of TKI-induced fatigue.2 CML patients require (potentially) lifelong treatment, which results in a persistent trigger for fatigue. Therefore, it is of the utmost importance to focus on prevention and management of fatigue in the care of CML patients. This is further supported by haematologica | 2021; 106(7)

Fatigue in CML patients

the finding that adverse events lead to lower TKI treatment adherence3 and therefore to poorer disease control.4 Although TKI-induced fatigue is one of the most frequently reported adverse effects,5,6 its actual prevalence is unknown because of the heterogeneity in measurement techniques used across studies and it has not been compared to that in the general population. Furthermore, clinicians are unable to identify patients at risk of fatigue since predictors have never been assessed in this specific population of patients. Although a variety of predictors of fatigue, such as gender, age and socioeconomic status have been described in literature,7-9 it is unknown whether these predictors, even when obtained in cancer populations, can be extrapolated to this unique group of CML patients on TKI therapy. Aside from these unmodifiable predictors of fatigue, physical activity has been identified as a modifiable predictor of fatigue in several patient populations.10 The aim of this multicenter observational study was threefold. First, to assess the prevalence of fatigue in CML patients on TKI therapy compared to that in the general population. Second, to identify predictors of fatigue in CML patients. Third, to objectively assess physical activity levels and compare these between fatigued and non-fatigued patients. In this way, we will facilitate the identification of patients at risk of fatigue and provide insight into the association between fatigue and physical activity in the CML population.

Methods CML patients aged ≥18 years who were receiving TKI therapy were invited to complete an online questionnaire to assess the prevalence and predictors of TKI-induced fatigue (Part 1). Control subjects were selected from a database consisting of over 20,000 subjects without CML who participated in previous research at the Department of Physiology at Radboud University Medical Center (Nijmegen, the Netherlands). Controls were matched for gender and age (±3 years) in a 1:2 ratio to the CML patients. A subgroup of CML patients was asked to wear an activity monitor in order to measure physical activity levels objectively (Part 2). Patients were recruited through the outpatient clinics at the Radboud University Medical Center and Amsterdam University Medical Center (Amsterdam, the Netherlands), and via CMyLife, a Dutch online platform for CML patients.11 Informed consent was obtained from all participants. This study was approved by the Medical Review Ethics Committee region Arnhem-Nijmegen and registered at The Netherlands Trial Registry with numbers NTR7308 (Part 1) and NTR7309 (Part 2).

Part 1: questionnaire Fatigue severity was measured by the Checklist Individual Strength subscale “subjective experience of fatigue” (CIS-fatigue), which is a validated fatigue questionnaire assessing fatigue over the preceding 2 weeks.12 A score of 35 or above was considered as severe fatigue. The following general characteristics were collected: age, gender, body mass index, education level, and marital status. Time since CML diagnosis, TKI type and dose, duration of TKI treatment, and disease control (major molecular response, defined as ≤0.1% BCR-ABL transcripts on the International Scale) were collected to assess CML-related medical history. The Charlson Comorbidity Index (CCI)13 was used to quantify participants’ medical comorbidities. Both over-the-counter and prescribed medication known to cause fatigue (e.g., benzodiazepines, opioids, βblockers, and metformin) were assessed. Lastly, potential lifestyle predictors were collected, including smoking, daily fluid and cafhaematologica | 2021; 106(7)

feine intake, alcohol consumption (beer and wine), and physical activity. Physical activity (defined as Metabolic Equivalent of Task [MET] min/week) was classified into four categories: inactive (<500 MET min/week), moderately active (500-1,499 MET min/week), vigorously active (1,500-2,999 MET min/week), and very vigorously active (>3,000 MET min/week).

Part 2: activity monitor Physical activity was measured with the activPAL3 micro (PAL Technologies Ltd., Glasgow, UK)14 in a subgroup of 143 CML patients. The sample size calculation was based on data from previous research published on differences in objectively assessed activity levels between fatigued and non-fatigued elderly subjects15 using a power of 80%, with a two-tailed a level of 0.05, an estimated effect size of 0.50 and a drop-out rate of 10%. Participants wore the activity monitor 24 hours per day for 7 consecutive days and were asked to maintain normal daily activities. In addition, employment status and total work time were reported. BCR-ABL transcript levels, hemoglobin concentration, white blood cell count and platelet count were extracted from the patients’ electronic records.

Statistical analysis Continuous data are reported as means ± standard deviation or median (interquartile range [IQR]) and categorical variables as counts and percentages. Logistic regression was performed to identify predictors of severe fatigue. Predictor variables with P values <0.10 in univariable analysis were selected for multivariable logistic regression analysis. Odds ratios (OR) with 95% confidence intervals (95% CI) were calculated to estimate the effect size. The optimal model was selected based on the discriminative ability, assessed by the area under the receiving operating characteristic curve, and calibration slope. Differences in activity patterns were tested using Student t tests for independent samples when data were normally distributed, and Wilcoxon rank sum tests when data were skewed. To correct for potential confounding factors, multivariable linear regression was used. All data were analyzed using SPSS (version 22.0, IBM, Armonk, NY, USA). Statistical significance was set at a P value <0.05. Detailed information on the questionnaire and activity monitor is provided in the Online Supplementary Methods S1.

Results A total of 357 participants were enrolled in the study, consisting of 247 CML patients and 110 controls. Figure 1 shows a schematic flowchart of participants in the two parts of the study. Two-hundred twenty CML patients (58% females, mean age 56 ± 13 years) and 110 gender- and age-matched controls completed the online questionnaire between May 2018 and May 2019 (Part 1). Of these 220 patients, 216 (98.2%) had no missing data and were included in the multivariable regression analysis. The activity monitor was worn by 143 CML patients, but five patients were excluded from analysis because of an invalid number of days registered by the activity monitor (Part 2).

Part 1: prevalence and predictors of severe fatigue The prevalence of severe fatigue was 55.5% in the CML patients and 10.9% in the matched controls (P<0.001). Reported QoL was significantly poorer in CML patients than in controls (mean QoL scores 6.9 ± 1.5 and 8.1 ± 1.0, respectively; P<0.001), and also in severely fatigued CML patients when compared to patients without severe fatigue 1877

L. Janssen et al.

(mean QoL 6.1 ± 1.3 and 7.8 ± 1.1, respectively; P<0.001). Table 1 shows the univariable and multivariable logistic regression analyses of the predictors of severe fatigue in CML patients. The final multivariable model included: age (OR 0.96, 95% CI: 0.93-0.99; P=0.004), female gender (OR 1.76, 95% CI: 0.92-3.34; P=0.09), CCI (OR 1.91, 95% CI: 1.16-3.13); P=0.011), the use of comedication known to cause fatigue (OR 3.43, 95% CI: 1.58-7.44; P=0.002), and physical activity at different intensities compared to physical inactivity, including moderately active (OR 0.43, 95% CI: 0.12-1.52; P=0.19), vigorously active (OR 0.22, 95% CI:

0.06-0.74; P=0.014), and very vigorously P<0.001). The area under the curve of the final model was 0.79 (95% CI: 0.73-0.85) and the calibration slope was 1.01. The predicted severe fatigue and associated observed fatigue are displayed in Figure 2, showing a positive predictive value of 73%, a negative predictive value of 68%, a sensitivity of 76%, and a specificity of 65%. Our model can be described by the following equation: log-odds = 1.98 – 0.04 * age + 0.56 * gender (male=0, female=1) + 0.65 * CCI + 1.23 * use of comedication known to cause fatigue (no=0, yes=1) – physical activity level (inactivity=0, moderately

Figure 1. Flowchart of the study. CML: chronic myeloid leukemia.

Figure 2. Predicted and observed severe fatigue. The predicted severe fatigue of the multivariable model is plotted against the observed severe fatigue. Each dot represents an individual.

1878

haematologica | 2021; 106(7)

Fatigue in CML patients

active=0.85, vigorously active=1.53, very vigorously active=2.58).

Part 2: activity patterns in fatigued and non-fatigued patients Table 2 shows the characteristics, including hematologic values, of the 138 CML patients who wore the activity monitor and were included in the final analysis. In line with the first part of the study, there was a higher propor-

tion of female patients in the severely fatigued group than in the non-fatigued group (61% vs. 44%, respectively; P=0.039). The employment status did not differ significantly across the groups (53% vs. 66% employed in the fatigued and non-fatigued groups, respectively; P=0.20). However, total work time was significantly shorter in both severely fatigued male and female patients than in non-fatigued patients (males: 29 ± 12 h/week vs. 38 ± 12 h/week, respectively; P=0.048; females: 15±8 h/week vs.

Table 1. Univariable and multivariable logistic regression analysis of general characteristics, medical history, and lifestyle factors as potential predictors of severe fatigue in patients with chronic myeloid leukemia.

Univariable analysis CML + severe CML + severe OR (95% CI) fatigue (N=122) fatigue (N=98)

P-value

Controls (N=110)

CML total (N=220)

56 ± 13 58 25.2 ± 3.9

56 ± 13 58 26.2 ± 4.3

55 ± 13 66 26.5 ± 4.6

58 ± 12 48 25.8 ± 4.0

0.98 (0.96-1.00) 2.14 (1.24-3.70) 1.04 (0.98-1.11)

0.07 0.006 0.24

10 36 55 81

14 42 44 83

11 49 40 81

17 34 49 85

REF 2.38 (1.03-5.50) 1.34 (0.59-3.05) 0.78 (0.38-1.59)

0.043 0.49 0.49

64 (27-131)

61 (27-116)

66 (25-140)

1.00 (0.99-1.00)

0.46

38 27 21 9 5

34 32 19 10 6

44 21 24 7 4

REF 1.90 (0.96-3.76) 1.05 (0.51-2.15) 1.80 (0.65-5.01) 1.83 (0.50-6.74)

0.07 0.90 0.26 0.36

400 (400-400) 100 (700-100) 600 (400-600) 400 (300-500) 30 (15-45)

400 (400-400) 100 (50-100) 600 (400-600) 400 (300-500) 30 (15-45)

400 (400-400) 100 (85-100) 600 (400-600) 400 (300-400) 30 (19-42)

1.00 (1.00-1.00) 0.98 (0.95-1.00) 1.00 (1.00-1.00) 1.00 (0.99-1.01) 0.99 (0.89-1.09)

0.77 0.13 0.45 0.58 0.77

36 (12 – 69)

24 (6-60)

36 (12-72)

1.00 (1.00-1.01)

0.48

80 2 (2-3)

75 2 (2-3)

86 2 (2-2)

0.47 (0.21 – 1.0) 1.67 (1.10-2.53)

8 1806 ± 1760 336 ± 197

5 1642 ± 545 326 ± 175

7 1601 ± 557 312 ± 175

6 19 28 47 33

16 25 30 30 46

Multivariable analysis OR (95% CI) P-value

General characteristics

Age, years Gender, % female BMI, kg/m2 Education level, % Low Mediate High Marital status, % married

0.96 (0.93-0.99) 1.76 (0.92-3.34)

0.004 0.09

0.05 0.017

1.91 (1.16-3.13)

0.011

2.89 (1.51-5.54)

0.001

3.43 (1.58-7.44)

0.002

2 1694 ± 528 344 ± 173

3.40 (0.71-16.4) 1.00 (1.00-1.00) 1.00 (1.00-1.00)

0.13 0.21 0.17

0.61 (0.28-1.34)

0.22

22 32 28 17 49

4 16 33 47 42

REF 0.39 (0.12 – 1.29) 0.16 (0.05 – 0.52) 0.07 (0.02 – 0.22) 1.35 (0.79 – 2.30)

Medical history

Time since diagnosis, months NA TKI type, % NA Imatinib Dasatinib Nilotinib Bosutinib Ponatinib TKI dose (mg/day) NA Imatinib Dasatinib Nilotinib Bosutinib Ponatinib Treatment duration of current TKI, months NA BCR-ABLIS transcript level, % MMR NA CCI 0 (0-0) Comedication known to cause fatigue*, % 7 Lifestyle

Smoking status, % smoker Fluid intake, mL/day Caffeine intake, mg/day Alcohol consumption >1U/day, % Physical activity#, % Inactive Moderately active Vigorously active Very vigorously active Not performing sports, %

0.12 0.43 (0.12 – 1.52) 0.19 0.002 0.22 (0.06 – 0.74) 0.014 <0.001 0.08 (0.02 – 0.26) <0.001 0.28

Data are presented as mean ± standard deviation, percentages or median (interquartile range). *Benzodiazepines, opioids, beta-blockers, and metformin. #Physical activity was classified into four categories: inactive (<500 metabolic equivalent of task [MET] min/week), moderately active (500-1,499 MET min/week), vigorously active (1,500-2,999 MET min/week), and very vigorously active (>3,000 MET min/week). CML: chronic myeloid leukemia; OR: odds ratio; 95% CI: 95% confidence interval; BMI: body mass index; REF: reference group; TKI: tyrosine kinase inhibitor; IS: International scale; MMR: major molecular response; CCI: Charlson Comorbidity Index; NA: not applicable.

haematologica | 2021; 106(7)

1879

L. Janssen et al.

28 ± 10 h/week, respectively; P=0.001). Figure 3 shows the daily activity patterns of fatigued and non-fatigued patients categorized into sleeping, sitting, light intensity physical activity, and moderate to vigorous physical activity. Severely fatigued CML patients slept significantly longer than patients without fatigue (8.8 h/day [IQR 8.3-9.7] vs. 8.4 h/day [IQR 7.9-9.1]; P=0.006). Sitting time was not significantly different between the groups (P=0.43). Severely fatigued CML patients were significantly less

active when compared to non-fatigued patients as shown by both lower physical activity of light intensity (3.9 h/day [IQR 3.1-5.0] vs. 4.8 h/day [IQR 3.6-5.6]; P=0.017) and moderate to vigorous physical activity (0.9 h/day [IQR 0.6-1.2] vs. 1.1 h/day [IQR 0.8-1.3]; P=0.009). In line, step counts were significantly lower in CML patients with severe fatigue (7,464 ± 3,486 steps/day) than in patients without fatigue (9,393 ± 4,200 steps/day; P=0.004). After adjustment for gender and age, sleeping time remained longer and

Table 2. Characteristics of chronic myeloid leukemia patients with and without severe fatigue in part 2 of the study.

Age, years Gender, % female BMI, kg/m2 Time since diagnosis, months TKI type, % Imatinib Dasatinib Nilotinib Bosutinib Ponatinib TKI dose, (mg/day) Imatinib Dasatinib Nilotinib Bosutinib Ponatinib Treatment duration of current TKI, months BCR-ABLIS transcript level, % MMR Hemoglobin, mmol/L Male Female WBC count (x 109/L) Platelet count (x 109/L)

CML + fatigue (N=67)

CML – fatigue (N=71)

P-value

55 ± 13 61 26.8 ± 4.7 67 (30-162)

55 ± 16 44 26.0 ± 4.9 89 (39-153)

0.88 0.039 0.37 0.46 0.12

37 27 16 15 5

23 28 32 10 7

400 (350-400) 100 (50-100) 600 (300-600) 350 (275-500) 30 (15-30) 35 (12-61) 87.3

400 (400-400) 95 (70-100) 600 (600-600) 400 (400-400) 30 (15-37.50) 36 (16-79) 94.0

0.17 0.99 0.16 0.88 0.79 0.55 0.18

8.5 ± 0.7 7.7 ± 0.9 6.3 ± 1.6 220 ± 61

8.7 ± 0.9 8.0 ± 0.8 6.1 ± 1.7 221 ± 68

0.24 0.11 0.49 0.93

CML: chronic myeloid leukemia; BMI: body mass index; TKI: tyrosine kinase inhibitor; IS: international scale; MMR: major molecular response; WBC: white blood cell.

Figure 3. Activity patterns in severely fatigued and non-fatigued chronic myeloid leukemia patients. Daily time spent sleeping, sitting, performing light intensity physical activity and moderate to vigorous intensity physical activity is shown for both severely fatigued and non-fatigued chronic myeloid leukemia patients. Box and whisker plots represent the median, interquartile range, and 2.5-97.5 percentiles (lower and upper whiskers, respectively). *P<0.05, **P<0.01, ns: not significant.

1880

haematologica | 2021; 106(7)

Fatigue in CML patients

physical activity remained lower in fatigued patients (mean difference 0.55 h/day, 0.67 h/day, 0.22 h/day, and 1,938 steps/day for sleeping time, light intensity physical activity, moderate to vigorous intensity physical activity, and step counts, respectively; all P<0.05). Physical activity patterns were also analyzed for week and weekend days separately. Although there was no difference between the fatigued and non-fatigued group in sitting time during week days (9.7 h/day [IQR 8.9-11.3] and 10.0 h/day [IQR 8.8-11.0], respectively; P=0.73), we found a trend towards longer sitting time during weekend days in fatigued patients compared to non-fatigued ones (9.6 h/day [IQR 8.6-10.9] and 9.2 h/day (IQR 8.1-10.5), respectively; P=0.06). Severely fatigued patients slept longer and performed less physical activity (of light as well as moderate to vigorous intensity) on both week and weekend days (all P<0.05).

Discussion This is the first study to assess the prevalence and predictors of severe fatigue in CML patients receiving TKI treatment and to provide insight into the relationship between severe fatigue and physical activity in this population. The prevalence of fatigue in our CML population was 55%. Using multivariable logistic regression, we built a model with good discriminative ability and found five significant predictors of severe fatigue in our population: younger age, female gender, higher CCI, the use of comedication known to cause fatigue, and physical inactivity. Using physical activity monitors, we objectively confirmed that severely fatigued CML patients are less physically active during the day, with regard to both light and moderate to vigorous intensity activity, on both week and weekend days. These findings suggest that: (i) there is a subset of CML patients particularly prone to TKI-induced fatigue, and (ii) severely fatigued patients have reduced levels of physical activity. Over half of the CML patients in the present study experienced severe fatigue, which was significantly greater than that in matched controls and compared to TKI-induced fatigue rates reported in literature. In patients receiving imatinib, the prevalence of fatigue varied across large clinical trials from 34.5% (IRIS16) to 15.5% (CML IV17), 10% (DASISION18), 22% (ENESTnd19), 47% (BFORE20), and 20% (EPIC21). However, these trials were not designed to assess fatigue, which may explain the wide range of prevalence rates. We did not find a difference in fatigue prevalence between patients taking different TKI, which is in agreement with the findings of these large trials.18-20 Interestingly, severe fatigue was neither independently associated with treatment-related factors, such as TKI therapy dose and duration, nor with disease control. Although fatigue is a common sign of severe anemia, hemoglobin levels did not differ between fatigued and non-fatigued patients. Low hemoglobin levels were also not identified as an independent predictor of fatigue in patients with other hematologic malignancies.22 We found that fatigue was more often present in younger and female patients. In line with this, Efficace et al. found that the largest differences in health-related QoL between CML patients and the general population was among younger subjects.2 Contradictory findings are reported in the literature regarding the association between age and fatigue in other populations. For example, chronic fatigue haematologica | 2021; 106(7)

was more often present in younger breast cancer survivors,7 while older age has been identified as a risk factor for fatigue in both hematologic23 and non-hematologic24,25 cancer patients. Several studies showed an association between fatigue and gender in line with our results, with a more prominent risk for female cancer patients.26,27 Interestingly, compared to the general population, female CML patients are more negatively affected than male CML patients in both mental and physical health.2 Sex-related differences in disease perception and anxiety may contribute to the higher prevalence of fatigue in women,28 although this aspect was beyond the scope of this study. Furthermore, we showed that patients with comorbidity were more often fatigued, as were patients taking comedication known to cause fatigue. This suggests the need to check patients’ medication records critically, and to stop or reduce the dose of any comedication known to cause fatigue if possible (e.g., benzodiazepines as sleep medication), especially in those CML patients who are prone to fatigue. Both subjective and objective assessments of physical activity showed that fatigued patients were more often inactive than were non-fatigued patients. More precisely, we found that compared to the non-fatigued patients, fatigued CML patients slept approximately 0.5 h/day longer, performed 1 h less of physical activity per day and took 2,000 fewer steps per day. Although it may seem self-evident that severely fatigued patients are less physically active as a result of fatigue, physically inactivity itself may contribute to the persistence of fatigue.9 Additionally, there is a significant body of evidence to support the beneficial effects of exercise interventions to reduce fatigue levels in various (post)-cancer patient populations.29 Interestingly, our study showed that the vast majority of the CML patients, both fatigued and nonfatigued, already met the recommended American College of Sports Medicine/American Heart Association guidelines for physical activity (i.e., 150-300 min of moderate-intensity or 75-150 min of vigorous intensity physical activity per week, or an equivalent combination). However, higher activity levels were associated with lower levels of fatigue in our CML population. Furthermore, the extra amount of physical activity that non-fatigued patients performed when compared to fatigued patients (~6.3 h of light intensity and 1.4 h of moderate to vigorous intensity physical activity per week) may yield additional health benefits.30 Consequently, it is of clinical relevance to focus on preventing and treating TKI-induced fatigue in clinical practice. This is further supported by our findings that fatigued CML patients have impaired QoL and work fewer hours when compared to non-fatigued patients. There are several limitations to this study. First, due to the cross-sectional design of the study we cannot distinguish between cause and effect. Although we found that a reduced level of physical activity is associated with the presence of fatigue, and thus is a predictor of fatigue, we cannot state that reduced physical activity is a risk factor for fatigue. However, regardless of whether or not there is a causal relationship between fatigue and reduced levels of physical activity, our results highlight the importance of combating fatigue and of examining whether exercise interventions are useful to counteract fatigue. Secondly, because of the inclusion of a heterogeneous study population, we observed a considerable variation in physical activity levels. However, the representative sam1881

L. Janssen et al.

ple in this population-based study allows translation of the findings to clinical practice. Lastly, we used a Likert scale for the assessment of QoL in order to reduce the length of our questionnaire to assess predictors of fatigue (Part 1) even though validated QoL questionnaires have been developed in the CML population. However, a simple Likert scale has been shown to measure QoL adequately in cancer patients.31 A major strength of our study is the objective assessment of physical activity, which ruled out response bias. Another strength is the relatively large sample size and the small amount of missing data (<3% in both parts of the study). In conclusion, we demonstrated that the majority of the CML patients receiving TKI therapy experienced severe fatigue and that severely fatigued patients have impaired QoL. Independent predictors of severe fatigue include: younger age, female gender, higher CCI, the use of comedication known to cause fatigue, and physical inactivity. Objective assessment of physical activity showed that, compared to patients without fatigue, severely fatigued CML patients sleep more and are less active during the day

References 1. Bower H, Bjorkholm M, Dickman PW, Hoglund M, Lambert PC, Andersson TM. Life expectancy of patients with chronic myeloid leukemia approaches the life expectancy of the general population. J Clin Oncol. 2016;34(24):2851-2857. 2. Efficace F, Baccarani M, Breccia M, et al. Health-related quality of life in chronic myeloid leukemia patients receiving longterm therapy with imatinib compared with the general population. Blood. 2011;118(17): 4554-4460. 3. Eliasson L, Clifford S, Barber N, Marin D. Exploring chronic myeloid leukemia patients' reasons for not adhering to the oral anticancer drug imatinib as prescribed. Leuk Res. 2011;35(5):626-630. 4. Marin D, Bazeos A, Mahon FX, et al. Adherence is the critical factor for achieving molecular responses in patients with chronic myeloid leukemia who achieve complete cytogenetic responses on imatinib. J Clin Oncol. 2010;28(14):2381-2388. 5. Efficace F, Baccarani M, Breccia M, et al. Chronic fatigue is the most important factor limiting health-related quality of life of chronic myeloid leukemia patients treated with imatinib. Leukemia. 2013;27(7):15111519. 6. Caldemeyer L, Dugan M, Edwards J, Akard L. Long-term side effects of tyrosine kinase inhibitors in chronic myeloid leukemia. Curr Hematol Malign Rep. 2016;11(2):71-79. 7. Winters-Stone KM, Bennett JA, Nail L, Schwartz A. Strength, physical activity, and age predict fatigue in older breast cancer survivors. Oncol Nurs Forum. 2008;35(5):815821. 8. Engberg I, Segerstedt J, Waller G, Wennberg P, Eliasson M. Fatigue in the general population- associations to age, sex, socioeconomic status, physical activity, sitting time and selfrated health: the northern Sweden MONICA study 2014. BMC Public Health. 2017;17(1): 654. 9. Bower JE. Cancer-related fatigue--mechanisms, risk factors, and treatments. Nat Rev Clin Oncol. 2014;11(10):597-609. 10. Oberoi S, Robinson PD, Cataudella D, et al.

1882

on both week and weekend days. These findings emphasize the importance of recognizing the reduction of fatigue as a treatment goal in CML care and the need for future studies to identify physical activity as a possible target to achieve this goal. Disclosures No conflicts of interest to disclose. Contributions LJ performed the research, analyzed data and wrote the manuscript with support from all authors. NB performed research and supervised the study. MD performed research and analyzed data. EB analyzed data. MN and JJ performed research. ST and MH supervised the study. Acknowledgments The authors would like to thank Carlijn Maasakkers from the Radboud University Medical Center (UMC) for her support in analyzing activity monitor data and Maaike de Ruijter from the Amsterdam UMC for assistance in recruiting participants.

Physical activity reduces fatigue in patients with cancer and hematopoietic stem cell transplant recipients: a systematic review and meta-analysis of randomized trials. Crit Rev Oncol Hematol. 2018;122:52-59. 11. Braspenning KAE HR, Nowee M, Janssen J, Westerweel PE, Blijlevens NMA. CMyLife: een eerste inzicht in kwaliteit van leven en verbetermogelijkheden bij CML-patiënten voorafgaand aan gebruik van CMyLife. Nederlands Tijdschrift voor Hematologie. 2017;4:183-189. 12. Vercoulen JH, Swanink CM, Fennis JF, Galama JM, van der Meer JW, Bleijenberg G. Dimensional assessment of chronic fatigue syndrome. J Psychosom Res. 1994;38(5):383392. 13. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987; 40(5):373-383. 14. Ryan CG, Grant PM, Tigbe WW, Granat MH. The validity and reliability of a novel activity monitor as a measure of walking. Br J Sports Med. 2006;40(9):779-784. 15. Egerton T, Chastin SF, Stensvold D, Helbostad JL. Fatigue may contribute to reduced physical activity among older people: an observational study. J Gerontol A Biol Sci Med Sci. 2016;71(5):670-676. 16. O'Brien SG, Guilhot F, Larson RA, et al. Imatinib compared with interferon and lowdose cytarabine for newly diagnosed chronic-phase chronic myeloid leukemia. N Engl J Med. 2003;348(11):994-1004. 17. Kalmanti L, Saussele S, Lauseker M, et al. Safety and efficacy of imatinib in CML over a period of 10 years: data from the randomized CML-study IV. Leukemia. 2015;29(5): 1123-1132. 18. Kantarjian H, Shah NP, Hochhaus A, et al. Dasatinib versus imatinib in newly diagnosed chronic-phase chronic myeloid leukemia. N Engl J Med. 2010;362(24):22602270. 19. Saglio G, Kim DW, Issaragrisil S, et al. Nilotinib versus imatinib for newly diagnosed chronic myeloid leukemia. N Engl J Med. 2010;362(24):2251-2259. 20. 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. 21. Lipton JH, Chuah C, Guerci-Bresler A, et al. Ponatinib versus imatinib for newly diagnosed chronic myeloid leukaemia: an international, randomised, open-label, phase 3 trial. Lancet Oncol. 2016;17(5):612-621. 22. Zordan R, Manitta V, Nandurkar H, ColeSinclair M, Philip J. Prevalence and predictors of fatigue in haemo-oncological patients. Intern Med J. 2014;44(10):1013-1017. 23. Ruffer JU, Flechtner H, Tralls P, et al. Fatigue in long-term survivors of Hodgkin's lymphoma; a report from the German Hodgkin Lymphoma Study Group (GHSG). Eur J Cancer. 2003;39(15):2179-2186. 24. Butt Z, Rao AV, Lai JS, Abernethy AP, Rosenbloom SK, Cella D. Age-associated differences in fatigue among patients with cancer. J Pain Symptom Manage. 2010;40(2): 217-223. 25. Tibubos AN, Ernst M, Brahler E, et al. Fatigue in survivors of malignant melanoma and its determinants: a register-based cohort study. Support Care Cancer. 2019;27(8):2809-2818. 26. Nacul LC, Lacerda EM, Pheby D, et al. Prevalence of myalgic encephalomyelitis/ chronic fatigue syndrome (ME/CFS) in three regions of England: a repeated cross-sectional study in primary care. BMC Med. 2011;9:91. 27. Jason LA, Richman JA, Rademaker AW, et al. A community-based study of chronic fatigue syndrome. Arch Intern Med. 1999;159(18): 2129-2137. 28. Bensing JM, Hulsman RL, Schreurs KM. Gender differences in fatigue: biopsychosocial factors relating to fatigue in men and women. Med Care. 1999;37(10):1078-1083. 29. Cramp F, Byron-Daniel J. Exercise for the management of cancer-related fatigue in adults. Cochrane Database Syst Rev. 2012; 11:CD006145. 30. Piercy KL, Troiano RP, Ballard RM, et al. The physical activity guidelines for Americans. JAMA. 2018;320(19):2020-2028. 31. Rogers MP, Orav J, Black PM. The use of a simple Likert scale to measure quality of life in brain tumor patients. J Neurooncol. 2001;55(2):121-131.

haematologica | 2021; 106(7)

ARTICLE

Hematopoiesis

Thrombopoietin maintains cell numbers of hematopoietic stem and progenitor cells with megakaryopoietic potential

Ferrata Storti Foundation

Aled O’Neill,1 Desmond Chin,1 Darren Tan,1 A’Qilah Banu Bte Abdul Majeed,1 Ayako Nakamura-Ishizu1,2 and Toshio Suda1,2 Cancer Science Institute, National University of Singapore, Singapore and 2 International Research Center for Medical Sciences, Kumamoto University, Kumamoto, Japan

ABSTRACT

Haematologica 2021 Volume 106(7):1883-1891

hrombopoietin has long been known to influence megakaryopoiesis and hematopoietic stem and progenitor cells, although the exact mechanisms through which it acts are unknown. Here we show that MPL expression correlates with megakaryopoietic potential of hematopoietic stem and progenitor cells and identify a population of quiescent hematopoietic stem and progenitor cells that show limited dependence on thrombopoietin signaling. We show that thrombopoietin is primarily responsible for maintenance of hematopoietic cells with megakaryocytic differentiation potential and their subsequent megakaryocyte differentiation and maturation. The loss of megakaryocytes in thrombopoietin knockout mouse models results in a reduction of megakaryocyte-derived chemokine platelet factor 4 (CXCL4/PF4) in the bone marrow and administration of recombinant CXCL4/PF4 rescues the loss of quiescence observed in these mice. CXCL4/PF4 treatment does not rescue reduced hematopoietic stem and progenitor cell numbers, suggesting that thrombopoietin maintains hematopoietic stem and progenitor cell numbers directly.

Introduction Hematopoietic stem cells (HSC) are defined by their ability to not only differentiate into all blood cell lineages, but also to self-renew and enter the non-proliferative quiescent state.1 Maintenance of this quiescent state is thought to rely on several spatially separated niche factors2 and there are many candidates for potential ex-vivo maintenance of HSC quiescence including CXCL12 and thrombopoietin (THPO).3 It has long been known that loss of THPO signaling results in reduced HSC numbers and loss of quiescence,4,5 although the mechanisms behind this phenomenon are yet to be identified. In addition to influencing quiescence, THPO is known to drive megakaryocyte differentiation and maturation as well as platelet production.6 The THPO receptor, myeloproliferative leukemia (MPL), is expressed on HSC and it has been shown that HSC require THPO for survival in vitro.7 Cells phenotypically similar to HSC in the hematopoietic compartment have also been shown to differentiate directly into megakaryocytes without cell division, suggesting a close relationship between the earliest megakaryocyte progenitors (MkP) and HSC.8 Determining how THPO is responsible for these seemingly different effects has been an object of much research, with several studies on mutated leukemia cell lines suggesting that the expression ratios between MPL and its downstream signal transducer JAK2 as well as extracellular THPO levels may determine whether hematopoietic stem and progenitor cells (HSPC) enter quiescent or proliferative pathways.9,10 Whether HSC in the bone marrow have different responses to THPO remains to be seen. We, therefore, set out to investigate the effects of THPO on various HSC both in vivo and in vitro.

Methods

Correspondence: TOSHIO SUDA sudato@keio.jp Received: October 23, 2019. Accepted: May 28, 2020. Pre-published: June 11, 2020 https://doi.org/10.3324/haematol.2019.241406

Experimental mice All experiments were performed on 8- to 12-week-old mice in the C57BL/6-NTac backhaematologica | 2021; 106(7)

1883

A. O’Neill et al.

ground in accordance with Institutional Animal Care and Use Committee protocols. Alb-Cre mice in a white background were kindly provided by Prof. Qingde Wang of the University of Pittsburgh Medical Center and were back-crossed with C57BL/6-NTac mice at least five times.

Bone marrow mononuclear cell isolation Mice were sacrificed by CO asphyxiation and femora, tibiae, pelves, humeri and vertebrae were crushed in phosphatebuffered saline (PBS). Red blood cells were lysed with NH Cl in PBS and bone marrow mononuclear cells (BMMNC) were stained as described below. 2

Flow cytometric analysis All flow cytometric analyses and sorting were performed on a BD FACS Aria II cell sorter. Antibodies used were CD4 (RM45, BD Biosciences), CD8a (53-6.7, eBioscience), B220 (RA36B2, Biolegend), CD11b (M1/70, BD Biosciences), Ly6G/C (RB6-8C5, Biolegend), Ter119 (TER-119, Biolegend), Sca1 (D7, Biolegend), MPL (AMM2, Immuno-Biological Laboratories), CD41 (MWReg30, BD Biosciences), CD150 (TC15-12F12.2, Biolegend), cKit (2B8, Biolegend), CD34 (RAM34, eBioscience), IL7R (A7R34, eBioscience), Flt3 (A2F10, eBioscience), CD16/32 (93, eBioscience) and CD105 (MJ7/18, Biolegend). Streptavidin (eBioscience) was used to resolve biotinylated

1884

Figure 1. Thrombopoietin from liver maintains hematopoietic stem and progenitor cells in the bone marrow. (A) Schematic of Thpo flox generation. Two guide RNA were used to simultaneously cut two sites flanking exons 2 and 3 of the Thpo gene. Single-stranded deoxyribo-oligonucleotides were then used to insert loxP sequences by homology directed repair. (B) Quantitative polymerase chain reaction analysis of Thpo mRNA from livers of Thpofl/fl and ThpoΔ/Δ Alb-Cre mice relative to B2M expression (n=3). (C) Flow cytometric gating system for analysis of hematopoietic stem cells (HSC) and megakaryocyte progenitors (MkP). Peripheral blood platelet counts were taken for ThpoΔ/Δ mice and Thpofl/fl littermate controls of PF4-Cre mice (D, n=5fl/fl /6Δ/Δ), Vav-Cre mice (E, n=5fl/fl /6Δ/Δ) Osx-Cre mice (F, n=5fl/fl /6Δ/Δ) and Alb-Cre mice (G, n=5fl/fl /6Δ/Δ) in addition to Thpo-/-mice and ThpoWT/WT littermate controls (H, n=4WT/WT/6/). Cell counts of MkP population in the bone marrow of ThpoΔ/Δ mice and Thpofl/fl littermate controls were analyzed in PF4Cre mice (I, n=6), Vav-Cre mice (J, n=6), Osx-Cre mice (K, n=6) and Alb-Cre mice (L, n=6) as well as Thpo-/-mice and ThpoWT/WT littermate controls (M, n=6). Cell counts of CD34-flt3-HSC populations in the bone marrow of ThpoΔ/Δ mice and Thpofl/fl littermate controls were analyzed in PF4-Cre mice (N, n=6), Vav-Cre mice (O, n=6), OsxCre mice (P, n=6) and Alb-Cre mice (Q, n=6) in addition to Thpo-/-mice and ThpoWT/WT littermate controls (R, n=6).

haematologica | 2021; 106(7)

Thrombopoietin maintains megakaryopoietic HSPC

antibodies and propidium iodide was added prior to analysis to identify live cells.

mRNA expression Cells were sorted into Trizol (Thermo Fisher Scientific) and RNA was isolated according to the manufacturer’s protocol. Complementary DNA was reverse transcribed by a SuperScript VILO cDNA Synthesis Kit (Thermo Fisher Scientific) and mRNA expression analyzed using TaqMan Fast Advanced Master Mix (Thermo Fisher Scientific) in a QuantStudio 3 Real-Time PCR system (Thermo Fisher Scientific). Expression levels were calculated relative to the expression of Gapdh.

haematologica | 2021; 106(7)

In vitro culture BMMNC were isolated as above and cKit enrichment was performed using magnetic activated cell sorting (MACS)-conjugated cKit antibodies (Miltenyi Biotec) and an AutoMACS cell sorter according to the manufacturer’s protocol. Enriched cells were stained with antibodies and sorted to StemSpanII (StemCell Technologies) containing stem cell factor (SCF, 20 ng/mL, Peprotech) and THPO (20 ng/mL, Peprotech), or SCF, THPO, interleukin-3 (IL3, 20 ng/mL, Peprotech), interleukin-6 (IL6, 20 ng/mL, Peprotech) and erythropoietin (EPO, 10 ng/mL, R&D Systems). On day 10 of culture colonies were stained with CD41, CD16/32, CD71 (C2, BD Biosciences), Ter119, CD11b,

Figure 2. Thrombopoietin maintains hematopoietic stem and progenitor cells with megakaryopoietic potential. (A, B) Mature B-cell counts were analyzed in the bone marrow of ThpoΔ/Δ Alb-Cre and Thpofl/fl control mice (A), as well as Thpo-/- and ThpoWT/WT mice (B). (C, D) Thymus CD4+ and CD8+ Tcell numbers were analyzed in the two models. (E-H) Peripheral blood white blood cells (E, F) and red blood cells (G, H) were quantified. (I-L) Lymphoid progenitors (I, J) and myeloid progenitors (K, L) were quantified In the bone marrow of the two murine models. (n=6) PB: peripheral blood; WBC: white blood cells; RBC: red blood cells; BMMNC: bone marrow mononuclear cells; CLP: common lymphoid progenitors; GMP: granulocyte/ macrophage progenitors; MegE: megakaryocute/erythroid progenitors; CFUE: colony-forming unit erythroid.

1885

A. O’Neill et al.

Ly6C/G and propidium iodide was added to identify live cells. Lineages were gated as shown in Online Supplementary Figure S1 and colonies were defined by the presence of >20 cells of any one of each lineage.

Cell cycle analysis cKit-enriched BMMNC were isolated as above and stained with PerCP-Cy5.5 lineage cocktail, FITC-conjugated CD34, PEconjugated CD41, PE-Cy7-conjugated Sca1, APC-conjugated CD150 and APC-Cy7-conjugated cKit. Cells were then fixed and permeabilized by IntraPrep Reagent (Beckman Coulter) in accordance with the manufacturer’s protocol. AlexaFluor700-conjugated Ki67 (SolA15, Invitrogen) antibodies and Hoechst 33342

For the rescue experiments, 1 mg CXCL4/PF4 (Peprotech) or PBS was injected intraperitoneally daily for 7 days and bone marrow analyzed on day 7. Romiplostim (Kyowa Kirin) or PBS as a control was administered intravenously at a dose of 100 mg/kg per day for 5 days.

Statistical analysis All values given are mean ± standard deviation. Statistical analyses were performed by two-tailed unpaired t-tests and the

1886

Rescue experiments

were added after permeabilization and cells were analyzed by flow cytometry.

Figure 3. Hematopoietic stem and progenitor cells show variable responses to loss of thrombopoietin/MPL signaling. (A) Flow cytometric gating system for hematopoietic stem and progenitor cell (HSPC) populations. (B-D) Cell counts of populations in bone marrow of ThpoΔ/Δ Alb-Cre, Thpofl/fl, Thpo-/- and ThpoWT/WT mice were analyzed for longterm hematopoietic stem cells (LT-HSC) (B), CD150- HSPC (C) and CD41+ HSPC (D) (n=6). (E) A representative plot of CD48 and MPL expression in LSK cells, with CD48 and MPL gating lines indicated. (F) Percentage of wild-type (WT) cells from each HSPC population that are CD48- (n=8). (G) Percentage of CD48- cells from each population expressing MPL (n=8). (H) MPL mean fluorescence intensity (MFI) of CD48MPL+ cells from each population (n=8). (I) MPL mRNA expression levels relative to GAPDH (n=3). BMMNC: bone marrow mononuclear cells.

haematologica | 2021; 106(7)

Thrombopoietin maintains megakaryopoietic HSPC

levels of statistical significance are represented by asterisks. *P<0.05, **P<0.01, ***P<0.001. NS represents P>0.05.

R). This would suggest that steady-state HSPC in the bone marrow mainly depend on THPO from the liver.

Results

Thromobopoietin maintains megakaryopoietic progenitor numbers

Hematopoietic stem and progenitor cells in bone marrow require thrombopoietin from liver To investigate the role of THPO in HSPC we first sought the source of THPO in the bone marrow. We generated a Thpoflox mouse using CRISPR/Cas9 to simultaneously insert two loxP sites flanking exons 2 and 3 of the Thpo gene (Figure 1A). Loss of Thpo mRNA expression in the liver of ThpoΔ/Δ Alb-Cre mice was confirmed by quantitative polymerase chain reaction analysis (Figure 1B). In agreement with Decker et al.11 we found that liver-specific knockout (KO) of THPO caused similar loss of platelets to that of Thpo-/-, while conditional KO models for megakaryo-cytes, HSPCs and bone marrow stromal cells (PF4, Vav and Osx respectively) did not show any significant difference (Figure 1D-H).11 Similarly, the numbers of bone marrow MkP were significantly reduced in ThpoΔ/Δ Alb-Cre mice and Thpo-/- mice, but not in other conditional KO models (Figure 1I-M). We also observed a similar pattern in the CD34-Flt3-LSK HSC population, with a reduction in cell numbers within the bone marrow of ThpoΔ/Δ AlbCre mice similar to that of Thpo-/- mice, but no reduction in bone marrow-specific conditional KO models (Figure 1N-

haematologica | 2021; 106(7)

We set out to find which other hematopoietic cells are influenced by THPO. Analysis of mature blood lineages revealed that B cells in the bone marrow (Figure 2A and B), T cells in thymus (Figure 2C and D) and white blood cells in peripheral blood (Figure 2E and F) were unaffected by loss of THPO in both Thpofl/flAlb-Cre and Thpo-/- models. No effect on peripheral blood red blood cells was seen in Thpofl/flAlb-Cre mice and although statistical analysis showed a significant reduction in Thpo-/- mice, this appears to be very low and may simply be an artefact of variance within the mouse strain (Figure 2G and H). To confirm that the loss of THPO did not affect mature lineages other than megakaryocytes, progenitor populations in bone marrow were analyzed. Common lymphoid progenitors in the bone marrow were unaffected by loss of THPO in both models (Figure 2I and J). Analysis of myeloid progenitors showed that only the previously defined Pre-GMP and Pre-MegE populations were affected by the loss of THPO (Figure 2K and L). Interestingly, these populations were previously shown to give rise to megakaryocytes in vitro while the other myeloid progenitors did not show megakaryopoietic potential,12 suggesting that only the MkP, Pre-GMP and

Figure 4. CD150- hematopoietic stem and progenitor cells have reduced thrombopoietin dependence and low megakaryopoietic potential. (A) Single cells of each hematopoietic stem and progenitor cell (HSPC) population were cultured in vitro with stem cell factor (SCF) or SCF and thrombopoietin (THPO) and colonies were counted on day 10. Bars show the percentage of sorted cells that gave rise to colonies (n=72 single cells across 3 experiments). (B) Single cells of each HSPC population were cultured in vitro and colonies analyzed by fluorescence activated cell sorting on day 10. Bars show the percentage of colonies that contain specific lineages (n=72 single cells across 3 experiments). (C, D) mRNA expression of Vwf (C) and Gfi1 (D) relative to GAPDH in HSPC (n=3). LT-HSC: long-term hematopoietic stem cell, Mk: megakaryocyte; MkE: megakaryocyte-erythroid; MkEGM: megakaryocyte-erythroid-granulocyte-macrophage; GM: granulocyte-macrophage.

1887

A. O’Neill et al.

Pre-MegE cells that give rise to megakaryocytes are affected by the loss of THPO. Analysis of MPL expression among these progenitor populations suggests that, while MPL was detectable in all populations, most cell populations showed baseline levels. Only those populations with megakaryocytic potential showed levels of expression significantly above the baseline (Online Supplementary Figure S2).

Thrombopoietin is required to maintain CD150+ hematopoietic stem cell numbers To further analyze the hematopoietic compartment the CD34-LSK HSC fraction was subdivided into three HSC populations based on CD41 and CD150 expression (Figure 3A). Previous studies have defined the CD34CD41-CD150+ LT-HSC population as being enriched for long-term repopulating HSC, while the CD150- HSPC population contains short-term repopulating HSC and lymphoid-biased HSC.13,14 The CD41+ HSPC population has been proposed to be enriched for common myeloid progenitors with the potential to repopulate the bone marrow of lethally irradiated mice.15 This method of defining populations was found to be optimal for distinguishing those LSK populations that express high levels of MPL from those with reduced MPL expression (Online Supplementary Figure S3) as well as for separating populations based on differentiation potential. Analysis revealed that the two CD150+ LSK populations, LT-HSC and CD41+ HSPC, showed significant loss of cell numbers with LT-HSC numbers reduced from 16.3±4.35 cells per 106 bone marrow mononuclear cells (BMMNC) to 1.8±0.89 in ThpoΔ/Δ Alb-Cre mice and from 19.9±8.21 to 1.4±0.94 in Thpo-/- mice (Figure 3B). CD41+ HSPC also decreased drastically from 9.4±3.92 to 0.7±0.52 in ThpoΔ/Δ Alb-Cre mice and from 10.5±3.68 to 0.2±0.08 in Thpo-/- mice, a reduction of over 90% (Figure 3D). The CD150- HSPC

population, however, showed no significant difference in either Thpofl/flAlb-Cre or Thpo-/- mice, suggesting that this population may have reduced dependence on THPO signaling (Figure 3C).

CD150- hematopoietic stem and progenitor cells have reduced MPL expression As CD48 is expressed by lineage-committed cells we analyzed CD48 expression in the HSPC populations. While a high proportion of LT-HSC and CD41+ HSPC did not express CD48, the majority of CD150- HSPC were CD48+ (Figure 3E and F). To investigate why the two CD150+ populations were highly affected by loss of THPO signaling while CD150- HSPC were not, we looked at the surface expression of MPL of the three populations. The CD48-CD150- HSPC expressed MPL, but a significantly smaller proportion of these cells expressed MPL compared to the other two HSC populations (Figure 3G). Among the MPL+CD48-CD150- HSPC the MPL mean fluorescence intensity was lower than in MPL+CD48- LT-HSC, suggesting that MPL is being downregulated as LT-HSC differentiate into CD150- HSPC (Figure 3H). To confirm that MPL expression is reduced we sorted CD48-MPL+ LT-HSC as well as CD48MPL+CD150- HSPC and CD48-MPL-CD150- HSPC and analyzed MPL mRNA expression (Figure 3I). We observed that even in the MPL+CD150- HSPC, MPL mRNA expression was significantly reduced compared to that in LT-HSC and was further reduced in MPLCD150- HSPC. This suggests that MPL is downregulated as LT-HSCs differentiate into CD150- HSPC and that CD150-HSPC may therefore be less dependent on THPO/MPL signaling. Analysis of MPL expression in the KO models showed that smaller proportions of LT-HSC and CD150- HSPC were expressing MPL in both Thpo-/mice and ThpoΔ/Δ Alb-Cre mice, while MFI of the MPL+ cells

Figure 5. Maintenance of cell number is independent of quiescence. (A) Quiescence was measured as the percentage of Ki67 negative (G ) cells within hematopoietic stem and progenitor cell (HSPC) populations of Thpo-/- and ThpoWT/WT control mice. (B) The concentration of CXCL4/PF4 in bone marrow of knockout mice as measured by enzyme-linked immunosorbent assay (n=4). (C-E) Percentage of cells in G at day 7 of intraperitoneal treatment with CXCL4/PF4 or phosphatebuffered saline (PBS) in Thpo-/- mice and ThpoWT/WT littermate controls (n=3). (F, G) Cell counts of HSPC populations at day 7 of intraperitoneal treatment with CXCL4/PF4 or PBS in Thpo/mice and ThpoWT/WT littermate controls (n=3). BMEF: bone marrow extracellular fluid; BMMNC: bone marrow mononuclear cells. 0

1888

haematologica | 2021; 106(7)

Thrombopoietin maintains megakaryopoietic HSPC

was not affected (Online Supplementary Figure S4). This suggests specific loss of MPL+ cells in the absence of THPO. Despite this, MPL expression in CD41+ HSPC was unaffected by loss of THPO, suggesting that this population of HSPC is more homogeneous than the other two HSPC populations and that MPL expression is intrinsic to CD41+ HSPC.

CD150- hematopoietic stem and progenitor cells have reduced dependence on thrombopoietin and low megakaryopoietic potential To examine whether the CD150- HSPC are dependent on THPO signaling for proliferation, single MPL+CD48CD150- HSPC and MPL-CD48-CD150- HSPC were cultured with THPO and SCF for 7 days. Proliferation of CD48-MPL+ LT-HSC and CD41+ HSPC was also analyzed (Figure 4A). A high proportion of the LT-HSC and CD41+ HSPC gave rise to colonies at day 10 in the presence of THPO, but in the absence of THPO few colonies were observed. This indicates that the two CD150+ populations are dependent on THPO for proliferation in vitro. Both MPL+CD150- HSPC and MPL-CD150- HSPC produced few colonies in the presence of THPO and while the MPL+CD150- HSPC produced slightly fewer colonies in the absence of THPO, the colony number among MPLCD150- HSPC was not significantly different. This suggests that the CD150- HSPC population has reduced dependence on THPO for proliferation in vitro, with the

MPL-CD150- HSPC population showing limited dependence on THPO/MPL signaling. Downregulation of MPL expression has previously been linked to lymphoid lineage commitment,16 suggesting that the CD150- HSPC would have reduced myeloid differentiation potential compared to the other LSK populations. To test this we utilized an adapted method of a previously described in vitro culture assay.17 The CD48- fractions of each population were cultured with SCF, THPO, EPO, IL3 and IL6 to assess their differentiation potentials (Figure 4B). At day 10 fewer than 50% of single cells from the two CD150HSPC populations produced colonies, suggesting that the myeloid differentiation cytokines in the media were insufficient to support the proliferation of many of the cells in these populations. Both MPL+CD48-CD150- HSPC and MPL-CD48-CD150- HSPC populations produced significantly fewer Mk colonies than LT-HSC and CD41+ HSPC. Previous studies have shown that MPL is downregulated as HSC become lymphoid-biased. Taken together this suggests that megakaryopoietic potential is lost as the CD150- HSPC become lymphoid-primed. Over 50% of CD48-MPL+ LT-HSC gave rise to colonies containing megakaryocytes, while approximately 90% of CD48MPL+CD41+ HSPC give rise to colonies with megakaryocytes. Subsequent mRNA expression analysis revealed that while LT-HSC and CD41+ HSPC both highly expressed the megakaryocyte marker, von Willebrand factor (Vwf), the two CD150- HSPC populations showed

Figure 6. Proposed model of maintenance of megakaryopoietic hematopoietic stem and progenitor cell number and quiescence. Thrombopoietin (THPO) from liver maintains megakaryopoietic cell numbers in the bone marrow. This preserves homeostatic levels of megakaryocyte-derived CXCL4, which induces quiescence in hematopoietic stem and progenitor cell (HSPC) populations. CD150- HSPC rely on CXCL4 for quiescence but are not dependent on THPO for maintenance of cell numbers. LT-HSC: long-term hematopoietic stem cell; MkP: megakaryocyte progenitor; MK: megakaryocyte.

haematologica | 2021; 106(7)

1889

A. O’Neill et al.

significantly lower expression, indicative of their reduction in megakaryopoietic potential (Figure 4C). Analysis of the expression of the neutrophil/lymphoid marker (Gfi1) revealed that the expression of Gfi1 was higher in the CD150- HSPC population than in LT-HSC, and its expression was reduced in CD41+ HSPC (Figure 4D). These data suggest that the CD150- HSPC may be lymphoid-primed at the expense of megakaryopoietic potential, while the CD41+ HSPC may become megakaryocyteprimed at the expense of lymphoid potential.

Loss of quiescence is not responsible for loss of cell number Previous studies have shown that THPO plays a role in maintaining HSC quiescence.4 In order to investigate this, we first looked at the quiescence levels of the HSC populations in Thpo-/- mice (Figure 5A). The frequency of quiescence was reduced in all three HSPC populations. Interestingly, quiescence was reduced in CD150- HSPC despite them having limited dependence on THPO/MPL signaling. Previous studies have suggested that megakaryocytes in the bone marrow are responsible for maintaining HSCs in quiescence through the megakaryocyte-specific cytokine CXCL4/PF4.18 We looked at CXCL4/PF4 levels in the bone marrow of Thpo-/- mice and found that CXCL4/PF4 concentration is decreased on loss of THPO (Figure 5B). This suggests that the loss of megakaryocytes in KO mice reduces bone marrow CXCL4/PF4 concentration. To assess whether administration of recombinant CXCL4/PF4 could rescue quiescence in Thpo-/- mice, CXCL4/PF4 was injected daily for 7 days. Treated KO mice showed a significant increase in the proportion of cells in G in all three HSPC populations (Figure 5C-E). The quiescence levels in treated KO mice were not significantly different from those of wild-type untreated mice, suggesting that CXCL4/PF4 treatment rescues quiescence back to wild-type levels in all three HSPC populations. Taken together these data would suggest that the HSPC require CXCL4/PF4 from megakaryocytes for maintenance of quiescence. Although quiescence is rescued by CXCL4/PF4 treatment the number of LT-HSCs and CD150- HSPC remained unchanged (Figure 5F and G). This would suggest that the loss of cell numbers is independent of loss of quiescence. To confirm this we treated Thpo-/- mice with the MPL receptor agonist romiplostim for 5 days via tail vein injection. On day 5 platelet counts in Thpo-/- mice had recovered to the level in wild-type untreated mice, suggesting that thrombopoietic potential – and by extension megakaryopoietic potential – is not lost in the absence of THPO signaling and can return to wild-type levels upon rescue of MPL signaling (Online Supplementary Figure S5A). 0

Discussion We have shown that loss of THPO signaling in bone marrow leads to a reduction of cell numbers throughout the hematopoietic compartment, primarily in cells that express high levels of MPL and show potential for megakaryopoietic differentiation. We show that the CD150- HSPC population has low MPL expression and reduced megakaryopoietic potential, findings that are consistent with previous reports that these cells are lymphoid-biased.14,16 Previous reports have defined the CD150- HSPC and CD41+ HSPC 1890

populations as HSC, although strictly speaking these populations contain predominantly hematopoietic progenitor cells, rather than true HSCs. In spite of this, it is important to note that these populations are both capable of repopulating the bone marrow of lethally irradiated mice, though not long-term, indicating a capacity for self-renewal. Both CD41+ HSPC and CD150- HSPC populations also show high levels of quiescence similar to LT-HSC and express cell surface markers very similar to those of LT-HSC, suggesting a very close relationship between the three populations. Indeed, previous studies have suggested that CD41+ HSPC and CD150- HSPCs represent the earliest branch point in the hematopoietic hierarchy, with CD41+ HSPC being the earliest myeloid branch and CD150- HSPC being the earliest lymphoid branch.14 The data in this study support this view. Despite their low levels of MPL expression and limited dependence on THPO for cell proliferation both in vivo and in vitro, CD150- HSPC show reduced quiescence in the absence of THPO in the bone marrow. Previous studies showed that loss of megakaryocytes in bone marrow results in reduced HSC quiescence,19,20 while a subsequent study showed that this effect is due to loss of megakaryocyte-derived CXCL4/PF4.18 We show that in the case of THPO KO mouse models, the loss of quiescence in HSPC results from the loss of CXCL4/PF4 signaling due to reduced megakaryocyte number in the bone marrow and that this can be rescued by administration of exogenous CXCL4/PF4. Interestingly, CXCL4/PF4 administration in wild-type mice does not produce an increase in quiescence, suggesting that quiescence is not dependent on CXCL4/PF4 signaling alone. Previous studies have shown that CXCL12 from CXCL12-abundant reticular (CAR) cells in the bone marrow also plays a role in HSPC quiescence.21 Other studies have provided evidence of dimerization of CXC ligands, including CXCL12.22 One theory is that CXCL4/PF4 and CXCL12 form heterodimers that induce quiescence in HSC and that without an increase in CXCL12, increased CXCL4/PF4 cannot induce further quiescence. Further research may clarify whether CXCL4/PF4-dependent quiescence is co-dependent on other signaling pathways. Our in vitro and in vivo data suggest that MPL expression in HSC correlates closely with megakaryocytic differentiation potential, indicating that THPO in the bone marrow is responsible for maintaining megakaryocytic differentiation potential in HSPC. A recent study from our own group has shown that increased THPO/MPL signaling leads to increased proliferation and megakaryocytic differentiation as well as mitochondrial activation in HSCs, suggesting that THPO drives cell division, rather than suppressing it.23 There are reports that THPO induces self-renewal division in HSC24 and it is possible that loss of this self-renewal division is responsible for the loss of cell numbers of megakaryopoietic HSPC within the bone marrow. Here we show that THPO is required for proliferation of HSPC with megakaryopoietic potential while previous studies showed that it is required for the maturation of megakaryocytes from MkP. Together, this would suggest that THPO plays a dual role in maintaining the megakaryocyte population and that both loss of megakaryocyte-producing HSPC and impairment of MkP maturation lead to the acute reduction of megakaryocytes seen in KO models. In humans, injection of THPO leads to an autoimmune response against the exogenous protein and even against endogenous THPO. For this reason, the artificial recombihaematologica | 2021; 106(7)

Thrombopoietin maintains megakaryopoietic HSPC

nant romiplostim and the small molecule eltrombopag have been approved as clinical MPL agonists. We showed that treatment of Thpo-KO mice with romiplostim can rescue platelet numbers, indicating that megakarypoietic potential is partially cell-intrinsic. Further investigation may reveal the mechanisms through which THPO/MPL signaling drives expansion of HSPC and its role in selfrenewal division of megakaryopoietic HSPC. This study addresses the many conflicting reports of the role of THPO in HSPC quiescence/self-renewal division as well as its role in megakaryocyte differentiation. It identifies specific HSPC populations that depend on THPO as well as showing that THPO may not be directly responsible for the loss of quiescence observed in THPO KO models, resolving previous contradictory findings.

References 1. Choi JS, Mahadik BP, Harley BA. Engineering the hematopoietic stem cell niche: frontiers in biomaterial science. Biotechnol J. 2015;10(10):1529-1545. 2. Birbrair A, Frenette PS. Niche heterogeneity in the bone marrow. Ann N Y Acad Sci. 2016;1370(1):82-96. 3. de Graaf CA, Metcalf D. Thrombopoietin and hematopoietic stem cells. Cell Cycle. 2011;10(10):1582-1589. 4. Qian H, Buza-Vidas N, Hyland CD, et al. Critical role of thrombopoietin in maintaining adult quiescent hematopoietic stem cells. Cell Stem Cell. 2007;1(6):671-684. 5. de Sauvage FJ, Carver-Moore K, Luoh SM, et al. Physiological regulation of early and late stages of megakaryocytopoiesis by thrombopoietin. J Exp Med. 1996;183(2):651-656. 6. Woolthuis CM, Park CY. Hematopoietic stem/progenitor cell commitment to the megakaryocyte lineage. Blood. 2016;127(10): 1242-1248. 7. Ema H, Takano H, Sudo K, Nakauchi H. In vitro self-renewal division of hematopoietic stem cells. J Exp Med. 2000;192(9):12811288. 8. Nishikii H, Kanazawa Y, Umemoto T, et al. Unipotent megakaryopoietic pathway bridging hematopoietic stem cells and mature megakaryocytes. Stem Cells. 2015;33(7): 2196-2207. 9. Besancenot R, Roos-Weil D, Tonetti C, et al. JAK2 and MPL protein levels determine

haematologica | 2021; 106(7)

Disclosures No conflicts of interest to disclose. Contributions AO’N conceived and designed the study, collected, analyzed and interpreted data and wrote the manuscript. DC wrote the manuscript; DT and A’QBBAM collected data. AI-N provided study materials and collected data. TS conceived and designed the study, organized financial support and gave final approval of the manuscript. Acknowledgments The authors thank Dr. Md. Zakir Hossain at CSI, NUS for assistance in the design and generation of the Thpo flox mice and Prof. Qingde Wang of University of Pittsburgh Medical Center for kindly providing Alb-Cre mice.

TPO-induced megakaryocyte proliferation vs differentiation. Blood. 2014;124(13):21042115. 10. Fleischman AG, Tyner JW. JAK2 V617F down-modulates MPL. Blood. 2012;119(20): 4579-4580. 11. Decker M, Leslie J, Liu Q, Ding L. Hepatic thrombopoietin is required for bone marrow hematopoietic stem cell maintenance. Science. 2018;360(6384):106-110. 12. Pronk CJ, Rossi DJ, Mansson R, et al. Elucidation of the phenotypic, functional, and molecular topography of a myeloerythroid progenitor cell hierarchy. Cell Stem Cell. 2007;1(4):428-442. 13. Oguro H, Ding L, Morrison SJ. SLAM family markers resolve functionally distinct subpopulations of hematopoietic stem cells and multipotent progenitors. Cell Stem Cell. 2013;13(1):102-116. 14. Ema H, Morita Y, Suda T. Heterogeneity and hierarchy of hematopoietic stem cells. Exp Hematol. 2014;42(2):74-82. 15. Yamamoto R, Morita Y, Ooehara J, et al. Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells. Cell. 2013;154(5):1112-1126. 16. Luc S, Anderson K, Kharazi S, et al. Downregulation of Mpl marks the transition to lymphoid-primed multipotent progenitors with gradual loss of granulocyte-monocyte potential. Blood. 2008;111(7):3424-3434. 17. Khoramian Tusi B, Socolovsky M. Highthroughput single-cell fate potential assay of

murine hematopoietic progenitors in vitro. Exp Hematol. 2018;60:21-29. 18. Bruns I, Lucas D, Pinho S, et al. Megakaryocytes regulate hematopoietic stem cell quiescence through CXCL4 secretion. Nat Med. 2014;20(11):1315-1320. 19. Nakamura-Ishizu A, Takubo K, Fujioka M, Suda T. Megakaryocytes are essential for HSC quiescence through the production of thrombopoietin. Biochem Biophys Res Commun. 2014;454(2):353-357. 20. Zhao M, Perry JM, Marshall H, et al. Megakaryocytes maintain homeostatic quiescence and promote post-injury regeneration of hematopoietic stem cells. Nat Med. 2014;20(11):1321-1326. 21. Greenbaum A, Hsu YM, Day RB, et al. CXCL12 in early mesenchymal progenitors is required for haematopoietic stem-cell maintenance. Nature. 2013;495(7440):227-230. 22. Veldkamp CT, Peterson FC, Pelzek AJ, Volkman BF. The monomer-dimer equilibrium of stromal cell-derived factor-1 (CXCL 12) is altered by pH, phosphate, sulfate, and heparin. Protein Sci. 2005;14(4):1071-1081. 23. Nakamura-Ishizu A, Matsumura T, Stumpf PS, et al. Thrombopoietin metabolically primes hematopoietic stem cells to megakaryocyte-lineage differentiation. Cell Rep. 2018;25(7):1772-1785. 24. Kovtonyuk LV, Manz MG, Takizawa H. Enhanced thrombopoietin but not G-CSF receptor stimulation induces self-renewing hematopoietic stem cell divisions in vivo. Blood. 2016;127(25):3175-3179

1891

ARTICLE Ferrata Storti Foundation

Hemophagocytosis

A pilot study of ruxolitinib as a front-line therapy for 12 children with secondary hemophagocytic lymphohistiocytosis

Qing Zhang,1* Ang Wei,2* Hong-Hao Ma,2 Li Zhang,2 Hong-Yun Lian,2 Dong Wang,2 Yun-Ze Zhao,2 Lei Cui,1 Wei-Jing Li,1 Ying Yang,2 Tian-You Wang,2 Zhi-Gang Li,1# and Rui Zhang2#

Laboratory of Hematologic Diseases, Beijing Pediatric Research Institute, Beijing Children’s Hospital, Capital Medical University, National Center for Children’s Health and 2Beijing Key Laboratory of Pediatric Hematology Oncology, National Key Discipline of Pediatrics, Key Laboratory of Major Diseases in Children, Ministry of Education, Department of Hematology Oncology Center, Beijing Children’s Hospital, Capital Medical University, National Center for Children’s Health, Beijing, China

Haematologica 2021 Volume 106(7):1892-1901

QZ and AW contributed equally as co-first authors.

ZL and RZ contributed equally as co-senior authors.

ABSTRACT

Correspondence: RUI ZHANG ruizh1973@126.com ZHI-GANG LI ericlzg70@hotmail.com Received: April 2, 2020. Accepted: July 21, 2020. Pre-published: July 30, 2020. https://doi.org/10.3324/haematol.2020.253781

1892

emophagocytic lymphohistiocytosis (HLH) is an immune-regulatory disorder characterized by excessive production of inflammatory cytokines. The treatment recommendations of the HLH1994 and HLH-2004 protocols have long been used in HLH therapy, but some patients still do not respond well to or have unacceptable side effects from conventional therapies. It is believed that cytokine-targeted strategies that directly target disease-driving pathways will be promising options for HLH. This prospective study aimed to investigate the efficacy and safety of ruxolitinib, a Janus kinase 1/2 inhibitor, as a front-line therapy in children with secondary HLH. Twelve newly diagnosed patients without previous treatment were enrolled in this study with a median follow-up of 8.2 (range, 7.1-12.0) months, including eight cases of Epstein-Barr virus associated HLH (EBV-HLH), two cases of autoinflammatory disorder (AID)- associated HLH, and two cases of unknown etiology. Patients received oral ruxolitinib dosed on 2.5 mg, 5 mg or 10 mg twice daily depending on the body weight for 28 consecutive days. The overall response rate at the end of treatment (day 28) was 83.3% (ten of 12), with 66.7% (eight of 12) in complete response (CR), 8.3% (one of 12) in partial response (PR), and 8.3% (one of 12) in HLH improvement. Among the patients achieving CR, 87.5% (seven of eight) maintained CR condition more than 6 months, and one patient with EBV-HLH relapsed following CR. For the EBV-HLH subgroup, all eight patients responded to ruxolitinib, with a CR rate of 75% and a PR rate of 25%. Two patients with AID-associated HLH had quite different responses, with one showing reversal of the HLH abnormalities soon and the other showing no improvement, as did the two cases of unknown etiology. Patients who had no response or discontinued ruxolitinib all responded well to the subsequent HLH-1994 regimen. The expected 6-month event-free survival rate was 58.3±10.2%. No serious adverse effects were reported. Our study provides further support for the possibility of ruxolitinib targeted therapy for secondary HLH in children. This study was registered in the Chinese Clinical Trials Registry Platform (http://www.chictr.org.cn/) as clinicaltrials gov. Identifier: ChiCTR2000029977.

Introduction Hemophagocytic lymphohistiocytosis (HLH) is an immune disorder characterized by uncontrolled T-lymphocyte and macrophage activation and excessive production of inflammatory cytokines. Patients present with multiple clinical features, including haematologica | 2021; 106(7)

Study of Ruxolitinib as a front-line therapy for pediatric HLH

fever, lymphadenopathy or hepatosplenomegaly, cytopenia, coagulopathy, and potentially life-threatening multisystem organ dysfunction. If left untreated, HLH can result in a high risk of death.1 Currently, HLH-1994 or HLH-2004 regimen is the standard HLH treatment strategy. In this regimen, etoposide and dexamethasone, with or without cyclosporine A, are used to treat active HLH. Long-term results of the HLH-1994 regimen showed that patients had a 5-year survival of 54±6%;2 compared with the HLH-1994 data, the HLH-2004 protocol did not improve HLH outcome significantly with a 5-year survival of 61%,3 which indicates that HLH treatment has not progressed significantly in the past few decades. Furthermore, based on the HLH-1994 or HLH-2004 regimen, intense cytotoxic chemotherapy can induce serious myelosuppressive or broadly immunosuppressive effects leading to severe infection and even death. In addition, the long-term side effect of etoposide in the regimen, secondary tumor risk, also deserves more attention.4-6 Therefore, prospective clinical trials investigating novel pharmacologic treatments for HLH are urgently needed. Recently, interleukin-1 (IL-1) inhibitors, interferon-γ (IFN-γ) monoclonal antibody and others have been suggested as possible treatment options, with varying clinical effects.7,8 Excessive production of cytokines, including IFN-γ, IL-10 and IL-6, contributes greatly to the pathogenesis of HLH. These overproduced cytokines bind to a broad array of specific receptors and activate the downstream JAK-STAT dependent signal pathway, which finally promotes the transcription of numerous downstream proinflammatory genes.9,10 Based on their essential roles in transmitting cytokine-induced signals, JAK inhibition might serve as a valid therapeutic approach in HLH. Recently, ruxolitinib (RUX), an oral selective JAK1/2 inhibitor, has shown promise in mouse models of primary and secondary HLH.11-13 When RUX is administered, cytokine production and tissue damage are decreased, leading to improved survival in mice. These data have led to interest in the use of RUX clinically for HLH treatment. Recently, RUX was used to treat refractory HLH patients in several case studies.14-16 There are also reports that RUX was used as a first-line treatment in one adult and one childhood HLH patient, and led to clinical remission.17, 18 In addition, a phase I clinical trial (clinicaltrails gov. Identifier: NCT02400463) using RUX in newly diagnosed adult HLH patients is ongoing and has published its preliminary data from the first five enrolled patients, which suggests that RUX is active and safe in that setting.19 However, since the etiology of HLH is complex and the severity of illness varies, the clinical outcome of RUX in the treatment of HLH still needs further investigation to identify the optimal dose and duration. Meanwhile, the specific subtypes of HLH which might show better sensitivity to RUX and the association between treatment response and risk stratification are still unclear. Therefore, we performed an open-label, single-arm, pilot study to investigate the efficacy and safety of RUX as a first-line agent in pediatric HLH and try to clarify the above uncertainties.

Methods Patients Patients who were enrolled in this study fulfilled the following criteria: (i) met HLH-2004 diagnostic criteria;20 (ii) had a new diagnosis of HLH; (iii) had no prior chemotherapy treatment for HLH haematologica | 2021; 106(7)

before screening; (iv) were male or female, less than 18 years of age and (v) signed an informed consent form before participating in the study. Patients who had any one of the following were ineligible: serious renal dysfunction (creatinine clearance <15 mL/min or glomerular filtration rate <15 mL/min), liver cirrhosis with a Model for End-stage Liver Disease (MELD) score >20, heart function above grade II (New York Heart Association), presence of a malignancy, parasitic infection, or a history of severe allergic, anaphylactic, or other hypersensitivity reactions to chemicals. Epstein-Barr virus-associated HLH (EBV-HLH) patients were defined as patients who met the HLH diagnosis criteria and whose EBV infection was confirmed by identifying increased EBV-DNA copy numbers (500 copies/L) in the peripheral blood or plasma (cell-free). Anti-EBV serological pattern showed the EBV status. In brief, EBNA-IgG-negative and EBCA-IgG (or IgM)-positive antibodies would indicate the first exposure to EBV, while positive EBNA-IgG, EBCA-IgG and EBCA-IgM antibodies would indicate EBV reactivation from a previous infection. Autoinflammatory disorder (AID)-HLH patients were defined as patients who met HLH diagnosis criteria and had excessive systemic inflammation, leading to recurrent fever, rashes, and IL-6 overproduction and no evidence of infection, tumor or specific antibody involvement. Patients were categorized into high- and low-risk groups. High risk was defined as any central nervous system involvement or accordance with at least three of the following criteria: i) age ≤2 years; ii) serious non-laboratory manifestations: severe hepatosplenomegaly, active bleeding, or icterus; iii) absolute neutrophil count <0.5×109/L; iv) soluble CD25 >25,000 pg/mL; v) ferritin >2,000 g/L; vi) a >10-fold increase in IFN-γ from normal levels; and vii) alanine aminotransferase (ALT) >200 U/L. Patients who do not meet the above criteria are considered low risk

Study design and ruxolitinib treatment protocol The study was registered in the Chinese Clinical Trials Registry Platform (clinicaltrials gov. Identifier: ChiCTR2000029977) and approved by the Ethics Committee of Beijing Children’s Hospital. This is a single-arm, open-label, pilot study to investigate the efficacy and safety of RUX as a front-line therapy in children with secondary HLH. Twelve patients were enrolled in this study. Patients in this study received oral RUX phosphate tablets on a 28-day cycle for one cycle. The dose was 2.5 mg, 5 mg or 10 mg twice daily depending on the body weight (≤10 kg, ≤20 kg or >20 kg, respectively). Therapy was changed immediately when there was no response after 3 days of treatment or at any time during treatment due to disease progression, relapse or toxic effects requiring drug discontinuation. The data cutoff for the primary analysis occurred when all patients completed day 28 of or discontinued therapy.

Study end points The primary efficacy endpoint was the overall response (OR) rate at day 28 of the last dose, including the proportion of patients achieving a complete response (CR), a partial response (PR) and HLH improvement. Secondary efficacy endpoints included the 6-month event-free survival (EFS, defined as the time from initial RUX treatment to the first occurrence of disease progression, relapse or death; for non-responders, EFS was defined as date of enrollment plus 1 day). Other end points included the durability of response, symptom reduction, dynamic changes of key biomarkers during treatment, the relationship between treatment response and risk stratification, and safety. In addition, for patients who discontinued and changed therapy regimens, we also counted the subsequent treatment responses and outcomes. 1893

Q. Zhang et al.

Assessment of treatment Efficacy was evaluated 7, 14 and 28 days after initiating RUX therapy. The assessment of treatment was mainly based on the

criterion previously described in studies for pediatric HLH.8, 21 A complete response was defined as normalization of all of the quantifiable symptoms and laboratory markers of HLH, including

Table 1. The main clinical features of enrolled patients at diagnosis.

Patient 1*

Patient 2+

Patient 3+

Patient 4

Patient 5*

13.2 / 41.5 Female 11 days IVIG ganciclovir corticosteroid (10 mg/qd, 2 days) EBV-HLH Yes Yes (5 cm) Yes 103 1.13 0.45 60 4.25 1.23 4375 36982 14.42 ↓ 5.44 ↓ –

11.2 / 35.2 Male 6 days

2.1 / 12.4 Female 4 days

1.3 / 9.1 Male 1.5 months

5.6 / 17.5 Female 21 days

antibiotics

–

EBV-HLH Yes Yes (3.9 cm) Yes 92 1.99 0.39 88 2.47 2.97 1039.8 22097 17.20 18.61 –

EBV-HLH Yes Yes (2 cm) Yes 78 2.87 0.76 41 5.14 2.33 1547 41075 16.04 11.20 –

IVIG antibiotics AID-HLH Yes No Yes 72 2.94 0.37 19 3.31 1.05 837 19326 15.32 22.79 –

7.3 / 23.4 Male 13 days ganciclovir IVIG antibiotics ganciclovir corticosteroid antibiotics (10 mg/qd, 3 days) EBV-HLH EBV-HLH Yes Yes Yes (4.4 cm) Yes (3 cm) Yes Yes 93 101 2.9 2.58 0.63 0.59 89 111 4.56 6.59 1.06 1.08 3196 29350 33950 71345 17.02 16.79 15.25 24.54 – –

Clinical features

Patient 7

Patient 8§

Patient 9*

Patient 10

Patient 11*

Patient 12#

Age(years)/body weight (kg) Sex Duration before diagnosis

2.9 / 12.5 Male 12 days

13.4 / 40.8 Female 3 months IVIG antibiotics ganciclovir corticosteroid (dose unclear） EBV-HLH Yes Yes (2.7 cm) Yes 73 1.5 0.49 42 2.69 2.57 3148 42885 17.88 22.53 –

3.3 / 13.6 Male 7 days

2.5 / 11.2 Male 21 days

4.7 / 16.8 Male 7 days

3.9 / 16.6 Male 7 days

antibiotics

antibiotics corticosteroid （5mg/qd, 4 days)

antibiotics

EBV-HLH Yes Yes (3.5 cm) Yes 101 4.13 1.41 39 4.68 0.8 139139 218875 16.20 19.08 –

Unclear Yes Yes (3.6 cm) Yes 76 2.29 1.27 112 3.52 3.61 4505 23132 16.66 13.51 –

EBV-HLH Yes Yes (2.8 cm) Yes 52 2.45 0.32 109 2.75 1.3 4419 44000 23.83 15.32 –

AIDs-HLH Yes No Yes 113 13.6 11 155 1.7 7.09 2525 16978 11.72 ↓ 12.26 –

Clinical features Age(years) /body weight (kg) Sex Duration before diagnosis Previous treatment

Aetiology Fever (>38.5 °C) Splenomegaly (size, under ribs) Hemophagocytosis in bone marrow Hemoglobin (g/L) White blood cell count (x109/L) Absolute neutrophil count (x109/L) Absolute platelet count (x109/L) Triglyceride (mmol/L) Fibrinogen (g/L) Ferritin concentration (mg/L) Soluble CD25(pg/mL) Natural killer cell activity (%) Natural killer cell CD107a (%) CNS involvement

Previous therapy

Etiology Fever (38.5°C) Splenomegaly (size, under ribs) Hemophagocytosis in bone marrow Hemoglobin (g/L) White blood cell count (x109/L) Absolute neutrophil count (x109/L) Absolute platelet count (x109/L) Triglyceride (mmol/L) Fibrinogen (g/L) Ferritin concentration (mg/L) Soluble CD25(pg/mL) Natural killer cell activity (%) Natural killer cell CD107a (%) CNS involvement

IVIG antibiotics corticosteroid （5mg/qd, 4 days) Unclear Yes No Yes 99 1 0.29 38 6.2 1.7 946 14219 28.34 11.67 –

Patient 6*

IVIG: intravenous immunoglobulin; EBV-HLH: Epstein-Barr virus-associated HLH; AID: autoinflammatory disorders; CNS: central nervous system; qd: once daily. Normal reference range: Ferritin concentration< 500 mg/L, soluble CD25≤6,400pg/mL, Natural killer cell activity ≥ 15.11%, Natural killer cell CD107a ≥ 10%. #Patient diagnosed as EBV-HLH in July 2017, treated by HLH-1994 protocol and had been off all therapy for > 2 years in good health before this diagnosis. +Patients who experienced the first exposure to EBV; *Patients who were undergoing EBV reactivation;§Patient who was identified as CAEBV.

1894

haematologica | 2021; 106(7)

Study of Ruxolitinib as a front-line therapy for pediatric HLH Table 2. Response outcomes.

Patient

Etiology

Risk Treatment stratification days

Reasons for discontinuation

Response from treatment to day28

Duration of response until data cutoff (days)

Follow-up treatment

Survival

Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Patient 7 Patient 8# Patient 9*

EBV-HLH EBV-HLH EBV-HLH AIDs-HLH EBV-HLH EBV-HLH Unclear CAEBV-HLH EBV-HLH

High low low High High High Low High High

28 28 28 28 28 28 28 28 5

/ / / / / / / / Critically ill; Potential CNS involvement

CR CR CR CR CR CR CR Relapse after CR HLH improvement

353 368 264 273 237 258 222 28 5

/ / / / / / / HLH-1994 HLH-1994

Yes Yes Yes Yes Yes Yes Yes No Yes

Patient 10* Patient 11* Patient 12*

Unclaer EBV-HLH AIDs-HLH

High High Low

3 7 3

Continuous progress Progress after PR Continuous progress

NR PR NR

1 7 1

HLH-1994 HLH-1994 HLH-1994

Yes Yes Yes

HLH: hemophagocytic lymphohistiocytosis; CR: complete response；PR: partial response；NR: no response; CNS: central nervous system; #: patients who were treated by HLH-1994 regimen after Ruxolitinib discontinuation, but reactivation occurred at the 4th week, and refused salvage treatment due to financial difficulty. *: patients who were treated by HLH1994 regimen after Ruxolitinib discontinuation, achieved complete remission, and stopped treatment after 8 weeks except patient-12 who still underwent maintenance treatment.

no fever (body temperature <37.5°C), no cytopenia (absolute neutrophil count ≥1.0×109/L and platelet count ≥100×109/L with the absence of granulocyte colony-stimulating factor (G-CSF) and transfusion support must be documented for ≥4 days), no evidence of coagulopathy (fibrinogen levels >1.50 g/L), normal levels of soluble CD25, ferritin and triglyceride, normal spleen size as measured by abdominal ultrasound, and no neurological and CSF abnormalities attributed to HLH. A partial response was defined as normalization of ≥3 of the aforementioned HLH abnormalities (including CNS abnormalities) and no progression of other aspects of HLH disease pathology. HLH improvement was defined as at least a 50% improvement in ≥3 HLH abnormalities from baseline. At least a 50% worsening in two or more signs or laboratory abnormalities was considered progressive disease. Three or more symptoms and laboratory markers developing into abnormalities after achieving complete response was defined as relapse.

Side effects and complications The safety population included all patients who received at least one dose of RUX. Adverse events were assessed according to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 3.0. (http://ctep.cancer.gov/protocol Development/electronic_applications/docs/ctcaev3.pdf). In detail, we monitored the enrolled patients for any signs of toxicity and complications every day for the first week of treatment, weekly during weeks 2-4, and monthly thereafter, including routine blood tests, coagulation, infections (i.e., latent tuberculosis, adenovirus, Epstein-Barr virus, cytomegalovirus, herpes zoster, Pneumocystis jirovecii and fungal infections), cardiac function (myocardial enzyme spectrum, electrocardiogram, ultrasound cardiogram), renal function (serum creatinine, urea nitrogen, and creatinine clearance), liver function (ALT, AST, GGT, ALB, TBIL, I-BIL, D-BIL), and other adverse drug reactions such as dizziness, headache, rash, dyspnea and gastrointestinal reaction.

Statistical analysis SPSS 20.0 software (SPSS, Chicago, IL, USA) was used for statistical analysis. All normally distributed data are represented as means ± standard deviations, and comparisons of multiple parameters between groups were performed by independent sample thaematologica | 2021; 106(7)

Figure 1. Event-free survival of patients defined as the time from the initial dose of ruxolitinib to the first occurrence of disease progression, relapse or death (event).

tests. All data that were not distributed normally are represented as medians and ranges, and comparisons of multiple parameters between groups were performed by Wilcoxon rank sum tests. Patient survival was estimated by the Kaplan-Meier method, and differences in survival between groups were estimated by the logrank test. In all analyses, P<0.05 was considered to denote a significant difference, and P<0.01 was considered very significant. For patients who discontinued RUX within 7 days, the last known values were used as the day 7 laboratory results.

Results Characteristics of the patients Twelve pediatric patients with newly diagnosed HLH were enrolled at Beijing Children’s Hospital from June 2019 to October 2019, including eight males and four females. The main clinical features of the enrolled patients at diagnosis are summarized in Table 1. The median age was 4.7 (range, 1.3-13.4) years. The median duration before HLH 1895

Q. Zhang et al.

Figure 2. Dynamics of hemophagocytic lymphohistiocytosis disease features during ruxolitinib treatment in eight patients who achieved complete response. (A) daily maximum temperature, within hours of receiving the first dose of ruxolitinib (RUX), most patient’s fever resolved; (B to D) The inflammatory markers soluble CD25 (sCD25), ferritin and interferon-γ (IFN-γ) cytokine level. Normal range values of sCD25, ferritin and IFN-γ are ≤6,400 pg/mL, ≤500 mg/L and ≤8 pg/mL respectively by the clinical laboratory; (E to G) absolute neutrophil count, hemoglobin concentration, absolute platelet count; (H) fibrinogen concentration; (I) triglyceride; (J) alanine aminotransferase (ALT) concentration. The dotted line on the x-axis of each graph indicate the start of RUX treatment.

1896

haematologica | 2021; 106(7)

Study of Ruxolitinib as a front-line therapy for pediatric HLH

diagnosis was 11.5 (range, 4-92) days. From onset to diagnosis, except for patient 3 who did not receive any treatment, the patients had been typically treated with antibiotics, ganciclovir or intravenous immunoglobulin (Ig). Five subjects had been given corticosteroids for 2-4 days, but all discontinued use before they were referred to our hospital due to a lack of response. Whole-exome sequencing (WES) was performed for all patients and their parents, and no pathogenic gene mutation associated with primary HLH was found. No patient had central nervous system (CNS) involvement at enrollment. All subjects had negative evaluations for malignant tumors after bone marrow or cervical lymph node biopsy. The primary disease could not be determined in two of the 12 HLH patients; in the other ten patients, eight cases were EBV-HLH, and two cases were AID-HLH. In the EBV-HLH subgroup, two patients experienced their first exposure to EBV, five involved EBV reactivation, and one case was clearly identified as chronic active EBV (CAEBV). We classified the patients according to our stratification criteria with four patients at low risk and eight at high risk.

Efficacy of ruxolitinib treatment Table 2 shows the response outcomes in detail. Among the entire cohort, eight of 12 patients completed the 28-day treatment protocol and achieved CR, but one relapsed at the time of the fourth scheduled response assessment (day 28), and was refractory to alternative HLH-1994 regimen and finally died. Four of 12 patients discontinued the therapy after 3-7 days of treatment and were adjusted to the HLH-1994 regimen for the following reasons: two subjects had no response to RUX and progressively deteriorated; one subject had a partial response for the first 7 days of therapy but progressed soon; one achieved obvious HLH improvement over the first 5 days of treatment; however, the patient remained critically ill and developed potential CNS involvement with drowsiness, although his brain MRI and cerebrospinal fluid (CSF) exam showed no abnormalities. According to the above, for the primary endpoint, the best OR rate to RUX at the end of treatment protocol was 83.4% (ten of 12 patients) , with 66.7% (eight of 12 patients) in CR, 8.3% (one of 12 patients) in PR, and 8.3% (one of 12 patients) in HLH improvement. Among the patients who achieved CR, 87.5% (seven of eight patients) maintained CR condition for >6 months. For the eight patients with EBV-HLH, the overall response was 100%, with 75% (six of eight patients) in CR and 25% (two of eight patients) in PR; however, 16.7% (one of six patients) relapsed after CR. Two AID-HLH patients had quite different responses, with one showing reversal of HLH abnormalities soon and the other showing no improvement, as did the two patients of unknown etiology. All patients who had no response or discontinued RUX responded well to the subsequent HLH-1994 regimen, achieved CR and remained off therapy except one (AID-HLH) still in treatment maintenance until data cutoff. The EFS time was calculated from the date of RUX therapy. All patients were followed up until death or June 7, 2020 (time of data cutoff), whichever occurred first, with a median follow-up of 8.2 (range, 7.1-12.0) months. A total of five of 12 patients had an event, and the expected 6-month EFS rate was 58.3%±14.2% (Figure 1). Patients in the highrisk group showed a tendency towards a worse EFS rate than those in the low-risk group, but there was no statistical significance in this small-scale analysis (50.0±17.7% vs. haematologica | 2021; 106(7)

Table 3. Adverse events observed in this study.

patient Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6

Dose Ruxolitinib Possible adverse Event led to (twice daily) treatment days event (Grade*) discontinuation 10 mg 10 mg 5 mg 2.5 mg 10 mg 10 mg

28 28 28 28 28 28

Patient 7 Patient 8

5 mg 10 mg

28 28

Patient 9 Patient 10 Patient 11 Patient 12

5 mg 5 mg 5 mg 5 mg

5 3 7 3

oral ulcer (1) Sweating (1) Elevated ALT (1) Elevated ALT (1) Nausea (2) Gastritis (1) Elevated ALT (1) Nausea (1) Decreased appetite (2) Constipation (1) -

– – – – – – – – – – – –

*Graded according to National Cancer Institute Adverse Event Common Toxicity Criteria version 3.0. No serious adverse events occurred, including cytopenia, secondary infections, acute liver failure, acute heart failure and acute renal failure. ALT: alanine aminotransferase.

75.0±21.7%, P=0.556) (Online Supplementary Figure S1 in the Online Supplementary Appendix).

