Haematologica, Volume 101, issue 8 by Haematologica

haematologica Journal of the European Hematology Association Published by the Ferrata Storti Foundation

Editor-in-Chief Jan Cools (Leuven)

Deputy Editor Luca Malcovati (Pavia)

Managing Director Antonio Majocchi (Pavia)

Associate Editors Hélène Cavé (Paris), Ross Levine (New York), Claire Harrison (London), Pavan Reddy (Ann Arbor), Andreas Rosenwald (Wuerzburg), Juerg Schwaller (Basel), Monika Engelhardt (Freiburg), Wyndham Wilson (Bethesda), Paul Kyrle (Vienna), Paolo Ghia (Milan), Swee Lay Thein (Bethesda), Pieter Sonneveld (Rotterdam)

Assistant Editors Anne Freckleton (English Editor), Cristiana Pascutto (Statistical Consultant), Rachel Stenner (English Editor), Kate O’Donohoe (English Editor)

Editorial Board Omar I. Abdel-Wahab (New York); Jeremy Abramson (Boston); Paolo Arosio (Brescia); Raphael Bejar (San Diego); Erik Berntorp (Malmö); Dominique Bonnet (London); Jean-Pierre Bourquin (Zurich); Suzanne Cannegieter (Leiden); Francisco Cervantes (Barcelona); Nicholas Chiorazzi (Manhasset); Oliver Cornely (Köln); Michel Delforge (Leuven); Ruud Delwel (Rotterdam); Meletios A. Dimopoulos (Athens); Inderjeet Dokal (London); Hervé Dombret (Paris); Peter Dreger (Hamburg); Martin Dreyling (München); Kieron Dunleavy (Bethesda); Dimitar Efremov (Rome); Sabine Eichinger (Vienna); Jean Feuillard (Limoges); Carlo Gambacorti-Passerini (Monza); Guillermo Garcia Manero (Houston); Christian Geisler (Copenhagen); Piero Giordano (Leiden); Christian Gisselbrecht (Paris); Andreas Greinacher (Greifswals); Hildegard Greinix (Vienna); Paolo Gresele (Perugia); Thomas M. Habermann (Rochester); Claudia Haferlach (München); Oliver Hantschel (Lausanne); Christine Harrison (Southampton); Brian Huntly (Cambridge); Ulrich Jaeger (Vienna); Elaine Jaffe (Bethesda); Arnon Kater (Amsterdam); Gregory Kato (Pittsburg); Christoph Klein (Munich); Steven Knapper (Cardiff); Seiji Kojima (Nagoya); John Koreth (Boston); Robert Kralovics (Vienna); Ralf Küppers (Essen); Ola Landgren (New York); Peter Lenting (Le Kremlin-Bicetre); Per Ljungman (Stockholm); Francesco Lo Coco (Rome); Henk M. Lokhorst (Utrecht); John Mascarenhas (New York); Maria-Victoria Mateos (Salamanca); Simon Mendez-Ferrer (Madrid); Giampaolo Merlini (Pavia); Anna Rita Migliaccio (New York); Mohamad Mohty (Nantes); Martina Muckenthaler (Heidelberg); Ann Mullally (Boston); Stephen Mulligan (Sydney); German Ott (Stuttgart); Jakob Passweg (Basel); Melanie Percy (Ireland); Rob Pieters (Rotterdam); Stefano Pileri (Milan); Miguel Piris (Madrid); Andreas Reiter (Mannheim); Jose-Maria Ribera (Barcelona); Stefano Rivella (New York); Francesco Rodeghiero (Vicenza); Richard Rosenquist (Uppsala); Simon Rule (Plymouth); Claudia Scholl (Heidelberg); Martin Schrappe (Kiel); Radek C. Skoda (Basel); Gérard Socié (Paris); Kostas Stamatopoulos (Thessaloniki); David P. Steensma (Rochester); Martin H. Steinberg (Boston); Ali Taher (Beirut); Evangelos Terpos (Athens); Takanori Teshima (Sapporo); Pieter Van Vlierberghe (Gent); Alessandro M. Vannucchi (Firenze); George Vassiliou (Cambridge); Edo Vellenga (Groningen); Umberto Vitolo (Torino); Guenter Weiss (Innsbruck).

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 European Hematology Association Published by 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 2016 are as following: Print edition

Institutional Euro 500

Personal Euro 150

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. Edoardo Ascari; Autorizzazione del Tribunale di Pavia n. 63 del 5 marzo 1955. Printing: Tipografia PI-ME, via Vigentina 136, Pavia, Italy. Printed in July 2016.

haematologica calendar of events

Journal of the European Hematology Association Published by the Ferrata Storti Foundation

EHA-SAH Hematology Tutorial on Thrombosis, Hemostasis & Myeloid Malignancies Chairs: R Foà, J Korin, G Kusminsky August 27-28, 2016 Buenos Aires, Argentina

EHA Scientific Conference on Bleeding Disorders Scientific Program Committee: C Balduini (Chair), A Falanga (Chair), F Rodeghiero, I Pabinger, M Makris September 14-17, 2016 Barcelona, Spain

2nd International Conference on New Concepts in B-Cell Malignancies European School of Haematology (ESH) Chairs: M Hallek, L Staudt, S Stilgenbauer, A ThomasTikhonenko September 9-11, 2016 Estoril, Portugal

12th Edition of the Educational Course of the EBMT Lymphoma Working Party on Treatment of Malignant Lymphoma: State-of-the Art and the Role of Stem Cell Transplantation European Group for Blood and Marrow Transplantation (EBMT) Chairs: S Montoto, P Dreger, A Sureda, E Vandenberghe September 21-23, 2016 Dublin, Ireland

10 Hodgkin Symposium University hospital of Cologne Chairs: A Engert, B von Treskow, B Böll October 22-25, 2016 Cologne, Germany th

Highlights of Past EHA - HOPE Dubai 2016 Chairs: R Foà, M Qari November 24-26, 2016 Dubai, UAE

EHA Scientific Meeting on Anemia Diagnosis and Treatment in the Omics Era Chair: A Iolascon February 2-4, 2017 Barcelona, Spain

EHA Hematology Tutorial on Lymphoid malignancies, Multiple myeloma and Bone Marrow Failure February 23-24, 2017 Colombo, Sri Lanka

EHA Hematology Tutorial on Lymphoid Malignancies March 17-18, 2017 Warsaw, Poland

EHA Scientific Meeting on Advances in Biology and Treatment of B Cell Malignancies with a Focus on Rare Lymphoma Subtypes Chairs: MJ Kersten and M Dreyling March 10-12, 2017 Barcelona, Spain

EHA Scientific Meeting on Aging and Hematology Chair: D Bron May 4-6, 2017 Location: TBC

22nd Congress of the European Hematology Association European Hematology Association June 22 - 25, 2017 Madrid, Spain

EHA Scientific Meeting on Challenges in the Diagnosis and Management of Myeloproliferative Neoplasms Chairs: JJ Kiladjian and C Harrison October 12-14, 2017 Location: TBC

EHA Scientific Meeting on Shaping the Future of Mesenchymal Stromal Cells Therapy Chair: W Fibbe November 23-25, 2017 Location: TBC

Calendar of Events updated on June 30, 2016

haematologica Journal of the European Hematology Association Published by the Ferrata Storti Foundation

Table of Contents Volume 101, Issue 8: August 2016 Cover Figure Megakaryocyte extending cytoplasmic projections to release platelets (Image created by www.somersault1824.com)

Editorials 891

Recent frustration and innovation in myelodysplastic syndrome Uwe Platzbecker and Pierre Fenaux

893

Innovative approach to older patients with malignant hemopathies Dominique Bron, et al.

Review Articles 896

The contribution of mouse models to the understanding of constitutional thrombocytopenia Catherine Léon, et al.

Articles Red Cell Biology & Its Disorders

909

Functional characterization of novel ABCB6 mutations and their clinical implications in familial pseudohyperkalemia Immacolata Andolfo, et al.

Myelodysplastic Syndromes

918

A randomized phase II trial of azacitidine +/- epoetin-β in lower-risk myelodysplastic syndromes resistant to erythropoietic stimulating agents Sylvain Thépot, et al.

Myeloproliferative Disorders

926

Antiplatelet therapy versus observation in low-risk essential thrombocythemia with a CALR mutation Alberto Alvarez-Larrán, et al.

Acute Myeloid Leukemia

932

Targeted positron emission tomography imaging of CXCR4 expression in patients with acute myeloid leukemia Peter Herhaus, et al.

Acute Lymphoblastic Leukemia

941

RNA sequencing unravels the genetics of refractory/relapsed T-cell acute lymphoblastic leukemia. Prognostic and therapeutic implications Valentina Gianfelici, et al.

Haematologica 2016; vol. 101 no. 8 - August 2016 http://www.haematologica.org/

haematologica Journal of the European Hematology Association Published by the Ferrata Storti Foundation Acute Lymphoblastic Leukemia

951

Deletions of the long arm of chromosome 5 define subgroups of T-cell acute lymphoblastic leukemia Roberta La Starza, et al.

Chronic Lymphocytic Leukemia

959

Different spectra of recurrent gene mutations in subsets of chronic lymphocytic leukemia harboring stereotyped B-cell receptors Lesley-Ann Sutton, et al.

Hodgkin Lymphoma

968

ENGAGE- 501: phase II study of entinostat (SNDX-275) in relapsed and refractory Hodgkin lymphoma Connie Lee Batlevi, et al.

Non-Hodgkin Lymphoma

976

Immune-checkpoint expression in Epstein-Barr virus positive and negative plasmablastic lymphoma: a clinical and pathological study in 82 patients Camille Laurent, et al.

Cell Therapy & Immunotherapy

985

Mesenchymal stromal cells from pooled mononuclear cells of multiple bone marrow donors as rescue therapy in pediatric severe steroid-refractory graft-versus-host disease: a multicenter survey Zyrafete KuĂ§i, et al.

Errata corrige 995

Treatment of relapsed and refractory multiple myeloma Pieter Sonneveld and Annemiek Broijl

Letters to the Editor Letters are available online only at www.haematologica.org/content/101/8.toc

e320

Immunohistochemical pattern of p53 is a measure of TP53 mutation burden and adverse clinical outcome in myelodysplastic syndromes and secondary acute myeloid leukemia Kathy L. McGraw et al. http://www.haematologica.org/content/101/8/e320

e324

Frontline therapy with high-dose imatinib versus second generation tyrosine kinase inhibitor in patients with chronic-phase chronic myeloid leukemia â&#x20AC;&#x201C; a propensity score analysis Koji Sasaki, et al. http://www.haematologica.org/content/101/8/e324

e328

Identification of a germline F692L drug resistance variant in cis with Flt3-internal tandem duplication in knock-in mice Oliver M. Dovey, et al. http://www.haematologica.org/content/101/8/e328

Haematologica 2016; vol. 101 no. 8 - August 2016 http://www.haematologica.org/

haematologica Journal of the European Hematology Association Published by the Ferrata Storti Foundation

e332

Imatinib-induced long-term remission in a relapsed RCSD1-ABL1-positive acute lymphoblastic leukemia Thomas Perwein, et al. http://www.haematologica.org/content/101/8/e332

e336

C-MYC is related to GATA3 expression and associated with poor prognosis in nodal peripheral T-cell lymphomas Rebeca Manso, et al. http://www.haematologica.org/content/101/8/e336

e339

Sepantronium bromide (YM155) improves daratumumab-mediated cellular lysis of multiple myeloma cells by abrogation of bone marrow stromal cell-induced resistance Sanne J. de Haart, et al. http://www.haematologica.org/content/101/8/e339

e344

Organ siderosis and hemophagocytosis during acute graft-versus-host disease Axel Nogai, et al. http://www.haematologica.org/content/101/8/e344

e348

Umbilical cord blood transplantation is a suitable option for consolidation of acute myeloid leukemia with FLT3-ITD Craig E. Eckfeldt, et al. http://www.haematologica.org/content/101/8/e348

e352

Unmanipulated haploidentical versus matched unrelated donor allogeneic stem cell transplantation in adult patients with acute myelogenous leukemia in first remission: a retrospective pair-matched comparative study of the Beijing approach with the EBMT database Yuqian Sun, et al. http://www.haematologica.org/content/101/8/e352

Haematologica 2016; vol. 101 no. 8 - August 2016 http://www.haematologica.org/

EDITORIALS Recent frustration and innovation in myelodysplastic syndrome Uwe Platzbecker1 and Pierre Fenaux2 1

Universitätsklinikum “Carl Gustav Carus“, Medizinische Klinik I, Dresden, Germany; and 2Service d’Hématologie Seniors, Hopital Saint Louis, Assistance Publique-Hôpitaux de Paris et Université Paris 7, Paris, France E-mail: uwe.platzbecker@uniklinikum-dresden.de doi:10.3324/haematol.2016.142836

The definition of innovation and what it could mean for MDS The term "innovation" in general means something completely original and more effective and, as a consequence, new, that "breaks into" a current system. In general terms, we would all agree that smartphones belong to recent innovations in human life. Concurrently, the field of hematology and oncology has been moving in a very dynamic fashion during the last decade. This is also true for myelodysplastic syndrome (MDS), but mainly in the field of pathophysiology and prognosis. In fact, since the introduction of azacitidine onto the market in 2009, there has been no registration of additional drugs for our patients with MDS. On the other hand, recent innovation in the MDS field involves a better understanding of how MDS might develop from an aging hematopoietic stem cell. This includes the characterization of the role and function of a complex network of molecular abnormalities which do not occur exclusively in MDS, but also in other hematological diseases. Herein, we intend to provide a short overview on recent innovations, but also on the challenges and frustrations within the field of MDS, especially regarding disease-specific therapies.

The new WHO classification A morphological assessment of the blood and bone marrow and standard metaphase cytogenetics are still the standard of care in the diagnosis of MDS.1 The current World Health Organization (WHO) defined criteria allow for a distinction between pure refractory anemia and refractory anemia with ringed sideroblasts, and patients with multi-lineage dysplasia and those with excess blasts of up to 20% in the bone marrow. Moreover, it transfers patients with an isolated deletion 5q into a separate category. This definition has been valuable as for the first time a genetic marker lesion also constituted a target for biological therapy (lenalidomide). A revision of the WHO classification has been recently published where some issues have been addressed.2 This includes the fact that patients with MDS and del(5q) can harbor one additional abnormality (e.g. +8) apart from chromosome 7, because only the latter confers a poor prognosis in these patients. Additionally, the term “refractory anemia” will be abandoned and replaced with "MDS with single lineage dysplasia" (formerly RA/RARS/RT/RN) or "MDS with excess blasts" (formerly RAEB). This is because patients with MDS often have anemia, but not exclusively in the majority of cases where other lineages are affected. However, we believe and anticipate that the WHO classification will be amended in the near future because of the recent advances in molecular data.