Changes in evaluation indicators before and after ruxolitinib therapy We analyzed 12 HLH-associated symptoms and laboratory parameters before, and 1, 2, 4 and 8 weeks after RUX treatment. Some indicators, such as temperature and routine blood, were tested every day within the first treatment week as much as possible. Rapid resolution of various indicators was observed with continued therapy. At day 14, a complete recovery in almost all parameters was achieved in eight patients with good response (Figure 2), whereas patients with no response or progress had a persistent abnormality in many indicators (Online Supplementary Figure S2). No relationship was found between patient characteristics at diagnosis and the treatment response (Online Supplementary Table 1). We further analyzed the data before and 1 week after RUX treatment in the CR and non-CR groups respectively. There was a significant difference between the two groups in the improvement of temperature, soluble CD25, ferritin, IFN-γ and neutrophil counts, suggesting that these parameters were important indicators for evaluating the treatment response of RUX in the early phase (Figure 3). Other key cytokines, including IL-6, IL-10, TNF-a and IL-18, showed a similar trend to IFN-γ (Online Supplementary Figure S3). In detail, within hours of receiving the first dose of RUX, ten of 12 patients became afebrile; however, for the patients who did not respond well to RUX, the antipyretic effect was not sustainable, and the fever recurred within 48-72 hours (Figure 2A; Online Supplementary Figure S2A). Dramatic improvements in soluble CD25 and serum ferritin were observed, falling rapidly to normal levels within approximately 1-2 weeks (Figure 2B and C); in contrast, for patients (patient 10 and 12) who were nonresponsive to RUX, soluble CD25 and serum ferritin increased to high levels (Online Supplementary Figure S2B and C). We also monitored the level of IFN-γ, a key cytokine in HLH. High 1897

Q. Zhang et al.

Figure 3. Comparison of clinical indicators changes before and 1 week after ruxolitinib therapy in the complete response group (CR, n=8) and non-CR group (n=4) respectively. For patients who discontinued ruxolitinib (RUX) within 7 days, the last known values were used as the week-1 laboratory results. For statistical analysis, the paired sample Wilcoxon signed rank test was applied.

1898

haematologica | 2021; 106(7)

Study of Ruxolitinib as a front-line therapy for pediatric HLH

inflammation was relieved in all well-responsive patients, with IFN-γ decreasing rapidly to normal levels at 2 weeks (Figure 2D). Cytopenia and coagulopathy progressively improved in all responding patients with blood transfusion support weaned off within 1 week. Platelet counts, in particular, increased to the normal range quickly within 1 week. Neutrophil count improvement varied, but all reached 1×109/L at 2 weeks and 2×109/L at 4 weeks. Hemoglobin and fibrinogen concentrations improved with slightly slower kinetics (Figure 2E and H). Triglyceride was not improved immediately with a rapid phase of improvement 1 week after starting RUX in most individuals (Figure 2I). Hepatic dysfunction resolved within 2 weeks (Figure 2J). In addition, EBV DNA load decreased dramatically in all EBV-HLH patients (Figure 4). Among them, EBV infections were resolved in six of eight patients 1-2 weeks after RUX treatment, with cell-free (plasma) EBV DNA becoming undetectable and having no recurrence after stopping treatment until the data cutoff (June 7, 2020) except in patient 8 (CAEBV). The median time that plasma EBV DNA was maintained negative in these patients was approximately 8.7 (range, 0.5-12.0) months. Meanwhile, EBV DNA in whole blood also fell rapidly from very high levels (105-107 copies/mL) to modest levels (103-104 copies/mL), and fluctuated at these modest levels during follow-up.

Side effects Treatment was well tolerated, with no toxicities leading to dose-reductions or interruptions of RUX observed. All possible adverse events are reported in Table 3. Although RUX has been reported to have a risk of thrombocytopenia or anemia in other settings, these side effects were not observed in our study. Cytopenia is a major clinical feature of HLH. However, patients who achieved a response to RUX had neutrophil counts, platelet counts and hemoglobin concentrations that improved gradually compared to pretreatment levels. No unusual infections were noted. Renal function was analyzed by creatinine clearance, and cardiac function tested by myocardial enzyme spectrum and ultrasound cardiogram was normal during the clinical course. Three individuals had moderately elevated ALT at 60 days after treatment, which may not have been due to RUX, as they had been off the drug for approximately 1 month. Two patients hadtwo or more grade 1-2 gastroin-

testinal adverse events (nausea, gastritis, decreased appetite) and were resolved soon by supportive therapies. In addition, three patients showed mild oral ulcers, sweating or constipation, and recovered without special treatment.

Discussion HLH is a disorder characterized by high inflammatory cytokine production induced by excessive immune activation. Recently, cytokine-targeted approaches have been suggested as possible treatment options, such as emapalumab.8, 22, 23 Rather than targeting a single cytokine, RUX can blunt numerous cytokines via inhibition of the JAK 1/2STAT1 pathway, which makes its clinical use for HLH more rational.11-13 Currently, RUX is often recommended for refractory/relapsed HLH as a salvage treatment regimen; however, its use as a first-line agent is still based on only a small number of case reports, clinician experience or small clinical studies (clinicaltrials gov. Identifier: NCT02400463). To date, the optimal duration of RUX therapy, the utility of concurrent corticosteroids, and the effect of disease etiology or risk stratification on treatment response remain uncertain. In this study, we explored the efficacy and safety of RUX monotherapy instead of chemotherapy for secondary HLH in children. RUX has been used for myelofibrosis,24 polycythemia,25 and graft-versus-host disease (GvHD)26, 27 and is generally dosed between 5 mg and 25 mg twice daily, demonstrating good clinical benefits and tolerance. In addition, in patients with relapsed or refractory myeloproliferative neoplasms especially those with JAK2 or CSF3R mutations, the RUX dose can be increased to 50-200 mg twice a day, and is reasonably tolerated overall.28, 29 The pediatric dosing of RUX in HLH is unestablished. The dose in this study (2.5-10 mg) was modified with reference to the lower dose used for GvHD in children,30 which, similar to HLH, is characterized by the production of high levels of proinflammatory cytokines. In order to eliminate the probable interference of corticosteroid use, we enrolled newly diagnosed patients who had not undergone previous corticosteroid treatment or who had used but discontinued corticosteroids at least 3 days before trial screening due to a lack of response.

Figure 4. Eppstein-Barr virus DNA levels in the whole blood and plasma after treatment with ruxolitinib. Reported Eppstein-Barr virus (EBV) DNA values are limited to 500 copies/L by the clinical laboratory. The dotted line on the x-axis of each graph indicate the start of RUX treatment.

haematologica | 2021; 106(7)

1899

Q. Zhang et al. Our results demonstrate that RUX is effective and safe in pediatric patients with secondary forms of HLH. The rapid resolution of clinical symptoms and normalization of clinical laboratory parameters were observed. The OR rate at the end of the treatment protocol (day 28) was 83.3% (ten of 12 patients), with 66.7% (eight of 12 patients) in complete response, and all remained in CR until the data cutoff except for one who relapsed. No serious adverse effects were reported. These findings suggested that RUX may serve as a potential first-line treatment option for secondary HLH, which can greatly reduce the toxic effects compared with intense chemotherapy. In addition, for patients who need biopsy analysis, RUX instead can provide more opportunities to find the original disease cause than corticosteroid-based therapy. Our study also showed that RUX had a quick effect, and for patients for whom RUX did not work well, we can probably identify this trend within approximately 3 days. Notably, the patients who had poor response to RUX after 3-7 days of treatment all responded well to the subsequent HLH-1994 regimen. This further suggests the possibility that we can attempt to use RUX first for approximately 3 days to determine the treatment response; for patients with poor efficacy, we can then combine it with chemotherapy regimen, which may not cause delayed treatment and poor prognosis. Future studies are warranted to determine the viability of this idea. However, this does not mean that the efficacy of RUX is superior to that of HLH-1994/2004 regimen, which is also used to treat a major proportion of patients with primary HLH. Our main purpose is to sort suitable patients to avoid chemotherapy when possible, and make treatment available for patients deemed unsuitable for chemotherapy. We believe that RUX and standard chemotherapy may be two complementary first-line therapy strategies, as there are patients who cannot be solved by either RUX or chemotherapy. The key point is to find the specific HLH settings sensitive to RUX to guide better treatment in the future. EBV-HLH accounts for approximately 60% of all pediatric HLH patients in China. EBV-HLH also accounts for a significant proportion of patients who are resistant to standard treatment and have a poor prognosis.31 No previous studies have focused on the association between etiology and RUX treatment response. In addition, current ongoing trials investigating the use of RUX in HLH have not yet enrolled EBV-HLH patients.19 Therefore, our study focused more on to the relationship between EBVHLH and RUX treatment. We enrolled eight EBV-HLH patients, the response rate was 100% (eight of eight), the CR rate was 75% (six of eight), and persistent EBV infection in plasma of these subjects resolved rapidly. This suggests the possibility of RUX for treating EBV-HLH. However, one subject relapsed soon after CR, experienced disease recurrence 1 month after the subsequent HLH-1994 therapy, and finally died. This patient had CAEBV disease that induced HLH. For CAEBV patients with HLH signs, initiating HLH-directed treatment is required to suppress the life-threatening inflammatory process that underlies HLH. In our patient, RUX could transiently control her HLH condition, but was ineffective for the CAEBV. Generally, in the absence of hematopoietic stem cell transplantation (HSCT), therapy for CAEBV, including other cytotoxic chemotherapy, can often delay the progression of disease at best, but over 1900

time, the patients become refractory and progress to an irreversible stage. It is interesting that patients in our study were able to clear the persistent EBV DNA in the plasma (initial concentration 7.28×106-4.38×104copies/L) within 1 week after RUX treatment and had no EBV recurrence except for the CAEBV subject. We are not sure about the duration of this EBV DNA decline, since T- or natural killer (NK)-cell infection is prone to reactivation, whereas all patients in this study exhibited multiple-cell-type EBV infection predominated by the T- or NK-cell subset. There are several explanations for this decrease in EBV DNA. First, it may be related to the EBV infection being either a first infection or a reactivation. Self-limiting is possible in patients with a first EBV infection like infectious mononucleosis. In addition, emapalumab (an anti– IFN-γ monoclonal antibody) has been reported to resolve persistent EBV infection in one patient, perhaps due to relief of immune paralysis caused by IFN-γ .23 As RUX can blunt numerous cytokines, including IFN-γ , the resolution of the EBV infection in this study may also be attributed to the relief of immune paralysis after RUX treatment, which primed the body immunity and ultimately resolved the EBV infection spontaneously. Moreover, evidence suggests that host cell stress such as oxidative stress, hypoxia and inflammation can induce EBV reactivation from latency.32 All of these factors may be removed after the excessive inflammation has been controlled by RUX, leading to a switch of EBV cycle from a lytic state to a latent state. In addition, we enrolled two patients with autoinflammatory diseases-associated HLH, but these two individuals had quite different responses to RUX, with one showing reversal of the HLH abnormalities soon and the other showing no improvement. Similarly, two patients of unknown etiology also showed two different kinds of responses. The RUX dose may not have been sufficient for these non-/poor-responders in this study. Moreover, the great heterogeneity of HLH, such as the etiology, clinical features and risk stratification, may also contribute to this difference. Therefore, further study is needed to optimize the clinical use of RUX, such as identifying some sensitive biomarkers. In summary, our preliminary study provides further support for the possibility of RUX-based targeted therapy in pediatric patients with secondary HLH. Since patients with known CNS involvement were not enrolled in this study, we were unable to make statements on the efficacy of ruxolitinib in the CNS. The number of patients in this study was small, and the observation time was relatively short. We are currently performing a large-scale, open-label, prospective trial on RUX monotherapy for pediatric HLH, which may answer more uncertainties and provide further evidence for RUX treatment as a firstline therapy in HLH. Disclosures No conflicts of interest to disclose. Contributions QZ conducted the data analysis and wrote the manuscript; RZ and ZGL contributed to the design of the study; AW, HHM, LZ, HYL, DW and YZZ performed the clinical aspects of the study; LC, WJL and YY performed laboratory tests and helped with data analysis; TYW helped with the study design. All authors read and approved the final manuscript. haematologica | 2021; 106(7)

Study of Ruxolitinib as a front-line therapy for pediatric HLH

Funding The authors would like to thank grants from the National Natural Science Foundation of China (No. 81800189); Beijing Municipal Administration of Hospitals’ Youth Programme (QML20181205); the Scientific Research Common Program of Beijing Municipal Commission of Education (No.

References 1. Thomas W, Veer MV, Besser M. Haemophagocytic lymphohistiocytosis: an elusive syndrome. Clin Med (Lond). 2016; 16(5):432-436. 2. Trottestam H, Horne A, Aricò M, et al. Chemoimmunotherapy for hemophagocytic lymphohistiocytosis: long-term results of the HLH-94 treatment protocol. Blood. 2011; 118(17):4577-4584. 3. Bergsten E, Horne A, Aricó M, et al. Confirmed efficacy of etoposide and dexamethasone in HLH treatment: long-term results of the cooperative HLH-2004 study. Blood. 2017;130(25):2728-2738. 4. Pui CH, Ribeiro RC, Hancock ML, et al. Acute myeloid leukemia in children treated with epipodophyllotoxins for acute lymphoblastic leukemia. N Engl J Med. 1991; 325(24):1682-1687. 5. Henter JI, Elinder G, Lübeck PO, Ost A. Myelodysplastic syndrome following epipodophyllotoxin therapy in familial hemophagocytic lymphohistiocytosis. Pediatr Hematol Oncol. 1993;10(2):163-168. 6. Kitazawa J, Ito E, Arai K, Yokoyama M, Fukayama M, Imashuku S. Secondary acute myelocytic leukemia after successful chemotherapy with etoposide for EpsteinBarr virus-associated hemophagocytic lymphohistiocytosis. Med Pediatr Oncol. 2001; 37(2):153-154. 7. Miettunen PM, Narendran A, Jayanthan A, Behrens EM, Cron RQ. Successful treatment of severe paediatric rheumatic disease-associated macrophage activation syndrome with interleukin-1 inhibition following conventional immunosuppressive therapy: case series with 12 patients. Rheumatology (Oxford). 2011;50(2):417-419. 8. Locatelli F, Jordan MB, Allen C, et al. Emapalumab in children with primary hemophagocytic lymphohistiocytosis. N Engl J Med. 2020;382(19):1811-1822. 9. Villarino AV, Kanno Y, Ferdinand JR, O'Shea JJ. Mechanisms of Jak/STAT signaling in immunity and disease. J Immunol. 2015;194(1):21-27. 10. Kiu H, Nicholson SE. Biology and significance of the JAK/STAT signalling pathways. Growth Factors. 2012;30(2):88-106.

haematologica | 2021; 106(7)

KM201910025011); the Special Fund of The Pediatric Medical Coordinated Development Center of Beijing Municipal Administration (No. XTZD20180202); National Science and Technology Key Projects (No. 2017ZX09304029003); Guangdong Province Key Laboratory of Popular High Performance Computers of Shenzhen University (SZU-GDPHPCL2017) .

11. Das R, Guan P, Sprague L, et al. Janus kinase inhibition lessens inflammation and ameliorates disease in murine models of hemophagocytic lymphohistiocytosis. Blood. 2016;127(13):1666-1675. 12. Maschalidi S, Sepulveda FE, Garrigue A, Fischer A, de Saint Basile G. Therapeutic effect of JAK1/2 blockade on the manifestations of hemophagocytic lymphohistiocytosis in mice. Blood. 2016;128(1):60-71. 13. Albeituni S, Verbist KC, Tedrick PE, et al. Mechanisms of action of ruxolitinib in murine models of hemophagocytic lymphohistiocytosis. Blood. 2019;134(2):147-159. 14. Sin JH, Zangardi ML. Ruxolitinib for secondary hemophagocytic lymphohistiocytosis: first case report. Hematol Oncol Stem Cell Ther. 2019;12(3):166-170. 15. Broglie L, Pommert L, Rao S, et al. Ruxolitinib for treatment of refractory hemophagocytic lymphohistiocytosis. Blood Adv. 2017;1(19):1533-1536. 16. Wang J, Wang Y, Wu L, et al. Ruxolitinib for refractory/relapsed hemophagocytic lymphohistiocytosis. Haematologica. 2020;105(5):e210-e212. 17. Slostad J, Hoversten P, Haddox CL, Cisak K, Paludo J, Tefferi A. Ruxolitinib as first-line treatment in secondary hemophagocytic lymphohistiocytosis: a single patient experience. Am J Hematol. 2018;93(2):E47-e49. 18. Zandvakili I, Conboy CB, Ayed AO, Cathcart-Rake EJ, Tefferi A. Ruxolitinib as first-line treatment in secondary hemophagocytic lymphohistiocytosis: a second experience. Am J Hematol. 2018;93(5):E123e125. 19. Ahmed A, Merrill SA, Alsawah F, et al. Ruxolitinib in adult patients with secondary haemophagocytic lymphohistiocytosis: an open-label, single-centre, pilot trial. Lancet Haematol. 2019;6(12):e630-e637. 20. Henter JI, Horne A, Aricó M, et al. HLH2004: diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis. Pediat Blood Cancer. 2007;48(2):124-131. 21. Marsh RA, Allen CE, McClain KL, et al. Salvage therapy of refractory hemophagocytic lymphohistiocytosis with alemtuzumab. Pediat Blood Cancer. 2013;60(1):101-109. 22. Marsh RA, Jordan MB, Talano JA, et al. Salvage therapy for refractory hemophago-

cytic lymphohistiocytosis: a review of the published experience. Pediat Blood Cancer. 2017;64(4). 23. Lounder DT, Bin Q, de Min C, Jordan MB. Treatment of refractory hemophagocytic lymphohistiocytosis with emapalumab despite severe concurrent infections. Blood Adv. 2019;3(1):47-50. 24. Verstovsek S, Mesa RA, Gotlib J, et al. A double-blind, placebo-controlled trial of ruxolitinib for myelofibrosis. N Engl J Med. 2012;366(9):799-807. 25. Vannucchi AM, Kiladjian JJ, Griesshammer M, et al. Ruxolitinib versus standard therapy for the treatment of polycythemia vera. N Engl J Med. 2015;372(5):426-435. 26. Przepiorka D, Luo L, Subramaniam S, et al. FDA approval summary: ruxolitinib for treatment of steroid-refractory acute graft-versushost disease. Oncologist. 2020;25(2):e328e334. 27. Zeiser R, Burchert A, Lengerke C, et al. Ruxolitinib in corticosteroid-refractory graftversus-host disease after allogeneic stem cell transplantation: a multicenter survey. Leukemia. 2015;29(10):2062-2068. 28. Pemmaraju N, Kantarjian H, Kadia T, et al. A phase I/II study of the Janus kinase (JAK)1 and 2 inhibitor ruxolitinib in patients with relapsed or refractory acute myeloid leukemia. Clin Lymphoma Myeloma Leuk. 2015;15(3):171-176. 29. Dao KT, Gotlib J, Deininger MMN, et al. Efficacy of ruxolitinib in patients with chronic neutrophilic leukemia and atypical chronic myeloid leukemia. J Clin Oncol. 2020; 38(10):1006-1018. 30. González Vicent M, Molina B, González de Pablo J, Castillo A, Díaz M. Ruxolitinib treatment for steroid refractory acute and chronic graft vs host disease in children: Clinical and immunological results. Am J Hematol. 2019;94(3):319-326. 31. Wang J, Wang Y, Wu L, Zhang J, Lai W, Wang Z. PEG-aspargase and DEP regimen combination therapy for refractory Epstein-Barr virus-associated hemophagocytic lymphohistiocytosis. J Hematol Oncol. 2016;9(1):84. 32. Li H, Liu S, Hu J, et al. Epstein-Barr virus lytic reactivation regulation and its pathogenic role in carcinogenesis. Int J Biol Sci. 2016; 12(11):1309-1318.

1901

ARTICLE Ferrata Storti Foundation

Hemostasis

Factor VIII activity and bleeding risk during prophylaxis for severe hemophilia A: a population pharmacokinetic model

Andreas Tiede,1 Faraizah Abdul Karim,2 Victor Jiménez-Yuste,3 Robert Klamroth,4 Sandra Lejniece,5 Takashi Suzuki,6 Andreas Groth7 and Elena Santagostino8

Haematologica 2021 Volume 106(7):1902-1909

1 Hannover Medical School (MHH), Hematology, Hemostasis, Oncology and Stem Cell Transplantation, Hannover, Germany; 2Haemophilia Centre, National Blood Centre, Kuala Lumpur, Malaysia; 3Hospital Universitario La Paz, Autónoma University, Madrid, Spain; 4Haemophiliezentrum, Klinik für Innere Medizin, Vivantes Klinikum im Friedrichshain, Berlin, Germany; 5Rîga East Clinical University Hospital, Chemotherapy and Hematology Clinic, Rîga, Latvia; 6Department of Laboratory Medicine, Tokyo Medical University, Tokyo, Japan; 7Novo Nordisk A/S, Søborg, Denmark and 8Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, IRCCS Cà Granda Foundation, Maggiore Hospital Policlinic, Milan, Italy

ABSTRACT

Correspondence: ANDREAS TIEDE tiede.andreas@mh-hannover.de Received: November 27, 2019. Accepted: April 22, 2020. Pre-published: April 23, 2020. https://doi.org/10.3324/haematol.2019.241554

uring factor VIII prophylaxis for severe hemophilia A, bleeding risk increases with time when factor VIII activity is below 1%. However, maintaining trough activity above 1% does not protect all patients from bleeding. The relationship between factor VIII activity during prophylaxis and bleeding risk has not been thoroughly studied. We investigated factor VIII activity and annualized bleeding rate for spontaneous bleeds during prophylaxis. A population pharmacokinetic model derived from three clinical trials was combined with dosing data and information on bleeding from patients’ diaries. Each patient’s time on prophylaxis was divided into five categories of predicted activity (0-1%, >1-5%, >5-15%, >15-50%, and >50%). Exposure time, mean factor VIII activity, and number of bleeds (from the patients’ diaries) were calculated for each activity category, and annualized bleeding rates estimated using negative binomial regression and a parametric model. Relationships between these bleeding rates and factor VIII activity were evaluated by trial phase (pivotal vs. extension) and age (adults/adolescents [≥12 years] vs. children [0-<12 years]). In total (n=187 patients; 815 patient-years’ exposure), factor VIII activity was predicted to be >1% for 85.64% of the time. The annualized bleeding rate decreased as factor VIII activity increased in each trial phase and age group. However, for a given activity level, bleeding rate differed substantially by trial phase and age. This suggests that bleeding risk can change over time and is influenced by factors independent of factor VIII pharmacokinetics and trough levels. When making decisions regarding target trough levels and the prophylactic regimen, the patients’ age, joint disease activity, and other bleeding risk determinants should be taken into consideration. Clinical trial registration numbers: NCT00840086; NCT01138501; NCT00984126

1902

Introduction Hemophilia is classified according to factor VIII (FVIII) plasma activity as “severe” (<1% of normal activity), “moderate” (1-5%), or “mild” (>5%-<40%).1,2 Severe hemophilia is characterized by spontaneous, recurrent bleeds into joints and muscles which can lead to chronic arthropathy, muscular atrophy, and deformities. Converting the clinical phenotype of hemophilia from severe to moderate has been the rationale for prophylaxis.3 In patients with severe hemophilia A, prophylaxis with replacement FVIII can prevent bleeds and structural joint damage when initiated at a young age,4 and can decrease bleeding frequency, slow joint disease progression, and improve quality of life - even when initiated in adults with established joint damage.5 haematologica | 2021; 106(7)

FVIII activity and bleeding risk in hemophilia A

In severe hemophilia A, the annualized bleeding rate (ABR) during prophylaxis was shown to correlate with time spent with FVIII activity (FVIII:C) below 1%, as predicted from the patient’s individual FVIII pharmacokinetics (PK).6 However, actual time below 1% or any other targeted trough level also depends on the prescribed prophylaxis regimen and the patient’s adherence to it. The risk of spontaneous bleeds during periods with FVIII:C ≥1% in subjects on prophylaxis is unknown, although an epidemiological study in individuals with mild/moderate hemophilia suggested that maintaining FVIII:C above 1% will not protect all patients from bleeding.7,8 Evaluation of the association between FVIII:C and bleeding pattern is important and can support evidence-based tailoring of prophylaxis independently of the factor concentrate used. We have used data from the guardian clinical trial program to examine bleeding risks according to FVIII:C with a recombinant FVIII (rFVIII) molecule during prophylaxis in patients with severe hemophilia A. The product, turoctocog alfa (NovoEight®, Novo Nordisk Health Care AG, Zürich, Switzerland), is a B-domain-truncated rFVIII. Data from three studies were combined: a pivotal trial in adults and adolescents (guardian 1; NCT00840086),9 a pivotal trial in children (guardian 3; NCT01138501),10 and an extension trial (guardian 2, which was an extension trial of guardian 1 and 3; NCT00984126).11 Estimated mean (95% confidence interval) spontaneous ABR in these studies were 4.32 (3.34-5.59) (pivotal trial in adults and adolescents), 1.69 (0.94-3.03) (pivotal trial in children), and 1.34 (1.07-1.68) (extension trial).9-11 The PK characteristics of turoctocog alfa, which have been extensively studied using standard PK assessments (based on plasma FVIII:C) and a population PK model, have been shown to be consistent over time, reproducible between lots and similar to those of other commercially available FVIII products.12 Defining the relationship between FVIII:C and bleeding pattern is important not only because bleeding frequency is an important outcome, but also to guide robust, individualized prophylactic treatment schedules. Given the importance of spontaneous bleeding in hemophilia, we assessed the relationship between FVIII:C and ABR, as a measure of bleeding frequency for spontaneous bleeds, including spontaneous joint bleeds, which were recorded in dosing diaries by patients with severe hemophilia A who received prophylaxis with turoctocog alfa in the guardian clinical trials.

Methods Trial design and patients’ eligibility The pivotal trials (guardian 1 and 3) were multinational, phase III trials assessing prophylaxis with turoctocog alfa in adults/adolescents (≥12 years) and pediatric patients (0-11 years), respectively. All patients had severe hemophilia A (FVIII ≤1%) without inhibitors and had already been exposed to FVIII for ≥150 days (adults/adolescents) or ≥50 days (children).9,10 Patients completing the pivotal trials or phase I PK trials could continue into the open-label extension trial (guardian 2).13 Patients received turoctocog alfa as prophylaxis (20-50 IU/kg every second day or 20-60 IU/kg three times weekly, depending on age), as well as for treatment of bleeds. All three trials are now complete. Patients in the trials reported their dosing and bleeding data in diaries. Baseline joint disease at pivotal study haematologica | 2021; 106(7)

entry was not assessed; joint disease in the current analysis was therefore inferred based on treatment history.

Pharmacokinetic assessment in the pivotal studies Consistent with the International Society of Thrombosis and Haemostasis FVIII product guidelines,14 the pivotal trials included a single-dose PK assessment in a subset of patients after the first dose and in adults/adolescents after 3-6 months’ treatment; this involved nine (adults/adolescents) or six (children) sampling time points up to 48 h after injection of the product.9,10,12 In addition, samples were taken from all patients before and after dosing at the routine visits. All plasma samples were analyzed at a central laboratory using a one-stage clot assay.12,15 and post-dose FVIII:C values were entered into the PK model.

Population pharmacokinetic modeling Population PK modeling was conducted with the nonlinear, mixed-effects modeling software NONMEM (v.7.1.2, ICON Development Solutions, Ellicott City, MD, USA). Variability between patients was quantified within the population PK model12 by including between-patient variability on PK parameters using a log-normal distribution with no correlation. Allometric scaling based on body weight was applied to clearance (CL) and volume of distribution (Vd) parameters, and age was estimated as a linear covariate on CL. A combined proportional and additive residual error model was used.

Exposure-response analysis: estimating the association between factor VIII activity and annualized bleeding rate The exposure-response analysis of the association between predicted FVIII:C and spontaneous bleeds, including joint bleeds, was conducted with the statistical software R3.2.3 (Comprehensive R Archive Network [CRAN] project, University of Münster, Germany). The population PK model estimates were applied to diaryrecorded dosing from all patients receiving prophylaxis with turoctocog alfa in the pivotal and extension trials who had appropriate records, regardless of whether patients reported a spontaneous bleed or not, to produce a predicted FVIII:C time course for each patient over the entire span of prophylaxis for that patient. These predicted profiles had peaks following each recorded dose and subsequent gradual decline, taking into account the contributions from the last three doses at any time point (Figure 1). Each patient’s time on prophylaxis was divided into five clinically meaningful categories of predicted FVIII:C: 0-1%, >1-5%, >5-15%, >15-50%, and >50%. Subsequently, the diary-reported bleeding (spontaneous, including joint bleeds) data were compared to the predicted FVIII:C–time-profiles to estimate the FVIII:C–bleeding pattern relationships. Patients were excluded from the analysis if they switched from prophylaxis to on-demand treatment during the extension trial, received only on-demand treatment throughout the extension trial, had missing diary returns, or failed to document time of bleeding for >50% of recorded bleeds. For the purposes of this analysis, we have described bleeds as those bleeds that patients reported in their diary returns. This is to acknowledge that the bleeds were subjectively evaluated by the patient. For each patient, exposure time to turoctocog alfa, mean FVIII:C, and number of spontaneous bleeds (including joint bleeds) requiring treatment (based on the patient’s diary entries and excluding re-bleeds within 48 h) were calculated for each predicted FVIII:C category. To assess the association between FVIII:C and ABR, bleeding data were used to estimate ABR for 1903

A. Tiede et al.

each of the five predicted FVIII:C categories for each patient, using negative binomial regression and predictions of a parametric model. Relationships between ABR and mean FVIII:C were evaluated for the overall dataset (pivotal and extension trials), and by trial phase (pivotal vs. extension) and patients’ age (adults/adolescents [≥12 years] vs. children [0-11 years]). The proportion of time in which patients were predicted to achieve a FVIII:C level was calculated by taking the patient-years of exposure (PYE) for a FVIII:C category range, dividing it by the overall PYE, and multiplying it by 100%. The risk of bleeding episodes and FVIII:C levels was also analyzed using survival analysis. For each patient and trial phase, the time on a given factor level to the first bleed was calculated. The time to first bleed is presented in Kaplan-Meier plots displaying the proportion of patients without a bleed at a given time. For each trial phase, the difference between the FVIII:C levels was tested using a log-rank test for statistical significance.

Results Patients and exposure to turoctocog alfa The demographics of the PK subgroups in the pivotal trials (n=50) and of the patients participating in these trials and from whom post-dose FVIII:C measurements were used in the analysis are shown in Table 1. The analysis population, which was limited to patients receiving prophylaxis, comprised 232 patients across the pivotal and extension trials. One patient was excluded from the analysis as over 50% of bleeds recorded by this patient lacked a time recording, leaving 231 patients (63 children, 168 adults). Total PYE for adults/adolescents and children in the pivotal trials were 77 and 23, respectively, and 498 and 217, respectively, in the extension trial.

Population pharmacokinetic modeling The total pool of PK data in the population PK model comprised the PK profiles from the PK subgroups in the pivotal trials supplemented with post-dose measurements from routine clinical visits for patients participating in these trials (n=231). The PK of turoctocog alfa was found to align well with a one-compartmental model (i.e., the elimination of turoctocog alfa after a single intravenous bolus dose was approximately log-linear) as described by the equation:

where D is dosage, t is the time since dose, V is the volume of distribution, CL is clearance, k is the elimination rate constant, and e is the base of the natural logarithm (~2.72). One-compartmental behavior is indicative of the compound being distributed mainly within the bloodstream itself. The population PK parameter estimates are shown in Table 2.12 Therefore, the one-compartmental model was selected as the basis for the population PK analysis, using turoctocog alfa data derived from the one-stage clot assay for all patients.12 The volume of distribution for an individual of weight W, V(W), was calculated according to the equation below, where Wref represents the body weight of the reference individual (70 kg):

Clearance for an individual with a weight W and age A (CL(W, A)), was calculated according to the formula below, where Aref represents the age of the reference individual age (20 years): 1904

Predicted factor VIII activity - bleeding relationship Overall, there were 1,237 spontaneous bleeds (324 during the pivotal phase and 913 during the extension phase), of which 1,063 occurred in adults/adolescents and 174 in children (Table 3). The vast majority (n=1,055; 85.29%) of spontaneous bleeds were joint bleeds; more adults/adolescents than children suffered joint bleeds. The proportion of time at FVIII:C levels >1% and the PYE that were used to calculate this parameter across patient populations and trial phases are shown in Table 3. It was predicted that overall FVIII:C levels ≥1% were achieved for 85.64% of the time. Mean spontaneous bleeds, including joint bleeds, decreased as FVIII:C increased (Tables 4 and 5), indicating an exposure–bleeding relationship. This relationship was evident in both the pivotal and extension trial phases for both types of bleed. However, for each FVIII:C activity category, lower ABR were observed in the extension phase than in the pivotal phase. A FVIII:C–bleeding relationship was also apparent for the two age groups of adults/adolescents and children, although children had a lower ABR than adults/adolescents within each FVIII:C activity category (Table 5). At low FVIII:C, ABR were lower during the extension phase than in the pivotal phase for both adults/adolescents and children (Figure 2). The difference in ABR between adults/adolescents and children was particularly evident for the pivotal phase; during the extension phase, the ABR for each age group was similar for spontaneous bleeds, with slightly lower ABR in children compared with adults/adolescents. Similar relationships between FVIII:C and ABR were evident for joint bleeds (Online Supplementary Figure S1). Using Kaplan-Meier plots, analyzed over a fixed period of time (60 days), the time to first spontaneous bleeding was longer during times at higher FVIII:C levels (Figure 3). A similar pattern was apparent with joint bleeds (Online Supplementary Figure S2). Hence, patients were more protected during times at higher FVIII:C levels, particularly during the extension phase. In general, patients remained bleed-free for longer at any level of FVIII:C in the extension phase, particularly at the lower FVIII:C levels. Log-rank calculations showed a significant difference between each FVIII:C category in the proportion of patients without a bleed during the pivotal phase (P<0.0001) and the extension phase (P<0.0001). .

Table 1. Demographics of patients included in the exposure–response analysis.

Parameters Number of patients Age on entering program, years; mean (SD) Body weight, kg; mean (SD)

Adults/adolescents PK All* subgroup

Children PK All* subgroup

22 24.00 (7.88)

168 28.98 (12.15)

28 5.96 (2.76)

63 6.08 (2.91)

71.84 (12.44)

73.5 (18.13)

24.43 (10.50)

24.6 (10.03)

*Adults/adolescents and children from the pivotal trials or patients who completed the phase I pharmacokinetic trials could enter the extension trial. Patients were excluded from the exposure–response analysis if they switched from prophylaxis to on-demand treatment during the extension trial, received only on-demand treatment throughout the extension trial, had missing diary returns, or failed to document time of bleeding for >50% of recorded bleeds. PK: pharmacokinetic; SD: standard deviation.

haematologica | 2021; 106(7)

FVIII activity and bleeding risk in hemophilia A

Figure 1. Patients’ time spent in different FVIII:C activity ranges predicted from dosing diaries. The first week of prophylaxis is illustrated for two representative patients on turoctocog alfa dosed three times weekly. Predicted profiles of FVIII:C activity were calculated from the reported patients’ diary information on doses and timing of doses. Both patients exhibited a dosing interval pattern of 2-2-3 days in their first week, although the last two doses were delayed and higher for patient 2 than for patient 1. Triangles represent the area under the curve (AUC) of FVIII:C contributing to the mean FVIII:C value for each range. The horizontal span of each triangle defines the time spent in the relevant FVIII:C range. Both the AUC and time spans were calculated across dosing intervals and patient years. Actual trough FVIII:C was measured inconsistently and locally, and was not generally used to guide treatment decisions.

Discussion We used a population PK model and diary-recorded dosing data to estimate FVIII levels in patients with severe hemophilia A undergoing prophylaxis with turoctocog alfa and assessed the relationship between FVIII:C and ABR using diary-recorded data on bleeds. This approach confirmed an association between higher FVIII activity levels and reduced risk of spontaneous bleeding, apparent across all ages and trial phases. Bleeding rates were lower in the extension phase than in the pivotal phase at the same FVIII:C levels, consistent with previous spontaneous ABR data from these studies.13 The FVIII exposure–bleeding relationship was also apparent for both adults/adolescents and children. This effect, however, disappeared during the extension phase; after a period of 6 months of prophylaxis during the pivotal studies, the bleeding risk in adult patients approached that of the pediatric population. This suggests that patients’ characteristics at baseline accounted for the difference in bleeding risk between the age groups. The use of a large number of data on outcomes and their potential modifiers, combined from patients in different age groups and trial phases, is a strength of our study. It is reasonable to assume that the observed relationship between FVIII:C and ABR, as well as the role of modifying factors, is largely independent of the factor concentrate used in the source studies. The main conclusions conveyed by our work are therefore for prophylaxis with any factor concentrate in severe hemophilia A. Although our analysis showed that ABR decreased with higher FVIII:C, it was not possible to identify a single FVIII:C level that would prevent bleeding in all patients. For example, an ABR of <2 was achieved when FVIII:C was >15% in the pivotal phase but had already been achieved when FVIII was >1% in the extension phase. We would speculate that, over time, the patients’ clinical condition (including joint disease activity) improved and the bleeding risk decreased. In addition, the effect of previous on-demand treatment on ABR would have diminished over time and may have contributed to the lower bleeding haematologica | 2021; 106(7)

Table 2. Population pharmacokinetic parameters.

Parameter

Population parameter estimate (CV)

CL70 kg, 20 y (mL/h)a V70 kg (L)b Allometric scaling exponent for body weight on clearance (eCL) Allometric scaling exponent for body weight on V (eV) Age effect on CL (1/year)

302 (0.32) 3.46 (0.22) 0.95 0.86 -0.01

a Reference value for a 70-kg, 20-year-old patient. bReference value for a 70-kg patient. CL: clearance; CV: inter-individual variability; V: volume of distribution.

rates in the extension phase. Over a third of adult/adolescent patients received only on-demand treatment prior to entering the pivotal trial,9 potentially affecting joint disease and bleeding risk at study entry, even at higher FVIII:C levels. In children (who commonly start prophylactic regimens early on), an ABR <2 was achieved at any FVIII:C level ≥1%, consistent with the notion that pristine joints have a lower risk of bleeding during prophylaxis. The lower bleeding rates in adults in the extension phase compared with the pivotal phase could therefore indicate a “calming-down” effect of prophylaxis on joint disease activity. These results support the notion that treatment outcomes (and study endpoints) such as ABR would not only be influenced by FVIII exposure, but also by patient-related factors such as age, activity level, or joint status. Different hemophilia products, including extended halflife factor concentrates, can transform FVIII exposure, but direct comparison of outcomes across studies may be confounded by such patient-related factors. Similar confounders may have to be considered when comparing results of non-factor replacement therapy or even gene therapy to results achieved with FVIII replacement. Our findings on the FVIII exposure-response relationship support and build on data from other modeling analyses using negative binomial distribution in patients with hemophilia A. Collins et al.6 established the importance of 1905

A. Tiede et al.

FVIII trough levels and protection from bleeding in patients with severe hemophilia A. Valentino et al.16 linked higher peak FVIII levels, higher area under the curve and time with FVIII >20 IU/dL (>20%) with increased protection against joint and non-joint bleeding. den Uijl et al.7

suggested that joint bleeds decreased to approximately zero at FVIII levels >12% in patients with mild to severe hemophilia A. As such, it would be expected that FVIII trough levels of 15% would prevent most bleeds, considering that patients with a FVIII trough of 15% are likely to

Table 3. Time spent at each FVIII:C range and bleeding characteristics after turoctocog alfa prophylaxis.

Parameters

Pivotal Adults/adolescents Children

PYE Total numbers of spontaneous bleeds Total numbers of spontaneous joint bleeds Proportion of time (%) in each FVIII activity range 0-1% >1-5% >5-15% >15-50% >50%

Extension Adults/adolescents Children

Overall

77 287 251

23 37 24

498 776 684

217 137 96

815 1237 1055

15.58 27.27 25.97 28.57 3.90

21.74 26.09 21.74 21.74 8.70

12.85 26.10 25.50 29.32 6.22

16.60 26.73 20.73 23.04 12.90

14.36 26.38 24.17 27.48 7.85

FVIII: factor VIII; FVIII:C: factor VIII activity; PYE: patient-years of exposure.

Table 4. FVIII:C ranges and annualized bleeding rate of spontaneous bleeds and spontaneous joint bleeds.

FVIII activity range

Patient years of exposure

Total number Total number of spontaneous of spontaneous bleeds joint bleeds

0-1%

116.6

303

241

>1-5%

214.5

396

335

>5-15%

197.1

371

337

>15-50%

223.7

154

133

>50%

63.6

Overall mean spontaneous ABR (negative binomial estimate, 95% CI)

Overall mean spontaneous joint ABR (negative binomial estimate, 95% CI)

4.16 (3.23-5.40) 2.65 (2.12-3.34) 2.14 (1.69-2.74) 0.76 (0.58-0.99) 0.21 (0.11-0.36)

3.44 (2.56-4.66) 2.28 (1.77-2.96) 1.99 (1.53-2.60) 0.67 (0.50-0.90) 0.15 (0.07-0.31)

ABR: annualized bleeding rate; 95% CI: 95% confidence interval; FVIII: factor VIII; FVIII:C: FVIII activity.

Table 5. FVIII:C ranges and annualized bleeding rate of spontaneous bleeds by subgroup.

FVIII activity range

0-1% >1-5% >5-15% >15-50% >50%

Mean spontaneous ABR (negative binomial estimate, 95% CI) Analyses population By trial phase By age group Pivotal Extension Adults/ Children adolescents 7.16 (5.15-10.02) 4.15 (3.02-5.76) 2.95 (2.06-4.25) 0.99 (0.60-1.61) 0.47 (0.08-1.44)

2.15 (1.58-2.97) 1.69 (1.30-2.23) 1.64 (1.23-2.23) 0.66 (0.49-0.89) 0.19 (0.09-0.34)

4.86 (3.64-6.52) 3.29 (2.61-4.17) 2.64 (2.04-3.44) 0.92 (0.69-1.22) 0.35 (0.19-0.59)

2.68 (1.61-4.60) 0.95 (0.49-1.90) 0.71 (0.40-1.27) 0.32 (0.16-0.64) 0.03 (0.00-0.15)

Mean spontaneous joint ABR (negative binomial estimate, 95% CI) Analyses population By trial phase By age group Pivotal Extension Adults/ Children adolescents 5.94 (4.01-8.94) 3.55 (2.46-5.22) 2.76 (1.87-4.10) 0.92 (0.55-1.52) 0.23 (0.01-1.02)

1.75 (1.23-2.56) 1.44 (1.07-1.98) 1.49 (1.09-2.08) 0.56 (0.40-0.79) 0.14 (0.06-0.31)

4.17 (3.02-5.83) 2.88 (2.22-3.78) 2.48 (1.89-3.28) 0.82 (0.60-1.13) 0.26 (0.12-0.50)

1.83 (0.95-3.73) 0.67 (0.32-1.49) 0.58 (0.27-1.29) 0.23 (0.10-0.55) 0.00 (0.00-0.00)

ABR: annualized bleeding rate; 95% CI: 95% confidence interval; FVIII: factor VIII; FVIII:C: FVIII activity.

1906

haematologica | 2021; 106(7)

FVIII activity and bleeding risk in hemophilia A

have a higher FVIII:C for the majority of time between doses. When we describe FVIII:C at 15% in our analysis, this reflects a momentary FVIII:C value between peaks and troughs, rather than a trough value (Figure 1). As such, bleeding risk will not be the same in a patient at a momentary FVIII:C level of 15%, if the trough in this patient is much lower, compared with a patient with a trough level of 15%, who would have much higher FVIII:C for most of the time. In addition to reporting negative binomial distribution estimates of ABR, we used Kaplan-Meier analysis and logrank statistics to assess the bleeding risk at different FVIII levels. This analysis provides a slightly different perspective on the data by depicting the proportion of bleed-free patients over a given period of observation, confirming the impact of FVIII levels on the risk of bleeding. It should be noted, however, that there was no plateau reached during the analysis period of 2 months in our study, and it can be expected that the proportion of bleed-free patients would decrease further over time. The proportion of bleed-free patients depicted in Figure 3 should not, therefore, be compared directly to proportions of bleed-free patients reported from other studies over different periods of time. Our analysis also has some potential sources of error such as: the misinterpretation of symptoms of pain as a

bleed; delayed recognition of bleeds resulting in a shift of the start of a bleed to some time after the next injection; and inaccuracy of injection records. Assuming these potential errors decreased during the guardian trials, this may provide some explanation for the decline in ABR over time. On the other hand, we cannot exclude underreporting of bleeds by patients during the extension phase, particularly when visits changed from every 2 months during the pivotal phase, to every 6 months during the extension phase. PK data from patients treated with FVIII products can be described using one- or two-compartmental PK models.17 Two-compartmental models may not be identifiable from PK data with less intensive sampling, such as data from children.17 Furthermore, one-compartmental models are less sensitive than two-compartmental models to the handling of observations below “limits of quantification”. This is because a smaller number of parameters are to be estimated from the given pool of data. While there is no well-established consensus regarding handling in population PK modeling,18 these choices can have profound effects for FVIII PK modeling.19 Overall, our data suggest that trough level targets need to be adapted according to a patient’s age and previous treatment and should be revised regularly over time. We hypothesize that higher levels may be required in patients

Figure 2. Estimates of annualized bleeding rates for spontaneous bleeds for five FVIII:C categories in the two trial phases and age groups. (A-D) Negative binomial estimates of annualized bleeding rate (ABR) for five FVIII:C categories (0-1%, >1-5%, >5-20%, >20-50%, and >50%) for spontaneous bleeds during the pivotal (A) and extension (B) phases for adults/adolescents, and during the pivotal (C) and extension (D) phases for children. FVIII:C: factor VIII activity.

haematologica | 2021; 106(7)

1907

A. Tiede et al.

Figure 3. Kaplan-Meier estimates of the proportion of patients without a spontaneous bleed according to predicted FVIII:C during the pivotal and extension trial phases. FVIII:C: FVIII factor VIII activity.

with active joint disease (particularly target joints and synovitis), and the intensity of prophylaxis may be reduced in those without active joint disease who have not bled for some time. Further prospective studies are needed to evaluate this approach. Bleeds are currently used as the main endpoint for studies in hemophilia, both for newly developed factor concentrates, as well as for non-factor replacement products. Characteristic features of novel factor concentrates include, but are not limited to, longer half-life and improved preservation of post-translational modification. It could be tempting to compare ABR across studies of these products, but it should be kept in mind that ABR is a multifactorial outcome. Our analysis illustrates that ABR changes over time and depends on patients’ characteristics. In conclusion, in patients of all ages with severe hemophilia A who were treated with turoctocog alfa, ABR for spontaneous bleeds, including joint bleeds, decreased as FVIII:C increased, indicating an exposure-response relationship. The data presented in the current study suggest that patient-related factors and treatment history influence the FVIII:C level needed to protect patients from bleeding. ABR reported in clinical trials not only reflect efficacy of the FVIII product, but also the characteristics of the population under study.7,20 The choice of prophylactic regimens to target certain trough levels should take into account the patient’s age, joint disease activity, and other determinants of bleeding risk. Disclosures AT has received research support, honoraria, or consultation fees from Alnylam, Bayer, Biogen Idec, Biotest, Bristol-Myers Squibb, Boehringer Ingelheim, CSL Behring, Leo Pharma, Novo Nordisk, Octapharma, Pfizer, Roche, Shire, and SOBI. VJ-Y has received reimbursement for attending symposia/congresses and/or honoraria for speaking and/or for consulting, and/or funds for research from Takeda, Bayer, CSL-Behring, Grifols, Novo Nordisk, Sobi, Roche, Octapharma and Pfizer. RK has received 1908

support, honoraria, or consultation fees from Bayer, Biogen Idec, Biotest, Bristol-Myers Squibb, Boehringer Ingelheim, CSL Behring, Grifols, Leo Pharma, Novo Nordisk, Octapharma, Pfizer, Roche, Shire, and SOBI. TS has received consultation and speaker fees from Bayer, Bioverativ, Chugai, CSL Behring, IL Japan, JB Pharma, Kaketsuken, Merck Sharp & Dohme, Nihon-Pharma, Novo Nordisk, Pfizer, Sekisui Medical, and Shire. AG is an employee of Novo Nordisk A/S. ES has attended and received funds for Advisory Boards from Bayer, Grifols, Kedrion, Novo Nordisk, Octapharma, Pfizer, Roche, Shire, and SOBI. She attended Speaker Bureaus for, and received funds from, Bayer, Bioverativ, CSL Behring, Grifols, Kedrion, Novo Nordisk, Octapharma, Pfizer, Roche, Shire, and SOBI. FAK and SL have no conflicts of interest to disclose. Contributions AG, representing Novo Nordisk A/S, provided the statistical analyses of the data; AT contributed to the study design and interpreted the results; all authors critically wrote or revised the intellectual content of the manuscript, reviewed and/or commented on each draft, and approved the final version for submission. Acknowledgments These trials were sponsored by Novo Nordisk A/S (Bagsværd, Denmark). The authors thank the patients with severe hemophilia A and their families/caregivers, as well as the investigators, pharmacists, nurses, and trial staff at each center for participating in these trials. The authors would also like to thank Brigitte Brand-Staufer (Novo Nordisk Health Care AG) and Lars Korsholm for their scientific advice and critical review of the manuscript. Medical writing support was provided by Jo Fetterman, PhD (Parexel, UK). Funding This work was funded by Novo Nordisk A/S (Bagsværd, Denmark). Novo Nordisk’s policy on data sharing may be found at https://novonordisk-ctts.app-trialscope.com/how-access-clinical-trial-datasets. haematologica | 2021; 106(7)

FVIII activity and bleeding risk in hemophilia A

References 1. Biggs R, MacFarlane RG. Haemophilia and related conditions: a survey of 187 cases. Br J Haematol. 1958;4(1):1-27. nd 2. White GC 2 , Rosendaal F, Aledort LM, Lusher JM, Rothschild C, Ingerslev J. Definitions in hemophilia. Recommendation of the scientific subcommittee on factor VIII and factor IX of the scientific and standardization committee of the International Society on Thrombosis and Haemostasis. Thromb Haemost. 2001; 85(3):560. 3. Nilsson IM, Hedner U, Ahlberg A. Haemophilia prophylaxis in Sweden. Acta Paediatr Scand. 1976;65(2):129-135. 4. Manco-Johnson MJ, Abshire TC, Shapiro AD, et al. Prophylaxis versus episodic treatment to prevent joint disease in boys with severe hemophilia. N Engl J Med. 2007; 357(6):535-544. 5. Srivastava A, Brewer AK, MauserBunschoten EP, et al. Guidelines for the management of hemophilia. Haemophilia. 2013;19(1):e1-47. 6. Collins PW, Blanchette VS, Fischer K, et al. Break-through bleeding in relation to predicted factor VIII levels in patients receiving prophylactic treatment for severe hemophilia A. J Thromb Haemost. 2009; 7(3):413-420. 7. den Uijl IE, Mauser Bunschoten EP, Roosendaal G, et al. Clinical severity of haemophilia A: does the classification of the 1950s still stand? Haemophilia. 2011; 17(6):849-853. 8. den Uijl IE, Fischer K, Van Der Bom JG, Grobbee DE, Rosendaal FR, Plug I. Analysis of low frequency bleeding data: the associ-

haematologica | 2021; 106(7)

ation of joint bleeds according to baseline FVIII activity levels. Haemophilia. 2011; 17(1):41-44. 9. Lentz SR, Misgav M, Ozelo M, et al. Results from a large multinational clinical trial (guardianTM 1) using prophylactic treatment with turoctocog alfa in adolescent and adult patients with severe haemophilia A: safety and efficacy. Haemophilia. 2013;19(5):691-697. 10. Kulkarni R, Karim FA, Glamocanin S, et al. Results from a large multinational clinical TM trial (guardian 3) using prophylactic treatment with turoctocog alfa in paediatric patients with severe haemophilia A: safety, efficacy and pharmacokinetics. Haemophilia. 2013;19(5):698-705. 11. Lentz SR, Janic D, Kavakli K, et al. Longterm safety and efficacy of turoctocog alfa in prophylaxis and treatment of bleeding episodes in severe haemophilia A: final results from the guardian 2 extension trial. Haemophilia. 2018;24(6):e391-e394. 12. Jimenez-Yuste V, Lejniece S, Klamroth R, et al. The pharmacokinetics of a B-domain truncated recombinant factor VIII, turoctocog alfa (NovoEight®), in patients with hemophilia A. J Thromb Haemost. 2015; 13(3):370-379. 13. Ozelo M, Misgav M, Abdul KF, et al. Longterm patterns of safety and efficacy of bleeding prophylaxis with turoctocog alfa ® (NovoEight ) in previously treated patients with severe haemophilia A: interim results TM of the guardian 2 extension trial. Haemophilia. 2015;21(5):e436-e439. 14. Lee M, Morfini M, Schulman S, and the Factor VIII/Factor IX Scientific and Standardization Committee of the International Society for Thrombosis and Haemostasis. Scientific and Standardization

Committee Communication. The design and analysis of pharmacokinetic studies of coagulation factors. https://www.isth.org/. Last update: 21 March 2001. Accessed in September, 2017. 15. Bolton-Maggs PH, Perry DJ, Chalmers EA, et al. The rare coagulation disorders-review with guidelines for management from the United Kingdom Haemophilia Centre Doctors' Organisation. Haemophilia. 2004;10(5):593-628. 16. Valentino LA, Pipe SW, Collins PW, et al. Association of peak factor VIII levels and area under the curve with bleeding in patients with haemophilia A on every third day pharmacokinetic-guided prophylaxis. Haemophilia. 2016;22(4):514-520. 17. Bjorkman S, Oh M, Spotts G, et al. Population pharmacokinetics of recombinant factor VIII: the relationships of pharmacokinetics to age and body weight. Blood. 2012;119(2):612-618. 18. Ahn JE, Karlsson MO, Dunne A, Ludden TM. Likelihood based approaches to handling data below the quantification limit using NONMEM VI. J Pharmacokinet Pharmacodyn. 2008;35(4):401-421. 19. Garmann D, McLeay S, Shah A, Vis P, Maas Enriquez M, Ploeger BA. Population pharmacokinetic characterization of BAY 818973, a full-length recombinant factor VIII: lessons learned - importance of including samples with factor VIII levels below the quantitation limit. Haemophilia. 2017; 23(4):528-537. 20. Fischer K, Chowdary P, Collins P, et al. Modelling FVIII levels for predictions of zero spontaneous-joint bleeding in a cohort of severe hemophilia A subjects with target joints initiated on tertiary prophylaxis. Blood. 2016;128(22):2576.

1909

ARTICLE Ferrata Storti Foundation

Myeloproliferative Disorders

Crucial role of hematopoietic JAK2 V617F in the development of aortic aneurysms

Tetsuro Yokokawa,1,2 Tomofumi Misaka,1,3 Yusuke Kimishima,1 Kento Wada,1 Keiji Minakawa,4 Koichi Sugimoto,1,2 Takafumi Ishida,1 Soji Morishita,5 Norio Komatsu,6 Kazuhiko Ikeda4 and Yasuchika Takeishi,1

Department of Cardiovascular Medicine, Fukushima Medical University, Fukushima; Department of Pulmonary Hypertension, Fukushima Medical University, Fukushima; 3 Department of Advanced Cardiac Therapeutics, Fukushima Medical University, Fukushima; 4Department of Blood Transfusion and Transplantation Immunology, Fukushima Medical University, Fukushima; 5Department of Transfusion Medicine and Stem Cell Regulation, Juntendo University Graduate School of Medicine, Tokyo and 6 Department of Hematology, Juntendo University Graduate School of Medicine, Tokyo, Japan. 1 2

Haematologica 2021 Volume 106(7):1910-1922

ABSTRACT

Correspondence: TOMOFUMI MISAKA misaka83@fmu.ac.jp KAZUHIKO IKEDA kazu-ike@fmu.ac.jp Received: June 24, 2020. Accepted: December 11, 2020. Pre-published: February 11, 2021. https://doi.org/10.3324/haematol.2020.264085

1910

AK2 V617F is the most frequent driver mutation in myeloproliferative neoplasms (MPN) and is associated with vascular complications. However, the impact of hematopoietic JAK2 V617F on aortic aneurysms (AA) remains unknown. Our cross-sectional study indicated that nine (23%) of 39 MPN patients with JAK2 V617F exhibited the presence of AA. In order to clarify whether the hematopoietic JAK2 V617F contributes to the AA, we applied bone marrow transplantation (BMT) with the donor cells from Jak2 V617F transgenic (JAK2V617F) mice or control wild-type (WT) mice into lethally irradiated apolipoprotein E-deficient mice. Five weeks after BMT, the JAK2V617F-BMT mice and WT-BMT mice were subjected to continuous angiotensin II infusion to induce AA formation. Four weeks after angiotensin II infusion, the abdominal aorta diameter in the JAK2V617F-BMT mice was significantly enlarged compared to that in the WT-BMT mice. Additionally, the abdominal AA-free survival rate was significantly lower in the JAK2V617F-BMT mice. Hematopoietic JAK2 V617F accelerated aortic elastic lamina degradation as well as activation of matrix metalloproteinase (MMP)-2 and MMP-9 in the abdominal aorta. The numbers of infiltrated macrophages were significantly upregulated in the abdominal aorta of the JAK2V617F-BMT mice accompanied by STAT3 phosphorylation. The accumulation of BM-derived hematopoietic cells carrying JAK2 V617F in the abdominal aorta was confirmed by use of the reporter green fluorescent proteintransgene. BM-derived macrophages carrying JAK2 V617F showed increases in mRNA expression levels of Mmp2, Mmp9, and Mmp13. Ruxolitinib decreased the abdominal aorta diameter and the incidence of abdominal AA in the JAK2V617F-BMT mice. Our findings provide a novel feature of vascular complications of AA in MPN with JAK2 V617F.

Introduction Myeloproliferative neoplasms (MPN) including polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF) are characterized by chronic proliferation of mature myeloid cells and extramedullary hematopoiesis. The prevalence of MPN is reported to be 0.4 to 2.8 per 100,000 persons and the incident rates are 3.1 to 10.9 per one million person-years, which are increasing with age.1,2 Major causes of morbidity and mortality in MPN patients are represented by vascular complications, progression to myelofibrosis, and transformation to acute leukemia.3 To date, vascular disorders in MPN are known as arterial and venous thrombosis and advanced atherosclerosis.3-5 Among the MPN, JAK2 V617F is the most frequent driver mutation, which is observed in over 95% of PV patients as well as in 50–60% of ET and PMF patients.6 Several murine studies haematologica | 2021; 106(7)

Hematopoietic JAK 2V617F in aortic aneurysms

have shown that expression of Jak2 V617F confers cytokine-independent growth of the myeloid cells, resulting in a variety of MPN phenotypes, like PV, ET, or PMF depending on the animal model.7,8 By contrast, the etiological link between hematopoietic JAK2 V617F and vascular disorders has not fully been elucidated in vivo. Aortic aneurysms (AA) are a life-threatening aortic disease, characterized by dilatation of the aorta.9 There are increasing numbers of patients with AA, with a prevalence in western countries of 1–5% in the adult population aged over 65 years old.10,11 Although small sized AA remain mostly asymptomatic, AA are slowly progressive, and large AA often lead to aortic rupture and/or sudden death when their diameters are >6.0 cm in thoracic AA (TAA) and >5.0 cm in abdominal AA (AAA). It has been reported that the incidence of rupture of AA is 21.3 per 100,000 people in the general population with a mortality of 85-90%.12 Thus, it is important to stratify the risk of AA rupture, and to discover a novel mechanism for the formation and progression of AA. It is known that circulating inflammatory cells play important roles in the development of AA, accompanied by the secretion of various inflammatory factors such as cytokines and chemokines.13,14 In this term, inflammatory AAA which characterizes an unusually thickened aneurysm wall and dense adhesions of adjacent intra-abdominal structures represent different features from atherosclerotic AAA.9 Although bone marrow (BM) progenitor cells contribute to the tissue inflammation and pathogenesis of cardiovascular diseases such as atherosclerotic diseases,15,16 the causal role of BM-derived cells on the onset and progression of AA has not yet been fully explored. Moreover, the prevalence of AA has not been studied in MPN patients with JAK2 V617F, who are more prone to vascular disorders than both the general population and MPN patients with driver mutations other than JAK2 V617F.17 In the present study, we hypothesized that hematopoietic JAK2 V617F might be causally associated with the development of AA. Strikingly, we demonstrated that the hematopoietic cells carrying JAK2 V617F promoted AAA progression in mice, as well as a common incidence of AA in MPN patients with JAK2 V617F.

mice with a C57BL/6J background were purchased from Japan SLC (Shizuoka, Japan). The JAK2V617F mice were crossed with the CAG-EGFP mice to generate JAK2V617F/CAG-EGFP double transgenic mice (JAK2V617F-GFP).21 WT littermates were used as controls (WT-GFP). All animal studies were approved by the Fukushima Medical University Animal Research Committee (approval ID, 2019078).

Bone marrow transplantation ApoE−/− mice or WT mice aged between 8 and 12 weeks (body weight range, 20-30 g) were lethally irradiated (9.0 Gy) 24 hours before bone marrow transplantation (BMT).19 Whole BM cells were harvested from the femurs and tibiae of the mice. BM cells (5.0×106) were injected into the ApoE−/− recipient mice or the WT recipient mice via the tail vein.

Angiotensin II-induced aortic aneurysm model Five weeks after BMT, the ApoE−/− recipient mice or the WT recipient mice were implanted with an osmotic pump (ALZET micro-osmotic pump MODEL 1004, DURECT Co., Cupertino, CA, USA)22 and angiotensin II (Ang II) (1900 ng/kg per min) or saline was continuously infused for 4 weeks.23,24

Futher methods The details of blood sampling, chimeric analysis, ultrasound imaging and determination of aortic aneurysms, measurements of blood pressure and heart rate, histological analysis, measurement of total cholesterol and triglyceride, gelatin zymography, preparation for BM-derived macrophages, western blot analysis and reverse transcription-quantitative polymerase chain reaction are described in the Online Supplementary Appendix.

Treatment with a JAK1/2 inhibitor For in vitro experiments, ruxolitinib (Novartis Pharmaceuticals, Basel, Switzerland) was used at a concentration of 250 nM for 24 hours prior to the RNA extraction. Dimethylsulfoxide was used as a control. For the in vivo study, ruxolitinib dissolved in 0.5% methylcellulose was administered to the mice orally at 60 mg/kg twice daily for 4 weeks.25,26 We used 0.5% methylcellulose as a vehicle control.

Statistical analysis Data are expressed as mean ± standard error of the mean (SEM). A value of P<0.05 was considered statistically significant.

Methods Detailed methods are provided in the Online Supplementary Appendix.

Results Prevalence of aortic aneurysms in patients with myeloproliferative neoplasms

Patients We enrolled 39 patients with JAK2 V617F-positive MPN whose computed tomography data were available in a series of our former studies.18,19 We evaluated the maximal diameter of the ascending aorta and abdominal aorta via computed tomography imaging. The protocol of the human studies was approved by the Institutional Ethics Committee of the Fukushima Medical University Hospital (approval ID, 29348 and 1242).

Animals We used male Jak2 V617F-expressing transgenic (JAK2V617F) mice.7,19 Wild-type (WT) littermates were used as controls. Male apolipoprotein E-deficient (ApoE−/−) mice on a C57BL/6J background were obtained from the Jackson Laboratory (JAX stock number, 002052, Bar Harbor, ME, USA).20 CAG-EGFP reporter haematologica | 2021; 106(7)

In order to investigate the incidence of AA in association with the presence of JAK2 V617F in the hematopoietic cells, we conducted a cross-sectional study that enrolled 39 MPN patients with the JAK2 V617F mutation (mean age, 68 ± 12 years; female, 54%) in a series of our former studies18,19 (Table 1). The allele burden of JAK2 V617F was 51±30%. Computed tomography revealed the presence of AA in nine (23%) of the 39 JAK2 V617F-positive MPN patients,9,27 including two patients (22%) who underwent graft replacement surgery of the aorta due to progression of AA during the MPN course. When we divided the patients into two groups based on the presence of prior histories of thrombotic events as a crucial risk factor in the conventional prognostic system regarding the incidence of arterial and venous thrombotic 1911

T. Yokokawa et al.

Table 1. Characteristics and prevalence of aortic aneurysms in JAK2 V617F-positive myeloproliferative neoplasm patients.

Age, years Female, n. (%) MPN diagnosis Polycythemia vera, n. (%) Essential thrombocythemia, n. (%) Primary myelofibrosis, n. (%) Laboratory data White blood cell count, ×109/L Red blood cell count, ×1012/L Hemoglobin concentration, g/dL Platelet count, ×109/L JAK2 V617F allele burden, % Comorbidities Hypertension, n. (%) Diabetes mellitus, n. (%) Dyslipidemia, n. (%) Smoking history, n. (%) Thrombosis history Arterial thrombosis, n. (%) Venous thrombosis, n. (%) Cytoreductive treatment Hydroxyurea, n. (%) Ruxolitinib, n. (%) Presence of aortic aneurysms Thoracic aortic aneurysm, n. (%) Abdominal aortic aneurysm, n. (%)

All patients (n = 39)

Thrombosis (+) (n = 10)

Thrombosis (-) (n = 29)

68 ± 12 21 (54)

73 ± 9 6 (60)

66 ± 13 15 (52)

0.118 0.468

16 (41) 12 (31) 11 (28)

2 (20) 5 (50) 3 (30)

14 (48) 7 (24) 8 (28)

0.141 0.130 0.591

13.5 ± 9.4 4.7 ± 1.5 14.0 ± 3.9 509 ± 427 51 ± 30

10.6 ± 5.5 4.6 ± 1.3 12.7 ± 3.0 500 ± 270 43 ± 33

14.5 ± 10.4 4.7 ± 1.6 14.5 ± 4.0 512 ± 473 53 ± 29

0.270 0.889 0.213 0.939 0.332

17 (44) 6 (15) 7 (18) 13 (33)

5 (50) 1 (10) 5 (50) 4 (40)

12 (41) 5 (17) 2 (7) 9 (31)

0.456 0.510 0.007 0.440

7 (18) 3 (8)

7 (70) 3 (30)

0 (0) 0 (0)

<0.001 0.013

17 (44) 1 (3) 9 (23) 7 (18) 2 (5)

1 (10) 0 (0) 5 (50) 4 (40) 1 (10)

16 (55) 1 (3) 4 (14) 3 (10) 1 (3)

0.014 0.744 0.032 0.057 0.452

events in PV and ET,28,29 the prevalence of AA was significantly higher in the JAK 2V617F-positive MPN patients with a history of thrombosis than in those without (Table 1 and Online Supplementary Table S1). Next, we examined the expression levels of genes known to be related to AA, such as MMP and cytokines,30 in circulating leukocytes. The MMP9, which is independently associated with aortic wall thickness and aortic luminal diameter during the process of AA,31 as well as TGFB3 and in interleukin 8 (IL8) were significantly increased in JAK2 V617F-positive MPN patients compared to the age- and sex-matched control subjects (Online Supplementary Figure S1 and Online Supplementary Table S2). These findings suggest that incidences of AA are not rare in JAK2 V617F-positive MPN patients, and that the presence of JAK2 V617F in leukocytes may lead to increases in the expression levels of genes involved in the pathogenesis of AA.

Hematopoietic JAK2 V617F promotes abdominal aortic aneurysms formation in angiotensin II-infused ApoE−/− mice In order to clarify whether or not hematopoietic Jak2 V617F could indeed play a mechanistic role in the development of AA, we applied the BMT strategy using the donor BM cells from the JAK2V617F mice7 into the lethally irradiated ApoE−/− mice24 (JAK2V617F-BMT mice) (Figure 1A). The irradiated ApoE−/− recipient mice transplanted with donor BM cells from WT littermates (WT-BMT mice) were used as controls. Five weeks after BMT, the donor cells were similarly engrafted as shown in the chimeric analysis of the peripheral blood cells between the WT-BMT mice and JAK2V617F-BMT mice (Figure 1B). 1912

The JAK2V617F-BMT mice displayed significantly higher white blood cell counts and spleen weights compared to the WT-BMT mice, indicating a hematological feature of MPN-like phenotype in the JAK2V617F-BMT mice (Figure 1C and Online Supplementary Figure S2). Next, the WTBMT mice and JAK2V617F-BMT mice were continuously infused with Ang II with 1900 ng/kg/min for 4 weeks, as a higher dose of Ang II is required to induce the AA formation when the irradiated ApoE−/− recipient mice are used.24 At 2 and 4 weeks following Ang II infusion, systolic, mean, and diastolic blood pressures were all significantly increased compared to those of the saline groups, but there were no differences in these parameters between the WT-BMT mice and JAK2V617F-BMT mice (Figure 1D; Online Supplementary Figure S3A and S3B). No significant difference was observed regarding the heart rate between the WT-BMT mice and JAK2V617F-BMT mice after Ang II infusion (Online Supplementary Figure S3C). Total cholesterol and triglyceride concentrations were decreased in the JAK2V617F-BMT mice compared to WTBMT mice (Online Supplementary Table S3), in agreement with the findings that MPN patients with JAK2 V617F often display low lipid profiles that correlate with advanced disease.32,33 Chimerism was not significantly different between the WT-BMT mice and JAK2V617F-BMT mice 4 weeks after saline or Ang II infusion (Online Supplementary Figure S3D). We monitored both the thoracic aorta and suprarenal abdominal aorta by ultrasonography. The presence of TAA and AAA was defined as an increase by ≥50% in aortic diameter in comparison to the baseline. We did not observe any significant differences in the thoracic aorta diameter or the incidence of haematologica | 2021; 106(7)

Hematopoietic JAK 2V617F in aortic aneurysms

Figure 1. Hematopoietic JAK2 V617F contributes to abdominal aortic aneurysm formation. (A) Schematic diagram of the experimental design of bone marrow (BM) transplantation (BMT). BM cells from control wild-type (WT) mice or JAK2V617F mice were injected into the lethally irradiated ApoE−/− recipient mice. Five weeks after BMT, the ApoE−/− recipient mice transplanted with WT BM cells (WT-BMT mice) or JAK2V617F BM cells (JAK2V617F-BMT mice) were subjected to saline or angiotensin II (Ang II, 1900 ng/kg per min) infusion for 4 weeks. (B) Chimerism of donor cells in the peripheral leukocytes (n=11–13) and (C) blood cell counts (n=9–11) in the WT-BMT mice and JAK2V617F-BMT mice at 5 weeks after BMT. *P<0.05 versus the WT-BMT mice by the unpaired Student’s t-test. (D) Systolic blood pressure (BP) at the baseline (n=17 each), 2 weeks (n =7–10) and 4 weeks (n=7–8) after saline or Ang II infusion. (E) Representative ultrasound images of abdominal aorta under saline or Ang II infusion for 4 weeks. Scale bars, 1.0 mm. (F) Abdominal aorta diameter at the baseline (n= 33–34), 2 weeks (n=7–24) and 4 weeks (n=7–24) after saline or Ang II infusion. (G) Abdominal aortic aneurysm (AAA)-free survival rate in the WT-BMT mice and JAK2V617F-BMT mice after saline or Ang II infusion (n=7–26) by log-rank test. All data are presented as mean ± standard error of the mean. *P<0.05 vs. the corresponding saline-infused mice and †P<0.05 vs. the corresponding WT-BMT mice by one-way ANOVA with Tukey post-hoc analysis. WT, ApoE−/−: mice transplanted with bone marrow cells from wild-type mice; JAK2V617F, ApoE−/−: mice transplanted with bone marrow cells from JAK2V617F mice; Ang II: angiotensin II; WBC: white blood cell count; RBC: red blood cell count; Hb: hemoglobin concentration; Plt: platelet count.

haematologica | 2021; 106(7)

1913

T. Yokokawa et al.

TAA between the WT-BMT mice and JAK2V617F-BMT mice after Ang II infusion (Online Supplementary Figure S4). In contrast, although abdominal aorta diameters in both WT-BMT mice and JAK2V617F-BMT mice were significantly increased at 2 and 4 weeks after Ang II infusion compared to the corresponding saline-infused groups, the abdominal aorta diameter in the JAK2V617F-BMT mice was significantly enlarged compared to that of WT-BMT mice at 4 weeks after Ang II infusion (Figure 1E and 1F). The incidence of AAA was 0% in both the saline-infused WT-BMT mice and JAK2V617F-BMT mice, while it was 18.5% of the WT-BMT mice (5 of 27) and 46.1% of the JAK2V617F-BMT mice (12 of 26) developed AAA after Ang II infusion. Accordingly, Kaplan-Meier analysis demonstrated that the AAA-free survival rate was significantly lower in the JAK2V617F-BMT mice than in WT-BMT mice (Figure 1G). No death was observed in either salineinfused WT-BMT mice or JAK2V617F-BMT mice, while 3.7% of the Ang II-infused WT-BMT mice (1 of 27) and 7.6% of the Ang II-infused JAK2V617F-BMT mice (2 of 26) died from AAA rupture during the study period. Although continuous Ang II infusion significantly increased heart weight in both the WT-BMT and the JAK2V617F-BMT mice, there was no significant difference between the two groups (Online Supplementary Figure S2B). When we used mice with a WT background as recipients, as opposed to the ApoE-/- recipient mice, the WT recipients transplanted with JAK2V617F BM cells did not exhibit AA formation (Online Supplementary Figure S5). These findings indicate that JAK2 V617F-mediated AAA development is required for an ApoE-deficient background experimentally, which is consistent with the previous work that the incidence of AAA in WT mice was much lower than in hyperlipidemic mice.34 Taken together, these data suggest that hematopoietic JAK2 V617F promoted the formation of AAA in response to continuous Ang II infusion independently of the levels of elevated blood pressure.

Hematopoietic JAK2 V617F induces aortic elastic lamina degradation and rupture accompanied by activation of matrix metalloproteinases In the morphometric analysis of aortas, a local enlargement of the abdominal aorta with a blood clot was observed in the Ang II-infused JAK2V617F-BMT mice (Figure 2A). Hematoxylin-eosin staining indicated thickening of the arterial wall as well as inflammatory cell infiltration accompanied by thrombus formation in the JAK2V617F-BMT mice after Ang II infusion (Figure 2B). Elastica-Masson staining revealed that discontinuity and breakage of elastin fibers in the aortic wall were detected in the Ang II-infused JAK2V617F-BMT mice (Figure 2B). The degree of elastin disruption in the abdominal aortic medial layers was more severe in the JAK2V617F-BMT mice than in WT-BMT mice after Ang II infusion based on the quantitative analysis by grading scores35 (Figure 2C). Activation of MMP, particularly MMP-2 and MMP-9, is required to produce AAA.36 Gelatin zymography assay using homogenates from the abdominal aorta demonstrated that either MMP-2 activity or pro MMP-9 expression levels were not different between saline-infused WT-BMT and JAK2V617F-BMT mice while both MMP-2 activity and pro MMP-9 expression were significantly elevated in the JAK2V617F-BMT mice compared to WT-BMT mice after Ang II infusion (Figure 2D and 2E). These data suggest that hematopoietic JAK2 1914

V617F contributes to elastin disruption accompanied by activation of MMP, resulting in enlargement of the abdominal aorta and AAA rupture.

Inflammatory responses are accelerated in the abdominal aortas in JAK2V617F-bone marrow transplantation mice with activation of STAT3 We next examined inflammatory responses in the AAA formation and progression.37 Immunohistochemical analysis showed that there were significant increases in CD45+ leukocytes, CD68+ macrophages, Ly6B.2+ neutrophils, and TER119+ erythrocytes in the abdominal arterial walls in the JAK2V617F-BMT mice compared to WTBMT mice after Ang II infusion (Figure 3A and 3B; Online Supplementary Figure S6). The phosphorylation levels of STAT3 in the abdominal aorta were not significantly different between saline-infused WT-BMT mice and JAK2V617F-BMT mice, but the levels in the JAK2V617F-BMT mice were higher than those in the WT-BMT mice in response to Ang II (Figure 4A). For the assessment of the inflammatory mediators related to AAA development, mRNA expression levels of Ccl6, an important macrophage chemokine, as well as Tgfb1, which is involved in collagen and elastin production, were increased in the abdominal aorta of the JAK2V617F-BMT mice in comparison to the levels in the WT-BMT mice after Ang II infusion. These results suggest that macrophages and neutrophils carrying JAK2 V617F induce inflammatory responses in abdominal aortic walls in response to Ang II.

Characterization of bone marrow-derived hematopoietic cells carrying JAK2 V617F in the abdominal aorta by use of GFP-transgene In order to further characterize the specific contribution of BM-derived circulating cells in the AAA formation in the Ang II-infused JAK2V617F-BMT mice, we generated double transgenic mice by crossing the JAK2V617F mice with CAG-EGFP mice (JAK2V617F-GFP).21 Then, we transplanted BM cells derived from JAK2V617F-GFP mice or control WT-GFP mice into lethally irradiated ApoE−/− mice. After BMT followed by Ang II infusion for 4 weeks, immunostaining showed that more GFP+ cells accumulated in the aorta of the JAK2V617F-GFP-BMT mice compared to WT-GFP BMT mice (Figure 5A). In the JAK2V617FGFP-BMT mice, most GFP+ cells expressed CD45, CD68, or Ly6B.2 in the abdominal aorta (Figure 5B). These data indicate that the accumulated inflammatory cells carrying JAK2 V617F are mobilized from BM, migrated, and engrafted into the aortic walls in response to Ang II stimulation.

Impact of bone marrow-derived JAK2V617F macrophages on the genesis of matrix metalloproteinase We next cultured mononuclear cells isolated from the BM in the presence of granulocyte-macrophage colony-stimulating factor to elucidate the effect of JAK2V617F macrophages on MMP. The BM cells were mostly differentiated into macrophages with proportions of CD68+ cells of over 80% both in the WT and JAK2V617F BM-derived cells (Figure 6A). The mRNA expression levels of Mmp2 and Mmp9 were significantly increased in the JAK2V617F BM-derived macrophages compared to the WT BM-derived macrophages (Figure 6B). Additionally, the expression levels of Mmp13, an activator of MMP-9,38 in the JAK2V617F BMhaematologica | 2021; 106(7)

Hematopoietic JAK 2V617F in aortic aneurysms

Figure 2. Hematopoietic JAK2 V617F induces elastin degradation and rupture of aortic aneurysm accompanied by matrix metalloproteinase activation. (A) Representative images of the aorta from the WT-BMT mice and JAK2V617F-BMT mice 4 weeks after saline or Ang II or infusion. Scale bars, 5 mm. The boxed area from the Ang II-infused JAK2V617F-BMT mice was highlighted at higher magnification. Scale bar, 2.5 mm. (B) Representative images of hematoxylin-eosin (HE)- and Elastica-Masson (EM)-stained sections. ‘T’ indicates thrombus in the aorta. Black arrow indicates the rupture site of the aorta. Scale bars, 50 mm. (C) Quantitative analysis of degradation grade in EM-stained sections (n=3–5). (D) A typical image of gelatin zymography using homogenates from abdominal aortas of the WT-BMT mice and JAK2V617F-BMT mice 4 weeks after saline or Ang II infusion. (E) Densitometric analysis of the gelatin zymography (n=5–6). The sum of matrix metalloproteinase (MMP)-2 and pro MMP-2 bands was evaluated as MMP-2 activity. All data are presented as mean ± standard error of the mean. *P<0.05 vs. the corresponding saline-infused mice and †P<0.05 vs. the Ang II-infused WT-BMT mice by one-way ANOVA with Tukey post-hoc analysis. WT, ApoE−/−: mice transplanted with bone marrow (BM) cells from wild-type (WT) mice; JAK2V617F, ApoE−/−: mice transplanted with BM cells from JAK2V617F mice; BMT: bonemarrow transplantation; Ang II: angiotensin II.

haematologica | 2021; 106(7)

1915

T. Yokokawa et al.

derived macrophages were higher than those in the WT BM-derived macrophages (Figure 6B). Ruxolitinib, a selective JAK1/2 inhibitor, significantly decreased both Mmp2 and Mmp9 expression levels in the JAK2V617F macrophages (Figure 6C). These data suggest that BM-derived macrophages carrying JAK2 V617F play important roles in the genesis of MMP.

Inhibition of JAK1/2 with ruxolitinib prevents hematopoietic JAK2 V617F-induced abdominal aortic aneurysm formation We investigated whether inhibition of JAK1/2 could attenuate hematopoietic JAK2 V617F-induced AAA formation (Figure 7A). Ruxolitinib treatment significantly decreased white blood cell counts and spleen weights in

Figure 3. Hematopoietic JAK2 V617F increases inflammatory cell infiltration in the abdominal aorta. (A) Representative immunohistochemical images of the aorta stained by anti-CD45, CD68, and Ly6B.2 antibodies in the WT-BMT mice and JAK2V617F-BMT mice 4 weeks after saline or angiotensin II infusion. Arrows indicate Ly6B.2-positive cells. Scale bars, 50 mm. (B) Quantitative analyses of the CD45+ (n=5–6), CD68+ (n=3–5), and Ly6B.2+ (n= 3–5) cells in the aortic sections. All data are presented as mean ± standard error of the mean. *P<0.05 vs. the corresponding saline-infused mice and †P <0.05 vs. the Ang II-infused WT-BMT mice by oneway ANOVA with Tukey post-hoc analysis. WT, ApoE−/−: mice transplanted with bone marrow (BM) cells from wild-type (WT) mice; JAK2V617F, ApoE−/−: mice transplanted with BM cells from JAK2V617F mice; BMT: BM transplantation; Ang II: angiotensin II.

1916

haematologica | 2021; 106(7)

Hematopoietic JAK 2V617F in aortic aneurysms

the Ang II-infused JAK2V617F-BMT mice (Figure 7B; Online Supplementary Figure 7A), but did not affect the levels of blood pressures or heart rate (Figure 7C; Online Supplementary Figure S7B to D). Total cholesterol and triglyceride concentrations were increased in the ruxolitinib-treated JAK2V617F-BMT mice compared to the vehicle-treated JAK2V617F-BMT mice, as previously reported in MPN patients (Online Supplementary Table S5). Importantly, abdominal aorta diameters were significantly decreased in the ruxolitinib-treated JAK2V617F-BMT mice compared to the vehicle-treated JAK2V617F-BMT mice at both 2 and 4 weeks after Ang II infusion (Figure 7D and 7E). The incidence of AAA was 64% of the vehicle-treated JAK2V617F-BMT mice (7 of 11) while 18% of the ruxolitinib-treated JAK2V617F-BMT mice (2 of 11). Accordingly, the Kaplan-Meier analysis demonstrated that the AAA-free survival rate was significantly higher in the ruxolitinib-treated JAK2V617F-BMT mice than in the vehicle-treated JAK2V617F-BMT mice (Figure 7F). Ruxolitinib decreased MMP-2 activity and pro MMP-9 expression in the abdominal aorta in Ang II-infused JAK2V617F-BMT mice (Figure 7G). In contrast, thoracic aorta diameters or incidence of TAA were not different between the ruxolitinib- and vehicle-treated JAK2V617FBMT mice (Online Supplementary Figure S7E and S7F). Taken together, these findings strongly suggest that hematopoietic JAK2 V617F plays a causal role in the onset and development of AAA.

Discussion The present study is the first to demonstrate that hematopoietic cells carrying JAK2 V617F are causally associated with AAA formation in mice. In response to Ang II stimulation, the infiltrated CD68+ macrophages with JAK2 V617F in the arterial walls may have contributed to the genesis and activation of matrix metalloproteinase, resulting in increases of elastin degradation in the abdominal aorta and the development of AAA. Our findings provide a novel feature of vascular complication of AA in MPN patients due to the constitutive activation of the hematopoietic JAK-STAT pathway. To date, vascular complications of MPN patients have been reported mainly in the arterial and venous thrombotic events from clinical and murine studies.3,39 History of thrombosis and age >60 years are the highest risk factors for thrombosis in MPN patients.40 Interestingly, JAK2 V617F-positive MPN patients with a history of thrombosis more frequently showed AA, suggesting that history of thrombosis may be a risk factor for AA as well as thrombosis, whereas age was not different between JAK2 V617F-positive MPN patients with AA and those without AA. Leukocytosis is another factor that increases the risks of thrombosis in MPN patients whereas the roles of increased number or functional alteration of platelets have been controversial in cases of thrombosis in MPN,3,41 indicating that abnormal MPN-clone-derived leukocytes

Figure 4. Hematopoietic JAK2 V617F increases STAT3 phosphorylation and cytokine expression in response to angiotensin II infusion in the abdominal aorta. (A) Western blot analysis on the STAT3 in the abdominal aorta. Aorta extracts from the WT-BMT mice or JAK2V617F-BMT mice were immunoblotted with the indicated antibodies. Phosphorylated STAT3 (p-STAT3) to total STAT3 (t-STAT3) ratios are shown in the bar graphs. The average value for the saline-infused WT-BMT mice was set to 1 (n=5–6). β actin was used as the loading control. (B) Relative mRNA expression levels of Ccl6 and Tgfb1 in the aorta. Actb was used for normalization. The average value for saline-infused WT-BMT mice was set to 1 (n=7–10). All data are presented as mean ± standard error of the mean. *P<0.05 vs. the corresponding saline-infused mice and †P<0.05 vs. the Ang II-infused WT-BMT mice by one-way ANOVA with Tukey post-hoc analysis. WT, ApoE−/−: mice transplanted with bone marrow (BM) cells from wild-type (WT) mice; JAK2V617F, ApoE−/−: mice transplanted with BM cells from JAK2V617F mice; BMT: BM transplantation; Ang II: angiotensin II.

haematologica | 2021; 106(7)

1917

T. Yokokawa et al. A

Figure 5. Characterization of bone marrow-derived JAK2 V617F hematopoietic cells in the abdominal aorta by use of GFP-transgene. (A) The lethally irradiated ApoE−/− mice were transplanted with bone marrow (BM) cells from the WT/CAG-EGFP (WT-GFP) mice or JAK2V617F/CAG-EGFP (JAK2V617F-GFP) double transgenic mice. The ApoE−/− recipient mice were subjected to saline or angiotensin II (Ang II) infusion for 4 weeks. The abdominal aortas were stained with an anti-GFP (green) antibody and DAPI (blue). Scale bars, 100 mm. (B) Representative immunofluorescence images of the aorta sections stained with anti-CD45 (red), anti-CD68 (red), or anti-Ly6B.2 (red) and anti-GFP (green) antibodies and DAPI (blue) in ApoE−/− mice transplanted with JAK2V617F-GFP BM cells. Scale bars, 50 mm. Ang II: angiotensin II; GFP: green fluorescent protein; DAPI: 4′,6-diamidino-2-phenylindole; WT: wild-type.

contribute to vascular diseases. In our cohort, 23% of the patients with JAK2 V617F-positive MPN exhibited the presence of AA. Given that the prevalence of AA is reported to be 1–5% in the general population above the age of 65 years,10,11 JAK2 V617F-positive MPN might be associated with a higher prevalence of AA. In combination with mouse studies, the presence of JAK2 V617F in leukocytes is likely to be a factor of AA development. The pathogenetic feature of AA is mediated through the degenerative process involving extracellular matrix remodeling in all layers of the arterial wall. Particularly, local activation of inflammatory responses by immune cells plays an important role in the process. Leukocyte accumulation in inflammatory aneurysms provides addi1918

tional evidence that inflammation represents variants along a spectrum of aneurysmal disease rather than distinct pathological entities of the inflammatory cells.14 It has been reported that plasma inflammatory cytokine levels in JAK2 V617F-positive MPN are significantly increased in the peripheral blood.42,43 Similarly, we showed that circulating JAK2 V617F leukocytes of MPN patients exhibited significant increases in the expression levels of the cytokines of TGFB3 as well as the chemokines of IL-8. These findings are supported by previous reports that inflammatory macrophages and neutrophils contributed to AAA development.44,45 It has been reported that macrophage NLRP3 inflammation activation in adventitia promotes inflammatory cytokine prohaematologica | 2021; 106(7)

Hematopoietic JAK 2V617F in aortic aneurysms

Figure 6. Matrix metalloproteinases are upregulated in bone marrow-derived macrophages with JAK2 V617F expression. (A) Mononuclear cells isolated from bonemarrow (BM) cells of wild-type (WT) mice or JAK2V617F mice were cultured in the presence of granulocyte-macrophage colony stimulating factor for 6 days. Representative immunofluorescence images of the cells stained with anti-CD68 (red) antibody and DAPI (blue). Scale bars, 100 mm. (B) Relative mRNA expression levels of Mmp2, Mmp9, and Mmp13 in the cultured mononuclear cells (n=7 each). (C) Mmp2 and Mmp9 mRNA expressions with and without ruxolitinib pretreatment at 250 nM for 24 hours in the JAK2V617F mononuclear cells (n=4 each). Dimethylsulfoxide (DMSO) was used for control. Actb was used for normalization. The average value for the cultured mononuclear cells from WT mice or DMSO group was set to 1. All data are presented as mean ± standard error of the mean. *P<0.05 vs. the WT group or DMSO group by the unpaired Student’s t-test. JAK2V617F: JAK2 V617F-expressing transgenic mice.

duction and MMP activation, and contributes to the development of AAA.46 Recruitments of macrophages and neutrophils, and expression of pro-inflammatory cytokines may characterize the degenerative process during AA formation. We observed that the JAK2V617F-BMT mice did not display significant enlargement of the ascending aorta, and therefore it remains to be elucidated whether TAA develops in this model and/or some other condition. It is known that matrix metalloproteases are closely involved in the pathogenesis of AA through elastin degradation.47 Studies have reported that the JAK-STAT signaling pathway is one of the regulators of matrix metalloproteases,48,49 and that macrophages are a significant source of matrix metalloproteinases.47,50 Both MMP-9 and MMP-2 are required and work in concert to produce AAA.36 We have demonstrated here that BM-derived macrophages with JAK2 V617F expression significantly upregulate Mmp2, Mmp9, and Mmp13. Likewise, substantially promoted macrophage infiltration was accompanied by increases in MMP-2 activity and pro MMP-9 expression in the abdominal aorta in the Ang II-stimulated JAK2V617F-BMT mice, suggesting that infiltration of JAK2 V617F-expressing macrophages with activation of matrix metalloproteinases plays causal roles in the pathohaematologica | 2021; 106(7)

genesis of AAA. In agreement, we found significantly higher mRNA levels of MMP9 in the peripheral leukocytes, containing macrophage-progenitor monocytes, of patients with JAK2 V617F-positive MPN compared to control subjects. MMP-9 plasma levels were independently associated with aortic wall thickness and aortic luminal diameter, but not with parameters of atherosclerosis.31 Further studies are needed to clarify the MMP-related mechanism for progression to AA in MPN patients. Currently, ruxolitinib is a clinically available JAK1/2 inhibitor for the treatment of patients with PMF and PV.51 Ruxolitinib has been shown to improve long-term survival in patients with PMF and reduced thrombotic events in patients with PV.52-54 Given our observations in the present study, ruxolitinib may be useful for preventing the onset and development of AAA in patients with MPN. It has recently been reported that clonal expansion of hematopoietic cells with somatic mutations, termed as clonal hematopoiesis, is common in the seemingly healthy older general population, and significantly associated with atherosclerotic cardiovascular diseases.55,56 Although JAK2 V617F has been identified as one of the recurrent somatic mutations in individuals with clonal hematopoiesis, it remains unclear whether clonal hematopoiesis is involved in AA in addition to stenotic 1919

T. Yokokawa et al.

atherosclerosis. However, given that our data from murine studies and clinical observations strongly indicate that hematopoietic JAK2 V617F plays crucial roles in the development of AA, we propose that JAK2 V617F-positive clonal hematopoiesis may be also involved in the

formation and progression of AA regardless of the hematological phenotypes of MPN.57 The limitation of the present study is the small number of patients with MPN included while thrombosis and non-thrombosis categories were also small subsets for

Figure 7. Ruxolitinib attenuates the formation of abdominal aortic aneurysms induced by hematopoietic JAK2 V617F. (A) Schematic diagram of the experimental design. The ApoE-/- recipients transplanted with JAK2V617F bone marrow cells (JAK2V617F-BMT) were administered with vehicle (0.5% methylcellulose) or ruxolitinib was orally at 60 mg/kg twice daily for 4 weeks along with continuous angiotensin II ( Ang II, 1900 ng/kg per min) infusion. (B) Blood cell counts at 4 weeks (n=6 each) and (C) systolic blood pressure (BP) at baseline (n=9–10), 2 weeks (n=9–10), and 4 weeks (n=9–10) in the vehicle- or ruxolitinib-treated JAK2V617F-BMT mice. (D) Abdominal aorta diameter at the baseline (n=11 each), 2 weeks (n=11 each), and 4 weeks (n=11 each). (E) Representative images from the aorta. Scale bars, 5 mm. (F) Abdominal aortic aneurysm (AAA)-free survival rate in the vehicle- or ruxolitinib-treated JAK2V617F-BMT mice (n=11 each) by log-rank test. (G) A typical image of gelatin zymography using homogenates from abdominal aortas. All data are presented as mean ± standard error of the mean. *P<0.05 vs. the vehicle group by the unpaired Student’s t-test. WBC: white blood cell count; RBC: red blood cell count; Hb: hemoglobin concentration; Plt: platelet count; BP: blood pressure; MMP: matrix metalloproteinase.

1920

haematologica | 2021; 106(7)

Hematopoietic JAK 2V617F in aortic aneurysms

comparison. Further studies are, thus, needed to validate our findings in larger cohorts. In conclusion, AA are a possible vascular complication of MPN and hematopoietic JAK2 V617F may causally be involved in the development of AA through macrophage accumulation and MMP activation. The present study also provides novel mechanisms underlying the pathogenesis of AA. Hematopoietic JAK-STAT signaling may be a potential therapeutic target for the development of AA. Contributions TY and TM designed the research, performed experiments, analyzed the results, and wrote the manuscript; YK, KW, and KM performed the experiments and analyzed the results; KS and TI interpreted the results and supervised the study; SM and NK analyzed the results and interpreted the results, and supervised the study; KI designed and supervised the research, analyzed the data, and wrote the manuscript; YT designed and supervised the research, and approved the final version of the manuscript.

References 1. Kutti J and Ridell B. Epidemiology of the myeloproliferative disorders: essential thrombocythaemia, polycythaemia vera and idiopathic myelofibrosis. Pathol Biol (Paris). 2001;49(2):164-166. 2. Srour SA, Devesa SS, Morton LM, et al. Incidence and patient survival of myeloproliferative neoplasms and myelodysplastic/myeloproliferative neoplasms in the United States, 2001-12. Br J Haematol. 2016;174(3):382-396. 3. Barbui T, Finazzi G and Falanga A. Myeloproliferative neoplasms and thrombosis. Blood. 2013;122(13):2176-2184. 4. De Stefano V, Ghirardi A, Masciulli A, et al. Arterial thrombosis in Philadelphia-negative myeloproliferative neoplasms predicts second cancer: a case-control study. Blood. 2020;135(5):381-386. 5. Hultcrantz M, Björkholm M, Dickman PW, et al. Risk for arterial and venous thrombosis in patients with myeloproliferative neoplasms: a population-based cohort study. Ann Intern Med. 2018;168(5):317-325. 6. Steensma DP, Dewald GW, Lasho TL, et al. The JAK2 V617F activating tyrosine kinase mutation is an infrequent event in both "atypical" myeloproliferative disorders and myelodysplastic syndromes. Blood. 2005;106(4):1207-1209. 7. Shide K, Shimoda HK, Kumano T, et al. Development of ET, primary myelofibrosis and PV in mice expressing JAK2 V617F. Leukemia. 2008;22(1):87-95. 8. Shide K. The role of driver mutations in myeloproliferative neoplasms: insights from mouse models. Int J Hematol. 2020;111(2):206-216. 9. Wanhainen A, Verzini F, Van Herzeele I, et al. Editor's choice - European Society for Vascular Surgery (ESVS) 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms. Eur J Vasc Endovasc Surg. 2019;57(1):8-93. 10. Wanhainen A, Hultgren R, Linne A, et al. Outcome of the swedish nationwide abdominal aortic aneurysm screening program. Circulation. 2016;134(16):1141-1148. 11. Oliver-Williams C, Sweeting MJ, Turton G, et al. Lessons learned about prevalence and growth rates of abdominal aortic

haematologica | 2021; 106(7)

Disclosures TY and KS have received financial support from Janssen Pharmaceutical K.K., Japan; TM has received financial support from Fukuda Denshi Co., Ltd., Japan; Ruxolitinib was provided by Novartis Pharmaceuticals to KI; these companies are, however, not associated with the contents of this study; all other authors declare no conflicts of interest. Acknowledgments The authors thank Ms Tomiko Miura and Ms Shoko Sato, from the Department of Cardiovascular Medicine, Fukushima Medical University, Japan, and Ms Chisato Kubo, from the Office for Gender Equality Support, Fukushima Medical University, Japan, for their technical assistance, as well as Prof. Kazuya Shimoda and Dr. Kotaro Shide, from the Department of Gastroenterology and Hematology, University of Miyazaki, Japan, for providing us with JAK2V617F mice. Funding This work was supported by JSPS KAKENHI (grant number: JP19K17532) to TY.

aneurysms from a 25-year ultrasound population screening programme. Br J Surg. 2018;105(1):68-74. 12. Gunnarsson K, Wanhainen A, Björck M, Djavani-Gidlund K and Mani K. Nationwide study of ruptured abdominal aortic aneurysms during twenty years (1994-2013). Ann Surg. 2019 Aug 14. [Epub ahead of print] 13. Dale MA, Ruhlman MK and Baxter BT. Inflammatory cell phenotypes in AAAs: their role and potential as targets for therapy. Arterioscler Thromb Vasc Biol. 2015;35(8):1746-1755. 14. Shimizu K, Mitchell RN and Libby P. Inflammation and cellular immune responses in abdominal aortic aneurysms. Arterioscler Thromb Vasc Biol. 2006;26(5): 987-994. 15. Dutta P, Courties G, Wei Y, et al. Myocardial infarction accelerates atherosclerosis. Nature. 2012;487(7407):325-329. 16. Moazzami K, Lima BB, Hammadah M, et al. Association between change in circulating progenitor cells during exercise stress and risk of adverse cardiovascular events in patients with coronary artery disease. JAMA Cardiol. 2020;5(2):147-155. 17. Barbui T, Finazzi G, Carobbio A, et al. Development and validation of an International Prognostic Score of thrombosis in World Health Organization-essential thrombocythemia (IPSET-thrombosis). Blood. 2012;120(26):5128-5133. 18. Harada-Shirado K, Ikeda K, Ogawa K, et al. Dysregulation of the MIRLET7/HMGA2 axis with methylation of the CDKN2A promoter in myeloproliferative neoplasms. Br J Haematol. 2015;168(3):338-349. 19. Ueda K, Ikeda K, Ikezoe T, et al. Hmga2 collaborates with JAK2V617F in the development of myeloproliferative neoplasms. Blood Adv. 2017;1(15):1001-1015. 20. Piedrahita JA, Zhang SH, Hagaman JR, Oliver PM and Maeda N. Generation of mice carrying a mutant apolipoprotein E gene inactivated by gene targeting in embryonic stem cells. Proc Natl Acad Sci U S A. 1992;89(10):4471-4475. 21. Okabe M, Ikawa M, Kominami K, Nakanishi T and Nishimune Y. 'Green mice' as a source of ubiquitous green cells. FEBS Lett. 1997;407(3):313-319.

22. Misaka T, Suzuki S, Miyata M, et al. Deficiency of senescence marker protein 30 exacerbates angiotensin II-induced cardiac remodelling. Cardiovasc Res. 2013;99(3): 461-470. 23. Ortega R, Collado A, Selles F, et al. SGLT-2 (Sodium-Glucose Cotransporter 2) inhibition reduces ang II (Angiotensin II)-induced dissecting abdominal aortic aneurysm in ApoE (Apolipoprotein E) knockout mice. Arterioscler Thromb Vasc Biol. 2019;39(8): 1614-1628. 24. Patel J, Douglas G, Kerr AG, Hale AB and Channon KM. Effect of irradiation and bone marrow transplantation on angiotensin II-induced aortic inflammation in ApoE knockout mice. Atherosclerosis. 2018;276:74-82. 25. Quintás-Cardama A, Vaddi K, Liu P, et al. Preclinical characterization of the selective JAK1/2 inhibitor INCB018424: therapeutic implications for the treatment of myeloproliferative neoplasms. Blood. 2010;115(15): 3109-3117. 26. Lee S, Shah T, Yin C, et al. Ruxolitinib significantly enhances in vitro apoptosis in Hodgkin lymphoma and primary mediastinal B-cell lymphoma and survival in a lymphoma xenograft murine model. Oncotarget. 2018;9(11):9776-9788. 27. Hiratzka LF, Bakris GL, Beckman JA, et al. 2010 ACCF/AHA/AATS/ACR/ASA/SCA/ SCAI/SIR/STS/SVM guidelines for the diagnosis and management of patients with Thoracic Aortic Disease: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgery, American College of Radiology, American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of Thoracic Surgeons, and Society for Vascular Medicine. Circulation. 2010;121(13):e266-369. 28. Carobbio A, Thiele J, Passamonti F, et al. Risk factors for arterial and venous thrombosis in WHO-defined essential thrombocythemia: an international study of 891 patients. Blood. 2011;117(22):5857-5859. 29. Barbui T, Tefferi A, Vannucchi AM, et al.

1921

T. Yokokawa et al. Philadelphia chromosome-negative classical myeloproliferative neoplasms: revised management recommendations from European LeukemiaNet. Leukemia. 2018; 32(5):1057-1069. 30. Lindholt JS and Shi GP. Chronic inflammation, immune response, and infection in abdominal aortic aneurysms. Eur J Vasc Endovasc Surg. 2006;31(5):453-463. 31. Grodin JL, Powell-Wiley TM, Ayers CR, et al. Circulating levels of matrix metalloproteinase-9 and abdominal aortic pathology: from the Dallas Heart Study. Vasc Med. 2011;16(5):339-345. 32. Tefferi A, Nicolosi M, Penna D, et al. Development of a prognostically relevant cachexia index in primary myelofibrosis using serum albumin and cholesterol levels. Blood Adv. 2018;2(15):1980-1984. 33. Fujita H, Hamaki T, Handa N, Ohwada A, Tomiyama J and Nishimura S. Hypocholesterolemia in patients with polycythemia vera. J Clin Exp Hematop. 2012;52(2):85-89. 34. Deng GG, Martin-McNulty B, Sukovich DA, et al. Urokinase-type plasminogen activator plays a critical role in angiotensin II-induced abdominal aortic aneurysm. Circ Res. 2003;92(5):510-517. 35. Satoh K, Nigro P, Matoba T, et al. Cyclophilin A enhances vascular oxidative stress and the development of angiotensin II-induced aortic aneurysms. Nat Med. 2009;15(6):649-656. 36. Longo GM, Xiong W, Greiner TC, Zhao Y, Fiotti N and Baxter BT. Matrix metalloproteinases 2 and 9 work in concert to produce aortic aneurysms. J Clin Invest. 2002;110 (5):625-632. 37. Quintana RA and Taylor WR. Cellular mechanisms of aortic aneurysm formation. Circ Res. 2019;124(4):607-618. 38. Hernández Ríos M, Sorsa T, Obregón F, et al. Proteolytic roles of matrix metalloproteinase (MMP)-13 during progression of chronic periodontitis: initial evidence for MMP-13/MMP-9 activation cascade. J Clin Periodontol. 2009;36(12):1011-1017.

1922

39. Edelmann B, Gupta N, Schnoeder TM, et al. JAK2-V617F promotes venous thrombosis through β1/β2 integrin activation. J Clin Invest. 2018;128(10):4359-4371. 40. Barbui T, Barosi G, Birgegard G, et al. Philadelphia-negative classical myeloproliferative neoplasms: critical concepts and management recommendations from European LeukemiaNet. J Clin Oncol. 2011;29(6):761-770. 41. Campbell PJ, MacLean C, Beer PA, et al. Correlation of blood counts with vascular complications in essential thrombocythemia: analysis of the prospective PT1 cohort. Blood. 2012;120(7):1409-1411. 42. Wehrle J, Seeger TS, Schwemmers S, Pfeifer D, Bulashevska A and Pahl HL. Transcription factor nuclear factor erythroid-2 mediates expression of the cytokine interleukin 8, a known predictor of inferior outcome in patients with myeloproliferative neoplasms. Haematologica. 2013;98(7):1073-1080. 43. Dutta A, Hutchison RE and Mohi G. Hmga2 promotes the development of myelofibrosis in Jak2(V617F) knockin mice by enhancing TGF-β1 and Cxcl12 pathways. Blood. 2017;130(7):920-932. 44. Hadi T, Boytard L, Silvestro M, et al. Macrophage-derived netrin-1 promotes abdominal aortic aneurysm formation by activating MMP3 in vascular smooth muscle cells. Nat Commun. 2018;9:5022. 45. Eliason JL, Hannawa KK, Ailawadi G, et al. Neutrophil depletion inhibits experimental abdominal aortic aneurysm formation. Circulation. 2005;112(2):232-240. 46. Usui F, Shirasuna K, Kimura H, et al. Inflammasome activation by mitochondrial oxidative stress in macrophages leads to the development of angiotensin II-induced aortic aneurysm. Arterioscler Thromb Vasc Biol. 2015;35(1):127-136. 47. Wang Y, Ait-Oufella H, Herbin O, et al. TGF-beta activity protects against inflammatory aortic aneurysm progression and complications in angiotensin II-infused mice. J Clin Invest. 2010;120(2):422-432.

48. Li WQ, Dehnade F and Zafarullah M. Oncostatin M-induced matrix metalloproteinase and tissue inhibitor of metalloproteinase-3 genes expression in chondrocytes requires Janus kinase/STAT signaling pathway. J Immunol. 2001;166(1):3491-3498. 49. Ghosh A, Pechota A, Coleman D, Upchurch GR, Jr. and Eliason JL. Cigarette smoke-induced MMP2 and MMP9 secretion from aortic vascular smooth cells is mediated via the Jak/Stat pathway. Hum Pathol. 2015;46(2):284-294. 50. Kothari P, Pestana R, Mesraoua R, et al. IL6-mediated induction of matrix metalloproteinase-9 is modulated by JAK-dependent IL-10 expression in macrophages. J Immunol. 2014;192(1):349-357. 51. 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. 52. Vannucchi AM and Harrison CN. Emerging treatments for classical myeloproliferative neoplasms. Blood. 2017;129(6):693-703. 53. Harrison CN, Vannucchi AM, Kiladjian JJ, et al. Long-term findings from COMFORTII, a phase 3 study of ruxolitinib vs best available therapy for myelofibrosis. Leukemia. 2016;30(8):1701-1707. 54. Verstovsek S, Vannucchi AM, Griesshammer M, et al. Ruxolitinib versus best available therapy in patients with polycythemia vera: 80-week follow-up from the RESPONSE trial. Haematologica. 2016;101(7):821-829. 55. 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. 56. Mas-Peiro S, Hoffmann J, Fichtlscherer S, et al. Clonal haematopoiesis in patients with degenerative aortic valve stenosis undergoing transcatheter aortic valve implantation. Eur Heart J. 2020;41(8):933-939. 57. 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.

haematologica | 2021; 106(7)

ARTICLE

Non-Hodgkin Lymphoma

Impact on survival through consolidation radiotherapy for diffuse large B-cell lymphoma: a comprehensive meta-analysis

Ferrata Storti Foundation

Martin D. Berger,1* Sven Trelle,2* Annina E. Büchi,3 Sabrina Jegerlehner,3 Codruta Ionescu,4 Thierry Lamy de la Chapelle5 and Urban Novak1

Department of Medical Oncology, Inselspital, Bern University Hospital, University of Bern, Switzerland; 2CTU Bern, University of Bern, Switzerland; 3Department of General Internal Medicine, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland; 4Department of Radiation Oncology, Inselspital, Bern University Hospital, University of Bern, Switzerland and 5Hematology Department, Rennes University Hospital, INSERM Research Unit 1236, Rennes, France

Haematologica 2021 Volume 106(7):1923-1931

*MDB and ST contributed equally as co-first authors.

ABSTRACT

ituximab has improved response rates and overall survival in diffuse large B-cell lymphoma. Radiotherapy is an effective treatment modality for lymphomas, but there is uncertainty on its use as consolidation after chemo-immunotherapy mainly in advanced stages. We evaluated its efficacy with a comprehensive meta-analysis and a systematic search of Pubmed, Embase, Cochrane, and abstracts from the American Society of Clinical Oncology, American Society of Hematology, European Society for Medical Oncology and American Society of Radiation Oncology published from June 1966 and December 2018. We identified 11 trials that evaluated consolidation radiotherapy following chemotherapy in a randomized fashion in 4,584 patients. The primary endpoint of this meta-analysis was progression-free survival (PFS). As three of the 11 trials were retracted, this data is based on 2,414 patients. For the primary endpoint, PFS, we found a hazard ratio (HR) 0.77 (95% Confidence Interval [CI]: 0.51-1.17), pooled (tau2: 0.25; I2: 85%), and a HR 0.80 (95% CI: 0.53-1.21), pooled in a bivariate metaanalysis and for the secondary endpoint, overall survival, a HR 0.93 (range, 0.61-1.40), pooled (tau2: 0.25; I2: 74%) and a HR 0.86 (95% CI: 0.58-1.27) in a bivariate meta-analysis. The lack of benefit did not change over time (P=0.95 (tau2: 0.32; I2: 88%), and was also absent for PFS when stratifying for i) chemotherapy, ii) the use of rituximab, iii) age, iv) the dose of radiotherapy and v) application to patients in complete remission with bulky disease. None of the trials used a positron emission tomography-guided approach. This meta-analysis revealed no survival benefit when consolidation radiotherapy is given to unselected diffuse large Bcell lymphoma patients following chemotherapy. These results need to be considered in future trials in the positron emission tomography-computed tomography era.

Correspondence: URBAN NOVAK urban.novak@insel.ch Received: February 9, 2020. Accepted: June 12, 2020. Pre-published: June 18, 2020. https://doi.org/10.3324/haematol.2020.249680

Introduction Comprising 35% of all non-Hodgkin lymphomas (NHL), diffuse large B-cell lymphoma (DLBCL) is the most common aggressive lymphoma in adults. The current standard therapy rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) cures two-thirds of patients.1,2 Several attempts with a variety of approaches including the addition of new drugs have so far failed to improve these results.3,4 Radiotherapy is an effective treatment option for patients with aggressive lymphomas. It was initially used as a primary modality for various lymphomas and was later used as consolidation when anthracycline-containing regimens became available in the 1980s. Radiotherapy is now commonly used in localized disease.5 As such, consolidation radiotherapy is part of the first line treathaematologica | 2021; 106(7)

1923

M.D. Berger et al.

ment of DLBCL in the European Society for Medical Oncology (ESMO)6 and National Comprehensive Cancer Network (NCCN) guidelines (https://www.nccn.org/professionals/physician_gls/pdf/b-cell.pdf (last access: Nov 28, 2019; for details see Online Supplementary Table S1). However, significant conceptual issues on its current use outside a clinical trial remain. They include different definitions of bulky disease, the use in advanced stages, and the recent implementation of positron emission tomography-computed tomography (PET-CT) in the clinical management. Albeit not limited to consolidation radiotherapy in DLBCL, treatment recommendations are often built on experience, clinical judgment and guidelines, but ideally should be based on data, preferably from randomized trials. Here we present a comprehensive meta-analysis to assess the impact of radiotherapy in addition to and after first-line chemo-immunotherapy of DLBCL based on the best currently available data by randomized controlled

trials. With this large meta-analysis, we aim to provide the rational basis for a future randomized trial on the use of consolidation radiotherapy in DLBCL.

Methods Literature search We performed a comprehensive search in electronic databases (Pubmed, Embase, Cochrane) in any language between June 1966 and December 2018 for randomized controlled trials. As the data presented on meetings may differ from the peer-reviewed publications,7 a manual search was done of abstracts from ASCO, ASH, ESMO, and ASTRO proceedings between 2009 and 2018. We used the following search strategy: (radiation therap*[Title] OR radiotherapy*[Title] OR radio-therap*[Title]) AND (non-hodgkin*[Title] OR non Hodgkin*[Title] OR nonhodgkin[Title] OR no Hodgkin*[Title] OR nhl[Title]) OR (lymphoma*[Title]) AND

Figure 1. Study selection. Flow diagram according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement to illustrate the search and selection process. DLBCL: diffuse large B-cell lymphoma.

1924

haematologica | 2021; 106(7)

Meta-analysis of radiotherapy for DLBCL

(aggressive[Title] OR malignant[Title] OR advanced[Title] OR histiocytic[Title] OR diffuse[Title] OR undifferentiated[Title] OR mixed[Title] OR high grade[Title] OR centroblastic [Title] OR immunoblastic[Title]).

Inclusion criteria and trial selection Three investigators independently screened the studies. The flow diagram according to the PRISMA statement8,9 depicted in Figure 1 illustrates the search and selection process. We aimed at identifying randomized trials that had enrolled at least 50 adult patients (≥18 years of age) per arm with newly diagnosed DLBCL (or aggressive lymphomas) at any stage according to the Ann Arbor classification. Patients had to be treated with a CHOP based chemotherapy (+/- rituximab), and randomized to subsequent consolidation radiotherapy or no radiotherapy. The 50-patients cut-off was chosen to exclude therapeutic exploratory trials. Although the cut-off is arbitrary, it is safe to assume that no confirmatory trials were excluded given the high (progression-free) survival rates observed in this population. Patients with previously treated or relapsed DLBCL were excluded. The full text report of identified trials was independently checked by the three investigators. Disagreements regarding trial selection were discussed until consensus was found. Each report was scrutinized to eliminate duplicates and to ensure that it was published as an original article.

Outcome measures Progression-free survival (PFS) and overall survival (OS) were the outcomes of interest. PFS was considered as tumor progression i.e., growth of the tumor during treatment, relapse i.e., growth after previous shrinkage or stabilization, or death. For trials that did not report outcome data that fit this definition, we used data of an outcome that was as close as possible to this definition e.g., event-free survival.

Data extraction Data was extracted in duplicate and disagreements were resolved by consensus. We used the Cochrane ‘Risk of Bias’ approach to assess methodological quality of trials.10 We used the data from the original publications, from intention-to-treat analyses, and from randomized patients only, and for the longest follow-up available for a particular outcome. The hazard ratio (HR) was used as effect measure for both outcomes. If HR and a measure of precision (standard error, variance, or 95% Confidence Interval [CI]) was not available, we digitized Kaplan-Meier curves, reconstructed the underlying time-toevent data, and calculated (log) HR and standard errors using a Cox regression model. Details on the outcome data of the 11 trials are shown in Table 2.

Statistical analysis Outcome data were pooled with a random-effects model using restricted maximum likelihood. We also did bivariate meta-analysis considering both outcomes in one analysis. Correlation between OS and PFS was estimated from two of the identified trials.11,12 We performed random-effects meta-regression for PFS over time using the mid of enrolment period as an independent covariate. Stratified analyses to explore possible reasons for heterogeneity were also done using meta-regression. Analyses were done using Stata (StataCorp. 2017. Stata Statistical Software: Release 16. College Station, TX, USA). Taking into account criticisms of meta-analysis,8,9 the Online Supplementary Appendix provides additional details on the analysis methods used and all outcome data. The latter, used in the meta-analysis, is provided in Table 2. haematologica | 2021; 106(7)

Results After deduplication, our search strategy generated 3,181 references (Figure 1). With the aim to identify clinical trials that assessed the role of consolidation radiotherapy in a randomized manner as part of the first-line therapy, our search revealed 11 trials amenable for this metaanalysis (details are listed in Table 1). Three of the four trials published by Aviles13-16 on this topic were later retracted.14-16 As of September 2019, these retracted papers have together received a total of 39 citations. Their data are provided in the respective figures, but were excluded from the meta-analyses. One trial was stopped early when the benefit of rituximab became evident,12 or as a result of a planned interim analysis.17 Older trials included lymphomas classified by the Kiel classification18 or included DLBCL according to the Working Formulation.19 Six of the trials included patients with localized disease only, but five of the 11 trials included also advanced stages. With the exception of the GELA LNH 93-111 where doxorubicin, cyclophosphamide, vindesine, bleomycin and prednisone (ACVBP) instead of CHOP was given in the comparator arm or SWOG,20 where the non-irradiated patients received eight cycles of CHOP (instead of three), the same chemotherapy was given to the randomized patients. The current standard R-CHOP was used in four of the 11 trials.15-17,21,22 Only the recent Lysa/GOELAMS 02 03 trial21 used PET, although not for guided treatment. Radiotherapy was given to both localized stages 1 and 2, but also advanced disease, and either to all or only patients in complete remission or bulky disease. GOELAMS 02 0321 was a non-inferiority trial whereas all other trials used for this meta-analysis used a superiority design. Seven trials with a total of 2,488 patients contributed to the analysis of the primary endpoint PFS (Figure 2). Data were extracted from the original publications. The UNFOLDER trial was presented in part at the 12th International Congress on Malignant Lymphomas,17 and again at the American Society of Clinical Oncology (ASCO) 2018,22 albeit with different endpoints. The latter have been used for this meta-analysis. Data from Engelhard18 was not available for the PFS analysis (Table 2). For PFS, the pooled HR was 0.77 (95% CI: 0.51-1.17), and in the pooled bivariate meta-analysis HR was 0.80 (95% CI: 0.53-1.21) (Figure 2). For OS, eight trials with a total of 2,744 patients were included. The pooled HR was 0.93 (95% CI: 0.61-1.40) and 0.86 (95% CI: 0.58-1.27) in the bivariate meta-analysis (Figure 3). Between-trial heterogeneity was high for both outcomes (PFS, tau2: 0.25, I2: 85%; OS, tau2: 0.25, I2: 74%). The total of 4,584 patients included in this meta-analysis were recruited between 1983 and 2013. However, the lack of benefit of the combined treatment modality remained stable over time, and time alone cannot explain the observed heterogeneity in the meta-analysis (P-value for time trend =0.95; tau2: 0.32; I2: 88%; Figure 4). Given the significant heterogeneity (see also Table 1), we analyzed the data by using the following stratifications: i) the applied chemotherapy was similar in both arms, ii) whether rituximab was used, iii) the dose of radiotherapy, and iv) whether the radiotherapy dose was given only in complete morphologic remission. In addition, we stratified according to the following trial population characteristics: v) mean age of the treated patients, 1925

M.D. Berger et al.

vi) whether the majority had advanced stage, and vii) whether the majority had bulky disease. As shown in Figure 5, we failed to explain between-trial heterogeneity by stratifying on any of these subgroups.

Discussion We here provide a large and comprehensive metaanalysis with the best currently available data from randomized trials on consolidation radiotherapy in the firstline treatment of aggressive lymphomas. In summary, we find no evidence for a survival benefit of an unselected consolidation radiotherapy for these patients, but uncertainty remains high. Our analysis extends the data from both retrospective and uncontrolled series in favor2,23-25 or against26 the use of consolidation radiotherapy in the first line setting. Our state-of-the art and updated meta-analysis that takes into account general concerns on the reproducibility of metaanalysis8,9 and significantly corroborates a previous metaanalysis on a limited number of trials.27 It also goes beyond extrapolations from data on particular extranodal sites,28 the common use of consolidation radiotherapy for

limited clinical stages only,23 pretreatment with different chemotherapy,29 or to treat bulky disease only.2 Collectively, the latter data are the basis for the current recommendations on the combined treatment modality also for patients with advanced stages. They have created an unsatisfactory uncertainty and rely on experts’ opinions on the use of radiotherapy when facing an individual patient. However, DLBCL is a disease in which cure, but also treatment-related toxicities and economic factors have to be considered. Unfortunately, our meta-analysis cannot provide data on costs, safety and long term risks of secondary malignancies related to radiation therapy. Overall, the data that could be used for this meta-analysis is of mixed quality (Tables 1 and 2). As an extreme, three of the four randomized trials by the same group all clearly supporting the added value of radiotherapy have later been retracted, the last one in early 2019.14-16 We display their results in our figures as they might have influenced the use of consolidation radiotherapy in routine practice or clinical trials before their retraction. The results of the important UNFOLDER trial is still not fully published.17,22 The trials used for this meta-analysis also harbor considerable conceptual heterogeneity: radiotherapy was given to shorten chemotherapy and its toxicity,

Table 1. Summary on the randomized trials used for the meta-analysis. The number of patients in the respective column indicates the actual number of patients for the individual trials that received consolidation radiotherapy in a randomized fashion. The retracted trials are highlighted in grey. The superscript number in the study column refers to the number of the references in the manuscript.

Trial (with reference)

Diagnosis

Patients (#)

Aviles et al.13 Engelhard et al.18 ECOG 148419 Avileset al.14 SWOG 8736 20 GELA 93-111 GELA 93-412 Aviles et al.15 Aviles et al.16 UNFOLDER17,22 GOELAMS 02 0321

DLCL high grade NHL diffuse aggressive NHL DLCL intermediate & high grade NHL aggressive NHL aggressive NHL PMBL DLBCL Largely DLBCL DLBCL

218 110 (of 548) 172 (of 399) 341 401 (of 442) 318 (of 647) 576 124 (of 182) 258 (of 612) 285 334

Trial 13

Aviles et al. Engelhard et al.18 ECOG 148419 Aviles et al.14 SWOG 873620 GELA 93-111 GELA 93-412 Aviles et al.15 Aviles et al.16 UNFOLDER17,22 GOELAMS 02 0321

Recruitment Mean age (y) Same period chemotherapy in both arms 1983-1988 1986-1989 1984-1992 1989-1995 1988-1995 1993-2000 1993-2002 2001-2004 2006-2010 2005-2012 2005-2013

59-61 56 59 53-57 59 46-47 68-69 32-35 53 44 56

yes yes yes yes no no yes yes yes no yes

Rituximab used

Radiation dose >30 Gy

no no no no no no no yes yes yes yes

yes yes no yes yes yes yes no no yes yes

Publication

Stages

Bulky disease

Randomized

Int J Radiat Biol 1994 Ann Oncol 1991 J Clin Oncol 2004 Leuk Lymphoma 2004 New Engl J Med 1998 New Engl J Med 2005 J Clin Oncol 2007 Int J Radiat Biol 2012 Hematology 2018 (12-ICML;a122); ASCO 2018;a7574 Blood 2018

advanced localized & advanced localized advanced localized localized localized localized advanced localized & advanced

all 19% initially; bulky not randomized 31% initially (tumor > 10cm) all number unknown, some initially 12% of RT pts.; 10% of non-RT pts 9% of RT pts; 8% of non-RT pts. 94% of RT pts. 30 % of RT pts. 76 % initially

localized

for non-bulky disease only

only CR and bulky disease only CR pts only CR pts only CR and bulky disease all all all only CR pts only CR and bulky disease initially 4 arms; random for RT only in CR pts. random at start, some PR pts. received RT

NHL: non-Hodgkin lymphoma; DLBCL: diffuse large B-cell lymphoma; CR: complete response; pts: points; RT: radiotherapy; y: years.

1926

haematologica | 2021; 106(7)

Meta-analysis of radiotherapy for DLBCL

to improve the outcome of the first-line treatment11,12,17,22 or as a salvage option for patients who achieved only a partial remission after chemotherapy.14,21 As the Korean “ASPIRE” trial was unfortunately later withdrawn (clinicaltrials gov. Indentifier: NCT02054559; three times RCHOP + radiotherapy vs. six times R-CHOP for stage 1 and 2 DLBCL), there is currently no randomized trial supporting the widely used, recently updated and safe approach to give less chemotherapy and PET-guided radiotherapy to patients with localized DLBCL.5,23,30-33 Also the data on limited stage DLBCL which accounts for 30 % of the cases, harbor significant variability as different definitions for limited stage, bulky disease as well as risk stratification and extrapolations were used.2,34 This ren-

ders the integration of all available results difficult. Furthermore, a detailed view goes beyond the possibilities of a meta-analysis analyzing population level data. Although we do not have information on the stage-modified-IPI20,35 for all trials included in our analysis, we assume that many patients with localized disease of this meta-analysis had a low risk disease. They have an excellent prognosis, regardless of radiotherapy.36 The FLYER trial established four cycles of R-CHOP to be sufficient for patients with favorable risk (and non-bulky) DLBCL.37 Radiotherapy in this trial was limited to the contralateral testis in case of testicular involvement. In the yet unpublished OPTIMAL>60 trial (clinicaltrials gov. Identifier: NCT014778542), radiotherapy (and two additional cycles

Figure 2. Effect of consolidation radiotherapy on progression-free survival. Circles are proportional to trial size i.e., number of patients; retracted trials are displayed with hollow circles.

Table 2. Outcome data of the individual trials used for the meta-analysis. Correlation between progression-free and overall survival were done for the GELA trial. The superscript number in the study column refers to the number of the references in the manuscript.

Trial (with reference)

Overall survival Hazard ratio

Progression-free survival ln HR (SE)

Hazard ratio

ln HR (SE)

0.33 2.09 0.81 0.35 0.64 1.98 1.08 0.21 0.28 1.2 0.52

-1.11 (0.44) 0.74 (0.58) -0.21 (0.28) -1.04 (0.25) -0.44 (0.23) 0.68 (0.20) 0.07 (0.14) -1.54 (0.29) -1.27 (0.32) 0.18 (0.38) -0.66 (0.45)

0.31 n/a 0.66 0.36 0.63 1.92 1.09 0.31 0.4 0.7 0.58

?-1.17 (0.36) n/a ?-0.41 (0.24) -1.02 (0.22) -0.46 (0.19) 0.65 (0.16) 0.09 (0.13) -1.16 (0.28) -0.92 (0.28) -0.36 (0.20) -0.54 (0.43)

Aviles et al. Engelhard et al.18 ECOG 148419 Aviles et al.14 SWOG 873620 GELA 93-111 GELA 93-4 12 Aviles et al.15 Aviles et al.16 UNFOLDER17,22 GOELAMS 02 0321

HR: hazard ratio; ln: natural logarithm; SE: standard error.

haematologica | 2021; 106(7)

1927

M.D. Berger et al.

of chemotherapy) is given just to PET-positive sites after four cycles of chemotherapy. According to an interim analysis, this can compensate the inferior outcome of this population.37 Furthermore, the authors of this trial communicated that radiotherapy to PET-negative bulky disease is not needed.38

In the latest ESMO guidelines, consolidation radiotherapy for DLBCL patients is recommended for both elderly and intermediate- and high-risk young patients with bulky disease.6 NCCN is less firm, and mainly restricts its recommendation to residual disease (partial remission or PET-positivity, Online Supplementary Table S1). These rec-

Figure 3. Effect of consolidation radiotherapy on overall survival. Circles are proportional to trial size i.e., number of patients; retracted trials are displayed with hollow circles. CI: Confidence Interval.

Figure 4. Time trend plot on the effect of consolidation radiotherapy. Circles are proportional to weight in analysis; dashed line shows the fitted linear regression; retracted trials are in grey. ln: natural logarithm.

1928

haematologica | 2021; 106(7)

Meta-analysis of radiotherapy for DLBCL

ommendations are not fully supported by the results of this meta-analysis, especially by the results of the stratified analysis provided in Figure 5. The International Lymphoma Radiation Oncology Group (ILROG) has recently updated its guidelines, albeit in the relapsed and refractory setting.39 The trials analyzed in our meta-analysis did not specifically include patients with extranodal DLBCL for which both ESMO40 and ILROG41 have published separate guidelines. Specifically, consolidative mediastinal radiotherapy is currently recommended in responding primary mediastinal B-cell lymphoma (PMBL) patients after treatment with standard-dose chemoimmunotherapy.40 However, extrapolation of the data of our meta-analysis on DLBCL not otherwise specified (NOS) to and from entities such as primary mediastinal lymphoma, primary central nervous system (CNS) or testicular lymphoma is discouraged. The safe omission of whole brain radiotherapy for CNS lymphomas is conceptually controversial.42,43 As the role of adjuvant mediastinal radiotherapy in PMBL patients with complete remission after chemotherapy is unclear and a large number of patients are cured by chemotherapy alone with DAEPOCH-R,44 it is important to note that accrual in IELSG37 (clinicaltrials gov. Identifier: NCT01599559) has recently been completed; this potentially practice changing randomized trial with a non-inferiority design has evaluated the role of consolidation radiotherapy in PETnegative patients. Our meta-analysis provides further evidence that

patients with a complete morphologic remission after chemotherapy or initial bulky disease are unlikely to particularly profit from consolidation radiotherapy.25,38 PET has become an integral part of the treatment of DLBCL patients, although the prognostic value of interim PET is limited,45,46 and a PET-based escalation of chemotherapy was unable to improve the outcome.47 None of the trials that we included in our meta-analysis used a truly PETguided treatment approach. This was applied in limited stage DLBCL in a retrospective32 and also a prospective,30 albeit non-randomized trial. In order not to add also radiotherapy to the recent painful flaws in clinical DLBCL research,3,4 our meta-analysis should be taken into account when a new trial is planned. We provide evidence on patients that we should rather not selectively irradiate, but we still do not know how to use consolidation radiotherapy. Besides its wide and established use in localized disease,5 we see the rationale use of radiotherapy in DLBCL patients analogous to the current situation in Hodgkin’s disease, e.g., for insufficient responses to chemo-immunotherapy. Considering retrospective trials,38,48 radiotherapy could be restricted to PET-positive rests. Among other unanswered questions, this would be practice changing. Ideally, this hypothesis needs corroboration in two separate prospective trials to randomly apply radiotherapy in trial 1 for patients with PET-negative, and trial 2 for patients with PET-positive rests. The first trial would be a non-inferiority trial to proof whether it is safe to not irradiate patients perceived

Figure 5. Stratified progression free survival analysis on the effect of consolidation radiotherapy. Circles are proportional to stratum size i.e., overall number of patients in stratum; color of circles reflects number of trials in stratum i.e., from black (seven trials) to light grey (one trial); age, stage, and bulky disease are characteristics of the study population and cannot be interpreted on the individual participant level (ecological fallacy). CI: Confidence Interval.

haematologica | 2021; 106(7)

1929

M.D. Berger et al.

to be cancer-free. Trial 2 would be a superiority design testing whether radiotherapy is able to improve the outcome residual DLBCL after chemo-immunotherapy. Assuming a 2-year PFS, an appropriate and pragmatic endpoint in DLBCL49,50 of 80%, an alpha of 0.025 for the non-inferiority (one-sided) and 0.05 (two-sided) for the superiority trial, a power of 80% and enrolment over 5 years, we calculated the following sample size: trial 1 (failure rate of 24% [HR 1.23]), would require 1,916 overall or 384 patients per year; trial 2 (and a HR of 0.75 or an improvement of the 2-year PFS to 85%) would need 1,098 patients or 220 patients per year. Assuming an endof-therapy PET-positivity of 25-30%,45 4,000 or 5,000 patients respectively have to be screened. Clearly, such numbers need a global and fully committed academic effort. However, otherwise the important question on the role of consolidation radiotherapy in DLBCL, which with the current data, regularly gives rise to unsatisfactory and futile discussions at lymphoma boards, will never be answered convincingly. Based on this meta-analysis and other data,21 we favor a superiority trial that first allocates a role of consolidation radiotherapy in DLBCL. Then, one may also test the use of smaller irradiation volumes according to the concept of involved node ver-

References 1. Coiffier B, Thieblemont C, Van Den Neste E, et al. Long-term outcome of patients in the LNH-98.5 trial, the first randomized study comparing rituximab-CHOP to standard CHOP chemotherapy in DLBCL patients: a study by the Groupe d'Etudes des Lymphomes de l'Adulte. Blood. 2010;116(12):2040-2045. 2. Pfreundschuh M, Kuhnt E, Trumper L, et al. CHOP-like chemotherapy with or without rituximab in young patients with goodprognosis diffuse large-B-cell lymphoma: 6year results of an open-label randomised study of the MabThera International Trial (MInT) Group. Lancet Oncol. 2011;12(11): 1013-1022. 3. Iacoboni G, Zucca E, Ghielmini M, Stathis A. Methodology of clinical trials evaluating the incorporation of new drugs in the firstline treatment of patients with diffuse large B-cell lymphoma (DLBCL): a critical review. Ann Oncol. 2018;29(5):1120-1129. 4. Goy A. Succeeding in Breaking the R-CHOP Ceiling in DLBCL: learning from negative trials. J Clin Oncol. 2017;35(31):3519-3522. 5. Persky DO, Unger JM, Spier CM, et al. Phase II study of rituximab plus three cycles of CHOP and involved-field radiotherapy for patients with limited-stage aggressive Bcell lymphoma: Southwest Oncology Group study 0014. J Clin Oncol. 2008;26(14):22582263. 6. Tilly H, Gomes da Silva M, Vitolo U, et al. Diffuse large B-cell lymphoma (DLBCL): ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2015;26(Suppl 5):v116-125. 7. Beyar-Katz O, Rowe JM, Townsend LE, Tallman MS, Hadomi R, Horowitz NA. Published abstracts at international meetings often over- or underestimate the initial response rate. Blood. 2017;129(16):23262328. 8. Wayant C, Page MJ, Vassar M. Evaluation of reproducible research practices in oncology

1930

sus involved site radiotherapy using modern techniques (intensity modulated radiotherapy [IMRT]) to reduce doses to organs at risk.31,51 New trials could also approach unanswered questions on the role of consolidation radiotherapy in other subpopulations like patients with interim PET positive disease, or in limited stage disease of high risk histologies such as double hit lymphomas although the prognosis of the later may be better than previously perceived.52 Disclosures No conflicts of interest to disclose. Contributions MDB performed research, analyzed data and wrote parts of the paper; ST analyzed data, contributed vital material and wrote parts of the paper; AEB and SJ performed research and analyzed data; CI analyzed data; TL contributed vital material; UN had the idea, designed research, analyzed data, contributed vital material, and wrote the paper. Acknowledgments We thank Doris Kopp for the literature search, as well as Matthias Egger and Emanuele Zucca for valuable comments.

systematic reviews with meta-analyses referenced by National Comprehensive Cancer Network Guidelines. JAMA Oncol. 2019;5(11):1550-1555. 9. Unger JM. Reproducible findings in systematic reviews and meta-analyses in oncology: verify, then trust. JAMA Oncol. 2019;5(11): 1545-1546. 10. Higgins JP, Altman DG, Gotzsche PC, et al. The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ. 2011;343:d5928. 11. Reyes F, Lepage E, Ganem G, et al. ACVBP versus CHOP plus radiotherapy for localized aggressive lymphoma. N Engl J Med. 2005;352(12):1197-1205. 12. Bonnet C, Fillet G, Mounier N, et al. CHOP alone compared with CHOP plus radiotherapy for localized aggressive lymphoma in elderly patients: a study by the Groupe d'Etude des Lymphomes de l'Adulte. J Clin Oncol. 2007;25(7):787-792. 13. Aviles A, Delgado S, Nambo MJ, Alatriste S, Diaz-Maqueo JC. Adjuvant radiotherapy to sites of previous bulky disease in patients stage IV diffuse large cell lymphoma. Int J Radiat Oncol Biol Phys. 1994;30(4):799-803. 14. Aviles A, Fernandez R, Perez F, et al. Adjuvant radiotherapy in stage IV diffuse large cell lymphoma improves outcome. Leuk Lymphoma. 2004;45(7):1385-1389. 15. Aviles A, Neri N, Fernandez R, HuertaGuzman J, Nambo MJ. Randomized clinical trial to assess the efficacy of radiotherapy in primary mediastinal large B-lymphoma. Int J Radiat Oncol Biol Phys. 2012;83(4):12271231. 16. Aviles A, Nambo MJ, Calva A, Neri N, Cleto S, Silva L. Retracted article: adjuvant radiotherapy in patients with diffuse large B-cell lymphoma in advanced stage (III/IV) improves the outcome in the rituximab era. Hematology. 2019;24(1):521-525. 17. Zwick C, Held G, Ziepert M, et al. The role of radiotherapy to bulky disease in elderly patients with aggressive B-cell lymphoma. Results from two prospective trials of the

DSHNHL. Hematol Oncol. 2013;31(Suppl 1):S96-150. 18. Engelhard M, Meusers P, Brittinger G, et al. Prospective multicenter trial for the response-adapted treatment of high-grade malignant non-Hodgkin's lymphomas: updated results of the COP-BLAM/IMVP-16 protocol with randomized adjuvant radiotherapy. Ann Oncol. 1991;2(Suppl 2):S177180. 19. Horning SJ, Weller E, Kim K, et al. Chemotherapy with or without radiotherapy in limited-stage diffuse aggressive nonHodgkin's lymphoma: Eastern Cooperative Oncology Group study 1484. J Clin Oncol. 2004;22(15):3032-3038. 20. Miller TP, Dahlberg S, Cassady JR, et al. Chemotherapy alone compared with chemotherapy plus radiotherapy for localized intermediate- and high-grade nonHodgkin's lymphoma. N Engl J Med. 1998;339(1):21-26. 21. Lamy T, Damaj G, Soubeyran P, et al. RCHOP 14 with or without radiotherapy in nonbulky limited-stage diffuse large B-cell lymphoma. Blood. 2018;131(2):174-181. 22. 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;360(Suppl 15):S7574. 23. Vargo JA, Gill BS, Balasubramani GK, Beriwal S. Treatment selection and survival outcomes in early-stage diffuse large B-cell lymphoma: do we still need consolidative radiotherapy? J Clin Oncol. 2015;33(32): 3710-3717. 24. Dabaja BS, Vanderplas AM, CrosbyThompson AL, et al. Radiation for diffuse large B-cell lymphoma in the rituximab era: analysis of the National Comprehensive Cancer Network lymphoma outcomes project. Cancer. 2015;121(7):1032-1039. 25. Held G, Murawski N, Ziepert M, et al. Role

haematologica | 2021; 106(7)

Meta-analysis of radiotherapy for DLBCL

of radiotherapy to bulky disease in elderly patients with aggressive B-cell lymphoma. J Clin Oncol. 2014;32(11):1112-1118. 26. Casasnovas RO, Ysebaert L, Thieblemont C, et al. FDG-PET-driven consolidation strategy in diffuse large B-cell lymphoma: final results of a randomized phase 2 study. Blood. 2017;130(11):1315-1326. 27. dos Santos LV, Lima JP, Lima CS, Sasse EC, Sasse AD. Is there a role for consolidative radiotherapy in the treatment of aggressive and localized non-Hodgkin lymphoma? A systematic review with meta-analysis. BMC Cancer. 2012;12:288. 28. Held G, Zeynalova S, Murawski N, et al. Impact of rituximab and radiotherapy on outcome of patients with aggressive B-cell lymphoma and skeletal involvement. J Clin Oncol. 2013;31(32):4115-4122. 29. Recher C, Coiffier B, Haioun C, et al. Intensified chemotherapy with ACVBP plus rituximab versus standard CHOP plus rituximab for the treatment of diffuse large B-cell lymphoma (LNH03-2B): an open-label randomised phase 3 trial. Lancet. 2011;378 (9806):1858-1867. 30. Persky DO, Li H, Stephens DM, et al. PETdirected therapy for patients with limitedstage diffuse large B-cell lymphoma - results of Intergroup NCTN Study S1001. Blood. 2019;134(Suppl 1):S349. 31. Campbell BA, Connors JM, Gascoyne RD, Morris WJ, Pickles T, Sehn LH. Limitedstage diffuse large B-cell lymphoma treated with abbreviated systemic therapy and consolidation radiotherapy: involved-field versus involved-node radiotherapy. Cancer. 2012;118(17):4156-4165. 32. Sehn LH, Scott DW, Villa D, et al. Long-term follow-up of a PET-guided approach to treatment of limited-stage diffuse large B-cell lymphoma (DLBCL) in British Columbia (BC). Blood. 2019;134(Suppl 1):S401. 33. Sehn LH. Chemotherapy alone for localized diffuse large B-cell lymphoma. Cancer J. 2012;18(5):421-426. 34. Persky DO. Limited-stage DLBCL: it's patient selection. Blood. 2018;131(2):155-156. 35. A predictive model for aggressive nonHodgkin's lymphoma. The International Non-Hodgkin's Lymphoma Prognostic Factors Project. N Engl J Med. 1993;329

haematologica | 2021; 106(7)

(14):987-994. 36. Miller TP. The limits of limited stage lymphoma. J Clin Oncol. 2004;22(15):29822984. 37. Poeschel V, Held G, Ziepert M, et al. Four versus six cycles of CHOP chemotherapy in combination with six applications of rituximab in patients with aggressive B-cell lymphoma with favourable prognosis (FLYER): a randomised, phase 3, non-inferiority trial. Lancet. 2020;394(10216):2271-2281. 38. Pfreundschuh M, Christofyllakis K, Altmann B, et al. Radiotherapy to bulky disease PETnegative after immunochemotherapy can be spared in elderly DLBCL patients: results of a planned interim analysis of the first 187 patients with bulky disease treated in the OPTIMAL > 60 study of the DSHNHL. J Clin Oncol. 2017;35(Suppl 15):S7506. 39. Ng AK, Yahalom J, Goda JS, et al. Role of radiation therapy in patients with relapsed/refractory diffuse large B-cell lymphoma: guidelines from the International Lymphoma Radiation Oncology Group. Int J Radiat Oncol Biol Phys. 2018;100(3):652669. 40. Vitolo U, Seymour JF, Martelli M, et al. Extranodal diffuse large B-cell lymphoma (DLBCL) and primary mediastinal B-cell lymphoma: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2016;27(Suppl 5):v91v102. 41. Yahalom J, Illidge T, Specht L, et al. Modern radiation therapy for extranodal lymphomas: field and dose guidelines from the International Lymphoma Radiation Oncology Group. Int J Radiat Oncol Biol Phys. 2015;92(1):11-31. 42. Thiel E, Korfel A, Martus P, et al. High-dose methotrexate with or without whole brain radiotherapy for primary CNS lymphoma (G-PCNSL-SG-1): a phase 3, randomised, non-inferiority trial. Lancet Oncol. 2010;11(11):1036-1047. 43. Ferreri AJM, Cwynarski K, Pulczynski E, et al. Whole-brain radiotherapy or autologous stem-cell transplantation as consolidation strategies after high-dose methotrexatebased chemoimmunotherapy in patients with primary CNS lymphoma: results of the second randomisation of the International

Extranodal Lymphoma Study Group-32 phase 2 trial. Lancet Haematol. 2017;4(11): e510-e523. 44. Dunleavy K, Pittaluga S, Maeda LS, et al. Dose-adjusted EPOCH-rituximab therapy in primary mediastinal B-cell lymphoma. N Engl J Med. 2013;368(15):1408-1416. 45. Mamot C, Klingbiel D, Hitz F, et al. Final results of a prospective evaluation of the predictive value of interim positron emission tomography in patients with diffuse large B-cell lymphoma treated with RCHOP-14 (SAKK 38/07). J Clin Oncol. 2015;33(23):2523-2529. 46. Kurtz DM, Scherer F, Jin MC, et al. Circulating tumor DNA measurements as early outcome predictors in diffuse large Bcell lymphoma. J Clin Oncol. 2018;36(28): 2845-2853. 47. Duhrsen U, Müller S, Hertenstein B, et al. Positron emission tomography-guided therapy of aggressive non-Hodgkin lymphomas (PETAL): a multicenter, randomized phase III trial. J Clin Oncol. 2018;36(20):2024-2034. 48. Freeman CF, Savage KJ, Villa D, et al. Longterm results of PET-guided radiation therapy in patients with advanced-stage diffuse large B-cell lymphoma treated with R-CHOP in British Columbia. Blood. 2017;130(Suppl 1):S823. 49. Maurer MJ, Habermann TM, Shi Q, et al. Progression-free survival at 24 months (PFS24) and subsequent outcome for patients with diffuse large B-cell lymphoma (DLBCL) enrolled on randomized clinical trials. Ann Oncol. 2018;29(8):1822-1827. 50. Shi Q, Schmitz N, Ou FS, et al. Progressionfree Survival as a surrogate end point for overall curvival in first-line diffuse large Bcell lymphoma: an individual patient-level analysis of multiple randomized trials (SEAL). J Clin Oncol. 2018;36(25):25932602. 51. Specht L, Yahalom J. The concept and evolution of involved site radiation therapy for lymphoma. Int J Clin Oncol. 2015;20(5):849854. 52. Torka P, Kothari SK, Sundaram S, et al. Outcomes of patients with limited-stage aggressive large B-cell lymphoma with highrisk cytogenetics. Blood Adv. 2020;4(2):253262.

1931

ARTICLE Ferrata Storti Foundation

Haematologica 2021 Volume 106(7):1932-1942

Non-Hodgkin Lymphoma

Outcomes of Burkitt lymphoma with central nervous system involvement: evidence from a large multicenter cohort study

Adam S. Zayac,1* Andrew M. Evens,2* Alexey Danilov,3 Stephen D. Smith,4 Deepa Jagadeesh,5 Lori A. Leslie,6 Catherine Wei,2 Seo-Hyun Kim,7 Seema Naik,8 Suchitra Sundaram,9 Nishitha Reddy,10 Umar Farooq,11 Vaishalee P Kenkre,12 Narendranath Epperla,13 Kristie A. Blum,14 Nadia Khan,15 Daulath Singh,16 Juan P. Alderuccio,17 Amandeep Godara,18 Maryam Sarraf Yazdy,19 Catherine Diefenbach,20 Emma Rabinovich,21 Gaurav Varma,22 Reem Karmali,23 Yusra Shao,5 Asaad Trabolsi,17 Madelyn Burkart,23 Peter Martin,22 Sarah Stettner,21 Ayushi Chauhan,19 Yun Kyong Choi,20 Allandria StrakerEdwards,15 Andreas Klein,18 Michael C. Churnetski,14 Kirsten M. Boughan,24 Stephanie Berg,16 Bradley M Haverkos,25 Victor M. Orellana-Noia,26 Christopher D'Angelo,12 David A Bond,13 Seth M. Maliske,11 Ryan Vaca,8 Gabriella Magarelli,6 Amy Sperling,4 Max J. Gordon,3 Kevin A. David,2 Malvi Savani,27 Paolo Caimi,24 Manali Kamdar,25 Matthew A. Lunning,28 Neil Palmisiano,29 Parameswaran Venugopal,7 Craig A Portell,26 Veronika Bachanova,27 Tycel Phillips,30 Izidore S. Lossos17 and Adam J. Olszewski1

Lifespan Cancer Institute, Alpert Medical School of Brown University, Providence, RI; Rutgers Cancer Institute of New Jersey, Robert Wood Johnson University Hospital, New Brunswick, NJ; 3Knight Cancer Institute, Oregon Health & Science University, Portland, OR; 4University of Washington/Fred Hutchinson Cancer Research Center, Seattle, WA; 5 Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH; 6John Theurer Cancer Center, Hackensack University Medical Center, Hackensack, NJ; 7Rush University Medical Center, Chicago, IL; 8Penn State Cancer Institute, Penn State University College of Medicine, Hershey, PA; 9Roswell Park Comprehensive Cancer Center, Buffalo, NY; 10 Vanderbilt University Medical Center, Nashville, TN; 11University of Iowa Carver College of Medicine, Iowa City, IA; 12University of Wisconsin Carbone Cancer Center, Madison, WI; 13The Ohio State University Comprehensive Cancer Center, Columbus, OH; 14Winship Cancer Institute, Emory University, Atlanta, GA; 15Fox Chase Cancer Center, Philadelphia, PA; 16Loyola University Medical Center, Loyola University Chicago, Maywood, IL; 17 Sylvester Comprehensive Cancer Center, University of Miami School of Medicine, Miami, FL; 18Tufts Medical Center, Boston, MA; 19Lombardi Comprehensive Cancer Center, Georgetown University Hospital, Washington, DC; 20New York University School of Medicine, Perlmutter Cancer Center, New York, NY; 21University of Illinois at Chicago, Chicago, IL; 22Weill Cornell Medical College, New York, NY; 23Northwestern University, Chicago, IL; 24University Hospitals Seidman Cancer Center, Cleveland, OH; 25University of Colorado Cancer Center, Aurora, CO; 26University of Virginia School of Medicine, Charlottesville, VA; 27University of Minnesota, Minneapolis, MN; 28University of Nebraska Medical Center, Omaha, NE; 29Sidney Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA and 30University of Michigan, Ann Arbor, MI, USA 1 2

Presented in part as an Oral Presentation at the 61st American Society of Hematology Meeting & Exposition, December 7-10, 2019, in Orlando, FL, USA.

Correspondence: ADAM J. OLSZEWSKI adam_olszewski@brown.edu Received: September 6, 2020. Accepted: December 22, 2020. Pre-published: February 4, 2021. https://doi.org/10.3324/haematol.2020.270876

1932

*ASZ and AME contributed equally as co-first authors.

ABSTRACT

entral nervous system (CNS) involvement in Burkitt lymphoma poses a major therapeutic challenge, and the relative ability of contemporary regimens to treat CNS involvement remains uncertain. We describe the prognostic significance of CNS involvement and the incidence of CNS recurrence/progression after contemporary immunochemotherapy using real-world clinicopathological data from adults with Burkitt lymphoma diagnosed between 2009 and 2018 in 30 institutions in the USA. We examined associations between baseline CNS involvement, patients’ characteristics, complete response rates, and survival. We also examined risk factors for CNS recurrence. Of 641 patients (aged 18 to 88 years), 120 (19%) had CNS involvement. CNS involvement was independently associated with human immunodeficiency virus infection, poor performance status, involvement of ≥2 extranodal sites, and bone marrow involvement. Selection of the first-line treatment regimen was unaffected by CNS involvement (P=0.93). Patients with CNS haematologica | 2021; 106(7)

Real-world CNS outcomes in Burkitt lymphoma

disease had significantly lower rates of complete response (59% vs. 77% for patients with and without CNS involvement, respectively; P<0.001), worse 3-year progression-free survival (adjusted hazard ratio [aHR]=1.53, 95% confidence interval [95% CI]: 1.14-2.06; P=0.004) and overall survival (aHR=1.62, 95% CI: 1.18-2.22; P=0.003). The 3-year cumulative incidence of CNS recurrence was 6% (95% CI: 4-8%) and was significantly lower among patients receiving other regimens (CODOX-M/IVAC, 4%, or hyperCVAD/MA, 3%) compared with DA-EPOCH-R (13%; adjusted sub-distribution HR=4.38, 95% CI:, 2.16-8.87; P<0.001). Baseline CNS involvement in Burkitt lymphoma is relatively common and portends inferior prognosis independently of the first-line treatment regimen selected. In real-world practice, regimens including intravenous systemic agents with pronounced CNS penetrance were associated with a lower risk of CNS recurrence. This finding may be influenced by observed suboptimal adherence to the strict CNS staging and intrathecal therapy procedures incorporated in the DA-EPOCH-R regimen.

Introduction Central nervous system (CNS) involvement is a serious complication of Burkitt lymphoma (BL), with an incidence ranging from 5% to 40%.1-7 Most first-line regimens employ dedicated CNS-directed strategies which typically include intrathecal and systemic chemotherapy capable of penetrating the blood-brain barrier. The need for multiple intrathecal injections and potential severe toxicities of high-dose methotrexate (HDMTX) pose challenges to effective CNSdirected therapy, requiring expertise from the clinician and strict adherence by the patient. It is uncertain whether the application and outcomes of CNS-directed treatments in routine clinical practice correspond to those in clinical trials. Short-cycle, dose-intensive immunochemotherapy regimens achieve progression-free survival (PFS) rates of 7080% in BL.8-11 Adverse effects associated with high-intensity regimens have limited the use of such regimens among the elderly or patients positive for human immunodeficiency virus (HIV), leading to interest in less aggressive options. The DA-EPOCH-R regimen (dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin, and rituximab) has demonstrated excellent survival and relatively low toxicity in single-arm trials, indicating that treatment for BL can be de-escalated without apparent loss of efficacy.12,13 Whereas DA-EPOCH-R involves the same or larger numbers of intrathecal chemotherapy administrations as those in common high-intensity regimens, it notably lacks classical CNS-penetrant systemic agents (HDMTX, cytarabine, ifosfamide) used for prophylaxis against CNS disease.14 DA-EPOCH-R requires strict CNS staging procedures (using flow cytometry of the cerebrospinal fluid [CSF] and brain imaging) and specific protocols for prophylactic or therapeutic intrathecal therapy to control the CNS invasion. Although the National Comprehensive Cancer Network (NCCN) guidelines recommend DA-EPOCH-R as a first-line option for BL along with R-CODOX-M/IVAC (rituximab, cyclophosphamide, doxorubicin, vincristine, and HDMTX, alternating with ifosfamide, etoposide, and cytarabine) and RhyperCVAD/MA (rituximab, cyclophosphamide, vincristine, doxorubicin, and dexamethasone alternating with HDMTX/cytarabine), there are limited data about the efficacy of DA-EPOCH-R in individuals with CNS involvement. The guidelines point out that high-risk patients presenting with symptomatic CNS disease should start treatment with the portion of systemic therapy that contains CNS-penetrating drugs, and that patients with parenchymal brain involvement were not included in the clinical trials of DA-EPOCH-R.15 haematologica | 2021; 106(7)

Our objective was to describe factors associated with CNS involvement in BL using a large, multi-institutional dataset designed to study practice patterns and outcomes of adult BL.16 We examined real-world practice of CNSdirected management, and also compared CNS-related outcomes among patients treated with DA-EPOCH-R or with ‘high-intensity’ regimens.

Methods Study cohort In this multicenter retrospective study, we included patients aged ≥18 years diagnosed with BL between 2009 and 2018 in 30 institutions throughout the USA. The study was approved by the institutional review boards of all the institutions and waiver of informed consent was accepted. Of 702 identified subjects, 641 had complete clinicopathological data and were entered into a centralized database. Sixty-one patients were excluded because of pathology inconsistent with BL (n=21), inadequate follow-up (n=15), treatment dates out of range (n=13), or lack of clinical details (n=12). The diagnosis was established by local review of pathology reports. BL was defined according to the World Health Organization (WHO) criteria,17 excluding other entities such as high-grade B-cell lymphoma, not otherwise specified, or double/triple-hit lymphoma. We included cases without known MYC rearrangement if they fulfilled other criteria for BL: smallcell morphology with tingible body macrophages, BCL2-negative, CD10/BCL6-positive immunophenotype, and Ki67 staining at ~100%. Staging evaluations and therapy were completed at the discretion of treating physicians.

Variables and endpoints Investigators collected data using a standardized protocol. Performance status (PS) was assigned according to the Eastern Cooperative Oncology Group (ECOG) scale. CNS involvement was classified as leptomeningeal (based on CSF evaluation, radiological invasion of meninges or cavernous sinus, or clinical cranial nerve palsy) or parenchymal (radiological or biopsyproven invasion of the brain, eye, or spinal cord), and always classified as stage 4 lymphoma. All CNS evaluations were performed according to institutional standards; specific use of imaging, CSF cytology, flow cytometry, or IGH polymerase chain reaction was not recorded. Serum lactate dehydrogenase (LDH) level was standardized relative to institutional upper limit of normal (ULN). Details about the location of CNS recurrence and schedules of intrathecal therapy were collected for a subset of patients who either experienced a CNS recurrence, or who received DA-EPOCH-R with known baseline CNS 1933

A.S. Zayac et al.

involvement. PFS was assessed locally as the time from diagnosis until disease progression, recurrence, or death.18 Overall survival (OS) was calculated from diagnosis until death or last follow-up.

Statistical analysis We compared clinicopathological characteristics between groups using Fisher exact tests and evaluated factors associated with baseline CNS involvement by univariate and multivariable logistic regression (reporting adjusted odds ratios, aOR). Associations with survival were examined in proportional hazard models, first univariate, and then stratified by general BL risk factors identified in the same dataset, reporting hazard ratios (HR).16 The cumulative incidence of CNS recurrence was studied in competing-risk models that accounted for other events such as systemic recurrence or death from any cause, reporting sub distribution hazard ratios (SHR).19 To address missing data on PS (7%), stage (2%), HIV positivity (2%), LDH (7%), hemoglobin (5%), and albumin values (9%), we averaged model coefficients and standard errors from 15 datasets using multiple imputation by chained equations.20 The imputation model included all covariates and outcomes. Estimates report 95% confidence intervals (in square brackets), and two-sided P values <0.05 were considered statistically significant.

Results The study included 641 patients with untreated BL diagnosed at a median age of 47 years (interquartile range [IQR], 34-59 years), who were predominantly male (76%) and had stage 4 disease (73%) (Table 1). The most common first-line regimens were CODOX-M/IVAC (30%), hyperCVAD/MA (30%), or DA-EPOCH-R (28%), and 90% of all patients received rituximab. Eight patients (1%) did not receive any chemotherapy. Intrathecal chemotherapy was given to 545 patients (85%) whereas 396 (62%) received systemic HDMTX as part of their first-line treatment regimen. The median follow-up was 45 months.

Baseline central nervous system involvement CNS involvement was present at diagnosis in 120 patients (19%), including 97 (15%) with leptomeningealonly disease, 20 (3%) with parenchymal disease (of whom 11 had concurrent leptomeningeal disease), and three (1%) with unspecified CNS involvement (Figure 1A). CSF was positive in 91 patients (14% of all cases, and 76% of those with CNS involvement), whereas ten patients had cavernous sinus involvement. Parenchymal disease included brain, ocular, and spinal cord invasion in

Table 1. Patient characteristics stratified by central nervous system involvement at diagnosis.

All

Numbers Age, years <40 ≥40 to 60 ≥60 Sex Male Female HIV infection Stage 4 B symptoms ECOG PS 2-4 Hemoglobin <11.5 g/dLa Albumin <3.5 g/dLa Lactate dehydrogenase > ULN >3x ULN >5x ULN ≥2 extranodal sites Extranodal involvement: Marrow Intestine Liver Pancreas Pleura/peritoneum Kidney/adrenal Testisa Uterus/ovarya Female breasta

Baseline CNS involvement No

Yes

(%)

641

(100)

521

(100)

120

(100)

233 257 151

(36) (40) (24)

195 204 122

(37) (39) (23)

38 53 29

(32) (44) (24)

485 156 142 462 304 144 264 254

(76) (24) (22) (72) (47) (23) (45) (40)

398 123 99 342 236 99 195 187

(76) (24) (19) (66) (45) (19) (37) (36)

87 33 43 120 68 45 69 67

(73) (28) (36) (100) (57) (38) (58) (56)

<0.001 N/A 0.02 <0.001 <0.001 <0.001

463 247 170 275

(72) (39) (27) (43)

361 180 115 194

(69) (35) (22) (37)

102 67 55 81

(85) (56) (46) (68)

<0.001 <0.001 <0.001 <0.001

222 112 88 27 88 54 12 14 14

(35) (18) (14) (4) (14) (8) (3) (9) (9)

146 103 64 24 76 35 8 9 10

(28) (20) (12) (5) (15) (7) (2) (7) (8)

76 9 24 3 12 19 4 5 4

(63) (8) (20) (3) (10) (16) (5) (15) (12)

<0.001 0.001 0.027 0.30 0.19 0.001 0.16 0.16 0.48

0.747

0.37

Cutoffs were empirically determined to provide optimal prognostic discrimination in the main study based on this dataset.16 bPercentages and P values calculated for men or women only, as pertinent. CNS: central nervous system; ECOG PS: Eastern Cooperative Oncology Group performance status; HIV: human immunodeficiency virus; N/A: not applicable; ULN: upper limit of normal.

1934

haematologica | 2021; 106(7)

Real-world CNS outcomes in Burkitt lymphoma

nine, seven, and four patients, respectively. The patients’ characteristics associated with baseline CNS involvement in univariate analysis included HIV infection, B symptoms, PS 2-4, low hemoglobin or albumin, high LDH, ≥2 non-CNS extranodal sites of involvement, and invasion of the bone marrow, liver, kidneys, or adrenal glands (Online Supplementary Table S1). Conversely, patients with extranodal involvement of the intestine were less likely to present with CNS disease. In a multivariate model, HIV infection (aOR: 1.84 [1.12-3.03], P=0.017), poor PS (aOR=2.13 [1.27-3.57], P=0.004), ≥2 extranodal sites (aOR=2.94 [1.75-4.94], P<0.001), and bone marrow involvement (aOR=2.80 [1.59-4.94], P<0.001) retained statistical significance, whereas intestinal involvement was consistently associated with a lower risk (aOR=0.34 [0.160.72], P=0.005). The use of first-line regimens did not differ significantly according to baseline CNS involvement (P=0.93) (Figure 1B). Furthermore, we observed no significant difference in receipt of any intrathecal chemotherapy (89% vs. 84%, respectively; P=0.16) or systemic HDMTX (67% vs. 61%, respectively; P=0.22). Radiation therapy was used somewhat more frequently in patients with CNS disease (11%

vs. 6%, P=0.05), but data on the specific radiation target were not available. Among 20 patients with parenchymal CNS involvement, four (20%) received DA-EPOCH-R, seven (35%) received CODOX-M/IVAC, eight (40%) received hyperCVAD/M, and one received a low-intensity regimen. Among 35 BL patients given DA-EPOCH-R who had CNS invasion, 29 (83%) had only leptomeningeal disease, four (11%) had parenchymal disease, and two (6%) had unspecified involvement. Thirty-four of these 35 patients received care in academic centers. Details of intrathecal administrations were available for 21 patients (60%). Although all 21 received intrathecal chemotherapy, only 45% followed the strict schedule from the original protocol (first twice-weekly, then weekly, and then monthly administration). The median number of intrathecal methotrexate administrations was eight (IQR, 5-12). Clearance of CSF disease was recorded in 89% of patients.

Outcomes of patients with baseline central nervous system involvement Patients with baseline CNS involvement had a significantly lower probability of attaining a complete response:

Figure 1. Baseline central nervous system involvement in Burkitt lymphoma. (A) Proportions of patients with leptomeningeal, parenchymal, or unspecified central nervous system (CNS) involvement; 11 patients with concurrent leptomeningeal and parenchymal disease are included in the last group. (B) Use of first-line chemotherapy regimens stratified by the presence of baseline CNS involvement. (C) Progression-free survival stratified by the presence of CNS involvement. (D) Overall survival stratified by the presence of CNS involvement. Shaded areas indicate 95% confidence interval bands; 3-year survival estimates and P-values from log-rank test are listed. NR: not reached.

haematologica | 2021; 106(7)

1935

A.S. Zayac et al.

59% versus 77% in patients with or without CNS disease, respectively (OR=0.45 [0.29-0.69], P<0.001; excluding untreated patients) as well as worse PFS (3-year estimate, 46% vs. 69%, respectively; HR=2.02 [1.52-2.67], P<0.001) (Figure 1C) and OS (49% vs. 74%, respectively; HR=2.18 [1.61-2.94], P<0.001) (Figure 1D). Patients with CNS involvement had a median PFS of 1.1 years [range, 0.5-4.2] and OS of 2.6 years [0.9 to not reached], whereas medians for those without CNS disease were not reached. CNS involvement remained an independent risk factor for PFS (adjusted HR=1.53 [1.14-2.06], P=0.004) and OS (adjusted HR=1.62 [1.18-2.22], P=0.003) after adjustment for other characteristics independently associated with poor outcomes: age ≥40 years, ECOG PS 2-4, and LDH level >3xULN.16 We observed no difference in PFS based on whether the CNS involvement was parenchymal or leptomeningeal only (log-rank P=0.90), but patients with CNS disease who were diagnosed at age ≥60 years had worse

survival (Online Supplementary Figure S1A-D). As expected, the four patients with parenchymal CNS disease who received DA-EPOCH-R had particularly poor outcomes (3year PFS 25% vs. 57% for those treated with CODOXM/IVAC and 56% for those given hyperCVAD/MA) (Online Supplementary Figure S1E and F). Inferior PFS with baseline CNS involvement was observed regardless of the first-line treatment regimen given (P =0.85) (Figure 2A-C) or whether rituximab was used (P =0.75). Similarly, complete response rates were lower with baseline CNS involvement across regimens (P =0.95) (Figure 2D) and independent of rituximab use (P =0.65). interaction

interaction

Central nervous system recurrence BL recurred in 167 patients (26%), with 39 (6%) presenting a CNS recurrence (21 with and 18 without CNS involvement at diagnosis) (Online Supplementary Figure S2A).

Figure 2. Prognostic significance of baseline central nervous system involvement among patients treated with specific first-line chemotherapy regimens. (A-C) Progression-free survival stratified by the presence of central nervous system (CNS) involvement at diagnosis, for patients treated with: (A) CODOX-M/IVAC (n=194); (B) hyper-CVAD/MA (n=195); and (C) DA-EPOCH-R (n=181). Shaded areas indicate 95% confidence interval bands; 3-year survival estimates and P-values from a logrank test are listed; the summary P-value for interaction between baseline CNS involvement and chemotherapy regimen was 0.85. (D) Proportions of patients achieving complete response (CR) to first-line therapy, stratified by specific regimen and baseline CNS involvement; the summary P-value for interaction between CNS involvement and chemotherapy regimen was 0.95.

1936

haematologica | 2021; 106(7)

Real-world CNS outcomes in Burkitt lymphoma

The median time from diagnosis to CNS recurrence was 5.8 months (IQR, 4.3-7.6). Thirty-four CNS recurrences (87%) occurred during the first year of follow-up and 32 (82%) involved the CNS alone, without systemic disease. Thirtyseven (95%) patients with CNS recurrence had stage 4 disease at initial diagnosis, and the only one with early-stage BL had CNS progression during treatment with RCHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone). The location of the CNS recurrence was only the leptomeninges in 52% of cases, only parenchyma in 22%, both compartments in 19%, and unspecified in 7%. Prior to CNS recurrence, 81% of patients had received rituximab, 97% had received intrathecal chemotherapy, and 44% had received HDMTX. Of note, 22 patients (56%) did not attain a complete response before progression in the CNS, and 18 of those had residual/recurrent disease limited only to the CNS. Following the CNS recurrence, 77% of patients received salvage treatment: systemic chemotherapy (64%), intrathecal chemotherapy (8%), or radiation (5%). Only five (13%) achieved a second complete response and the median survival after CNS recurrence was 2.8 months (range, 2.0-4.2) (Online Supplementary Figure S2B). All five survivors received HDMTX-based salvage therapy. Four of them had received first-line DA-EPOCH-R, one underwent consolidative autologous stem cell transplantation, and one received an allogeneic stem cell transplant. Analysis of the cumulative incidence of CNS recurrence was conducted in the subcohort of 570 patients who received one of the NCCN-recommended first-line regimens: CODOX-M/IVAC, hyperCVAD/MA, or DAEPOCH-R. The cumulative incidence of CNS recurrence in this subcohort was 6% [4-8] at 1 year (Figure 3A). In univariate analysis, the strongest associations we observed were between CNS recurrence and stage 4 disease (Figure 3B), baseline CNS involvement (Figure 3C), and testicular involvement, this last being very rare (Table 2). Other significant factors included HIV infection, poor PS, ≥2 extranodal sites, LDH level >3xULN, and bone marrow involvement (Online Supplementary Figure S3). In a two-variable model, stage 4 disease (adjusted SHR=7.68 [1.01-58.40], P=0.049) and baseline CNS involvement (adjusted SHR=4.04 [2.08-7.87], P<0.001) were cumulatively associated with CNS recurrence. Due to the limited number of events, we did not explore more complex multivariate models. Achievement of complete response after first-line therapy was associated with a lower risk of CNS recurrence (SHR=0.30 [0.15-0.57], P<0.001), whereas receipt of rituximab was not (SHR=1.51 [0.36-6.39], P=0.58).

Central nervous system recurrence according to first-line treatment regimen We examined the risk of CNS recurrence according to first-line treatment regimen, comparing regimens that contain high-dose, CNS-penetrant systemic agents (CODOXM/IVAC or hyperCVAD/MA; n=389) with DA-EPOCH-R (n=181). There was no significant difference between these two groups in the prevalence of baseline CNS involvement (P=0.93), stage 4 disease (P=0.79), LDH level >3xULN (P=0.31), ≥2 extranodal sites (P=0.27), or testicular involvement (P=0.20). However, patients selected for DAEPOCH-R were older (median age 54 vs. 44 years; P<0.001), and more likely to have poor PS (30% vs. 18%; P=0.002), HIV infection (33% vs. 20%; P<0.001), or low haematologica | 2021; 106(7)

Figure 3. Cumulative incidence of central nervous system recurrence. (A-C) The cumulative incidence of central nervous system (CNS) recurrence in: (A) all patients (n=570) who received chemotherapy with CODOX-M/IVAC, hyperCVAD/MA, or DA-EPOCH-R; (B) stratified by disease stage; and (C) stratified by the presence of baseline CNS involvement at diagnosis. Shaded areas indicate 95% confidence interval bands; P values are from univariate competing risk models; estimates at 3 years are presented.

1937

A.S. Zayac et al. Table 2. Risk factors for central nervous system recurrence in Burkitt lymphoma.

Variable

With %

Age ≥40 years Age ≥60 years Female sex HIV infection Stage 4 B symptoms ECOG PS 2-4 Hemoglobin <11.5 g/dL Albumin <3.5 g/dL LDH > ULN LDH >3x ULN LDH >5x ULN ≥2 extranodal sites Involvement at diagnosis: CNS Bone marrow Intestine Liver Pancreas Pleura/peritoneum Kidney/adrenal Testisa Uterus/ovarya Female breasta

Cumulative incidence at 3 years Without 95% CI % 95% CI

SHR

Univariate model 95% CI

6 5 9 11 9 7 11 9 9 7 10 10 9

(4-9) (2-11) (5-14) (6-17) (6-12) (4-11) (6-17) (6-14) (5-13) (5-10) (6-14) (6-16) (6-14)

7 7 6 5 1 6 5 4 5 6 4 5 4

(4-11) (5-9) (4-8) (3-8) (0-3) (4-9) (3-7) (2-7) (3-8) (2-11) (3-7) (3-8) (2-7)

0.92 0.86 1.57 2.04 13.47 1.25 2.31 2.54 1.84 1.48 2.30 2.04 2.13

(0.47-1.79) (0.38-1.95) (0.78-3.13) (1.05-4.00) (1.83-98.9) (0.65-2.40) (1.14-4.67) (1.26-5.11) (0.92-3.66) (0.62-3.55) (1.17-4.50) (1.05-3.97) (1.09-4.16)

0.80 0.72 0.20 0.036 0.011 0.50 0.019 0.009 0.08 0.38 0.016 0.036 0.027

18 9 5 8 9 5 4 26 8 8

(11-26) (6-14) (2-11) (3-16) (2-24) (2-11) (1-12) (6-52) (1-31) (1-29)

4 5 7 6 6 7 7 5 9 9

(2-6) (3-7) (5-9) (4-9) (4-9) (5-9) (5-9) (3-8) (5-15) (5-15)

5.73 2.14 0.74 1.30 1.31 0.76 0.61 5.93 0.97 0.90

(2.98-11.0) (1.09-4.17) (0.29-1.88) (0.54-3.14) (0.32-5.30) (0.27-2.15) (0.15-2.58) (1.74-20.2) (0.13-7.35) (0.11-7.22)

<0.001 0.026 0.52 0.55 0.70 0.60 0.51 0.004 0.97 0.92

a Men or women only, as pertinent. 95% CI: 95% confidence interval; CNS: central nervous system; ECOG PS: Eastern Cooperative Oncology Group performance status; HIV: human immunodeficiency virus; LDH: lactate dehydrogenase; SHR: subhazard ratio; ULN: upper limit of normal.

albumin concentration (48% vs. 34%; P<0.001) (Online Supplementary Table S2). The 3-year risk of CNS recurrence was significantly lower after CODOX-M/IVAC (4% [2-8]) or hyperCVAD/MA (3% [1-6]) than after DA-EPOCH-R (13% [8-18]; SHR=3.57 [1.83-6.97]; P<0.001) (Figure 4A and B). Furthermore, recurrences involved the CNS more frequently after DA-EPOCH-R (40%) than after the other two regimens (16%, P<0.001). The risk was higher regardless of baseline CNS involvement (P =0.99) (Figure 4C and D), age (P =0.94), PS (P =0.12), or HIV status (P =0.86) (Online Supplementary Figure S4A-F). Among patients with baseline CNS involvement receiving DA-EPOCH-R, the 3year incidence of CNS recurrence reached 35% [20-51%]. The risk did not differ significantly between patients treated with CODOX-M/IVAC or hyperCVAD/MA within any subset. The increased risk of CNS recurrence after DAEPOCH-R persisted adjusting for age, PS, stage 4 disease, HIV positivity, baseline CNS involvement, and testicular involvement (adjusted SHR=4.38 [2.16-8.87], P<0.001). All patients with CNS recurrence after DA-EPOCH-R treatment had received intrathecal chemotherapy during their initial therapy with a median of six (IQR, 5-12) doses of methotrexate and a median of four (IQR, 0-5) doses of cytarabine, but strict adherence to the protocol-defined schedule was reported in only 57% of this subgroup. CNS recurrence after DA-EPOCH-R treatment was leptomeningeal in 63%, intraparenchymal in 25%, and in both compartments in 12% of cases. Seven patients (6 with baseline leptomeningeal disease) received prophylactic HDMTX (± cytarabine) upon completion of DA-EPOCH-R,

but BL recurred in five, including three with a CNS recurrence. Among all patients receiving DA-EPOCH-R, factors significantly associated with CNS recurrence included baseline CNS involvement (SHR=5.97 [2.59-13.79]; P<0.001), marrow involvement (SHR=2.57 [1.07-6.14]; P=0.034), LDH level >3xULN (SHR=2.53 [1.08-5.94]; P=0.033), and ≥2 extranodal sites (SHR=3.29 [1.28-8.44]; P=0.013). However, adjusting for baseline CNS involvement, no other variable retained statistical significance.

interaction

1938

interaction

DIscussion In this large, multi-institutional dataset of patients with BL treated with contemporary immunochemotherapy regimens, CNS involvement at diagnosis was an independent prognostic factor for PFS and OS, and a strong predictor of subsequent CNS recurrence. CNS recurrences developed early and almost exclusively in stage 4 BL. No patient with stage 1 or 2 disease experienced a CNS recurrence after treatment with NCCN-recommended regimens. The risk of CNS recurrence was significantly lower with regimens containing high-dose systemic CNS-penetrating agents. However, we observed that among patients with baseline CNS involvement who were treated with DA-EPOCH-R, intrathecal therapy was frequently (in 55%) not administered according to the schedule in the published protocol, which may have contributed to the high observed incidence of CNS recurrence (35%). Our results have important implications for the management of BL in clinical practice, given the rarity of the disease and the paucity of randomized trials. haematologica | 2021; 106(7)

Real-world CNS outcomes in Burkitt lymphoma

Figure 4. Cumulative incidence of central nervous system recurrence according to first-line chemotherapy regimen. (A-D) The cumulative incidence of central nervous system (CNS) recurrence in: (A) all patients (n=570) treated with CODOX-M/IVAC, hyper-CVAD/MA, or DA-EPOCH-R; (B) patients with stage 4 disease (n=413); (C) patients without CNS involvement at diagnosis; and (D) patients with CNS involvement at diagnosis. Subhazard ratios were derived from univariate competing-risk models comparing hyperCVAD versus CODOX-M/IVAC, or DA-EPOCH-R versus both high-intensity regimens combined. 95% CI: 95% confidence interval; SHR: subjazard ratios.