Molecular markers and clonal hematopoiesis In fact, important developments in molecular technologies have recently led to significant strides in the understanding of the potential molecular pathogenesis of MDS. Analyses of large non-MDS populations without cytopenias have even revealed that somatic mutations in hematopoietic cells can be acquired during a human lifetime, and are seen in >10% of people over 70 years of age.3 The most frequently mutated genes are haematologica | 2016; 101(8)

DNMT3A, ASXL1 and TET-2, and those healthy individuals generally carry only one mutation. The presence of clonal hematopoiesis (in the absence of cytopenia) is associated with an increased risk of subsequent myeloid (MDS and AML) but also lymphoid malignancies, in addition to an increased mortality risk from several other causes (especially cardiovascular). However, most individuals with age-associated clonal hematopoiesis will never develop MDS. Therefore, this clinical state has been recently defined as "clonal hematopoiesis of indeterminate potential" (CHIP), but it is still unclear if it is analogous to monoclonal gammopathy of undetermined significance (MGUS) and clonal lymphocytes of unknown origin (CLUS),4 because it does not only specifically increase the risk of one disorder, like MGUS or CLUS. In proven MDS cases, most of the recently discovered mutations (e.g. RUNX1, ASXL1, TP53) have a negative impact on prognosis, while SF3B1 has a positive impact on prognosis. Mutations may add prognostic value to existing scoring systems (IPSS, IPSS-R) for different types of treatment,5 and may soon be incorporated into the revised IPSS (IPSS-R), thus forming the “IPSS-Rm”. Nevertheless, the value of these molecular testing systems other than for diagnostic purposes (to confirm a suspected clonal disease) in daily practice, especially for treatment choice, remains debatable in the absence of large prospective studies.

Old and novel drugs and a European network on clinical trials (EMSCO) Treatment with ESAs (i.e. recombinant erythropoietin (EPO) or darbepoetin) can induce erythroid responses in patients with lower-risk MDS (LR-MDS). Although several trials, including small phase III studies, have been performed with ESAs, and despite the fact that they are widely used and accepted in the medical community, as of yet no ESA is currently approved by health agencies (EMA, FDA) for the treatment of MDS. However, two prospective placebo-controlled randomized trials (clinicaltrials.gov identifier: 01381809 and clinicaltrials.gov identifier: 01362140) have recently been completed and the results are expected soon. At least one of these studies may lead to the registration of this class of drugs. Nevertheless, response to an ESA is generally transient, and therapeutic options are needed for LR-MDS patients with anemia not responding to or relapsing after a response. Lenalidomide is a potentially active drug in patients with non-del(5q) MDS, with erythroid responses seen in a quarter of unselected patients.6 This phase II study by Raza et al. was the rationale to further investigate lenalidomide in non-del(5q) patients with anemia refractory to ESAs in a phase III placebocontrolled study.7 Overall, 27% of patients treated with lenalidomide achieved transfusion independency ≥8 weeks compared with 2.5% of placebo-treated patients (P<0.001). The median duration of response was 8.2 months (range 5.2 to 17.8 months). The main adverse events were grade 3–4 neutropenia and thrombocytopenia, although the frequency of these events was lower in non-del(5q) patients (49.3% vs. 73.7% for neu891

Editorials

tropenia and 37.3% vs. 64.2% for thrombocytopenia, respectively).8 Recently, in a randomized trial, Toma et al. showed that in LR-MDS patients with ESA resistant anemia, the combination of LEN and EPO significantly improved erythroid response compared to LEN alone.9 Activin receptor inhibitors antagonizing TGFβ and SMAD signaling currently arise as promising targets to treat anemia in MDS patients. One of these emerging compounds is ACE 011 (sotatercept), an ActRIIA ligand trap consisting of the extracellular domain of hActRIIA linked to the hIgG1 Fc domain. Originally, ACE 011 was developed to increase bone mineral density in bone diseases. Data from a phase II trial with ACE 011 in patients with anemia refractory to ESA and low- and and intermediate-1-risk MDS have shown considerable responses, especially in patients with a low transfusion burden (clinicaltrials.gov identifier: 01736683).10 A parent compound is ACE 536 (luspatercept), which consists of a modified extracellular domain of hActRIIB fused to the Fc domain of hIgG1. In vivo studies with a mouse analog, RAP-536, showed a rapid and robust dose-dependent increase in hematocrit, hemoglobin, red blood cell and reticulocyte counts, and was hence able to both reduce and prevent anemia in the NUP98-HOXD13 MDS mouse model for over seven months.11 SMAD2/3 activation was reduced and erythroid hyperplasia and ineffective erythropoiesis were strikingly corrected. A phase I study of ACE 536 in healthy postmenopausal women demonstrated a sustained increase in hemoglobin levels, beginning seven days after the initiation of treatment, and which could be maintained for several weeks.12 Clinical phase II studies evaluating the erythroid response are ongoing in anemic low or intermediate-1 MDS patients who are treatment naive for hypomethylating agents (clinicaltrials.gov identifier: 01749514). Patients receive the drug SC every three weeks for up to five cycles. Preliminary data already suggest clinical activity, with the majority of patients demonstrating increased hemoglobin levels and/or decreased transfusion requirement accompanied by a favorable safety profile. Higher response rates were observed in patients with ring sideroblasts and/or SF3B1 mutations.13 As a result, a placebo-controlled registration trial (Medalist) in patients with either RARS or RCMDRS has recently been started (clinicaltrials.gov identifier: 02631070). The results of the study mentioned above show that the heterogeneity and diversity of MDS is about to increase further. Strong clinical networks are required in the future to allow a fast and efficient recruitment of selected subgroups of patients into clinical trials. It has been an ambition of the ELN and the EHA SWG since 2012 to start a European initiative on clinical trials (EMSCO) in order to foster academic-driven clinical research in the field of MDS. Several clinical trials have been launched or promoted by this platform (SINTRA, EUROPE, DACOTA), and more are about to come. The experience so far has demonstrated the great heterogeneity in regulations among different countries in the EU. It is hoped that the new EU regulations (EU 536/2014), which are anticipated to be in place in 2018, will lead to a harmonization within Europe.

Mode of action of hypomethylating agents Azacitidine, and to a lesser extent decitabine, have become the standard therapeutic approach for older patients with higher-risk disease. Responses are observed in roughly 892

40 to 50% of patients, including “classical” complete remission as seen in AML with conventional chemotherapy. The mode of action of HMAs has not been completely understood, but is thought to involve both a classical cytoreductive (“chemo-like”) effect and also a so called “epigenetic” modulation of hematopoiesis. Recently, a study showed that clinical responses can be achieved independently of changes in the molecular burden of patients with CMML.14 By combining serial whole-exome and whole-genome sequencing, Merlevede et al. showed that the response to a HMA is associated with changes in gene expression and DNA methylation, without any decrease in the mutation allele burden nor prevention of the occurrence of new genetic alterations. This is an important observation highlighting the epigenetic effects of HMAs, which might differ depending on the disease type as well as molecular background. Therefore, HMAs seem to be able to restore a disturbed hematopoiesis without affecting the size of the mutated clone. This observation mirrors clinical experience where therapy with HMAs is maintained even in the absence of classical CR or PR, but in patients achieving a hematological improvement only.

Therapeutic options for hypomethylating agents failure in patients Although HMAs are active in many higher-risk MDS patients, the majority do not respond or lose response after an initial response. The subsequent outcome is poor with a median survival of less than 6 months, no drug having demonstrated a survival advantage at this stage, while allogeneic HSCT remains the only potentially curative option for a small subset of medically fit patients. Other patients should be offered clinical trials testing new drugs. The first randomized phase III study in that area compared standard of care (mostly supportive care only) with rigosertib, a cellcycle inhibitor, which induces mitotic arrest in vitro in leukemic cells by inhibiting several kinases, including PLKs. Small phase II studies suggested that the drug has cytoreductive activity in the bone marrow of MDS patients with advanced disease. Unfortunately, the primary endpoint of this phase III trial, i.e. demonstrating a survival advantage with rigosertib, was not met (with a median survival of 8.2 versus 5.9 months, respectively).15 A subgroup analysis identified that patients for whom HMAs fail within the first 9 months of therapy and those with unfavorable karyotypes seemed to benefit most from this drug. Therefore, a new trial with rigosertib in this subgroup of higher-risk MDS patients has recently been launched (clinicaltrials.gov identifier: 02562443).

Response criteria in MDS The heterogeneity of MDS has challenged the evaluation of response to a given treatment. In 2006, an International Working Group (IWG) proposed a revision of standardized response criteria (IWG 2000) to evaluate clinical responses in MDS. The IWG 2006 criteria have been used in many clinical trials and served as a valuable tool for the standardization of clinically meaningful response measures in MDS. Recent clinical experiences, however, have shown that there are still some pitfalls when adopting these criteria in clinical practice, which can lead to the misinterpretation of outcome, especially with regards to erythroid response. Therefore, the ELN and EHA MDS groups, together with EMSCO, have haematologica | 2016; 101(8)

Editorials

started an initiative to revise these criteria. We believe that the modifications (IWG 2016) will be crucial and will allow for more individualized pre- and on-study assessment and, therefore, provide the MDS community with an improved tool in terms of response evaluation.

Summary

MDS is a moving target with maximum innovation in the understanding of the complex molecular pathways during the last decade. Compared to other “chronic” hematological malignancies like myeloma or CLL this has, unfortunately, not yet been translated into novel treatment options. Given the actual developments in the field, we are optimistic that recent frustrations will be overcome and that new treatment opportunities will soon be available for our patients.

References 1. Malcovati L, Hellström-Lindberg E, Bowen D, et al. Diagnosis and treatment of primary myelodysplastic syndromes in adults: recommendations from the European LeukemiaNet. Blood. 2013;122(17): 29432964. 2. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revision to the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia. Blood. 2016 Apr 11. [Epub ahead of print] 3. Genovese G, Jaiswal S, Ebert BL, McCarroll SA. Clonal hematopoiesis and blood-cancer risk. N Engl J Med. 2015;372(11):1071–1072. 4. 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. 5. Platzbecker U, Fenaux P. Personalized medicine in myelodysplastic syndromes: wishful thinking or already clinical reality? Haematologica. 2015;100(5):568–571. 6. Raza A, Reeves JA, Feldman EJ, et al. Phase 2 study of lenalidomide in

10.

11. 12. 13.

14. 15.

transfusion-dependent, low-risk, and intermediate-1 risk myelodysplastic syndromes with karyotypes other than deletion 5q. Blood. 2008;111(1):86–93. Santini V, Almeida A, Giagounidis A, et al. Efficacy and Safety of Lenalidomide (LEN) Versus Placebo (PBO) in RBC-Transfusion Dependent (TD) Patients (Pts) with IPSS Low/Intermediate (Int-1)-Risk Myelodysplastic Syndromes (MDS) without Del(5q) and Unresponsive or Refractory to Erythropoiesis-Stimul. Blood. 2014;124(21):409 abstr. Almeida A, Santini V, Gröpper S, et al. Safety of Lenalidomide (LEN) 10mg in Non-Del(5q) Versus Del(5q) in the Treatment of Patients (Pts) with Lower-Risk Myelodysplastic Syndromes (MDS): Pooled Analysis of Treatment-Emergent Adverse Events (TEAEs). Blood. 2015;126(23): 2880 abstr. Toma A, Kosmider O, Chevret S, et al. Lenalidomide with or without erythropoietin in transfusion-dependent erythropoiesis-stimulating agent-refractory lower-risk MDS without 5q deletion. Leukemia. 2016;30(4):897–905. Komrokji RS, Garcia- Manero G, Ades L, et al. An Open-Label, Phase 2, Dose-Finding Study of Sotatercept (ACE-011) in Patients with Low or Intermediate-1 (Int-1)-Risk Myelodysplastic Syndromes (MDS) or Non-Proliferative Chronic Myelomonocytic Leukemia (CMML) and Anemia Requiring Transfusion. Blood. 2014;124(21):3251 abstr. Suragani RN, Cadena SM, Cawley SM, et al. Transforming growth factor-β superfamily ligand trap ACE-536 corrects anemia by promoting late-stage erythropoiesis. Nat Med. 2014;20(4):408–414. Attie KM, Allison MJ, McClure T, et al. A phase 1 study of ACE-536, a regulator of erythroid differentiation, in healthy volunteers. Am J Hematol. 2014;89(7):766–770. Platzbecker U, Germing U, Giagounidis A, et al. Biomarkers of Ineffective Erythropoiesis Predict Response to Luspatercept in Patients with Low or Intermediate-1 Risk Myelodysplastic Syndromes (MDS): Final Results from the Phase 2 PACE-MDS Study. Blood. 2015;126(23):2862 abstr. Merlevede J, Droin N, Qin T, et al. Mutation allele burden remains unchanged in chronic myelomonocytic leukaemia responding to hypomethylating agents. Nat Commun. 2016;7:10767. Garcia-Manero G, Fenaux P, Al-Kali A, et al. Rigosertib versus best supportive care for patients with high-risk myelodysplastic syndromes after failure of hypomethylating drugs (ONTIME): a randomised, controlled, phase 3 trial. Lancet Oncol. 2016 Mar 8. [Epub ahead of print]

Innovative approach to older patients with malignant hemopathies Dominique Bron,1 Pierre Soubeyran,2 and Tamas Fulop,3 on behalf of the SWG “Aging and Hematology” of the EHA 1

Department of Clinical and Experimental Hematology, Institut Jules Bordet (ULB), Brussels, Belgium; 2Hematological Research Group, Bergonié Cancer Institut, Bordeaux, France; and 3Department Geriatry, Centre Hospitalier Universitaire du Sherbrooke, Quebec, Canada. E-mail: dbron@ulb.ac.be doi:10.3324/haematol.2016.142810

Introduction Aging represents a significant health problem since nobody can escape this natural process. Though not a disease per se, aging progressively leads to organ dysfunctions and represents a major risk factor for most cancers and diseases. Indeed, with the aging of the population, a 50% increase in new cancer cases is expected over the next 20 years. Since adult stem cells are responsible for maintaining tissue homeostasis, an attractive theory is that age-related degenerative changes may be due to alterations in tissue stem cells, particularly in the hematopoietic stem cells (HSCs). Extensive research is currently underway; it demonstrates a progressive waning in our immune defenses and concomitantly, genetic and epigenetic modifications of the hematopoietic stem cells and their microenvironment. In addition, older patients of a similar age are an extremely heterogeneous population in terms of fitness. Thus, chronological age does not adequately guide clinicians in choosing their treatment. haematologica | 2016; 101(8)

A better understanding of the cellular and molecular changes involved in the aging process, combined with a better assessment of the “fitness” status of older patients, will definitely help optimize and personalize therapeutic approaches in this older population in order to achieve the primary objective: healthy aging and not only prolonged survival.