The prevalence of baseline CNS involvement in this real-world BL cohort (19%) is higher than that in phase II trials (10-14%)8,9,11-13,21-24 or in a recent retrospective study of 264 patients treated with immunochemotherapy (8%),5 but lower than that in a phase III trial (25%).10 These differences may reflect both selection bias in smaller trials, and less rigorous CNS staging in routine practice (compared with the phase III setting), in which the use of CSF cytology or flow cytometry is not standardized. Our study had sufficient power to demonstrate independent impacts of CNS involvement on rates of complete response, PFS, and OS in BL, as suggested by some trials,8,13,22,25 but not by others.9,10,21,24,26 Interestingly, the choice of using an immunochemotherapy regimen did not differ according to CNS involvement, and CNS involvement was prognostically unfavorable regardless of the use of first-line rituximab or any specific regimen, including those that contained HDMTX. Collecting granular data on CNS-directed intrathecal therapy proved challenging in this retrospective study, but intensive administration schedules designed for patients haematologica | 2021; 106(7)

with CNS involvement appeared difficult to execute in real-world practice. All regimens used in our cohort (hyperCVAD/MA, CODOX-M/IVAC, and DA-EPOCHR) involve intensified intrathecal regimens for patients with CNS disease, which requires thorough staging to identify subclinical CNS invasion and expertise in the delivery of intrathecal agents, including the use of intraventricular devices. We could not determine whether these schedules were correctly applied in most patients, and whether barriers to effective CNS-directed therapy were related to physicians’ preference, patients’ refusal, or system-level factors. Furthermore, we did not have data on the use of intraventricular reservoirs, although their availability may affect the efficacy of intrathecal therapy. Both poor prognosis with CNS involvement and suboptimal delivery of CNS-directed therapy indicate that patients with CNS disease need more efficacious and practicable treatment approaches that can be consistently administered in routine clinical practice. CNS recurrence after standard immunochemotherapy regimens was uncommon (6%) and exclusive to patients 1939

A.S. Zayac et al.

with stage 4 disease. Of note, over half of CNS recurrences emerged in patients who did not attain complete response, suggesting primary refractory disease or inadequate control of baseline CNS involvement. Clinical trials have reported lower incidences (0-4%) of CNS recurrence,8,12,13,21,22,27,28 which may reflect selective enrollment of lower-risk patients in trials, or suboptimal CNS-directed therapy in our real-world sample. Despite high rates of intrathecal therapy, one in five patients with baseline CNS involvement succumbed to a CNS recurrence, which was predominantly leptomeningeal. Survival after CNS recurrence was dismal and consistent with generally poor outcomes of recurrent BL.7,29-31 HDMTX salvaged a few patients, but mainly those who did not receive it as part of their initial treatment regimen. An emerging finding was the lower risk of CNS recurrence among patients treated with regimens that contain high-dose systemic CNS-penetrant agents compared with DA-EPOCH-R, which relies solely on intrathecal chemotherapy for CNS management. Our observation could not be explained by stage distribution or prevalence of baseline CNS involvement, which were similar among patients receiving these strategies. Although patients selected for DA-EPOCH-R were older and had a worse PS, possibly compromising the intensity of treatment and adherence to CNS prophylaxis, the increased risk persisted after adjustment for these factors. In this context, it is important to note that we could not ascertain specific CNS staging procedures performed in our cohort. Patients enrolled in the clinical trials of DA-EPOCH-R uniformly underwent CSF evaluations using flow cytometry.12,13 Underdiagnosis of occult leptomeningeal disease might result in suboptimal intrathecal treatment, as DA-EPOCHR relies on an intensive intrathecal regimen (starting with twice-weekly administration) in cases with CSF involvement, whether detected by cytology or flow cytometry. The NCI-9177 study enrolled no patients with parenchymal CNS disease, and out of 11 subjects with CSF involvement, six (55%) experienced relapse or death.12,13 Considering this experience and the outcomes observed in our series, DA-EPOCH-R should likely be avoided for treatment of BL with parenchymal CNS involvement. Among 81 high-risk patients without baseline CSF involvement, two developed a parenchymal CNS recurrence despite prophylaxis with eight intrathecal injections. In contrast, we observed CNS recurrence in 18% of patients with stage 4 BL receiving DA-EPOCH-R and in 35% among those with baseline CNS involvement. We could not discern the reasons for the lower intensity of the intrathecal administration schedule during DA-EPOCH-R (median 6 doses), which possibly contributed to the high failure rate of CNS control; adherence was poor even in academic centers. Because of unavoidable immortal-time bias, we could not compare outcomes based on degree of adherence in our retrospective data. It is possible that with stricter adherence the difference in CNS recurrence between patients treated with the different regimens would not have been significant. However, 37% of recurrences involved the parenchymal CNS compartment, which may not be prevented with intrathecal therapy. On the other hand, reassuringly, there were no instances of CNS recurrence in patients with disease stages 1-3, supporting the observation of 100% PFS in low-risk BL after three cycles of DA-EPOCH-R without any CNS prophylaxis.13 Because most patients in our study received 1940

intrathecal chemotherapy, we could not identify a subgroup that could omit it. We highlight an unmet need for prospective studies of augmentation of DA-EPOCH-R for patients with CNS involvement using a CNS-directed strategy that is more efficacious and feasible to execute in routine clinical practice. The few attempts at “consolidative” HDMTX (± cytarabine) after DA-EPOCH-R observed in our sample do not enable any true interpretation of treatment effect, as these patients had CNS involvement and were likely at particularly high perceived risk of recurrence. The unfavorable outcomes of this approach may reflect the delayed HDMTX administration, lack of added benefit beyond that of intensive intrathecal therapy in leptomeningeal disease, or simply a matter of bias. The need to adjust doses based on degree of cytopenias, early emergence of CNS recurrences, and high failure rate of “consolidative” HDMTX prophylaxis in our admittedly small sample illustrate the challenges of designing such an augmentation.32,33 We point out that an analysis of this dataset, as well as other retrospective studies, does not suggest significant differences in the overall PFS or OS according to the first-line treatment regimen used in BL.4,5,16 The strengths of this study include its multicenter scope encompassing academic and community-based practices, HIV-associated BL, and a large subset with CNS involvement to allow for an in-depth analysis. Limitations include the retrospective design, rarity of CNS recurrence, and variation in CNS-directed staging or treatment, particularly limited information on the use of CSF flow cytometry, intensity of intrathecal regimens and mode of delivery (intraventricular catheter or lumbar puncture). We also lacked molecular data that could provide insights into the biology of CNS invasion and inform future treatment strategies. In conclusion, despite the success of immunochemotherapy in BL, patients with CNS involvement constitute a high-risk group in need of better management. In this large real-world dataset, CNS involvement was associated with worse prognosis regardless of first-line regimen applied, and independently of other factors. Regimens incorporating high-dose systemic CNS-penetrant agents were associated with a lower risk of CNS recurrence when administered to patients with baseline CNS involvement in routine clinical practice. Selective use of DA-EPOCH-R for patients with advanced age and worse PS, who could not maintain the intrathecal administration schedule inherent to this protocol, may have influenced our findings. An important insight from our analysis is that when clinicians apply the DA-EPOCH-R regimen, they should strictly follow the pre-specified CNS staging procedures (including CSF flow cytometry) and adequately tailored intrathecal administration schedules, as frequent deviations from the protocol may lead to suboptimal CNS control. CNS-related outcomes from the ongoing HOVON-127BL randomized trial comparing DA-EPOCH-R with CODOX-M/IVAC (EudaraCT 2013-004394-27) will be critical to further define optimal therapy, provided that sufficient numbers of patients with baseline CNS involvement are enrolled. Further prospective studies are needed to optimize realistic delivery of CNS-directed prophylaxis with all standard regimens and to mitigate the incidence of CNS recurrence. Disclosures AE has received research funding from Takeda and Merck; has received honoraria from Research to Practice; has received honoraria from and provided consultancy services for Seattle Genetics, haematologica | 2021; 106(7)

Real-world CNS outcomes in Burkitt lymphoma

Verastem, Affimed and Bayer; and has received honoraria from and sat on a DMC for Pharmacyclics. AD has received research funding from Aptose Biosciences, Gilead Sciences, Takeda Oncology and Bristol-Myers Squibb; has received research funding and provided consultancy services for AstraZeneca, Genentech, Bayer Oncology and Verastem Oncology; has acted as a consultant for Beigene. TG Therapeutics, Celgene, Nurix and Rigel Pharmaceuticals; and is a Leukemia and Lymphoma Society Scholar in Clinical Research. SDS has received research funding from Incyte Corporation, Seattle Genetics, Portola Pharmaceuticals, Pharmacyclics, Acerta Pharma BV, Genentech and Denovo Biopharma; he has received research funding from and acted as a consultant for Merck Sharp & Dohme Corp; and he has been a member of Astra Zeneca’s Board of Directors or advisory committees and has received research funding from this company; a Astra Zeneca. An immediate family member has received research funding from Ignyta, Bristol-Myers Squibb and Ayala. SN has provided advisory board services for Celgene and Sanofi. NR has acted as a consultant for KITE Pharma, Abbvie and Celgene; has received research funding from Genentech; and has provided consultancy services for and received research funding from BMS. UF has received honoraria from Celgene and research funding from Kite Pharma. NE has received honoraria from Pharmacyclics and taken part in a speakers bureau for Verastem Oncology. NK has been a member of the Board of Directors or an advisory committee of Seattle Genetics and Abbvie; has received research funds from Bristol Myers; and has delivered educational content or symposia for Janssen. JPA has received honoraria from Targeted Oncology; and acted as a consultant for OncLive. An immediate family member has had contacts with Puma Biotechnology, Agios, Inovio Pharmaceuticals and Foundation Medicine. MSY has received honoraria from Bayer; has received research funding from Genentech; and has acted as a consultant for Octapharma and Abbvie. CD has received research funding from Denovo, Incyte, LAM Therapeutics, MEI, Millenium/Takeda and Trillium; and has acted as consultant for and received research funding from Bristol-Myers Squibb, Genentech, Merck and Seattle Genetics. RK has sat on speakers’ bureaus for Takeda, AstraZeneca and BeiGene; has supplied advisory board services

References 1. Zayac AS, Olszewski AJ. Burkitt lymphoma: bridging the gap between advances in molecular biology and therapy. Leuk Lymphoma. 2020;61(8):1784-1796. 2. Sariban E, Edwards B, Janus C, Magrath I. Central nervous system involvement in American Burkitt's lymphoma. J Clin Oncol. 1983;1(11):677-681. 3. Boehme V, Zeynalova S, Kloess M, et al. Incidence and risk factors of central nervous system recurrence in aggressive lymphoma--a survey of 1693 patients treated in protocols of the German High-Grade NonHodgkin's Lymphoma Study Group (DSHNHL). Ann Oncol. 2007;18(1):149157. 4. Oosten LEM, Chamuleau MED, Thielen FW, et al. Treatment of sporadic Burkitt lymphoma in adults, a retrospective comparison of four treatment regimens. Ann Hematol. 2018;97(2):255-266. 5. Jakobsen LH, Ellin F, Smeland KB, et al. Minimal relapse risk and early normalization of survival for patients with Burkitt lymphoma treated with intensive immunochemotherapy: an international study of 264 real-world patients. Br J

haematologica | 2021; 106(7)

for Karyopharm; and has sat on speakers’ bureaus and advisory boards and received research support from Kite, Gilead, BMS, Celgene and Juno. PM has acted as a consultant for Celgene, Teneobio, Karyopharm, Janssen, Sandoz and I-MAB. GM has participated in speakers’ bureaus for Tevan Oncology. MK has acted as a consultant for Pharmacyclics, AstraZeneca and Celgene; has sat on speakers’ bureaus for Seattle Genetics; and is an employee of the University of Colorado. ML has acted as a consultant for Spectrum, Seattle Genetics, Portola, OncLive, Novartis. Kite, Gilead Sciences, Inc., DAVA, Bayer, AbbVie, VANIUM and Verastem; has received research funding from MiRagen and Curis; and has acted as a consultant for and received research funding from TG Therapeutics, Juno Therapeutics and Janssen Scientific Affairs, LLC. CP has received research funding from Xencor, Roche/Genentech, Infinity, TG Therapeutics, AbbVie and Acerta/AstraZeneca; has acted as a consultant for Pharmacyclics, Janssen, Amgen and Bayer, and has provide consultancy services for and received research funding from Genentech, BeiGene and Kite. ISL has been a member of the Board of Directors or an advisory committee of Janssen Scientific and Seattle Genetics; and has received research funding from the NIH. AJO has received research funding from Genentech, Adaptive Biotechnologies, TG Therapeutics and Spectrum Pharmaceuticals. The remaining authors report no significant conflicts of interest. Contributions ASZ, AME, ISL and AJO designed the research, performed the analysis, and wrote the manuscript draft. ASZ and AME contributed equally. AD, SDS, DJ, LAL, CW, S-HK, SN, SSu, NR, UF, VPK, NE, KAB, NK, DS, JPA, AG, SY, CD, ER, GV, RK, YS, AT, MB, PM, SSt, AC, YKC, AS-E, AK, MCC, KMB, SB, BMH, VMO-N, CD'A, DAB, SMM, RV, GM, AS, MJG, KAD, MS, PC, MK, MAL, NP, PV, CAP, VB and TP contributed patients’ data, and edited and approved the manuscript. Funding This work was conducted with voluntary support of the contributing institutions and researchers, without any external funding.

Haematol. 2020;189(4):661-671. 6. Thomas DA, Cortes J, O'Brien S, et al. Hyper-CVAD program in Burkitt's-type adult acute lymphoblastic leukemia. J Clin Oncol. 1999;17(8):2461-2470. 7. Woessmann W, Zimmermann M, Meinhardt A, et al. Progressive or relapsed Burkitt lymphoma or leukemia in children and adolescents after BFM-type first-line therapy. Blood. 2020;135(14):1124-1132. 8. Thomas DA, Faderl S, O'Brien S, et al. Chemoimmunotherapy with hyper-CVAD plus rituximab for the treatment of adult Burkitt and Burkitt-type lymphoma or acute lymphoblastic leukemia. Cancer. 2006;106(7):1569-1580. 9. Rizzieri DA, Johnson JL, Byrd JC, et al. Improved efficacy using rituximab and brief duration, high intensity chemotherapy with filgrastim support for Burkitt or aggressive lymphomas: cancer and Leukemia Group B study 10 002. Br J Haematol. 2014;165(1):102-111. 10. Ribrag V, Koscielny S, Bosq J, et al. Rituximab and dose-dense chemotherapy for adults with Burkitt's lymphoma: a randomised, controlled, open-label, phase 3 trial. Lancet. 2016;387(10036):2402-2411. 11. Lacasce A, Howard O, Lib S, et al. Modified

magrath regimens for adults with Burkitt and Burkitt-like lymphomas: preserved efficacy with decreased toxicity. Leuk Lymphoma. 2004;45(4):761-767. 12. Dunleavy K, Pittaluga S, Shovlin M, et al. Low-intensity therapy in adults with Burkitt's lymphoma. N Engl J Med. 2013;369(20):1915-1925. 13. Roschewski M, Dunleavy K, Abramson JS, et al. Multicenter study of risk-adapted therapy with dose-adjusted EPOCH-R in adults with untreated Burkitt lymphoma. J Clin Oncol. 2020;38(22):2519-2529. 14. Alderuccio JP, Lossos IS. DA-EPOCH-R for adult Burkitt's lymphoma: pros and cons. J Oncol Pract. 2018;14(11):676-678. 15. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®): B-cell lymphomas. Version 4.2020 - August 13, 2020. Available from: https://www.nccn.org/professionals/physician_gls/pdf/b-cell.pdf. 16. Evens AM, Danilov AV, Jagadeesh D, et al. Burkitt lymphoma in the modern era: real world outcomes and prognostication across 30 US cancer centers. Blood. 2021;137 (3):374-386. 17. Swerdlow SH, Campo E, Pileri SA, et al. The 2016 revision of the World Health

1941

A.S. Zayac et al. Organization classification of lymphoid neoplasms. Blood. 2016;127(20):2375-2390. 18. Cheson BD, Pfistner B, Juweid ME, et al. Revised response criteria for malignant lymphoma. J Clin Oncol. 2007;25(5):579586. 19. Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999;94(446):496509. 20. White IR, Royston P, Wood AM. Multiple imputation using chained equations: issues and guidance for practice. Stat Med. 2011;30(4):377-399. 21. Noy A, Lee JY, Cesarman E, et al. AMC 048: modified CODOX-M/IVAC-rituximab is safe and effective for HIV-associated Burkitt lymphoma. Blood. 2015;126(2):160-166. 22. Hoelzer D, Walewski J, Dohner H, et al. Improved outcome of adult Burkitt lymphoma/leukemia with rituximab and chemotherapy: report of a large prospective multicenter trial. Blood. 2014;124(26):38703879. 23. Corazzelli G, Frigeri F, Russo F, et al. RDCODOX-M/IVAC with rituximab and intrathecal liposomal cytarabine in adult Burkitt lymphoma and 'unclassifiable'

1942

highly aggressive B-cell lymphoma. Br J Haematol. 2012;156(2):234-244. 24. Mead GM, Barrans SL, Qian W, et al. A prospective clinicopathologic study of dose-modified CODOX-M/IVAC in patients with sporadic Burkitt lymphoma defined using cytogenetic and immunophenotypic criteria (MRC/NCRI LY10 trial). Blood. 2008;112(6):2248-2260. 25. Wasterlid T, Brown PN, Hagberg O, et al. Impact of chemotherapy regimen and rituximab in adult Burkitt lymphoma: a retrospective population-based study from the Nordic Lymphoma Group. Ann Oncol. 2013;24(7):1879-1886. 26. Mead GM, Sydes MR, Walewski J, et al. An international evaluation of CODOX-M and CODOX-M alternating with IVAC in adult Burkitt's lymphoma: results of United Kingdom Lymphoma Group LY06 study. Ann Oncol. 2002;13(8):1264-1274. 27. Divine M, Casassus P, Koscielny S, et al. Burkitt lymphoma in adults: a prospective study of 72 patients treated with an adapted pediatric LMB protocol. Ann Oncol. 2005;16(12):1928-1935. 28. Patte C, Auperin A, Michon J, et al. The Societe Francaise d'Oncologie Pediatrique LMB89 protocol: highly effective multia-

gent chemotherapy tailored to the tumor burden and initial response in 561 unselected children with B-cell lymphomas and L3 leukemia. Blood. 2001;97(11):3370-3379. 29. Short NJ, Kantarjian HM, Ko H, et al. Outcomes of adults with relapsed or refractory Burkitt and high-grade B-cell leukemia/lymphoma. Am J Hematol. 2017;92(6):E114-E117. 30. Decker DP, Egan PC, Zayac AS, Treaba DO, Olszewski AJ. Treatment strategies and risk of central nervous system recurrence in high-grade B-cell and Burkitt lymphoma. Leuk Lymphoma. 2020;61(1):198201. 31 Maramattom LV, Hari PN, Burns LJ, et al. Autologous and allogeneic transplantation for burkitt lymphoma outcomes and changes in utilization: a report from the center for international blood and marrow transplant research. Biol Blood Marrow Transplant. 2013;19(2):173-179. 32. Gastwirt JP, Roschewski M. Management of adults with Burkitt lymphoma. Clin Adv Hematol Oncol. 2018;16(12):812-822. 33. Dunleavy K. Approach to the diagnosis and treatment of adult Burkitt's lymphoma. J Oncol Pract. 2018;14(11):665-671.

haematologica | 2021; 106(7)

ARTICLE

Plasma Cell Disorders

Halting the vicious cycle within the multiple myeloma ecosystem: blocking JAM-A on bone marrow endothelial cells restores angiogenic homeostasis and suppresses tumor progression Antonio G. Solimando,1,2,3 Matteo C. Da Viá,1,4,5 Patrizia Leone,3 Paola Borrelli,6 Giorgio A. Croci,7,8 Paula Tabares,1,9 Andreas Brandl,1,9 Giuseppe Di Lernia,3 Francesco P. Bianchi,10 Silvio Tafuri,10 Torsten Steinbrunn,1 Alessandra Balduini,11,12 Assunta Melaccio,3 Simona De Summa,13 Antonella Argentiero,2 Hilka Rauert-Wunderlich,14 Maria A. Frassanito,3 Paolo Ditonno,2 Erik Henke,15 Wolfram Klapper,7 Roberto Ria,3 Carolina Terragna,16 Leo Rasche,1 Andreas Rosenwald,14 K. Martin Kortüm,1 Michele Cavo,16 Domenico Ribatti,17 Vito Racanelli,3 Hermann Einsele,1 Angelo Vacca3 and Andreas Beilhack1,9

Department of Medicine II, University Hospital of Würzburg, Würzburg, Germany; IRCCS Istituto Tumori "Giovanni Paolo II", Bari, Italy; 3Department of Biomedical Sciences and Human Oncology, Unit of Internal Medicine “Guido Baccelli”, University of Bari Aldo Moro Medical School, Bari, Italy; 4Hematology Unit, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy; 5Department of Oncology and Hemato-Oncology, University of Milan, Milan, Italy; 6Unit of Biostatistics and Clinical Epidemiology, Department of Public Health, Experimental and Forensic Medicine, University of Pavia, Pavia, Italy; 7Department of Pathology, Hematopathology Section and Lymph Node Registry, University of Kiel/University Hospital Schleswig-Holstein, Kiel, Germany; 8Pathology Unit, Department of Pathophysiology and Transplantation, University of Milan and Fondazione IRCCS, Ca' Granda, Milan, Italy; 9Interdisciplinary Center for Clinical Research Laboratory, University Hospital of Würzburg, Würzburg, Germany; 10Section of Hygiene, Department of Biomedical Science and Human Oncology, University of Aldo Moro Medical School, Bari, Italy; 11Department of Molecular Medicine, University of Pavia, Pavia, Italy; 12Department of Biomedical Engineering, Tufts University, Medford, MA, USA; 13Molecular Diagnostics and Pharmacogenetics Unit, IRCCS Istituto Tumori "Giovanni Paolo II", Bari, Italy; 14Institute of Pathology, University of Würzburg, Würzburg, Germany; 15Institute of Anatomy and Cell Biology, Julius-Maximilians Universität Würzburg, Würzburg, Germany; 16Institute of Hematology "L. and A. Seràgnoli", Bologna, Italy and 17Department of Basic Medical Sciences, Neurosciences and Sensory Organs, University of Bari Aldo Moro Medical School, Bari, Italy

Ferrata Storti Foundation

Haematologica 2021 Volume 106(7)1943-1956

1 2

ABSTRACT

nteractions of malignant multiple myeloma (MM) plasma cells with the microenvironment control MM plasma-cell growth, survival, drug-resistance and dissemination. As microvascular density increases in the bone marrow in MM, we investigated whether bone marrow MM endothelial cells control disease progression via the junctional adhesion molecule-A (JAM-A). Membrane and cytoplasmic JAM-A levels were upregulated in MM endothelial cells in 111 patients with newly diagnosed MM and in 201 with relapsed/refractory MM compared to the levels in patients with monoclonal gammopathy of undetermined significance and healthy controls. Elevated membrane expression of JAM-A on MM endothelial cells predicted poor clinical outcome. Mechanistically, addition of recombinant JAM-A to MM endothelial cells increased angiogenesis, whereas inhibition of this adhesion molecule impaired angiogenesis and MM growth in two-dimensional and three-dimensional in vitro cell cultures and chorioallantoic membrane assays. To corroborate these findings, we treated MM-bearing mice with a JAM-A-blocking monoclonal antibody and demonstrated impaired MM progression, corresponding to decreased MM-related vascularity. These findings support the concept that JAM-A is an important mediator of MM progression through facilitating MM-associated angiogenesis. Elevated haematologica | 2021; 106(7)

Correspondence: ANDREAS BEILHACK beilhack_a@klinik.uni-wuerzburg.de Received: October 7, 2019. Accepted: April 28, 2020. Pre-published: June 4, 2020. https://doi.org/10.3324/haematol.2019.239913

1943

A.G. Solimando et al.

JAM-A expression on bone marrow endothelial cells is an independent prognostic factor for the survival of both patients with newly diagnosed MM and those with relapsed/refractory MM. Blocking JAM-A restricts angiogenesis in vitro, in utero and in vivo and represents a suitable druggable molecule to halt neo-angiogenesis and MM progression.

Introduction Junctional adhesion molecule-A (JAM-A), also known as JAM-1, CD321, and F11R, belongs to the immunoglobulin superfamily.1 In healthy tissues, JAM-A regulates cell growth and differentiation, while its aberrant expression or deregulation confers a more aggressive phenotype with poor prognosis in different types of human cancers,1 including multiple myeloma (MM),2 breast, lung, brain, and head and neck cancers.3 Overactivation of JAM-A results either from upregulation or aberrant dimerization, driving the receptor in a state of constitutive signal transmission, or from excessive release of JAM-A ligands by normal and tumor cells into the microenvironment.4 Membrane-bound JAM-A and its soluble form (sJAM-A) can form homophilic interactions and also heterophilic interactions1 with lymphocyte functionassociated antigen 1 (LFA-1), afadin (AFDN), calcium/calmodulin-dependent serine protein kinase (CASK) and tight junction protein-1 (TJP1) with high receptor/ligand binding affinities.5 These interactions trigger JAM-A downstream signaling pathways involved in the regulation of tumor cell survival, growth, angiogenesis and dissemination.6 JAM-A inhibition can be achieved directly by blocking the ligand-binding site on the extracellular receptor domain with monoclonal antibodies7 and indirectly with small-molecule inhibitors.8 Moreover, neutralizing the sJAM-A9 released into the microenvironment can prevent JAM-A activation.10 JAM-A plays a pivotal role in endothelial cell physiology6 and pathology.2 Although the function of JAM-A in tumorigenesis has been investigated in solid tumors,3 and its angiogenic role has been shown in pancreatic islet carcinoma,11 data on JAM-A-related angiogenesis in hematologic neoplasms remain elusive. Since bone marrow (BM) neovascularization favors the progression of MM,12 we investigated whether JAM-A can drive angiogenesis in MM,13 contributing to progression of the disease.2 We quantified JAM-A surface expression on BM-derived endothelial cells (MMEC) from 312 patients with MM and demonstrated that JAM-Ahigh MMEC correlate strongly with poor survival both in newly diagnosed (NDMM) and relapsed/refractory (RRMM) patients. Mechanistically, adding recombinant JAM-A protein to MM plasma cells (MM-cells) increased angiogenesis in both two-dimensional (2D) and three-dimensional (3D) models. Conversely, blocking JAM-A impaired MM-related angiogenesis. To corroborate these findings, we treated MM-bearing mice with JAM-A-blocking monoclonal antibodies and observed impaired MM progression and decreased MM vascularity.

(n=201) and subjects with monoclonal gammopathy of undetermined significance (MGUS) (n=35) were included in this study. The patients’ characteristics and genetic risk stratification are provided in Online Supplementary Tables S1 and S2. The study was approved by the Ethical Committees of Bari and Würzburg University Hospitals (reference numbers 5145 and 76/13), and all patients provided informed consent to participation in the study, in accordance with the Declaration of Helsinki (details are given in the Online Supplementary Methods).

Cell lines and cultures procedures RPMI-8226, OPM-2 and human umbilical vein endothelial cells were cultured as described elsewhere.3 MM-cells were cocultured with MMEC (4×105) at 1:1 and 1:5 cell ratios for 24 hours (h) with or without an inserted transwell (0.4 mm pore size; Costar, Cambridge, MA, USA). Details are provided in the Online Supplementary Methods.

Chick chorioallantoic membrane assay Fertilized chicken eggs were incubated at 37°C at a constant humidity. On day 8, sterilized gelatin sponges imbued with MMEC conditioned medium or medium obtained by treatment of MMEC with sJAM-A (100 ng/mL), with or without anti-JAMA monoclonal antibodies were implanted on the top of the chick chorioallantoic membrane (CAM) as described in more detail in the Online Supplementary Methods.

Multiple myeloma xenograft mouse models Twenty female 8- to 10-week-old NOD.CB17Prkdcscid/NCrHsd mice (NOD-SCID; Envigo, Huntingdon, UK) were injected intratibially with 2×105 RPMI-8226 cells suspended in phosphate-buffered saline. Mice were treated with the anti-JAM-A monoclonal antibody (Sigma-Aldrich, St. Louis, MO, USA, mouse monoclonal clone J10.4) recognizing the distal membrane extracellular domain of JAM-A. Twenty female 6- to 8-week-old NOD-SCID mice were injected subcutaneously (s.c.), into the right flank, with 1×107 RPMI-8226 cells suspended in 200 mL RPMI-1640 medium and 200 mL MatrigelTM as described previously16 and detailed in the Online Supplementary Methods.

Functional in vitro assays Wound-healing and MatrigelTM angiogenesis assays were performed as previously described and detailed in the Online Supplementary Methods.

Protein expression studies and polymerase chain reaction analyses Western blots, enzyme-linked immunosorbent assays, human angiogenesis array and real-time reverse transcriptase polymerase chain reactions were performed according to the manufacturers’ instructions (detailed in the Online Supplementary Methods).

Methods Immunohistochemistry and in silico analysis Patients Patients fulfilling the International Myeloma Working Group diagnostic criteria14 for NDMM (n=111), patients with RRMM15 1944

Details of the immunohistochemical studies and the in silico analysis, using the CoMMpass study dataset, are supplied in the Online Supplementary Methods. haematologica | 2021; 106(7)

Blocking JAM-A on BM endothelial cells in MM

Results

Statistical analysis A descriptive analysis was carried out using the median and interquartile range for the quantitative variables and percentage values for the qualitative ones. The normality of the distribution of data was tested using the Shapiro-Wilk test. The levels of JAMA expression on membrane MMEC, determined by mean fluorescence intensity (MFI) in fluorescence activated cell sorting (FACS), were dichotomized into two classes, JAM-Ahigh and JAM-Alow, choosing the median as the class boundary (Online Supplementary Methods). Moreover, for further confirmation, survival was analyzed using a model based on quartile ranges (Online Supplementary Methods).

Elevated JAM-A expression on bone marrow primary multiple myeloma endothelial cells correlates with poor prognosis in both newly diagnosed and relapsed/refractory multiple myeloma First, we compared JAM-A expression in MMEC and MGUS-derived endothelial cells (MGEC) (Figure 1A). JAM-A mRNA expression in MMEC significantly exceeded JAM-A levels in MGEC (1.8-fold difference, P<0.0001) and in endothelial cells from healthy subjects (Online Supplementary Figure S1A). Subsequent western blot analy-

Figure 1. Elevated JAM-A expression on bone marrow primary multiple myeloma endothelial cells in newly diagnosed patients correlates with poor overall survival. (A) Relative mRNA expression level of JAM-A of endothelial cells from patients with multiple myeloma (MMEC) (n=73) or monoclonal gammopathy of undetermined significance (MGEC) (n=73) by real-time reverse transcriptase polymerase chain reaction. ****P<0.0001, Mann-Whitney test. (B) Western blot densitometric analysis of basal protein expression of JAM-A of MGEC and MMEC lysates normalized to β-actin (MGEC from 24 patients with MGUS; MMEC from 24 patients with NDMM). Results are presented as mean ± standard deviation, ****P<0.0001, Mann-Whitney test. (C) FACS analysis of JAM-A cell surface expression from representative patient-derived MMEC, identified as CD45/138/38neg/31pos cells. (D) JAM-A overexpression colocalizes with higher vessel density on bone marrow biopsies. Vessel density (as highlighted by CD34 [red] staining) was higher in bone marrow spaces infiltrated by JAM-Ahigh (brown) neoplastic plasma cells, as compared to JAM-Alow cases. Magnification x 200. Scale bar=50 mm. (continued on the next page)

haematologica | 2021; 106(7)

1945

A.G. Solimando et al. E

Figure 1. (continued from the previous page). (E) Kaplan-Meier estimator of OS, by level of surface MMEC JAM-A expression. The median OS estimated in subjects with JAM-Alow MMEC cells at FACS was not reached whereas in subjects with JAM-Ahigh MMEC, the median OS was 78 months (HR=9.14, 95% CI: 2.8-29.76, P<0.0001; χ2LR=20.11; P<0.0001, upper panel). Uni- and multivariate analysis (lower panel). BM: bone marrow; MMEC: endothelial cells from patients with multiple myeloma; MGEC: endothelial cells derived from patients with monoclonal gammopathy of undetermined significance; NDMM: newly diagnosed multiple myeloma; Pts: patients. OS: overall survival; NR: not reached. R-ISS: Revised International Staging System; Hb: hemoglobin.

sis confirmed that JAM-A protein expression was significantly upregulated in MMEC in comparison to the levels in MGEC (P<0.0001) (Figure 1B). Because JAM-A had been previously proven to be a prominent adhesion molecule on MM cells,2 and is also known to form homophilic interactions,1 we investigated whether JAM-A expression in the vascular microenvironment affects disease outcome. To this end, we enrolled 312 patients, 111 with NDMM and 201 with RRMM. Employing flow cytometry on MMEC we divided the patients with NDMM on the basis of JAM-Ahigh and JAM-Alow MMEC surface expression (Figure 1C). Immunohistochemical analyses of BM trephines corroborated the findings (Figure 1D, Online Supplementary Figure S1B and C). Notably, overall survival was significantly shorter in patients with JAM-Ahigh MMEC than in patients with JAM-Alow MMEC: median not reached versus 78 months, respectively (hazards ratio [HR]=9.14, 95% con1946

fidence interval [95% CI]: 2.80-29.76;, P<0.001; χ2LR=20.11; P<0.0001) (Figure 1E upper panel). Strikingly, these results maintained significance also in the multivariate analysis (HR=9.11, 95% CI: 2.79-29.76; P<0.001) (Figure 1E lower panel). Concerning progression-free survival, only renal impairment displayed a significant impact in univariate as well as in multivariate analyses (HR=1.64, 95% CI: 1.09-2.47, P=0.017). The level of MMEC JAM-A expression did not influence risk of progression in NDMM (data not shown). Thus, JAM-A overexpression on MMEC represents a risk factor for shorter overall survival in patients with NDMM. Next, we performed flow cytometry on samples from a cohort of 201 patients with RRMM. Within these patients, JAM-Ahigh MMEC represented an independent poor prognostic factor for both overall survival and progression-free survival (Figure 2A and B). Survival differed significantly in patients according to the level of JAM-A expression on haematologica | 2021; 106(7)

Blocking JAM-A on BM endothelial cells in MM

Figure 2. Elevated JAM-A expression on bone marrow primary endothelial cells predicts poor prognosis in relapsed refractory multiple myeloma. Kaplan-Meier estimator of OS (A) and PFS (B), by level of surface MMEC JAM-A expression. (C) Cox model set on OS and PFS analyses. The median follow-up was 53 months (4-262 months) for OS and 23 months (1-119 months) for PFS. *Cox models adjusted for sex and age. **Cox stratified hazards regression by chronic kidney disease. OS: overall survival; PFS: progression-free survival; MMEC: bone marrow primary multiple myeloma endothelial cells. Pts: patients. NR: not reached. R-ISS: Revised International Staging System; Hb: hemoglobin.

MMEC. The median overall survival was 130 months in patients with JAM-Ahigh MMEC and was not reached in those with JAM-Alow MMEC (HR=2.96, 95% CI: 1.36– 6.37, P<0.006; χ2LR=8.52; P=0.0035) (Figure 2A). In patients with JAM-Alow MMEC cells, the estimated median progression-free survival was 27 months, while, in subjects with JAM-Ahigh MMEC the median progressionfree survival was only 18.3 months (HR=1.41, 95% CI: 1.05-1.88; P=0.019; χ2LR=5.78; P=0.0162) (Figure 2B). Multivariate analyses confirmed that JAM-Ahigh MMEC was an independent significant risk factor for low overall survival (HR=2.39, 95% CI: 1.09-5.28; P=0.030) in much the same way that Revised International Staging System (R-ISS) stage II (HR=5.34, 95% CI: 1.24-22.97; P=0.024) and stage III disease (HR=0.57, 95% CI: 1.25-34.54; P=0.026) and chronic kidney disease (HR=2.12, 95% CI: 1.00-4.52; P=0.049) were independent significant risk factors (Figure 2C). A Cox stratified model implemented for progression-free survival confirmed only high levels of membrane MMEC JAM-A as a statistically significant risk factor (HR=1.35, 95% CI: 1.00–1.81; P=0.044) stratified by chronic kidney disease (Figure 2C). Interestingly, only JAM-Ahigh MMEC remained significant in the multivariate model. Moreover, we found a significant association in the setting of RRMM between JAM-Ahigh MMEC and R-ISS stage II and III disease (χ2=17.4, P<0.0001) and the risk of extramedullary dishaematologica | 2021; 106(7)

semination (χ2=7.04, P=0.008). Thus, JAM-A surface expression on BM endothelial cells derived from MM patients exerted a strong and independent effect, with a linear trajectory, on overall survival in both cohorts and an additive poor prognostic impact on progression-free survival in the RRMM cohort. Additionally, we divided MMEC JAM-A surface expression of the entire cohort (Online Supplementary Figure S1D) into quartiles (JAMAQ1 to JAM-AQ4) and then compared the outcomes of patients in the lowest quartile (JAM-AQ1) to those of patients in the highest quartile (JAM-AQ4). Strikingly, overall survival differed significantly: the median overall survival was 88 months in JAM-AQ4 patients and was not reached in JAM-AQ1 patients (HR=8.24, 95% CI: 3.220.9, P<0.0001; χ2LR=28.15; P<0.0001). Interestingly, JAM-AQ4 MMEC maintained significance in the multivariate Cox-model (HR=6.36, 95% CI: 2.30-17.63; P<0.001). This comparison further corroborated the role of JAM-A-positive vs. JAM-A-negative MMEC in predicting poor clinical outcome in our cohort (Online Supplementary Figure S1E, upper and lower panels). The absence of a statistically significant impact on progression-free survival of patients with NDMM is likely due to a more pronounced effect of JAM-A MMEC expression in more advanced stages of MM. This suggests the importance of JAM-A within the BM microenvironment during disease progression. 1947

A.G. Solimando et al.

Figure 3. Bone marrow primary multiple myeloma endothelial cells enhance JAM-A expression on multiple myeloma cells. (A, B) Experimental design depicted at the top. RPMI-8226 cells were cultured alone or cocultured with MMEC at a 1:5 ratio (RPMI8266:MMEC) in inserted transwells and analyzed for JAM-A expression by western blotting (A) and flow cytometry (B). (C) Experimental design shown at top, RPMI-8226 cells were maintained for 24 h in CM from MMEC or MGEC. Cells were harvested and lysed and the extracted proteins immunoblotted for JAM-A expression. Overall densitometric analyses are reported. (D) RPMI-8226 cells were also analyzed by FACS after culture for 24 h in MGEC or MMEC CM. Results are presented as mean ± standard deviation (MGEC from 24 patients with MGUS; MMEC from 24 patients with NDMM), ****P<0.0001, MannWhitney test. BM: bone marrow; MMEC: endothelial cells from patients with multiple myeloma; MGEC: endothelial cells derived from patients with monoclonal gammopathy of undetermined significance; NDMM: newly diagnosed MM; CM: conditioned medium; MFI: mean fluorescence intensity.

1948

haematologica | 2021; 106(7)

Blocking JAM-A on BM endothelial cells in MM

Figure 4. Legend on next page.

haematologica | 2021; 106(7)

1949

A.G. Solimando et al. C

Figure 4. Pivotal role of JAM-A in multiple myeloma-associated angiogenesis in two- and three-dimensional conditions. (A) Upper panel. Confluent monolayers of MMEC underwent a scratch wound-healing assay. Three hours after scratching photographs were taken of MMEC that had been maintained in MMEC CM alone (CTRL) or previously supplemented for 12 h with sJAM-A at 100 ng/mL and treated with isotype control (+sJAM-A +ISO) or anti-JAM-A (+sJAM-A +a-JAM-A) blocking antibody. Lower panel: Migrating cells in each wound were counted. Counts of proliferating and migrating cells of six independent experiments. ****P<0.0001, Mann-Whitney test. (B) Photographs at 3 hours of newly-formed capillary networks after MMEC were seeded on a MatrigelTM layer. Direct comparison of MMEC in CM vs. MMEC treated with 100 ng/mL sJAM-A (upper left panel). Independent experiment to assess JAM-A inhibition in MMEC treated with an anti-JAM-A blocking antibody or isotype control antibody (upper right panel). Independent experiment to assess JAM-A knock-down in MMEC comparing treatment with JAM-A specific siRNA vs. non-specific scrambled siRNA without addition of sJAM-A (lower left panel). Independent experiment to assess the effect of blocking JAM-A after addition of 100 ng/mL of sJAM-A by comparing capillary formation after MMEC treatment with sJAM-A and anti-JAM-A blocking antibody vs. sJAM-A and isotype control antibody (lower right quadrant). Representative pictures of three biological replicates. Skeletonization of the meshes were analyzed and branching points measured. Data are normalized to control. Scale bar=100 μm. (C) Chorioallantoic membrane assay with the gelatin sponge loaded with MMEC CM alone (CTRL) or with MMEC CM supplemented with sJAM-A (+sJAM-A), in the presence or absence of 0.5 mg/mL anti-JAM-A monoclonal antibody. On day 12, pictures were taken in ovo. One representative experiment is shown at 50X magnification. Newly formed vessels were counted. Mann-Whitney test. (D) An array of 55 human angiogenesis-related proteins was performed on MMEC CM after sJAM-A treatment without and with blocking with the anti-JAM-A monoclonal antibody. Array spots were analyzed with ImageJ Lab v. 1.51 software and normalized to positive control signal intensities. Graph bars represent the pixel density of the detected angiogenesis-related cytokines in two independent experiments. Values are expressed as mean ± standard deviation of ten independent experiments. Mann-Whitney test. *P<0.05; ****P<0.0001, versus SFM as control. See Online Supplementary Figure S3 and the main text for more details. MMEC: bone marrow primary multiple myeloma endothelial cells; a-JAM-A: monoclonal antibody against JAM-A; sJAM-A: soluble JAM-A; CM: conditioned medium; CTRL: control; SFM: serum-free medium; n.s.: not significant. CAM: chick chorioallantoic membrane. ADAMTS1: human metalloproteinase with thrombospondin type 1 motifs; PLG: plasminogen; FGF-2: fibroblast growth factor-2; IGFBP1: insulinlike growth factor binding protein 1; IL8: interleukin-8; TIMP1: tissue inhibitor matrix metalloprotease 1; VEGFA: vascular endothelial growth factor A; VEGFC: vascular endothelial growth factor C.

Multiple myeloma endothelial cells enhance JAM-A expression on multiple myeloma plasma cells To address how interactions with MMEC functionally influence MM-cell biology, we performed indirect and direct co-culture experiments of MMEC with MM-cell lines. JAM-A expression levels increased on MM-cells when co-cultured with MMEC (Figure 3A and B). We next exposed MM-cells to primary MMEC- or MGEC-derived culture media. JAM-A protein expression was higher on MM-cells after exposure to MMEC medium than after exposure to MGEC medium (Figure 3C and D, respectively). Results confirmed JAM-A upregulation upon direct coculture experiments (Online Supplementary Figure S2A). Notably, only after direct co-culture did MMEC recapitulate the same behavior as that of MM-cells (Online 1950

Supplementary Figure S2B). Consistently, sJAM-A levels also increased after co-culture of MM-cells with MMEC (Online Supplementary Figure S2C). Similarly to RPMI-8226 cells, OPM-2 cells upregulated JAM-A after direct co-culture with MMEC (Online Supplementary Figure S2D), but not after indirect culture (data not shown). These data indicate that both cell-cell contact and soluble factors released by MMEC into the BM microenvironment upregulated JAM-A expression on MM-cells. MMEC JAM-A upregulation parallels this dynamic process, suggesting a vicious cycle, promoting MM growth by supporting angiogenesis.

JAM-A enhances angiogenesis in two- and three-dimensional conditions We hypothesized that JAM-A upregulation during the haematologica | 2021; 106(7)

Blocking JAM-A on BM endothelial cells in MM

Figure 5. JAM-A inhibition reduces myeloma proliferation and vasculature in an intratibial in vivo multiple myeloma model. NOD/SCID mice (n=20) bearing RPMI8226 intratibial xenografts were repeatedly treated with a JAM-A blocking monoclonal antibody (a-JAM-A) or isotope control IgG (ISO) for 3 weeks. (A) Upper panel. From left to the right. Ki67/CD138 and JAM-A staining: CD138 and JAM-A (red) reactivity appears to be more represented on the smaller neoplastic plasma cells, whereas the more pleomorphic/anaplastic component shows less reactivity; the opposite staining distribution is observed for Ki67 nuclear staining (brown), which is more prominent in the larger cells. CD31/JAM-A double and CD31 staining (brown) highlight endothelia-lined thin-walled microvessels; lumina appear to be only slightly dilated. JAM-A (red) stains a fraction of neoplastic plasma cells, with a cytoplasmic pattern. Lower panel. Decreased Ki67 expression in specimens treated with anti-JAM-A. Within the CD31-stained non-involved bone marrow lacunae (see CD31+ megakaryocytes) from the anti-JAM-A treated group the vessels are more distended, and endothelia display a thin, inconspicuous cytoplasmic rim. (B) From left to the right, differences in terms of MM proliferation, JAM-A, CD31 positivity on endothelial cells and vessel counts, assessed by two pathologists. Data shown are mean ± standard deviation from ten individual mice for each group. ****P<0.0001 versus controls, Mann-Whitney test. Scale bar=100 mm.

progression of MM may enhance angiogenesis. To study this, we treated MMEC with increasing concentrations of human recombinant sJAM-A and measured different parameters of angiogenesis.17,18 To examine whether JAMA directly affects spontaneous MMEC migration, we performed experiments in two-dimensional (2D) and threedimensional (3D) environments. Enhanced spontaneous MMEC migration was observed after 12 h of sJAM-A treatment in a 2D scratch assay in which migrating MMEC were counted (Figure 4A, upper and lower panels). Blocking JAM-A abolished the enhanced MMEC migration (Figure 4A, upper panel) and reduced the numbers of migrating MMEC (Figure 4A, lower panel). In a 2D angiogenesis assay, sJAM-A treatment increased endothelial structural complexity in terms of branching points and haematologica | 2021; 106(7)

vessel length, which are parameters of angiogenesis. Three hours after seeding, sJAM-A treatment resulted in a structured capillary network, while the control remained in a rudimentary stage of organization with small clumps of cells distributed on the MatrigelTM layer (Figure 4B, upper left quadrant). Furthermore, we tested the effect of JAM-A inhibition in MMEC by both siRNA and an antibody blocking JAM-A without adding sJAM-A (Figure 4B, lower left and upper right quadrants, respectively). Consistently, blocking JAM-A with a monoclonal antibody impaired the capillary network formation and resulted in poorly skeletonized structures (Figure 4B, lower right quadrant). The observed down-modulation of MMEC migration, reduced number of branching points and shorter vessel length occurred independently of cytotoxic 1951

A.G. Solimando et al.

Figure 6. JAM-A inhibition restricts angiogenesis and tumor growth in subcutaneous multiple myeloma xenograft model. NOD/SCID mice (n=20) bearing RPMI-8226 subcutaneous xenografts were repetitiously treated with a JAM-A blocking monoclonal antibody (α-JAM-A), or isotope control IgG (ISO) or with vehicle only for 40 days for 3 days/week. (A) Immunohistochemistry staining: JAM-A (red) reactivity is stronger in the smaller neoplastic cells whereas it is lower in the more pleomorphic/anaplastic components. The opposite staining distribution is observed for Ki67 nuclear staining (brown), which is more clear-cut in the larger cells. CD31 staining shows focal positivity in the control group and is absent in the group treated with the JAM-A blocking antibody. (B) Treatment was continued for 3 days/week for 40 days and tumor volumes were measured every 2 days with a caliper. (C) Hemoglobin values, Ki-67 positivity, vessel area and number of vessels expressed as mean ± standard deviation of three independent experiments. *P<0.05 versus vehicle-treated control. Scale bar=50 mm.****P<0.0001 versus controls; Mann-Whitney test.

effects, since JAM-A neutralization did not affect MMEC survival (Online Supplementary Figure S3A, left and right panels). Based on the 2D observations, we investigated whether JAM-A could influence structured MM-associated angiogenesis in a 3D CAM assay. CAM were implanted with gelatin sponges imbued with either MMEC conditioned medium as the control (CTRL) or MMEC conditioned medium with sJAM-A (+sJAM-A), in the presence or absence of a monoclonal antibody blocking JAM-A. MMEC conditioned medium stimulated new vessel formation in CAM,19 and this effect was markedly enhanced by the addition of sJAM-A. The effect could be selectively inhibited by treatment with a sJAM-A blocking antibody as treatment with a cocktail containing an isotype IgG1 control antibody (sJAM-A + ISO, middle panel) did not reduce vessel formation (Figure 4C). To explore potential factors that enhance JAM-A-mediated MM angiogenesis, we compared conditioned media 1952

from MMEC supplemented with sJAM-A before and after treatment with the JAM-A blocking antibody with an angiogenesis array (Figure 4D, Online Supplementary Figure S3B). sJAM-A strongly reduced anti-angiogenic and increased pro-angiogenic factors secreted by MMEC,16 such as plasminogen (PLG), fibroblast growth factor 2 (FGF-2), insulin-like growth factor binding protein 1 (IGFBP1) and vascular endothelial growth factors A and C (VEGFA, VEGFC). Reverse transcriptase polymerase chain reaction analysis corroborated the proteomic findings and revealed sJAM-A-induced transcriptional upregulation of these factors and ligands (PLG and ENO1, JAM-A with LFA-1 and TJP1) (Online Supplementary Figure S3C-L). Moreover, because JAM-A can form homophilic interactions with JAM-A itself1 as well as heterophilic interactions with LFA-1, TJP1, CAV1 and CASK, we investigated whether the expression of these ligands correlated with MM-MMEC interactions. Direct co-culture of RPMI-8226 cells and MMEC significantly increased LFA-1 and CAV1 haematologica | 2021; 106(7)

Blocking JAM-A on BM endothelial cells in MM

C C

D D

Figure 7. JAM-A boosts multiple myeloma-related angiogenesis in the bone marrow microenvironment. (A) Molecular interactions between MM plasma cells and MMEC: cell-adhesion mediated changes via trans-homo/heterophilic JAM-A interactions. (B) FGF-2 mediates the pro-angiogenic and proliferative roles of JAM-A and the release of monomeric JAM-A from the ternary complex through an unknown mechanism. We speculate that once monomeric JAM-A is available at a membrane level, it forms homodimers that mediate downstream signaling and is also susceptible for cleavage and shedding via ADAM17. (C) JAM-A-mediated cytoskeleton rearrangement via TJP1 downregulation and cell function modification. Depending on myeloma cell-mediated interactions, the endothelial cells can lose their tight junction and thus vascular permeability can increase. (D) An interaction between PLG and ENO1 triggers angiogenesis and microenvironmental modifications via JAM-A. Caveolin1 modulates the transport of cytosolic ENO-1 to the cell surface. α-JAM-A: monoclonal antibody against JAM-A. BM: bone marrow; MMEC: bone marrow primary multiple myeloma endothelial cells; FGF-2: fibroblast growth factor-2; ADAM17: ADAM metallopeptidase domain 17; TJP1: tight junction protein-1; PLG: plasminogen; ENO1: enolase 1; LFA-1: lymphocyte function-associated antigen 1; MAPK: mitogen-activated protein kinase; avβ3: integrin alpha V beta 3; CD9: CD9 molecule; VEGFA: vascular endothelial growth factor A; ADAMTS1: human metalloproteinase with thrombospondin type 1 motifs. AURKA: aurora kinase A.

on MM-cells, whereas TJP1, CASK and ADAM17 expression levels decreased (Online Supplementary Figure S3G-K). We therefore investigated whether the induced gene expression was non-random, by studying other molecules involved in neoplastic angiogenesis processes,12 namely VEGFA, VEGFC, HGF, FGF16 and Aurora kinase A (AURKA);20 also in this case we found significant VEGFA and AURKA gene upregulation after MM-MMEC co-culture (Online Supplementary Figure S3F and L). These data support the concept that MM-MMEC interactions enhance angiogenesis. Thus, we investigated whether MM cells participate actively in the angiogenesis program in a reciprocal interaction with the BM microenvironment in patients and whether a pro-angiogenic gene signature could identify patients with worse progressionfree and overall survival. We therefore studied 646 NDMM patients enrolled in the CoMMpass trial, comparing two different cohorts, based on survival outcome (alive vs. dead for overall survival and progressed vs. ongoing disease for progression-free survival) performing a supervised analysis based on the gene expression of the pro-angiogenic factors contained in the angiogenesis array and other well-known JAM-A interactors. Strikingly, these two cohorts differed significantly: JAM-A, ENO-1, VEGFA and AURKA were all overexpressed in patients experiencing shorter progression-free and overall survival. Conversely, reduced TJP1 expression in patients correlated with poor survival (Online Supplementary Table S3). These haematologica | 2021; 106(7)

gene expression data confirmed the protein expression results from our cohort of patients. Exogenous JAM-A modulated the secretory profile of MMEC, favoring angiogenesis, and highlighted the tight connection with an angiogenic environment, comprising key angiogenic factors, such as PLG, FGF-2, IGFBP1, VEGFA and VEGFC.

JAM-A inhibition impairs angiogenesis and inhibits tumor growth in vivo To investigate whether JAM-A inhibition may affect in vivo angiogenesis and in turn MM-cell growth, we employed two different mouse models. To mimic advanced MM21 we injected RPMI-8226 cells intratibially into NOD-SCID mice and analyzed bone specimens after anti-JAM-A treatment.2 Blocking JAM-A reduced MM-cell proliferation and angiogenesis (Figure 5A). The difference was statistically significant regarding numbers and percentages of Ki-67high proliferating MM-cells (79.87±1.242 and 35.38±0.3455 in the groups treated with isotype control and anti-JAM-A, respectively; P<0.0001) and vessels/mm2 field (9.3 and 7.1 in the groups treated with isotype control and anti-JAM-A, respectively P<0.0001) (Figure 5B). The group treated with the antibody against JAM-A expressed lower JAM-A levels (79.78±1.443 and 36.98±0.466 in the groups treated with isotype control and anti-JAM-A, respectively; P<0.0001), lower CD31% (5.58±1.34 and 3.48±0.646 in the groups treated with isotype control and anti-JAM-A, respectively; P<0.0001) and 1953

A.G. Solimando et al.

lower vessel density (calculated as number of vessels/mm2) than the group treated with the isotype control, which had a higher number of vessels with well-lit lumina (9.77±2,63 and 6.48±0,631 in the groups treated with isotype control vs. anti-JAM-A, respectively; P<0.0001). To assess the activity of the anti-JAM-A blocking antibody on angiogenesis on a solitary plasmacytoma in vivo and to monitor MM-cell growth at extra-osseous sites noninvasively with a caliper, we employed a subcutaneous MM xenograft model. This approach enables dissection of the endothelial bystander effect on BM-independent extramedullary MM.21 Thus, we employed a second in vivo xenograft model engrafting RPMI-8226 cells subcutaneously into the flanks of NOD/SCID mice.21 The animals were randomized at day 3 after engraftment and treated with either anti-JAM-A monoclonal antibody or a non-specific isotype control antibody intraperitoneally for 3 days/week for 40 days. Subsequently, we measured the vascular area, tumor volume and hemoglobin content of the MM mass. Blocking JAM-A reduced the vascular area in the soft tissue MM masses compared to that in animals treated with an isotype control (difference between medians -0.015; P<0.0001). No adverse events occurred upon continuous anti-JAM-A treatment. Notably, after 40 days, the vascular area increased significantly in tumors and MM disease progressed more in controls than in the group treated with the monoclonal antibody against JAM-A (Figure 6A, CD31 staining, and 6B). In isotype-treated control mice, tumors grew exponentially, contrasting with the only limited tumor growth in antiJAM-A-treated animals (Figure 6B, Online Supplementary Figure S4A). Lower hemoglobin content confirmed poor MM vascularization in the anti-JAM-A-treated mice (8.4±0.04 in the isotype-treated control mice vs. 5.5±0.04 in the anti-JAM-A-treated group; P<0.0001, 95% CI: -3.02 to -2.8 (Figure 6C). Ki-67-staining, vascular area and vessel count confirmed that blocking JAM-A strongly reduced MM vascularity and disease progression. Furthermore, JAM-A blocking significantly reduced pro-angiogenic factors such as FGF-2 and VEGF-A in the peripheral blood of MM-bearing mice (Online Supplementary Figure S4B-D).

Discussion Angiogenic switching is a key process during transition from premalignant asymptomatic MGUS to full-blown MM. Angiogenic parameters in the BM at the time of diagnosis were widely considered to be able to predict MM progression.22 In solid tumors, such as breast, lung, head and neck, and brain cancers, JAM-A activation promotes tumor progression, while its inhibition by anti-JAM-A2 agents reduces tumor growth.11 We demonstrated in four independent experimental settings that JAM-A essentially stimulates MM-associated angiogenesis. In the CAM assay, a monoclonal antibody neutralizing JAM-A caused a strong reduction of the number of vessels, implying that JAM-A exerts an essential angiogenic stimulus that could not be replaced by any other compensating factor contained in the MMEC conditioned medium.23 Our new findings pinpoint JAM-A as an attractive target in MM patients. JAM-A appears pivotal in MM evolution, which can be explained by several angiogenic mechanisms.24,25 First, we demonstrated significantly higher JAM-A levels on MMEC from NDMM patients than on MGEC. Furthermore, we 1954

could link the high JAM-A surface expression on MMEC with a significantly shorter overall survival in both NDMM and RRMM and, at even more advanced disease stages, higher JAM-A expression levels also correlated with reduced progression-free survival. We therefore examined the pathophysiological basis responsible for favoring MM progression. As already described for the HGF/cMET axis,26,27 JAM-A acts within the BM microenvironment, sustaining the neoplastic clone and promoting MM-related angiogenesis both directly and indirectly by priming MMEC. Thus, JAM-A and its soluble isoform sJAM-A appear to feed into a vicious cycle involving MMEC, generating a malignant environment favorable for MM progression. Although JAM-A is expressed in several solid cancers,3 to our knowledge this is the first report of the role of endothelial JAM-A expression in the MM tumor microenvironment. Homophilic interactions between recombinant sJAM-A and membrane JAM-A have been demonstrated biochemically.10 Homophilic JAM-A interactions can be inhibited by an anti-JAM-A monoclonal antibody that binds to an epitope close to the N-terminus of the mature protein10 as well as by a peptide that corresponds to the Nterminal 23 residues of the mature protein.28 This suggests that the homophilic trans-interaction is mediated through the membrane-distal V-type Ig-like JAM-A domain at the N-terminus of the molecule. Targeting this domain of the JAM-A molecule on MMEC in our in vitro co-culture systems suggested that this type of interaction mediates the MM-MMEC crosstalk. In line with previous reports about MMEC sustaining MM growth,29,30 our disease models showed that during the transition from the pre-angiogenic to the angiogenic phase, proliferation of tumor cells and neovascularization intensely involve over-expression of JAM-A on MMEC. MMEC were responsive to the presence of sJAM-A in the surrounding microenvironment, which increased their JAM-A surface protein expression. sJAM-A directly and indirectly upregulated JAM-A on the bystander MM-cells, independently of their basal JAM-A expression status. These observations support the concept that cellular components of MM BM, including MMEC, can release JAM-A to sustain disease progression and prepare a tumor-"friendly" niche, exerting significant modulation on FGF-2, VEGF-A and PLG/ENO1 downstream effects. JAM-A has been described to interact with CD9, a wellknown driver of MM-related drug resistance31 and clinical prognosis.32 We found significant expression of FGF-2, a potent stabilizer and activator of a ternary complex involving JAM-A, CD9 and avβ3 integrin, a novel potential therapeutic target.33 Peddibhotla et al. described that the aggregation of this ternary complex can activate downstream pathway cascades to induce proliferation, migration and an angiogenic stimulus to endothelial cells.34 Our in silico validation shed more light on this pathophysiological process. ENO1 encodes a-enolase, which in the cytoplasm works as a plasminogen receptor and has been described to show upregulated membrane expression in several types of cancer.35,36 Of note, plasminogen upregulation had been correlated with tumor invasion and angiogenesis;37 its activation, derived from the interaction with a-enolase, prompted activation of downstream signaling such as the MEK-ERK pathway, which was able to promote cell invasion and angiogenesis further. a-enolase can also modulate antitumor immune responses. Cappello et al. described that aenolasehigh myeloid-derived suppressor cells could not haematologica | 2021; 106(7)

Blocking JAM-A on BM endothelial cells in MM

adhere to tumor necrosis factor-a-primed endothelial cells in the presence of an anti-ENO-1 monoclonal antibody. Consequently, decreased migration of the myeloid-derived suppressor cells reduced in situ immunosuppression, enhancing T-cell-mediated immunity against malignant cells.38,39 We found that JAM-A overexpression in MMEC strongly correlated with the expression of ADAMTS1, a regulator of angiogenesis40 and immune-surveillance,41 which appears to play a central role in preparing a favorable BM milieu.42 We also identified that the expression of JAM-A on the surface of MMEC was inversely correlated with ADAM17 expression. Conversely, sJAM-A release correlated directly with ADAM17 upregulation, a mechanism described for endothelial cells in inflammation.9 ADAM17 upregulation has also been observed in MM in the context of fractalkine release,43 which identifies this system as a potential novel therapeutic target in MM patients to disrupt a vicious circle enhancing the MM niche. JAM-A levels also correlated strongly with AURKA levels in MM patients. This finding may link JAM-A-mediated cell adhesion to MM resilience and drug resistance.20 Indeed, proteasome engulfment-derived proteotoxicity44 and invasiveness through epithelial-mesenchymal-transition and cell adhesion45 are complex biological events that affect prognosis.46 We also demonstrated that the interaction between MMEC and MM-cells affects the expression of JAM-A and other fundamental molecules, such as TJP1 and LFA-1. This reciprocal “education” parallels the invasive behavior of the MM-cells towards the endothelial counterpart, instructing the vasculature to interact actively with the malignant cells, potentially driving their survival and drug resistance.47,48 In line with this, increased JAM-A endothelial levels correlated strongly with unfavorable and resistant MM stages such as high R-ISS disease stages and the risk of extramedullary development. A network of interactions between JAM-A and a-enolase49 emphasizes the strong communication with the MM niche environment to allow persistence and sustaining proliferative signaling. Therefore, JAM-A may represent a key factor in the nurturing substrate50 supporting the evolution of MM. Moschetta et al.13 previously described the interderpendency of endothelium and MM-cells: endothelial progenitor cell trafficking enhances MM progression, particularly at an early disease stage. Rajkumar et al. highlighted a progressive increase in BM angiogenesis along the spectrum of plasma cell disorders from MGUS to advanced MM.12 Integrating the prognostic relevance of JAM-A expressed by MMEC and our experimental data led us to propose JAMA as a key player in coordinating the interactions with the MM milieu enabling a permissive BM ecosystem during the aggressive disease evolution from NDMM to RRMM. Indeed, the anti-MM effect of blocking JAM-A also relies on a complex antiangiogenic effect, especially in critical transition phases of MM progression, such as the passage from MGUS to symptomatic MM and from responsive to drug-refractory disease, interfering with a main proangiogenic factor and with MM-cell proliferation. Our 2D and 3D models showed that, mechanistically, JAM-A drives MM-associated angiogenesis via a homophilic interaction and through identified downstream targets, namely FGF2, VEGF-A and PLG/ENO1, in the BM microenvironment. Moreover, the clinical impact demonstrated in a large cohort of consecutive individuals pinpoints the JAM-A axis haematologica | 2021; 106(7)

as a new player in MM-associated angiogenesis able to refine the prognostic stratification of patients, especially those with more advanced disease. The close link between MM and the BM microenvironment appears paradigmatic for MM evolution and disease progression. We connected the interaction of MMEC with MM-cells via the adhesion molecule JAM-A. Our data point towards a vicious cycle of JAM-A overexpression on MMEC reflected by a higher JAM-A expression on the tumoral counterpart. Shed from the cell surface, sJAM-A enhances the establishment of homophilic JAM-A complexes fostering MM niche formation. Finally, our results may lead to the development of JAM-A-based therapeutic strategies directed against MM-interactions with the tumor microenvironment (Figure 7A-D). Clearly, these findings need to be confirmed in a larger population of patients in a carefully designed, prospective, clinical study.

Disclosures No conflicts of interest to disclosure. Contributions AGS, MCDV, VR, HE, AV and ABe designed and performed research, analyzed and interpreted data and wrote the manuscript; AGS, MCDV, PL, GC, PTG, GDL, ABr, EH, HR-W, DR performed research and analyzed data; AGS, MCDV, AA, PB, GC, FPB, AM, SDS, TS, ABa, MAF, WK, AR and ST analyzed data; AGS, HE, RR, LR, PD, KMK, VR and AV provided patients’ samples; AGS, MCDV, CT, MC, VR, AV, HE and ABe designed research, interpreted data, and edited the manuscript; all authors revised the manuscript and approved its submission. VR and HE contributed equally to this study. AV and ABe contributed equally to the conception and design of this study. Acknowledgments The authors acknowledge the Multiple Myeloma Research Foundation for providing an updated and comprehensive real-life MM dataset for the international scientific community. The in silico analysis and the relative clinical correlation were generated as part of the Multiple Myeloma Research Foundation Personalized Medicine Initiative. We thank the members of the laboratories of Beilhack and Vacca for lively discussions. We also thank Drs. Julia Delgado Tascón, Claudia Siverino and Marco Metzger, and Mss Katharina Schmiedgen, Charlotte Botz-Von Drathen, Hannah Manz and Vittoria Musci for technical support and valuable discussions. Public consulted datasets are available at https://research.mmrf.org, CoMMpass longitudinal, prospective observational study (release IA12). Funding This work was supported by the Bayerische Forschungsstiftung consortium FortiTher (WP2TP3), the Deutsche mBone consortium (2084/1, Forschungsgemeinschaft 401253051), by the Italian Association for Cancer Research (AIRC) through an Investigator Grant (n. 20441 to VR), by the GLOBALDOC Project - CUP H96J17000160002 approved with A.D. n. 9 from Puglia Region, financed under the Action Plan for Cohesion approved with Commission decision C (2016) 1417 to AGS. This research project was supported in part by the Apulian Regional Project “Medicina di Precisione” to AGS. AR was supported by the Wilhelm Sander-Stiftung (grant n. 2014.903.1).

1955

A.G. Solimando et al.

References 1. Steinbacher T, Kummer D, Ebnet K. Junctional adhesion molecule-A: functional diversity through molecular promiscuity. Cell Mol Life Sci. 2018;75(8):1393-1409. 2. Solimando AG, Brandl A, Mattenheimer K, et al. JAM-A as a prognostic factor and new therapeutic target in multiple myeloma. Leukemia. 2018;32(3):736-743. 3. Leech AO, Cruz RGB, Hill ADK, Hopkins AM. Paradigms lost-an emerging role for over-expression of tight junction adhesion proteins in cancer pathogenesis. Ann Transl Med. 2015;3(13):184. 4. Severson EA, Parkos CA. Mechanisms of outside-in signaling at the tight junction by junctional adhesion molecule A. Ann N Y Acad Sci. 2009;1165:10-18. 5. Ebnet K, Schulz CU, Meyer Zu Brickwedde MK, Pendl GG, Vestweber D. Junctional adhesion molecule interacts with the PDZ domain-containing proteins AF-6 and ZO-1. J Biol Chem. 2000;275(36):27979-27988. 6. Ebnet K. Junctional adhesion molecules (JAMs): cell adhesion receptors with pleiotropic functions in cell physiology and development. Physiol Rev. 2017;97(4): 1529-1554. 7. Terral G, Champion T, Debaene F, et al. Epitope characterization of anti-JAM-A antibodies using orthogonal mass spectrometry and surface plasmon resonance approaches. mAbs. 2017;9(8):1317-1326. 8. Vellanki SH, Cruz RGB, Jahns H, et al. Natural compound tetrocarcin-A downregulates junctional adhesion molecule-A in conjunction with HER2 and inhibitor of apoptosis proteins and inhibits tumor cell growth. Cancer Lett. 2019;440-441:23-34. 9. Koenen RR, Pruessmeyer J, Soehnlein O, et al. Regulated release and functional modulation of junctional adhesion molecule A by disintegrin metalloproteinases. Blood. 2009;113(19):4799-4809. 10. Bazzoni G, Martinez-Estrada OM, Orsenigo F, Cordenonsi M, Citi S, Dejana E. Interaction of junctional adhesion molecule with the tight junction components ZO-1, cingulin, and occludin. J Biol Chem. 2000;275(27):20520-20526. 11. Murakami M, Francavilla C, Torselli I, et al. Inactivation of junctional adhesion molecule-A enhances antitumoral immune response by promoting dendritic cell and T lymphocyte infiltration. Cancer Res. 2010;70(5):1759-1765. 12. Rajkumar SV, Mesa RA, Fonseca R, et al. Bone marrow angiogenesis in 400 patients with monoclonal gammopathy of undetermined significance, multiple myeloma, and primary amyloidosis. Clin Cancer Res. 2002;8(7):2210-2216. 13. Moschetta M, Mishima Y, Kawano Y, et al. Targeting vasculogenesis to prevent progression in multiple myeloma. Leukemia. 2016;30(5):1103-1115. 14. 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):e538-548. 15. Rajkumar SV, Harousseau J-L, Durie B, et al. Consensus recommendations for the uniform reporting of clinical trials: report of the International Myeloma Workshop Consensus Panel 1. Blood. 2011; 117(18):4691-4695. 16. Rao L, De Veirman K, Giannico D, et al. Targeting angiogenesis in multiple myeloma by the VEGF and HGF blocking DARPin® protein MP0250: a preclinical

1956

study. Oncotarget. 2018;9(17):13366-13381. 17. Jridi I, Catacchio I, Majdoub H, et al. The small subunit of Hemilipin2, a new heterodimeric phospholipase A2 from Hemiscorpius lepturus scorpion venom, mediates the antiangiogenic effect of the whole protein. Toxicon. 2017;126:38-46. 18. Nowak-Sliwinska P, Alitalo K, Allen E, et al. Consensus guidelines for the use and interpretation of angiogenesis assays. Angiogenesis. 2018;21(3):425-532. 19. Ribatti D, Nico B, Vacca A, Presta M. The gelatin sponge-chorioallantoic membrane assay. Nat Protoc. 2006;1(1):85-91. 20. Qin Y, Zhang S, Deng S, et al. Epigenetic silencing of miR-137 induces drug resistance and chromosomal instability by targeting AURKA in multiple myeloma. Leukemia. 2017;31(5):1123-1135. 21. Mitsiades CS, Anderson KC, Carrasco DR. Mouse models of human myeloma. Hematol Oncol Clin North Am. 2007; 21(6):1051-1069, viii. 22. Rajkumar SV, Leong T, Roche PC, et al. Prognostic value of bone marrow angiogenesis in multiple myeloma. Clin Cancer Res. 2000;6(8):3111-3116. 23. Ria R, Catacchio I, Berardi S, et al. HIF-1a of bone marrow endothelial cells implies relapse and drug resistance in patients with multiple myeloma and may act as a therapeutic target. Clin Cancer Res. 2014;20(4): 847-858. 24. Fukuhara T, Kim J, Hokaiwado S, et al. A novel immunotoxin reveals a new role for CD321 in endothelial cells. PloS One. 2017; 12(10):e0181502. 25. Stamatovic SM, Sladojevic N, Keep RF, Andjelkovic AV. Relocalization of junctional adhesion molecule A during inflammatory stimulation of brain endothelial cells. Mol Cell Biol. 2012;32(17):3414-3427. 26. Moschetta M, Basile A, Ferrucci A, et al. Novel targeting of phospho-cMET overcomes drug resistance and induces antitumor activity in multiple myeloma. Clin Cancer Res. 2013;19(16):4371-4382. 27. Ferrucci A, Moschetta M, Frassanito MA, et al. A HGF/cMET autocrine loop is operative in multiple myeloma bone marrow endothelial cells and may represent a novel therapeutic target. Clin Cancer Res. 2014;20(22):5796-5807. 28. Babinska A, Kedees MH, Athar H, et al. F11receptor (F11R/JAM) mediates platelet adhesion to endothelial cells: role in inflammatory thrombosis. Thromb Haemost. 2002;88(5):843-850. 29. Cooke VG, Naik MU, Naik UP. Fibroblast growth factor-2 failed to induce angiogenesis in junctional adhesion molecule-A-deficient mice. Arterioscler Thromb Vasc Biol. 2006;26(9):2005-2011. 30. Leone P, Di Lernia G, Solimando AG, et al. Bone marrow endothelial cells sustain a tumor-specific CD8+ T cell subset with suppressive function in myeloma patients. Oncoimmunology. 2019;8(1):e1486949. 31. Hu X, Xuan H, Du H, Jiang H, Huang J. Down-regulation of CD9 by methylation decreased bortezomib sensitivity in multiple myeloma. PloS One. 2014;9(5):e95765. 32. De Bruyne E, Bos TJ, Asosingh K, et al. Epigenetic silencing of the tetraspanin CD9 during disease progression in multiple myeloma cells and correlation with survival. Clin Cancer Res. 2008;14(10):29182926. 33. Wallstabe L, Mades A, Frenz S, Einsele H, Rader C, Hudecek M. CAR T cells targeting avβ3 integrin are effective against advanced cancer in preclinical models. Adv Cell Gene Ther. 2018;1(2):e11.

34. Peddibhotla SSD, Brinkmann BF, Kummer D, et al. Tetraspanin CD9 links junctional adhesion molecule-A to avβ3 integrin to mediate basic fibroblast growth factor-specific angiogenic signaling. Mol Biol Cell. 2013;24(7):933-944. 35. Tu S-H, Chang C-C, Chen C-S, et al. Increased expression of enolase alpha in human breast cancer confers tamoxifen resistance in human breast cancer cells. Breast Cancer Res Treat. 2010;121(3):539553. 36. Yu Y-Q, Wang L, Jin Y, et al. Identification of serologic biomarkers for predicting microvascular invasion in hepatocellular carcinoma. Oncotarget. 2016;7(13):1636216371. 37. Rossignol P, Ho-Tin-Noé B, Vranckx R, et al. Protease nexin-1 inhibits plasminogen activation-induced apoptosis of adherent cells. J Biol Chem. 2004;279(11):10346-10356. 38. Cappello P, Tonoli E, Curto R, Giordano D, Giovarelli M, Novelli F. Anti-α-enolase antibody limits the invasion of myeloid-derived suppressor cells and attenuates their restraining effector T cell response. Oncoimmunology. 2016;5(5):e1112940. 39. Castella B, Foglietta M, Riganti C, Massaia M. Vγ9Vd2 T cells in the bone marrow of myeloma patients: a paradigm of microenvironment-induced immune suppression. Front Immunol. 2018;9:1492. 40. Casal C, Torres-Collado AX, Plaza-Calonge MDC, et al. ADAMTS1 contributes to the acquisition of an endothelial-like phenotype in plastic tumor cells. Cancer Res. 2010;70(11):4676-4686. 41. Hope C, Foulcer S, Jagodinsky J, et al. Immunoregulatory roles of versican proteolysis in the myeloma microenvironment. Blood. 2016;128(5):680-685. 42. Li J, Zou K, Yu L, et al. MicroRNA-140 inhibits the epithelial-mesenchymal transition and metastasis in colorectal cancer. Mol Ther Nucleic Acids. 2018;10:426-437. 43. Marchica V, Toscani D, Corcione A, et al. Bone marrow CX3CL1/fractalkine is a new player of the pro-angiogenic microenvironment in multiple myeloma patients. Cancers. 2019;11(3):321. 44. Savitski MM, Zinn N, Faelth-Savitski M, et al. Multiplexed proteome dynamics profiling reveals mechanisms controlling protein homeostasis. Cell. 2018;173(1):260-274.e25. 45. Ren B-J, Zhou Z-W, Zhu D-J, et al. Alisertib induces cell cycle arrest, apoptosis, autophagy and suppresses EMT in HT29 and Caco-2 cells. Int J Mol Sci. 2015;17(1):41. 46. Noll JE, Vandyke K, Hewett DR, et al. PTTG1 expression is associated with hyperproliferative disease and poor prognosis in multiple myeloma. J Hematol Oncol. 2015;8:106. 47. Waldschmidt JM, Simon A, Wider D, et al. CXCL12 and CXCR7 are relevant targets to reverse cell adhesion-mediated drug resistance in multiple myeloma. Br J Haematol. 2017;179(1):36-49. 48. Zhang X-D, Baladandayuthapani V, Lin H, et al. Tight junction protein 1 modulates proteasome capacity and proteasome inhibitor sensitivity in multiple myeloma via EGFR/JAK1/STAT3 signaling. Cancer Cell. 2016;29(5):639-652. 49. Ramroop JR, Stein MN, Drake JM. Impact of phosphoproteomics in the era of precision medicine for prostate cancer. Front Oncol. 2018;8:28. 50. Mundy GR. Metastasis to bone: causes, consequences and therapeutic opportunities. Nat Rev Cancer. 2002;2(8):584-593.

haematologica | 2021; 106(7)

ARTICLE

Plasma Cell Disorders

Lenalidomide before and after autologous stem cell transplantation for transplant-eligible patients of all ages in the randomized, phase III, Myeloma XI trial Graham H. Jackson,1 Faith E. Davies,2 Charlotte Pawlyn,3 David A. Cairns,4 Alina Striha,4 Corinne Collett,4 Anna Waterhouse,4 John R. Jones,5 Bhuvan Kishore,6 Mamta Garg,7 Cathy D. Williams,8 Kamaraj Karunanithi,9 Jindriska Lindsay,10 David Allotey,11 Salim Shafeek,12 Matthew W. Jenner,13 Gordon Cook,14 Nigel H. Russell,8 Martin F. Kaiser,3 Mark T. Drayson,15 Roger G. Owen,16 Walter M. Gregory4 and Gareth J. Morgan2 for the UK NCRI Haematological Oncology Clinical Studies Group

Ferrata Storti Foundation

Haematologica 2021 Volume 106(7):1957-1967

Northern Institute for Cancer Research, Newcastle University, Newcastle upon Tyne, UK; Perlmutter Cancer Center, NYU Langone Health, New York, NY, USA; 3The Institute of Cancer Research and The Royal Marsden Hospital NHS Foundation Trust, London, UK; 4 Clinical Trials Research Unit, Leeds Institute of Clinical Trials Research, University of Leeds, Leeds, UK; 5King’s College Hospital NHS Foundation Trust, London, UK; 6Heart of England NHS Foundation Trust, Birmingham, UK; 7Leicester Royal Infirmary, Leicester, UK; 8 Centre for Clinical Haematology, Nottingham University Hospital, Nottingham, UK; 9 University Hospital of North Midlands, Stoke-on-Trent, UK; 10East Kent Hospitals University NHS Foundation Trust, Canterbury, UK; 11Derby Teaching Hospitals NHS Foundation Trust, Derby, UK; 12Worcestershire Acute Hospitals NHS Trust, Worcester, UK; 13University Hospital Southampton NHS Foundation Trust, Southampton, UK; 14Section of Experimental Haematology, Leeds Institute of Cancer and Pathology, University of Leeds, Leeds, UK; 15 Clinical Immunology, School of Immunity and Infection, University of Birmingham, Birmingham, UK and 16St James's University Hospital, Haematological Malignancy Diagnostic Service (HMDS), Leeds, UK 1 2

ABSTRACT

he optimal way to use immunomodulatory drugs as components of induction and maintenance therapy for multiple myeloma is unresolved. We addressed this question in a large phase III randomized trial, Myeloma XI. Patients with newly diagnosed multiple myeloma (n=2,042) were randomized to induction therapy with cyclophosphamide, thalidomide, and dexamethasone (CTD) or cyclophosphamide, lenalidomide, and dexamethasone (CRD). Additional intensification therapy with cyclophosphamide, bortezomib, and dexamethasone (CVD) was administered before autologous stem-cell transplantation to patients with a suboptimal response to induction therapy using a response-adapted approach. After receiving high-dose melphalan with autologous stem cell transplantation, eligible patients were further randomized to receive either lenalidomide alone or observation alone. Co-primary endpoints were progression-free survival (PFS) and overall survival (OS). The CRD regimen was associated with significantly longer PFS (median: 36 vs. 33 months; hazard ratio [HR], 0.85; 95% confidence interval [CI]: 0.75-0.96; P=0.0116) and OS (3-year OS: 82.9% vs. 77.0%; HR, 0.77; 95% CI: 0.63-0.93; P=0.0072) compared with CTD. The PFS and OS results favored CRD over CTD across all subgroups, including patients with International Staging System stage III disease (HR for PFS, 0.73; 95% CI: 0.58-0.93; HR for OS, 0.78; 95% CI: 0.56-1.09), high-risk cytogenetics (HR for PFS, 0.60; 95% CI: 0.43-0.84; HR for OS, 0.70; 95% CI: 0.42-1.15) and ultra-high-risk cytogenetics (HR for PFS, 0.67; 95% CI: 0.411.11; HR for OS, 0.65; 95% CI: 0.34-1.25). Among patients randomized to lenalidomide maintenance (n=451) or observation (n=377), maintenance therapy improved PFS (median: 50 vs. 28 months; HR, 0.47; 95% CI: 0.370.60; P<0.0001). Optimal results for PFS and OS were achieved in the patients who received CRD induction and lenalidomide maintenance. The trial was registered with the EU Clinical Trials Register (EudraCT 2009010956-93) and ISRCTN49407852. haematologica | 2021; 106(7)

Correspondence: GRAHAM H. JACKSON graham.jackson@newcastle.ac.uk Received: January 20, 2020. Accepted: May 28, 2020. Pre-published: June 4, 2020. https://doi.org/10.3324/haematol.2020.247130

1957

G.H. Jackson et al.

Introduction The introduction of novel agents, such as immunomodulatory drugs and proteasome inhibitors, has contributed to the recent dramatic improvements in outcomes observed for patients with multiple myeloma.1-3 Following induction, high-dose melphalan-based chemotherapy with autologous stem cell transplantation (ASCT) remains the standard of care for eligible patients.4-9 The optimal approach to induction therapy prior to ASCT and consolidation or maintenance after ASCT in this new era has not yet been defined. However, several principles have been established, including the value of using at least triplet combinations of agents that can induce deeper, longer remissions by targeting different clonal populations.10,11 The efficacy of immunomodulatory drugs in multiple myeloma has been linked to their mode of action. These drugs target the cereblon ubiquitin ligase complex, which leads to both tumoricidal effects early on and immunomodulatory effects beneficial for long-term tumor control.12-15 The immunomodulatory drugs thalidomide and lenalidomide are recognized as effective treatment options in both the induction7,9,10,16-18 and maintenance settings.6,19-21 Lenalidomide has fewer side effects than thalidomide, enabling long-term treatment and disease control.19-21 We have addressed how to optimize the use of these agents between induction and maintenance for patients receiving ASCT in a large, randomized trial (UK National Cancer Research Institute [NCRI] Myeloma XI).

Methods The Myeloma XI study had a multifactorial design enabling the investigation of a number of pertinent clinical questions with adequate statistical control and power. Importantly, the influence of one phase of treatment or question on another could be separated and controlled for. This was achieved by stratifying the consolidation and maintenance randomizations for earlier treatment allocations. This report concentrates on induction and its interaction with maintenance therapy in the transplant-eligible population of patients within the trial. The other questions posed by the study are addressed in separate manuscripts.

cyclophosphamide, lenalidomide, and dexamethasone (CRD) or cyclophosphamide, thalidomide, and dexamethasone (CTD) (induction randomization), stratified according to certain factors (Online Supplementary Methods). Patients received a minimum of four cycles in the absence of progressive disease, and treatment continued until maximum response was achieved. Additional intensification therapy before ASCT was administered to patients with a suboptimal response to induction therapy using a response-adapted approach: patients with stable disease after induction therapy or those with progressive disease at any time during induction therapy received a maximum of eight cycles of cyclophosphamide, bortezomib, and dexamethasone (CVD); patients with a minimal or partial response were randomized (1:1) to CVD or no CVD. Patients with a very good partial response or complete response received no additional therapy before ASCT. The results of the intensification randomization have been published elsewhere.22 Three months after ASCT, eligible patients were randomized to observation or to maintenance therapy with lenalidomide alone, or in combination with vorinostat until unacceptable toxicity or progressive disease. Patients were excluded from maintenance randomization if they did not respond to CRD induction, had no response to any prior study treatment, had progressive disease, or relapsed after achieving a complete response. Randomized patients were stratified according to treatment center and previous randomization group(s). The results of the maintenance randomization have been published elsewhere.23 Further details on the dose and schedule of all study treatments are provided in Online Supplementary Table S1, and a flow diagram of the CRD and CTD groups of patients is shown in Online Supplementary Figure S1. The study was approved by the national ethics review board (National Research Ethics Service, London, UK), institutional review boards of the participating centers, and the competent regulatory authority (Medicines and Healthcare Products Regulatory Agency, London, UK). All patients provided written informed consent. The trial was registered with the EU Clinical Trials Register (EudraCT number, 2009-010956-93) and ISRCTN49407852.

Study endpoints and statistical analysis The co-primary endpoints were progression-free survival (PFS) and overall survival (OS). Secondary endpoints included PFS-Two (PFS2), response, and safety. Further details regarding the statistical analysis are provided in the Online Supplementary Material.

Study design and eligibility criteria The Myeloma XI trial was a phase III, open-label, parallelgroup, multi-arm, adaptive design trial with three randomization stages conducted at 110 National Health Service hospitals in England, Wales, and Scotland (see Online Supplementary Data for a list of study sites with principal investigators and number of patients recruited). Eligible patients were aged ≥18 years and newly diagnosed with multiple myeloma. Exclusion criteria included previous or concurrent malignancies (including myelodysplastic syndromes), grade ≥2 peripheral neuropathy, acute renal failure (unresponsive to up to 72 h of rehydration, characterized by creatinine >500 mmol/L or urine output <400 mL/day or requiring dialysis), and active or prior hepatitis C infection. The trial design included an intensive treatment pathway for transplant-eligible patients and a less intensive treatment pathway for transplant-ineligible patients. Strict age limits were deliberately avoided so that fit, older patients could receive intensive therapy and ASCT. The decision of which treatment pathway to undertake was made on an individual patient basis taking into account performance status, clinician judgment, and patient preference. Transplant-eligible patients were randomized on a 1:1 basis to 1958

Results Patients Between May 26, 2010 and April 20, 2016, 2,042 transplant-eligible patients underwent induction randomization (Online Supplementary Figure S1). Baseline characteristics were well balanced between the two treatment groups (Table 1). The median age of all the patients was 61 years (range, 28-75 years), 60% of the patients were male, and 24% had International Staging System (ISS) stage III disease. Of the 836 (40.9%) patients for whom genetic risk could be calculated, 266 (31.8%) had high-risk and 111 (13.3%) had ultra-high-risk cytogenetics. The median duration of follow-up since study entry was 36.3 months (interquartile range [IQR], 23.0-48.5 months).

Induction randomization results Progression-free survival and overall survival Disease progression or death occurred in 456 patients in haematologica | 2021; 106(7)

Lenalidomide induction and maintenance in MM

Table 1. Patients’ characteristics according to induction regimen.

Characteristic Median age (range), years Age group, n (%) ≤65 years >65 years Sex, n (%) Male Female Ethnicity, n (%) White Black (Black Caribbean, Black African, other) Asian (Indian, Pakistani, Bangladeshi, other) Other Unknown WHO performance status, n (%) 0 1 2 ≥3 Unknown Immunoglobin subtype, n (%) IgG IgA IgM IgD Light chain only Non-secretor Unknown ISS stage, n (%) I II III Unknown Median serum creatinine (range), mol/L Unknown, n (%) Median lactate dehydrogenase (range), IU/L Unknown, n (%) CVD randomization after MR/PR, n (%) Allocated to CVD Allocated to no CVD Received CVD after SD/PD, n (%) Maintenance treatment, n (%) Lenalidomide Lenalidomide plus vorinostat Observation Cytogenetic data available, n (%) Cytogenetic lesions, n (% of those with data available) t(4;14) t(14;16) t(14;20) del(17p) gain(1q) Cytogenetic risk category, n (% of those with data available) Standard High* Ultra-high†

CRD (n=1,021)

CTD (n=1,021)

61 (28-75)

61 (29-74)

772 (75.6) 249 (24.4)

754 (73.8) 267 (26.2)

610 (59.7) 411 (40.3)

611 (59.8) 410 (40.2)

938 (91.9) 21 (2.1) 28 (2.7) 10 (0.9) 24 (2.4)

937 (91.8) 14 (1.4) 27 (2.6) 14 (1.4) 29 (2.8)

421 (41.2) 363 (35.6) 119 (11.7) 53 (5.2) 65 (6.4)

439 (43.0) 367 (35.9) 135 (13.2) 34 (3.3) 46 (4.5)

633 (62.0) 220 (21.5) 4 (0.4) 12 (1.2) 139 (13.6) 6 (0.6) 7 (0.7)

600 (58.8) 269 (26.3) 4 (0.4) 9 (0.9) 127 (12.4) 7 (0.7) 5 (0.5)

301 (29.5) 392 (38.4) 246 (24.1) 82 (8.0) 85.0 (28.0-825.0) 9 (8.8) 262.0 (3.0-2519.0) 228 (22.3)

306 (30.0) 388 (38.0) 253 (24.8) 74 (7.2) 83.0 (30.0-897.0) 7 (6.9) 273.0 (0.0-3550.0) 215 (21.1)

85 (8.3) 82 (8.0) 35 (3.4)

98 (9.6) 102 (10.0) 38 (3.7)

230 (22.5) 103 (10.1) 190 (18.6) 414 (40.5)

221 (21.6) 93 (9.1) 187 (18.3) 422 (41.3)

56 (13.5) 8 (1.9) 3 (0.7) 31 (7.5) 137 (33.1)

70 (16.6) 12 (2.8) 2 (0.5) 42 (10.0) 136 (32.2)

223 (53.9) 149 (36.0) 42 (10.1)

236 (55.9) 117 (27.7) 69 (16.4)

CRD: cyclophosphamide, lenalidomide, and dexamethasone; CTD: cyclophosphamide, thalidomide, and dexamethasone; CVD: cyclophosphamide, bortezomib, and dexamethasone; Ig: immunoglobulin; ISS: International Staging System; MR: minimal response; PD: progressive disease; PR: partial response; SD: stable disease; WHO: World Health Organization. *High risk defined as the presence of any one of t(4;14), t(14;16), t(14;20), del(17p), or gain(1q). †Ultra-high risk defined as the presence of more than one lesion.

haematologica | 2021; 106(7)

1959

G.H. Jackson et al.

the CRD group and in 509 patients in the CTD group. The CRD regimen was associated with significantly longer PFS than the CTD regimen (hazard ratio [HR], 0.85; 95% confidence interval [CI]: 0.75-0.96; P=0.0116) (Figure 1A). The median PFS was 36 months (95% Cl: 33-39) with CRD and 33 months (95% CI: 31-35) with CTD. The median overall survival has not yet been reached with the current follow-up. Death occurred in 185 patients in the CRD group and in 230 patients in the CTD group. There was a statistically significant difference in OS favoring CRD (HR, 0.77; 95% CI: 0.63-0.93; P=0.0072) (Figure 1B). The 3-year OS rate was 82.9% (95% Cl: 80.2-85.7) with CRD and 77.0% (95% CI: 73.9-80.0) with CTD.