Assessment of Immunosenescence Cellular “senescence” refers to the specific phenomenon wherein a proportion of competent cells undergoes permanent growth arrest in response to various cellular stresses, translating in a replicative limit in culture, while being metabolically very active. The definition of “immunosenescence” is still a controversial issue, but is commonly accepted as the decrease in immune function associated with aging; it combines immune deficiencies (changes in innate immune functions, shrinking of naïve T- and B-cell compartments, reduced Tand B-cell receptor diversity, decreased T-cell receptor sensi893

Editorials

tivity to stimuli) and an age-related pro-inflammatory state (excess production of inflammatory cytokines such as IL-6 and TNF, the production of autoantibodies). This leads to an increased sensitivity to infections, autoimmune disorders, chronic inflammatory diseases and cancer development.1-2 Due to their impaired immune defenses, older cancer patients are more vulnerable to life-threatening side effects of hematotoxic and immunosuppressive drugs. A comprehensive care program, including vaccinations, nutritional supplements, primary prophylaxis with granulocyte colony-stimulating factors and IV immunoglobulins, if required, constitutes the current recommended approach to this population.3

Table 1.

Genetic and epigenetic changes in HSCs

Assessment of “physiological” age

The functional decline in hematopoiesis in the elderly, which involves a progressive reduction in the immune response and an increased incidence of malignancies, is partly linked to HSC aging. Understanding the molecular processes controlling hematopoietic stem cell survival, selfrenewal and commitment to specific differentiated cell lineages is indeed crucial to determine the drivers and effectors of age-associated stem cell dysfunction, which remain poorly elucidated to this day. The aging phenotype is partly explained by damages in DNA integrity resulting in poor DNA repair, telomere shortening, chromosomal instability, altered intercellular communication and senescent environment, and loss of apoptosis-regulating genes. Moreover, recent observations suggest that small changes in the epigenetic landscape can lead to significant alterations in the expression patterns (either directly by loss of regulatory control, or through indirect additive effects, ultimately leading to transcriptional changes of the stem cells). These changes can also play a key role in modulating the functional potential of HSCs. The two best characterized epigenetic changes are DNA methylation and histone modifications. However, non-coding RNAs could also play a role in regulating HSC function in aging.4 The aging of HSCs has long been thought to be an intrinsic irreversible process. Mouse model studies have shown that aging is associated with elevated activity of the Rho GTPase Cdc42 in HSCs which causes loss of polarity. This results in a symmetric distribution of epigenetic markers that is responsible for functional deficits of aged HSCs, whereas in dividing young HSCs, distribution is mainly asymmetric. This work suggests that the inhibition of Cdc42 activity in aged HSCs may reverse a number of phenotypes associated with HSC aging. These findings support the hypothesis that the functional decline of aged HSCs may be reversed by pharmacological intervention of agealtered signaling pathways and epigenetic modifications.5-6 Such restorative interventions hold promise for the treatment of many diseases, including sarcopenia, heart failure and neurodegeneration. Besides the molecular mechanisms associated with the aging of hematopoietic stem cells, poor homing capacity and the aging of stem cell niches are currently being further investigated.7 Such knowledge will be essential to develop therapies to slow, and perhaps reverse, age-related degenerative changes and to enhance the regenerative capacity of organs, thus favoring healthy aging.8

The older population with cancer is a heterogeneous cohort in terms of physical performance, physiological functions, psycho-cognitive functions and socio-economic environment. Chronological age does not adequately guide physicians in proposing optimal therapeutic approaches. In contrast with younger populations, the management of these older patients deserves a multi-step procedure: besides the accurate assessment of the tumor’s prognosis and the patient’s risk of dying from it, clinicians have to take into account the biological reserves, the patient’s life expectancy and their capacity to tolerate the treatment. Additionally, the patient’s wishes and their capacity to understand the therapeutic approach should be fully integrated in the geriatric assessment.9 A modern approach thus consists in the assessment of the patient’s physiological age.10 In this setting, geriatricians are essential collaborators, proposing various tools to evaluate physical performance (PS, Up and Go test, ADL, IADL, etc.), physiological status (Comorbidity index, polypharmacy, nutritional status, etc.), psycho-cognitive functions (GDS, MMSE, MOCA, etc.) and socio-economic environment (income, caregivers, etc.). A test which could be used to evaluate the dynamic physiological reserve would be a helpful tool in this approach. Although clinicians can reliably evaluate physical fitness, it has now been demonstrated that depression and cognitive impairment are completely underestimated and the socioeconomic environment is also poorly explored. Yet, poor cognitive functions and a disadvantaged socio-economic environment are correlated with worse survival, and deserve specific attention in “clinically fit” patients. Thus, geriatric assessment not only helps to identify older patients with a higher risk of morbidity/mortality, but also allows for better management of their vulnerabilities.11-12 However, such a comprehensive geriatric assessment is not applicable on a routine basis outside centers with oncogeriatric nurses.13 Additional simple tools are still needed to further assess the risk/benefit ratio for a specific patient receiving a specific anticancer therapy that could potentially compromise their long-term functionality and quality of life.14 Attention should be drawn to very old patients. Few reports and even fewer randomized trials are published in this population which, despite a significant reduction in treatment posology, experiences early life-threatening grade 3/4 toxicities. A prephase treatment has been demonstrated to significantly reduce the first chemotherapy cycle’s toxicity, and is now recommended in frail patients suffering from diffuse large B-cell lymphoma.15-16

894

TAKE HOME MESSAGES Older patients are an extremely heterogeneous population requiring a deep and multidimensional evaluation of physiological reserves Unsuspected cognitive impairment deserves specific attention because of its significant impact on survival Geriatric assessment should take into account the reversible (disease-related) character of the complaints and wishes of the patient Reliable biomarkers of frailty are urgently needed

haematologica | 2016; 101(8)

Editorials

The G8 questionnaire represents a simple screening test to rapidly identify oncological patients requiring a full geriatric assessment.17-18 However, nutritional and psychological problems have a major impact on the total G8 score, and since these issues may be disease-related in patients with malignant hemopathies, some authors propose delaying this screening test in order to eliminate the bias due to these problems, which could be reversible after a few days of treatment.19 Furthermore, the increased mortality in elderly patients is not only related to their frailty and poor tolerance to chemotherapy. Indeed, oncologists tend to reduce the doses of treatment in older patients in order to avoid potentially fatal side effects such as febrile neutropenia, thereby decreasing the chances of therapeutic success. Additionally, patients and their families, fearing a loss of autonomy, will also push physicians to cut back on the doses of treatment. These patients in poor physical or psychological conditions are too often excluded from prospective studies, even though they represent the population we most often have to face in our daily practice.

Biomarkers of frailty In addition to age and diagnosis, the most frequently reported “clinical” items correlated with shortened overall survival are impaired functional and nutritional status.12 For “clinically fit” patients receiving chemotherapy, a mild cognitive impairment is correlated with worse overall survival.20-21 Besides shorter overall survival, unacceptable outcomes in the eyes of clinicians, patients and their relatives, are early toxic death, loss of autonomy and unexpected hospitalization. In recent large retrospective analyses, early toxic deaths (within 6 months of treatment) are correlated with poor nutritional status (MNA<24) and low physical performance (Up and Go test >20s).17 Loss of autonomy is correlated with psychological distress (GDS>5) and abnormal daily functioning (IADL<8), and the increased risk of hospitalization is correlated with poor nutritional status (MNA<24).22-23 Biological cellular or molecular biomarkers of frailty are still currently under investigation (CRP, IL-6, IL-10, etc.), and require validation in hematological malignancies based on a large series in the general population which tend to show their potential predictive value.24 The expression of p16 in circulating T lymphocytes, a known biomarker of senescence, not only correlates with age25 but also with chemotherapyrelated aging.26 However, despite using the best available geriatric assessment, some clinically fit patients, referred to receive full-dose chemotherapy, presented unexpected treatment-related, and sometimes life-threatening side effects, whereas some patients deemed clinically vulnerable tolerated full-dose treatment. Thus, more accurate biomarkers that dynamically test the physiological reserve are urgently needed to better identify the patients who will benefit from standard treatment.

Conclusion Although the multidisciplinary approach brings together the concerns of scientists, geriatricians, home practitioners and onco-hematologists (Table 1), the additional involvement of the patients themselves should result in optimized haematologica | 2016; 101(8)

and personalized patient care, focusing not only on overall survival, but also on improved qualitative survival.

References 1. Fulop T, Le Page A, Fortin C, Witkowski JM, Dupuis G, Larbi A. Cellular signaling in the aging immune system. Curr Opin Immunol. 2014;29:105-11. 2. Fulop T. Biological research into aging: from cells to clinic. Biogerontology. 2016;17(1):1-6. 3. Hurria C, Browner I, Cohen HJ, et al. Senior Adult Oncology. J Natl Compr Canc Netw. 2012;10(2):162-209. 4. Geiger H, Denkinger M, Schirmbeck R. Hematopoietic stem cell aging. Curr Opin Immunol. 2014;29:86-92. 5. Geiger H, de Haan G, Florian MC et al. The ageing hematopoietic stem cell compartment. Nat Rev Immunol. 2013;(13):376-389. 6. Florian MC, Dörr K, Niebel A, et al. CDC42 activity regulates hematopoietic stem cell aging and rejuvenation. Cell Stem Cell. 2012;(10):520-530. 7. Kusumb AP, Ramasamy SK, Itkin T, et al. Age-dependent modulation of vascular niches for haematopoietic stem cells. Nature 2016;532(7599): 380-384. 8. Oh J, Lee YD, Wagers AJ. Stem cell aging: mechanisms, regulators and therapeutic opportunities. Nat Med. 2014;20(8):870–880. 9. Bron D, Ades L, Fulop T, Goede V. Aging and malignant hemopathies. Haematologica. 2015;100(5):571-574. 10. Wildiers H, Heeren P, Puts M, et al. International Society of Geriatric Oncology consensus on geriatric assessment in older patients with cancer. J Clin Oncol. 2014;20;32(24):2595-2603. 11. Liu JJ, Extermann M. Comprehensive geriatric assessment and its clinical impact in oncology. Clin Geriatr Med. 2012;28(1):19-31. 12. Hamaker ME, Prins MC, Stauder R. The relevance of a geriatric assessment for elderly patients with a haematological malignancy--a systematic review. Leuk Res. 2014;38(3):275-283. 13. Kenis C, Heeren P, Bron D, et al. Multicenter implementation of geriatric assessment in Belgian patients with cancer: a survey on treating physicians' general experiences and expectations. J Geriatr Oncol. 2014;5(4):431-438. 14. Hamaker ME, Stauder R, van Munster BC. Ongoing clinical trials in elderly patients with a hematological malignancy: are we addressing the right end-points? Ann Oncol. 2014;(3):675-681. 15. Pfreundschuh M. How I treat elderly patients with diffuse large B-cell lymphoma. Blood. 2010;9;116(24):5103-5110. 16. Handforth C, Clegg A, Young C, et al. The prevalence and outcomes of frailty in older cancer patients: a systematic review. Ann Oncol. 2015;26(6):1091-1101. 17. Soubeyran P. From suboptimal to optimal treatment in older patients with cancer. J Geriatr Oncol. 2013;4(3):291-293. 18. Hamaker ME, Mitrovic M, Stauder R. The G8 screening tool detects relevant geriatric impairments and predicts survival in elderly patients with a haematological malignancy. Ann Hematol. 2014;93(6):1031-1040. 19. Petit-Monéger A, Rainfray M, Soubeyran P, Bellera CA, MathoulinPélissier S. Detection of frailty in elderly cancer patients: Improvement of the G8 screening test. J Geriatr Oncol. 2016;7(2):99-107. 20. Klepin HD, Geiger AM, Tooze JA, et al. The feasibility of inpatient geriatric assessment for older adults receiving induction chemotherapy for acute myelogenous leukemia. J Am Geriatr Soc. 2011;59(10):18371846. 21. Dubruille S, Libert Y, Maerevoet M, et al. Prognostic value of neuro-psychological and biological factors in clinically fit older patients with hematological malignancies admitted to receive chemotherapy. Geriatr Oncol. 2015;6(5):362-369. 22. Hoppe S, Rainfray M, Fonck M, et al. Functional decline in older patients with cancer receiving first-line chemotherapy. J Clin Oncol. 2013;1;31(31):3877- 3882 23. Hurria A, Togawa K, Mohile SG, et al. Predicting chemotherapy toxicity in older adults with cancer: a prospective multicenter study. J Clin Oncol. 2011;29(25):3457-3465. 24. Collerton J, Martin-Ruiz C, Davies K, et al. Frailty and the role of inflammation, immunosenescence and cellular ageing in the very old: cross-sectional findings from the Newcastle 85+study. Mech Age Dev. 2012;133:456-466. 25. Nelson JA, Krishnamurthy J, Menezes P, et al. Expression of p16(INK4a) as a biomarker of T-cell aging in HIV-infected patients prior to and during antiretroviral therapy. Aging Cell. 2012;11:916-918. 26. Sanoff HK, Deal AM, Krishnamurthy J, et al. Effect of cytotoxic chemotherapy on markers of molecular age in patients with breast cancer. J Natl Cancer Inst. 2014;106(4):dju057.

895

REVIEW ARTICLE EUROPEAN HEMATOLOGY ASSOCIATION

Ferrata Storti Foundation

The contribution of mouse models to the understanding of constitutional thrombocytopenia

Catherine Léon,1,2,3,4 Arnaud Dupuis,1,2,3,4 Christian Gachet,1,2,3,4 and François Lanza1,2,3,4

Haematologica 2016 Volume 101(8):896-908

UMR_S949, INSERM, Strasbourg; 2Etablissement Français du Sang-Alsace (EFS-Alsace), Strasbourg; 3Université de Strasbourg; and 4Fédération de Médecine Translationnelle de Strasbourg (FMTS), France

ABSTRACT

Correspondence: catherine.leon@efs.sante.fr

Received: March 16, 2016. Accepted: May 4, 2016. Pre-published: No prepublication

onstitutional thrombocytopenias result from platelet production abnormalities of hereditary origin. Long misdiagnosed and poorly studied, knowledge about these rare diseases has increased considerably over the last twenty years due to improved technology for the identification of mutations, as well as an improvement in obtaining megakaryocyte culture from patient hematopoietic stem cells. Simultaneously, the manipulation of mouse genes (transgenesis, total or conditional inactivation, introduction of point mutations, random chemical mutagenesis) have helped to generate disease models that have contributed greatly to deciphering patient clinical and laboratory features. Most of the thrombocytopenias for which the mutated genes have been identified now have a murine model counterpart. This review focuses on the contribution that these mouse models have brought to the understanding of hereditary thrombocytopenias with respect to what was known in humans. Animal models have either i) provided novel information on the molecular and cellular pathways that were missing from the patient studies; ii) improved our understanding of the mechanisms of thrombocytopoiesis; iii) been instrumental in structure-function studies of the mutated gene products; and iv) been an invaluable tool as preclinical models to test new drugs or develop gene therapies. At present, the genetic determinants of thrombocytopenia remain unknown in almost half of all cases. Currently available high-speed sequencing techniques will identify new candidate genes, which will in turn allow the generation of murine models to confirm and further study the abnormal phenotype. In a complementary manner, programs of random mutagenesis in mice should also identify new candidate genes involved in thrombocytopenia.

doi:10.3324/haematol.2015.139394

Article summary Check the online version for the most updated information on this article, online supplements, and information on authorship & disclosures: www.haematologica.org/content/101/8/896

©2016 Ferrata Storti Foundation Material published in Haematologica is covered by copyright. All rights are reserved to the Ferrata Storti Foundation. Copies of articles are allowed for personal or internal use. Permission in writing from the publisher is required for any other use.