Subgroup analyses indicated that PFS and OS were better with CRD than with CTD across all subgroups (Figure 2). In the subset of patients with ISS stage III disease, CRD was superior to CTD for PFS (HR, 0.73; 95% CI: 0.580.93) and there was a trend toward improved OS (HR, 0.78; 95% CI: 0.56-1.09). In each case, there was no evidence of heterogeneity of treatment effect (PFS: P=0.2645; OS: P=0.7606) (Figure 2). Similar results were seen in the subgroup of patients with high-risk cytogenetics (HR for PFS, 0.60; 95% CI: 0.43-0.84; HR for OS, 0.70; 95% CI: 0.42-1.15) and ultra-high risk cytogenetics (HR for PFS, 0.67; 95% CI: 0.41-1.11; P=0.6164; HR for OS, 0.65; 95% CI: 0.34-1.25; P=0.8131), with no significant heterogeneity

Figure 1. Outcomes according to induction regimen. (A) Progression-free survival and (B) overall survival, with dashed line showing the median. CRD: cyclophosphamide, lenalidomide, and dexamethasone; CTD: cyclophosphamide, thalidomide, and dexamethasone; 95% CI: 95% confidence interval.

1960

haematologica | 2021; 106(7)

Lenalidomide induction and maintenance in MM

of treatment effect observed (Figure 2). PFS2, a secondary endpoint, was also analyzed. CRD was associated with significantly longer PFS2 than was CTD (HR, 0.76; 95% CI: 0.65-0.90; P=0.001) (Online Supplementary Figure S2). The median PFS2 was 59 months (95% Cl: 55-63) with CRD and 54 months (95% CI: 4960) with CTD.

Response After induction triplet therapy, the proportion of patients with a very good partial response or better was significantly higher with CRD than with CTD (60.4% vs. 52.9%; P=0.0006) (Table 2). The odds ratio (OR) of 1.37 (95% CI: 1.15-1.65) indicates a 37% increase in the odds of achieving a deep remission in the CRD group than in the CTD group. After ASCT, the proportion of patients achieving a very good partial response or better remained higher in the CRD group than in the CTD group, but the difference was not statistically significant (81.5% vs. 76.9%; OR, 1.25; 95% CI: 0.94-1.66; P=0.1277) (Table 2). Due to the lower induction response rate with CTD compared with CRD, more patients underwent CVD intensification as per protocol (CRD, 11.8% vs. CTD, 13.3%). The interaction between induction therapy and CVD was therefore examined further. Counterfactual estimates of the survivor function if CVD rescue treatment was not received by any patients maintained differences in median PFS (CRD: 36 months [95% CI: 33-39] vs. CTD: 33 months [95% CI: 30-34]) (Online Supplementary Figure S3A) and 3-year OS rate (CRD: 82.9% [95% CI: 80.0-85.5] vs. CTD: 76.3% [95% CI: 73.0-79.2]) (Online Supplementary Figure S3B). Similar counterfactual estimates obtained in the scenario in which patients randomized to no CVD after partial/minimal response were treated with CVD provided similar estimates for median PFS (CRD: 36 months [95% CI: 33-39] vs. CTD: 33 months [95% CI: 3136]) (Online Supplementary Figure S3C) and 3-year OS rate (CRD: 83.1% [95% CI: 80.2-85.6] vs. CTD: 77.3% [95% CI: 74.1-80.2]) (Online Supplementary Figure S3D). After adjustment for the effect of CVD treatment in a counterfactual analysis, the hazard ratios for PFS and OS were 0.82 (95% CI: 0.69-0.96) and 0.69 (95% CI: 0.53-0.91), respectively. This suggests a greater treatment effect of

CRD induction treatment on PFS and particularly OS than apparent with the unadjusted intention-to-treat analysis (Online Supplementary Figure S3A and S3B, respectively). The full results of the CVD intensification randomization have been presented elsewhere.22

Safety The median number of cycles of induction therapy delivered was five (range, 1-18) for CRD and five (range, 1-12) for CTD. The median percentage of minimum protocol-defined delivered dose of lenalidomide and thalidomide during induction therapy was 116.7% (IQR, 96.4150.0) and 100.0% (IQR, 71.4-128.6), respectively. Lenalidomide dose modifications occurred in 391 (38.3%) patients who received CRD induction therapy, and thalidomide dose modifications occurred in 751 (73.6%) patients who received CTD induction therapy. The rate of discontinuation of induction therapy due to adverse events was similar with CRD and CTD (51 patients [5.0%] and 68 patients [6.7%], respectively). Overall, 64.4% of patients proceeded to ASCT following induction with or without intensification. There was no difference in the proportion of patients undergoing ASCT between those receiving CTD (63.3%) or CRD (65.5%) induction suggesting this was not due to induction-related toxicity. The most common reason for not proceeding was “Patient not fit/clinician’s decision” in 36.1% of cases. Differences in the safety profile of CRD and CTD were consistent with the known side effects of lenalidomide and thalidomide (Table 3). In general, CRD was associated with a higher rate of grade ≥3 neutropenia (22.3% vs. 11.7%) and diarrhea (2.6% vs. 1.0%), whereas CTD was associated with a higher rate of grade ≥3 peripheral sensory neuropathy (1.5% vs. 0.6%) and constipation (1.9% vs. 0.8%). The incidence of deep vein thrombosis was 5.7% in the CRD group and 4.8% in the CTD group; pulmonary embolism was reported in 3.2% and 4.9% of patients, respectively. The 3-year cumulative incidence of invasive second primary malignancies (SPM) was low and comparable between CRD and CTD (2.9% [95% CI: 1.7-4.1] vs. 1.5% [95% CI: 0.6-2.4]; HR, 1.60 [95% CI: 0.87-2.93]; P=0.1311). The SPM incidence rate per 100 patient-years

Table 2. Response rates after induction and autologous stem-cell transplantation.

Response, n (%) CR or VGPR CR CR w/o BM VGPR PR or MR PR MR SD or PD SD PD Death within 100 days after ASCT Unknown

Response following induction therapy CRD CTD (n=1,021) (n=1,021) 617 (60.4) 87 (8.5) 297 (29.1) 233 (22.8) 297 (29.1) 261 (25.6) 36 (3.5) 32 (3.1) 8 (0.8) 24 (2.4) 13 (1.3) 57 (5.6)

540 (52.9) 61 (6.0) 223 (21.8) 256 (25.1) 348 (34.1) 301 (29.5) 47 (4.6) 43 (4.2) 8 (0.8) 35 (3.4) 17 (1.7) 61 (6.0)

Response following ASCT CRD CTD (n=628) (n=603) 512 (81.5) 149 (23.7) 218 (34.7) 145 (23.1) 95 (15.1) 94 (15.0) 1 (0.2) 11 (1.8) 0 (0.0) 11 (1.8) 1 (0.2) 9 (0.9)

464 (76.9) 122 (20.2) 214 (35.5) 128 (21.2) 102 (16.9) 98 (16.3) 4 (0.7) 10 (1.7) 0 (0.0) 10 (1.7) 6 (1.0) 21 (2.1)

ASCT: autologous stem cell transplantation; CR: complete response; CR w/o BM: complete response by immunological criteria without confirmation by bone marrow; CRD: cyclophosphamide, lenalidomide, and dexamethasone; CTD: cyclophosphamide, thalidomide, and dexamethasone; MR: minimal response; PD: progressive disease; PR: partial response; SD: stable disease; VGPR: very good partial response.

haematologica | 2021; 106(7)

1961

G.H. Jackson et al.

was 1.2 (95% CI: 0.8-1.7) in the CRD group and 0.9 (95% CI: 0.6-1.3) in the CTD group. The incidence of serious adverse events during induction was similar with CRD and CTD (59.0% vs. 57.7%). Infection accounted for nearly half of all serious adverse events reported during induction (45.2% for CRD vs. 46.4% for CTD). Fatal adverse events occurred in six patients in the CRD group and in three patients in the CTD group. Of the nine patients with grade 5 adverse events, one had three concurrent events (renal failure, liver failure, and sepsis), one had two concurrent events (small bowel obstruction and sepsis), and the remaining seven

patients had one event each (pneumonia [n=2]; sepsis [n=2]; collapse/syncope [n=2]; lower respiratory tract infection [n=1]; hepatitis encephalopathy [n=1]).

Interaction of lenalidomide induction and maintenance Following ASCT, patients were randomized between maintenance lenalidomide and observation, giving us the opportunity to explore the interaction between induction and maintenance agents in this setting. Of the 2,042 transplant-eligible patients who entered the first randomization, 1,024 entered the maintenance phase and were randomized to lenalidomide alone (n=451), to lenalidomide

Figure 2. Outcomes according to induction regimen in selected subgroups. (A) Progression-free survival and (B) overall survival; Hazard ratio <1.00 favors CRD. *Likelihood ratio test for heterogeneity of effect among patients with subgroup data available. CI: confidence interval; CRD: cyclophosphamide, lenalidomide, and dexamethasone; CTD: cyclophosphamide, thalidomide, and dexamethasone; het: heterogeneity; HiR: high risk; HR: hazard ratio; ISS: International Staging System; SR: standard risk; UHiR: ultra-high risk.

1962

haematologica | 2021; 106(7)

Lenalidomide induction and maintenance in MM

plus vorinostat (n=196, not included in this further analysis), or to observation (n=377). The baseline characteristics of patients undergoing maintenance randomization were well balanced between the two treatment groups (Online Supplementary Table S2). Approximately half of patients in both the lenalidomide and observation groups had received CRD as induction therapy (230 of 451 [51.0%] in

the lenalidomide group; 190 of 377 [50.4%] in the observation group). Lenalidomide maintenance was associated with significantly longer PFS and OS compared with observation in transplant-eligible patients (median: 50 vs. 28 months; HR, 0.47; 95% CI: 0.37-0.60; P<0.0001 at a median follow-up of 27.2 [IQR, 12.8-42.0] months). In a post-hoc exploratory analysis, the longest PFS was

Figure 3. Outcomes according to induction and maintenance treatment. (A) Progression-free survival and (B) overall survival. CRD: cyclophosphamide, lenalidomide, and dexamethasone; CTD: cyclophosphamide, thalidomide, and dexamethasone; Obs: observation; Len: lenalidomide.

haematologica | 2021; 106(7)

1963

G.H. Jackson et al.

observed in patients who received CRD induction and lenalidomide maintenance. The median PFS in this group was not reached, while it was 49 months in those who received CTD and lenalidomide maintenance, 32 months in those who received CTD and observation, and 24 months in those who received CRD and observation (Figure 3A). Similarly, the longest OS was observed in patients who received CRD induction and lenalidomide maintenance. The median OS was not reached in any group, but 3-year OS rates were 92.3% for those who received CRD induction with lenalidomide maintenance, 89.0% in those who received CTD and lenalidomide maintenance, 86.0% in those who received CTD and observation, and 90.3% in those who received CRD and observation (Figure 3B).

Discussion This is the largest study to evaluate the CRD regimen as induction therapy before ASCT in patients with multiple myeloma. We show that it is associated with excellent efficacy and safety data and the results are consistent with prior studies evaluating either CTD,17,18,24 CRD as induction therapy,25 or CRD as treatment in the relapsed/refractory disease setting.26 A direct comparison of thalidomide and lenalidomide as the immunomodulatory component of induction therapy has not been previously undertaken in the context of a randomized trial for transplant-eligible newly diagnosed myeloma patients. Our results demonstrate the superiority of lenalidomide over thalidomide both in terms of efficacy and tolerability in the context of combination with an alkylating agent (cyclophosphamide), supporting the findings of previous non-randomized analyses.27,28 Previous randomized studies in patients not eligible for stem-cell transplant have compared thalidomide to lenalidomide in combination with the alkylating agent melphalan.29,30 In these studies no difference between lenalidomide and thalidomide in terms of response, progression-free or overall survival was identified. The differences between these prior studies and the finding from Myeloma XI might be explained by the different patient population or the different alkylating agent, cyclophosphamide, which may be better tolerated than melphalan. Response rates obtained with CRD in the current study were good: 60% of patients achieved at least a very good partial response after induction and 82% did so after ASCT. This compares favorably with other novel-agent-based triplet induction therapies, including bortezomib, doxorubicin, and dexamethasone (VAD),31,32 CVD,32 bortezomib, thalidomide, and dexamethasone (VTD),5,10,33,34 and even the immunomodulatory drug/proteasome inhibitor regimen bortezomib, lenalidomide, and dexamethasone (VRD)9,35 (Online Supplementary Table S3). However there are many caveats when trying to compare results across trials. Particularly in comparing response rates it should be noted that patients in Myeloma XI received induction until maximum response rather than for a fixed duration and this may have led to deeper responses prior to transplantation than in other studies. Although immunomodulatory drug and proteasome inhibitor combinations (e.g. VTD/VRD) have recently become widely used in the European Union and 1964

USA this was not the situation when the study was initially implemented. At that time either an immunomodulatory-based regimen or a proteasome inhibitor-based regimen (e.g., MPV or VD) was used. The standard of care in the UK, as in a number of other countries, was CTD. The addition of a proteasome inhibitor to induction regimens offers the potential to target immunomodulatory agent-resistant subclones of disease with a second novel agent. This concept was explored in the intensification randomization aspect of the study which has been previously reported22 and demonstrated that intensification treatment with CVD significantly improved PFS in patients with newly diagnosed multiple myeloma and a suboptimal response to immunomodulatory induction therapy compared with no intensification treatment. The combination of a fourth agent with a different mechanism of action to induction, such as daratumumab plus VTD (Dara-VTD) investigated in the recently published Cassiopeia trial, is able to induce even deeper responses, with 83% of patients achieving at least very good partial response.36 PFS was longer in patients treated with Dara-VTD than in those treated with VTD alone, suggesting the addition of further agents to active triplets can improve outcomes yet further. In contrast, however, CRD offers an all oral regimen requiring only one hospital visit per month and including only one more expensive agent, lenalidomide. As such it is comparatively easier to deliver and likely to be cheaper in terms of both drug and administration costs. The lower incidence of peripheral neuropathy seen with CRD than that seen with combinations including bortezomib and/or thalidomide may also be beneficial for some patients. The Myeloma XI data support the continued use of ASCT, since in a previous study of CRD without ASCT,7 the median PFS was 28.6 months, which is shorter than that achieved with CRD and ASCT in the Myeloma XI trial (36 months). Similarly, in the IFM 2009 study comparing VRD with or without ASCT, the combination of VRD and ASCT led to significantly better PFS than VRD alone (median: 50 vs. 36 months; P<0.001).9 The median OS in that study was similar in both groups, likely due to the fact that 79% of patients assigned to VRD alone received salvage ASCT at relapse and the short current follow-up. These findings and data from several other studies suggest a complementary role for novel agents and ASCT. We have shown that treatment with lenalidomide maintenance therapy after ASCT is associated with improved PFS and OS, a finding consistent with other reports.6,19,20,37 We show that in Myeloma XI, the efficacy of lenalidomide maintenance was not diminished by prior exposure to lenalidomide; in fact, the best outcomes were achieved when lenalidomide was given as both induction and maintenance. This is similar to results seen in previous lenalidomide maintenance studies, which showed significant heterogeneity of effect of lenalidomide maintenance with outcomes favoring those who had received lenalidomide induction.20,38 This suggests that patients with disease sensitive to immunomodulation with lenalidomide will continue to benefit from its continued use, perhaps as the maintenance therapy targets quiescent cells as they come out of cycle. We noted that patients receiving CRD and observation appeared to have slightly inferior PFS than patients receiving CTD and observation. This was not due to any apparent difference in early discontinuation of therapy or dose haematologica | 2021; 106(7)

Lenalidomide induction and maintenance in MM

Table 3. Adverse events according to induction regimen (safety population*).

Grade ≥3 AE, n (%)

CRD (n=1,010)

CTD (n=1,004)

Neutropenia Anemia Thrombocytopenia Diarrhea Constipation Peripheral sensory neuropathy Peripheral motor neuropathy

225 (22.3) 97 (9.6) 46 (4.5) 26 (2.6) 8 (0.8) 6 (0.6) 5 (0.5)

117 (11.7) 67 (6.7) 17 (1.7) 10 (1.0) 19 (1.9) 15 (1.5) 14 (1.4)

AEs of interest (any grade), n (%)

CRD (n=1,010)

CTD (n=1,004)

Peripheral sensory neuropathy Peripheral motor neuropathy Deep vein thrombosis Pulmonary embolism Other thrombosis/embolism

251 (24.9) 87 (8.6) 58 (5.7) 32 (3.2) 8 (0.8)

452 (45.0) 163 (16.2) 48 (4.8) 49 (4.9) 11 (1.1)

*The safety population included all randomly assigned patients who received one or more doses of the induction or maintenance regimen. AE: adverse event; CRD: cyclophosphamide, lenalidomide, and dexamethasone; CTD: cyclophosphamide, thalidomide, and dexamethasone.

modifications and so is difficult to explain. The PFS difference is small, not statistically significant and may have occurred by chance. In the analysis of OS the reverse pattern was seen with patients receiving CRD and observation having an apparent improved OS compared to those receiving CTD and observation. The results of Myeloma XI are likely to reflect the true impact of the CRD combination in clinical practice because of the limited exclusion criteria for the study population. Notably, there were no age restrictions for the intensive pathway, allowing older but fit patients to undergo ASCT. The median age in this group was 61 years, and patients up to the age of 75 years were included. In contrast, most previous studies of ASCT have excluded patients aged over 65 or 70 years. Evidence suggests that fit patients aged >65 years can benefit from ASCT, especially when combined with regimens containing novel agents.3,39,40 Our approach may also explain the relatively lower proportion of patients proceeding to ASCT in this study than in other studies of induction therapy which are usually limited only to patients under the age of 65. The most common reason for patients not proceeding to stem-cell transplant was given as “patient not fit/clinician’s decision” suggesting that clinicians may have initially entered patients in the transplant-eligible pathway of the study as a ‘trial of fitness’ so as not to limit their options prior to withdrawing the patient nearer the time of transplantation. In addition, the proportion of patients with ISS stage III disease (24%) in the present study was slightly higher than that in some recent studies of induction therapy.9,10,31,35 Cytogenetic abnormalities, such as t(4;14), t(14;16), and del(17p), are important prognostic markers, and should therefore be investigated in all patients with multiple myeloma according to the International Myeloma Working Group molecular classification.41 Although cytogenetic data were only available for 41% of patients in our study, this percentage is comparable to that in other trials of patients with newly diagnosed multiple myeloma.42 haematologica | 2021; 106(7)

While three-drug induction regimens are generally more effective than two-drug regimens, they may also be more toxic.10,11 In the Myeloma XI trial, the safety results for CRD and CTD were consistent with the known safety profiles of these agents. Notably, rates of peripheral neuropathy were lower with CRD than with CTD. An important safety concern with lenalidomide treatment in patients with newly diagnosed multiple myeloma is the risk of SPM.43 In this population of transplant-eligible patients, the overall 3-year cumulative incidence of invasive SPM was low (2.2%; 95% CI: 1.5-3.0) and the type of induction therapy used did not appear to affect the SPM incidence rate. Safety results for lenalidomide maintenance compared to observation, including SPM incidence, have been previously published.23,24 Despite the risks associated with continued active therapy, registry data suggest that health-related quality of life of patients receiving lenalidomide maintenance is similar to that of patients receiving no maintenance.45 In summary, induction therapy with CRD improved PFS and OS compared with CTD in transplant-eligible patients with newly diagnosed multiple myeloma. The best results were achieved when patients received both lenalidomide-based induction therapy and lenalidomide maintenance. Disclosures GHJ has received honoraria from and provided consultancy and speakers bureau services for Roche, Amgen, Janssen, and Merck Sharp and Dohme; and has received honoraria, travel support and research funding from and provided consultancy and speakers bureau services for Celgene Corporation and Takeda. FED reports consultancy for and honoraria from Amgen, AbbVie, Takeda, Janssen and Roche; and consultancy for and honoraria and research funding from Celgene Corporation. CP reports consultancy for and travel support from Amgen and Takeda Oncology; honoraria and travel support from Janssen; and consultancy for and honoraria and research funding from Celgene Corporation. DAC, AS, CC and AW have received research funding from Celgene Corporation, Amgen, Merck Sharp and Dohme. JRJ has received honoraria and research funding from Celgene Corporation. BK reports consultancy and speakers bureau services for and travel support from Celgene Corporation, Takeda, and Janssen. MG has received travel support and research funding from and participated in speakers bureau for Janssen; has received travel support from Takeda; and travel support and research funding from Novartis. CDW has received honoraria and travel support from and participated in speakers bureau for Takeda, Janssen and Celgene Corporation; has received honoraria from Novartis; and has participated in speakers bureau for and received honoraria from Novartis. KK has received travel support and research funding from Celgene Corporation and Janssen. JL has provided consultancy services for Janssen; received travel support from Novartis; honoraria and travel support from Takeda; has acted as a consultant for and received travel support from Bristol-Myers Squibb; and reports consulting for and honoraria and travel support from Celgene Corporation. DA and NHR have nothing to disclose. SS reports travel support from Celgene Corporation and Janssen; speaker services for Pfizer; and meeting sponsorship from AbbVie. MWJ reports consultancy for and honoraria, travel support, and research funding from Janssen; consultancy for and honoraria and research funding from Celgene Corporation; consultancy for and honoraria from Novartis; and consultancy for and honoraria and travel support from Takeda and Amgen GC repports con1965

G.H. Jackson et al.

sultancy for and honoraria from Glycomimetics and BristolMyers Squibb; consultancy and speakers bureau services for and honoraria from Sanofi; and consultancy and speakers bureau services for and honoraria and research funding rom Takeda, Celgene Corporation, Janssen and Amgen. MFK reports consulting for Chugai; consultancy for and honoraria from Janssen and Amgen; consultancy for and travel support from Takeda and Bristol-Myers Squibb; and consultancy for and honoraria and research funding from Celgene Corporation. MTD has equity ownership or membership on the board of directors or advisory committees of Abingdon Health. RGO has received honoraria and travel support from Takeda; provided consultancy services for and received travel support from Janssen; and acted as a consultant for and received honoraria and research funding from Celgene Corporation. WMG reports consultancy services for and research funding from Celgene Corporation; research funding from Amgen, Merck Sharp and Dohme and honoraria from Janssen. GJM reports research funding from Janssen; consultancy for and honoraria from Bristol-Myers Squibb, Takeda, Roche, Amgen, GlaxoSmith Kline and Karyopharm; and consultancy for and honoraria and research funding from Celgene Corporation. Contributions GHJ, FED, NHR and GJM were the chief investigators; GHJ, FED, NHR, WMG and GJM designed the trial and developed the protocol; DAC, AS and WMG developed and carried out the statistical analysis; GHJ, FED, CP, JRJ, BK, MG, CDW, KK, JL, DA, SS, MWJ, GC, NHR, MFK, RGO, and GJM participated in recruitment of patients; MFK, MTD, RGO, and GJM coordinated the central laboratory investigations; CC and AW coordinated the data collection and regulatory and governance requirements; GHJ, FED, CP, DAC, AS, MFK, MTD, RGO, WMG and GJM analyzed and interpreted the data; GHJ, FED, CP, DAC, AS and GJM developed the first drafts of the manuscript. All authors contributed to the review and amendments of the manuscript for important intellectual content and approved the final version for submission. Acknowledgments We thank all the patients at centers throughout the UK whose willingness to participate made this study possible. We are grateful to the UK National Cancer Research Institute

References 1. Kumar SK, Rajkumar SV, Dispenzieri A, et al. Improved survival in multiple myeloma and the impact of novel therapies. Blood. 2008;111(5):2516-2520. 2. Kristinsson SY, Anderson WF, Landgren O. Improved long-term survival in multiple myeloma up to the age of 80 years. Leukemia. 2014;28(6):1346-1348. 3. Kumar SK, Dispenzieri A, Lacy MQ, et al. Continued improvement in survival in multiple myeloma: changes in early mortality and outcomes in older patients. Leukemia. 2014;28(5):1122-1128. 4. Child JA, Morgan GJ, Davies FE, et al. High-dose chemotherapy with hematopoietic stem-cell rescue for multiple myeloma. N Engl J Med. 2003;348(19):1875-1883. 5. Moreau P, Avet-Loiseau H, Facon T, et al. Bortezomib plus dexamethasone versus reduced-dose bortezomib, thalidomide plus dexamethasone as induction treatment before autologous stem cell transplantation

1966

Haematological Oncology Clinical Studies Group, UK Myeloma Research Alliance, and to all principal investigators, sub-investigators, and local center staff for their dedication and commitment to recruiting patients to the study. We thank the members of the Myeloma XI Trial Steering Committee and Data Monitoring and Ethics Committee. The support of the Clinical Trials Research Unit at the University of Leeds was essential to the successful running of the study; we thank all their staff who have contributed, past and present. Central laboratory analysis was performed at the Institute of Immunology and Immunotherapy, University of Birmingham; the Institute of Cancer Research, London; and the Haematological Malignancy Diagnostic Service, St James’s University Hospital, Leeds. We are very grateful to the laboratory teams for their contribution to the study. We also acknowledge support from the National Institute of Health Biomedical Research Centre at the Royal Marsden Hospital and the Institute of Cancer Research. The authors received editorial support from Excerpta Medica, funded by the University of Leeds. Funding The primary financial support was from Cancer Research UK [C1298/A10410]. Unrestricted educational grants from Celgene Corporation, Amgen, and Merck Sharp and Dohme, and funding from Myeloma UK supported trial coordination and laboratory studies. The authors are solely responsible for study design, data collection, data analysis and interpretation, writing, and decisions about publication submission; no funder had any role in these aspects of the trial. Trial data were accessible to all authors. The corresponding author had full access to all of the data and the final responsibility to submit for publication. Data sharing statement De-identified participant data will be made available when all trial primary and secondary endpoints have been met. Any requests for trial data and supporting material (data dictionary, protocol, and statistical analysis plan) will be reviewed by the trial management group in the first instance. Only requests which have a methodologically sound proposal and whose proposed use of the data has been approved by the independent trial steering committee will be considered. Proposals should be directed to the corresponding author in the first instance; to gain access, data requestors will need to sign a data access agreement.

in newly diagnosed multiple myeloma. Blood. 2011;118(22):5752-5758. 6. 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. 7. Gay F, Oliva S, Petrucci MT, et al. Chemotherapy plus lenalidomide versus autologous transplantation, followed by lenalidomide plus prednisone versus lenalidomide maintenance, in patients with multiple myeloma: a randomised, multicentre, phase 3 trial. Lancet Oncol. 2015; 16(16):1617-1629 8. Stadtmauer EA, Pasquini MC, Blackwell B, et al. Autologous transplantation, consolidation, and maintenance therapy in multiple myeloma: results of the BMT CTN 0702 trial. J Clin Oncol. 2019;37(7):589-597. 9. 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. 10. Cavo M, Tacchetti P, Patriarca F, et al.

Bortezomib with thalidomide plus dexamethasone compared with thalidomide plus dexamethasone as induction therapy before, and consolidation therapy after, double autologous stem-cell transplantation in newly diagnosed multiple myeloma: a randomised phase 3 study. Lancet. 2010;376(9758):2075-2085. 11. Davies FE, Wu P, Jenner M, et al. The combination of cyclophosphamide, velcade and dexamethasone induces high response rates with comparable toxicity to velcade alone and velcade plus dexamethasone. Haematologica. 2007;92(8):1149-1150. 12. Lopez-Girona A, Mendy D, Ito T, et al. Cereblon is a direct protein target for immunomodulatory and antiproliferative activities of lenalidomide and pomalidomide. Leukemia. 2012;26(11):2326-2335. 13. Gandhi AK, Kang J, Havens CG, et al. Immunomodulatory agents lenalidomide and pomalidomide co-stimulate T cells by inducing degradation of T cell repressors Ikaros and Aiolos via modulation of the E3

haematologica | 2021; 106(7)

Lenalidomide induction and maintenance in MM

ubiquitin ligase complex CRL4. Br J Haematol. 2014;164(6):811-821. 14. Kronke J, Udeshi ND, Narla A, et al. Lenalidomide causes selective degradation of IKZF1 and IKZF3 in multiple myeloma cells. Science. 2014;343(6168):301-305. 15. Bjorklund CC, Lu L, Kang J, et al. Rate of CRL4(CRBN) substrate Ikaros and Aiolos degradation underlies differential activity of lenalidomide and pomalidomide in multiple myeloma cells by regulation of c-Myc and IRF4. Blood Cancer J. 2015;5:e354. 16. Wu P, Davies FE, Horton C, et al. The combination of cyclophosphomide, thalidomide and dexamethasone is an effective alternative to cyclophosphamide - vincristine - doxorubicin - methylprednisolone as induction chemotherapy prior to autologous transplantation for multiple myeloma: a case-matched analysis. Leuk Lymphoma. 2006;47(11):2335-2338. 17. Morgan GJ, Davies FE, Gregory WM, et al. Cyclophosphamide, thalidomide, and dexamethasone as induction therapy for newly diagnosed multiple myeloma patients destined for autologous stem-cell transplantation: MRC Myeloma IX randomized trial results. Haematologica. 2012;97(3):442450. 18. Morgan GJ, Davies FE, Gregory WM, et al. Long-term follow-up of MRC Myeloma IX Trial: survival outcomes with bisphosphonate and thalidomide treatment. Clin Cancer Res. 2013;19(21):6030-6038. 19. 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. 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. Morgan GJ, Gregory WM, Davies FE, et al. The role of maintenance thalidomide therapy in multiple myeloma: MRC Myeloma IX results and meta-analysis. Blood. 2012; 119(1):7-15. 22. Jackson GH, Davies FE, Pawlyn C, et al. Response-adapted intensification with cyclophosphamide, bortezomib, and dexamethasone versus no intensification in patients with newly diagnosed multiple myeloma (Myeloma XI): a multicentre, open-label, randomised, phase 3 trial. Lancet Haematol. 2019;6(12):e616-e629. 23. Jackson GH, Davies FE, Pawlyn C, et al. Lenalidomide maintenance versus observation for patients with newly diagnosed multiple myeloma (Myeloma XI): a multicentre, open-label, randomised, phase 3 trial. Lancet Oncol. 2019;20(1):57-73. 24. Morgan GJ, Davies FE, Gregory WM, et al. Long-term follow-up of MRC Myeloma IX trial: survival outcomes with bisphosphonate and thalidomide treatment. Clin Cancer Res. 2013;19(21):6030-6038.

haematologica | 2021; 106(7)

25. Kumar SK, Lacy MQ, Hayman SR, et al. Lenalidomide, cyclophosphamide and dexamethasone (CRd) for newly diagnosed multiple myeloma: results from a phase 2 trial. Am J Hematol. 2011;86(8):640-645. 26. Morgan GJ, Schey SA, Wu P, et al. Lenalidomide (Revlimid), in combination with cyclophosphamide and dexamethasone (RCD), is an effective and tolerated regimen for myeloma patients. Br J Haematol. 2007;137(3):268-269. 27. Gay F, Hayman SR, Lacy MQ, et al. Lenalidomide plus dexamethasone versus thalidomide plus dexamethasone in newly diagnosed multiple myeloma: a comparative analysis of 411 patients. Blood. 2010;115(7):1343-1350. 28. Luo J, Gagne JJ, Landon J, et al. Comparative effectiveness and safety of thalidomide and lenalidomide in patients with multiple myeloma in the United States of America: a population-based cohort study. Eur J Cancer. 2017;70:22-33. 29. Zweegman S, van der Holt B, Mellqvist UH, et al. Melphalan, prednisone, and lenalidomide versus melphalan, prednisone, and thalidomide in untreated multiple myeloma. Blood. 2016;127(9):11091116. 30. Stewart AK, Jacobus S, Fonseca R, et al. Melphalan, prednisone, and thalidomide vs melphalan, prednisone, and lenalidomide (ECOG E1A06) in untreated multiple myeloma. Blood. 2015;126(11):1294-1301. 31. Sonneveld P, Schmidt-Wolf IG, van der Holt B, et al. Bortezomib induction and maintenance treatment in patients with newly diagnosed multiple myeloma: results of the randomized phase III HOVON-65/ GMMG-HD4 trial. J Clin Oncol. 2012;30(24):2946-2955. 32. Mai EK, Bertsch U, Dürig J, et al. Phase III trial of bortezomib, cyclophosphamide and dexamethasone (VCD) versus bortezomib, doxorubicin and dexamethasone (PAd) in newly diagnosed myeloma. Leukemia. 2015;29(8):1721-9. 33. Rosinol L, Oriol A, Teruel AI, et al. Superiority of bortezomib, thalidomide, and dexamethasone (VTD) as induction pretransplantation therapy in multiple myeloma: a randomized phase 3 PETHEMA/GEM study. Blood. 2012;120(8):15891596. 34. Cavo M, Pantani L, Pezzi A, et al. Bortezomib-thalidomide-dexamethasone (VTD) is superior to bortezomibcyclophosphamide-dexamethasone (VCD) as induction therapy prior to autologous stem cell transplantation in multiple myeloma. Leukemia. 2015;29(12):2429-2431. 35. Kumar S, Flinn I, Richardson PG, et al. Randomized, multicenter, phase 2 study (EVOLUTION) of combinations of bortezomib, dexamethasone, cyclophosphamide, and lenalidomide in previously

untreated multiple myeloma. Blood. 2012; 119(19):4375-4382. 36. 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, openlabel, phase 3 study. Lancet. 2019; 394(10192):29-38. 37. McCarthy PL, Holstein SA, Petrucci MT, et al. Lenalidomide maintenance after autologous stem-cell transplantation in newly diagnosed multiple myeloma: a metaanalysis. J Clin Oncol. 2017;35(29):32793289. 38. Holstein SA, Jung SH, Richardson PG, et al. Updated analysis of CALGB (Alliance) 100104 assessing lenalidomide versus placebo maintenance after single autologous stem-cell transplantation for multiple myeloma: a randomised, double-blind, phase 3 trial. Lancet Haematol. 2017; 4(9): e431-e442. 39. Ozaki S, Harada T, Saitoh T, et al. Survival of multiple myeloma patients aged 65-70 years in the era of novel agents and autologous stem cell transplantation. A multicenter retrospective collaborative study of the Japanese Society of Myeloma and the European Myeloma Network [Multicenter Study]. Acta Haematol. 2014;132(2):211219. 40. Minoia C, Pisapia G, Palazzo G, et al. Impact of novel agents followed by autologous hematopoietic stem cell transplantation for multiple myeloma patients aged 65 years or older: a retrospective single Institutional analysis. Bone Marrow Transplant. 2015;50(11):1486. 41. Fonseca R, Bergsagel PL, Drach J, et al. International Myeloma Working Group molecular classification of multiple myeloma: spotlight review. Leukemia. 2009; 23(12):2210-2221. 42. Avet-Loiseau H, Facon T. Front-line therapies for elderly patients with transplantineligible multiple myeloma and high-risk cytogenetics in the era of novel agents. Leukemia. 2018;32(6):1267-1276. 43. 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. 44. Jones JR, Cairns DA, Gregory WM, et al. Second malignancies in the context of lenalidomide treatment: an analysis of 2732 myeloma patients enrolled to the Myeloma XI trial. Blood Cancer J. 2016;6(12):e506. 45. Abonour R, Wagner L, Durie BGM, et al. Impact of post-transplantation maintenance therapy on health-related quality of life in patients with multiple myeloma: data from the Connect® MM Registry. Ann Hematol. 2018;97(12):2425-2436.

1967

ARTICLE Ferrata Storti Foundation

Platelet Biology & its Disorders

Antiplatelet properties of Pim kinase inhibition are mediated through disruption of thromboxane A2 receptor signaling Amanda J. Unsworth,1,2* Alexander P. Bye,1 Tanya Sage,1 Renato S. Gaspar,1 Nathan Eaton,3,4 Caleb Drew,3 Alexander Stainer,1 Neline Kriek,1 Peter J. Volberding,3,4,5 James L. Hutchinson,6 Ryan Riley,2 Sarah Jones,2 Stuart J. Mundell,6 Weiguo Cui,3,4,5 Hervé Falet3,4 and Jonathan M. Gibbins1*

Haematologica 2021 Volume 106(7):1968-1978

Institute for Cardiovascular and Metabolic Research, School of Biological Sciences, University of Reading, Reading, UK; 2Department of Life Sciences, Faculty of Science and Engineering, Manchester, Metropolitan University, Manchester, UK; 3Blood Research Institute, Versiti, Milwaukee, WI, USA; 4Department of Cell Biology, Neurobiology, and Anatomy, Medical College of Wisconsin, Milwaukee, WI, USA; 5Department of Microbiology and Immunology, Medical College of Wisconsin. Milwaukee, WI, USA and 6School of Physiology, Pharmacology & Neuroscience, Bristol, UK 1

*AJU and JMG contributed equally to this work.

ABSTRACT

Correspondence: AMANDA J. UNSWORTH a.unsworth@mmu.ac.uk Received: April 1, 2019. Accepted: May 27, 2020. Pre-published: May 28, 2020. https://doi.org/10.3324/haematol.2019.223529

1968

im kinases are upregulated in several forms of cancer, contributing to cell survival and tumor development, but their role in platelet function and thrombotic disease has not been explored. We report for the first time that Pim-1 kinase is expressed in human and mouse platelets. Genetic deletion or pharmacological inhibition of Pim kinase results in reduced thrombus formation but is not associated with impaired hemostasis. Attenuation of thrombus formation was found to be due to inhibition of the thromboxane A2 receptor as effects on platelet function were non-additive to inhibition caused by the cyclo-oxygenase inhibitor indomethacin or the thromboxane A2 receptor antagonist GR32191. Treatment with Pim kinase inhibitors caused reduced surface expression of the thromboxane A2 receptor and resulted in reduced responses to thromboxane A2 receptor agonists, indicating a role for Pim kinase in the regulation of thromboxane A2 receptor function. Our research identifies a novel, Pim kinase-dependent regulatory mechanism for the thromboxane A2 receptor and represents a new targeting strategy that is independent of cyclo-oxygenase-1 inhibition or direct antagonism of the thromboxane A2 receptor that, while attenuating thrombosis, does not increase bleeding.

Introduction The family of Pim (proviral insertion in murine lymphoma) kinases, Pim-1, -2, and -3, are highly homologous serine/threonine kinases that are widely expressed across several cell types, and are highly expressed in hematopoietic cells. Pim kinases are constitutively active and are linked with cancer progression,1,2 with overexpression and upregulation of Pim kinase activity associated with both hematologic cancers and solid tumors. They function by phosphorylating their target proteins on serine/threonine residues located within the common consensus sequence ARKRRHPS*GPPTA.1 A number of proteins that have important roles in the regulation of cellular proliferation and survival have been identified as phosphorylation targets of the Pim kinases.3-6 Expressed as a short (32 kDa) or long (44 kDa) variant, the longer variant of Pim-1 kinase, Pim-1L, kinase has also been found to regulate adenosine triphosphate-binding cassette drug transporters.7-9 Pim-1 phosphorylates both BCRP/ABCG2 and Pgp transporters enabling, through different mechanisms, the formation of drug efflux pumps.7,9 Pim kinases are highly expressed in hematopoietic cells where they are important for differentiation and development of blood cells and blood cell precursors including megakaryocytes10 and platelets.11 Whether Pim kinases are involved in the regulation of platelet function has not been explored. Analysis of the mouse haematologica | 2021; 106(7)

Pim kinase regulates TP receptor signaling

megakaryocyte transcriptome database12 identified multiple tags for both Pim-1 and Pim-2 kinases and the mRNA transcripts for all three Pim kinases have been identified in the human platelet transcriptome.13,14 Interestingly although triple knockout mice deficient in all three Pim kinase isoforms are viable, they have been shown to have altered hematopoiesis, but there is some dispute as to whether disruption of all three isoforms results in alteration of platelet count;10,11 however, platelet counts appear to be unaffected by alteration of Pim-1 expression levels in mice.15,16 Platelets rely on G protein-coupled receptors (GPCR) such as the thromboxane A2 receptor (TPaR), ADP receptors (P2Y1 and P2Y12) and the thrombin receptors (PAR1 and PAR4) to mediate platelet activation in response to vessel damage. All platelet GPCR are regulated in some way by receptor cycling/internalization from the platelet surface as well as desensitization.17 Pim-1 kinase has also been shown to have a role in the regulation of GPCR function, through modulation of surface levels of the CXCR4 receptor.18,19 Inhibition of Pim kinase prevents Pim kinase-dependent phosphorylation of CXCR4 at Ser339 and modification of the CXCR4 intracellular C terminal domain, resulting in reduced surface expression and signaling. In this study we report the presence of Pim-1 in human and mouse platelets, and reduced thrombosis in Pim-1 null mice, and following pharmacological inhibition of Pim kinase, but with no associated effect on hemostasis. We describe a novel mechanism of action by which Pim kinase inhibitors negatively regulate TPaR signaling.

Methods Procedures and experiments using human blood were approved by the University of Reading Research Ethics Committee and protocols involving mice were performed according to the National Institutes of Health and Medical College of Wisconsin Institutional Animal Care and Use Committee guidelines and as following procedures approved by the University of Reading Research Ethics Committee. Platelet isolation, thrombus formation assays, tail bleeding experiments, platelet function tests, aggregometry, granule secretion, flow cytometry, calcium imaging, immunoblotting, image analysis, statistical analyses and materials used are described in the Online Supplementary Methods.

Results Expression of Pim kinase in human and mouse platelets Pim kinases are highly expressed in hematopoietic cells.10,11 mRNA transcripts for all three Pim kinases have been identified in the human and mouse platelet transcriptomes13,14 and HaemAtlas mRNA expression profiles in hematopoietic cells20 show high expression of Pim-1 in megakaryocytes and moderate expression in platelets (Figure 1, Online Supplementary Figure S1).21 Western blot analysis of platelet lysates identified a protein band of 44 kDa apparent molecular mass in both human and mouse platelet lysates indicating the expression of the larger Pim-1 variant (Pim-1L). A protein band at 32 kDa in mouse platelets also suggested expression of the smaller haematologica | 2021; 106(7)

Pim-1S. K562 and Jurkat cell lines were included as positive controls8,18,22 (Figure 1B).

Reduced thrombus formation in Pim-1-deficient mice To determine whether Pim-1 plays a role in the regulation of platelet function and thrombosis, we measured the ability of Pim-1-deficient mouse platelets, taken from constitutive Pim-1-deficient mice, to form thrombi on collagen under arterial flow in vitro. Constitutive Pim1–/– mice were as described previously15,16 and global deletion of Pim-1 was confirmed by polymerase chain reaction analysis of genomic DNA (Online Supplementary Figure S2A). Whole blood from Pim-1-/- or Pim-1+/- mice was perfused over collagen-coated (100 mg/mL) Vena8 biochips for 4 min at an arterial shear rate of 1000 s-1. Thrombus formation was significantly attenuated in blood from Pim-1-/- mice compared to controls, indicating that Pim-1 plays a positive role in the regulation of platelet function and thrombus formation on collagen (Figure 2A). Constitutive Pim-1-/- mice show unaltered platelet counts and no difference in expression levels of major platelet adhesion receptors GPIba, GPIbβ, GPIX, GPV, GPVI and integrins β1 and β3 was observed in Pim-1-/- platelets compared to the levels in controls (Online Supplementary Figure S2B). Interestingly, despite the reduced ability to form thrombi, Pim-1-deficient mice showed no alteration in hemostasis as tail bleeding was unaffected compared to that of littermate controls (Figure 2B).

Pim kinase inhibitors reduce thrombus formation but do not disrupt hemostasis As genetic deletion of Pim-1 in mice resulted in reduced in vitro thrombus formation, we assessed the effects of the Pim kinase inhibitor AZD1208 (100 mM) on thrombus formation in human whole blood. The effect of the Pim kinase inhibitor AZD1208 (100 mM) on thrombus formation on collagen in human whole blood in vitro was also assessed. Human whole blood was pre-incubated with vehicle control or AZD1208 and perfused over collagencoated (100 mg/mL) Vena8 biochips at either an arterial shear rate (20 dynes/cm3 for 10 min) or a pathological shear rate (135 dynes/cm3 for 5 min). Similarly to what had been observed in Pim-1-deficient mice, a reduction in thrombus formation and stability on collagen under flow in vitro was also observed following AZD1208 treatment in comparison to that seen in vehicle-treated controls under arterial shear conditions (Figure 2C). While thrombus size and stability appeared reduced at arterial flow rates, the early stages of thrombus formation, including initial adhesion, appeared unaffected by AZD1208 treatment. This was further supported by the lack of inhibition of platelet adhesion and spreading on collagen caused by AZD1208 under static conditions (Online Supplementary Figure S3), indicating that the initial adhesion to collagen is not affected by Pim kinase inhibition. Interestingly, enhanced inhibition of thrombus formation on collagen was observed following treatment with AZD1208 at pathological shear rates (~80% inhibition) (Figure 2D) in comparison to the inhibition observed at an arterial shear rate (~50% inhibition). In contrast to arterial and pathological shear rates, a slight (but not significant) reduction in thrombus formation was observed in AZD1208-treated platelets compared to vehicle-treated controls under venous shear conditions (Figure 2E). Inhibitor-treated platelets appeared to form ‘woolly’ or 1969

A.J. Unsworth et al. A

Figure 1. Expression of Pim kinase in human and mouse platelets. (A) HaemAtlas analysis of Pim kinase mRNA expression levels. Pim kinase mRNA levels were quantified in human megakaryocytes and a range of blood cells by analysis of gene array data. Megakaryocytes (MK), human erythroblasts (EB), human umbilical vein endothelial cells (HUVEC), monocytes (CD14), granulocytes (CD66), mature B cells (CD19), natural killer cells (CD56), cytotoxic T cells (CD*) and helper T cells (CD4); 10+ (lighter colors) was deemed high expression. (B) Human and mouse washed platelets (three preparations) were lysed in SDS PAGE Laemmli sample buffer, separated on SDS PAGE gels and transferred to PVDF membranes before immunoblotting (IB) with anti-Pim-1 antibody. K562 and Jurkat cell lysates were included as positive controls. Actin was included as a loading control. Representative blots are shown.

‘loose’ aggregates compared to the aggregates formed in vehicle-treated controls, but no difference in fluorescence intensity of platelets adhered was observed. These findings indicate that Pim kinase inhibition does not significantly alter platelet adhesion and thrombus formation at low shear rates compared to the effects observed at higher shear rates. To examine whether Pim kinase inhibition could regulate thrombosis in vivo, we performed intravital microscopy following ferric-chloride-induced injury in mice pretreated with AZD1208 (100 µM) or vehicle control (Figure 2F). As with in vitro thrombus formation, treatment with AZD1208 resulted in a dramatic attenuation in the ability of platelets to form thrombi in vivo, at the site of ferric-chloride-induced injury with significantly prolonged occlusion times observed in AZD1208-treated mice (1463±37 s) compared to those in vehicle-treated controls (697±72 s). This supports a role for Pim kinase in the positive regulation of platelet function, and the antiplatelet properties of Pim kinase inhibitors. Interestingly, despite the dramatic attenuation in the ability of platelets to form thrombi in vivo, as with genetic deletion, pharmacological inhibition of Pim kinase was not associated with altered hemostasis. Tail bleeding assays performed in mice indicated that treatment with AZD1208 (100 mM) did not cause any significant increase in bleeding (Figure 2G), suggesting that despite the observed inhibitory effect on thrombus formation, normal hemostasis is not compromised following inhibition or genetic deletion of Pim kinase.

Pim kinase inhibitors reduce platelet aggregation To determine how Pim kinase plays a role in the regulation of platelet function, human washed platelets were pretreated for 10 min with a range of concentrations of the pan-Pim kinase inhibitor AZD1208 before stimulation with platelet agonists. As shown in Figure 3A-E pretreatment of platelets with AZD1208 inhibited aggregation stimulated by collagen (1 mg/mL) or the thromboxane A2 (TxA2) mimetic U46619 (0.3 mM). A slight inhibition was also observed in CRP-XL stimulated platelets (0.3 mg/mL). In contrast no inhibition of thrombin- (0.03 U/mL) or ADP- (10 mM) induced platelet aggregation was observed following treatment with increasing concentra1970

tions of AZD1208 (up to 10 mM). Treatment with four other structurally unrelated Pim kinase inhibitors, PIM447 (LGH-447), SGI-1776, SMI-4a and CX6258, inhibited platelet aggregation stimulated by either collagen or U46619 but not thrombin, recapitulating findings with AZD1208 and supporting a Pim kinase-dependent mode of action (Online Supplementary Figure S4). Pim kinases have been shown previously to play important roles in cell survival as Pim-2 is known to phosphorylate and inactivate Bcl-2-associated death promoter (BAD).23 Inhibition of Pim kinase enables activation of BAD and initiation of apoptosis.5 If Pim kinase inhibitors activate apoptosis in platelets, this could cause the observed reduction in aggregation. To investigate whether Pim kinase inhibition triggers apoptosis in platelets, phosphatidylserine exposure (a marker of membrane flippage) was determined by measuring annexin V binding by flow cytometry following treatment for 2 h with AZD1208 or the BCL-2 inhibitor ABT-263, an activator of apoptosis in platelets (Online Supplementary Figure S5A). ABT-263 treatment caused an increase in annexin V binding but no difference was observed in AZD1208treated platelets compared to vehicle-treated controls over the same incubation time. Furthermore, caspase cleavage did not occur in platelets following 2 h of treatment with Pim kinase inhibitors: AZD1208 (10 mM), SGI1776 (10 mM), SMI-4a (30 mM) or CX6258 (10 mM), but did occur following treatment with the BCL-2 inhibitor ABT-263 (Online Supplementary Figure S5B). Together these observations suggest that Pim kinase inhibition does not initiate platelet apoptosis.

Pim kinase inhibitors reduce thromboxane A2-mediated a-granule secretion and integrin activation

Alpha granule secretion and activation of integrin aIIbβ3 are critical events in platelet activation and aggregation. We investigated the effects of AZD1208 (10 µM) on agranule secretion and aIIbβ3 activation by measuring surface P-selectin exposure and fibrinogen binding, respectively. AZD1208 (10 mM) inhibited a-granule secretion and integrin activation evoked by collagen, CRP-XL and U46619 compared to the effects in vehicle-treated conhaematologica | 2021; 106(7)

Pim kinase regulates TP receptor signaling

A (i)

C (i)

D (i)

E (i)

(ii)

F (i)

(ii)

F (ii)

Figure 2. Genetic deletion and pharmacological inhibition of Pim-1 kinase attenuates thrombus formation on collagen, but does not cause bleeding. (A) DiOC6 loaded mouse whole blood from Pim1+/- (black) or Pim1-/- (red) mice was perfused through collagen-coated (100 mg/mL) Vena8Biochips at a shear rate of 1500 s-1. (i) Representative images taken at the end of recording are shown. (ii) Thrombus formation was determined after 4 min by comparing the percentage area covered. (B) Tail bleeding in Pim1+/- or Pim1-/- mice represented as time to cessation of bleeding (s). (C-E) DiOC6 loaded human whole blood was pretreated with vehicle (black) or 100 mM AZD1208 (red) for 10 min before perfusion through collagen-coated (100 mg/mL) Vena8Biochips at (C) an arterial shear rate of 20 dyn/cm2, (D) a pathological shear rate of 135 dyn/cm2, or (E) a venous shear rate of 4.5 dyn/cm2. The concentration of 100 mM AZD1208 was chosen because of the reduced bioavailability of AZD1208 in plasma. Thrombus formation was determined after 5 min (pathological shear) or 10 min (arterial and venous shear) by comparing percentage of maximum vehicle-treated fluorescence intensity, which measures both surface area coverage and thrombus size, in the vehicle and treated samples. (i) Representative images taken at the end of recording are shown. (ii) Data expressed as percentage of maximum vehicle-treated fluorescence intensity. (F) Thrombus formation was determined in vivo following ferric chloride-induced injury in mice pretreated for 10 mins with vehicle or 100 µM AZD1208. DiOC6 was used to enable visualization of platelets. (i) representative images taken at 0, 300, 600 and 900 s. (ii) Data expressed as time to occlusion (s). (G) Tail bleeding determined as time to cessation of bleeding (s) in mice pretreated with vehicle or 100 mM AZD1208 for 10 min. Results are mean ± standard error of mean for n≥3, *P≤0.05, **P≤0.01 ****P≤0.001 in comparison to vehicle control; where normalized data are shown statistics were performed prior to normalization.

trols (Online Supplementary Figure S6). This suggests that Pim kinase inhibitors inhibit aggregation by reducing both integrin activation and secretion of granule contents.

Pim kinase inhibitors reduce platelet activation to GPVI via reduced TPaR signaling TxA2 is synthesized and released following platelet activation by several platelet agonists. It acts as a secondary mediator, boosting platelet responses to other agonists, and is essential for the amplification of platelet activation and thrombus formation. Further investigation of the inhibitory actions of AZD1208 identified significant inhibition of U46619-evoked aggregation with half maximal inhibitory concentration (IC50) values <10 mM, concentrations similar to those achieved in plasma in patients taking AZD1208,24 following stimulation by low concentrations of U46619 (0.03-1 mM) (Figure 4A). In contrast haematologica | 2021; 106(7)

the inhibitory activity of AZD1208 could be overcome at higher concentrations of U46619 (3 and 10 mM). Collagen-induced platelet activation is known to be dependent on the release of secondary mediators. To test whether the inhibition of collagen-induced platelet aggregation by AZD1208 was a result of reduced activation of the TPaR, platelet aggregation in response to a range of collagen concentrations was measured following treatment with AZD1208 (10 mM). We performed the experiments in the presence or absence of indomethacin (10 mM), a cyclo-oxygenase (COX) inhibitor that prevents TxA2 synthesis, or a TPaR antagonist GR32191 (100 nM) to investigate whether inhibition was additive. As shown in Figure 4B, both indomethacin and GR32191 caused an inhibition of collagen-induced platelet aggregation, but this was not additive to the inhibition caused by AZD1208, suggesting that AZD1208 shares a common 1971

A.J. Unsworth et al. A (i)

B (i)

C (i)

D (i)

(ii)

(iii)

E (i)

Figure 3. AZD1208 inhibits platelet aggregation. (A-E, i and ii) Washed human platelets were pretreated with increasing concentrations of AZD1208 (0.1, 1, 10 µM) prior to stimulation with (A) collagen (1 mg/mL), (B) CRP-XL (0.3 mg/mL), (C) U46619 (0.3 mM), (D) thrombin (0.03 U/mL) or (E) ADP (10 mM) and aggregation monitored using optical light transmission aggregometry. (i) Representative traces and (ii) quantified data are shown. (A-E, iii) Human washed platelets were pretreated with 10 mM AZD1208 (red) or vehicle (black) prior to stimulation with (A) collagen (0.01-10 mg/mL), (B) CRP-XL (1-3 µg/mL), (C) U46619 (3 nM - 3 mM), (D) thrombin (0.01-1 U/mL) or (E) ADP (0.1-100 µM) and aggregation was monitored after 5 min using an optical light transmission plate-based aggregometry assay; quantified data are shown. Results are mean ± standard error mean for n≥3, *P≤0.05 ***P≤0.005 in comparison to vehicle control; where normalized data are shown statistics were performed prior to normalization.

TPaR.-dependent mechanism. The inhibition observed in collagen-stimulated platelets is likely due to the loss of TxA2 signaling, indicating that Pim kinase has a positive regulatory role in the regulation of TPaR signaling. Similar results were also observed following treatment with structurally unrelated Pim kinase inhibitors SGI1776, PIM-447 (LGH447), SMI-4a and CX6258 in the presence or absence of indomethacin (Online Supplementary Figure S4D-G). In further support of this, we also observed that AZD1208-mediated inhibition of thrombus formation on collagen under flow occurred via a TPaR-dependent mechanism, as indomethacin (10 mM) did not cause further inhibition of thrombus formation when combined with AZD1208 treatment (Figure 4C). In support of the effects of AZD1208 being mediated via Pim kinase, and a role for Pim kinase in the regulation of platelet TPaR signaling, platelets from Pim1-/- mice (red) displayed reduced fibrinogen binding, compared to controls (black), following stimulation with U46619 (10 mM) or CRP-XL (10 mg/mL) but not following stimulation with thrombin (0.01 U/mL) (Figure 4D).

Pim kinase inhibitors inhibit platelet function independently of COX1 The synthesis of TxA2 in platelets is dependent on activation of COX1 following platelet stimulation. To determine whether the inhibitory effects of Pim kinase inhibitors were due to altered COX activity, thromboxane B2 (TxB2) levels (a marker of TxA2 release) were determined following stimulation of platelets with 1972

arachidonic acid (a direct substrate for COX1 and the TxA2 synthesis pathway) or collagen and compared to the levels following treatment with indomethacin, a COX1 inhibitor, in the presence of the TPaR antagonist GR32191 to remove secondary TPaR feedback mechanisms. While indomethacin caused almost complete ablation of both collagen- and arachidonic acid-induced platelet TxB2 generation, AZD1208 did not cause any significant alterations in TxB2 generation indicating that it does not regulate COX1 activity (Figure 5A). In further support of AZD1208 mediating its inhibitory actions via inhibition of TPaR receptor signaling and not inhibition of cyclo-oxygenase activity, while AZD1208 causes inhibition of U46619-mediated platelet aggregation, concentrations of indomethacin that cause ablation of platelet responses to arachidonic acid are unable to inhibit platelet activation by the TPaR agonist U46619 (Figure 5A).

Pim kinase inhibitors reduce signaling events downstream of TPaR The TPaR receptor is coupled to both Gq and Ga13 proteins. Gq couples the TPaR to phospholipase C which in turn regulates calcium mobilization and the activation of protein kinase C (PKC), key mediators of granule secretion and activation of integrin aIIbβ3. Ga13 regulates the Rho/Rho-kinase signaling pathway which regulates the phosphorylation of myosin IIa and is important for the regulation of cytoskeletal rearrangements and platelet shape change. haematologica | 2021; 106(7)

Pim kinase regulates TP receptor signaling

C (i)

B (i)

C (ii)

(ii)

Figure 4. Inhibition of thromboxane A2 signaling underlies inhibition of collagen-induced aggregation and thrombus formation by Pim kinase inhibitors. (A, B) Human washed platelets were treated with (A) increasing concentrations of AZD1208 (1-100 mM) or vehicle control prior to stimulation with increasing concentrations of U46619 (0.03-10 mM) or (B) 10 mM AZD1208 in the presence or absence of (i) indomethacin (10 µM) or (ii) the thromboxane A2 receptor antagonist GR32191 (100 nM). Platelet aggregation was monitored after 5 min of stimulation by collagen (0.01-10 mg/mL) using a 96-well plate based aggregometry assay. (C) DiOC6 loaded human whole blood was pretreated with vehicle (black) or 100 mM AZD1208 (red), in the presence of 10 mM indomethacin for 10 min before perfusion through collagen coated (100 mg/mL) Vena8Biochips at a shear rate of 20 dyn/cm2. Thrombus formation was determined over 10 min by comparing fluorescence intensity in the vehicle and treated samples. (i) Representative images taken at 10 min. (ii) Data expressed as the percentage of maximum fluorescence of vehicle treated cases and normalized to an untreated (no indomethacin treatment) control, where the maximum fluorescence observed in untreated platelets is considered to be 100% thrombus formation. (D) Fibrinogen binding in washed platelets from control (Pim1+/-) or Pim1-/- mice was determined following stimulation with thrombin (0.01 U/mL), U46619 (10 mM) or CRP (10 mg/mL) and expressed as percentage of positive cells. Results are mean + standard error of mean for n≥3, *P≤0.05 ***P≤0.005 in comparison to vehicle control; where normalized data are shown, statistics were performed prior to normalization.

To determine whether Pim kinase regulates processes downstream of TPaR coupled G proteins, levels of intracellular calcium, PKC activity and myosin light chain (MLC) phosphorylation were monitored following stimulation with U46619. As shown in figure 5B, treatment of platelets with AZD1208 caused a significant reduction in calcium mobilization following stimulation with U46619 (0.3 mM) compared to vehicle-treated control platelets. In contrast no significant difference in calcium mobilization was observed following stimulation with ADP (10 mM) supporting a specific role for Pim kinase in the regulation of TPaR signaling. AZD1208 also caused a reduction both in PKC activity and MLC (S19) phosphorylation compared to vehicle controls following stimulation with U46619 (1 mM) (Figure 5C and D). Ga13 is also associated with the regulation of integrin aIIbβ3 outside in signaling.25 Phosphorylation of Y773 on the integrin β3 tail which is essential for propagation of outside-in signaling was reduced in U46619 stimulated platelets pretreated with AZD1208 compared to vehicle-treated control platelets (Figure 5E). Taken together these results support a role for Pim kinase in the positive regulation of Gq and Ga13 signaling downstream of TPaR activation.

AZD1208 is not a competitive antagonist of the TPaR As components of both Gq and G13 signaling, pathways immediately downstream of TPaR activation, were haematologica | 2021; 106(7)

found to be inhibited following treatment with AZD1208 we hypothesized that the inhibitor was having direct effect on the function of the TPaR itself. One potential mechanism of action could be that Pim kinase inhibitors such as AZD1208 act as antagonists of the TPaR and inhibit the TPaR directly, independently of Pim kinase. To determine whether AZD1208 acts as a competitive TPaR antagonist, platelet aggregation was measured following treatment with increasing concentrations of AZD1208 and stimulation with a range of U46619 concentrations so that the concentration relationship between inhibitor and antagonist could be quantified by Schild analysis. As shown in Figure 6A the inhibitory effect of AZD1208 became saturated by 10 mM, with higher concentrations (30, 50, 100 mM) unable to achieve greater levels of inhibition. In contrast, increasing concentrations of the competitive TPaR antagonist GR32191 (Figure 5) caused non-saturable inhibition of aggregation stimulated by U46619 and generated a linear Schild plot with a pA2 of 8.9 which was consistent with the reported properties of this competitive antagonist acting at the TPaR.26 In contrast, the Schild plot for AZD1208 was linear up to a concentration of 10 mM at which point the inhibitory effects were saturated and increasing concentrations of AZD1208 no longer altered the apparent EC50 of U46619. These results indicate that the concentrationresponse relationship of GR32191 and U46619 in the 1973

A.J. Unsworth et al. A (i)

B (i)

D (i)

(ii)

C (i)

(ii)

E (i)

(ii)

Figure 5. AZD1208 inhibits thromboxane receptor signaling. (A) Resting and stimulated human washed platelets were treated with 10 mM AZD1208 for 10 min and stimulated with (i) collagen (1 mg/mL) or arachidonic acid (1 mM) or (ii) U46619 (0-3 mM) in the presence and absence of indomethacin (10 mM). (i) TxB2 levels (EnzoLife Sciences ELISA) or (ii) aggregation were monitored after 5 min of shaking. (B) Mobilization of intracellular calcium was determined in FURA-2 AM loaded platelets following stimulation with U46619 (300 nM) or ADP (10 mM). (i) Representative traces and (ii) quantified data are shown, with data expressed as the change in [Ca2+] (nM). (C-E) Human platelets were pre-incubated with vehicle or AZD1208 (10 mM) for 10 min and stimulated with U46619 (1 mM) for 30 s or 3 min before lysis in SDS Laemmli sample buffer. (C) Protein kinase C (PKC) activity was determined by blotting these samples and using a phospho-site specific antibody (for the PKC substrate recognition sequence) that detects PKC substrate phosphorylation. (D) Myosin light chain (MLC) phosphorylation at Ser19 was determined using a phospho-specific antibody that recognizes the phosphorylated MLC. (E) Phosphorylation of the integrin β3 subunit at Y773 was determined using a phospho-specific antibody. Actin was used to confirm equal loading. (i) Representative blots and (ii) quantified data are shown. Levels of total phosphorylation were quantified and expressed as a percentage of the maximum phosphorylation observed in vehicle-treated, stimulated controls. Results are mean + standard error of mean for n≥3, *P≤0.05 in comparison to vehicle controls.

aggregation assay was consistent with that of a TP receptor agonist and antagonist competing to bind to the orthosteric site, while the results with AZD1208 do not conform to this model.

AZD1208 alters TPaR receptor surface expression It has been described previously that Pim kinase modulates levels of the CXCR4 receptor at the surface of chronic lymphocytic leukemia cells.18,19 We hypothesized that Pim kinase inhibitors could modulate TPaR function via a similar mechanism in platelets and measured expression levels of TPaR following treatment with AZD1208 using flow cytometry to investigate this further. A TPaR antibody that recognizes the N-terminal (extracellular) region of the TPaR was used to determine surface expression levels of TPaR on platelets (Figure 6C). As previously described,27 stimulation of platelets with U46619 was associated with a reduction in cell surface levels of TPaR, compared to the levels on unstimulated platelets, due to receptor internalization. Resting platelets treated with AZD1208 (10 mM) showed reduced surface levels of the TPaR receptor compared to the levels in vehicle-treated 1974

controls with total levels of TPaR unaffected by AZD1208 treatment. U46619- (1 mM) stimulated platelets pretreated with AZD1208 also showed reduced surface expression levels of TPaR compared to both those of vehicle-treated controls and of platelets treated with U46619 only. These findings indicate that reduced U46619 signaling following treatment with AZD1208 is linked to reduced surface expression levels of TPaR.

AZD1208 inhibits CXCR4 signaling in human platelets CXCR4 is expressed in platelets and is able to signal following stimulation with its ligand SDF-1a.28-30 To determine whether inhibition of Pim kinase alters CXCR4 signaling in human platelets, aggregometry following stimulation with SDF-1a (200 ng/mL) was performed in vehicle- and AZD1208-treated human platelet-rich plasma. SDF-1a caused a modest level of platelet aggregation (~50%) in vehicle-treated platelets which was significantly reduced in AZD1208- (100 mM) treated samples (~20% aggregation) (Figure 6D). These observations indicate that in addition to regulation of the TPaR, Pim kinase also regulates platelet CXCR4 receptor function. haematologica | 2021; 106(7)

Pim kinase regulates TP receptor signaling

A (i)

C (i)

D (i)

(ii)

B (i)

(ii)

(iii)

(ii)

Figure 6. AZD1208 reduces TPaR surface expression and signaling and does not act as a competitive antagonist of the TPaR. (A, B) Human washed platelets were pre-treated with: (A) AZD1208 (1, 3, 10, 30, 50 and 100 mM) or (B) GR32191 (1, 2, 3, 5, 10, 30, 100 nM) prior to stimulation with U46619 (3 nM–3 mM) and aggregation was monitored after 5 min using an optical light transmission plate-based aggregometry assay. Quantified data are shown. (i) Percentage aggregation. (ii) EC50 values from the aggregation dose-response curves determined following incubation with AZD1208 were used to plot a Schild regression plot to determine whether AZD1208 acts as an antagonist for the TxA2 receptor. (C) Platelets were treated with AZD1208 (10 mM; 10 mins) or vehicle control. Surface expression of TPaR was assessed using an antibody that recognizes the extracellular portion of the TPaR and detected by flow cytometry. Samples were diluted in HBS and not fixed to avoid disruption of the membrane. Anti-DOK6 antibody was included as a negative control. (i) A representative histogram from the flow cytometer. (ii) Data are expressed as median fluorescent intensity (MFI). (iii) Total cellular TPaR was detected by western blotting. (D) Human platelet-rich plasma was pretreated with AZD1208 (100 mM) or vehicle, as a control, for 10 min prior to stimulation with SDF-1a (200 ng/mL) and aggregation was monitored using optical light transmission aggregometry for 5 min. (i) Representative trace and (ii) quantified data are shown. Results are mean + standard error of mean for n≥3, *P≤0.05, **P≤0.01, ***P≤0.005 in comparison to vehicle-treated control; where normalized data are shown, statistics were performed prior to normalization.

Discussion Development of kinase inhibitors as potential therapeutics for solid tumors and hematologic malignancies has been fueled by the recent successes of kinase inhibitor therapy for cancers.1,2,31,32 Pim kinase is known to enhance cancer progression and drug resistance, and loss of all three Pim kinase isoforms does not affect embryo viability, indicating that inhibition of Pim activity is likely to be tolerable.10,11,33 In addition to its well-established role in the regulation of cycle progression and prevention of cellular apoptosis,3-6,23 Pim kinase has also been shown to play roles in cell migration, potentially contributing to metastasis and cell invasion18,19 and is also implicated in drug resistance through activation of multidrug resistance transporters.7-9 Pim kinase is therefore seen as a promising potential drug target. haematologica | 2021; 106(7)

Pim kinases have been shown to be highly expressed in hematopoietic cells with important roles in the development and differentiation of megakaryocytes10 and platelets.11,15 Deletion of Pim-1 alone has no effect on the hematopoietic system,16 possibly indicating a level of redundancy between the Pim kinase family members. Kinases, however, often have broad expression profiles across several different cell types, which increases the risk of kinase inhibitors having unwanted side effects. Platelets in particular rely heavily on kinase-driven signaling cascades to enable them to function effectively in response to vascular damage. Several kinase inhibitors have been reported that affect the ability of platelets to activate and are associated with an increased risk of bleeding.34-36 Western blot analysis identified expression of both the 1975

A.J. Unsworth et al.

44 kDa and 32 kDa variants of Pim-1 kinase in human and mouse platelets. While Pim-2 and Pim-3 were not identified in our assays, expression of either paralog in human and mouse platelets cannot be ruled out. Comparison of thrombus formation on collagen under flow in vitro established that platelets from Pim-1-deficient mice showed significant attenuation in comparison to that in wild-type controls, highlighting a role for Pim1 kinase in the regulation of platelet function and thrombus formation under arterial flow on collagen. Despite the reduction in thrombotic potential, deletion of Pim-1 was not associated with altered hemostasis, indicating that drugs targeting Pim kinase activity could offer an antithrombotic therapeutic strategy that is not associated with the increased bleeding risk usually observed with other antiplatelet agents. In support of the antithrombotic potential of Pim kinase inhibitors we determined that several structurally different Pim kinase inhibitors, AZD1208, PIM-447, SGI1776, SMI-4a and CX6258, which are pan-Pim kinase inhibitors that target all three Pim kinases, all caused inhibition of platelet functional responses, including aggregation in response to GPVI agonists collagen and CRP-XL, CXCR4 ligand SDF-1a (CXCL12) and TxA2 mimetic U46619, but not to other GPCR agonists including thrombin, TRAP6 and ADP, with an inhibitory profile similar to that observed by Lordkipanidzé et al. in a patient with a mutation in the TP receptor.27,37 Furthermore inhibition of collagen-induced platelet responses was found to be due to an inhibition of TPaR signaling, as the level of platelet aggregation observed in the presence of indomethacin or the TPaR antagonist GR32191 was not further decreased following treatment with AZD1208 (10 mM). Similarly to the reduced thrombus formation observed in whole blood from Pim-1-deficient mice, AZD1208 caused significant attenuation of thrombus formation on collagen under flow in vitro at both arterial and pathological shear rates and a dramatic inhibition of thrombus formation in vivo (Figure 2). Interestingly, however, no effect on platelet adhesion and thrombus formation at venous flow rates was observed following treatment with AZD1208; this, combined with the lack of effect on bleeding in mice following damage to the tail vein, suggests that although Pim kinase inhibition or deficiency reduces thrombus formation under high arterial shear, this does not alter thrombus formation or hemostasis at venous or low shear. This absence of effect is likely due to the lack of inhibition of PAR or P2Y receptor-mediated platelet activation by Pim kinase inhibitors. Previous studies have shown that while GPVI deficiency or inhibition has varying effects in tail bleeding assays,38,39 PAR or P2Y12 deficiency or P2Y1 inhibition results in significant increases in tail bleeding and alteration of hemostasis in mice.39-41 As initial adhesion to collagen is unaffected following treatment with AZD1208, maintenance of PAR and P2Y receptor responses may compensate for the lack of TP receptor signaling, allowing for normal hemostasis. The lack of bleeding effect following treatment with AZD1208 is consistent with the lack of reported bleeding-related adverse effects in patients in a recent phase I clinical study investigating the efficacy of AZD1208 in solid and hematologic cancers.24 This provides further evidence that Pim kinase inhibitors may not be associated with drug-induced platelet dysfunction-related bleeding events and points to future use of Pim kinase inhibitors as 1976

a possible treatment strategy for individuals with increased risk of cardiovascular disease and atherosclerosis, conditions associated with pathological shear rates. Investigation into how Pim kinases elicit their inhibitory effects on TPaR signaling, revealed that AZD1208 inhibited TPaR downstream signaling events, pointing to upstream regulation of TPaR signaling most likely via direct regulation of the TPaR. It has previously been described that Pim-1 kinase regulates CXCR4 activity in Jurkat and chronic lymphocytic leukemia cells via regulation of surface expression levels of the receptor.18,19 Inhibition or deletion of Pim kinase reduces surface expression levels of CXCR4 and inhibits CXCL12/SDF-1 signaling.29,30 Similarly, in our study we identified reduced surface expression levels of TPaR in platelets following treatment with AZD1208 compared to the levels in vehicle-treated controls. The TPaR signaling pathway is a key target to reduce cardiovascular disease-related thrombotic events and inflammation in patients. Despite widespread use, currently available GPCR-targeted therapies are associated with variable outcomes and adverse side effects in patients. Aspirin is the ‘gold standard’ antiplatelet agent for the prevention of arterial thrombosis. Aspirin targets platelet TPaR signaling via an indirect route, by inhibiting COX1, the enzyme that controls synthesis of TxA2. Aspirin, however, has dose-limiting off-target effects on COX2, an enzyme involved in synthesis of endogenous inhibitors of platelet function that can increase the risk of thrombosis when repressed. Aspirin is also less effective in patients suffering from diabetes, hypertension and obesity, and is associated with an increased risk of severe bleeding42 with one recent meta-analyses suggesting that in low-risk individuals the harms of aspirin outweigh the cardiovascular benefits.43-45 Our findings suggest that Pim kinases may be a safer target to control thrombosis. The mechanism underlying the regulation of TPaR signaling by Pim-1 kinase requires further investigation. In other cell types the 44 kDa variant of Pim-1 kinase, Pim1L is localized to the plasma membrane and associated with the phosphorylation and regulation of several membrane proteins, while the smaller 32 kDa variant Pim-1S predominantly localizes to the cytosol and nucleus.7,46 One possibility is that Pim kinase phosphorylates the TPαR. controlling its surface expression levels, similar to the mechanism observed for CXCR4 in chronic lymphocytic leukemia cells. Sequence alignments of the Pim kinase substrate recognition sequence have identified four putative Pim kinase phosphorylation sites within the TPaR sequence, including one within the first intracellular loop, a region that has previously been shown to be associated with TPaR surface expression and receptor function.27 Pim kinase inhibition may therefore disrupt phosphorylation of the receptor and reduce surface expression via increasing its internalization or preventing the dynamic process of receptor recycling to the surface. It is also possible that Pim kinase does not phosphorylate the receptor directly and instead orchestrates interactions with other proteins that regulate receptor surface expression.18,19 This work identifies a novel, Pim kinase-dependent regulatory mechanism for the TPaR and represents a new targeting strategy that is independent of COX1 inhibition or direct antagonism of the TPaR that, while reducing thrombosis, does not increase the risk of bleeding. haematologica | 2021; 106(7)

Pim kinase regulates TP receptor signaling

Disclosures No conflicts of interest to disclose.

Mitchell and Mike Fry, University of Reading, for their help with the work and preparation of this manuscript.

Contributions AJU designed the research, performed experiments, analyzed results and wrote the paper; APB, TS, RSG, NE, CD, AS, NK, PJV, LH, RR, SJ, SM and WC performed experiments and analyzed results; HF and JMG designed the research and wrote the paper. Acknowledgments The authors would like to thank Gemma Little, Joanne

References 1. Pogacic V, Bullock AN, Fedorov O, et al. Structural analysis identifies imidazo[1,2b]pyridazines as PIM kinase inhibitors with in vitro antileukemic activity. Cancer Res. 2007;67(14):6916-6924. 2. Shah N, Pang B, Yeoh KG, et al. Potential roles for the PIM1 kinase in human cancer a molecular and therapeutic appraisal. Eur J Cancer. 2008;44(15):2144-2151. 3. Bachmann M, Moroy T. The serine/threonine kinase Pim-1. Int J Biochem Cell Biol. 2005;37(4):726-730. 4. Bachmann M, Kosan C, Xing PX, Montenarh M, Hoffmann I, Moroy T. The oncogenic serine/threonine kinase Pim-1 directly phosphorylates and activates the G2/M specific phosphatase Cdc25C. Int J Biochem Cell Biol. 2006;38(3):430-443. 5. Macdonald A, Campbell DG, Toth R, McLauchlan H, Hastie CJ, Arthur JS. Pim kinases phosphorylate multiple sites on Bad and promote 14-3-3 binding and dissociation from Bcl-XL. BMC Cell Biol. 2006;7:1. 6. Aho TL, Sandholm J, Peltola KJ, Mankonen HP, Lilly M, Koskinen PJ. Pim-1 kinase promotes inactivation of the pro-apoptotic Bad protein by phosphorylating it on the Ser112 gatekeeper site. FEBS Lett. 2004;571(13):43-49. 7. Xie Y, Xu K, Linn DE, et al. The 44-kDa Pim-1 kinase phosphorylates BCRP/ABCG2 and thereby promotes its multimerization and drug-resistant activity in human prostate cancer cells. J Biol Chem. 2008;283(6):3349-3356. 8. Darby RA, Unsworth A, Knapp S, Kerr ID, Callaghan R. Overcoming ABCG2-mediated drug resistance with imidazo-[1,2-b]pyridazine-based Pim1 kinase inhibitors. Cancer Chemother Pharmacol. 2015; 76(4):853-864. 9. Xie Y, Burcu M, Linn DE, Qiu Y, Baer MR. Pim-1 kinase protects P-glycoprotein from degradation and enables its glycosylation and cell surface expression. Mol Pharmacol. 2010;78(2):310-318. 10. Mikkers H, Nawijn M, Allen J, et al. Mice deficient for all PIM kinases display reduced body size and impaired responses to hematopoietic growth factors. Mol Cell Biol. 2004;24(13):6104-6115. 11. An N, Kraft AS, Kang Y. Abnormal hematopoietic phenotypes in Pim kinase triple knockout mice. J Hematol Oncol. 2013;6:12. 12. Senis YA, Tomlinson MG, Garcia A, et al. A comprehensive proteomics and genomics analysis reveals novel transmembrane proteins in human platelets and mouse

haematologica | 2021; 106(7)

Funding This work was supported by the British Heart Foundation programme grant RG/15/2/31224 (to JMG), British Heart Foundation project grant PG/2019/34798 (to AJU), National Institutes of Health R01 grants HL126743 (to HF) and AI125741 (to WC), the Centre for Biosciences, Manchester Metropolitan University and Manchester Metropolitan University RKE Internal Funding grant 343846 (to AU)..

megakaryocytes including G6b-B, a novel immunoreceptor tyrosine-based inhibitory motif protein. Mol Cell Proteomics. 2007;6(3):548-564. 13. Weyrich AS, Zimmerman GA. Evaluating the relevance of the platelet transcriptome. Blood. 2003;102(4):1550-1551. 14. Rowley JW, Oler AJ, Tolley ND, et al. Genome-wide RNA-seq analysis of human and mouse platelet transcriptomes. Blood. 2011;118(14):e101-111. 15. An N, Lin YW, Mahajan S, et al. Pim1 serine/threonine kinase regulates the number and functions of murine hematopoietic stem cells. Stem Cells. 2013;31(6):12021212. 16. Laird PW, van der Lugt NM, Clarke A, et al. In vivo analysis of Pim-1 deficiency. Nucleic Acids Res. 1993;21(20):4750-4755. 17. Li D, D'Angelo L, Chavez M, Woulfe DS. Arrestin-2 differentially regulates PAR4 and ADP receptor signaling in platelets. J Biol Chem. 2011;286(5):3805-3814. 18. Grundler R, Brault L, Gasser C, et al. Dissection of PIM serine/threonine kinases in FLT3-ITD-induced leukemogenesis reveals PIM1 as regulator of CXCL12CXCR4-mediated homing and migration. J Exp Med. 2009;206(9):1957-1970. 19. Decker S, Finter J, Forde AJ, et al. PIM kinases are essential for chronic lymphocytic leukemia cell survival (PIM2/3) and CXCR4-mediated microenvironmental interactions (PIM1). Mol Cancer Ther. 2014;13(5):1231-1245. 20. Watkins NA, Gusnanto A, de Bono B, et al. A HaemAtlas: characterizing gene expression in differentiated human blood cells. Blood. 2009;113(19):e1-9. 21. Simon LM, Edelstein LC, Nagalla S, et al. Human platelet microRNA-mRNA networks associated with age and gender revealed by integrated plateletomics. Blood. 2014;123(16):e37-45. 22. Lin YW, Beharry ZM, Hill EG, et al. A small molecule inhibitor of Pim protein kinases blocks the growth of precursor T-cell lymphoblastic leukemia/lymphoma. Blood. 2010;115(4):824-833. 23. Yan B, Zemskova M, Holder S, et al. The PIM-2 kinase phosphorylates BAD on serine 112 and reverses BAD-induced cell death. J Biol Chem. 2003;278(46):4535845367. 24. Cortes J, Tamura K, DeAngelo DJ, et al. Phase I studies of AZD1208, a proviral integration Moloney virus kinase inhibitor in solid and haematological cancers. Br J Cancer. 2018;118(11):1425-1433. 25. Gong H, Shen B, Flevaris P, et al. G protein subunit Galpha13 binds to integrin alphaIIbbeta3 and mediates integrin "outside-in" signaling. Science. 2010;

327(5963):340-343. 26. Lumley P, White BP, Humphrey PP. GR32191, a highly potent and specific thromboxane A2 receptor blocking drug on platelets and vascular and airways smooth muscle in vitro. Br J Pharmacol. 1989;97(3):783-794. 27. Nisar SP, Lordkipanidze M, Jones ML, et al. A novel thromboxane A2 receptor N42S variant results in reduced surface expression and platelet dysfunction. Thromb Haemost. 2014;111(5):923-932. 28. Akbiyik F, Ray DM, Gettings KF, Blumberg N, Francis CW, Phipps RP. Human bone marrow megakaryocytes and platelets express PPARgamma, and PPARgamma agonists blunt platelet release of CD40 ligand and thromboxanes. Blood. 2004; 104(5):1361-1368. 29. Walsh TG, Harper MT, Poole AW. SDF1alpha is a novel autocrine activator of platelets operating through its receptor CXCR4. Cell Signal. 2015;27(1):37-46. 30. Clemetson KJ, Clemetson JM, Proudfoot AE, Power CA, Baggiolini M, Wells TN. Functional expression of CCR1, CCR3, CCR4, and CXCR4 chemokine receptors on human platelets. Blood. 2000; 96(13):4046-4054. 31. Keeton EK, McEachern K, Dillman KS, et al. AZD1208, a potent and selective pan-Pim kinase inhibitor, demonstrates efficacy in preclinical models of acute myeloid leukemia. Blood. 2014;123(6):905-913. 32. Kirschner AN, Wang J, van der Meer R, et al. PIM kinase inhibitor AZD1208 for treatment of MYC-driven prostate cancer. J Natl Cancer Inst. 2015;107(2):dju407. 33. Din S, Konstandin MH, Johnson B, et al. Metabolic dysfunction consistent with premature aging results from deletion of Pim kinases. Circ Res. 2014;115(3):376-387. 34. Bye AP, Unsworth AJ, Vaiyapuri S, Stainer AR, Fry MJ, Gibbins JM. Ibrutinib inhibits platelet integrin alphaIIbbeta3 outside-in signaling and thrombus stability but not adhesion to collagen. Arterioscler Thromb Vasc Biol. 2015;35(11):2326-2335 35. Gratacap MP, Martin V, Valera MC, et al. The new tyrosine-kinase inhibitor and anticancer drug dasatinib reversibly affects platelet activation in vitro and in vivo. Blood. 2009;114(9):1884-1892. 36. Levade M, Severin S, Gratacap MP, Ysebaert L, Payrastre B. Targeting kinases in cancer therapies: adverse effects on blood platelets. Curr Pharm Des. 2016; 22 (16):2315-2322. 37. Lordkipanidze M, Lowe GC, Kirkby NS, et al. Characterization of multiple platelet activation pathways in patients with bleeding as a high-throughput screening option: use of 96-well Optimul assay. Blood.

1977

A.J. Unsworth et al. 2014;123(8):e11-22. 38. Nieswandt B, Watson SP. Platelet-collagen interaction: is GPVI the central receptor? Blood. 2003;102(2):449-461. 39. Bynagari-Settipalli YS, Cornelissen I, Palmer D, et al. Redundancy and interaction of thrombin- and collagen-mediated platelet activation in tail bleeding and carotid thrombosis in mice. Arterioscler Thromb Vasc Biol. 2014;34(12):2563-2569. 40. Dorsam RT, Kunapuli SP. Central role of the P2Y12 receptor in platelet activation. J Clin Invest. 2004;113(3):340-345. 41. Karim ZA, Vemana HP, Alshbool FZ, et al.

1978

Characterization of a novel function-blocking antibody targeted against the platelet P2Y1 receptor. Arterioscler Thromb Vasc Biol. 2015;35(3):637-644. 42. Hankey GJ, Eikelboom JW. Aspirin resistance. Lancet. 2006;367(9510):606-617. 43. Warner TD, Nylander S, Whatling C. Antiplatelet therapy: cyclo-oxygenase inhibition and the use of aspirin with particular regard to dual anti-platelet therapy. Br J Clin Pharmacol. 2011;72(4):619-633. 44. Desborough MJR, Keeling DM. The aspirin story - from willow to wonder drug. Br J Haematol. 2017;177(5):674-683.