896

• Constitutional thrombocytopenias of hereditary origin have long been poorly studied due to the difficulties of obtaining invasive marrow samples. • Genetic engineering has allowed the generation of numerous mouse models for all these pathologies, providing invaluable information to understanding these diseases and serving as preclinical models to test new therapies. Constitutional thrombocytopenias result from genetic mutations affecting platelet production. These rare diseases are still underdiagnosed, especially in adults, because they remain little-known and have a highly variable expression. Autoimmune thrombocytopenia is still often wrongly diagnosed, thereby leading to the inadequate management of patients, with occasionally inappropriate splenectomy. The diagnosis of congenital thrombocytopenia relies on cytological and functional platelet analyses performed almost exclusively in specialized laboratories. Furthermore, about 50% of thrombocytopenias, associated or not with a thrombopathy, still remain of unknown origin.1 In cases where platelet studies orient the diagnosis to a known disease, the detection of mutations in the suspected genes can haematologica | 2016; 101(8)

Mouse models of constitutional thrombocytopenia

confirm the pathology. Our understanding of the pathophysiological mechanisms leading to congenital thrombocytopenia has long been based only on the observation of the megakaryocytes present in bone marrow. The need for invasive marrow samples and the rarity of these cells (less than 1% of bone marrow cells) have for a long time hampered such studies. Although obtaining culture of megakaryocytes from circulating hematopoietic progenitors is now possible, it remains confined to research laboratories, and only a few patients have been investigated in this manner. The recent development of the genetic reprogramming of iPS cells and their megakaryocytic differentiation has enabled some progress, but these in vitro systems are imperfect and do not faithfully reproduce all the steps leading to the formation of platelets. The development of tools to genetically manipulate mice now allows us to generate in vivo models mimicking these various pathologies, enabling the assessment of the impact of mutations on platelet production and function. Targeted mutagenesis and transgenesis now offer a wide range of such models. "Total knockout mice" are generated by the inactivation of the gene in the whole organism. "Conditional knockout mice" permit inactivation of a gene in the megakaryocytic line tissue, or at a specific stage of development through the use of cre recombinase expressed under the control of the promoter of interest. The Mx-cre system has been most widely employed to excise a portion of DNA at a selected developmental stage. Almost all recombination in the megakaryocytic lineage has been obtained by using the Pf4-Cre system developed by Radek Skodaâ&#x20AC;&#x2122;s group.2,3 And finally, "knock-in mice" allow the introduction of point mutations or insertions/deletions through homologous recombination at the locus of interest. These models faithfully reproduce the mutations present in humans and represent the best approach to mimic the pathology. In addition to this genetic recombination toolkit, several chemical mutagenesis programs based on the treatment of gametes with N-nitroso-N-ethylurea followed by systematic phenotypic screening have been established to increase the frequency of mutations potentially targeting the hematopoietic system.4-6 This approach might allow one to direct the screening of novel genes in patients with unidentified congenital thrombocytopenias. Mouse models thus represent an essential tool to deepen our understanding of the mechanisms involved in platelet formation. The objective of this review is to focus on the contribution which mouse models have made to the elucidation and treatment of these diseases. We will briefly discuss the steps and key points of platelet formation, with emphasis on the roles played by proteins whose mutations are responsible for congenital thrombocytopenia. We will then describe the various constitutional thrombocytopenias where the contribution of animal models has been essential for their elucidation and/or treatment. For a more detailed description of the human pathologies, the reader may refer to three excellent reviews7-9 and the OMIM database (input numbers in Table 1).

Platelet information Megakaryocyte differentiation and maturation In humans, as in mice, blood platelets are mainly formed from multipotent hematopoietic stem cells (HSCs) located within the bone marrow. The process is complex and haematologica | 2016; 101(8)

involves a series of steps of proliferation and differentiation of progenitors and maturation of megakaryocytes. During their maturation, the megakaryocytic progenitors undergo several endomitotic cycles resulting in the formation of giant (>50 microns) and polyploid cells (up to 128N). In parallel, the cytoplasm considerably enlarges and a network of extremely complex and structured internal membranes develops. This network, called the demarcation membrane system (DMS), serves as a reserve for the membranes of long projections called proplatelets, which after release into the circulation form preplatelets, before being remodeled to give platelets (Figure 1). Cytoskeletal proteins play a key role in the extension of proplatelets and platelet formation, as in the case of the microtubule polymerization and sliding which condition the elongation of proplatelets and the generation of discoid platelets. Actomyosin also plays a crucial role in megakaryocyte differentiation and proplatelet formation. Several congenital thrombocytopenias result from mutations in genes directly or indirectly related to the cytoskeleton10 (Table 1).

Regulation of megakaryocytopoiesis Megakaryocytopoiesis is regulated through various steps controlling the proliferation of progenitor cells, their commitment and differentiation into mature megakaryocytes and the release of platelets. This involves the concerted action of a number of factors, including cytokines and growth factors, of which thrombopoietin (TPO) plays a major role, and a number of transcription factors (Figure 2). Many of these actors are the sites of mutations responsible for thrombocytopenia (Table 1). TPO acts through the MPL receptor, leading to the modulation of the expression of genes involved at different points of the process of megakaryocytic differentiation/maturation. It plays a part especially in the survival of HSCs and their differentiation and commitment to the megakaryocytic pathway. It is therefore not surprising that patients with mutations in the TPO-MPL pathway, according to whether these mutations cause loss or gain of function, develop particularly severe thrombocytopenia or thrombocytosis. Transcription factors play an important role throughout megakaryopoiesis in regulating both the commitment of progenitor cells to their lineages and the transcriptional program determining megakaryocyte maturation. Megakaryocytes derive from a bipotent megakaryocyteerythroid progenitor (MEP) through the coordinated activation of specific megakaryocytic genes and selective inactivation of transcription factors of the erythroid lineage (Figure 2), although recent data propose the existence of megakaryocyte-biased progenitors.11 Some key transcription factors mutated in congenital thrombocytopenia include: GATA1, a zinc finger protein which forms a complex with the common factor FOG1 (Friend of GATA1). A number of thrombocytopenias result from mutations in the GATA1 gene, while a point mutation in the GATA1 binding site in the promoter of the gene encoding GPIbÎ˛ (GPIBB) causes a form of Bernard-Soulier syndrome. RUNX1 (AML1), a RUNT family transcription factor which acts with its cofactor CBFB (core-binding factor, Î˛ subunit), and plays an important part in megakaryocytic differentiation through its role in regulating the expression of cytoskeletal proteins and platelet components. Some mutations in RUNX1 lead to thrombocytopenia associated with a high risk of developing leukemia and myelodys897

C. Léon et al. Table 1.

Disease (abbreviation)

Transmission #N°OMIM

Anomalies of the cytoskeleton MYH9 related AD diseases (MYH9-RD) #600208 #155100 #152640 #605249 #153650 Bernard-Soulier #231200 syndrome (BSS) - Biallelic (AR) AR -Monoallelic (AD) AD

Gene (chromosome)

Human Symptoms (platelet number±SD) (x109/L)

Mouse models

Mouse phenotype

MYH9 (22q12-13)

Giant platelets (35±25.9) Leukocyte inclusions and/or nephropathy, cataract, hearing loss

1. Total inactivation13,14 2. Tissue-specific inactivation 15 3. Knock-in 20,112

1. Embryonic lethal (E7.5) 2. Defective megakaryocytes/ macrothrombocytopenia 3. Recapitulate human phenotype

Giant platelets (41±34.6) Large platelets (87±30.2)