45. Antithrombotic Trialists (ATT) Collaboration; Baigent C, Blackwell L, Collins R, et al. Aspirin in the primary and secondary prevention of vascular disease: collaborative meta-analysis of individual participant data from randomised trials. Lancet. 2009;373(9678):1849-1860. 46. Xie Y, Xu K, Dai B, et al. The 44 kDa Pim1 kinase directly interacts with tyrosine kinase Etk/BMX and protects human prostate cancer cells from apoptosis induced by chemotherapeutic drugs. Oncogene. 2006;25(1):70-78.

haematologica | 2021; 106(7)

ARTICLE

Red Cell Biology & its Disorders

In vitro and in vivo induction of fetal hemoglobin with a reversible and selective DNMT1 inhibitor

Aidan G. Gilmartin,1 Arthur Groy,1 Elizabeth R. Gore,1 Charity Atkins,1 Edward R. Long,1 Monica N. Montoute,1 Zining Wu,1 Wendy Halsey,1 Dean E. McNulty,1 Daniela Ennulat,1 Lourdes Rueda,1 Melissa B. Pappalardi,1 Ryan G. Kruger,1 Michael T. McCabe,1 Ali Raoof,2 Roger Butlin,2 Alexandra Stowell,2 Mark Cockerill,2º Ian Waddell,2º Donald Ogilvie,2º Juan Luengo,1º Allan Jordan2º and Andrew B. Benowitz1

GlaxoSmithKline, Collegeville, Pennsylvania, PA, USA and 2Drug Discovery Unit, Cancer Research UK Manchester Institute, University of Manchester, Alderley Park, Manchester, UK

Ferrata Storti Foundation

Haematologica 2021 Volume 106(7):1979-1987

MC current address: MediTech Media, Manchester, UK; IW current address: Charles River Laboratories, Saffron Walden, UK; DO current address: Framingham Consulting Limited, Manchester, UK; JL current address: Prelude Therapeutics, Newark, DE, USA; AJ current address: Sygnature Discovery Limited, Nottingham, UK

ABSTRACT

harmacological induction of fetal hemoglobin (HbF) expression is an effective therapeutic strategy for the management of β-hemoglobinopathies such as sickle cell disease. DNA methyltransferase (DNMT) inhibitors 5-azacytidine (5-aza) and 5-aza-2′-deoxycytidine (decitabine) have been shown to induce HbF expression in both preclinical models and clinical studies, but are not currently approved for the management of hemoglobinopathies. We report here the discovery of a novel class of orally bioavailable DNMT1-selective inhibitors as exemplified by GSK3482364. This molecule potently inhibits the methyltransferase activity of DNMT1, but not DNMT family members DNMT3A or DNMT3B. In contrast with cytidine analog DNMT inhibitors, the DNMT1 inhibitory mechanism of GSK3482364 does not require DNA incorporation and is reversible. In cultured human erythroid progenitor cells, GSK3482364 decreased overall DNA methylation resulting in derepression of the γ-globin genes HBG1 and HBG2 and increased HbF expression. In a transgenic mouse model of sickle cell disease, orally administered GSK3482364 caused significant increases in both HbF levels and in the percentage HbF-expressing erythrocytes, with good overall tolerability. We conclude that in these preclinical models, selective, reversible inhibition of DNMT1 is sufficient for the induction of HbF, and is well-tolerated. We anticipate that GSK3482364 will be a useful tool molecule for the further study of selective DNMT1 inhibition both in vitro and in vivo.

Correspondence: ANDREW B. BENOWITZ Andrew.B.Benowitz@GSK.com Received: January 30, 2020. Accepted: June 18, 2020. Pre-published: June 25, 2020. https://doi.org/10.3324/haematol.2020.248658

Introduction Adult hemoglobin (HbA) is a tetramer composed of two a-globin and two β-globin polypeptide chains (a β ) with four coordinated heme molecules, which is encoded by genes HBA1, HBA2 and HBB. Various mutations in the β-globin gene HBB cause the β-hemoglobinopathies sickle cell disease (SCD) and β-thalassemia, the most common heritable blood disorders in the world.1 In sickle cell anemia, the primary form of SCD, a missense mutation in both alleles of HBB results in an E6V substitution, producing sickle hemoglobin (a βs ; HbS). In its deoxygenated state, the E6V mutant β-globin proteins in the HbS tetramer enable hydrophobic interactions with mutant β-globin proteins in neighboring HbS tetramers, resulting in hemoglobin aggregates. These aggregates grow into rods that distort the cell into a characteristic sickle shape, increase erythroid cell rigidity, and ultimately result in cell membrane damage and 2

haematologica | 2021; 106(7)

1979

A.G. Gilmartin et al.

hemolysis. These changes in the sickle erythrocytes produce a cascade of effects that result in anemia, impaired blood flow, and painful vaso-occlusive events that ultimately cause tissue ischemia and long-term damage.2 During fetal development and until shortly after birth, erythrocytes preferentially express an alternative hemoglobin tetramer termed fetal hemoglobin (a γ ; HbF) that is composed of two γ-globin chains paired with a-globin chains rather than β-globin chains. The genes encoding for γ-globin, HBG1 and HBG2, lack the mutation that causes SCD. Consequently, symptoms of SCD first manifest several months after birth following the “hemoglobin switch”, the transition from HbF to HbA, or to HbS in the case of SCD patients.3 During the transition from HbF to HbA/HbS, the genes encoding for γ-globin, HBG1 and HBG2, are repressed by transcriptional complexes that include GATA1, TR2/TR4, MYB, KLF1, Sox6, BCL11A, LRF, DNMT1, and HDAC1/2.4-6 The repressor complexes cause significant chromatin remodeling, controlled in part through increased DNA methylation of HBG1 and HBG2 gene promoters and demethylation of the HBB gene promoter. 7, 8 Although HbF typically decreases to a few percent of total hemoglobin shortly after birth, HbF levels can remain elevated in a rare condition called hereditary persistence of HbF (HPFH) in which mutations prevent the normal repression of γ-globin.9 When HPFH co-occurs with the mutations that cause SCD, elevated levels of HbF can prevent the aggregation of HbS and protect erythrocytes from sickling, significantly ameliorating the disease.10 To date, the most important pharmacological agent for the management of SCD remains the ribonucleotide reductase inhibitor hydroxyurea (HU), which benefits patients through increasing HbF expression and reducing the incidence of vaso-occlusive crises. Although HU mitigates the clinical severity of disease for many SCD patients, there are important limitations to the clinical utility of HU. Importantly, there is typically a narrow therapeutic window between the efficacious dose of HU for beneficial HbF induction and the maximum tolerated dose typically defined by acceptable myelosuppression. As a consequence, there are variable pharmacological responses to HU in many patients.11-13 There is therefore a desire to identify alternative agents that safely and consistently induce HbF to therapeutic levels for the treatment of SCD. The hypomethylating agent (HMA) 5-azacytidine (5-aza) is a cytidine analog that was first demonstrated to induce HbF in an anemic baboon model.14 It was subsequently confirmed to increase HbF in investigational studies of patients with SCD and β-thalassemia15-18 as well as in patients with myelodysplastic syndrome and acute myeloid leukemia.19, 20 Low doses of decitabine were also confirmed to increase HbF levels in SCD patients, in some cases exceeding the maximal HbF levels observed with HU.18 Decitabine and 5aza are inhibitors of DNA methyltransferases (DNMT), enzymes that establish and maintain the epigenetic pattern of DNA methylation that functions in chromatin condensation and gene silencing. The catalytically active members of the DNMT family are DNMT3A, DNMT3B, and DNMT1. DNMT3A and DNMT3B establish the de novo pattern of DNA methylation, while DNMT1 is the primary maintenance methyltransferase that propagates the pattern of DNA methylation to daughter cells during cell division.21 In cultured human erythroid progenitor cells (EPC)22-24 and in vivo models with monkeys,25, 26 treatments with either decitabine or 5-aza decreased methylation of multiple CpG 2

1980

sites in the promoters of HBG1 and HBG2, resulting in increased γ-globin expression and elevated HbF levels. While they are effective inducers of HbF, the decitabine and 5-aza mechanism of action relies on incorporation into DNA, and they both carry drug label warnings for genotoxicity and cytotoxicity. Decitabine and 5-aza are currently approved only for use in myelodysplastic syndromes and acute myeloid leukemia, and they are not currently approved to treat β-hemoglobinopathies. In the current work, we describe the identification of a novel class of orally-dosed, reversible DNMT1-selective inhibitors, exemplified by GSK3482364. This molecule caused decreased DNA methylation in cultured human EPC, resulting in increased γ-globin gene expression and increased HbF. In a murine model of SCD, orally dosed GSK3482364 decreased DNA methylation in bone marrow, and increased HbF expression in erythrocytes. Notably, although GSK3482364 and decitabine were comparable in their maximal effects on DNA methylation in cells, GSK3482364 treatment resulted in lower cytotoxicity in cultured cells as well as improved in vivo tolerability in preclinical models. These results indicate that selective, reversible inhibition of DNMT1 is sufficient for the induction of HbF, is well-tolerated in vivo, and that neither irreversible DNMT1 inhibition nor inhibition of DNMT3A or DNMT3B is required for this effect.

Methods Erythroid progenitor cell culture Cryopreserved human bone marrow CD34+ cells (AllCells) were confirmed to be sourced ethically, and their research use was in accord with the terms of the informed consents under an Institutional Review Board/Ethics Commity approved protocol. Cells were cultured according to previously described methods23 for 7 days to generate EPC. For compound treatment studies, cell culture plates were typically incubated for 3-5 days unless otherwise indicated. In order to investigate potential drug effects on erythroid maturation to reticulocytes, CD34+ cells were cultured for 19 days in a three stage protocol that models maturation into reticulocytes with continuous compound treatment. Details for cell culture and methods for cellular assays can be found in the Online Supplemental Appendix.

Methylation-sensitive restriction endonuclease assays Genomic DNA was extracted from EPC or bone marrow using Zymo Quick-gDNA kits (Zymo Research). Total DNA was measured on a NanoDrop (ThermoFisher), diluted, and split into tubes containing reaction buffer +/- methylation-sensitive HpaII (New England Biolabs) for a 1-hour reaction. Reaction products were then quantitated in a 50 mL SYBR Green quantitative polymerase chain reaction (qPCR) (Applied Biosystems). -53 base pair: Primer 1: 5ˈ-GAACTGCTGAAGGGTGCT-3ˈ, Primer 2: 5ˈ-GACAAGGCAAACTTGACCAATAG-3ˈ.

In vivo studies All studies were conducted in accordance with the GlaxoSmithKline (GSK) Policy on the Care, Welfare and Treatment of Laboratory Animals and were reviewed by the Institutional Animal Care and Use Committee either at GSK or by the ethical review process at the institution where the work was performed. Male and female human hemoglobin transgenic mice [B6;129-HBAtm1(HBA)Tow/HBBtm2(HBG1,HBB*)Tow/J Mice] (Jackson Laboratories) were 6-8 weeks of age and grouped into haematologica | 2021; 106(7)

Fetal hemoglobin induction with a DNMT1 inhibitor

approximately sex balanced groups of five to six mice. Unless otherwise noted, mice were administered vehicle (10% DMA/90% PEG400) or GSK3482364 twice daily (b.i.d.) by oral gavage on weekdays with no doses on the intervening 2-day weekend, typically for a 12 days study. Alternatively, mice were administered vehicle (phospahate buffered saline [PBS]) or decitabine thrice weekly (Monday, Wednesday, Friday) subcutaneously (s.c.) for a total of six doses. At the end of the dosing period, blood was collected into EDTA tubes for analysis, and HbF was analyzed as detailed below. For bone marrow analysis, femurs from mice were flushed with Dulbecco’s PBS, cells were centrifuged briefly to pellet, and cell pellets were processed for DNA methylation analysis (as above) or for RNA analysis (detailed in the Online Supplemental Appendix). For bone marrow histology assessment, sternums from treated animals were formalin-fixed, paraffin embedded, sectioned and stained with hematoxylin and eosin. Complete blood counts were conducted with an Advia Hematology Analyzer (Siemens).

Fetal hemoglobin analysis by high-performance liquid chromatography or flow cytometry Percentage HbF was determined by high-performance liquid chromatography (HPLC) using the D-10 Hemoglobin Analyzer (Bio-Rad). Percentage F-cells was determined by flow cytometry with a FACSCanto I (BD BioSciences) using a mouse monoclonal anti-human HbF antibody conjugated to allophycocyanin (APC) (Life Technologies). For analysis of human cell cultures, the nuclear stain Syto16 (Life Technologies) was used to distinguish EPC (Syto16high) from enucleated reticulocytes (Syto16low). For analysis of mouse whole blood samples, Syto16 was used to distinguish mature red blood cells (Syto16negative) from reticulocytes (Syto16low). Flow cytometry data were analyzed with FlowJo v7 software (Tree Star). One-way ANOVA of data was employed to determine significance of changes relative to vehicle treated samples (Graphpad Prism v7).

activity. Briefly, approximately 1.8 million compounds were assayed in a scintillation proximity assay measuring the transfer of a radioactive methyl group by recombinant DNMT1 to a hemi-methylated 40-mer DNA substrate (Stowell, A. et al. manuscript in preparation; Pappalardi, M. et al. manuscript submitted). Screening hits were further profiled to eliminate compounds that were also inhibitors of DNMT3A or DNMT3B or that were non-specific DNA binders. From this screen and a subsequent medicinal chemistry campaign that employed non-radioactive breaklight format methyltransferase assays 27 (see the Online Supplementary Appendix), a class of potent biochemical inhibitors of DNMT1 that does not inhibit either DNMT3A or DNMT3B was identified, as exemplified by GSK348236428 (Figure 1A and B). GSK3482364 was confirmed to be a reversible inhibitor of DNMT1 using a jump dilution protocol in which a preincubated complex of DNMT1 and compound was rapidly diluted 100-fold with the addition of substrates, and recovery of DNMT1 activity was established (Pappalardi, M. et al.). Since the mechanism of action of decitabine and 5-aza requires incorporation into DNA before covalently trapping the DNMT proteins,29,30 these compounds do not show inhibition of DNMT1 in this biochemical assay. In cultured EPC, GSK3482364 treatment produced dosedependent DNA hypomethylation. Bone marrow derived CD34+ cells were expanded and differentiated over 7 days into EPC, previously characterized as expressing high CD71 and increasing CD235.23 Day 7 EPC were treated with a dose range of GSK3482364 for 5 additional days, after which DNA was harvested. Compound effects on EPC DNA methylation were measured by enzymatically digesting DNA into nucleosides and measuring the ratio of methylcytosine to total cytosine by mass spectrometry. GSK3482364 treatment reduced 5-methylcytosine levels in a dose-dependent manner with an IC of 0.24 mM (Figure 1C). An additional orthogonal assay was also developed to measure the level of methylation on specific cytosines located at positions -53 basepair (bp) relative to transcription start sites of both HBG1 and HBG2. As has been previously reported, these are among a number of highly 50

Results A high-throughput screen was conducted to identify novel biochemical inhibitors of DNMT1 methyltransferase

Figure 1. Biochemical and cellular inhibition of DNMT1 with GSK3482364. (A) Structure of DNMT1 inhibitor GSK3482364. (B) Representative data for GSK3482364 activity in biochemical methyltransferase assays with DNMT1 (black), DNMT3A (white), or DNMT3B (gray). (C) Effect of GSK3482364 on 5-methylcytosine (black), HBG1/HBG2 -53bp methylation (gray), and HBG1/HBG2 mRNA (red) in erythroid progenitor cells treated for 5 days.

haematologica | 2021; 106(7)

1981

A.G. Gilmartin et al.

methylated cytosines in the promoters of HBG1 and HBG2 that become hypomethylated in response to DNMT1 gene silencing in EPC.31,38 We have confirmed the hypomethylation of a number of the same methylcytosines in response to DNMT1 inhibition by bisulfite sequencing, including the -53 bp methylcytosine for HBG1 (Online Supplementary Figure S1). Consistent with the findings in the global methylcytosine assay, methylation of the -53 bp HBG1/ HBG2 cytosine residues was decreased for cells treated with GSK3482364 with an IC of 0.33 mM (Figure 1C). In order to determine the effect of decreased HBG1 and HBG2 promoter methylation on gene expression, mRNA from treated cells was measured by reverse transcritase qPCR (RT-qPCR) with an assay that detects both HBG1 and HBG2 mRNA (99% genetic identity). After 5 days of treatment, GSK3482364 caused a dose-dependent increase in HBG1/HBG2, raising levels 5.3-fold compared to vehicle 50

treated cells (Figure 1C). Increases in globin gene expression correlated inversely with DNA methylation levels; at the cellular IC50 of the -53 bp methylcytosine assay, 0.33 mM GSK3482364 treatment caused more than a 2-fold increase in HBG1/HBG2 mRNA. In order to further characterize the effect of GSK3482364 on cultured EPC, HbF protein expression was measured by enzyme-linked immunosorbent assay (ELISA). Day 7 EPC cultured in the presence of GSK3482364 for 5 additional days increased HbF 2-fold at 0.56 mM, and up to a maximum of 300% of vehicle-treated levels (Figure 2A). In comparison, decitabine treatment increased HbF 2-fold at 0.04 mM, and up to a maximum of 250% of vehicle. Notably, decitabine treatment reproducibly generated a bell-shaped curve response in the ELISA assay, consistent with cytotoxicity at high concentrations which was confirmed in parallel cell growth assays (Figure 2B). At high concentrations

Figure 2. Effect of GSK3482364 and decitabine on fetal hemoglobin, cell growth, and caspase activation in erythroid progenitor cells. Erythroid progenitor cells (EPC) were treated for 5 days with GSK3482364 (A) or decitabine (B) and then assayed for change in fetal hemoglobin (HbF) by enzyme-linked immunosorbent assay (green) or cell growth (blue); data represent the mean +/- standard deviation for n=20 assays. Concentrations resulting in 200% HbF and 50% growth inhibition are indicated with light blue and gray lines, respectively. (C) Caspase-Glo assay measure caspase 3/7 activation in EPC after 3 days treatment with GSK3482364 (black) or decitabine (red).

Figure 3. Effects of GSK3482364 and decitabine on fetal hemoglobin levels and reticulocyte differentiation after 18 days in cell culture. CD34+ bone marrow hematopoietic stem cells were cultured for 18 days in the presence of GSK3482364 or decitabine, with three phases of media exchange promoting erythroid differentiation into reticulocytes. Day 18 cells were stained with Syto16 (nucleic acids) and anti-fetal hemoglobin (anti-HbF) (APC) antibody. Erythroid progenitor cells are Syto16high events that retain nuclei, and reticulocytes are Syto16low events that lack nuclei.

1982

haematologica | 2021; 106(7)

Fetal hemoglobin induction with a DNMT1 inhibitor

representing 25x the concentrations that increased HbF 2fold, GSK3482364 and decitabine caused, respectively, 58% and 90% cell growth inhibition of EPC over 5 days of cell culture. In order to better characterize the relative acute cytotoxicity of GSK3482364 and decitabine, caspase 3/7 activity was measured in EPC treated with compounds for 2 days. GSK3482364 and decitabine increased cleavage of a caspase 3/7 substrate by 1.5-fold and 16.5-fold, respectively (Figure 2C) at the highest test concentration of 10 mM. Comparing at equivalent relative potencies 18-fold above the concentration that causes 2-fold HbF induction, GSK3482364 (10 mM) and decitabine (0.7 mM) increased caspase activity by 50% and 120%, respectively. In agreement with the cell growth assays, these data indicate potentially differentiated cytotoxicity between the two compounds. It has previously been observed that treatment with cytidine analog HMA including 5-aza, decitabine, and zebularine cause destabilization of DNMT1 and to a lesser extent DNMT3A and DNMT3B, through a ubiquitin-dependent mechanism.32-36 There is a debate whether the effects of the cytidine analog HMA on DNMT3A and DNMT3B protein levels are due to direct inhibition and DNA adduct formation, or are instead an indirect outcome of DNMT1 inhibition. However, evidence that zebularine-incorporated oligonucleotides can trap and biochemically inhibit DNMT1, DNMT3A, and DNMT3B37, 38 suggests that cytidine analog HMA may not be DNMT1-selective. In our study of EPC treated for 24 hours with decitabine, both DNMT1 and to a lesser extent DNMT3A levels were shown to decrease while DNMT3B was largely unaffected (Online Supplementary Figure S2). Treatment with GSK3482364 also caused a decrease in DNMT1 protein, although less than decitabine, and had no effect on either DNMT3A or DNMT3B levels. The mechanism by which GSK3482364 causes DNMT1 protein levels to decrease remains a matter of investigation. In order to characterize longer treatment effects of GSK3482364 on cellular expansion and differentiation, bone marrow CD34+ cells were expanded and differentiated in a three-phase erythroid cell differentiation protocol39 over 18 days in the presence of GSK3482364, decitabine, or vehicle. Fresh compound or vehicle were added at each stage of media exchange. On day 18, cells were stained with Syto16 to label nucleic acids and to distinguish mature enucleated reticulocytes from less mature erythroblasts. Cells were also stained with an anti-HbF antibody and analyzed by flow cytometry. On day 18, it was found that 31% of the vehicle-treated cells had matured into enucleated reticulocytes (Syto16low), distinguishable from the less mature erythroblasts (Syto16high). Consistent with HbF

ELISA results, GSK3482364 treatment (1.0 mM) for 18 days caused a >4-fold increase in HbF-positive cells compared to vehicle (Figure 3). Moreover, the resulting fraction of cells maturing into reticulocytes was comparable to or slightly higher than vehicle treatment, indicating that treated cells were not arrested earlier in erythropoiesis. Decitabine (0.1 mM) treatment caused a similar increase in HbF-positive cells as GSK3482364 (1 mM) when compared to vehicle, but also caused a marked decrease in the proportion of cells maturing into reticulocytes, reflecting delayed or arrested cellular maturation at this concentration. In order to measure the in vivo HbF induction activity of GSK3482364, the Townes mouse model of SCD40 was employed in which mouse a- and β-globin genes were replaced with human genes HBA1, HBG1, and HBB including the E6V sickle mutation. Mice in this model express HbF during fetal development, repress HbF shortly after birth, and experience cell sickling and multiple organ pathologies analogous to sickle cell disease. GSK3482364 was administered orally (p.o.) to 6-8 week old mice b.i.d. at 10 or 50 mg/kg for 12 days (weekday dosing only). At the end of dosing, HbF levels were measured in whole blood by an HPLC method, and F-cells were measured by flow cytometry. Compared with vehicle treated animals, both dose levels of GSK3482364 caused significant increases in both HbF and F-cells (Figure 4). At the 50 mg/kg dose, HbF increased 10.3-fold and F-cells increased 8.4-fold relative to vehicle treatment. In a subsequent study of murine SCD model mice treated with GSK3482364, blood was analyzed for HbF, and femoral bone marrow was harvested on day 12 to measure changes in DNA methylation and RNA expression. Treatment with GSK3482364 caused dose-dependent increases in HbF protein in whole blood by up to 9-fold versus vehicle treatment (Figure 5A). No further increase in HbF was observed between 33.3 mg/kg and 100 mg/kg doses, suggesting a plateau of activity that is consistent with a plateau of the exposure of GSK3482364 (data not shown). In bone marrow samples, globin gene expression and HBG1 promoter methylation were evaluated. In order to assess effects on DNA methylation, the -53 bp HBG1 methylationsensitive restriction endonuclease assay was employed. Bone marrow from treated animals showed a dose-dependent decrease in methylation at this site by 25% compared to vehicle-treated animals (Figure 5A). In the same samples, HBG1 mRNA levels increased in all dose groups by up to 29fold (Figure 5B). In contrast, HBB and HBA1 mRNA levels were not significantly changed in dose groups compared to baseline; this is consistent with the low baseline promoter methylation of these highly expressed genes. In a parallel study, transgenic mice were dosed subcutaneously with decitabine at doses of 0.2, 0.4, or 0.8 mg/kg

Table 1. Effects of GSK3482364 twice daily in vivo on blood cell counts and bone marrow DNA methylation.

Study 1GSK3482364 Vehicle 3.7 mg/kg 11.1 mg/kg 33.3 mg/kg 100 mg/kg

RBC (x106/µl)

Platelets (x103/µl)

Neutrophils (x103/µl)

Lymphocytes (x103/µl)

Monocytes (x103/µl)

% Change DNA Methylation

7.7 ± 1.0 7.4 ± 0.3 8.1 ± 1.0 7.0 ± 1.0 7.6 ± 1.4

819 ± 145 757 ± 150 854 ± 107 733 ± 90 678 ± 210

3.7 ± 2.1 2.9 ± 2.4 2.4 ± 0.8 3.2 ± 2.3 2.3 ± 1.2

14.9 ± 3.9 12.6 ± 5.6 10.7 ± 3.3 9.7 ± 1.8 11.0 ± 4.9

0.7 ± 0.5 0.5 ± 0.2 0.5 ± 0.4 0.4 ± 0.2 0.3 ± 0.3

0 ± 4.0 -2.2 ± 2.7 -14.7 ± 5.7** -25.4 ± 8.1** -23.9 ± 9.5**

Mean +/- standard deviation are shown. Asterisks indicate significance by 1-way ANOVA. (**P<0.001)

haematologica | 2021; 106(7)

1983

A.G. Gilmartin et al.

thrice weekly for a total of six doses and terminated after the final dose on day 12. In previous studies, lower doses of decitabine administered daily resulted in no significant HbF induction in a 2-week timeframe, while higher doses administered daily or every other day were not tolerated (Online Supplemental Table S1). In the current decitabine study, mean HbF levels were increased by up to 2-fold at the 0.4 mg/kg dose, and by up to 4.4-fold at the 0.8 mg/kg (Figure 5A). However, 0.8 mg/kg was not tolerated based on the death of two of six animals. Bone marrow samples from surviving decitabine-treated animals exhibited decreased -53 bp DNA methylation by 4%, 9%, and 13% respectively in the 0.2, 0.4, and 0.8 mg/kg dose groups (Figure 5A). As further characterization of the effects of dosing with GSK3482364 in the SCD mouse model, complete blood counts were taken at all doses. Compared to vehicle-treated animals, GSK3482364-treated SCD mice had no remark-

able changes in peripheral blood cell counts or red cell indices (Table 1). Consistent with peripheral blood cell counts, histopathology of sternal bone marrow from the high dose group of animals treated with GSK3482364 (100 mg/kg) demonstrated no overt abnormalities in marrow cellularity or hematopoietic cell composition (six of six mice; Figure 6). Notably, in the comparable 12 day study, dose levels of decitabine that caused HbF induction (≥0.2 mg/kg ) also caused dose-dependent decreases in multiple blood cell counts (Online Supplemental Table S2), indicating that these dose regimens were all cytotoxic to some extent in the SCD mice. While lower doses of decitabine (0.1-0.2 mg/kg) administered either daily or every other day were poorly effective at inducing HbF in our previous 12 day studies (Online Supplemental Table S1), we acknowledge that alternative non-cytotoxic dose regimens or longer study durations might better optimize the efficacy and tolerability of decitabine in these mice.

Figure 4. Effects of GSK3482364 and decitabine in vivo. (A) Percentage fetal hemoglobin (%HbF) by highperformance liquid chromatography (HPLC) in whole blood of transgenic sickle cell mice treated with GSK3482364 for 12 days. Bars represent mean %HbF +/- standard deviation from 4-6 mice. (B) Representative histograms from HPLC assay; HbF peaks are indicated with arrowheads. (C) %F-cells for blood samples from treated mice were determined by flow cytometry. (D) Representative flow cytometry staining of HbF and nucleic acid content for blood cells. Asterisks indicate significance by 1-way ANOVA. (*P<0.01; **P<0.001). b.i.d.: twice daily; p.o.: orally.

1984

haematologica | 2021; 106(7)

Fetal hemoglobin induction with a DNMT1 inhibitor

In order to further assess the effect of GSK3482364 on alternative dose schedules, SCD mice were dosed as previously described or on a reduced frequency schedule (b.i.d for 3 weekdays). Induction of HbF was observed to be approximately equivalent for mice dosed on both schedules (Online Supplemental Figure S3A). Since both dose regimens were equally efficacious, we next tested the effect of dosing GSK3482364 on the reduced frequency schedule for either 2 or 4 weeks to determine the effect of longer dose duration on efficacy and tolerability. Both 2- and 4week dosing regimens were well-tolerated with no effects on body weight compared to control mice, and all mice

exhibited significant gain in HbF (Online Supplemental Figure S3B). Following 4 weeks of dosing, there was a small increase in percentage HbF induction compared to animals dosed for only 2 weeks, but this increase was not found to be significant. As with previous studies, peripheral blood cell counts measured after 4 weeks indicated no remarkable differences from vehicle treated animals in any parameters. These results demonstrate that GSK3482364 can robustly induce HbF in vivo with intermittent dosing, and that it is well-tolerated for up to 4 weeks without causing evidence of myelosuppression in mice.

Figure 5. Effects of GSK3482364 and decitabine in vivo. (A) Bars represent mean percentage fetal hemoglobin (%HbF) +/- standard deviation (SD), measured by high-performance liquid chromatography (HPLC), in whole blood of sickle cell mice treated with GSK3482364 (twice daily [p.o.]) or decitabine (thrice weekly [s.c.]) for 12 days. Red circles represent mean +/- SD of DNA methylation (-53bp HBG1 methylcytosine) from bone marrow of 4-6 mice. (B) Bone marrow was further evaluated for HBG1, HBA, and HBB mRNA levels by reverse transcriptase quantitative polymerase chain reaction. In all cases levels are normalized to vehicle-treated control levels. Asterisks indicate significance by 1-way ANOVA (*P<0.01; **P<0.001).

Figure 6. Effects of GSK3482364 on bone marrow cellularity. Representative histopathology images (15x) of sternum bone marrow for mice following 12 days dosing with vehicle or GSK3482364. b.i.d.: twice daily; p.o.: orally.

haematologica | 2021; 106(7)

1985

A.G. Gilmartin et al.

Discussion As potential agents for the management of SCD, the DNMT inhibitors 5-aza and decitabine have demonstrated significant induction of HbF levels in preclinical and clinical studies.14, 18, 23, 41-45 The robust HbF induction caused by these compounds is explained both by the high proliferative index of bone marrow progenitor cells and by transitional changes in DNA methylation that they undergo during erythropoiesis. Changes in DNA methylation occur during key determining events in hematopoiesis.46 During erythropoiesis a rapid decrease in global DNA methylation marks a commitment toward erythropoietin dependence and the expression of erythroid specific master regulators GATA1 and KLF1.8, 47 As hematopoietic stem cells commit to erythropoiesis, DNMT1 becomes the dominantly expressed DNMT with the primary role of maintaining DNA methylation and regulating globin gene transcription.8, 31, 48 For EPC in adult bone marrow, key cytosine residues in the promoter region of HBG1 and HBG2 become highly methylated while the β-globin HBB gene promoter is largely unmethylated, corresponding to the increased expression of β-globin and the repression of γ-globin expression.31, 49 Although the known HMA decitabine and 5-aza have proven to be valuable probes to study the biology of DNMT methylation, and partial and complete DNMT1 knockout studies in animals have established a critical role for DNMT1 in cellular differentiation and stem cell maintenance, it has not been previously possible to study the in vitro and in vivo cellular and hematopoietic effects of reversible and selective DNMT1 inhibition. GSK3482364 represents a novel class of DNMT1-selective inhibitors that are mechanistically distinct from other HMA. In cellular studies, GSK3482364 treatment caused DNA hypomethylation and HbF induction with maximal effects that were approximately equivalent to decitabine treatment. However, at concentrations of GSK3482364 and of decitabine that produced equivalent HbF induction, decitabine was observed to consistently cause more cell growth inhibition. In an in vitro model of erythropoiesis, expanding and differentiating from CD34+ cells to enucleated reticulocytes, GSK3482364 and decitabine caused equivalent increases in HbF-cells, but GSK3482364 treatment resulted in a larger proportion of cells maturing into HbF expressing reticulocytes. In a transgenic mouse model of SCD, the effects of orallydosed GSK3482364 on bone marrow DNA methylation and erythrocyte HbF elevation exceeded the corresponding effects of decitabine at tolerated doses over a 12-day period. Examination of complete blood counts and bone marrow cellularity from in vivo studies with GSK3482364 suggests that the effects of this compound on DNA methylation in the bone marrow were well-tolerated without evidence of other adverse hematological effects. Chronic intermittent pharmacologic inhibition of

References 1. Weatherall D. Beginnings: The Molecular Pathology of Hemoglobin. Molecular Hematology: Third Edition, 2010:1-18. 2. Kato GJ, Piel FB, Reid CD, et al. Sickle cell

1986

DNMT1 does not appear to fully phenocopy the effects of permanent DNMT1 deletion. In our mouse studies, the limited impact on bone marrow cellularity or blood cell populations appears to indicate that repeat daily doses of GSK3482364 were tolerated by hematopoietic stem and progenitor cells. Notably, in our in vivo studies we did not observe the significant increase in platelets that has been reported in clinical studies with low dose decitabine and that is attributed to effects of hypomethylation in promoting megakaryocyte maturation.18, 45, 50 Since the increase in platelets does not appear to be captured in our mouse model, we cannot currently draw any conclusions about potential differentiating effects on platelets for GSK3482364 or related compounds. Future studies in non-human primate models, where multiple HbF inducers were initially characterized, are warranted to address this and other questions about the optimal dosing regimens for this class of DNMT1 selective inhibitors. The differential cellular biology and in vivo pharmacology observed with GSK3482364 as compared to decitabine suggest that this may be a useful tool molecule to study the selective, reversible inhibition of DNMT1 in hematopoiesis and for the elevation of HbF in erythrocytes. Disclosures AGG, AG, EG, CA, EL, MM, ZW, WH, DM, DE, LR, MP, MTM, RGK, JL, and ABB are current or former employees of GlaxoSmithKline; AR, RB, AS, MC, DO, IW, and AJ are current or former employees of CRUK Manchester Institute; patent US20190194166 pertains to compounds discussed in this manuscript. Contributions AGG, AG, EG, CA, EL, MNM, ZW, WH, DM, DE, AS, MP, and ABB designed or performed experiments; AR, RB, MC, AJ, LR, DO,IW, MTM, RGK, and JL were involved in the identification and synthesis of the lead compound; AGG wrote the manuscript and all authors read and approved the final version of the manuscript. Acknowledgments The authors wish to thank Karen Evans, Rakesh Nagilla, Elizabeth Rivera, Karen Lynch, Denise Depagnier, Chris Traini, Leonard Azzarano, and Shanker K. Sundaram for contributions to this work. Additionally, we wish to thank Jerry Adams, Dirk Heerding, Chris Carpenter, Steve Pessagno, Philip Chapman, Charlotte Burt, Martyn Bottomley, Kristin Goldberg, and Chris Kershaw for their contributions to the identification of GSK3482364. Work conducted at the Cancer Research UK Manchester Institute was wholly funded by Cancer Research UK (Grant numbers C480/A1141 and C5759/A17098). Funding Work conducted at the Cancer Research UK Manchester Institute was wholly funded by Cancer Research UK (Grant numbers C480/A1141 and C5759/A17098).

disease. Nat Rev Dis Primers. 2018;4:18010. 3. Sankaran VG, Xu J, Orkin SH. Advances in the understanding of haemoglobin switching: Review. Br J Haematol. 2010;149(2): 181-194. 4. Sankaran VG, Orkin SH. The switch from

fetal to adult hemoglobin. Cold Spring Harb Perspect Med. 2013;3(1):a011643. 5. Suzuki M, Yamamoto M, Engel JD. Fetal globin gene repressors as drug targets for molecular therapies to treat the β-globinopathies. Mol Cell Biol. 2014;34(19):

haematologica | 2021; 106(7)

Fetal hemoglobin induction with a DNMT1 inhibitor 3560-3569. 6. Masuda T, Wang X, Maeda M, et al. Transcription factors LRF and BCL11A independently repress expression of fetal hemoglobin. Science. 2016;351(6270):285-289. 7. Mabaera R, West RJ, Conine SJ, et al. A cell stress signaling model of fetal hemoglobin induction: what doesn't kill red blood cells may make them stronger. Exp Hematol. 2008;36(9):1057-1072. 8. Shearstone JR, Pop R, Bock C, et al. Global DNA demethylation during mouse erythropoiesis in vivo. Science. 2011;334(6057):799802. 9. Higgs DR, Wood WG. Genetic complexity in sickle cell disease. Proc Natl Acad Sci U S A. 2008;105(33):11595-11596. 10. Perrine RP, Brown MJ, Clegg JB, Weatherall DJ, May A. Benign sickle-cell anaemia. Lancet. 1972;2(7788):1163-1167. 11. Lanzkron S, Strouse JJ, Wilson R, et al. Systematic review: hydroxyurea for the treatment of adults with sickle cell disease. Ann Intern Med. 2008;148(12):939-955. 12. Steinberg MH, Barton F, Castro O, et al. Effect of hydroxyurea on mortality and morbidity in adult sickle cell anemia: Risks and benefits up to 9 years of treatment. JAMA. 2003;289(13):1645-1651. 13. Maier-Redelsperger M, de Montalembert M, Flahault A, et al. Fetal hemoglobin and F-cell responses to long-term hydroxyurea treatment in young sickle cell patients. Blood. 1998;91(12):4472-4479. 14. DeSimone J, Heller P, Hall L, Zwiers D. 5Azacytidine stimulates fetal hemoglobin synthesis in anemic baboons. Proc Natl Acad Sci U S A. 1982;79(14):4428-4431. 15. Dover GJ, Charache SH, Boyer SH, Talbot J, Smith KD. 5-Azacytidine increases fetal hemoglobin production in a patient with sickle cell disease. Prog Clin Biol Res. 1983;134:475-488. 16. Ley TJ, Anagnou NP, Noguchi CT, et al. DNA methylation and globin gene expression in patients treated with 5-azacytidine. Prog Clin Biol Res. 1983;134:457-474. 17. Lowrey CH, Nienhuis AW. Brief report: treatment with azacitidine of patients with end-stage +¦- thalassemia. N Engl J Med. 1993;329(12):845-848. 18. Saunthararajah Y, Hillery CA, Lavelle D, et al. Effects of 5-aza-2′-deoxycytidine on fetal hemoglobin levels, red cell adhesion, and hematopoietic differentiation in patients with sickle cell disease. Blood. 2003;102(12): 3865-3870. 19. Reinhardt D, Haase D, Schoch C, et al. Hemoglobin F in myelodysplastic syndrome. Ann Hematol. 1998;76(3-4):135-138. 20. Lubbert M, Ihorst G, Sander PN, et al. Elevated fetal haemoglobin is a predictor of better outcome in MDS/AML patients receiving 5-aza-2'-deoxycytidine (Decitabine). Br J Haematol. 2017;176(4): 609-617. 21. Lyko F. The DNA methyltransferase family: a versatile toolkit for epigenetic regulation. Nat Rev Genet. 2018;19(2):81-92. 22. Bradner JE, Mak R, Tanguturi SK, et al. Chemical genetic strategy identifies histone deacetylase 1 (HDAC1) and HDAC2 as ther-

haematologica | 2021; 106(7)

apeutic targets in sickle cell disease. Proc Natl Acad Sci U S A. 2010;107(28):1261712622. 23. Li H, Xie W, Gore ER, et al. Development of phenotypic screening assays for gamma-globin induction using primary human bone marrow day 7 erythroid progenitor cells. J Biomol Screen. 2013;18(10):1212-1222. 24. Torrealba-de Ron AT, Papayannopoulou T, Knapp MS, et al. Perturbations in the erythroid marrow progenitor cell pools may play a role in the augmentation of HbF by 5azacytidine. Blood. 1984;63(1):201-210. 25. Akpan I, Banzon V, Ibanez V, et al. Decitabine increases fetal hemoglobin in Papio anubis by increasing gamma-globin gene transcription. Exp Hematol. 2010; 38(11):989-993.e1. 26. Chin J, Singh M, Banzon V, et al. Transcriptional activation of the gammaglobin gene in baboons treated with decitabine and in cultured erythroid progenitor cells involves different mechanisms. Exp Hematol. 2009;37(10):1131-1142. 27. Wood RJ, McKelvie JC, Maynard-Smith MD, Roach PL. A real-time assay for CpGspecific cytosine-C5 methyltransferase activity. Nucleic Acids Res. 2010;38(9):e107. 28. Adams ND, Benowitz AB, Rueda Benede ML, et al. inventors. Substituted Pyridines as Inhibitors of DNMT1. Patent WO/2017/ 216727. 2017. 29. Santi DV, Norment A, Garrett CE. Covalent bond formation between a DNA-cytosine methyltransferase and DNA containing 5azacytosine. Proc Natl Acad Sci U S A. 1984; 81(22):6993-6997. 30. Oka M, Meacham AM, Hamazaki T, et al. De novo DNA methyltransferases Dnmt3a and Dnmt3b primarily mediate the cytotoxic effect of 5-aza-2′-deoxycytidine. Oncogene. 2005;24(19):3091-3099. 31. Mabaera R, Richardson CA, Johnson K, et al. Developmental- and differentiation-specific patterns of human γ- and β-globin promoter DNA methylation. Blood. 2007;110(4):13431352. 32. Ghoshal K, Datta J, Majumder S, et al. 5Aza-Deoxycytidine induces selective degradation of DNA methyltransferase 1 by a proteasomal pathway that requires the KEN box, bromo-adjacent homology domain, and nuclear localization signal. Mol Cell Biol. 2005;25(11):4727-4741. 33. Robaina MC, Mazzoccoli L, Arruda VO, et al. Deregulation of DNMT1, DNMT3B and miR-29s in Burkitt lymphoma suggests novel contribution for disease pathogenesis. Exp Mol Pathol. 2015;98(2):200-207. 34. Palii SS, Van Emburgh BO, Sankpal UT, Brown KD, Robertson KD. DNA methylation inhibitor 5-aza-2′-deoxycytidine induces reversible genome-wide DNA damage that is distinctly influenced by DNA methyltransferases 1 and 3B. Mol Cell Biol. 2008;28(2):752-771. 35. Cheng JC, Yoo CB, Weisenberger DJ, et al. Preferential response of cancer cells to zebularine. Cancer Cell. 2004;6(2):151-158. 36. Patel K, Dickson J, Din S, et al. Targeting of 5-aza-2′-deoxycytidine residues by chromatin-associated DNMT1 induces proteaso-

mal degradation of the free enzyme. Nucleic Acids Res. 2010;38(13):4313-4324. 37. Zhang Z-M, Lu R, Wang P, et al. Structural basis for DNMT3A-mediated de novo DNA methylation. Nature. 2018;554(7692):387391. 38. Sledziewski A, Devos T, Kole R (inventors). Oligonucleotide inhibitors of DNA methyltransferases and their use in treating diseases. Patent WO2014011573. 2015. 39. Hu J, Liu J, Xue F, et al. Isolation and functional characterization of human erythroblasts at distinct stages: implications for understanding of normal and disordered erythropoiesis in vivo. Blood. 2013;121(16): 3246-3253. 40. Wu LC, Sun CW, Ryan TM, et al. Correction of sickle cell disease by homologous recombination in embryonic stem cells. Blood. 2006;108(4):1183-1188. 41. Charache S, Dover G, Smith K, et al. Treatment of sickle cell anemia with 5-azacytidine results in increased fetal hemoglobin production and is associated with nonrandom hypomethylation of DNA around the gamma-delta-beta-globin gene complex. Proc Natl Acad Sci U S A. 1983;80(15):48424846. 42. Koshy M, Dorn L, Bressler L, et al. 2-Deoxy 5-azacytidine and fetal hemoglobin induction in sickle cell anemia. Blood. 2000;96(7):2379-2384. 43. Nienhuis AW, Ley TJ, Humphries RK, Young NS, Dover G. Pharmacological manipulation of fetal hemoglobin synthesis in patients with severe beta-thalassemia. Ann N Y Acad Sci. 1985;445:198-211. 44. Letvin NL, Linch DC, Beardsley GP, et al. Influence of cell cycle phase-specific agents on simian fetal hemoglobin synthesis. J Clin Invest. 1985;75(6):1999-2005. 45. Molokie R, Lavelle D, Gowhari M, et al. Oral tetrahydrouridine and decitabine for non-cytotoxic epigenetic gene regulation in sickle cell disease: a randomized phase 1 study. PLoS Med. 2017;14(9):e1002382. 46. Van Der Ploeg LHT, Flavell RA. DNA methylation in the human gamma delta beta-globin locus in erythroid and nonerythroid tissues. Cell. 1980;19(4):947-958. 47. Pop R, Shearstone JR, Shen Q, et al. A key commitment step in erythropoiesis is synchronized with the cell cycle clock through mutual inhibition between PU.1 and Sphase progression. PLoS Biol. 2010;8(9): e1000484. 48. Gautier E-F, Ducamp S, Leduc M, et al. Comprehensive proteomic analysis of human erythropoiesis. Cell Rep. 2016;16(5): 1470-1484. 49. Mavilio F, Giampaolo A, Care A, et al. Molecular mechanisms of human hemoglobin switching: selective undermethylation and expression of globin genes in embryonic, fetal, and adult erythroblasts. Proc Natl Acad Sci U S A. 1983;80(22):6907-6911. 50. van den Bosch J, Lubbert M, Verhoef G, Wijermans PW. The effects of 5-aza-2'deoxycytidine (Decitabine) on the platelet count in patients with intermediate and high-risk myelodysplastic syndromes. Leuk Res. 2004;28(8):785-790.

1987

LETTERS TO THE EDITOR Mixed myeloid chimerism and relapse of myelofibrosis after allogeneic stem cell transplantation Allogeneic stem cell transplantation (allo-SCT) remains the only potentially curative treatment for myelofibrosis (MF).1,2 The 5-year overall survival (OS) rates in MF patients after allo-SCT have improved over the past several years, ranging from 47% to 62%.3-7 However, relapse after transplantation remains a frequent cause of death, with relapse rates ranging from 10-43% after alloSCT.1,8 Strategies for detecting early relapse and improving outcomes in these high-risk patients represent an unmet need. Assessing response and confirming early relapse after transplant for MF is often challenging based on the clinical criteria,9 and it usually takes a few months for fibrosis to resolve and the bone marrow morphologic remission to be achieved.10 Detecting the JAK2 V617F mutation after SCT in MF is a strong predictor of relapse and a potential marker for guiding adoptive immunotherapy.7,1113 Few studies have focused on the potential role of mixed chimerism in predicting early relapse, particularly in those with mixed myeloid chimerism (MMC).14-18 The influence of MMC on posttransplantation relapse of MF has not been well studied. Therefore, we aimed to determine the role of MMC in predicting relapse in MF patients after allo-SCT. We further explored the correlation between myeloid chimerism and molecular relapse of MF. Eighty-two consecutive patients with primary or secondary MF who underwent their first allo-SCT at The University of Texas MD Anderson Cancer Center from January 2005 to July 2015 were identified. Patients with double cord or haploidentical donors or with primary graft failure were excluded. MF relapse was defined as any evidence of persistent or recurrent morphologic disease. Molecular relapse was defined as any patient with persistent and/or reappearance of pretransplant molecular genetic abnormalities (JAK2 V617F, CALR and MPL). In this cohort, all patients with molecular relapse had evidence of morphologic persistent and/or relapsed bone marrow disease, except for one patient. Patients were determined to have MMC if they had less than 95% of donor myeloid cells at any time after day 30 after transplantation. This study was approved by the institutional review board. Chimerism testing was performed using eight highly polymorphic microsatellite markers. It included lineagespecific analysis via separation of myeloid cell and T-lymphocyte populations as described previously.19 For the majority of patients, peripheral blood chimerism testing and molecular testing for MF were performed routinely per institutional policies at months 1, 3, 6, 9, 12, 18, 24, and 36 after transplant, and more frequently as indicated at the discretion of the treating physician. The primary end point was the frequency of MMC among patients with evidence of morphologic and/or molecular relapse. Secondary end points were the progression-free survival (PFS) and OS rates. Treatment responses were defined as described previously.20 Additionally, molecular remission was defined as JAK2 V617F or CALR/MPL negativity in patients previously positive for them. Survival estimates were calculated for all patients and according to myeloid chimerism status using the Kaplan-Meier method Forty-four of the 82 patients were male, and the median age at allo-SCT was 57.5 years. Fourty-seven patients (57%) were positive for a molecular marker before trans1988

Table 1. Patient, disease, and transplant characteristics of study cohort. Number of patients 82 Sex Male 44 Female 38 Median age in years at transplant (range) 57.5 (27-74) Diagnosis Primary MF 53 Secondary MF 25 Other MPN 4 Molecular mutation status Jak2 V617F positive 41 CALR positive 5 MPL positive 1 Not available 5 Donor type Sibling 38 Unrelated 44 Source of stem cells Marrow 10 Peripheral blood 72 Conditioning regimen Myeloablative 66 Reduced intensity 16 Mixed myeloid chimerism Yes 35 No 47 MF: myelofibrosis; MPN: myeloproliferative neoplasm.

plantation. Thirty-five patients (43%) developed MMC, of which 24 patients received a myeloablative conditioning regimen. Twenty-nine of these 35 patients had initial full donor myeloid chimerisn after transplant before they developed MMC. Table 1 summarizes the patient, disease, and transplant characteristics of study population. During the study period, a total of 34 patients (41%) had persistent or relapsed disease after SCT. The flow charts in Figure 1 and the Online Supplementary Figure S1 show the study patients and their disease outcomes according to the MMC status and by molecular status. Only one patient with full donor myeloid chimerism (n=47) experienced disease progression during the study period (Figure 1). In contrast, all but two patients with MMC had morphologic and/or molecular relapse either at the time when MMC was detected or soon afterwards. When we analyzed the study patients with relapsed disease (n=34), all but one patient had MMC. Among the 47 patients with pretransplant positive molecular marker, 21 patients developed MMC of whom 95% (n=20) had concomitant molecular relapses. The exception was a patient with complete conversion to full chimerism after immunosuppression reduction. Similarly, for the 30 patients with negative molecular testing before transplantation, 13 of 14 patients (93%) with MMC eventually experienced morphologic relapse. The exception was a patient with the successful conversion to full donor chimerism after immunosuppression reduction. The Online Supplementary Table S1 and Online Supplementary Figure S2 summarize the patient characteristics, disease outcomes, and interventions done of the 35 patients with MMC. The most common cause of death for these patients was persistent/recurrent disease. Thirteen of the 18 deaths were attributed to progressive disease, of which seven had transformed acute myeloid haematologica | 2021; 106(7)

Letters to the Editor

Figure 1. Flow chart of myelofibrosis patient cohorts and outcomes according to myeloid chimerism and molecular disease status. *These two patients underwent immunosuppression reduction before any evidence of relapse. One was converted to full donor chimerism, and the other had progressive loss of donor myeloid chimerism and graft failure. **This patient with mixed T-cell chimerism had a molecular relapse (recurrent JAK2 mutation) that responded to a tacrolimus dose reduction (converted to full donor chimerism and had molecular remission).

leukemia and one patient had accelerated MF with 19% blasts. One patient died from complications of graft-versus-host disease (GvHD) (after immunosuppression reduction) and persistent disease. The other four patients died while in complete remission (CR); two patients achieved CR with immunosuppression reduction alone (died from GvHD complications), and one patient each achieved CR after donor leukocyte infusion and second allogeneic SCT (both died of second malignancy). For the 17 surviving patients, nine responded to immunosuppression reduction alone and converted to full donor myeloid chimerism and complete remission, seven patients were salvaged with second allogeneic SCT, and one patient was not evaluable at time of last follow-up. The majority of the patients underwent molecular testing for both chimera and clonal molecular markers on the same date. MMC and molecular relapse were concurrently detected in all but five patients. Among these five patients, molecular relapse was preceded by MMC in three patients, and one patient had an initially positive JAK2 V617F molecular relapse followed shortly by MMC. The fifth patient did not undergo concomitant clonal molecular testing at the initial presentation when found to have MMC but had a molecular relapse 6 months later. We assessed survival in the whole patient group and according to MMC status. With a median follow-up of 49 months (range: 3-105), the 4-year PFS and OS rates in all study patients were 32% and 51%, respectively. When stratified according to chimerism status, the 4-year PFS rate was 4% in those with MMC versus 60% in those with full donor myeloid chimerism (P<0.0001) (Online Supplementary Figure S3). Similarly, patients with MMC had a 4-year OS of 47% compared to 59% in those with full chimerism. However, this difference was not statistically significant, likely because of the small sample size and salvage treatment interventions. Highly sensitive molecular testing is increasingly being used for early detection of relapse and guidance of therapies. Approximately 60% of patients with MF harbor the JAK2 mutation.21 Other molecular markers (MPL and CALR) are increasingly being used, but not yet validated as strong predictors of relapse after allo-SCT. Hence, the need for a better universal marker to predict relapse because morphologic findings are not helpful in many haematologica | 2021; 106(7)

cases. Loss of donor chimerism has long been correlated with increased relapse incidence after allo-SCT in various hematologic neoplasms. Furthermore, Thiede et al.17 proved that MMC not only predicts clinical relapse of chronic myeloid leukemia but is also associated with reappearance of BCR-ABL1 translocation transcripts. The present study is one of the largest to demonstrate a strong association between MMC and morphologic and molecular relapse in patients with MF. The finding in a few patients that loss of myeloid chimerism may precede early molecular relapse is worth further investigation. Patients who never developed MMC rarely relapse. We propose that myeloid chimerism testing alone or in combination with other clonal molecular markers can be one of the earliest and most accurate methods of predicting relapse, particularly early after transplant where clinical and morphologic bone marrow findings are frequently not useful in confirming relapse. Our findings suggest the unmet need for a revised definition of MF relapse after allogeneic SCT to account for the role of MMF and molecular data in the posttransplant setting. Early intervention with immunosuppression reduction alone in this high-risk population with MMC is feasible and worth further investigation. Samer A. Srour,1 Amanda Olsonr,1 Stefan O. Ciurear,1 Parth Desair,2 Qaiser Bashirr,1 Betul Oranr,1 Prithviraj Boser,3 Rohtesh Mehtar,1 Keyur P. Patelr,4 Naveen Pemmarajur,3 Naval Daverr,3 Srdan Verstovsekr,3 Richard E. Champlinr1 and Uday R. Popatr1 1 Department of Stem Cell Transplantation and Cellular Therapy, The University of Texas MD Anderson Cancer Center, Houston; 2 Department of Medicine, The University of Texas Health Science Center at San Antonio, San Antonio; 3Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston and 4 Department of Hematopathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA Correspondence: UDAY R. POPAT upopat@mdanderson.org doi:10.3324/haematol.2019.223503 Received: April 1, 2019. Accepted: July 10, 2019. Pre-published: July 11, 2019. Disclosures: no conflicts of interest to disclose. 1989

Letters to the Editor

Contributions: SAS and URP conceived and designed the research; SAS performed statistical analysis; SAS, PD and URP analyzed and interpreted data; SAS and URP wrote the manuscript; and SAS, AO, SOC, PD, QB, BO, PB, RM, KPP, NP, ND, SV, REC and URP critically reviewed and edited the manuscript for important intellectual content.

References 1. Kroger NM, Deeg JH, Olavarria E, et al. Indication and management of allogeneic stem cell transplantation in primary myelofibrosis: a consensus process by an EBMT/ELN international working group. Leukemia. 2015;29(11):2126-2133. 2. Reilly JT, McMullin MF, Beer PA, et al. Guideline for the diagnosis and management of myelofibrosis. Br J Haematol. 2012;158(4):453471. 3. Gupta V, Malone AK, Hari PN, et al. Reduced-intensity hematopoietic cell transplantation for patients with primary myelofibrosis: a cohort analysis from the center for international blood and marrow transplant research. Biol Blood Marrow Transplant. 2014;20(1):89-97. 4. Ditschkowski M, Elmaagacli AH, Trenschel R, et al. Dynamic International Prognostic Scoring System scores, pre-transplant therapy and chronic graft-versus-host disease determine outcome after allogeneic hematopoietic stem cell transplantation for myelofibrosis. Haematologica. 2012;97(10):1574-1581. 5. Scott BL, Gooley TA, Sorror ML, et al. The Dynamic International Prognostic Scoring System for myelofibrosis predicts outcomes after hematopoietic cell transplantation. Blood. 2012;119(11):2657-2664. 6. Alchalby H, Yunus DR, Zabelina T, et al. Risk models predicting survival after reduced-intensity transplantation for myelofibrosis. Br J Haematol. 2012;157(1):75-85. 7. Alchalby H, Badbaran A, Zabelina T, et al. Impact of JAK2V617F mutation status, allele burden, and clearance after allogeneic stem cell transplantation for myelofibrosis. Blood. 2010;116(18):35723581. 8. van den Brink MR, Porter DL, Giralt S, et al. Relapse after allogeneic hematopoietic cell therapy. Biol Blood Marrow Transplant. 2010;16(Suppl 1):S138-145. 9. Tefferi A, Cervantes F, Mesa R, et al. Revised response criteria for myelofibrosis: International Working Group-Myeloproliferative Neoplasms Research and Treatment (IWG-MRT) and European LeukemiaNet (ELN) consensus report. Blood. 2013;122(8):1395-1398. 10. Kroger N, Kvasnicka M, Thiele J. Replacement of hematopoietic system by allogeneic stem cell transplantation in myelofibrosis patients induces rapid regression of bone marrow fibrosis. Fibrogenesis

1990

Tissue Repair. 2012;5(Suppl 1):S25. 11. Kroger N, Badbaran A, Holler E, et al. Monitoring of the JAK2-V617F mutation by highly sensitive quantitative real-time PCR after allogeneic stem cell transplantation in patients with myelofibrosis. Blood. 2007;109(3):1316-1321. 12. Steckel NK, Koldehoff M, Ditschkowski M, Beelen DW, Elmaagacli AH. Use of the activating gene mutation of the tyrosine kinase (VAL617Phe) JAK2 as a minimal residual disease marker in patients with myelofibrosis and myeloid metaplasia after allogeneic stem cell transplantation. Transplantation. 2007;83(11):1518-1520. 13. Klyuchnikov E, Holler E, Bornhauser M, et al. Donor lymphocyte infusions and second transplantation as salvage treatment for relapsed myelofibrosis after reduced-intensity allografting. Br J Haematol. 2012;159(2):172-181. 14. Bader P, Beck J, Frey A, et al. Serial and quantitative analysis of mixed hematopoietic chimerism by PCR in patients with acute leukemias allows the prediction of relapse after allogeneic BMT. Bone Marrow Transplant. 1998;21(5):487-495. 15. Thiede C, Bornhauser M, Ehninger G. Strategies and clinical implications of chimerism diagnostics after allogeneic hematopoietic stem cell transplantation. Acta Haematol. 2004;112(1-2):16-23. 16. Mackinnon S, Papadopoulos EB, Carabasi MH, et al. Adoptive immunotherapy evaluating escalating doses of donor leukocytes for relapse of chronic myeloid leukemia after bone marrow transplantation: separation of graft-versus-leukemia responses from graft-versus-host disease. Blood. 1995;86(4):1261-1268. 17. Thiede C, Lutterbeck K, Oelschlagel U, et al. Detection of relapse by sequential monitoring of chimerism in circulating CD34+ cells. Ann Hematol. 2002;81(Suppl 2):S27-28. 18. Bornhauser M, Oelschlaegel U, Platzbecker U, et al. Monitoring of donor chimerism in sorted CD34+ peripheral blood cells allows the sensitive detection of imminent relapse after allogeneic stem cell transplantation. Haematologica. 2009;94(11):1613-1617. 19. Lee HC, Saliba RM, Rondon G, et al. Mixed T lymphocyte chimerism after allogeneic hematopoietic transplantation is predictive for relapse of acute myeloid leukemia and myelodysplastic syndromes. Biol Blood Marrow Transplant. 2015;21(11):1948-1954. 20. Tefferi A, Barosi G, Mesa RA, et al. International Working Group (IWG) consensus criteria for treatment response in myelofibrosis with myeloid metaplasia, for the IWG for Myelofibrosis Research and Treatment (IWG-MRT). Blood. 2006;108(5):1497-1503. 21. Tefferi A, Lasho TL, Finke CM, et al. CALR vs JAK2 vs MPL-mutated or triple-negative myelofibrosis: clinical, cytogenetic and molecular comparisons. Leukemia. 2014;28(7):1472-1477.

haematologica | 2021; 106(7)

Letters to the Editor

Symptom burden in transplant-ineligible patients with newly diagnosed multiple myeloma: a population-based cohort study

Table 1. Characteristics of multiple myeloma patients who reported at least one Edmonton Symptom Assessment System score in the first 12 months following diagnosis.

Multiple myeloma (MM), a neoplasm characterized by the clonal proliferation of malignant plasma cells, is associated with major morbidity and mortality. The median age at diagnosis is 70 years, making MM a disease of older adults. While much progress has been made in the therapeutics for newly-diagnosed multiple myeloma (NDMM) patients, including transplant-ineligible patients, MM remains an incurable malignancy. Both the disease of MM itself, as well as the treatments initiated, likely impact patients’ quality of life and their burden of symptoms. To date, there has been no large population study conducted in adults with NDMM, specifically transplant-ineligible patients, examining the effect of symptom burden over time and associated factors. In 2007, as part of an initiative to improve symptom management, routine prospective collection of a patientreported outcome, the Edmonton Symptom Assessment System (ESAS) score, during all outpatient cancer clinic visits was started in Ontario, Canada. The ESAS is a validated and reliable patient-reported outcome tool that is used to assess common cancer-associated symptoms.1 It consists of nine symptoms, namely tiredness, impaired well-being, pain, drowsiness, loss of appetite, anxiety, shortness of breath, depression and nausea, which are scored by the patients on a numerical rating scale from 0 (no symptoms) to 10 (worst possible symptoms). We conducted a longitudinal study of ESAS data in order to examine symptom trajectory and determine factors associated with moderate to severe symptoms in the first year following diagnosis among transplant-ineligible adults with NDMM receiving treatment between 2007-2018. Multiple administrative healthcare databases in the universal, single-payer, publicly funded system in Ontario, Canada were linked using a unique encrypted patient identifier and analyzed at ICES (formerly known as the Institute for Clinical Evaluative Sciences). ICES is an independent, non-profit research institute whose legal status allows it to collect and analyze healthcare and demographic data without consent for health system evaluation and improvement. The study was approved by the ethics committee of McMaster University and followed data confidentiality and privacy guidelines of ICES. All adults (age ≥18 years) with a new diagnosis of MM (International Classification of Diseases for Oncology, 3rd Edition, histology code 9732) between January 2007 and December 2018 were identified. Patients’ demographics (age, sex) were extracted. Baseline co-morbidities were recorded using the modified Charlson-Deyo Comorbidity Index (CCI) within 1 year prior to the date of diagnosis.2 Socioeconomic status was determined using a validated modified Ontario Marginalization Index.3 Myelomarelated end-organ damage was defined as previously described by Fiala et al.4 Treatment center (non-teaching vs. teaching) was defined as the center in which the patient first received MM treatment. Treatment receipt was defined as not having received a transplant but having received therapy with one or more of the following agents: oral cyclophosphamide/melphalan (often used in combination with steroids) or novel agents (thalidomide, lenalidomide or bortezomib often used in combination as thalidomide/melphalan/prednisone, lenalidomide/dexamethasone, bortezomib/melphalan/prednisone or cyclophosphamide/bortezomib/dexamethasone) within

Age, years; median (IQR) Male; n (%) Geographic region; n (%) Urban Rural Socioeconomic status, poor; n (%) Charlson Co-morbidity Index; n (%) ≤1 ≥2 Year of diagnosis; n (%) 2007-2012 2013-2018 Myeloma end-organ damage*; n (%) Anemia Hypercalcemia Bone disease Renal failure Time from diagnosis to treatment, days; median (IQR) Time from diagnosis to first recorded ESAS score; median (IQR) Hospital type; n (%) Teaching Non-teaching Novel agents; n (%) Proteasome inhibitor Immunomodulatory agent Patients alive at 1 year; n (%)

haematologica | 2021; 106(7)

Characteristics

(N=2876) 74 (70-80) 1,649 (57.3) 2,473 (86.0) 403 (14.0) 464 (16.1) 2,478 (86.2) 398 (13.8) 953 (33.1) 1,923 (66.9) 911 (31.7) 130 (4.5) 304 (10.6) 752 (26.2) 39 (20-87) 35 (14-79)

676 (23.5) 2,200 (76.5) 2,146 (74.6) 688 (23.9) 2,349 (81.7)

*Patients could have more than one ‘CRAB’ feature. IQR: interquartile range; ESAS: Edmonton Symptom Assessment System.

1 year of diagnosis; the above anti-myeloma drugs encompass all funded cancer drugs available to patients during the study period. Symptoms were assessed using the ESAS score which consists of the nine symptoms described previously. We considered ESAS scores of ≥4 (moderate or severe) as being clinically relevant because they have been previously shown to identify a clinically significant burden.5 For the final cohort creation, only transplant-ineligible patients who received MM treatment as defined above and reported at least one ESAS assessment in the 12 months following the date of diagnosis were included. All analyses were conducted using the Statistical Analysis System (SAS version 9.4). A total of 4,610 transplant-ineligible patients who received treatment for NDMM were identified, of whom 2,876 (62.3%) completed at least one ESAS assessment following diagnosis and were included in the final cohort. In this final cohort, a total of 27,701 unique ESAS assessments were captured. Patients recorded a median of 15 ESAS assessments in the first year (interquartile range [IQR] 9-29). The baseline characteristics of the cohort are detailed in Table 1. The trajectory of moderate to severe symptoms in each month following diagnosis is shown in Figure 1. For most symptoms, a high proportion of the cohort reported moderate to severe symptoms at diagnosis, with tiredness (64%) and impaired well-being (60%) being among 1991

Letters to the Editor

Figure 1. Trajectory of symptom burden in patients with newly diagnosed multiple myeloma. Proportion of the cohort reporting moderate-to-severe Edmonton Symptom Assessment System (ESAS) scores ≥4 for (A) tiredness, well-being, pain, drowsiness and loss of appetite and (B) anxiety, shortness of breath, depression and nausea, by month following diagnosis.

the most prevalent and nausea being the least prevalent (13%). Most symptoms decreased over the first year, with the largest decline happening in the first 3 months. One year following diagnosis, there continued to be a substantial burden of symptoms, with over 25% of the cohort reporting moderate-severe levels of each of the following symptoms: tiredness, impaired well-being, pain, drowsiness and loss of appetite. Additionally, whereas physical symptoms such as pain improved over time, psychosocial symptoms of anxiety/depression showed minimal improvement with generally flat scores. The odds of reporting moderate to severe symptoms during the first year are listed in Table 2 with a higher odds ratio for a co-variate being ‘worse’ for each specified symptom. Increasing age was associated with a slightly lower burden of pain, depression and nausea with an odds ratio that was borderline at 0.98. Female sex was associated with a 1.19 to 1.59 higher odds of reporting moderate to severe symptoms for all categories except shortness of breath. An urban geographic location was

1992

associated with higher anxiety and nausea. Socioeconomic status did not have an impact on most symptoms except for tiredness. Increased co-morbidity status did not show any clear correlation with increased symptoms. A more recent year of diagnosis was associated with slightly lower odds of pain, loss of appetite and nausea. Receiving treatment at a non-teaching hospital was associated with higher odds of reporting pain and depression. Myeloma-related end-organ damage, specifically bone disease, was associated with higher symptom burden; the effect size ranging from 1.52 to 2.65 times for impaired well-being and pain respectively. In summary, our study demonstrates that transplantineligible patients with NDMM experience a substantial burden of symptoms following diagnosis. From a patient’s perspective, knowledge of these symptoms and how they change over time may enhance communication regarding expected trajectory and informed shared decision-making. From the oncology team’s perspective, understanding the considerable burden of symptoms haematologica | 2021; 106(7)

Letters to the Editor

Table 2. The odds ratio of reporting moderate to severe Edmonton Symptoms Assessment System score (≥4) during the first year following diagnosis using multivariable logistic regression.#

Tiredness

Well being

Pain

Drowsiness

Loss of appetite

Anxiety

Shortness Depression of breath

Age 1.01 1.00 0.98 0.99 1.00 0.99 1.00 0.98 (per year increase) (0.99-1.02) (0.99-1.02) (0.98-0.99) (0.99-1.01) (0.99-1.01) (0.98-1.00) (0.99-1.01) (0.87-0.99) Sex 1.36 1.35 1.28 1.19 1.43 1.59 0.92 1.32 (female vs. male) (1.11-1.68) (1.12-1.63) (1.09-1.52) (1.02-1.39) (1.22-1.67) (1.35-1.85) (0.79-1.06) (1.12-1.52) Location 0.85 0.99 1.00 0.96 1.04 1.29 0.99 1.06 (urban vs. rural) (0.63-1.15) (0.76-1.30) (0.79-1.27) (0.76-1.20) (0.83-1.32) (1.05-1.61) (0.79-1.22) (0.85-1.32) SES 1.31 0.95 0.93 1.10 0.92 1.18 1.15 0.89 (poor vs. non-poor) (1.01-1.69) (0.74-1.22) (0.75-1.16) (0.89-1.36) (0.74-1.14) (0.96-1.45) (0.94-1.40) (0.72-1.09) CCI 1.20 1.09 1.02 1.20 1.02 1.02 1.10 1.02 (≥ 2 vs. ≤ 1) (0.88-1.64) (0.83-1.44) (0.80-1.29) (0.95-1.51) (0.81-1.29) (0.82-1.28) (0.89-1.37) (0.81-1.27) Year of diagnosis 0.98 0.97 0.96 0.99 0.92 0.97 0.99 0.99 (per year increase) (0.94-1.01) (0.94-1.01) (0.93-0.99) (0.96-1.02) (0.89-0.95) (0.95-1.00) (0.97-1.02) (0.87-1.02) Anemia 1.05 1.05 0.96 1.00 0.95 1.23 1.19 1.11 (yes vs. no) (0.84-1.31) (0.86-1.28) (0.81-1.14) (0.84-1.18) (0.80-1.13) (1.05-1.45) (1.01-1.40) (0.95-1.31) Hypercalcemia 1.86 1.10 1.64 1.36 1.25 1.14 1.10 1.35 (yes vs. no) (0.98-3.53) (0.68-1.78) (1.06-2.54) (0.90-2.04) (0.83-1.88) (0.79-1.66) (0.76-1.58) (0.94-1.96) Bone disease 1.23 1.52 2.65 1.29 1.72 1.07 1.022 1.22 (yes vs. no) (0.87-1.73) (1.10-2.12) (1.92-3.67) (0.99-1.67) (1.31-2.27) (0.84-1.37) (0.80-1.30) (0.96-1.56) Renal Disease 1.26 1.31 0.85 1.12 1.36 1.07 1.09 1.081 (yes vs. no) (0.99-1.62) (1.05-1.64) (0.71-1.03) (0.93-1.35) (1.13-1.65) (0.89-1.27) (0.92-1.30) (0.91-1.29) Hospital 0.91 1.09 1.27 0.93 0.94 1.11 1.05 1.23 (Non-teaching vs. (0.72-1.17) (0.88-1.35) (1.05-1.54) (0.88-1.11) (0.78-1.14) (0.93-1.33) (0.88-1.27) (1.03-1.47) teaching) Novel drugs 1.01 0.96 1.20 0.94 0.96 0.99 0.88 0.95 (yes vs. no) (0.73-1.39) (0.71-1.30) (0.93-1.55) (0.73-1.20) (0.75-1.25) (0.78-1.26) (0.69-1.11) (0.75-1.21)

Nausea 0.98 (0.97-0.99) 1.56 (1.33-1.85) 1.35 (1.02-1.72) 1.06 (0.85-1.32) 1.07 (0.85-1.36) 0.95 (0.92-0.98) 0.99 (0.83-1.18) 1.18 (0.81-1.73) 1.12 (0.87-1.45) 1.07 (0.89-1.30) 1.08 (0.88-1.30) 1.12 (0.87-1.45)

Data are presented as odds ratios (95% confidence interval). A higher odds ratio is ‘worse’, indicating that the covariate is associated with a higher odds of reporting moderate to severe symptom. Bold indicates statistically significant values (P<0.05). SES: socioeconomic status; CCI: Charlson Comorbidity Index.

both at diagnosis and over time may be an essential first step in incorporating multidisciplinary teams, with a specific focus on managing psychosocial symptoms, which are known to be often unaddressed by oncology teams.6 Several patients’ characteristics were identified in our study as being associated with an increased odds of experiencing high symptom burden. Although increasing age was associated with a slight decrease in symptom burden, given the minimal change in odds ratio, the clinical impact of this is unknown. Compared to males, females had a higher symptom burden for nearly all recorded symptoms in our study, similarly to previously published literature in oncology which also reported sex differences in symptom burden and quality of life.7,8 Additionally, although increased co-morbidities has been previously shown to be correlated with a higher symptoms burden,9 we did not detect a statistically significant relationship in our retrospective study and future prospective studies are needed for further investigation. Non-teaching hospital was associated with higher rates of pain and depression and although the exact reason cannot be determined from our data, clinical outcomes are known to be different between teaching and non-teaching sites.10 Lastly, while the majority of patients in our study were taking novel drugs, these drugs were not associated with a decrease in symptom burden, suggesting that symptom management may require both effective anti-myeloma therapy as well as optimal supportive care services. Comparison of our results to those of other studies on transplant-ineligible patients is difficult due to the heterohaematologica | 2021; 106(7)

geneous populations included which encompass newlydiagnosed, relapsed as well as palliative patients.11,12 Despite these differences, the findings of our study are consistent with the results of a systematic review which also showed that severe fatigue and pain were common symptoms, with the pooled prevalence reaching more than 40% with anxiety/depression present in nearly 25% of the patients.13 A major strength of our study is that it is the largest, longitudinal study done on symptom burden among this group of patients using patient-reported outcomes. Our study utilizes real-world data with a focus on older adults with MM who are under-represented in clinical trials.14 Our study has several limitations. Our administrative database does not contain myeloma-specific variables, such as stage, response, or frailty status, which may be associated with symptom burden. Additionally, we did not capture symptoms during specific lines of treatment which may also be important as shown in a recent prospective study in which quality of life deteriorated with increasing line of treatment.15 The myeloma-defining ‘CRAB’ features may also be under-reported or overreported compared to those in prospectively conducted studies16 because of the limitations of diagnosis and billing codes in the administrative database. Similarly, retrospective collection of co-morbidities may also have led to under-reporting as noted in previous studiess.17 ESAS scores were only recorded during outpatient visits and potentially severe symptoms during hospital admissions may have been missed in our study. While the ESAS 1993

Letters to the Editor

has been validated for general cancer symptoms, we were not able to capture symptoms specifically related to MM or the impact of symptom burden on the disruption of functional and social activities of the patients or their caregivers. In conclusion, transplant-ineligible patients with MM experience a substantial burden of symptoms in the first year following diagnosis. Future prospective studies both in clinical trials and in real-world patients are needed to further evaluate factors associated with high symptom burden longitudinally while simultaneously evaluating interventional supportive care strategies to alleviate this burden. Hira S. Mian,1 Gregory R. Pond,2 Tanya M. Wildes,3 Branavan Sivapathasundaram,2 Jonathan Sussman1 and Hsien Seow1 1 Juravinski Cancer Center, Department of Oncology, McMaster University, Hamilton, Ontario, Canada; 2McMaster University, Hamilton, Ontario, Canada and 3Washington University School of Medicine, St. Louis, MO, USA Correspondence: HIRA S. MIAN - hira.mian@medportal.ca doi:10.3324/haematol.2020.267757 Received: September 11, 2020. Accepted: December 10, 2020. Pre-published: December 23, 2020. Disclosures: HSM has received consultancy fees and/or honoraria from Celgene, Takeda, Janssen, Amgen, and Sanofi; GRP has a close family member who is employed by Roche Canada, and who owns Roche Canada stock; he has also received consultancy fees from AstraZeneca and an honorarium from Takeda. TMW reports having received research funding from Janssen and is a consultant for Carevive Systems and Seattle Genetics. BS, JS and HS have no conflicts of interest to disclose. Contributions: HSM, GRP, HS and JS were responsible for the conception and design of the study; BS and HSM collected the data; HSM, GRP, BS, TMW, HS, and JS analyzed and interpreted the data. All authors were involved in writing the manuscript and approved its final version. Acknowledgments: the authors would like to thank Dr. Mark Levine for manuscript revision. Parts of the material are based on data and/or information compiled and provided by the Canadian Institute for Health Information (CIHI). However, the analyses, conclusions, opinions and statements expressed in the material are those of the authors, and not necessarily those of CIHI. We acknowledge Cancer Care Ontario for access to the Ontario Cancer Registry (OCR), Cancer Activity Level Reporting (ALR), Symptom Management Database (ESAS) and The New Drug Funding Program (NDFP). We thank IQVIA Solutions Canada Inc. for the use of their Drug Information Database. Funding: this investigation was supported by a grant from the Juravinski Cancer Centre Foundation and Myeloma Canada. The study was supported by ICES, which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care (MOHLTC). The opinions, results and conclusions reported in this arti-

1994

cle are those of the authors and are independent of the funding sources. No endorsement by ICES or the Ontario MOHLTC is intended or should be inferred.

References 1. Bruera E, Kuehn N, Miller MJ, Selmser P, Macmillan K. The Edmonton Symptom Assessment System (ESAS): a simple method for the assessment of palliative care patients. J Palliat Care. 1991;7(2):6-9. 2. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45(6):613-619. 3. Matheson FI, White HL, Moineddin R, Dunn JR, Glazier RH. Neighbourhood chronic stress and gender inequalities in hypertension among Canadian adults: a multilevel analysis. J Epidemiol Community Health. 2010;64(8):705-713. 4. 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. 5. Selby D, Cascella A, Gardiner K, et al. A single set of numerical cutpoints to define moderate and severe symptoms for the Edmonton Symptom Assessment System. J Pain Symptom Manage. 2010;39(2):241-249. 6. Hallet J, Davis LE, Isenberg-Grzeda E, et al. Gaps in the management of depression symptoms following cancer diagnosis: a populationbased analysis of prospective patient-reported outcomes. Oncologist. 2020;25(7):e1098-e1108. 7. Laghousi D, Jafari E, Nikbakht H, Nasiri B, Shamshirgaran M, Aminisani N. Gender differences in health-related quality of life among patients with colorectal cancer. J Gastrointest Oncol. 2019;10(3):453-461. 8. Cheung WY, Le LW, Gagliese L, Zimmermann C. Age and gender differences in symptom intensity and symptom clusters among patients with metastatic cancer. Support Care Cancer. 2011;19(3):417-423. 9. Ritchie CS, Zhao F, Patel K, et al. Association between patients' perception of the comorbidity burden and symptoms in outpatients with common solid tumors. Cancer. 2017;123(19):3835-3842. 10. Burke LG, Frakt AB, Khullar D, Orav EJ, Jha AK. Association between teaching status and mortality in US hospitals. JAMA. 2017;317(20):2105-2113. 11. Ramsenthaler C, Gao W, Siegert RJ, Edmonds PM, Schey SA, Higginson IJ. Symptoms and anxiety predict declining health-related quality of life in multiple myeloma: a prospective, multi-centre longitudinal study. Palliat Med. 2019;33(5):541-551. 12. Mols F, Oerlemans S, Vos AH, et al. Health-related quality of life and disease-specific complaints among multiple myeloma patients up to 10 yr after diagnosis: results from a population-based study using the PROFILES registry. Eur J Haematol. 2012;89(4):311-319. 13. Ramsenthaler C, Kane P, Gao W, et al. Prevalence of symptoms in patients with multiple myeloma: a systematic review and metaanalysis. Eur J Haematol. 2016;97(5):416-429. 14. Duma N, Azam T, Riaz IB, Gonzalez-Velez M, Ailawadhi S, Go R. Representation of minorities and elderly patients in multiple myeloma clinical trials. Oncologist. 2018;23(9):1076-1078. 15. Engelhardt M, Ihorst G, Singh M, et al. Real-world evaluation of health-related quality of life in patients with multiple myeloma from Germany. Clin Lymphoma Myeloma Leuk. 2021;21(2):e160-e175. 16. Kyle RA, Rajkumar SV. Criteria for diagnosis, staging, risk stratification and response assessment of multiple myeloma. Leukemia. 2009; 23(1):3-9. 17. Engelhardt M, Dold SM, Ihorst G, et al. Geriatric assessment in multiple myeloma patients: validation of the International Myeloma Working Group (IMWG) score and comparison with other common comorbidity scores. Haematologica. 2016;101(9):1110-1119.

haematologica | 2021; 106(7)

Letters to the Editor

Mixed-lineage leukemia protein modulates the loading of let-7a onto AGO1 by recruiting RAN The mixed-lineage leukemia (MLL) proto-oncogenic protein, as the founding member of human TrxG proteins, was originally identified through its association with both acute lymphoblastic leukemia and acute myeloid leukemia.1 MLL is a histone H3 lysine 4 (H3K4) methyltransferase that can execute methylation on a subset of target genes through its evolutionarily conserved SET domain, an activity that is essential for normal MLL function.2 MLL is proteolytically cleaved into two distinct subunits: MLLC180 and MLLN320, which non-covalently interact to assemble an intramolecular complex involved in epigenetic transcriptional regulation.2 MLL is routinely regarded as a nuclear protein. Interestingly, however, our recent research revealed that the MLLC180 subunit alone can localize to cytoplasmic processing bodies (P-bodies),3,4 where microRNA (miRNA)-mediated gene silencing takes place,5 and affect the function of a subset of miRNA, as exemplified by the let-7a family.3,4 The dysregulated function of let-7a resulting from the reduced expression of MLLC180 was very important for maintaining a high level of MYC in MLL leukemia.4 Thus, our work uncovered an unexpected role for MLL in miRNA-mediated translational repression. However, how MLL participates in the regulation of miRNA function remains elusive. We therefore sought to uncover the underlying mechanisms of how MLL participates in miRNA-mediated translational repression. In this study, we demonstrated that MLL was required to recruit let-7a and miR-10a to the miRNA-induced silencing complex (miRISC), partly through its binding partner RAN. The methods and datasets are available as Online Supplementary Information files. Most miRNA are loaded onto Argonaute (AGO) proteins in the miRISC and act as post-transcriptional regulators of their target mRNA.6 Unfortunately, how these miRNA are selectively loaded onto AGO proteins still remains poorly understood.6 Among miRISC-associated factors, AGO1 plays a predominant and specific role in miRNA-mediated translational repression.7 Our immunofluorescence results demonstrated that AGO1 and MLL were localized in the same cytoplasmic foci, which was disrupted upon MLL depletion (Figure 1A and B, Online Supplementary Figure S1A-C), suggesting an interaction between AGO1 and MLL. Using a specific P-body marker DCP1A, we further confirmed that MLL and AGO1 colocalized in the cytoplasmic P-bodies (Online Supplementary Figure S1D). Previous studies showed that Argonaute proteins could accumulate in stress granules in addition to P-bodies when cells were subjected to stress.8 We observed that upon arsenite treatment MLL, together with AGO1, could co-localize to stress granules, as indicated by the specific stress granule marker G3BP1 (Online Supplementary Figure S1E). These results are consistent with those of our previous study showing that MLL was present not only in P-bodies but also in stress granules.3 Co-immunoprecipitation experiments showed that MLLC180 but not MLLN320 interacts with AGO1 (Figure 1C). Additionally, we demonstrated that the interaction between MLLC180 and AGO1 preferentially occurs in the cytoplasm, and not in the nucleus (Figure 1D). Interestingly, the interaction between MLL and AGO1 decreased dramatically after RNase A treatment, as revealed by co-immunoprecipitation assays, indicating that this interaction was an RNA-dependent indirect interaction, rather than a direct protein-protein interachaematologica | 2021; 106(7)

tion (Figure 1E, Online Supplementary Figure S1F). Indeed, the interaction between MLL and AGO1 was enhanced by co-transfected let-7a (Figure 1F, Online Supplementary Figure S1G), indicating that miRNA might play a critical role in the MLL and AGO1 axis. miRNA-mediated gene silencing requires miRNA to associate with AGO proteins and other silencing factors to form a functional miRISC to repress target mRNA.6 Given that miRNA may fully function in mediating gene silencing even without the existence of microscopically visible P-bodies,9 functional miRISC may still be formed upon MLL depletion. We thus further examined whether the depletion of MLL would impair the recruitment of miRNA to form the functional miRISC. We focused on let-7a and miR-10a, which were two MLL-binding miRNA reported in our previous studies.3 We performed anti-AGO1 RNA immunoprecipitation (RIP) experiments and the results showed that MLL depletion resulted in the loss of binding of let-7a and miR-10a to AGO1 (Figure 1G, Online Supplementary Figure S1H-K). A pull-down assay using biotinylated let-7a further validated that the binding of AGO1 to let-7a was reduced in Mll knockout (Mll-/-) murine embryo fibroblasts (MEF) (Figure 1H). In addition, the recruitment of let-7a and miR-10a target mRNA, MYC, HRAS and HOXA1, to AGO1 was largely impaired in the MLL-depleted cells (Figure 1I and Online Supplementary Figure S1L and S1M). Notably, AGO1 expression was not affected by the knockdown of MLL (Online Supplementary Figure S1H), suggesting that this impaired recruitment of miRNA and its target mRNA to AGO1 was likely not caused by the reduced AGO1 protein levels. Together with the above-mentioned data showing that the interaction between MLL and AGO1 was RNA-dependent, these results indicated that MLL and miRNA may require each other in order to be efficiently recruited by AGO1 and form a functional miRISC. To further investigate the role of MLL in the recruitment of miRNA to miRISC, we reintroduced shRNAresistant MLLN320, MLLC180 or full-length MLL (MLLFL) into MLL knockdown 293T cells or Mll knockout (Mll-/-) MEF cells and found that the recruitment of let-7a and miR-10a to miRISC was rescued by exogenous MLLC180 (Figure 1J and K, Online Supplementary Figure S1P-S). Collectively, these results indicated that MLL plays a causal role in targeting miRNA and their target mRNA to AGO1 to form a translationally repressed miRISC complex, highlighting the importance of MLL in the control of miRNA-mediated expression. MLLC180 itself does not possess any predictable RNA recognition motif, so we reasoned that MLL might recruit RNA components indirectly through its binding partners. Our proteomics data showed that RAN, a small GTPase involved in the import of cargo through nuclear pore complexes,10 was one of the proteins displaying strong interactions with MLL in the cytoplasm (Figure 2A). In line with a previous study,11 we found that RAN interacted with MLL in an RNA-independent manner (Figure 2B). We further confirmed that RAN could pull down MLLC180, indicating a direct interaction between MLLC180 and RAN (Figure 2C). Moreover, immunofluorescence data showed that upon arsenite treatment, MLL together with RAN co-localized to stress granules, as revealed by a stress granule marker eIF3 (Figure 2D), suggesting a potential role of RAN in regulating mRNA translation besides involving the import of cargo. RAN and XPO5 can form a complex which plays a critical role in nucleocytoplasmic transport of pre-miRNA molecules.10 Unlike XPO5, which dissociates from pre-miRNA in the cytoplasm, RAN could still associate with pre-miRNA in the cyto1995

Letters to the Editor

Figure 1. MLL is required for the loading of let-7a onto AGO1. (A) 293T cells were transfected with GFP-AGO1. Immunofluorescence experiments were performed to visualize the localization of GFP-AGO1 and MLL. MLL-CT antibody, which recognizes MLLC180 (aa2829-2883), was used to detect MLL. Scale bar, 5 mm. (B) Mll wild-type (Mll+/+) and Mll knockout (Mll-/-) MEF cells were transfected with GFP-AGO1. Immunofluorescence experiments were performed to visualize the localization of GFP-AGO1 and MLL. Arrowheads show the localization of MLL with the GFP-AGO1. Scale bar, 5 mm. (C) 293T cells were transfected with FLAGtagged full-length MLL (MLLFL), MLLC320, MLLC180 or empty vector. Cell lysates were prepared and subjected to anti-FLAG immunoprecipitation assays. The interaction between MLL and AGO1 was analyzed by western blot assays using indicated antibodies. (D) The cytosolic and nuclear fractions of 293T cells were separated and subjected to immunoprecipitation using anti-MLL antibodies. Co-purified proteins were examined by immunoblots using the indicated antibodies. (E) 293T cell lysates were treated with RNase A followed by anti-MLL immunoprecipitation. Western blots were performed using the indicated antibodies. (F) The interaction between MLL and AGO1 was assessed after let-7a transfection. Anti-MLL immunoprecipitation assays were performed, results were analyzed by immunoblots with indicated antibodies. (G) Extracts of 293T-shScr and 293T-shMLL cells were subjected to RNA immunoprecipitation (RIP) analysis using antiAGO1 antibody, and pulled down RNA were analyzed by quantitative reverse transcription polymerase chain reaction (qRT-PCR) using specific primers for let-7a. (H) Mll+/+ and Mll-/- MEF cellular lysates were subjected to a biotinylated- let-7a RNA pull-down assay. Then let-7a-immunoprecipitated AGO1 proteins were subjected to western blot analysis. Scrambled miRNA were used as a negative control. (I) 293T-shScr and 293T-shMLL cells were transfected with Agomir-negative control (NC) and Agomir- let-7a mimic (let-7a) followed by anti-AGO1 RIP experiments at 24 h post-transfection. Total RNAs were isolated to analyze the MYC mRNA level by qRT-PCR. (J, K) 293T-shScr and 293T-shMLL cells with the latter being rescued by exogenous shRNA-resistant MLLN320, MLLC180 or MLLFL were performed with anti-AGO1 RIP experiments at 24 h after transfection. Total RNA were isolated to analyze the let-7a (J) and MYC (K) levels by qRT-PCR using specific primers. NS, no significant difference. *P<0.05, **P<0.01, ***P<0.001. Data represent mean and standard error fo mean of three independent experiments.

1996

haematologica | 2021; 106(7)

Letters to the Editor

E D

Figure 2. Legend on following page.

haematologica | 2021; 106(7)

1997

Letters to the Editor

Figure 2. MLL contributes to the loading of let-7a onto AGO1 through interacting with RAN. (A) List of MLL-associated proteins identified by mass spectrometric analysis. 293T cells transfected with MLL were harvested and subjected to the nuclear-cytoplasmic fractionation. The cytoplasmic fractions were prepared for the immunoprecipitation assays followed by mass spectrometric analysis. (B) 293T cell lysates were treated with RNase A followed by anti-MLL immunoprecipitation. Western blots were performed using the indicated antibodies. (C) Direct interaction between MLLC180 and GST-RAN was examined. Left panels: western blots showing the inputs of purified GST-RAN and Myc-MLLC180. Right panels: the pull-down immunoblots were shown with GST-RAN as the bait and the pulled MLLC180 detected by an anti-Myc antibody. (D) 293T cells were untreated (upper panels) or treated with arsenite (0.5 mM, 45 min) (lower panels), then fixed and stained with the indicated antibodies. Note that eIF3 is specific for stress granules. Arrowheads show the localization of MLL with RAN and eIF3. Scale bar, 5 mm. (E) The RPISeq tool was used to predict the interactions between RAN and let-7a or pre-let-7a. The random forest (RF) classifier and support vector machine (SVM) classifier represent the confidence of the prediction. In performance evaluation experiments, predictions with probabilities >0.5 were considered “positive”. (F) 293T cellular lysates were prepared and anti-RAN RIP experiments were performed. Pulled down RNA were isolated, pre-let-7a and mature let-7a were analyzed by qRT-PCR using specific primers. (G) 293T cellular lysates were subjected to biotinylated- let-7a RNA pull-down assays. Then let-7a-immunoprecipitated RAN proteins were subjected to western blot analysis. Scrambled miRNA were used as negative controls. (H) 293T-shScr and shRAN cells transfected with Agomir-negative control (NC) or Agomir- let7a mimic (let-7a) were subjected to dual luciferase reporter assays. The ratio of luciferase activity was measured and normalized to the value of the cells transfected with the control reporter and NC. (I) Extracts of 293T-shScr and shRAN cells, with the latter being rescued by shRNA-resistant RAN, were subjected to anti-AGO1 RIP assays. Pulled-down RNA were analyzed by qRT-PCR using specific primers for let-7a. (J) 293T-shScr and 293T-shRAN cellular lysates were subjected to biotinylated-let-7a RNA pull-down assays. Then let-7a-immunoprecipitated AGO1 proteins were subjected to western blot analysis. Scrambled miRNA were used as negative controls. (K) 293T cell lysates were treated with RNase A followed by anti-AGO1 immunoprecipitation. Western blots were performed using the indicated antibodies. (L) Extracts of 293T-shScr and 293T-shMLL cells were collected and co-immunoprecipitation assays were performed and analyzed using the indicated antibodies. (M) The proposed mechanism through which MLL and RAN are involved in the loading of let-7a onto AGO1. MLL is required for the loading of let-7a onto AGO1 via a direct interaction with RAN. Thus, RAN serves as a molecular adaptor for the assembly of MLL-associated miRISC. NS, no significant difference. *P<0.05, **P<0.01, ***P<0.001. Data represent the mean and standared error of mean of three independent experiments.

plasm.12 According to the RPISeq online tool,13 RAN could bind the pre-miRNA and the mature miRNA (Figure 2E, Online Supplementary Figure S2A). Next, our RIP assay confirmed that RAN could bind not only the pre-miRNA but also the mature miRNA (Figure 2F, Online Supplementary Figure S2B). Moreover, RNA pull-down results showed a higher RAN expression in the let-7a or miR-10a-biotinylated group compared to that in the control group (Figure 2G, Online Supplementary Figure S2C). These results were consistent with a previous finding that RAN was an RNA-binding protein,14 suggesting that RAN may be involved in the later steps of miRNA processing and function. We next probed whether RAN is required to mediate gene silencing of miRNA targets. As shown in Figure 2H and Online Supplementary Figure S2D,E, luciferase activity in RAN-depleted cells was increased compared with that in control cells, indicating that the loss of RAN impaired the let-7a and miR-10a silencing functions. RIP experiments showed that the binding of both let-7a, miR-10a and MYC, HOXA1 to AGO1 was decreased in RANdepleted cells, an effect that could be recovered by the reintroduction of RAN (Figure 2I, Online Supplementary Figure S2F-I). Our previous studies demonstrated that MLLC180 plays a causal role in the miRNA functional deficiency,3,4 so we then investigated the role of RAN-binding in MLLC180-regulated miRNA function. We found that MLLC180 failed to rescue the miRNA activity when Ran was depleted, an effect that could be recovered by MLLC180 together with reintroduction of RAN, suggesting that RAN was required for the MLLC180-mediated miRNA regulation (Online Supplementary Figure S2J-K). Given the fact that RAN is a small GTPase involved in nucleocytoplasmic transport,10 we determined whether the GTPase activity of RAN is required for the functional interaction of the MLL-miRISC complex. As revealed in Online Supplementary Figure S2L, both wild-type RAN (RANWT) and GTPase-deficient mutant (RANQ69L) could partially reverse the deficits in the binding of let-7a to AGO1 caused by loss of endogenous RAN. We also observed that depletion of RAN significantly impaired the interaction between MLL and AGO1, which could be recovered by RANWT or RANQ69L re-expression (Online Supplementary Figure S2M), suggesting that the GTPase activity of RAN was not required for the function of the MLL-miRISC complex. Additionally, the binding of AGO1 to let-7a or miR-10a was decreased in RAN-depleted cells as revealed by a pull-down assay using biotinylated let-7a or miR-10a (Figure 2J, Online Supplementary Figure S2N). These results 1998

indicated that RAN, beyond pre-miRNA export, was required for miRNA-mediated gene silencing. To decipher the role of RAN in the function of miRISC, we tested the interaction between RAN and AGO1. We observed that AGO1 had an RNA-dependent indirect interaction with RAN (Figure 2K). Importantly, coimmunoprecipitation experiments revealed that besides AGO1, DDX6 a key P-body protein specifically involved in miRNA-mediated translational repression,15 interacts with RAN, but these interactions decreased significantly upon MLL depletion (Figure 2L), indicating that MLL is accountable for these interactions. To further strengthen our findings, we explored how RAN behaves in a leukemic context. Co-immunoprecipitation assays performed in three leukemia cell lines, JM1, REH and U937, showed that MLL interacts with RAN (Online Supplementary Figure S2O). In REH and U937 cells, MLL together with RAN co-localized to stress granules following arsenite treatment, as illustrated by immunofluorescence assay (Online Supplementary Figure S2P). Additionally, we found that the binding of let-7a to AGO1 was decreased in RAN-depleted REH and U937 cells, an effect that could be restored by the reintroduction of RAN (Online Supplementary Figure S2Q-R). Consistent with the results obtained from 293T cells, we observed that AGO1 had an RNA-dependent indirect interaction with RAN in REH cells (Online Supplementary Figure S2S). Moreover, the interaction between AGO1 and RAN was impaired in MLL-depleted REH cells (Online Supplementary Figure S2T-V). As expected, the binding of RAN to AGO1 was reduced in MLL leukemic cells due to the downregulation of MLLC180 (Online Supplementary Figure S2W). Collectively, in the present study, we demonstrated that MLL was required for recruiting let-7a and its target mRNA to the miRISC, partly through its direct binding partner RAN (Figure 2M), unraveling an unexpected role for RAN in the loading of miRNA onto AGO1. Our findings provide an alternate mechanism and expanded the functional scope of RAN in the miRNA processing pathway. Thus, the discovery of interplay between MLL and miRNA represents a new regulatory layer, and an additional level of complexity, in the control of gene expression. Shouhai Zhu,1,2* Zhihong Chen,1* Dan Jiang,3* Ruiheng Wang,1 Xiaoyan Cheng,1 Dan Li,1 Qiongyu Xv,1 Fei Zhao,2 Wootae Kim,2 Guijie Guo,2 Chunjun Zhao,1 Zhenkun Lou2 and Han Liu1 haematologica | 2021; 106(7)

Letters to the Editor

1 Shanghai Institute of Hematology, State Key Laboratory of Medical Genomics, National Research Center for Translational Medicine at Shanghai, Ruijin Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China; 2Department of Oncology, Mayo Clinic, Rochester, MN, USA and 3Department of Endocrinology, Affiliated Hospital of Jiangsu University, Zhenjiang, Jiangsu, China. *SZ, ZC and DJ contributed equally as co-first authors. Correspondence: ZHENKUN LOU - lou.zhenkun@mayo.edu HAN LIU - liuhan68@sjtu.edu.cn doi:10.3324/haematol.2020.268474 Received: July 29, 2020. Accepted: December 11, 2020. Pre-published: December 17, 2020. Disclosures: no conflicts of interest to disclose. Contributions: SHZ, ZHC and DJ designed and performed most of the experiments, analyzed the data and wrote the draft manuscript; RHW, DL, QYX, FZ, GJG and WK provided technical assistance for the immunofluorescence experiments and data analyses; CJZ performed some experiments and provided expertise and extensively edited the manuscript; HL and ZKL contributed grant support, designed the entire project, wrote the manuscript and supervised the project. All authors discussed the results and commented on the manuscript. Acknowledgments: we would like to thank all the members of Liu’s laboratory for their technical assistance. We also thank the Core Facility and Technical Service Center (Shanghai Institute of Hematology) for generous support with cell imaging. We apologize for not citing all the relevant references due to space limitations. Funding: this work was supported by the National Key Research and Development Program of China (2018YFA0107802), the National Natural Science Foundation of China (81973996, 81900107 and 81570119), the Program of Shanghai Academic/Technology Research Leader (19XD1402500), the Shanghai Municipal Education Commission Gaofeng Clinical Medicine grant (20161304), the Shanghai Municipal Health Commission (2019CXJQ01), the Shu Guang project (14SG15), the Collaborative Innovation Center of Hematology, and the Samuel Waxman Cancer Research Foundation.

haematologica | 2021; 106(7)

References 1. Tkachuk DC, Kohler S, Cleary ML. Involvement of a homolog of Drosophila trithorax by 11q23 chromosomal translocations in acute leukemias. Cell. 1992;71(4):691-700. 2. Dou Y, Milne TA, Tackett AJ, et al. Physical association and coordinate function of the H3 K4 methyltransferase MLL1 and the H4 K16 acetyltransferase MOF. Cell. 2005;121(6):873-885. 3. Zhu S, Chen Z, Wang R, et al. MLL is required for miRNA-mediated translational repression. Cell Discov. 2019;3(5):43. 4. Zhu S, Cheng X, Wang R, et al. Restoration of microRNA function impairs MYC-dependent maintenance of MLL leukemia. Leukemia. 2020;34(9):2484-2488. 5. Sen GL, Blau HM. Argonaute 2/RISC resides in sites of mammalian mRNA decay known as cytoplasmic bodies. Nat Cell Biol. 2005; 7(6):633-636. 6. Ha M, Kim VN. Regulation of microRNA biogenesis. Nat Rev Mol Cell Biol. 2014;15(8):509-524. 7. Peters L, Meister G. Argonaute proteins: mediators of RNA silencing. Mol Cell. 2007;26(5):611-623. 8. Leung AK, Calabrese JM, Sharp PA. Quantitative analysis of Argonaute protein reveals microRNA-dependent localization to stress granules. Proc Natl Acad Sci U S A. 2006;103(48):18125-18130. 9. Leung AKL. The whereabouts of microRNA actions: cytoplasm and beyond. Trends Cell Biol. 2015;25(10):601-610. 10. Yi R, Qin Y, Macara IG, Cullen BR. Exportin-5 mediates the nuclear export of pre-microRNAs and short hairpin RNAs. Genes Dev. 2003; 17(24):3011-3016. 11. Nakamura T, Mori T, Tada S, et al. ALL-1 is a histone methyltransferase that assembles a supercomplex of proteins involved in transcriptional regulation. Mol Cell. 2002;10(5):1119-1128. 12. Wang X, Xu X, Ma Z, et al. Dynamic mechanisms for pre-miRNA binding and export by Exportin-5. RNA. 2011;17(8):1511-1528. 13. Muppirala UK, Honavar VG, Dobbs D. Predicting RNA-protein interactions using only sequence information. BMC Bioinformatics. 2011;12:489. 14. Brannan KW, Jin W, Huelga SC, et al. SONAR discovers RNA-binding proteins from analysis of large-scale protein-protein interactomes. Mol Cell. 2016;64(2):282-293. 15. Jonas S, Izaurralde E. Towards a molecular understanding of microRNA-mediated gene silencing. Nat Rev Genet. 2015;16(7):421433.

1999

Letters to the Editor

Glenzocimab does not impact glycoprotein VI-dependent inflammatory hemostasis Glycoprotein VI (GPVI), the main platelet receptor for collagen, has emerged as a new target for antithrombotic therapy because its genetic deficiency or pharmacological blocking inhibits platelet aggregation and experimental thrombosis without increasing bleeding time.1–5 While these data have stimulated the development of new antiplatelet drugs targeting GPVI, recent findings have indicated that GPVI is essential for repair of neutrophilinduced vascular injury in various inflamed organs and tissues.6–9 It thus appears important to assess and anticipate the yet uninvestigated risk of inflammation-induced bleeding under GPVI antagonists, especially considering that inflammation is a component of various thrombotic diseases. In that respect, it is worth noting that neutrophil mobilization is a predictor of hemorrhagic transformation of ischemic stroke10 and contributes to intraplaque hemorrhage,11 which is known to precipitate plaque rupture and the clinical expression of atherosclerosis. Among the newly developed drugs targeting GPVI, ACT017 (Glenzocimab, Acticor Biotech) is a humanized antibody fragment (Fab) that has already completed its phase I clinical trial in healthy volunteers12 and has just entered a phase II trial in stroke patients (Acute Ischemic Stroke Interventional Study “ACTIMIS”, clinicaltrials gov. Identifier: NCT03803007). ACT017 binds to human GPVI and inhibits the procoagulant activity and aggregation of collagen-stimulated platelets, as well as platelet adhesion and thrombus formation onto collagen surfaces under arterial flow conditions.1,13,14 The inhibitory action of ACT017 occurs without causing thrombocytopenia or depletion of GPVI, and is not associated with spontaneous bleeding events or increased bleeding time.14 Nevertheless, whereas preclinical bleeding time tests can help evaluate the risk of bleeding associated with trauma or surgery, they may not predict the risk of bleeding associated with inflammation.15 Here, using the cutaneous reverse passive Arthus reaction (rpA) as a model situation where GPVI plays a major role in inflammatory hemostasis, we investigated whether ACT017 increases the risk of inflammation-induced bleeding. We first assessed the contribution of GPVI to the prevention of inflammation-induced bleeding by platelets in the brain and lungs. In agreement with previous results obtained with an antibody causing depletion of mouse Gpvi,16,17 there was no cerebral hemorrhage in any of the Gpvi-/- mice subjected to transient (90 minutes) middle cerebral artery occlusion (Figure 1A). In contrast, cerebral hemorrhage occurred in all mice that had been rendered severely thrombocytopenic by the mean of a plateletdepleting antibody (Figure 1A). Genetic deficiency in GPVI was not associated with an increased bleeding risk in the model of acute lung injury induced by inhalation of Pseudomonas aeroginosa endotoxin either (Figure 1B). In the cutaneous rpA, as predicted by previous reports,6,7,9 GPVI-/- mice developed skin bleeding at the inflammatory reaction site, a bleeding phenotype that was seen neither in GPVI+/+ mice nor in GPVI+/- mice, which expressed half of normal GPVI surface levels (Figure 1C and D). Taken together, these results are consistent with evidence that GPVI is dispensable for hemostasis in the inflamed brain and lungs6,16–18 but primarily involved in the prevention of bleeding in the rpA-inflamed skin. Notably, they further indicate that 50% of normal GPVI surface levels are sufficient for hemostasis during the cutaneous rpA. 2000

The ability of ACT017 to inhibit collagen/GPVI interactions and their functional consequences has been previously demonstrated in humans and in nonhuman primates.14 However, it has not been tested in hGPVI mice. We thus verified the activity of ACT017 against GPVI from hGPVI mice. Like its murine precursor Fab 9O121, ACT017 added to whole blood from hGPVI mice caused a drastic reduction in platelet adhesion and aggregation onto collagen fibers under arterial and venous flow conditions (Figure 2A and B; Online Supplementary Movie). We next tested whether hGPVI mice treated with therapeutic (16 mg/kg) or higher doses of ACT017 (32 and 64 mg/kg) were sensitized to inflammation-induced bleeding during the cutaneous rpA. No bleeding occurred in ACT017treated hGPVI mice, whatever the dose of ACT017 used (data not shown, Figure 2C and D). There was no bleeding either when ACT017 at the highest dose tested (64 mg/kg) was given through a continuous infusion over the 4 hours of rpA (data not shown). The absence of bleeding in hGPVI mice treated with ACT017 was in contrast to the petechial bleeding observed in GPVI-/- mice (Figure 1) and in platelet-depleted GPVI+/+ and hGPVI mice (Figures 1C and D, and 2C and D), which is known to be a consequence of neutrophil recruitment.7 Absence of bleeding in rpA-challenged hGPVI mice treated with ACT017 was not due to altered neutrophil recruitment as this was comparable to that in hGPVI mice (Figure 2E). Interestingly, the latter result indicates that ACT017 does not impact neutrophil recruitment, at least in this model. Importantly, ACT017 did not alter platelet recruitment to the inflamed skin either (Figure 2F). The latter result underscores a major difference between the impact of genetic deficiency in GPVI and that of GPVI blocking by ACT017. In fact, bleeding in rpA-challenged Gpvi-/- mice was previously shown to be associated with reduced platelet recruitment at the reaction site.7 Solid tumors represent another inflammatory situation in which platelets were shown to continuously prevent leukocyte-induced bleeding and recent data have suggested that GPVI could be central to this function.19 Like in the cutaneous rpA, acute treatment of hGPVI mice bearing skin tumors with ACT017 did not cause tumor bleeding. The absence of effect of ACT017 on tumor vessel stability was in contrast to the effect of acute depletion of platelets, which caused tumor bleeding (Figure 2F). In conclusion, in addition to confirming that GPVI is not required for inflammation-associated hemostasis in the brain and lungs, our results show that pharmacological blockade of GPVI by ACT017 does not impair GPVIdependent inflammatory hemostasis. There are several non-exclusive reasons that could explain why pharmacological inhibition of GPVI by ACT017 does not impair the vasculoprotective recruitment of platelets during the cutaneous rpA. First, it was shown previously that GPVI can co-operate with other platelet receptors like integrin a2β1 to provide residual collagen-dependent platelet activation when its collagen binding site is blocked pharmacologically.20 Furthermore, while ACT017 blocks the interactions between GPVI and collagen, it remains unknown whether ACT017 has similar blocking effects towards the other ligands of GPVI. Besides collagen, fibrin(ogen) and a number of adhesive proteins of the vessel wall have been reported as GPVI ligands (e.g., laminin, fibronectin and vitronectin) and could thus provide redundant binding mechanisms. Moreover, were collagen to be one of the ligands supporting the adhesion of platelets to inflamed skin vessels, it is interesting to note that despite a drastic reduction in platelet adhesion and haematologica | 2021; 106(7)

Letters to the Editor

Figure 1. Contribution of glycoprotein VI to inflammation-associated hemostasis. The contribution of glycoprotein VI (GPVI) to inflammation-associated hemostasis was determined in three different models of acute inflammation. (A) Representative images of brain sections taken 24 hours after GPVI+/+, GPVI-/-, GPVI-/-, and platelet-depleted mice were subjected to 90 minutes transient middle cerebral artery occlusion (tMCAO). Note that tMCA0 caused bleeding only in platelet-depleted mice. The images are representative of n=6 mice per group. (B) Representative images of the bronchoalveolar lavage fluid from Gpvi+/+, Gpvi/, and platelet-depleted mice collected 24 hours after lipopolysaccharide inhalation. The images are representative of n=8 mice per group. (C and D) Effect of partial or complete GPVI deficiency on inflammatory bleeding during the cutaneous reverse passive Arthus reaction (rpA). (C) Representative images of the skin of GPVI+/+, GPVI-/-, GPVI-, and platelet-depleted mice after 4 hours of rpA. The images are representative of n=7-10 mice per group. Bar =500 mm. (D) Skin hemoglobin content after 4 hours of rpA. # indicates a significant difference (P<0.05) from the rpA GPVI+/+ group, n=14-20 skin biopsies per group. Inset: Representative histogram of flow cytometry analysis of GPVI surface levels in GPVI+/+, GPVI+/-, and GPVI-/- mice, as assessed using the JAQ1 antibody to representative mouse GPVI.

haematologica | 2021; 106(7)

2001

Letters to the Editor

Figure 2. Impact of ACT017 on glycoprotein VI-dependent hemostasis. (A and B) Citrated whole blood from hGPVI mice was labeled with the fluorochrome DiOC6, incubated or not with ACT017 (80 mg/mL) for 10 minutes, and perfused at a wall shear rate of 1,500 s–1 or 100 s–1 for 3 minutes over a collagen-coated surface. Bar=50 mm. (A) Representative images of platelet coverage at the end of the perfusion. (B) Mean surface areas covered by platelets calculated from 20 different fields taken with a 20x objective along channels from four different runs (five fields per run). (C) Representative images of the skin of hGPVI mice treated or not with ACT017 (64 mg/kg) after 4 hours of reverse passive Arthus reaction (rpA). The images are representative of n=4-9 mice per group. Bar=500 mm. (D) Skin hemoglobin content after 4 hours of rpA. # indicates a significant difference (P<0.05) from the rpA hGPVI group, n=6-18 skin biopsies per group. (E and F) Skin myeloperoxidase (E) and platelet factor 4 (PF4) (F) content after 2 hours of rpA, as assessed by enzyme-linked immunosorbent assay; n=12 skin biopsies per group.

aggregation onto fibrillar collagen in the presence of ACT017 (Figure 2A and B; Online Supplementary Movie), residual platelet adhesion was observed at both arterial and venous blood flow. Considering that ACT017 has no effect on platelet recruitment during cutaneous rpA and that previous results have shown that individual platelets and platelet monolayers ensure hemostasis at sites of mild inflammatory vascular injury,15 such residual interactions with collagen could be sufficient for inflammatory hemostasis. Previous studies have also shown 2002

that platelets are particularly efficient in maintaining vascular integrity in inflamed organs, as platelet counts as low as 10% can support this function.15 Consistent with this notion, Gpvi+/- mice with half of normal GPVI surface levels showed normal hemostasis during the cutaneous rpA (Figure 1C and D). All in all, our results indicate that the highly favorable safety profile of ACT017 suggested by previous results in bleeding time assays and by the absence of adverse bleeding events in the phase I clinical trial12 also applies to inflammatory situations. Whether haematologica | 2021; 106(7)

Letters to the Editor

the safety profile of ACT017 still holds true when combining it with other drugs like recombinant tissue-type plasminogen activator remains to ascertain, but the absence of effect of ACT017 on platelet recruitment to the inflamed vasculature suggests there is a realistic chance for it to be maintained. Soumaya Jadoui,1* Ophélie Le Chapelain,1* Véronique Ollivier,1 Ali Mostefa-Kara,1 Lucas Di Meglio,1 Sébastien Dupont,1 Angèle Gros,1 Mialitiana Solo Nomenjanahary,1 Jean-Philippe Desilles,1,2 Mikaël Mazighi,1,2 Bernhard Nieswandt,3 Stéphane Loyau,1 Martine Jandrot-Perrus,1 Pierre H. Mangin4 and Benoit Ho-Tin-Noé1 1 Université de Paris, LVTS, INSERM U1148, Paris, France; 2 Department of Interventional Neuroradiology, Rothschild Foundation Hospital, Paris, France; 3University Hospital Würzburg, Rudolf Virchow Center for Experimental Biomedicine, Würzburg, Germany and 4Université de Strasbourg, INSERM, EFS Grand-Est, BPPS UMR-S1255, FMTS, Strasbourg, France *SJ and OLC contributed equally as co-first authors. Correspondence: BENOIT HO-TIN-NOE’ - benoit.ho-tin-noe@inserm.fr doi:10.3324/haematol.2020.270439 Received: August 25, 2020. Accepted: December 21, 2020. Pre-published: December 30, 2020. Disclosures: MJ-P is the founder of Acticor Biotech; all other authors have no conflicts of interest to disclose. Contributions: SJ, OLC, VO, MSN, AM,LDM, SD, AG, and SL performed research and data analysis; JPD, MM, MJP, BN, and PHM provided critical feedback and helped shape the research; BTHN supervised the study, performed research and data analysis, and wrote the manuscript with support of SJ, OLC, VO, SD and MJP; all authors discussed the results and contributed to the final manuscript. Funding: this work was supported by grants from INCA (2016-09/435/NI-KA), La Fondation ARC (PJA 20151203107), La Fondation pour la Recherche Médicale (grant #DPC20171138959), and by public grants overseen by the French National Research Agency (ANR) as part of the Investments for the Future program (PIA) under grant agreement No. ANR-18-RHUS-0001 (RHU Booster) and BPI (CMI2 project TherAVC2.0); SJ was awarded a grant from la Société Française d’Hématologie.

References 1. Mangin PH, Tang CJ, Bourdon C, et al. A humanized glycoprotein VI (GPVI) mouse model to assess the antithrombotic efficacies of anti-GPVI agents. J Pharmacol Exp Ther. 2012;341(1):156-163. 2. Dütting S, Bender M, Nieswandt B. Platelet GPVI: a target for

haematologica | 2021; 106(7)

antithrombotic therapy?! Trends Pharmacol Sci. 2012;33(11):583590. 3. Jiang P, Jandrot-Perrus M. New advances in treating thrombotic diseases: GPVI as a platelet drug target. Drug Discov Today. 2014; 19(9):1471-1475. 4. Stegner D, Haining EJ, Nieswandt B. Targeting glycoprotein vi and the immunoreceptor tyrosine-based activation motif signaling pathway. Arterioscler Thromb Vasc Biol. 2014;34(8):1615-1620. 5. Rayes J, Watson SP, Nieswandt B. Functional significance of the platelet immune receptors GPVI and CLEC-2. J Clin Invest. 2019; 129(1):12-23. 6. Rayes J, Jadoui S, Lax S, et al. The contribution of platelet glycoprotein receptors to inflammatory bleeding prevention is stimulus and organ dependent. Haematologica. 2018;103(6):e256-e258. 7. Gros A, Syvannarath V, Lamrani L, et al. Single platelets seal neutrophil-induced vascular breaches via GPVI during immune-complex-mediated inflammation in mice. Blood. 2015;126(8):1017-1026. 8. Hillgruber C, Pöppelmann B, Weishaupt C, et al. Blocking neutrophil diapedesis prevents hemorrhage during thrombocytopenia. J Exp Med. 2015;212(8):1255-1266. 9. Boulaftali Y, Hess PR, Getz TM, et al. Platelet ITAM signaling is critical for vascular integrity in infammation. J Clin Invest. 2013; 123(2):908-916. 10. Maestrini I, Strbian D, Gautier S, et al. Higher neutrophil counts before thrombolysis for cerebral ischemia predict worse outcomes. Neurology. 2015;85(16):1408-1416. 11. Michel JB, Martin-Ventura JL, Nicoletti A, Ho-Tin-Noé B. Pathology of human plaque vulnerability: mechanisms and consequences of intraplaque haemorrhages. Atherosclerosis. 2014;234(2):311-319. 12. Voors-Pette C, Lebozec K, Dogterom P, et al. Safety and tolerability, pharmacokinetics, and pharmacodynamics of ACT017, an antiplatelet GPVI (glycoprotein VI) Fab: first-in-human healthy volunteer trial. Arterioscler Thromb Vasc Biol. 2019;39(5):956-964. 13. Lecut C, Feeney LA, Kingsbury G, et al. Human platelet glycoprotein VI function is antagonized by monoclonal antibody-derived fab fragments. J Thromb Haemost. 2003;1(12):2653-2662. 14. Lebozec K, Jandrot-Perrus M, Avenard G, Favre-Bulle O, Billiald P. Design, development and characterization of ACT017, a humanized Fab that blocks platelet’s glycoprotein VI function without causing bleeding risks. MAbs. 2017;9(6):945-958. 15. Ho-Tin-Noé B, Boulaftali Y, Camerer E. Platelets and vascular integrity: how platelets prevent bleeding in inflammation. Blood. 2018;131(3):277-288. 16. Kraft P, Schuhmann MK, Fluri F, et al. Efficacy and safety of platelet glycoprotein receptor blockade in aged and comorbid mice with acute experimental stroke. Stroke. 2015;46(12):3502-3506. 17. Kleinschnitz C, Pozgajova M, Pham M, Bendszus M, Nieswandt B, Stoll G. Targeting platelets in acute experimental stroke: impact of glycoprotein Ib, VI, and IIb/IIIa blockade on infarct size, functional outcome, and intracranial bleeding. Circulation. 2007;115(17):23232330. 18. Claushuis TAM, de Vos AF, Nieswandt B, et al. Platelet glycoprotein VI aids in local immunity during pneumonia-derived sepsis caused by gram-negative bacteria. Blood. 2018;131(8):864-876. 19. Volz J, Mammadova-Bach E, Gil-Pulido J, et al. Inhibition of platelet GPVI induces intratumor hemorrhage and increases efficacy of chemotherapy in mice. Blood. 2019;133(25):2696-2706. 20. Nieswandt B, Brakebusch C, Bergmeier W, et al. Glycoprotein VI but not a2β1 integrin is essential for platelet interaction with collagen. EMBO J. 2001;20(9):2120-2130.

2003

Letters to the Editor

Comparison of CD38 antibodies in vitro and ex vivo mechanisms of action in multiple myeloma CD38, a transmembrane glycoprotein, is widely expressed on multiple immune cell populations.1,2 High expression of CD38 on myeloma cells makes it a target of choice for therapeutic antibodies targeting cell surface molecules in multiple myeloma (MM).2 CD38 functions as a receptor for CD31 and as an ectoenzyme catalyzing the reaction between NAD+ and NADP+ to generate cyclic ADP ribose (ADPR), NAADP, and ADPR.3 Daratumumab, a human IgG1κ monoclonal antibody (mAb) targeting CD38, eliminates MM cells through several direct mechanisms: antibody-dependent cellular phagocytosis (ADCP), complement-dependent cytotoxicity (CDC), antibody-dependent cell-mediated cytotoxicity (ADCC), and apoptosis,4-7 as well as immunomodulatory mechanisms.2,8 Other CD38-targeting mAb in clinical development, isatuximab (ISA) and TAK-079, are reported to act similar to daratumumab.9,10 It remains unclear how the pleiotropic mechanisms of CD38-targeting mAb collectively impact tumor cytolysis and exhibit anti-tumor effects in a comprehensive ex vivo immune milieu. Here, we report results of mechanistic comparison studies of three CD38-targeting mAb: daratumumab and analogs of ISA and TAK-079 (generated based on the published antigen-binding fragment sequences for ISA11 and TAK-079,12 respectively). In order to assess antibody binding, CD38-expressing Daudi and LP-1 tumor cells were coated with daratumumab, ISA analog, or TAK-079 analog antibodies at varying concentrations. Cells were washed and stained with Live/Dead® (Invitrogen, Carlsbad, CA, USA) and Alexa Fluor 647–conjugated goat anti-human Fc (Jackson ImmunoResearch, West Grove, PA, USA), and binding was analyzed by flow cytometry on a fluorescence-activated cell sorting (FACS) Celesta instrument (BD Biosciences, San Diego, CA, USA). CD38 expression was measured using CD38 (clone HIIT2) PerCp-Cy5.5 (BioLegend, San Diego, CA, USA). All three CD38 mAb (daratumumab, ISA analog, and TAK-079 analog) demonstrated similar relative binding to the target cells, which is in line with earlier findings of the binding properties of the three mAb.5 CDC activity of the three CD38 mAb was tested on multiple cell lines with a range of CD38 surface expression and CDC sensitivity levels (Online Supplementary Figure S1A).13 In Daudi, LP-1, and MOLP-8 cells, daratumumab resulted in higher levels of CDC activity compared with the other CD38 mAb, with a more pronounced difference seen in Daudi cells (Online Supplementary Figure S1B). In contrast, LP-1 and MOLP8 cells were susceptible to CDC activity with all three CD38 mAb. However, in LP-1 cells, daratumumab exhibited higher maximal cytotoxicity versus ISA and TAK-079 analogs and lower half maximal effective concentration (EC50) versus TAK-079 (Online Supplementary Figure S1C). Similarly, in MOLP-8 cells, daratumumab exhibited higher maximal cytotoxicity versus ISA analog and lower EC50 versus TAK-079 analog. In ADCC assays (E:T ratio, 50:1) using peripheral blood mononuclear cells as effectors, all three CD38 mAb induced similar levels of target cell death (Online Supplementary Figure S2A). Compared to Daudi cells, MOLP-8 and LP-1 cells were less susceptible to ADCC activity. In ADCP assays (E:T ratio 4:1) using monocytederived M2c macrophages as effectors, all three CD38 mAb induced similar levels of target cell phagocytosis as 2004

detected by pHrodo labeling of target cells after 4 hours (Online Supplementary Figure S2B). Daudi cells and MOLP-8 cells were phagocytosed by M2c macrophages by all three CD38 antibodies. However, LP-1 cells were relatively the most resistant to phagocytosis. Apoptosis was assessed in the presence and absence of Fc receptor (FcR) crosslinking. Phosphatidylserine translocation to the cell surface was induced by the ISA analog in the absence of FcR crosslinking as measured by annexin V staining, similar to previously published reports (Figure 1A).9 Neither daratumumab nor the TAK079 analog could elicit annexin V staining in the absence of crosslinking. However, in the presence of crosslinking, all three CD38 mAb could induce annexin V staining in Daudi cells. In order to probe the level of cell death over time, we used a 5-day cytotoxicity assay to detect metabolically active cells (Figure 1B and C). All three CD38 mAb elicited comparably high levels of cell death in the presence of the FcR crosslinker and low levels in its absence. Minimal activation-induced cell death (AICD) was observed with LP-1 (Figure 1D) or MOLP-8 cells (Figure 1D). Daratumumab has been shown to reduce CD38 expression levels partly by trogocytosis.14 In order to assess trogocytosis, we utilized Daudi cells as targets and THP-1 cells as effectors. Time-dependent loss of CD38 mAb staining on Daudi cells was correlated with a gain of signal on THP-1 cells (Online Supplementary Figure S3A). Daratumumab with a silent Fc did not mediate loss of CD38 on cell surfaces, indicating the effect is Fc dependent. Membrane dye was transferred in addition to CD38 (Online Supplementary Figure S3B), and imaging showed comparable efficiency of target transfer from Daudi to THP-1 cells among all three CD38 mAb (Online Supplementary Figure S3C). In the absence of THP-1, there was a negligible loss of signal for all three mAb observed on Daudi cells. The lysosome-associated membrane protein 1 (CD107a), co-localized with CD38 in effector cells, suggesting that CD38 is degraded after trogocytosis (Online Supplementary Figure S3D). All three CD38 mAb showed comparable results, suggesting that all can mediate trogocytosis. We aimed to compare cytotoxicity of the CD38 mAb ex vivo utilizing all mechanisms of action (MOA). The cumulative effect of the CD38 mAb was compared using europium-labeled LP-1 and MOLP-8 cells in the presence of whole blood from healthy donors containing both effector cells and complement. Within the assay, daratumumab demonstrated a significantly higher maximal cytotoxicity than comparator mAb in LP-1 cells (P<0.0001 for both comparisons) and MOLP-8 cells (P=0.0016 for both comparisons; Figure 2). Moreover, the EC50 was significantly lower for daratumumab versus the TAK-079 analog in both cell lines (P<0.0001) and was lower than the ISA analog in MOLP-8 cells (P=0.0008). Similar trends were seen with a 24-hour assay using flow cytometry as a read-out. Bone marrow samples from untreated newly diagnosed patients, containing tumor cells and autologous immune effector cells, were obtained commercially to compare the cumulative impact of the mode of action (MOA) of CD38 mAb ex vivo. Depletion of the CD19–CD20–CD38+ CD138+ MM cells was measured by flow cytometry after 3 days in the presence of CD38 mAb and human complement (Figure 3A). Daratumumab elicited higher percent cytotoxicity of the CD38+CD138+ MM cells compared with ISA and TAK079 analogs (Figure 3B). CD38 was detected using HuMab, which was developed to not compete with haematologica | 2021; 106(7)

Letters to the Editor

Figure 1. Legend on next page. haematologica | 2021; 106(7)

2005

Letters to the Editor

Figure 1. Activation-induced cell death activity of CD38 monclonal antibodies, in the absence and presence of crosslinker. Analyzed at 24 hoursa in Daudi cells by (A) annexin V staining and (B) at 5 daysb by CellTiter Glo Assay; no activity was seen in (C) LP-1 cells at 5 daysb by CellTiter Glo Assay and (D) MOLP-8 cells at 48 hoursa by annexin V staining and at 5 daysb by CellTiter Glo Assay. Daudi cells were plated with or without crosslinker (AffiniPure F(ab')₂ fragment goat antihuman IgG, Fcγ fragment specific, in which endotoxin was removed; Jackson ImmunoResearch, West Grove, PA, USA). Antibody dilutions were added and incubated for 24 hours (annexin V staining) or 5 days (CellTiter Glo Luminescent Cell Viability Assay). After 24 hours, cells were washed and stained with Live/Dead (Invitrogen) followed by annexin V (BioLegend, San Diego, CA, USA) and analyzed by flow cytometry (FACSCanto). After 5 days, CellTiter Glo reagent was prepared according to the manufacturer’s recommendations and added to a second plate to assess viability. Luminescence was measured and normalized to “plate max” and “plate min”. This analysis was also conducted using MOLP-8 cells, with 48-hour (annexin V staining) or 5-day (CellTiter Glo) incubation. aData are a summary of three independent experiments performed in duplicate; bdata are a summary of three independent experiments performed in triplicate. ISA: isatuximab; XL: crosslinking.

Figure 2. Daratumumab demonstrated higher cytotoxicity than comparator CD38 monoclonal antibodies in whole blood assays with (A) LP-1 and (B) MOLP8 cells. Multiple myeloma cell lines LP-1 or MOLP-8 were labeled with DELFIA® europium solution (PerkinElmer, Pittsburgh, PA, USA) according to the manufacturer’s protocol. Healthy donor blood samples were added with a titration of daratumumab, isatuximab (ISA) analog, TAK-079 analog, or isotype control. Europium release was measured after 3-hour incubation. Percent cytotoxicity = [(experimental lysis – min lysis)/(max lysis – min lysis)] x 100. Daudi cells did not effectively label with europium in this assay, which is a phenomenon that has been previously described. Data are shown as representative experiments (LP-1, n=6 donors; MOLP-8, n=5 donors).

daratumumab. It also did not compete with the ISA analog or TAK-079 analog. Data were similar with gating on CD19–CD20–CD138+CD27dim cells (Figure 3C) The results from this study add to the literature on the MOA of CD38 mAb and can provide insight into potential clinical differences that may be seen among the agents. We confirmed that all three mAb bind to CD38 at a similar level. Additionally, all three mAb demonstrated CDC, ADCC, ADCP, AICD, and trogocytosis MOA. Although most mechanisms were similar among the three mAb, daratumumab demonstrated higher CDC activity and, in the presence of human serum (which allows all possible MOA for antibody activity), showed stronger depletion of MM cells. The cell lines tested varied in their sensitivity to different effector functions, partly due to differing expression levels of CD38 and complement inhibitory proteins. Regardless, daratumumab had greater CDC activity across cell lines compared to the ISA analog and TAK-079 analog. In contrast to a report published by Jiang et al.,9 we did not observe AICD in MOLP-8 cells. Although it is unclear why this was observed, one possibility is that the MOLP-8 cells used in our study had lower CD38 levels or were more resistant to AICD. The difference between the results from the 24-hour annexin V staining and the 5-day CellTiter Glo Assay in the absence of crosslinking suggests that the impact of AICD over time without crosslinking is minimal; cells may be able to recover and continue to proliferate. However, in the presence of crosslinking, AICD resulted in more 2006

effective tumor killing by all three CD38 mAb in Daudi cells. The structural differences between daratumumab and ISA have previously been hypothesized to account for the differing interactions with FcR crosslinking.15 Neither MOLP-8 nor LP-1 cells were susceptible to AICD in this study. The ex vivo assay using healthy donor blood and MM cell lines was performed within 3 hours. At this timepoint, ADCC, CDC, and ADCP were the major mechanisms responsible for MM cell ablation. Whole blood contains endogenous complement, natural killer (NK) cells, and monocytes, which function as effector cells. This assay was repeated at 24 hours with similar results using absolute cell counts by flow cytometry of the labeled MM cell lines. In the MM patient samples, which contain endogenous NK cells and monocytes, the superiority of daratumumab killing was maintained even after 3 days. It is likely that daratumumab had the highest maximal cytotoxicity because of its superior CDC activity. Our study has several limitations. First, daratumumab was compared with analogs of comparators, ISA and TAK-709. Second, not all anti-tumor mechanisms of CD38 mAb, including direct inhibition of enzymatic activity, were tested in this study.5 Last, observed differences in CDC were tested in blood, and findings may vary in the setting of a bone marrow compartment in MM patients. In conclusion, daratumumab and surrogate analogs of ISA and TAK-079 have generally similar MOA. It haematologica | 2021; 106(7)

Letters to the Editor

remains to be determined in clinical trials if these in vitro differences lead to differences in clinical benefit. Michelle Kinder,1 Nizar J. Bahlis,2 Fabio Malavasi,3 Bart de Goeij,4 Alexander Babich,1 Jocelyn Sendecki,1 Joshua Rusbuldt,1 Kevin Bellew,1 Colleen Kane1 and Niels W.C.J. van de Donk5 1 Janssen Research & Development, LLC, Spring House, PA, USA; 2University of Calgary, Arnie Charbonneau Cancer Institute, Calgary, Canada; 3Department of Medical Science, University of Turin and Fondazione Ricerca Molinette, Turin, Italy; 4 Genmab B.V., Utrecht, the Netherlands and 5Department of Hematology, Amsterdam University Medical Center, Vrije Universiteit Amsterdam, Cancer Center Amsterdam, Amsterdam, the Netherlands Correspondence: NIELS W.C.J. VAN DE DONK n.vandedonk@amsterdamumc.nl doi:10.3324/haematol.2020.268656 Received: August 4, 2020. Accepted: December 21, 2020. Pre-published: January 14, 2021. Disclosures: MK was an employee of Janssen and owns stock/shares in Johnson & Johnson; NJB received research funding from Celgene, served in a consulting or advisory role for AbbVie, Amgen, Celgene, Janssen, Sanofi, and Takeda, and had travel, accommodations, or other expenses paid or reimbursed by Celgene and Janssen; FM received research support from Janssen, Celgene, Tusk Therapeutics, and Centrose, and served on advisory boards for Centrose, Tusk Therapeutics, Janssen, Takeda, and Sanofi; BdG is an employee of Genmab; JS, JR, and CK are employees of Janssen and own stock/shares in Johnson & Johnson; AB and KB are employees of Janssen; NWCJvdD received research funding from Amgen, Bristol Myers Squibb, Celgene, Janssen, and Novartis, and haematologica | 2021; 106(7)

Figure 3. Daratumumab depletes multiple myeloma cells in patient samples as depicted by (A) counts of CD38+CD138+, (B) percent depletions of CD138+CD38+, and (C) percent depletion of CD27dimCD138+ multiple myeloma cells.a Peripheral blood mononuclear cells or bone marrow mononuclear cells from multiple myeloma (MM) patients were obtained from MM patients according to the guidelines of the Ethics Committee of the Discovery Life Sciences (Huntsville, AL, USA) and in compliance with Declaration of Helsinki protocols. Cells were thawed and measured for viability/density, and 200,000 live cells were seeded to assay plates. MM patient cells were treated with daratumumab, isatuximab (ISA) analog, or TAK-079 analog at specified concentrations in the presence of 10% human complement. After 3 days, MM cell numbers were measured using Precision Count BeadsTM (BioLegend) and by gating on live CD19-CD20–CD138+CD38+ (HuMab; does not compete with tested CD38 monoclonal antibodies). The percent cytotoxicity was determined relative to the corresponding IgG1 control. Complement was present in each experiment. Data are shown as representative experiment at 10 mg/mL treatment. aPeripheral blood mononuclear cells or bone marrow–derived macrophages from MM patients (n=5 donors).

served in a consulting or advisory role for Amgen, Bayer, Bristol Myers Squibb, Celgene, Janssen, Novartis, Servier, and Takeda. Contributions: all authors developed the manuscript, provided final submission approval, and confirmed that the data were accurate and complete. Acknowledgmemts: the authors would like to thank Amy Wong, Amy Axel, and Mi Ta of Oncology at Janssen Research & Development, LLC, for providing additional experimental support. Funding: the analysis was funded by Janssen Global Services, LLC; editorial and medical writing support were provided by Tara Abraham, PhD, and Grace Wang, PharmD, of MedErgy, and were funded by Janssen Global Services, LLC.

References 1. van de Donk NWCJ, Richardson PG, Malavasi F. CD38 antibodies in multiple myeloma: back to the future. Blood. 2018;131(1):1329. 2. Krejcik J, Casneuf T, Nijhof IS, et al. Daratumumab depletes CD38+ immune-regulatory cells, promotes T-cell expansion, and skews T-cell repertoire in multiple myeloma. Blood. 2016;128(3):384-394. 3. Konen JM, Fradette JJ, Gibbons DL. The good, the bad and the unknown of CD38 in the metabolic microenvironment and immune cell functionality of solid tumors. Cells. 2020;9(1):52. 4. de Weers M, Tai YT, van der Veer MS, et al. Daratumumab, a novel therapeutic human CD38 monoclonal antibody, induces killing of multiple myeloma and other hematological tumors. J Immunol. 2011;186(3):1840-1848. 5. Lammerts van Bueren J, Jakobs D, Kaldenhoven N, et al. Direct in vitro comparison of daratumumab with surrogate analogs of CD38 antibodies MOR03087, SAR650984 and Ab79. Blood. 2014; 124(21):3474. 6. Overdijk MB, Verploegen S, Bogels M, et al. Antibody-mediated phagocytosis contributes to the anti-tumor activity of the therapeutic antibody daratumumab in lymphoma and multiple myelo-

2007

Letters to the Editor

ma. MAbs. 2015;7(2):311-321. 7. Overdijk MB, Jansen JH, Nederend M, et al. The therapeutic CD38 monoclonal antibody daratumumab induces programmed cell death via Fcγ receptor-mediated cross-linking. J Immunol. 2016; 197(3):807-813. rd 8. Adams HC 3 , Stevenaert F, Krejcik J, et al. High-parameter mass cytometry evaluation of relapsed/refractory multiple myeloma patients treated with daratumumab demonstrates immune modulation as a novel mechanism of action. Cytometry A. 2019;95(3):279-289. 9. Jiang H, Acharya C, An G, et al. SAR650984 directly induces multiple myeloma cell death via lysosomal-associated and apoptotic pathways, which is further enhanced by pomalidomide. Leukemia. 2016;30(2):399-408. 10. Moreno L, Perez C, Zabaleta A, et al. The mechanism of action of the anti-CD38 monoclonal antibody isatuximab in multiple myeloma. Clin Cancer Res. 2019;25(10):3176-3187. 11. Rojkjaer L, Boxhammer R, Endell J, Winderlich M, Samuelsson C, MORPHOSYS AG. (2011) WO2012/041800. Anti-CD38 antibody and lenalidomide or bortezomib for the treatment of mul-

2008

tiple myeloma and NHL. Available from: https:// patentscope.wipo.int/search/en/detail.jsf?docId=WO2012041800 &recNum=5&office=&queryString=WO2012%2F041800+%28 MOR03087%29&prevFilter=&sortOption=Pub+Date+Desc&ma xRec=5. Accessed in June, 2020. 12. Elias KA, Landes G, Singh S, et al. (2012). WO2012/092612. AntiCD38 antibodies. Available from: https://patentscope.wipo.int/ search/en/detail.jsf?docId=WO2012092612&recNum=113&doc An=US2011068235&queryString=%20&. Accessed in June, 2020. 13. Nijhof IS, Casneuf T, van Velzen JF, et al. CD38 expression and complement inhibitors affect response and resistance to daratumumab therapy in myeloma. Blood. 2016;128(7):959-970. 14. Krejcik J, Frerichs KA, Nijhof IS, et al. Monocytes and granulocytes reduce CD38 expression levels on myeloma cells in patients treated with daratumumab. Clin Cancer Res. 2017;23(24):74987511. 15. Malavasi F, Faini AC. Mechanism of action of a new anti-CD38 antibody: enhancing myeloma immunotherapy. Clin Cancer Res. 2019; 25(10):2946-2948.

haematologica | 2021; 106(7)

Letters to the Editor

Immunophenotypic changes of leukemic blasts in children with relapsed/refractory B-cell precursor acute lymphoblastic leukemia who have been treated with blinatumomab Targeting the B-lineage surface antigen CD19 in B-cell precursor acute lymphoblastic leukemia (BCP-ALL) is one of the most successful examples of T-cell-based immunotherapies.1,2 The CD3/CD19 bispecific T-cell engager, blinatumomab, produces excellent responses in adult and pediatric patients with relapsed/refractory BCP-ALL.1-3 However, a significant proportion of patients do not respond to therapy or relapse.1,3 Acting as a strong selective factor, CD19-directed immunotherapy can drive the specific, but still not completely understood immune escape mechanism by the loss of CD19 expression on leukemic blasts, thereby leading to CD19-negative relapses.3,4 This loss of CD19 creates significant challenges to the application of multicolor flow cytometry (MFC) for monitoring minimal residual disease (MRD). Moreover, the expression of other markers can also change, and the frequency of these changes is still unclear. As accurate detection of residual tumor cells has emerged as a key tool in evaluating efficacy and predicting failures after CD19-directed therapies,5 these obstacles to flow cytometry are highly significant. The current report briefly summarizes our data on MFC-MRD and

relapse detection in patients treated with blinatumomab with emphasis on changes in the expression of markers that are relevant for MFC-MRD investigations. We carried out a retrospective review of 90 pediatric patients with relapsed/refractory ВCP-ALL who received blinatumomab between December 2015 and August 2020. The characteristics of the patients, including their cytogenetic data, are presented in Online Supplementary Table S1. All patients, except for two in whom treatment was interrupted because of disease progression, received at least one 28-day course of blinatumomab (median number of courses 1; range, 1-4). Blinatumomab was kindly provided by Amgen as part of a named-patient extended-access program. Blinatumomab treatment was administered to 67 of the 90 studied patients as a “bridge therapy” to allogeneic stem cell transplantation (Online Supplementary Table S1). Morphological examination and MFC-MRD detection in bone marrow aspirates were performed before and after each cycle of blinatumomab, while, for patients who underwent hematopoietic stem cell transplantation, MFC-MRD evaluation was carried out on days +30, +90, +120, +180, +360 after transplantation or in cases of suspected relapse. All patients underwent routine diagnostic immunophenotyping and MRD detection by eight- or ten-color flow cytometry according to the standard protocols of the Moscow-Berlin group.6,7 During the study period MFC was performed on FACS Canto II, FACS Celesta (both

Figure 1. Outcome after blinatumomab treatment in the studied patients (n=90) with emphasis on the expression of CD19. Patients who achieved complete multicolor flow cytometry (MFC)-minimal residual disease (MRD)-negative remission and those with bone marrow (BM) MFC-MRD-negativity, but with progression of extramedullary disease are grouped together and named “Lack of leukemic cells in BM”. Remaining cases (n=48) include resistant ones (n=8), relapses (21 CD19-positive, 6 CD19-negative and 3 switches to acute myeloid leukemia) and patients with blasts detected by MFC at MRD-level in bone marrow at least once (n=10).

haematologica | 2021; 106(7)

2009

Letters to the Editor

from Becton Dickinson, BD; USA), CytoFlex and Navios (both from Beckman Coulter, BC; USA) flow cytometers. EuroFlow guidelines for machine performance monitoring were used.8 The immunophenotype of the tumor cells was analyzed with focus on markers applicable for MFCMRD investigation.4,9 Online Supplementary Table S2 provides a list of monoclonal antibodies used for MFC-MRD monitoring. CD22 and CD24 were additionally studied, mainly after the blinatumomab courses.10 Expression of surface antigens was deemed positive if the antigen was expressed on more than 20% of tumor cells.6 An increase or decrease of expression of each single antigen was defined as a change of the percentage of positive cells by more than 25%. Proportions of cases with stable and changed expression of each single antigen between CD19-negative and CD19-positive relapses were compared using the Fisher exact test.

The treatment outcomes of the studied patients are summarized in Figure 1. Thirty-nine patients achieved complete MFC-MRD-negative remission and three achieved bone marrow MFC-MRD-negativity, but with progression of extramedullary disease. These patients never had detectable leukemia in the bone marrow during follow-up, so they were excluded from analysis of immunophenotypic changes. Overall, modulation of antigen expression was studied in 48 patients with tumor blasts detectable in bone marrow at least once after a course of treatment with blinatumomab. We focused separately on the status of CD19 expression on leukemic cells (Figure 1), since this is the sole possible immunophenotypic change directly linked to the administration of blinatumomab. Thirty patients experienced relapse (>5% of blasts cells by MFC). In 21 cases, leukemic cells at relapse were CD19-positive and in six

Figure 2. Frequency of changes in immunophenotype of leukemic blasts. (A-C) Frequency of changes in immunophenotype of leukemic blasts in relapsed cases (A), cases with detectable blasts in bone marrow only at a minimal residual disease level (B), and in resistant cases (C). Cases of “lineage switch” are not shown. *Differences in frequencies of antigen expression changes are statistically significant (P<0.05). MRD: minimal residual disease; MFC: multicolor flow cytometry.

2010

haematologica | 2021; 106(7)

Letters to the Editor

cases they were CD19-negative. Three children (2 with KMT2A gene rearrangements and 1 with germline KMT2A) developed relapses through “lineage switch” to CD19-negative acute myeloid leukemia (n=1), mixedphenotype acute leukemia (n=1) and acute unclassifiable leukemia (n=1). Eight patients were resistant to blinatumomab therapy, and the expression of CD19 on blast cells was retained in the bone marrow of seven of them while one patient had CD19-negative leukemic blasts after immunotherapy. Blast cells of patients who had leukemic cells in bone marrow only at the level of MFCMRD (<5% by MFC) (n=10), were either CD19-positive (n=7) or CD19-negative (n=3). The immunophenotypic changes of leukemic cells in 27 relapsed patients (3 cases of lineage-switch excluded) are presented in Figure 2A. For all antigens applicable for MFC-MRD assessment, except CD58, changes in expression, either up- or down-modulation, were demonstrated in substantial proportions of cases. We found different frequencies of changes in the expression of two markers among CD19-negative and CD19-positive relapses: CD45 and CD38 were less stable on CD19-negative blasts (Figure 2A). Expression of CD22 and CD24, which are suggested as candidates to replace CD19,10 was studied at relapse in 24 and 19 patients, respectively. Total positivity (≥90%) for these antigens was found in 20 and 17 cases, respectively. We analyzed ten patients, who did not relapse, but had leukemic cells at a MRD level by MFC in bone marrow at least once during the follow-up period (Online Supplementary Table S2). Besides the understandable downmodulation of CD19 in three out of ten patients, expression of CD10, CD20 and CD34 had changed in five, four and three cases, respectively (Figure 2B). Leukemic cells in eight resistant patients had rather

stable immunophenotypic profiles with only very rare changes in antigen expression (Figure 2C). Since cytometric residual leukemia detection is based on investigation of the B-cell compartment, CD19 is a vital antigen for conventional MFC-MRD monitoring.9 Possible loss of this marker during CD19-negative relapse could break the well-established algorithm of MFC-MRD gating.10 If modulation in the expression of other antigens also occurs, cytometric MRD studies could become very tricky. The implication is that assessment of immunophenotypic changes is crucial for improving antibody panels and gating algorithms in patients who undergo CD19directed treatment. Overall, leukemic blasts were CD19-negative in 27.1% of patients with relapses, progression or MFC-MRD-positivity after blinatumomab. It was shown previously that CD19 is completely lost in 10-25% of relapses.3,4 In our series lack of the targeted antigen was noted in nine out of a total of 30 relapses (30.0%) including three “switches” to acute myeloid leukemia. Contrary to data published by Jabbour et al.,11 not all patients resistant to blinatumomab preserved high CD19 expression: in one of eight cases (12.5%) leukemic cells became completely CD19negative. Although CD19-positive relapses were mainly preceded by CD19-positive MFC-MRD and CD19-negative relapses were mainly preceded by CD19-negative MFC-MRD, we observed some exceptions, demonstrating that this is not a strict pattern (Online Supplementary Figure S1). Studying the expression of other antigens commonly used for MFC-MRD evaluation (CD10, CD20, CD34, CD45, CD58, CD38), we also observed frequent changes both in the percentage of positive cells and the distribution of the level of positivity. Phenotypic shifts between

Figure 3. Possible algorithm for searching for residual leukemic cells in patients with B-cell precursor acute lymphoblastic leukemia after CD19-targeted therapy. Various ways of B-lineage gating are shown with their limitations indicated. Consecutive investigation of all these B-cell compartments together with the CD45-defined blast region, precursor regions and aberrantly expressed antigens can help to overcome changes in expression of CD19 and other antigens, applicable for minimal residual disease monitoring.

haematologica | 2021; 106(7)

2011

Letters to the Editor

diagnostic and relapse samples have been reported with a frequency of up to 70% in patients with BCP-ALL.12 Moreover, corticosteroid-mediated changes of antigen expression profile have been observed during remission induction.13 In our series, CD58 demonstrated outstanding stability: no cases of reduced expression of this antigen were noted. All remaining markers, usually useful for MFC-MRD detection, underwent either increased or decreased expression in substantial proportions of relapses and MRD-positive patients. Nevertheless, we were not able to point to any trend in immunological changes. Frequent loss of CD19 expression under selective pressure of CD19-targeted therapy leads to weakness of the application of CD19 as the main gating antigen in searches for neoplastic B-cell precursors. As suggested by Cherian et al., CD22 and CD24 could be added to aid in monitoring BCP-ALL if CD19-negativity develops.10 However, both of these markers could be negative on leukemic cells particularly when KMT2A gene rearrangement occurs (12% of cases in current study).14 We have found total positivity for these antigens in the majority but not in all patients who developed relapse after blinatumomab treatment. Other antigens could also be used (Figure 3) for primary gating,4 although their application might be based on initially detected expression. Our data show that not only CD19 could be downmodulated under the pressure of blinatumomab. Expression of almost all other markers that are useful for MFC-MRD monitoring in BCP-ALL could be changed between ALL diagnosis, MRD and relapse. This suggests that MFCMRD monitoring after CD19 targeting should be based on a sophisticated approach with combinations of multiple markers and flexible gating strategies (Figure 3) in order to minimize the possibility of false negative results. In fact, more than a half of patients with disease progression or reappearance preserved CD19 expression, thus it has no sense to exclude this conventional antigen from tumor-cell gating. However, if residual leukemia is not found among CD19-positive cells, other B-cell compartments should be studied with consideration of the blast immunophenotype detected before CD19 targeting (Figure 3).4,10,15 Moreover, taking into account possible myeloid switching under the selective pressure of blinatumomab therapy, the distribution of cells according to CD45 expression and light scatter should also be investigated. Thus, large and relatively individualized panels of antibodies with additional B-lineage and aberrant markers (myeloid antigens, NG2, etc.) should be applied to increase the effectiveness of MFC-MRD detection in BCP-ALL patients after CD19-directed treatment. Ekaterina Mikhailova, Evgeny Gluhanyuk, Olga Illarionova, Elena Zerkalenkova, Svetlana Kashpor, Natalia Miakova, Yulia Diakonova, Yulia Olshanskaya, Larisa Shelikhova, Galina Novichkova, Michael Maschan, Alexey Maschan and Alexander Popov Dmitry Rogachev National Medical Research Center of Pediatric Hematology, Oncology and Immunology, Moscow, Russian Federation Correspondence: ALEXANDER POPOV - uralcytometry@gmail.com doi:10.3324/haematol.2019.241596

2012

Received: October 28, 2019. Accepted: December 22, 2020. Pre-published: December 30, 2020. Disclosures: no conflicts of interest to disclose. Contributions: EM and AP designed the study, collected cytometric data and wrote the paper. EG, NM and YuD collected clinical data. OI and SK collected cytometric data. EZ and YuO collected cytogenetic and molecular genetic data and wrote the paper. LSh collected clinical data and wrote the paper. GN, MM and AM designed the study and wrote the paper. All authors revised the final version of the manuscript Funding: the KMT2A rearrangement assessment study was supported by RFBR grant n. 17-29-06052 and Presidential grant n. MK-1645.2020.7 (n. 075-15-2020-338)

References 1. Kantarjian H, Stein A, Gokbuget N, et al. Blinatumomab versus

chemotherapy for advanced acute lymphoblastic leukemia. N Engl J Med. 2017;376(9):836-847. 2. von Stackelberg A, Locatelli F, Zugmaier G, et al. Phase I/phase II study of blinatumomab in pediatric patients with relapsed/refractory acute lymphoblastic leukemia. J Clin Oncol. 2016;34(36):4381-4389. 3. Topp MS, Gokbuget N, Zugmaier G, et al. Phase II trial of the antiCD19 bispecific T cell-engager blinatumomab shows hematologic and molecular remissions in patients with relapsed or refractory Bprecursor acute lymphoblastic leukemia. J Clin Oncol. 2014;32(36):4134-4140. 4. Mejstrikova E, Hrusak O, Borowitz MJ, et al. CD19-negative relapse of pediatric B-cell precursor acute lymphoblastic leukemia following blinatumomab treatment. Blood Cancer J. 2017;7(12):659. 5. Jabbour E, Short NJ, Jorgensen JL, et al. Differential impact of minimal residual disease negativity according to the salvage status in patients with relapsed/refractory B-cell acute lymphoblastic leukemia. Cancer. 2017;123(2):294-302. 6. Novikova I, Verzhbitskaya T, Movchan L, et al. Russian-Belarusian multicenter group standard guidelines for childhood acute lymphoblastic leukemia flow cytometric diagnostics. Oncohematology. 2018;13(1):73-82. 7. Popov A, Belevtsev M, Boyakova E, et al. Standardization of flow cytometric minimal residual disease monitoring in children with Bcell precursor acute lymphoblastic leukemia. Russia–Belarus multicenter group experience. Oncohematology. 2016;11(4):64-73. 8. Kalina T, Flores-Montero J, Lecrevisse Q, et al. Quality assessment program for EuroFlow protocols: summary results of four-year (2010-2013) quality assurance rounds. Cytometry A. 2015;87(2):145156. 9. Karawajew L, Dworzak M, Ratei R, et al. Minimal residual disease analysis by eight-color flow cytometry in relapsed childhood acute lymphoblastic leukemia. Haematologica. 2015;100(7):935-944. 10. Cherian S, Miller V, McCullouch V, et al. A novel flow cytometric assay for detection of residual disease in patients with B-lymphoblastic leukemia/lymphoma post anti-CD19 therapy. Cytometry B Clin Cytom. 2018;94(1):112-120. 11. Jabbour E, Dull J, Yilmaz M, et al. Outcome of patients with relapsed/refractory acute lymphoblastic leukemia after blinatumomab failure: no change in the level of CD19 expression. Am J Hematol. 2018;93(3):371-374. 12. Borowitz MJ, Pullen DJ, Winick N, et al. Comparison of diagnostic and relapse flow cytometry phenotypes in childhood acute lymphoblastic leukemia: implications for residual disease detection: a report from the children's oncology group. Cytometry B Clin Cytom. 2005;68(1):18-24. 13. Dworzak MN, Gaipa G, Schumich A, et al. Modulation of antigen expression in B-cell precursor acute lymphoblastic leukemia during induction therapy is partly transient: evidence for a drug-induced regulatory phenomenon. Results of the AIEOP-BFM-ALL-FLOW-MRDStudy Group. Cytometry B Clin Cytom. 2010;78(3):147-153. 14. De Zen L, Bicciato S, te Kronnie G, Basso G. Computational analysis of flow-cytometry antigen expression profiles in childhood acute lymphoblastic leukemia: an MLL/AF4 identification. Leukemia. 2003;17(8):1557-1565. 15. Cherian S, Stetler-Stevenson M. Flow cytometric monitoring for residual disease in B lymphoblastic leukemia post T cell engaging targeted therapies. Curr Protoc Cytom. 2018;86(1):e44.

haematologica | 2021; 106(7)

Letters to the Editor

Whole genome CRISPR screening identifies TOP2B as a potential target for IMiD sensitization in multiple myeloma Thalidomide analogues (IMiD), such as lenalidomide (LEN) and pomalidomide (POM) have significantly improved survival in patients with multiple myeloma (MM).1 However, many patients relapse despite continued IMiD exposure, and IMiD-resistance remains a significant clinical problem. IMiD engage Cereblon (CRBN), an adaptor for the CUL4A-DDB1-RBX1 E3 ligase complex to promote proteasome-dependent degradation of neosubstrates IKZF1 (Ikaros) and IKZF3 (Aiolos). The degradation of these transcription factors is both directly toxic to MM cells and immunostimulatory to T cells.2–4 Interrogation of the molecular events driving IMiD-mediated engagement of CRBN neosubstrates has greatly improved our mechanistic understanding of these agents. Acquired or intrinsic IMiD-resistance may occur through several mechanisms including loss of expression of CRBN and/or associated E3 ligase factors.5,6 However, deeper insight into IMiD-resistance mechanisms may inform novel therapeutic approaches. Here, genome-wide CRISPR-Cas9 screening was employed to characterize resistance mechanisms and resensitization factors in isogenic IMiD-sensitive and -resistant MM lines. Loss of DNA topoisomerase IIβ (TOP2B) resensitized IMiDrefractory cells to IMiD and its inhibition with the cardio-

protective drug, dexrazoxane (DXZ), potentiated IMiD activity. Collectively, these findings identify TOP2B as a potential new therapeutic target in MM. LEN-resistant MM1.S (MM.1Sres) cells were previously derived by culturing MM.1S cells in presence of increasing doses of LEN.7 MM.1Sres cells displayed no significant increase in cell death upon prolonged (7 days) LEN exposure, while isogenic MM.1S cells were sensitive to LEN-induced cell death (Figure 1A). MM.1Sres cells also exhibited resistance to the anti-proliferative effects of LEN and POM as demonstrated by CellTrace Violet (CTV) labeling (Figure 1B). As previously reported,7 MM.1Sres cells showed reduced CRBN expression and reduced IKZF3 degradation upon IMiD treatment as compared to MM.1S (Figure 1C). In order to determine genes and pathways required for IMiD anti-myeloma activity in MM.1S cells, a genomescale CRISPR knockout screen was performed in MM.1SCas9 cells treated with LEN, POM or dimethyl sulfoxide (DMSO) (Figure 1D). Genetic dependencies of MM.1SCas9 cells were identified by loss-of-representation of short guide RNA (sgRNA) in (DMSO)-treated cells and identified genes such as MYC, IRF4, IKZF1 and IKZF3 (Online Supplementary Figure S1A). MM.1S-Cas9 cells were dependent on essential processes such as RNA metabolism and DNA-damage related pathways (Online Supplementary Figure S1B and C). Deletion of CRBN and members of the COP9 signalosome (CSN), a 9-protein

Figure 1. Resistance to lenalidomide and pomalidomide is mediated by loss of Cereblon and the COP9 signalosome. (A) Bar plot representing the percentage of propidium iodide negative (PI-, viable) MM.1S and lenlidomide (LEN)-resistant MM1.S (MM.1Sres) MM.1Sres cells treated with dimethyl sulfoxide (DMSO) or LEN (2 mM) for 7 days. Results are aggregated from n=3 independent experiments with two technical replicates per experiment. Error bars represents the mean ± standard error of the mean of the three biological replicates with their respective technical replicates; *P<0.0001. (B) CellTrace™ Violet/propidium iodide proliferation assay comparing proliferation of MM.1S and MM.1Sres cells in presence of LEN (2 mM) or pomalidomide (POM) (500 nM) (day 7 timepoint) in comparison to vehicle (DMSO). Results are representative of three independent experiments. (C) Immunoblot of IKZF3 and Cereblon (CRBN) levels in DMSO-, LEN(2 mM) or POM- (500 nM) treated MM.1S and MM.1Sres cells for 16 hours (blots are representative of n=3 independent experiments). (D) Schematic representation of the CRISPR genome-scale resistance screen workflow. Approximately 500x106 MM.1Sres-Cas9 cells were transduced with the Brunello genome-wide library, selected with puromycin and then treated with DMSO, LEN (2 µM) or POM (500 nM) for approximately 8 weeks. Genomic DNA was extracted for library preparation and Illumina sequencing. (E) Scatter plot showing hits overlapping between LEN and POM that are significant for adjusted P<0.05 and hits unique to LEN-treated cells significant for adjusted P<0.05.

haematologica | 2021; 106(7)

2013

Letters to the Editor

complex involved in protein turnover regulation, were the most significantly enriched sgRNA in the presence of continued IMiD exposure (Figure 1E; Online Supplementary Figure S1D to E), validating previously published data.5,6 Gene ontology (GO) analysis and protein-

protein interaction (PPI) networking revealed that CSN may also modulate CUL4-DDB1 functions in response to DNA damage (Online Supplementary Figure S1F and G). Notably, some sgRNA (NCOR1, EDC4, SCAP, UBE2G1, MBTPS1/2) conferred selective resistance to LEN but not

Figure 2. TOP2B deletion synergizes with Thalidomide analogues (IMiD) in MM.1S and MM.1S resistant cells. (A) Schematic representation of the CRISPR genome-scale dropout screen workflow. Approximately 500x106 MM.1Sres-Cas9 cells were transduced with the Brunello genome-wide library, selected with puromycin and then treated with dimethyl sulfoxide (DMSO), lenalidomide (LEN) (2 mM) or pomalidomide (POM) (500 nM) for 3 weeks. Genomic DNA (gDNA) was extracted for library prepraration and Illumina sequencing. (B) Scatter plot overlapping the -log (P-value) of negatively enriched guides in MM.1Sres-Cas9 cells treated with LEN and POM at the end of the screen. (C) Normalized counts (in log ) at the end of the screen (T-End) of short guide RNA (sgRNA) specific for TOP2B in MM.1Sres. In the DMSO condition, the counts in T-End are compared to a timepoint zero (T-0) reference. In the LEN and POM conditions, the counts for sgTOP2B in T-End are plotted relatively to the average counts in the DMSO T-End condition. (D) Schematic representation of the competitive proliferation assay workflow. MM.1S- or MM.1Sres-Cas9 scramble-BFP cells were grown in competition with MM.1Sres-Cas9 sgTOP2B-GFP at a 1:1 ratio in DMSO, LEN (2 mM) or POM (500 nM) to validate the data observed in the genome-wide dropout screen. (E) Representative experiment of n=3 independent replicates showing the change in percentage of MM.1Sres-Cas9 scramble- or sgTOP2B-GFP+ cells grown in competition with MM.1Sres-Cas9 Scramble-BFP+ cells treated with DMSO, LEN (2 mM) or POM (500 nM) for 21 days. Each dot with relative error bars represent the mean ± standard deviation of values from two technical replicates. (F) Representative experiment of three independent replicates showing the change in percentage of MM.1S-Cas9 scramble- or sgTOP2B-GFP+ cells grown in competition with MM.1S-Cas9 scramble-BFP+ cells (at a 1:1 ratio) treated with DMSO, LEN (2 mM) or POM (500 nM) for 10 days. Each dot with relative error bars represent the mean ± standard deviation of values from two technical replicates. GFP: green fluorescent protein. 10

2014

haematologica | 2021; 106(7)

Letters to the Editor

Figure 3. TOP2 inhibitor dexrazoxane displays anti-myeloma properties and combinatorial activity with lenalidomide. (A) Immunoblot showing TOP2B expression in MM.1S cells treated with dimethyl sulfoxide (DMSO) or dexrazoxane (DXZ) (20 mM or 50 mM) across 24, 48 and 72 hours. Tubulin expression is provided as a loading control. The experiment is representative of three biological replicates. (B) Bar graphs demonstrating Nicoletti cell cycle profiling of MM.1S and MM.1S resistant (MM1Sres) cells exposed to DMSO, lenalidomide (LEN) (2 mM), DXZ (20 mM) or combined LEN/DXZ (combo). The plot represents an aggregate of three independent experiments with three technical replicates each. Error bars represent mean ± standard error of the mean of n=3 independent experiments. v: significant vs. vehicle (DMSO); l: significant vs. LEN; d: significant vs. DXZ (P<0.05 or less). (C) Heatmaps representing the variation of zero interaction potency (ZIP) synergy score in MM.1S and MM.1Sres cells treated with increasing concentration of LEN and DXZ in combination. The average percentage of PI negative cells (viable) in each drug combination of three independent experiments was employed to compute synergy. The R package synergyfinder v2.2.4 was employed to perform the analysis. (D) Average viable cell count at 3 and 7 days for MM.1S and MM.1Sres treated with DMSO, LEN (2 mM), DXZ (5 mM or 20 mM depending on the cell line) or combined LEN/DXZ (combo). (E) Immunoblot showing expression of IKZF1, IKZF3 after 24h of treatment with DMSO, LEN (2 mM), DXZ (20 µM) or combined treatment and expression of IRF4 and MYC after 72 hours of treatment in the same conditions. HSP90 is provided as loading control. The experiment is representative of three biological replicates. (F) Viable cell count at 3 and 7 days of OPM2, RPMI-8226 and JJN3 cells treated with DMSO, LEN (2 mM), DXZ (20 mM) or combo. (G) Percentage of PI negative (viable) OPM2 cells treated with DMSO, LEN (2 mM), DXZ (20 mM) or combo and assessed by flow cytometry. (H) Percentage of PI negative (viable) RPMI-8226 cells treated with DMSO, LEN (2 mM), DXZ (5 mM) or combo and assessed by flow cytometry. (I) Percentage of PI negative (viable) JJN3 cells treated with DMSO, LEN (2 mM), DXZ (20 mM) or combo and assessed by flow cytometry. In (D), (F), (G), (H) and (I), error bars represent mean ± standard error of the mean of n=3 independent experiments. v: significant vs. vehicle (DMSO); l: significant vs. LEN; d: significant vs. DXZ (P<0.05 or less).

haematologica | 2021; 106(7)

2015

Letters to the Editor

POM (Figure 1E), reflecting either a difference of potency and/or discrepant substrate specificity between the two IMiD. The identification of genes that when deleted restore IMiD-sensitivity in MM.1Sres cells was achieved by a loss-of-representation CRISPR screen in 21-day LEN and POM-treated MM.1Sres-Cas9 cells relative to DMSO control (Figure 2A). The dependencies of DMSO-treated MM.1Sres-Cas9 cells partially overlapped and correlated with those of MM.1S-Cas9 cells, suggesting that acquired IMiD-resistance did not globally alter gene dependencies in these cells (Online Supplementary Figure S2A to F). In order to identify selective resentitization mechanisms in the presence of IMiD, the DMSO endpoint was compared with the matched timepoint in IMiD-treated MM.1Sres-Cas9 cells. This revealed that deletion of ATXN7, TOP2B, MIER3, YPEL5, MAEA and MED13L sensitized MM.1Sres cells to both LEN and POM (Figure 2B and C). ATXN7 is part of the deubiquitination module of STAGA, a multisubunit entity involved in transcriptional regulation and DNA repair.8 MED13L and MIER3 also modulate transcription.9,10 Interestingly, YPEL5 and MAEA co-operate in an E3 ligase complex targeting gluconeogenesis enzymes.11 However, TOP2B was selected for further study due to its potential tractability as a drug target. TOP2B is an enzyme which resolves topological DNA constraints during replication, transcription and repair.12 In order to validate that TOB2B loss resensitizes to IMiD, competitive proliferation assays were performed in the presence and absence of LEN and POM (Figure 2D). MM.1Sres-Cas9 cells expressing two independent TOP2B sgRNA with a GFP reporter were mixed at a 1:1 ratio with MM.1Sres-Cas9 cells expressing a non-targeting sgRNA with a BFP reporter (Figure 2D to E; Online Supplementary Figure S2G). Consistent with the CRISPR-screen results, a competitive loss of GFP+ relative to BFP+ cells was observed following IMiD treatment (Figure 2E) confirming that TOP2B deletion confers IMiD-sensitivity in MM.1Sres-Cas9 cells. Analogous assays in MM.1S-Cas9 cells revealed that TOP2B loss further sensitized these cells to the anti-tumour effects of IMiD (Figure 2F), indicating that TOP2B deletion enhances IMiD activity in both IMiD-naïve and resistant contexts. TOP2B deletion had little or no effect on CRBN expression or subsequent IKZF3 degradation following IMiD treatment (Online Supplementary Figure S3A and B). MYC levels appeared to be modestly more downregulated in IMiD-treated MM.1Sres-Cas9 compared to MM.1S-Cas9 cells while the IMiD-induced downregulation of IRF4 was similar in MM.1Sres-Cas9 compared to MM.1S-Cas9 cells ((Online Supplementary Figure S3C). These observations indicate that the resensitization to IMiD treatment following deletion of TOP2B likely does not depend on further effects on the IKZF1/3-IRF4-MYC axis. Having discovered the IMiD-sensitizing effects of genetic TOP2B depletion in MM.1Sres-Cas9 cells, orthogonal assays using DXZ, a chemical inhibitor of TOP2 that induces selective degradation of TOP2B protein, were used to validate this observation. DXZ also posseses iron-chelating activity and is Food and Drug Administration-approved for prevention of anthracycline-induced cardiotoxicity.13 In order to investigate whether DXZ-mediated TOP2B degradation would phenocopy genetic deletion, MM.1Sres and MM.1S cells were treated with DXZ alone or in combination with LEN. TOP2B degradation in DXZ-treated MM.1S cells was evident 24 hours (hrs) after drug exposure and remained low in treated cells for up to 72 hrs (Figure 3A) at clinically relevant concentrations.14 Cell cycle analysis 2016

demonstrated that concentrations of DXZ sufficient to induce TOP2B degradation were cytostatic in MM.1S and MM.1Sres cells, with accumulation of cells in SubG1 and >2N, suggesting apoptosis-induction and failure of cytokinesis (Figure 3B; Online Supplementary S3D to F). Combinatorial effects of LEN and DXZ were observed in MM.1S and MM.1Sres cells with DXZ alone inducing cytostasis in MM.1Sres cells (Figure 3C and D). Similar to the effects observed following TOP2B deletion, DXZ did not modulate CRBN expression or alter IKZF3 degradation (Online Supplementary Figure S3G). However, an effect of the LEN and DXZ combination on cMYC, IKZF1, IKZF3 and IRF4 expression (Figure 3E) was evident. These findings may indicate that the effects of LEN and DXZ converge on the IRF4-MYC axis in MM.1Sres cells downstream or in parallel to canonical CRBN-neosubstrate interactions. Subsequently, DXZ combination treatments were performed in IMiD-sensitive OPM2 cells and IMiD-resistant RPMI-8226 and JJN3 MM cell lines (Figure 3D). In OPM2 cells, DXZ and LEN alone induced robust growth inhibition with a combinatorial effect observed following treatment with both agents (Figure 3F and G). DXZ induced death of RPMI-8226 and JJN3 cells, with mild additivity in the presence of LEN (Figure 3F, H and I). CRISPR-based dissection of the genetic dependencies of MM.1S cells provided additional insight into IMiD biology and acquired IMiD resistance. Consistent with the initial description of MM.1Sres,7 downregulation of CRBN expression and attenuation of neosubstrate degradation appears to be the major mechanism of IMiD resistance in these cells. Synthetic generation of IMiD resistance using gene deletion in MM.1S cells recapitulated prior studies identifying CRBN and elements of the CSN.5,6 Gene ontology analysis of these hits revealed their importance in transcription-coupled nucleotide excision repair (TC-NER). Moreover, CUL4 and DDB1 have been demonstrated to participate with CSN in DNA repair pathways such as NER and TC-NER.15 IMiD-sensitivity in MM.1Sres cells was rescued by knockout of TOP2B, a gene that modulates DNA repair, chromatin stability and gene expression.12 However, LEN did not induce a DNA damage response or synergize with etoposide in MM.1Sres cells, suggesting that DNA damage induction is not the primary re-sensitization mechanism (Online Supplemnentary Figure S3H and I). Genetic deletion of TOP2B was not lethal to MM.1Sres cells, however these cells were sensitized to IMiD-induced death. This phenotype seemed to be independent of an increase in CRBN activity or expression changes within the IKZF1/3IRF4-MYC axis. The biology underpinning re-sensitization of MM.1Sres cells to IMiD through loss of TOP2B remains to be defined. DXZ had demonstrable anti-MM properties and additional activity in combination with LEN, especially in IMiD-sensitive cells. Since TOP2B deletion did not induce growth inhibition, the single-agent activity of DXZ may depend upon TOP2A inhibition rather than TOP2B degradation or through other effects. Deletion or depletion of TOP2A can have deleterious effects on the growth and/or survival of cancer cells (Online Supplemnentary Figure S3J) but we do not have evidence clearly demonstrating that the anti-MM activity of DXZ is through effects on TOP2A. Given that DXZ did not appear to demonstrably impinge on the IKZF1/3-IRF4MYC axis, exactly how DXZ confers anti-MM activity, either alone or in combination with IMiD, remains unknown. The IC of DXZ across the MM lines tested spanned from 5 to 20 mM (data not shown), which is significantly lower than the peak plasma concentration 50

haematologica | 2021; 106(7)

Letters to the Editor

reached after a cardioprotective 500 mg/m2 dose (36.5 mg/mL or 135 mM).14 This suggests that DXZ could be repurposed as a TOP2-targeting anti-MM agent as part of a combinatorial approach, however its posology is not well suited to recurrent or chronic administration. We are unaware of any other selective small-molecule TOP2B inhibitors. Greater understanding of the structure-activity relationship between DXZ and TOP2B may allow the rational development of related chemotypes for drug therapy. Further investigation of these mechanisms by which TOP2B inhibition leads to anti-MM activity could reveal alternative pathways to IMiD potentiation. Matteo Costacurta,1,2 Stephin J Vervoort,1,2 Simon J Hogg,1,2 Benjamin P Martin,1,2 Ricky W Johnstone1,2# and Jake Shortt1,2,3,4# 1 Translational Hematology Program, Peter MacCallum Cancer Center, Melbourne; 2Peter MacCallum Department of Oncology, The University of Melbourne, Parkville, Melbourne; 3Monash Hematology, Monash Health, Clayton, Melbourne and 4Blood Cancer Therapeutics Laboratory, School of Clinical Sciences at Monash Health, Monash University, Clayton, Melbourne, Victoria, Australia. # RWJ and JS contrituted equally as co-senior authors. Correspondence: RICKY W. JOHNSTONE - ricky.johnstone@petermac.org JAKE SHORTT - jake.shortt@monash.edu doi:10.3324/haematol.2020.265611 Received: July 3, 2020. Accepted: December 22, 2020. Pre-published: December 30, 2020. Disclosures: JS sits on advisory boards and received speakers fees from Celgene and BMS outside of the published work; The Johnstone laboratory receives funding support from Roche, BMS, Astra Zeneca, and MecRx; RWJ is a paid scientific consultant and shareholder in MecRx; all other authors declare no conflicts of interest. Contributions: MC conducted experimental work, planned experiments, analyzed the data and wrote the manuscript; SV helped with experimental planning, analyzed sequencing data and contributed with manuscript writing; SH analyzed sequencing data; BM helped with screening experiments; RJ and JS supervised the project, planned experiments and wrote the manuscript. Funding: JS is supported by an Australian Medical Research Future Fund Next Generation Clinician Researcher Fellowship; this research was funded by NHMRC project Ggrant 1127387; RWJ was supported by the Cancer Council Victoria, National Health and Medical Research Council of Australia (NHMRC), and The Kids’ Cancer Project; The Victorian Center for Functional Gnomics, the Molecular Genomics Core and the Flow Cytometry Core Facilities at the Peter

haematologica | 2021; 106(7)

MacCallum Cancer Center provided excellent technical support; The Peter MacCallum Foundation and Australian Cancer Research Foundation provide generous support for equipment and core facilities. Acknowledgments: We acknowledge the FACS facility and the Molecular Genomic Core at the Peter MacCallum Cancer Center for contributing to this work.

References 1. Kumar SK, Rajkumar SV, Dispenzieri A, et al. Improved survival in multiple myeloma and the impact of novel therapies. Blood. 2008;111(5):2516-2520. 2. Lu G, Middleton RE, Sun H, et al. The myeloma drug lenalidomide promotes the Cereblon-dependent destruction of Ikaros proteins. Science. 2014;343(6168):305-309. 3. Licht JD, Shortt J, Johnstone R. From anecdote to targeted therapy: the curious case of thalidomide in multiple myeloma. Cancer Cell. 2014;25(1):9-11. 4. Krönke J, Udeshi ND, Narla A, et al. Lenalidomide causes selective degradation of IKZF1 and IKZF3 in multiple myeloma cells. Science. 2014;343(6168):301-305. 5. Liu J, Song T, Zhou W, et al. A genome-scale CRISPR-Cas9 screening in myeloma cells identifies regulators of immunomodulatory drug sensitivity. Leukemia. 2019;33(1):171-180. 6. Sievers QL, Gasser JA, Cowley GS, Fischer ES, Ebert BL. Genomewide screen identifies cullin-RING ligase machinery required for lenalidomide-dependent CRL4CRBN activity. Blood. 2018; 132(12):1293-1303. 7. Zhu YX, Braggio E, Shi C-X, et al. Cereblon expression is required for the antimyeloma activity of lenalidomide and pomalidomide. Blood. 2011;118(18):4771-4779. 8. Ramachandran S, Haddad D, Li C, et al. The SAGA deubiquitination module promotes DNA repair and class switch recombination through ATM and DNAPK-mediated H2AX formation. Cell Rep. 2016;15(7):1554-1565. 9. Poss ZC, Ebmeier CC, Taatjes DJ. The Mediator complex and transcription regulation. Crit Rev Biochem Mol Biol. 2013;48(6):575608. 10. Derwish R, Paterno GD, Gillespie LL. Differential HDAC1 and 2 recruitment by members of the MIER family. PLoS One 2017;12(1):e0169338. 11. Lampert F, Stafa D, Goga A, et al. The multi-subunit GID/CTLH E3 ligase promotes proliferation and targets the transcription factor Hbp1 for degradation. eLife. 2018;7e35528. 12. Chen SH, Chan N-L, Hsieh T. New mechanistic and functional insights into DNA topoisomerases. Biochemistry. 2013;82(1):139170. 13. Lyu YL, Kerrigan JE, Lin C-P, et al. Topoisomerase II mediated DNA double-strand breaks: implications in doxorubicin cardiotoxicity and prevention by dexrazoxane. Cancer Res. 2007;67(18): 8839-8846. 14. Brier ME, Gaylor SK, McGovren JP, Glue P, Fang A, Aronoff GR. Pharmacokinetics of dexrazoxane in subjects with impaired kidney function. J Clin Pharmacol. 2011;51(5):731-738. 15. Fischer ES, Scrima A, Böhm K, et al. The molecular basis of CRL4DDB2/CSA ubiquitin ligase architecture, targeting, and activation. Cell. 2011;147(5):1024-1039.

2017

Letters to the Editor

Myocardial injury and coronary microvascular disease in sickle cell disease Myocardial infarction and microvascular ischemic damage in the heart are one of the least well-described entities in the sickle cell disease (SCD) spectrum.1 Over the last few decades, some autopsy studies and case reports have described myocardial infarction without obstructive coronary artery disease suggesting microvascular ischemic injury in SCD patients.2 Impaired myocardial perfusion reserve in SCD has been demonstrated using different modalities including contrast echocardiography, nuclear myocardial perfusion scans, single-photon emission computerized tomography (SPECT) as well as cardiac magnetic resonance imaging (MRI).3 During a state of physiological stress or acute crisis, it can lead to myocardial injury which can be detected by serum troponin measurement. The role of troponin in microvascular disease in sickle cell disease has not been well defined. troponin-I was elevated (>0.4 ng/mL) in two of 32 patients 24 hours after the onset of acute crisis, with chest pain and electrocardiogram findings of sinus tachycardia and non-specific ST-T wave changes.4 In another study of six patients, troponin-T was normal 24 hours after admission for a sickle cell crisis in all patients.5 Cardiac magnetic resonance (CMR) is a non-invasive diagnostic tool that can be used for myocardial tissue characterization and assessment of coronary microvascular disease (CMD).6 Due to high spatial resolution with first-pass perfusion imaging, it can visualize diffuse subendocardial perfusion abnormality resulting from CMD during the administration of a vasodilating drug such as adenosine.7 Late gadolinium enhancement (LGE) imaging visualizes decreased clearance of gadoliniumbased contrast agents from the extracellular space in areas of damaged myocardium. LGE has become the in

vivo gold standard for visualization of myocardial injury from a variety of causes;8 subendocardial enhancement indicates an ischemic injury, whereas midwall enhancement indicates fibrosis of non-ischemic myocardial disease and epicardial enhancement is consistent with inflammatory damage.9 Our study aims to estimate the prevalence of myocardial injury defined by elevated troponin-I levels. We will also define the prevalence of coronary microvascular disease and other myocardial abnormalities in an SCD cohort clinically referred for cardiac MRI. We conducted a retrospective study of the SCD patients seen at The Ohio State University Wexner Medical Center over a period of 10 years from July 2005 to July 2015. Patients age 18 years or above, with troponin-I level elevation (level >0.11 ng/mL) and/or cardiac MRI were included in the initial cohort. Coronary microvascular disease (CMD) on CMR was defined by the presence of either subendocardial damage by LGE or impaired myocardial perfusion by adenosine stress perfusion imaging. All other abnormalities were categorized as non-CMD due to a lack of specificity. Clinical and laboratory variables closest to the peak troponin elevation and cardiac MRI were recorded, if available within 4 weeks. For all patients with troponin level measurement and cardiac MRI, the date of death confirmed by chart review by June 2019 was recorded. Out of 373 SCD patients, 69 had either troponin level measurement or cardiac MRI, or both done. The median age was 34 years (range, 19-67 years) with 30% of patients over the age of 40 years. Thirty-four (49%) patients were female. Seventy-five percent of the patients were hemoglobin (Hb) SS, and the rest 25% were other genotypes (SC 15%, S-βthal 9%). Median baseline Hb (defined by Hb level at a steady-state within the preceding year) was 8.0 g/dL (range, 4.5-13 g/dL) and median current Hb at the time of the event (either troponin ele-

Figure 1. Overall survival comparison between coronary microvascular disease and non-CMD group. CI: Confidence Interval; Y: CMD present; N: no CMD; NE: not estimable.

2018

haematologica | 2021; 106(7)

Letters to the Editor

vation or cardiac MRI) was 7.6 g/dL (2.9-12.6 g/dL). Only 5 (7.2%) patients had a history of diabetes mellitus and (8.6%) patients had a family history of myocardial infarction. A majority of these patients (n=57, 83%) had a history of acute chest syndrome (ACS) (Table 1). Median initial troponin level was 0.14 ng/dL (range, 0.01-38.09) and peak troponin of 0.44 ng/dL (range, 0.0119.97). Patients with troponin elevation were more likely to have acute chest syndrome (40% vs. 10%; P=0.02) and acute kidney injury (0% vs. 18%; P=0.0002). There was no difference in median Hb in both groups; however, the troponin elevation group tended to have lower platelet count (272 vs. 165 K/uL, P=0.007), higher lactate dehydrogenase (LDH) level (392 vs. 498 U/L, P=0.37) (Table 2). In an expanded cohort of 239 patients with troponin measurement, 42 (18%) had elevated troponin-I at one or multiple instances. Troponin elevation significantly increased the likelihood of death with a hazard ratio of 2.6 (95% Confidence Interval [CI]: 1.4-4.9; P=0.0003). There were 47 patients with cardiac MRI performed over the 10-year period for various indications, most common being chest pain and/or troponin elevation (n=26) and iron overload. CMD was present in 15 patients (32%), only five patients had normal CMR and the rest of them had other non-ischemic findings. There was no statistically significant difference between the CMD and non-CMD groups in terms of baseline characteristics, clinical or laboratory variables (Table 2). Cardiac catheterization showed no epicardial vessel obstruction in eight patients, one patient had triple vessel disease. Overall survival was similar in both groups (P=0.42; Figure 1). Patients were treated with the following medications either alone or in combination: low dose aspirin (n=9), long-acting nitrate (n=5), beta-blockers (n=4), or angiotensin-converting enzyme inhibitors (ACEI) (n=3), clopidogrel (n=1). One patient, a 33-year-old male, presented with ST-elevation MI with a peak troponin-I level of 38.09 ng/dL, cardiac catheterization with clean coronaries and microvascular disease on MRI was started on aspirin, clopidogrel, beta-blockers, and ACEI. Two patients received a simple transfusion, exchange transfusion was recommended in one patient but was deferred later due to clinical improvement. In 22 patients with troponin level measurement within 30 days before cardiac MRI, troponin elevation predicted the presence of CMD with a sensitivity of 87.5% (95% CI: 47-100) and specificity of 57% (95% CI: 29-82). There was no correlation between the degree of troponin elevation and CMD, the difference between median baseline and peak troponin was similar in both groups (0.08 ng/mL, range, 0-25.92 in CMD vs. 0.07 ng/mL, range, 0-1.04 in non-CMD; P=0.31). After restricting the analysis to troponin elevation within 14 days prior, the proportion of CMD was higher in elevated troponin group in patients without ACS (n=14, 100% vs. 43%, P=0.07) or AKI (n=19, 73% vs. 50%, P=0.38). Our study highlights the fact that myocardial injury and coronary microvascular disease is indeed prevalent in SCD. The presence of troponin elevation in 18% of the SCD patients suggests that they suffer some degree of myocardial injury when presenting with chest pain or in an acute crisis. This number differs from a previous report by Aslam et al., in which 6.2% patients had troponin elevation. However, the sample size was smaller and a higher troponin cutoff (0.4 ng/mL vs. 0.11 ng/mL) was used.4 Since troponin-I is well-established to be a very sensitive and specific marker of ongoing myocardial injury,10 lower sensitivity and specificity in our cohort might be explained by the timing of the troponin testing. haematologica | 2021; 106(7)

Table 1. Patient characteristics.

Variable Age (years) Median (range) ≤ 20 ≤40 >40 Sex Male Female Genotype SS SC Sβ-thal History of SCD related complications ACS AVN Stroke Retinopathy Leg ulcer PHT Body Mass Index (kg/m2) Median (range) Coronary artery disease related factors Diabetes mellitus Chronic kidney disease History of aspirin use Family history of myocardial infarction Indications for cardiac MRI Chest pain and/or troponin elevation Iron overload Cardiomyopathy Arrhythmias† Right atrial mass Pulmonary stenosis Right to left shunt Unknown

Number, out of total 69 patients (%) 34 (19-67) 3 (4) 45 (65) 21 (30) 35 (51) 34 (49) 51 (75) 11 (16) 6 (9) 57 (83) 25 (36) 17 (25) 10 (14) 5 (7) 22 (33) 23.9 (12.3-46.2)

5 (7) 14 (20) 19 (28) 6 (9) 26 11 2 3 1 1 1 1

SS: homozygous sickle cell with hemoglobin SS; SC: homozygous sickle cell with hemoglobin SC; Sβ-thal: sickle cell β thalassemia; SCD: sickle cell disease; ACS: acute chest syndrome; AVN: avascular necrosis; PHT: pulmonary hypertension. †included ventricular tachycardia and AV-nodal re-entrant tachycardia.

As it is not a provocative test, troponin-I will not capture chronic ischemic changes, infarct scaring, and impaired myocardial perfusion reserve unless there is acute ischemia leading to myocardial damage at the time of testing. These findings can be elicited by CMR with contrast and stress testing with great precision and accuracy. Patients with CMD will have impaired perfusion reserve at rest and ischemia on stress testing, it may or may not translate into myocardial injury to cause troponin elevation. To our knowledge, this is the first study to assess the effect of myocardial injury and CMD on mortality for SCD patients. Myocardial injury was associated with a 2.6-fold increase in all-cause mortality and there was a statistically insignificant trend towards lower survival in the CMD group. These findings can be a potential explanation of the high frequency of sudden death, especially of cardiac cause, in otherwise healthy patients presenting in acute crisis.11 Myocardium in distress with underlying ischemia can act as an arrhythmogenic substrate leading to fatal arrhythmias.12 There is no randomized controlled data on coronary 2019

Letters to the Editor

Table 2. Association between troponin elevation and other variables, and between coronary microvascular disease on cardiac magnetic resonance imaging and other variables.

Troponin data (N=63)† Normal Elevated Troponin Troponin (n=21) (n=42) Clinical variables, number (%) ACS at time of event No Yes AKI at the time event No Yes EKG changes No Yes missing OME use 24 hour prior to peak troponin elevation, median (range) missing OME use 24 hour after peak troponin elevation, median (range) Lab variables, median (range) WBC count (×109/L) Hemoglobin (g/dL) Platelet count (×109/L) Serum creatinine (mg/dL) LDH (U/L)

Cardiac MRI data (N=47)† CMD Non-CMD (n=15) (n=32)

19 (90) 2 (10)

25 (60) 17 (40)

0.02

12 (80) 3 (20)

26 (81) 6 (19)

21 (100) 0 (0)

24 (57) 18 (43)

0.0002

5 (100) 0 (0) 16 75 (75-75) 20

27(73) 10(27) 5 64 (0-1524) 19

0.31

1.00

7 (78) 2 (22) 6 NA

11 (85) 2 (15) 19 NA

119 (0-3200)

12.2 (5.2-61) 13.7 (4.5-34.7) 7.7 (5.6-12) 7.7 (2.9-12.6) 272 (127-633) 165.5 (41-824) 0.7 (0.1-1.4) 1.4 (0.4-7.6) 391.5 (210-1584) 498 (142-18500)

0.39 0.93 0.007 0.0002 0.37

11.9 (8-61) 13.6 (5.2-26.8) 7.8 (5.6-11.9) 7.6 (2.9-12) 282 (140-664) 259 (119-633) 0.8 (0.5-1.6) 0.7 (0.1-7.6) 375 (142-1066) 486.5 (210-2647)

1.00

0.35 0.34 0.30 0.41 0.29

CMD: coronary microvascular disease; MRI: magnetic resonance imaging; CMR: cardiac MRI; ACS: acute chest syndrome; AKI: acute kidney injury; EKG: electrocardiography; OME: oral morphine equivalents; WBC: white blood cell; LDH: lactate dehydrogenase; NA: not applicable; n: number. *Fisher’s exact test for categorical variables; Wilcoxon rank sum test for continuous variables. The number in parenthesis represents percentage of total patients for categorical variables and range for continuous variables, which are reported as median. †Troponin level was not available for six of 69 patients and CMR was not available for 22 of 69 patients.

microvascular disease management in SCD patients. Patients are usually managed with standard acute coronary syndrome management with anticoagulation and antiplatelet agents. In patients with microvascular ischemia in general, aspirin, nitrate, beta-blockers, statins, and other coronary artery disease management interventions have been used but specific data for SCD is not available.13 Exchange transfusion with a goal HbS <30% has been reported to be effective in a patient with recurrent myocardial infarction.14 In a recent study evaluating the effect of hydroxyurea on skeletal and cardiac muscles, SCD patients treated with hydroxyurea had higher resting myocardial perfusion as compared to those without hydroxyurea.15 Newer therapies that target Hb modification could also have a role in the management, but prospective trials are needed to assess the effectiveness of various treatment strategies. We suggest that cardiac MRI with stress testing can be considered as a screening tool for patients with evidence of myocardial damage and recurrent chest pain. This could potentially identify patients who need to be started on medical therapy or considered for exchange transfusions. Improving access to cardiac MRI will be essential since a lack of its universal availability would be a hurdle in adopting this practice widely. Kiranveer Kaur,1 Ying Huang,2 Subha V. Raman,3 Eric Kraut2 and Payal Desai2 1 Division of Medical Oncology, The Ohio State University Wexner Medical Center, Columbus, OH; 2Division of 2020

Hematology, The Ohio State University Wexner Medical Center, Columbus, OH and 3Krannert Institute of Cardiology, Indiana University School of Medicine, Indianapolis, IN, USA Correspondence: PAYAL DESAI - payal.desai@osumc.edu doi:10.3324/haematol.2020.271254 Received: September 2, 2020. Accepted: January 12, 2021. Pre-published: January 21, 2021. Disclosures: no conflicts of interest to disclose. Contributions: PD is the principal investigator of this study and developed the study design, contributed to data interpretation, reviewed, and edited the manuscript; KK contributed to the data collection, analysis, interpretation, and manuscript writing; YH contributed to the statistical analysis and interpretation of the data; SR and EH contributed to study design and manuscript editing.

References 1. Gladwin MT, Sachdev V. Cardiovascular abnormalities in sickle cell disease. J Am Coll Cardiol. 2012;59(13):1123-1133. 2. Martin CR, Johnson CS, Cobb C, et al. Myocardial infarction in sickle cell disease. J Natl Med Assoc. 1996;88(7):428-432. 3. Raman SV, Simonetti OP, Cataland SR, et al. Myocardial ischemia and right ventricular dysfunction in adult patients with sickle cell disease. Haematologica. 2006;91(10):1329-1335. 4. Aslam AK, Rodriguez C, Aslam AF, et al. Cardiac troponin I in sickle cell crisis. Int J Cardiol. 2009;133(1):138-139. 5. Lippi G, De Franceschi L, Salvagno GL, et al. Cardiac troponin T during sickle cell crisis. Int J Cardiol. 2009;136(3):357-358.

haematologica | 2021; 106(7)

Letters to the Editor

6. West AM, Kramer CM. Comprehensive cardiac magnetic resonance imaging. J Invasive Cardiol. 2009;21(7):339-345. 7. Gerber BL, Raman SV, Nayak K, et al. Myocardial first-pass perfusion cardiovascular magnetic resonance: history, theory, and current state of the art. J Cardiovasc Mag Reson. 2008;10(1):18. 8. Niss O, Taylor MD. Applications of cardiac magnetic resonance imaging in sickle cell disease. Blood Cells Mol Dis. 2017;67:126-134. 9. Hunold P, Schlosser T, Vogt FM, et al. Myocardial late enhancement in contrast-enhanced cardiac MRI: distinction between infarction scar and non-infarction-related disease. AJR Am J Roentgenol. 2005; 184(5):1420-1426. 10. Thygesen K, Alpert JS, Jaffe AS, et al. Fourth universal definition of myocardial infarction. J Am Coll Cardiol. 2018;72(18):2231-2264. 11. Platt OS, Brambilla DJ, Rosse WF, et al. Mortality in sickle cell dis-

haematologica | 2021; 106(7)

ease. Life expectancy and risk factors for early death. N Engl J Med. 1994;330(23):1639-1644. 12. James TN, Riddick L, Massing GK. Sickle cells and sudden death: morphologic abnormalities of the cardiac conduction system. J Lab Clin Med. 1994;124(4):507-520. 13. Samim A, Nugent L, Mehta PK, et al. Treatment of angina and microvascular coronary dysfunction. Curr Treat Options Cardiovasc Med. 2010;12(4):355-364. 14. Khalique Z, Pavlu J, Lefroy D, et al. Erythrocytapheresis in the prevention of recurrent myocardial infarction in sickle cell disease. Am J Hematol. 2010;85(1):91-91. 15. Sachdev V, Sidenko S, Wu MD, et al. Skeletal and myocardial microvascular blood flow in hydroxycarbamide-treated patients with sickle cell disease. Br J Haematol. 2017;179(4):648-656.

2021

Letters to the Editor

Entospletinib and obinutuzumab in patients with relapsed/refractory chronic lymphocytic leukemia and B-cell malignancies Therapeutic resistance and intolerance of Bruton tyrosine kinase (BTK) inhibitors is an emerging unmet medical need in chronic lymphocytic leukemia (CLL).1,2 Entospletinib is a small molecule inhibitor of spleen tyrosine kinase (SYK), which is integral to the activation of BTK and the B-cell receptor (BCR) signaling cascade.3 SYK is overexpressed in CLL and its inhibition induces apoptosis of CLL cells in pre-clinical models.4,5 We have shown that BAFF-mediated SYK activation triggered BCR signaling and rendered protection of CLL cells from spontaneous apoptosis in vitro.5 Single agent entospletinib was efficacious in treatment of patients with relapsed/refractory (R/R) CLL who had progressed following chemoimmunotherapy.6 Here we report the results of a single arm, open label, investigator-initiated phase I/II clinical trial which evaluated safety and efficacy of entospletinib in combination with obinutuzumab, a glycoengineered monoclonal anti-CD20 antibody, in patients with R/R CLL and non-Hodgkin lymphoma (NHL) (clinicaltrails gov. Identifier: NCT03010358). Patients enrolled in the phase I dose-escalation portion of the trial included adult patients with CLL7 or NHL (phase I part of the study only) after ≥1 prior therapy. Participants had an Eastern Cooperative Oncology Group (ECOG) performance status ≤2, aspartate transaminase (AST) and alanine transaminase (ALT) <2.5x, bilirubin <2x upper limit of normal and creatinine clearance (CrCl)

≥50 mL/min. Complex karyotype (CK) was determined as presence of ≥3 cytogenetic abnormalities on a CpG-stimulated karyotype. Gene mutations were identified using a 76gene next-generation sequencing panel (GeneTrails®). Bone marrow examinations were required to confirm complete response (CR), with minimal residual disease (MRD) assessment using 8-color flow cytometry (MRDundetectable at <10-4). Patients were enrolled at two dose levels to receive entospletinib 200 mg (dose level 1 [DL1]) or 400 mg (dose level 2 [DL2]) twice-daily orally according to a standard 3+3 design (Online Supplementary Table S1). The primary endpoint for the phase I portion of the study was to determine the maximum tolerated dose (MTD) and/or the recommended phase II dose (RP2D) of the combination. All patients received single agent entospletinib as part of a 7-day run-in. Thereafter, patients received entospletinib on days 1-28 of each 28-day cycle continuously, and obinutuzumab intravenously in standard doses for 6 cycles. Adverse events were graded according to Common Terminology Criteria for Adverse Events (NCI CTCAE) v4.03 and international workshop CLL (iwCLL) criteria.7 Dose limiting toxicity (DLT) was defined as grade≥3 nonhematological toxicity (except grade 3 nausea, vomiting, diarrhea or asymptomatic grade 3-4 laboratory abnormalities reversible within 72 hours; grade 3 infusion reactions, grade 3 tumor lysis syndrome); grade 4 neutropenia lasting >7 days or febrile neutropenia; grade 4 thrombocytopenia/anemia or grade 3 thrombocytopenia with bleeding.

Figure 1. Efficacy of entospletinib and obinutuzumab in patients with chronic lymphocytic leukemia. (A) Redistribution lymphocytosis. Peripheral blood absolute lymphocyte count (ALC) prior to (day -7) and after 7 days of treatment with entospletinib single agent, prior to introduction of obinutuzumab (cycle 1 day 1 [C1D1]). **P<0.01. (B) Nodal response (maximum percent change in the sum of the product [SPD] of the longest perpendicular dimensions) in patients with chronic lymphocytic leukemia evaluable lymphadenopathy. (C) Event-free survival. (D) Duration of treatment.

2022

haematologica | 2021; 106(7)

Letters to the Editor

Table 1. Adverse events.

Adverse events, N (%) Adverse Events - All Hematologic Adverse Events Neutrophil count decreased Platelet count decreased Anemia Febrile neutropenia Lymphocyte count increased

All Patients (N=23) Any Grade Grade ≥ 3 22 (95.7)

15 (65.2)

11 (47.8) 5 (21.7) 3 (13.0) 3 (13.0) 1 (4.3)

10 (43.5) 4 (23.5) 1 (4.3) 3 (13.0) 1 (4.3)

Non-hematologic Adverse Events Fatigue 11 (47.8) Infusion related reaction 10 (43.5) Nausea 8 (34.8) Diarrhea 7 (30.4) Fever 7 (30.4) Alanine aminotransferase increased 5 (21.7) Aspartate aminotransferase increased 5 (21.7) Other Infection 5 (21.7) Creatinine increased 4 (17.4) Pneumonia 3 (13.0) Chills 3 (13.0) Dizziness 3 (13.0) Vomiting 2 (8.7) Tremor 2 (8.7) Tumor lysis syndrome 2 (8.7)

2 (8.7) 4 (17.4) 1 (4.3) 0 0 3 (13.0) 3 (13.0) 1 (4.3) 0 1 (4.3) 1 (4.3) 0 1 (4.3) 1 (4.3) 1 (4.3)

Once the RP2D was determined, a phase II study enrolled patients with R/R CLL only. CR was the primary endpoint.7 Secondary endpoints included objective response rate (ORR), event-free survival (EFS), (defined as the interval between the first study treatment and objective signs of disease recurrence, subsequent antileukemic therapy, or death), and safety. Patients with CLL enrolled at the RP2D level in the phase I were included in the phase II part of the study. The planned sample size of 17 patients in the phase II part of the study provided 83% power to detect a difference in CR of 0.30 (0.20 vs. 0.50) using a one-sided binomial test (P<0.05). The null hypothesis (H0)=2 was based on CLL11 study data (CR 21%).8 Patients who received at least one dose of the study therapy were evaluable for safety. Response rates and exact 95% Confidence Interval (CI) were estimated by Clopper-Pearson method and EFS by the Kaplan-Meier method. Data cut-off date for analysis was January 1, 2020. The study was approved by the Oregon Health and Science University Institutional Review Board and conducted in accordance with the principles of the Declaration of Helsinki. A total of 24 patients (n=22 CLL and n=2 follicular lymphoma [FL]) were enrolled between 08/2017 and 10/2018 (Online Supplementary Table S2). One patient had CLL in Richter’s transformation leading to ineligibility on study entry and was removed from subsequent analysis. Twelve patients were enrolled in the phase I part of the study. The phase II part of the study included 17 patients with CLL (of which six received entospletinib at DL2 on the phase I part of the study). Among the 23 evaluable patients, the median follow-up was 17 months (range, 7-28 months). The median relative dose intensity of entospletinib was 96%. The median treatment duration was 16 months (range, 2-26). Four patients (17.4%) experienced permanent dose reductions due to toxicities. An additional nine patients had temporary dose holds/reductions. Thirteen patients (56.5%) discontinued study therapy due to: progressive disease (n=8), haematologica | 2021; 106(7)

Table 2. Clinical characteristics of patients with chronic lymphocytic leukemia.

Characteristics Age < 65 years ≥ 65 years Sex Male Female ECOG performance status 0 1 2 β2-microglobulin (n=16) ≤3.5 mg/L >3.5 mg/L IgG* (n=20) <700 mg/dL >700 mg/dL FISH Deletion 13q Trisomy 12 Deletion 11q Deletion 17p Complex Karyotype (CK, ≥3 abnormalities) CK + TP53 aberration ≥5 abnormalities Mutational Status TP53 Notch1 SF3B1 High Risk CLL** Prior Treatments 1 2 ≥3 Fludarabine Bendamustine Ibrutinib Idelalisib

All patients N=21 (%) 7 14

(33.3) (66.7)

15 6

(71.4) (28.6)

8 12 1

(38.1) (57.1) (4.8)

5 11

(31.3) (68.7)

16 4

(80.0) (20.0)

14 3 2 5 6 5 1

(66.7) (14.2) (9.5) (23.8) (28.6) (23.8) (4.8)

9 2 4 13

(42.9) (9.5) (19.0) (61.9)

6 8 7 12 14 7 1

(26.8) (38.1) (33.3) (57.1) (66.7) (33.3) (4.8)

*Normal range: 700–1,600 mg/dL. **High-Risk CLL defined as complex karyotype (CK), TP53 aberration, NOTCH1 and SF3B1 mutations. CLL: chronic lymphocytic leukemia: ECOG: Eastern Cooperative Oncology Group; FISH: fluorescense in situ hybridization.

adverse event related to entospletinib (n=1; recurrent AST/ALT abnormalities); a new diagnosis of breast cancer unrelated to study treatment (n=1), withdrawal of consent (n=2), and achievement of CR with undetectable MRD in the bone marrow after 12 cycles of therapy (n=1). Twelve patients enrolled in the phase I part of the study received a median of 18 cycles (range, 7-28) of study therapy. Of six patients (four CLL, two FL) who received entospletinib 200 mg (DL1), one patient experienced a DLT (grade 3 asymptomatic AST/ALT abnormalities) attributed to entospletinib. No DLT were observed among the six patients who received entospletinib 400 mg (DL2). Thus, entospletinib 400 mg twice-daily was determined to be the RP2D in combination with obinutuzumab. Treatment-related AE were reported in 95.7% of patients (Table 1). Grade 3 or higher AE occurred in 65.2%. The most common hematologic AE observed across all patients were neutropenia (47.8%), thrombocy2023

Letters to the Editor

topenia and anemia, as in earlier studies of SYK inhibitors.6,9,10 Neutropenia (43.5%; including four patients [17.4%] who had transient grade 4 neutropenia attributed to obinutuzumab), thrombocytopenia and anemia were the most common grade ≥3 hematologic toxicities. The median onset of neutropenia was 7 days after the first obinutuzumab infusion, median duration was 28 days. In six of 11 patients, first onset occurred during cycle one of therapy. Growth factor support was not required. The most frequently occurring non-hematological AE of all grades were fatigue, infusion-related reaction and nausea. The most frequent grade 3-4 non-hematological AE were: infusion-related reaction (17.4%; all attributed to obinutuzumab) and increased AST/ALT abnormalities (13.0%), the latter adverse event reported in 10-20% of patients receiving SYK inhibitors.6,10-12 Despite the fact that patients had received a median of two prior therapies, including fludarabine and bendamustine, few patients (13%) developed febrile neutropenia or pneumonia. There were no grade 5 AE. Of 23 patients, only one patient discontinued therapy due to AE (recurrent AST/ALT abnormalities which resolved upon cessation of entospletinib). This discontinuation rate (4.3%) compares favorably with those seen in early-phase trials and real-world analyses of BTK inhibitors over a similar follow-up period.1,13-15 Efficacy of entospletinib+obinutuzumab was analyzed in the 21 patients with CLL, of which 17 received entospletinib at RP2D (400 mg twice daily). Among an additional four patients who were initially treated with 200 mg entospletinib, two escalated to RP2D after 12 and 13 treatment cycles. Patients with CLL had a median age of 66 years (range, 48-76; Table 2). Ten patients (47.6%) had either complex karyotype (CK; n=6) or a TP53 aberration (n=9). Including CK, TP53 aberration, NOTCH1 and SF3B1 mutations, 13 patients (61.9%) had unfavorable cytogenetic and molecular characteristics, defined as “high-risk CLL”. The median number of prior therapies was two (range, 1-6). Seven patients had received prior ibrutinib and one patient received the phosphoinotiside-3 kinase (PI3K) inhibitor idelalisib. Of those, four patients discontinued ibrutinib due to intolerance, one per their discretion, and three patients discontinued ibrutinib and idelalisib for progressive disease. Baseline median absolute neutrophil count was 2,700/mL, hemoglobin 10.9 g/dL, platelet count 150,500/mL, immunoglobulin (Ig) G level 494 mg/dL (range, 176-1475 mg/dL). Treatment with entospletinib for 7 days during run-in resulted in redistribution lymphocytosis (Figure 1A). Among the 21 efficacy-evaluable participants with CLL, the ORR was 66.7% (95% CI: 43.0–85.4). Three patients (14.3%, 95% CI: 3.1–36.3) achieved a CR, and 11 patients (52.4%) had a partial response (PR). The remaining seven patients had stable disease as their best response. Two of three patients with confirmed CR had undetectable MRD in the bone marrow. The median time to CR was 10.7 months (range, 7.0-17.9 months) and time to PR was 5.8 months (range, 2.8-6.5 months; Online Supplementary Table S3). Nineteen patients with evaluable lymphadenopathy demonstrated a reduction in lymph node sizes (Figure 1B). The low CR rate is consistent with that seen with BCR-signaling inhibitors.13,15 Median EFS was 24 months, treatment duration – 17 months (95% CI: 16–28; Figure 1C and D). Nine patients with CLL (eight out of nine discontinued due to progressive disease) started subsequent therapy (six with BTK 2024

inhibitors, two with venetoclax and one with PI3K inhibitor). All but one patient remains alive. Thirteen patients with high-risk disease genetics had an ORR of 53.8% (five PR and two CR). Five patients remain on treatment, with a median EFS of 24 months. Among the eight patients who had previously received kinase inhibitors, ORR was 62.5% (all PR) and median EFS was 17 months. Among the 17 patients with CLL who received entospletinib at RP2D in phase II of the study ORR was 82.4% (64.7% PR, 17.6% CR). The median EFS has not been reached, and the estimated 2-year EFS is 64%. This compares favorably with entospletinib monotherapy in patients with R/R CLL (ORR 61%, median PFS 13.8 months).6 Thus, the combination of entospletinib and obinutuzumab has an acceptable safety profile and shows a strong efficacy signal warranting continued exploration in CLL. Adam S. Kittai,1° Scott Best,1 Bria Thurlow,1 Vi Lam,1 Taylor Hashiguchi,1 Shaun Goodyear,1 Daniel O. Persky,2 Craig Okada,1 Byung Park,1 Stephen E. Spurgeon1 and Alexey V. Danilov1,3 1 Knight Cancer Institute, Oregon Health & Science University, Portland, OR; 2University of Arizona Cancer Center, Tucson, AZ and 3City of Hope Comprehensive Cancer Center, Duarte, CA, USA °Current address: Ohio State University, Columbus, OH, USA Correspondence: ALEXEY V. DANILOV - adanilov@coh.org doi:10.3324/haematol.2020.270298 Received: August 20, 2020. Accepted: January 18, 2021. Pre-published: January 28, 2021. Disclosures: AVD received research funding from Aptose Biosciences, AstraZeneca, Gilead Sciences, Takeda Oncology, Genentech, Bayer Oncology, Verastem Oncology and Bristol-Myers Squibb, and consulted for Astra Zeneca, Abbvie, Beigene, Bayer Oncology, Bristol-Meyers-Squibb, Genentech, Karyopharm, Pharmacyclics, TG Therapeutics, Nurix and Rigel Pharmaceuticals; SES received research fundingfrom Janssen, Bristol-Myers Squibb, Acerta, Velos-Bio, Gilead Sciences, Genentech and Astra Zeneca as well as consultancy fees from Janssen. Contributions: AVD designed the study, provided oversight at all stages of the study, performed research and analyzed data; ASK, SB, BT, FX, VL, TH, DO, CO, BP and SES performed research and analyzed data; ASK, BT, SG and AVD wrote the manuscript; all authors participated in drafting, revising, and approving the final manuscript. Acknowledgements: we extend our thanks to the patients and their families. We would also like to thank Sarah Dreumont, Basak Gokcora, Renee MacKinnon, Kirsten Orand, Bethany Wollam, Nan Subbiah, Stephen Monette, Lacy Moore, Debbie Ryan and the rest of the lymphoma team at Oregon Health and Science University for their contribution to this study. Funding: this Investigator-sponsored study was funded by Gilead Sciences Inc and Roche/Genentech; AVD was supported in part by the Leukemia & Lymphoma Society Scholar in Clinical Research Award #2319-19.

References 1. Mato AR, Nabhan C, Thompson MC, et al. Toxicities and outcomes of 616 ibrutinib-treated patients in the United States: a real-world analysis. Haematologica. 2018;103(5):874-879.

haematologica | 2021; 106(7)

Letters to the Editor

2. Fletcher L, Nabrinsky E, Liu T, et al. Cell death pathways in lymphoid malignancies. Curr Oncol Rep. 2020;22(1):10. 3. Currie KS, Kropf JE, Lee T, et al. Discovery of GS-9973, a selective and orally efficacious inhibitor of spleen tyrosine kinase. J Med Chem. 2014;57(9):3856-3873. 4. Gobessi S, Laurenti L, Longo PG, et al. Inhibition of constitutive and BCR-induced Syk activation downregulates Mcl-1 and induces apoptosis in chronic lymphocytic leukemia B cells. Leukemia. 2009; 23(4):686-697. 5. Paiva C, Rowland TA, Sreekantham B, et al. SYK inhibition thwarts the BAFF - B-cell receptor crosstalk and thereby antagonizes Mcl-1 in chronic lymphocytic leukemia. Haematologica. 2017;1 02(11):1890-1900. 6. Sharman J, Hawkins M, Kolibaba K, et al. An open-label phase 2 trial of entospletinib (GS-9973), a selective spleen tyrosine kinase inhibitor, in chronic lymphocytic leukemia. Blood. 2015;1 25(15):2336-2343. 7. Hallek M, Cheson BD, Catovsky D, et al. Guidelines for the diagnosis and treatment of chronic lymphocytic leukemia: a report from the International Workshop on Chronic Lymphocytic Leukemia updating the National Cancer Institute-Working Group 1996 guidelines. Blood. 2008;111(12):5446-5456. 8. Goede V, Fischer K, Busch R, et al. Obinutuzumab plus chlorambucil in patients with CLL and coexisting conditions. N Engl J Med. 2014; 370(12):1101-1110. 9. Friedberg JW, Sharman J, Sweetenham J, et al. Inhibition of Syk with fostamatinib disodium has significant clinical activity in non-

haematologica | 2021; 106(7)

Hodgkin lymphoma and chronic lymphocytic leukemia. Blood. 2010;115(13):2578-2585. 10. Awan FT, Thirman MJ, Patel-Donnelly D, et al. Entospletinib monotherapy in patients with relapsed or refractory chronic lymphocytic leukemia previously treated with B-cell receptor inhibitors: results of a phase 2 study. Leuk Lymphoma. 2019;60(8):1972-1977. 11. Danilov AV, Herbaux C, Walter HS, et al. Phase Ib study of tirabrutinib in combination with idelalisib or entospletinib in previously treated chronic lymphocytic leukemia. Clin Cancer Res. 2020;2 6(12):2810-2818. 12. Bussel J, Arnold DM, Grossbard E, et al. Fostamatinib for the treatment of adult persistent and chronic immune thrombocytopenia: results of two phase 3, randomized, placebo-controlled trials. Am J Hematol. 2018;93(7):921-930. 13. Byrd JC, Furman RR, Coutre SE, et al. Targeting BTK with ibrutinib in relapsed chronic lymphocytic leukemia. N Engl J Med. 2013; 369(1):32-42. 14. Gordon MJ, Churnetski M, Alqahtani H, et al. Comorbidities predict inferior outcomes in chronic lymphocytic leukemia treated with ibrutinib. Cancer. 2018;124(15):3192-3200. 15. Danilov AV, Persky DO. Incorporating acalabrutinib, a selective next-generation Bruton tyrosine kinase inhibitor, into clinical practice for the treatment of haematological malignancies. Br J Haematol. 2021;193(1):15-25.

2025

Letters to the Editor

An international retrospective study for tolerability of 6-mercaptopurine on NUDT15 bi-allelic variants in children with acute lymphoblastic leukemia 6-mercaptopurine (6-MP) is one of the essential chemotherapeutic agents for treatment of acute lymphoblastic leukemia (ALL) in children and adults.1 Bone marrow suppression is the main dose-limiting toxicity of 6-MP, and the sensitivity to 6-MP is strongly affected by germline variants in genes regulating thiopurine metabolism.2 Recently, the NUDT15 variant c.415C>T has been identified as a genetic cause for 6-MP intolerability,3 which could explain the majority of thiopurine-induced myelosuppression in Asians that are also common in Hispanics.2 So far, multiple NUDT15 haplotypes with various combination of variants are known to exist (Figure 1A). Several researchers have reported that these variants had decreased NUDT15 activity,4, 5 and bi-allelic variants caused extremely intolerance to 6-MP.6 However, individual studies included a limited number of patients with bi-allelic variants, which significantly hindered the comprehensive analysis of the exact clinical course of 6-MP toxicity and development of evidencebased recommendations. Therefore, in this international collaborative study, we comprehensively evaluated the actual 6-MP tolerable dose, frequencies of 6-MP-induced toxicity, and outcomes in ALL patients with bi-allelic variants of NUDT15. We asked collaborators from Japan, Singapore, Malaysia, Taiwan, China, and Thailand, about their experience of cases with NUDT15 bi-allelic variants, which led to the identificationof 37 ALL cases, most of the which were genotyped due to intolerance to 6-MP. Clinical information of the cases was retrospectively collected, focusing on 6-MP dosing and toxicity. Patients with NUDT15 bi-allelic variants were enrolled in this study, including some patients in prior case reports or small case series.6, 7 NUDT15 was genotyped by Sanger sequencing4. Thiopurine methyltransferase (TPMT) genotype information was available for 20 cases, and no case had hypomorphic variants which also confer 6-MP sensitivity. The treatment of maintenance therapy typically started with 40 to 60 mg/m2/day of 6-MP (Online

Supplementary Table S1) and 20 to 40 mg/m2/week of methotrexate (MTX); these dosages were adjusted to maintain the target leukocyte count at 1,500 to 3,000/mL. Toxicities were graded by the Common Terminology Criteria for Adverse Events (CTCAE) version 4.0, and those rates were estimated by cumulative incidence. The tolerated dosages of 6-MP and MTX were defined as the average (mean) of the doses per day or per week, respectively, during the entire duration of maintenance therapy. The dose for bi-allelic variant was compared with the dose for wild-type and mono-allelic variant in our previous report.6, 7 The average dose in each NUDT15 genotype was estimated by the Kruskal-Wallis test. The interruption duration between 6-MP initial doses was estimated by the Mann Whitney U-test. Four-year overall survival (OS) and event-free survival (EFS) from start of maintenance therapy were estimated by the log-rank test. The statistical analysis was conducted using R statistical software (version 3.4.1; http://www.r-project.org/). Patient characteristics for the 37 cases are shown in Table 1. Patients with bi-allelic variant had intolerance to 6-MP, and reduction was required mainly due to myelosuppression (Online Supplementary Figure S1). The average 6-MP dose of these patients during maintenance therapy was 5.2 (range, 1.1–25.6) mg/m2/day, and the 6-MP dose by each diplotype is shown in Figure 1B. Comparatively, the average MTX dose was 10.4 (range, 1.9–44.6) mg/m2/week (Online Supplementary Figure S2). This 6-MP dose was significantly lower compared with the average dose for the NUDT15 wild-type (n=138, 41.7 mg/m2, P= 3.9×10–14) and mono-allelic variant (n=47, 33.6 mg/m2, P=2.7×10–13) in Japanese patients reported previously (Figure 2).6, 7 Most of the cases showed intolerance to 6MP, and 10 mg/m2 or less was sufficient to maintain the target leukocyte range for 32 (86.4%) of the 37 cases. The median 6-MP average dose for *2/*2, *2/*3, and *3/*3 (poor metabolizer [PM]) were 5.2 mg/m2/day, and the average dose was not different among these three diplotypes (P=0.29, Figure 1B). NUDT15 haplotypes other than PM showed heterogeneous sensitivity to 6-MP, although the average 6-MP dose as a group was not statistically different from PM (Online Supplementary Table S2, P=0.53).

Table 1. Patient characteristics Total, n Male/Female, n Median age, years (range) Immunotype (BCP/T), n Median 6-MP initial dose, mg/m2 (range) NCI/Rome criteria Standard/High risk, n NUDT15 genotype, n *2/*2 *2/*3 *2/*5 *2/*6 *2/*7 *3/*3 *3/*5 *5/*5

Japan

Singapore

Taiwan

China

Thailand

20 9/11 6 (3-15) 18/2 17.2 (1.9-51.3)

7 3/4 6 (3-14) 7/0 10.9 (3.5-17.5)

6 5/1 9 (3-16) 5/1 11.4(5.0-39.5)

3 1/2 6 (4-7) 2/1 20.6 (4.3-29.8)

1 1/0 5 1/0 24.4

15/5

3/4

4/2

2/1

1/0

1 5 1 0 0 10 2 1

0 1 0 2 0 4 0 0

3 0 0 0 1 2 0 0

0 1 0 0 0 1 1 0

0 0 0 0 0 1 0 0

BCP: B-cell precursor; T: T-cell; 6-MP: 6-mercaptopurine; n: number; 6-MP: 6-mercaptopurine; NCI: National Cancer Institute.

2026

haematologica | 2021; 106(7)

Letters to the Editor

Thirty-two of the 37 patients (86.5%) required interruption of maintenance therapy, and the median duration of interruption for all patients was 47 days (range, 0– 148 days). In patients with a 6-MP initial dose <10 mg/m2, the days of interruption during whole maintenance therapy was significantly shorter than in patients with a 6-MP initial dose of 10 mg/m2 or more (P=0.042) (Online Supplementary Figure S3). When limited to the interruption within the first 8 weeks of maintenance therapy,8 the effect of the initial dose was more remarkable (Figure 1C). In terms of toxicities, 36 of the 37 patients were observed to have grade 3 or worse neutropenia. Grade 4 leukopenia and grade 4 neutropenia were observed in 16 (43.2%) and 32 (86.4%) patients, respectively, and the median observation times of leukopenia and neutropenia were 33 days (range, 19–662 days) and 37 days (range, 9–

139 days), respectively, from start of the maintenance therapy (Figure 1D). We, thus, confirmed that the doselimiting toxicity of 6-MP in patients with NUDT15 biallelic variant was neutropenia. Moreover, during the consolidation therapy (most of the protocol adopted early consolidation with 6-MP, so called "IB"), severe myelosuppression was observed in 21 of these patients (Online Supplementary Table S3). Conversely, grade 3 or worse liver enzyme elevation was observed in only 10 patients. The median duration of follow-up was 1,398 days (range, 84–5,357 days) from the start of maintenance therapy. One patient relapsed during maintenance therapy and five patients relapsed at 772 to 2,659 days from the start of maintenance therapy. Three of these six patients died at 499 to 720 days after relapse. The causes

Figure 1. Tolerability and efficacy for patients with NUDT15 bi-allelic variants. (A) Major haplotypes of NUDT15. (B) Average 6-mercaptopurine (6-MP) dose in each NUDT15 bi-allelic variant. (C) The association between initial 6-MP dose and therapy interruption for 56 days for start of therapy in maintenance therapy in patients with NUDT15 bi-allelic variant. Black circles and white circles show starting dose for patients with bi-allelic variant of exon 3 and others, respectively. (D) Toxicity during maintenance therapy.

haematologica | 2021; 106(7)

2027

Letters to the Editor

of death were relapse of leukemia, second malignancy, or complications related to bone marrow transplantation. OS and EFS were 91%± 6% and 82%±7%, respectively (Online Supplementary Figure S4). This Asian international study showed that most patients with NUDT15 PM required a reduced 6-MP dose to <10 mg/m2 during maintenance therapy. These findings were concordant with the recommendations by the Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines.2 NUDT15 c.52G>A and c.36_37insGGAGTC are defined as an uncertain function allele in the CPIC guidelines,2 and a patient with *5/*5 can tolerate as high as 18.3 mg/m2. However, three cases with *3/*5 had intolerance to 6-MP at <10 mg/m2, pointing to a compound heterozygous effect. Additionally, cases with bi-allelic variant with *6 (only c.36_37insGGAGTC) might be more tolerant to 6-MP than those with c.415C>T. Moriyama et al. defined *3 as low, and *5 and *6 as intermediate activity in vitro.4 Our results demonstrate that diplotypes of intermediate/intermediate tolerate moderate intensity, but that intermediate/low is extremely sensitive to 6-MP. These heterogeneous sensitivities in bi-allelic variants of NUDT15 highlight the importance of precise diplotyping analysis. Twenty-seven patients started maintenance therapy with the reduced 6-MP dose to less than 30 mg/m2, mainly because they experienced severe toxicities during consolidation and their NUDT15 variants had already been genotyped. As shown in the Online Supplementary Figure S1, typical cases with NUDT15 bi-allelic variants showed a sudden crash of the leukocyte count after an approximately 2-week exposure to 6-MP, and required a long time to attain recovery of leukocyte counts. These observations are concordant with the findings of previous reports.8, 9 Accordingly, adjustment of the 6-MP dose is often difficult in most cases as the 6-MP dose fluctuated dramatically and treatment interruption was common. With a reduced starting dose of 6-MP, dose fluctuation was not observed and maintenance therapy could be given continuously. However, some researchers reported that patients with the NUDT15 c.415C>T variant developed thiopurine-induced leukopenia within 2 months from initiation of therapy.7, 10 Regarding tolerability to MTX, some studies reported that the average MTX dose was not different in NUDT15 genotypes.6, 11 However, some cases had reduced MTX dose, probably due to myelosuppression caused by 6-MP and, thus, the optimal MTX dose in NUDT15 bi-allelic cases needs to be established in future studies. Patients with the NUDT15 variant experienced thiopurine-induced hematological toxicity for several months regardless of the disease or race.9 The majority of patients with NUDT15 bi-allelic variant experienced grade 4 neutropenia. This finding was in line with previous reports that Nudt15–/– mice, which demonstrated significantly decreased neutrophil counts upon thiopurine exposure.12 Neutrophils were more sensitive than other leukocytes to thiopurine with deficient NUDT15. For patients with biallelic variants, neutrophil counts should be carefully monitored, as well as total leukocyte counts, during 6MP treatment. Given the risk of severe infectious complications, pre-emptive NUDT15 genotyping for all patients with ALL should be performed and dose modification in cases with bi-allelic variants must be considered. This study has some limitations. First, TPMT genotype information is insufficient because routine screening for TPMT variants, another determinant of 6-MP sensitivity, was not performed. However, considering variant distribution of NUDT15 and TMPT2, variant allele frequency 2028

Figure 2. Average 6-mercaptopurine dose during maintenance therapy.

of TPMT in those with NUDT15 bi-allelic variant is extremely low as observed in our limited data. Therefore, we can select, according to each racial background, which of the two major genetic determinants of 6-MP should be genotyped. However, considering recent racial mixture and advances in genomic analysis technology, comprehensive genotyping information responsible for drug sensitivity for all cases should be obtained to provide a precise medical approach. Second, most of our cases were identified as having NUDT15 variants because of their intolerance to 6-MP, and , thus, the tolerable dose of NUDT15 bi-allelic cases may be overestimated, which underpins the importance of upfront genotyping. Third, the number of cases with some haplotypes (such as *6 or *7) were small, and tolerability of those patients with these rare haplotypes still needs to be determined by future studies. In conclusion, bi-allelic NUDT15 variants conferred extreme intolerance to 6-MP. Pre-emptive NUDT15 genotyping for all patients with ALL should be performed and dose modification in cases with bi-allelic variants must be considered. Precise upfront genotyping and a reduction of the 6-MP dose to less to than 10 mg/m2 is recommended to avoid the risk of severe complications and therapy interruption. Yoichi Tanaka,1 Allen Eng Juh Yeoh,2 Takaya Moriyama,3 Chi-Kong Li,4 Ko Kudo,5 Yuki Arakawa,6 Jassada Buaboonnam,7 Hui Zhang,8 Hsi-Che Liu,9 Hany Ariffin,10 Zhiwei Chen,2 Shirley K.Y. Kham,2 Rina Nishii,3 Daisuke Hasegawa,11 Junya Fujimura,12 Dai Keino,13 Kensuke Kondoh,13 Atsushi Sato,14 Takahiro Ueda,15 Masaki Yamamoto,16 Yuichi Taneyama,17 Moeko Hino,18 Masatoshi Takagi,19 Akira Ohara,20 Etsuro Ito,5 Katsuyoshi Koh,7 Hiroki Hori,21 Atsushi Manabe,22 Jun J. Yang,3 and Motohiro Kato,23,24,25 1 Division of Medicinal Safety Science, National Institute of Health Sciences, Kawasaki, Japan; 2VIVA-NUS Center of Translational Research in Acute Leukemia (Molecular), Department of Pediatrics, haematologica | 2021; 106(7)

Letters to the Editor

Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 3Department of Pharmaceutical Sciences, St. Jude Children's Research Hospital, Memphis, TN, USA; 4Hong Kong Children’s Hospital, The Chinese University of Hong Kong, Shatin, Hong Kong, Special Administrative Region, China; 5Department of Pediatrics, Hirosaki University Graduate School of Medicine, Hirosaki, Japan; 6Department of Hematology/Oncology, Saitama Children’s Medical Center, Saitama, Japan; 7Hematology/Oncology Division, Department of Pediatrics, Faculty of Medicine, Siriraj Hospital Mahidol University, Thailand; 8Department of Hematology and Oncology, Women and Children's Medical Center, Guangzhou, China; 9Division of Pediatric Hematology-Oncology, Mackay Children's Hospital and Mackay Memorial Hospital, Taipei, Taiwan; 10 Department of Pediatrics, University of Malaya Medical Center, Kuala Lumpur, Malaysia; 11Department of Pediatrics, St. Luke's International Hospital, Tokyo, Japan; 12Department of Pediatrics, Juntendo University, School of Medicine, Tokyo, Japan; 13Department of Pediatrics, St. Marianna University School of Medicine Hospital, Kawasaki, Japan; 14Department of Hematology and Oncology, Miyagi Children’s Hospital, Sendai, Japan; 15Department of Pediatrics, Nippon Medical School, Tokyo, Japan; 16Department of Pediatrics, Sapporo Medical University School of Medicine, Sapporo, Japan; 17Department of Hematology and Oncology, Chiba Children’s Hospital, Chiba, Japan; 18Department of Pediatrics, Chiba University Graduate School of Medicine, Chiba, Japan; 19Department of Pediatrics, Tokyo Medical and Dental University, Tokyo, Japan; 20 Department of Pediatrics, Toho University School of Medicine, Tokyo, Japan; 21Department of Pediatrics, Mie University, Tsu, Mie, Japan; 22 Department of Pediatrics, Hokkaido University Graduate School of Medicine, Sapporo, Japan; 23Department of Transplantation and Cell Therapy, Children’s Cancer Center, National Center for Child Health and Development, Tokyo, Japan; 24Department of Pediatric Hematology and Oncology Research, National Center for Child Health and Development, Tokyo, Japan and 25Department of Pediatrics, University of Tokyo, Tokyo, Japan. Correspondence: MOTOHIRO KATO - katom-tky@umin.ac.jp doi:10.3324/haematol.2020.266320 Received: July 8, 2020. Accepted: January 18, 2021. Pre-published: January 28, 2021. Disclosures: no conflicts of interest to disclose Contributions: MK is the principal investigator and takes primary responsibility for the paper—he designed this study, interpreted data, wrote the manuscript, and gave final approval; YT, AY, TM, RN, AM, and JJY collected, analyzed and interpreted data, and wrote the manuscript; CKL, K Kudo, YA, JB, HCL, HA, ZC, SK, DH, JF,

haematologica | 2021; 106(7)

DK, KKondoh, AS, TU, M., YTaneyama, MH, MT, AO, EI, KKoh, and HH evaluated patients and collected data; all authors discussed the results and critically reviewed the manuscript. Acknowledgements: the authors would like to thank Ms. Etsuko Mochizuki for her technical assistance. The authors wish to thank the medical editor from the Department of Education for Clinical Research of the National Center for Child Health and Development for editing this manuscript. Funding: this research was supported by a grant from the NIH (R01GM118578), by AMED under grant numbers JP20ck0106467, and JP20kk0305014, and by the Japan Society for the Promotion of Science (JSPS) through a Grant-in-Aid for Scientific Research (grant numbers 17K10129 and 18K07836).

References 1. Kato M, Manabe A. Treatment and biology of pediatric acute lym-

phoblastic leukemia. Pediatr Int. 2018;60(1):4-12. 2. Relling MV, Schwab M, Whirl-Carrillo M, et al. Clinical Pharmacogenetics Implementation Consortium Guideline for thiopurine dosing based on TPMT and NUDT15 genotypes: 2018 Update. Clin Pharmacol Ther. 2019;105(5):1095-1105. 3. Yang JJ, Landier W, Yang W, et al. Inherited NUDT15 variant is a genetic determinant of mercaptopurine intolerance in children with acute lymphoblastic leukemia. J Clin Oncol. 2015;33(11):1235-1242. 4. Moriyama T, Nishii R, Perez-Andreu V, et al. NUDT15 polymorphisms alter thiopurine metabolism and hematopoietic toxicity. Nat Genet. 2016;48(4):367-373. 5. Valerie NC, Hagenkort A, Page BD, et al. NUDT15 hydrolyzes 6thio-deoxyGTP to mediate the anticancer efficacy of 6-thioguanine. Cancer Res. 2016;76(18):5501-5511. 6. Tsujimoto S, Osumi T, Uchiyama M, et al. Diplotype analysis of NUDT15 variants and 6-mercaptopurine sensitivity in pediatric lymphoid neoplasms. Leukemia. 2018;32(12):2710-2714. 7. Tanaka Y, Kato M, Hasegawa D, et al. Susceptibility to 6-MP toxicity conferred by a NUDT15 variant in Japanese children with acute lymphoblastic leukaemia. Br J Haematol. 2015;171(1):109-115. 8. Sutiman N, Chen S, Ling KL, et al. Predictive role of NUDT15 variants on thiopurine-induced myelotoxicity in Asian inflammatory bowel disease patients. Pharmacogenomics. 2018;19(1):31-43. 9. Schaeffeler E, Jaeger SU, Klumpp V, et al. Impact of NUDT15 genetics on severe thiopurine-related hematotoxicity in patients with European ancestry. Genet Med. 2019;21(9):2145-2150. 10. Yang SK, Hong M, Baek J, et al. A common missense variant in NUDT15 confers susceptibility to thiopurine-induced leukopenia. Nat Genet. 2014;46(9):1017-1020. 11. Khera S, Trehan A, Bhatia P, Singh M, Bansal D, Varma N. Prevalence of TPMT, ITPA and NUDT 15 genetic polymorphisms and their relation to 6MP toxicity in north Indian children with acute lymphoblastic leukemia. Cancer Chemother Pharmacol. 2019;83(2):341-348. 12. Nishii R, Moriyama T, Janke LJ, et al. Preclinical evaluation of NUDT15-guided thiopurine therapy and its effects on toxicity and antileukemic efficacy. Blood. 2018;131(22):2466-2474.

2029

CASE REPORTS A mutation in the iron-responsive element of ALAS2 is a modifier of disease severity in a patient suffering from CLPX associated erythropoietic protoporphyria Porphyrias are a group of eight genetically distinct disorders, each resulting from a partial deficiency or gain-offunction of a specific enzyme in the heme biosynthetic pathway.1 Porphyrias are inherited as autosomal dominant, autosomal recessive or X-linked traits.2 Erythropoietic protoporphyria (EPP) is a constitutive hematological disorder characterized by protoporphyrin IX (PPIX) accumulation in erythrocytes and other tissues resulting in acute skin photosensitivity, mild microcytic anemia, and rarely, severe liver disease. The majority of the patients with EPP present the autosomal EPP form (OMIM #177000) due to a partial deficiency of ferrochelatase (FECH), the last enzyme of the heme biosynthetic pathway.1,2 In most EPP patients, the clinical expression requires the coinheritance of a FECH mutation, that abolishes or markedly reduces FECH activity, in trans to an hypomorphic FECH allele (rs2272783, NM_000140.3; c.[315-48T>C]) carried by about 11% of Caucasians.3 In Europe and the USA, 4-10% of EPP patients have been reported to harbor a gain-of-function mutation in the 11th exon of the erythroid d-aminolevulinic synthase gene (ALAS2).4 Rare cases of EPP have also been reported in few reference centers without any mutations in the FECH or ALAS2 genes (personal communication from JC Deybach). EPP due to gain-of-function ALAS2 mutations are inherited as an X-linked trait and result in a distinct biochemical feature not only with overproduction and accumulation of free PPIX but also zinc protoporphyrin (ZnPP) (XLPP) (OMIM #300751). In these patients, the FECH enzyme is functional and utilizes all available iron for heme production. The excess PPIX is used to make ZnPP in a reaction catalyzed by FECH.3,5 Moreover, gain-of-function missense mutations altering the C-terminal part of ALAS2 exacerbate congenital erythropoietic porphyria, suggesting that ALAS2 is a gatekeeper of erythroid heme biosynthesis and may function as a modifier gene.6

The 5’ untranslated region (UTR) of ALAS2 mRNA contains a cis-regulatory iron-responsive element (IRE) that confers iron-dependent posttranscriptional regulation by the iron regulatory proteins (IRP).7 IRE mutations are known to cause human diseases. IRE mutations in ferritin L mRNA cause hereditary hyperferritinemia with cataract syndrome (HHCS) (OMIM #600886),8-9 and a single point mutation in the IRE of ferritin H is responsible for an autosomal dominant iron overload phenotype (OMIM #615517).10 These cases suggest a possible role for IRE mutations in the ALAS2 gene in modifying the severity of hematologic diseases. Here, we describe an 18 year-old Caucasian female proband (III:2, Figure 1A) referred to the French Center of Porphyria because of early onset (9 months) acute photosensitivity characterized by painfully phototoxic reactions suggesting EPP. An incomplete genetic characterization of this case was previously reported to harbor a gain of function in the mitochondrial unfoldase gene, CLPX.11 Written informed consent was obtained for all participants. This study was approved through the local ethical committees in accordance with the World Medical Association Declaration of Helsinki ethical principles for medical research involving human subjects and its subsequent amendments (R162-16-7 and 145-15-4 French ethical agreement). In the proband, a high level of free erythrocyte PPIX and ZnPP confirmed the diagnosis of EPP (Table 1). At the age of diagnosis, the proband also presented with a microcytic iron deficiency anaemia (Table 1). FECH enzyme activity was normal (Table 1). No point mutation or large FECH gene deletion were identified, and chromosome 18, where FECH gene is located, was excluded by linkage and comparative genomic hybridization (CGH) array analyses. Moreover, the proband did not harbor the FECH low-expressed allele rs2272783, NM_000140.3; c.[31548T>C] (IVS3-48C). The father (II.4) and one uncle (II.2) of the proband presented with zinc- and free PP accumulation in erythrocytes that were associated with a mild photosensitivity, but without symptoms of EPP (Figure 1 and Table 1). Whole-exome sequencing (WES) analysis showed that proband (III.2), the father (II.4) and the uncle (II.2) carried a heterozygous single

Table 1. Clinical and biochemical data in affected subjects of the erythropoietic protoporphyria proband’s family.

Pedigree

II:2

II:4

III:2

II:5

I.3

II:6

Sex Photosensitivity IRE ALAS2 mutation Age at the visit (years) Total porphyrins in plasma (nmol/L) Erythroid PPIX (mmol/L RBC): Free PPIX (%) ZnPP (%) FECH activity (nmol/mg prot. h) Hb (g/dL) Serum iron (mmol/L) Serum ferritin (mg/L)

M mild 56 54 30.4 33 67 4.4 15.5 25.0 196

M mild 46 52 26.7 38 62 3.9 15.9 20.0 171

F Severe + 18 20 936 780 140.9 89.5 71 65 29 35 3.6 3.9 10.9 12.1 4.0 11.0 5 11

F no + 43 22 2.1 20 80 4.3 12.2 21.0 12

F no + 78 19 1.9 17 83 4.2 13.3 14.0 137

F no + 51 20 2.0 22 78 4.2 14.0 20.0 85

Transferrin saturation (%) Soluble Transferrin receptor (mg/L)

31 1.24

32 1.41

5 3.95

31 1.15

24 1.26

36 1.19

15 1.75

Normal values

< 20 <1.9 < 28 > 72 >3.5 11.5 – 16.0 12.0 - 26.0 15 – 250♂ 8 – 150♀ 20 - 45 0.76-1.76

The proband (III:2) was examined twice: first at the initial visit associated with iron deficiency and one year later after oral iron supplementation.PPIX: protoporphyrin IX; ZnPP: Zinc protoporphyrin; Hb: hemoglobin; FECH: ferrochelatase; RBC: red blood cells; m: male; f: female.

2030

haematologica | 2021; 106(7)

Case Reports

Figure 1. ALAS2 iron-responsive element mutation as modifier for erythropoietic protoporphyria clinical severity. (A) Pedigree of the studied family. Filled black symbol indicates the patient with clinical overt erythropoietic protoporphyria (EPP). Grey symbols indicate relatives (II.2 and II.4) with protoporphyrin IX (PPIX) over-production in red blood cells (RBC) associated with mild photosensitivity without EPP; the grandmother (I.1) presented with a suspected mild photosensitivity but no biochemical data were available; barred symbols indicate deceased subjects. Asterisks indicate subjects characterized at the molecular level. Partial chromatograms of ALAS2 iron-responsive element (IRE) sequences where the -38T>C mutation is located. Reference sequence for ALAS2: NM_000032.4. (B) Non-radioactive competitive electrophoretic mobility shift assays (EMSA) with wild-type (wt) and mutated (mut) ALAS2 IRE. Graphic representation of the ferritin H (FTH1) IRE sequences (nucleotides 60-85 in NM_002032.2) wt or mut -165ΔC, ALAS2 IRE sequence (nucleotides 93-125 in NM_000032.4) wt or the -38U>C mut, used for the synthesis of RNA probes in competitive EMSA experiment. RNA Watson-Crick pairs are depicted as A-U; U-A; C-G or G-C; wobble pairings are shown as U.G and not possible pairings are depicted as CxA. Fluorescent-labeled FTH1 IRE wt probe was incubated with increasing molar excess (2x, 5x, 10x, 20x, 40x and 80x) of unlabeled competitors corresponding to the FTH1 IRE wt type sequence (lanes 3-8; lanes 31-26), or the FTH1 IRE mutant -165ΔC (lanes 9-14; lanes 37-42), ALAS2 IRE wt sequence (lanes 17-22; lanes 37-42) or the ALAS2 -38U>C mutation (lanes 23-28; lanes 51-56). Samples were incubated with recombinant IRP1 (upper panel) or IRP2 (bottom panel) and resolved on acrylamide gels. Quantification of the signals of the shifted bands was performed using the Odyssey Infrared Imaging System (LI-COR Biosciences) and compared to the signal in lane N, set as 100%. Means ± standard deviation of at least three independent experiments. Statistical analysis by Student’s t-test (two-tailed) compares the signal given by the mutated IRE of FTH1 or ALAS2 sequences to the signal given by the corresponding wt IRE sequences, at each molar concentration *P<0.05, **P<0.01, ***P<0.001. F: free probe; N: no competitor added. (C) Proteins from CD34+ erythroid cells from proband’s mother (II.5) and the proband (III.2) cultured and differentiated over 14 days were extracted. Twenty micrograms were separated (4-12% NuPAGE gel) with MES NuPAGE buffer. Proteins were transferred to nitrocellulose membrane. ALAS2 was detected by chemiluminescence using specific antibody. The lower band is ALAS2 whereas the upper band, indicated with an asterix, is an unknown non-specific protein (Agios Pharmaceutical). The membrane was then probed with an anti-human beta-actin antibody as a loading control.

nucleotide substitution (c.1102G>A; p.G298D) at the exon 7 of CLPX gene.11 The proband did not harbor mutations in the 11th exon of ALAS2 gene demonstrating that she is affected by an unusual form of EPP. Sequencing of the rest of the ALAS2 gene revealed an heterozygous T>C change at position -38 in the first exon, at the 5’ UTR and located inside the ALAS2 IRE motif (NM_000032.4; c.[38T>C];[=]) (Figure 1A and B). The mutation is absent in the 6,500 sequenced exomes in the Exome Variant Server database, excluding the possibility of being a neutral polymorphism. This variation was predicted to disturb the stability of the IRE, since the IRE position below the C8 bulge is critical to maintain closed the lower stem (IRE ALAS2 wild-tytpe [wt] predicted minimum free energy ΔG=-7.00 Kcal/mol vs. IRE ALAS2 mutant predicted minimum free energy ΔG=-5.60 haematologica | 2021; 106(7)

Kcal/mol).12 In order to confirm the functional consequence of the -38U>C change in ALAS2 IRE, we performed non-radioactive competitive electrophoretic mobility shift assays (EMSA) with recombinant IRP1 or IRP213 (Figure 1B). As expected, the ferritin H (FTH1) wt positive control showed efficient competition, while its corresponding negative control, the FTH1 mutant 165ΔC could not compete with the labeled FTH1 wt probe (Figure 1B, compare lanes 3-8 to 9-14 in upper panel and 31-36 to 37-42 in bottom panel). In addition, the ALAS2 -38U>C mutation totally abolished the ability of the ALAS2 IRE to compete with FTH1 wt probe for IRP1 or IRP2 binding, in contrast to the wt ALAS2 IRE sequence (Figure 1B, compare lanes 17-22 to 23-28 in upper panel and 45-50 to 51-56 in bottom panel). Therefore, the -38U>C mutation in ALAS2 IRE abolishes IRP1 and IRP2 binding affinity in vitro, which would 2031

Case Reports

impair ALAS2 mRNA regulation in response to iron levels. We hypothesized that under iron limiting conditions, IRP would not repress ALAS2 mRNA translation, with consequent overproduction of ALAS2 protein. This predicted increase in ALAS2 protein levels was indeed observed by immunoblotting for ALAS2 from cell lysates obtained from cultured and differentiated CD34+ erythroid precursors from the proband and the mother (II.5) who only bears the ALAS2 IRE mutation (Figure 1C). Therefore, our data supports the hypothesis that the 38U>C mutation in the ALAS2 IRE increases ALAS2 protein expression and modifies the severity of EPP caused by a confirmed CLPX mutation. In contrast, the ALAS2 gain-of-function in XLPP is caused by increased enzymatic activity in the protein.14 Sequencing of the ALAS2 gene in other family members showed that this mutation does not co-segregate with porphyrin accumulation (Figure 1A and data not shown). The proband inherited the -38U>C change from the maternal branch (II:5; II:6; I3, Figure 1A). Those relatives have higher than normal ZnPP levels and are asymptomatic (Table 1). Methylation analysis confirmed the absence of skewed X inactivation in relatives (II.5 and II.6) and the proband (III.2) who carried the IRE mutation. Previously we and others described the important role of CLPX in heme biosynthesis through its regulation of ALAS2 turnover and enzyme activity.11,15 The G298D mutation in the CLPX gene, present in heterozygosity in this family (proband, father and paternal uncle), decreases the proteolytic activity of the CLPXP protease, causing accumulation of ALAS2.11 However, in comparison to the mild photosensitivity presented by the paternal family, the proband presents with a full and severe EPP phenotype due to the combination of the paternal CLPX G298D mutation and the maternal ALAS2 -38T>C mutation present in the 5’ IRE of ALAS2 gene. Collectively, our data suggests that the ALAS2 IRE mutation is the modifier necessary for the clinical presentation of EPP in the proband. The IRE/IRP-dependent down regulation of ALAS2 during iron deficiency is impaired and contributes to ALA overproduction and increases PPIX level in erythrocytes especially during iron deficiency. This postulate was supported clinically by the beneficial effect of the oral iron therapy when the patient was anemic (Table 1). In summary, we have identified the first described mutation in the IRE of ALAS2 gene that contributes to porphyrin accumulation in erythrocytes above a threshold leading to overt EPP. These data strongly support the role of ALAS2 as an important modifier gene triggering the clinical manifestations in erythropoietic disorders and extend the involvement of its IRE/IRP system to human heme metabolism and erythropoietic porphyria. Sarah Ducamp,1,2* Sara Luscieti,3* Xènia Ferrer-Cortès,4,5 Gaël Nicolas,1,2 Hana Manceau,2,6 Katell Peoc’h,2,6 Yvette Y. Yien,7 Caroline Kannengiesser,1,3 Laurent Gouya,1,2,6 Herve Puy,1,2,6# and Mayka Sanchez4,5# 1 Centre de Recherche sur l’inflammation, INSERM U1149 CNRS ERL Université Paris Diderot, site Bichat, Sorbonne Paris Cité, Paris, France; 2Laboratory of Excellence, GR-EX, Paris, France; 3Institute of Predictive and Personalized Medicine of Cancer (IMPPC), Badalona, Barcelona, Spain; 4Universitat Internacional de Catalunya (UIC), Department of Basic Sciences, Iron metabolism: Regulation and Diseases, Sant Cugat del Vallès, Barcelona, Spain; 5BloodGenetics S.L. Diagnostics in Inherited 2032

Blood Diseases, Esplugues de Llobregat, Barcelona; Spain; Centre Français des Porphyries, AP-HP, Hôpital Louis Mourier, Colombes, France and 7Department of Biological Sciences, University of Delaware, Newark, DE, USA *SD and SL contributed equally as co-first authors. # HP and MS contributed euqally as co-senior authors. Correspondence: HERVE’ PUY - herve.puy@aphp.fr MAYKA SANCHEZ - msanchez@uic.es doi:10.3324/haematol.2020.272450 Received: September 23, 2020. Accepted: January 22, 2021. Pre-published: Fabruary 18, 2021. Disclosures: no conflicts of interest to disclose. Contributions: MS, LG and HP developed the study concept and designed the research; KP, LG and HP recruited family members and collected patients’ data; SD and HM performed patients DNA sequencing; Ckdid the X inactivation and CGH array experiments; SL performed the IRE/IRP functional EMSA; HM did the erythroid progenitor cell culture; GN perfomed immunoblot experiments; XF-C, YYY, HP and MS wrote the manuscript and all authors participated in data discussion, analyzed data, read and approved the manuscript. Acknowledgments: we are very grateful to all parent family who kindly contributed to this study. We thank Drs. Carole Beaumont, Zoubida Karim and Saïd Lyoumi for helpful feedback, discussions, and editorial assistance. Funding: this study was partially supported by grant SAF2015-70412-R, and grant RTI-2018-101735-B-I100, MCI/AEI/FEDER, EU from the Spanish Secretary of Research, Development and Innovation (MINECO); grant DJCLS-R-14/04 from the Deutsche Josep Carreras Leukämie-Stiftung, 2014 SGR225 (GRE) from Generalitat de Catalunya to MS; SD and HM were supported by the Laboratory of excellence, BR-Ex, Paris, France; the labex GR-Ex (reference ANR-11- LABX-0061) is funded by the program “Investissements d’avenir” of the French National Research Agency (reference ANR-11-IDEX-0005-02); YYY is supported by the National Institutes of Health (R03DK118307). 6

References 1. Puy H, Gouya L, Deybach JC. Porphyrias. Lancet. 2010;375(9718):924-937. 2. Balwani M, Desnick RJ. The porphyrias: advances in diagnosis and treatment. Blood. 2012;120(23):4496-4504. 3. Gouya L, Martin-Schmitt C, Robreau AM, et al. Contribution of a common single-nucleotide polymorphism to the genetic predisposition for erythropoietic protoporphyria. Am J Hum Genet. 2006;78(1):2-14. 4. Whatley SD, Ducamp S, Gouya L, et al. C-terminal deletions in the ALAS2 gene lead to gain of function and cause X-linked dominant protoporphyria without anemia or iron overload. Am J Hum Genet. 2008;83(3):408-414. 5. Ducamp S, Schneider-Yin X, de Rooij F, et al. Molecular and functional analysis of the C-terminal region of human erythroid-specific 5-aminolevulinic synthase associated with X-linked dominant protoporphyria (XLDPP). Hum Mol Genet. 2013;22(7):12801288. 6. To-Figueras J, Ducamp S, Clayton J, et al. ALAS2 acts as a modifier gene in patients with congenital erythropoietic porphyria. Blood. 2011;118(6):1443-1451. 7. Dandekar T, Stripecke R, Gray NK, et al. Identification of a novel iron-responsive element in murine and human erythroid deltaaminolevulinic acid synthase mRNA. EMBO J. 1991;10(7):19031909. 8. Beaumont C, Leneuve P, Devaux I, et al. Mutation in the iron responsive element of the L ferritin mRNA in a family with dom-

haematologica | 2021; 106(7)

Case Reports

inant hyperferritinaemia and cataract. Nat Genet. 1995;11(4): 444-446. 9. Girelli D, Corrocher R, Bisceglia L, et al. Molecular basis for the recently described hereditary hyperferritinemia-cataract syndrome: a mutation in the iron-responsive element of ferritin Lsubunit gene (the "Verona mutation"). Blood. 1995;86(11):40504053. 10. Kato J, Fujikawa K, Kanda M, et al. A mutation, in the ironresponsive element of H ferritin mRNA, causing autosomal dominant iron overload. Am J Hum Genet. 2001;69(1):191-197. 11. Yien YY, Ducamp S, van der Vorm LN, et al. Mutation in human CLPX elevates levels of d-aminolevulinate synthase and protoporphyrin IX to promote erythropoietic protoporphyria. Proc Natl Acad Sci U S A. 2017;114(38): e8045-e8052.

haematologica | 2021; 106(7)

12. Campillos M, Cases I, Hentze MW, Sanchez M. SIREs: searching for iron-responsive elements. Nucleic Acids Res. 2010;38 (Suppl):W360-367. 13. Luscieti S, Tolle G, Aranda J, et al. Novel mutations in the ferritin-L iron-responsive element that only mildly impair IRP binding cause hereditary hyperferritinaemia cataract syndrome. Orphanet J Rare Dis. 2013;8(1):30. 14. Fratz EJ, Clayton J, Hunter GA, et al. Human erythroid 5aminolevulinate synthase mutations associated with X-linked protoporphyria disrupt the conformational equilibrium and enhance product release. Biochemistry. 2015;54(36):5617-5631. 15. Whitman JC, Paw BH, Chung J. The role of ClpX in erythropoietic protoporphyria. Hematol Transfus Cell Ther. 2018;40(2):182188.

2033

Case Reports

Light chain proteinuria revealing mu-heavy chain disease: an atypical presentation of Waldenström macroglobulinemia in two cases Heavy chain diseases (HCD) are rare mature B-cells proliferative disorders first described in 19641 and characterized by the production of a paraprotein consisting of truncated heavy chains devoid of bound light chains. Normal immunoglobulin is composed of two heavy chains and two light chains joined by disulfide bonds at the heavy chain constant domain 1 (CH1). In the absence of light chains, the heat shock protein BiP binds to CH1 and retains the heavy chain in the endoplasmic reticulum.2 In HCD various mutations are responsible of splicing error leading to complete or partial deletion of CH1 and preventing therefore the binding of heavy chains to light chains as well as BiP.3,4 Three HCD involving the main immunoglobulin (Ig) classes have been described: a-HCD, γ-HCD and m-HCD which is the least common. A single case of d-HCD has been reported. m-HCD is often associated with a B-cell lymphoid disorder such as chronic lymphocytic leukemia with hepatosplenomegaly. It has also been described in association with myelodysplasia, cirrhosis and auto-immune disease.5 Waldenström macroglobulinemia (WM) is a lymphoplasmacytic lymphoma secreting monoclonal IgM, mainly κ, which is strongly associated with the MYD88 L265P somatic mutation.6 Patients may be asymptomatic or may present symptoms related either to bone marrow infiltration and/or to IgM gammopathy physico-chemical properties (including hyperviscosity, auto-immune

hemolytic anemia, cryoglobulinemia, anti-MAG neuropathy). Serum-free light chains (sFLC) rarely reach high levels in WM and complications related to light chains, like nephropathy or amyloidosis, are uncommon7 compared to multiple myeloma. We report here two cases with IgM κ monoclonal gammopathy visible as a small peak on serum protein electrophoresis (SPEP) and in contrast to a high level of sFLC revealing m-HCD associated with WM. Case 1. In December 2018, a 79 year-old man with chronic renal failure of unknown cause presented with acute renal failure (creatinine 1,160 mmol/L) associated with nephrotic syndrome (proteinuria 2,9 g/24 hours and albumin 27 g/L) leading to end-stage renal failure requiring hemodialysis and normocytic non-regenerative anemia. The blood count was as following: hemoglobin 6 g/dL, platelets 208,000/mm3, neutrophils 3,100/mm3, lymphocytes 900/mm3. Kidney biopsy showed interstitial fibrosis and tubular atrophy associated with linear Congo red-negative deposits of κ-light chains along the basement membrane suggestive of Randall-type monoclonal immunoglobulin deposition disease (MIDD). sFLC-κ were elevated at 2,585 mg/L with a κ/l ratio of 57/36. γ globulins were at 6,5 g/L and no peak was detected on SPEP but immunofixation was positive for monoclonal IgM κ (Figure 1A). Serum immuno-selection confirmed the presence of m-heavy chain (m-HC) (Figure 1B). Bone marrow aspiration and biopsy showed lympho-plasmocytic infiltration with 19% of lymphoid and plasma cells on aspiration and 25% of CD19+ CD20+ cells with monotypic expression of κ-light chain on flow cytometry. Immuno-histochemistry on biopsy identified

Figure 1. Immunological tests. On the left side, serum protein electrophoresis and immuno-fixation for patient 1 (A) and patient 2 (C) who had both monoclonal IgM κ. On the right side, immuno-electrophoresis with immuno-selection for patient 1 (B) and patient 2 (D). Black arrow indicate precipitine line consisting of m-heavy chain (continuus arrow for patients, dotted arrow for positive control)

2034

haematologica | 2021; 106(7)

Case Reports a κ monotypic population consisting of 20% of mature plasma cells CD138+ and lymphoid cells CD20+ CD79a+ CD5+ CD10– CD23–. The screening for L265P mutation of MYD88 was positive. No adenomegaly nor splenomegalia was found on whole-body computed tomography scan. Cardiac markers were increased (troponin 199 ng/L and NT-pro BNP 13,664 ng/L) and echocardiography suggested infiltrative cardiomyopathy confirmed by cardiac magnetic resonance imaging. Although imaging features could not distinguish between amyloid and Randall-type light chain deposition (LCD), endomyocardial biopsy was not performed because of histological evidence of MIDD in the kidneys. This infiltrative cardiomyopathy was attributed to probable Randall-type LCD. A treatment combining rituximab 375 mg/m² + cyclophosphamide 750 mg/m² + dexamethasone 20 mg was initiated allowing partial hematological response after seven cycles. Weekly subcutaneous injections of bortezomib 1.3 mg/m² were then added, leading to a κ/l ratio normalization after one cycle of bortezomib, so that cyclophosphamide was discontinued. At date of follow-up in October 2020, the patient had received 13 cycles of rituximab and eight courses cycles of bortezomib and had achieved a sustained complete hematological response. Case 2: In March 2019, a 74 year-old woman presented at a rheumatologic clinic for diffuse pains associated with joints swelling leading to the diagnosis of articular chondrocalcinosis. In the context of osteo-articular pains, a SPEP was performed revealing hypogammaglobulinemia at 2,7 g/L. sFLC-κ were elevated at 3,260 mg/L with a κ/l ratio of 14/61. The blood count was as following: hemoglobin 12 g/dL, platelets 374,000/mm3, neutrophils 7,660/mm3, lymphocytes 2,790/mm3. There was no hypercalcemia. Urine protein-to-creatinine ratio was at 29 mg/mmol (corresponding to 0,29 g/24 hours of proteinuria) and renal function was normal. Urine protein immuno-electrophoresis revealed Bence Jones proteinuria. β-2-microglobuline was 2,6 mg/L. Free light chain multiple myeloma was suspected and bone marrow aspiration was performed which revealed lymphocytic infiltration consisting of 54% of mature lymphocytes associated with 3% plasma cells frequently containing vacuoles. This cytologic aspect of low-grade lymphoma was suggestive of Waldenström disease. Flow cytometry on bone marrow confirmed this diagnosis with a large monoclonal κ B-cell population CD19+, CD20+, CD22+, FMC7+, CD200+, CD5–, CD23–, CD10–, CD43–, CD38–. L265P mutation of MYD88 was detected. A whole-body computed tomography scan and 18fluorodeoxyglucose-positron emission tomography scan showed no lytic bone lesion or hepatosplenomegaly or lymphadenopathy. Monoclonal IgM κ was detected on serum immunofixation (Figure 1C) and immuno-selection confirmed the presence of m-HC (Figure 1D). In the absence of clinical impact, no specific treatment was introduced other than sodium bicarbonate to prevent cast nephropathy. Few cases of γ-HCD at diagnosis or during the evolution of WM have been described8 and Wahner-Roedler et al. reported in 1992 three cases of WM among the 27 first cases of m-HCD.11 Unlike γ-HCD and a-HCD, mHCD is characterized by secretion of sFLC most often κ in the urine in one-half to two-thirds of patients with a risk of cast nephropathy or amyloidosis.4,5 However, the abnormal Ig is not detected by SPEP in two-thirds of mHCD.5 These two new cases illustrated an uncommon presentation of WM like light-chain multiple myeloma with haematologica | 2021; 106(7)

hypogammaglobulinemia, elevated sFLC and proteinuria revealing finally m-HCD. The dissociation between the sFLC level and the absence of a peak on SPEP was unusual. Moreover, the presence of vacuolated plasma cells in the bone marrow was highly suggestive of m-HCD. In order to detect heavy chains devoid of light chains on immuno-electrophoresis, immuno-selection techniques and the use of specific anti-light chains anti-serum are required. The serum samples were electrophoresed in agar containing anti-κ and anti-l antibodies trapping free lights chains and complete immunoglobulins. The throughs contained anti-m antiserum revealing mobile free µ-HC through precipitin line (Figure 1B and D). Bone marrow aspiration, immune phenotyping of B cells and the presence of monoclonal IgM on immunofixation easily clarified the diagnosis of WM, in conjunction with the MYD88 mutation. Of note, this is to our knowledge the first report of such a mutation in patients with m-HCD. The association of these two conditions raises the question of the underlying pathophysiology and may suggest a continuum between WM and m-HCD: the secretion of truncated monoclonal IgM would be secondary to alteration of immunoglobulin gene within lympho-plasmocytic cells. Unfortunately we did not have sufficient biological sample to perform DNA sequencing. sFLC are part of the monitoring of multiple myeloma especially oligo-secretory myeloma and light-chain myeloma as well as amyloid light-chain amyloidosis. It has been recently suggested that sFLC could be a reliable marker in WM for prognosis and therapeutic response10,11 but currently, the routine use is not recommended in WM. Although m-HCD is a rare condition and renal complications are even more infrequent, it could be costeffective to screen for proteinuria or even to measure sFLC and light chain proteinuria at diagnosis of lymphoplasmatic lymphoma, especially if the paraprotein is not detected on electrophoresis, because of the possible harmful renal and systemic consequences of sFLC increase. Indeed, cases of cast nephropathy12 and systemic amyloidosis13 associated with m-HCD have been reported and here we described the first case of MIDD. Patient 1 presented Randall-type LCD disease with no HCD disease. Even though both conditions, HCD disease and HCD are due to CH1 deletion, m-HC protein never causes kidney or another organ damage. This difference could be explained by the fact that in m-HCD, the CH1 deletion is associated with deletions of a variable region which seems to be involved in tissue precipitation. Indeed, it has been reported that sequence analysis of HCD disease proteins revealed amino acid substitutions in the variable region responsible for charge and hydrophobicity modifications.4 Because of the rarety of this condition, there is no prospective studies and therefore no guidelines for the management of m-HCD which is based on case reports. In asymptomatic patients such as patient 2, simple monitoring seems reasonable. For symptomatic patients, the chemotherapy targets the underlying clone as proposed in the monoclonal gammopathy of clinical significance field.15 In this report the use of rituximab associated with bortezomib and cyclophosphamide + dexamethasone allowed a complete response in patient 1. In summary, low levels of IgM protein with the presence of light chain proteinuria and high level of sFLC in WM patients are highly suggestive of m-HCD, even more if bone marrow examination reveals vacuolated plasma cells, and should alert physicians to the possibility of kidney damage. Finally, our report suggests that m-HCD associated with lymphoplasmatic proliferation and MYD 2035

Case Reports

mutation can be regarded as particular subgroup of WM. Hélène Vergneault,1 Djaouida Bengoufa,2 Aline Frazier-Mironer,3 Isabelle Brocheriou,4 Samuel Bitoun,1 Camille Villesuzanne,1 Alexis Talbot,1,5 Stéphanie Harel,1 Bertrand Arnulf1 and Bruno Royer1 1 Immuno-hematology Department, Saint-Louis Hospital, APHP; 2 Immunology Laboratory, Saint-Louis Hospital, APHP; 3Rheumatology Department, Lariboisière Hospital, APHP; 4Pathology Laboratory, La Pitié Salpêtrière Hospital, APHP and 5INSERM U976 Équipe 5, Institut de Recherche Saint Louis, Université de Paris, Paris, France Correspondence: BRUNO ROYER - bruno.royer@aphp.fr doi:10.3324/haematol.2020.277137 Received: December 16, 2020. Accepted: February 4, 2021. Pre-published: February 18, 2021. Disclosures: no confilcts of interest to disclose. Contributions: HV, BA and BR initiated the study, analyzed the data and wrote the manuscript; DB performed immunological analyzes; IB interpreted renal biopsy; HV, AT, SB, CV, BA, AFM, SH and BR took care of patients; and all authors analyzed the data, reviewed and approved the final manuscript.

References 1. Franklin EC, Lowenstein J, Bigelow B, Meltzer M. Heavy chain disease. A new disorder of serum γ-globulins: report of the first case. Am J Med. 1964;37(3):332-350. 2. Hendershot L, Bole D, Köhler G, Kearney JF. Assembly and secretion of heavy chains that do not associate posttranslationally with immunoglobulin heavy chain-binding protein. J Cell Biol. 1987; 104(3):761-767. 3. Bakhshi A, Guglielmi P, Siebenlist U, Ravetch JV, Jensen JP, Korsmeyer SJ. A DNA insertion/deletion necessitates an aberrant

2036

RNA splice accounting for a mu heavy chain disease protein. Proc Natl Acad Sci U S A. 1986;83(8):2689-2693. 4. Seligmann M, Mihaesco E, Preud’homme J-L, Danon F, Brouet J-C. Heavy chain diseases: current findings and concepts. Immunol Rev. 1979;48(1):145-167. 5. Fermand J-P, Brouet J-C. Heavy-chain diseases. Hematol Oncol Clin North Am. 1999;13(6):1281-1294. 6. Treon SP, Xu L, Yang G, et al. MYD88 L265P somatic mutation in Waldenström’s macroglobulinemia. N Engl J Med. 2012;367(9):826833. 7. Uppal NN, Monga D, Vernace MA, et al. Kidney diseases associated with Waldenström macroglobulinemia. Nephrol Dial Transplant. 2019;34(10):1644-1652. 8. Presti BC, Sciotto CG, Marsh SG. Lymphocytic lymphoma with associated γ heavy chain and IgM-γ paraproteins: an unusual biclonal gammopathy. Am J Clin Pathol.1990;93(1):137-141. 9. Wahner-Roedler DL, Kyle RA. m-heavy chain disease: presentation as a benign monoclonal gammopathy. Am J Hematol. 1992; 40(1): 56-60. 10. Leleu X, Xie W, Bagshaw M, et al. The role of serum immunoglobulin free light chain in response and orogression in Waldenstrom macroglobulinemia. Clin Cancer Res. 2011;17(9):3013-3018. 11. Itzykson R, Le Garff-Tavernier M, Katsahian S, Diemert M-C, Musset L, Leblond V. Serum-free light chain elevation is associated with a shorter time to treatment in Waldenstrom’s macroglobulinemia. Haematologica. 2008;93(5):793-794. 12. Preud’homme JL, Bauwens M, Dumont G, Goujon JM, Dreyfus B, Touchard G. Cast nephropathy in mu heavy chain disease. Clin Nephrol. 1997;48(2):118-121. 13. Kinoshita K, Yamagata T, Nozaki Y, et al. Mu-heavy chain disease associated with systemic amyloidosis. Hematol Amst Neth. 2004; 9(2):135-137. 14. Khamlichi AA, Aucouturier P, Preud’homme JL, Cogné M. Structure of abnormal heavy chains in human heavy-chain-deposition disease. Eur J Biochem. 1995;229(1):54-60. 15. Witzig TE, Wahner-Roedler DL. Heavy chain disease. Curr Treat Options Oncol. 2002;3(3):247 254.

haematologica | 2021; 106(7)

haematologica — Vol. 106 n. 7 — July 2021 —17791-2036