1. Total inactivation Gp1ba23 2. Total inactivation Gp1bb24,25 3. Knock-in Gp1ba with IL4R extracellular part28

1. Defective proplatelet formation and 2. macrothrombocytopenia 3. Partial recovery of the WT phenotype 1. Thrombocytopenia with normal sized platelets 2. Mimic most features of complete WASp-deficiency Macrothrombocytopenia Increased platelet clearance Macrothrombocytopenia with defective marginal band -

GP1BA (17p13) GP1BB (22q11) GP9 (3q21)

Wiskott-Aldrich syndrome (WAS) (THC1)

X-linked #301000 #313900

WAS (Xp11)

Small platelets (61±64.7) Severe immunodeficiency

1. Total inactivation35-37 2. Knock-in Y293113

Thrombocytopenia linked to filamin A Thrombocytopenia linked to tubulin 1 Thrombocytopenia linked to α-actinin Thrombocytopenia linked to DIAPH1

X-linked #none AD #nd AD #615193 AD #none

FLNA (Xq28)

Large platelets (34±12.7)

Tissue-specific inactivation47

TUBB1 (6p21.3)

Giant platelets (82±44.7)

Total inactivation48

ACTN1 (14q24.1)

Large platelets (87±31.7)

No mouse model

DIAPH1 (5q31)

Total inactivation115

No thrombocytopenia Normal platelets

Thrombocytopenia linked to PRKACG

AR #616176

PRKACG (9q21.11)

Large platelets (range 69-147)114 Mild neutropenia Severe early onset hearing loss Thrombocytopenia with giant platelets (range 5-8)116 Decreased level of FLNA Impaired platelet activation

No mouse model

Mutations of the TPO-cMPL pathway Congenital amegakaryocytic AR thrombocytopenia (CAMT) #604498 Mutations of transcription factors GATA1-related diseases X-linked (GATA1-RD) #300367 #314050 Congenital AD thrombocytopenia with #605432 radio-ulnar synostosis (CTRUS) Familial platelet disorder AD and predisposition #601399 to acute myeloid leukaemia (FPD/AML) Paris-Trousseau thrombocytopenia (TCTP) Jacobsen syndrome (JBS)

MPL (1p34)

Normal sized platelets (13±4.7) Evolution to medullary aplasia

Total inactivation51,52 Tissue-specific inactivation61

Severe thrombocytopenia

GATA1 (Xp11)

Large platelets (24±9.8) Anemia

HOXA11 (7p15-14)

Normal sized platelets (30) radio-ulnar synostosis ± other defects Normal sized platelets (103±35.9) Risk of leukemia or MDS

1. Total inactivation65 2. Knock-in in the promoter region (Gatalow)68 Total inactivation73,74

1. Embryonic lethal (E10.5-11.5) 2. Severe thrombocytopenia due to defective MK maturation Abnormal

1. Total inactivation78,79 2. Conditional inactivation (Mx-Cre)80-83 3. Conditional inactivation (Pf4-cre)87 1. Total inactivation Fli189 2. Knock-in with truncated Fli-192

1. Embryonic lethal (E11.5-12.5) 2. Defective MK and lymphocyte maturation 3. Defective MK maturation

RUNX1 (21q22)

Deletion (11q23-ter) Normal sized platelets (49±9.9) including FLI-1

Thrombocytopenia linked to GFI1B

AD #188025 #600588 #147791 AD #187900

GFI1B (9q34.13)

Large platelets (97±24.7) Decrease in α-granules number

Thrombocytopenia linked to ETV6 ; THC5

AD #616216

ETV6 (12p13.2)

Normal sized platelets (range 44-132) Erythrocyte macrocytosis Predisposition to leukemia

1. Embryonic lethal (E11.5) 2. Decreased viability Thrombocytopenia Functional platelet defects 95 1. Total inactivation 1. Embryonic lethal (E14.5) 2. Conditional inactivation (Mx-cre)96 2. Expansion of functional 3. Conditional inactivation HSC no longer quiescent (inducible by doxycycline)96 3. Lethal within 3 weeks Decreased Hb level and platelets Total inactivation102 Embryonic lethal (E10.5-11.5)

continued on the next page

898

haematologica | 2016; 101(8)

Mouse models of constitutional thrombocytopenia continued from the previous page

Other mutations Gray platelet syndrome

AR #139090

NBEAL2 (3p21.1)

Thrombocytopenia linked to ANKRD26 / Familial thrombocytopenia (THC2) Thrombocytopenia with absent radii (TAR)

AD #313900

ANKRD26 (10p2)

AR #274000

RBM8A (1q21.1)

Thrombocytopenia linked to CYCS Thrombocytopenia linked to ITGA2B/ITGB3

AD #612004 AD

CYCS (7p15.3) ITG2B/ITGB3 (17q21.31/17q21.32)

Thrombocytopenia which worsens Total inactivation104-107 with age (55±21.3) Myelofibrosis and splenomegaly Normal sized platelets (43±28.4) Total inactivation109,110,117 Risk of leukemia or MDS

Macrothrombocytopenia Absence of α-granules in platelets Obesity, gigantism due to hyperphagia No thrombocytopenia

Thrombocytopenia which normalizes in adults (19) Normal sized platelets Bilateral radial aplasia±other malformations Thrombocytopenia (109x109/L) Knock-in mouse expressing (range 73–167)118,119 a mutant Cyt c120 Macrothrombocytopenia121 No mouse model

No mouse model

plasia. FLI-1, a factor of the ETS family (E26 transformation-specific or E-twenty-six), whose binding site is present in most megakaryocytic promoters. Deletions in the FLI-1 gene are responsible for Paris-Trousseau syndrome and Jacobsen’s disease. And finally, HOXA11, a member of the family of homeobox genes mutated in patients with radioulnar synostosis with amegakaryocytic thrombocytopenia (RUSAT). The genes responsible for several forms of thrombocytopenia have been identified, and mouse models exist for most of the diseases listed to date (Table 1).

Changes affecting the cytoskeleton MYH9-related diseases The MYH9 gene encodes the heavy chain of non-muscle myosin IIA (NMMHC-IIA, non-muscle myosin heavy chain IIA) and mutations in this gene are responsible for several syndromes, including May-Hegglin anomaly, Sebastian syndrome, Fechtner syndrome, Epstein syndrome and some forms of Alport syndrome. Since the early 2000s, when the gene was identified, these diseases have been collectively renamed "MYH9 syndrome" and are the most common causes of constitutional thrombocytopenia. This is an autosomal dominant disease characterized by thrombocytopenia, platelet macrocytosis and the presence or absence of Döhle-like inclusions in leukocytes. Depending on the type of mutation, the macrothrombocytopenia may be accompanied by the gradual onset of kidney failure, cataracts and/or sensorineural hearing loss. A genotype-phenotype comparison has enabled a correlation of the presence of mutations in the N-terminal motor domain with the likelihood of developing non-hematological events and the severity of thrombocytopenia. The latter is also directly correlated with the severity of hemorrhagic manifestations, which are weak in most cases but require transfusion in a minority of patients.12 There are three isoforms of non-muscle myosin type II (IIA, IIB and IIC), but only myosin IIA persists in the late stages of megakaryocytic differentiation haematologica | 2016; 101(8)

Embryonic or perinatal death -

and in platelets, explaining the congenital platelet phenotype. During megakaryocytic differentiation, the expression of myosin IIB is normally repressed by the RUNX1 transcription factor. The persistence of myosin IIB thus characterizes thrombocytopenic patients with mutations in RUNX1. Several mouse models have been developed to study MYH9-related diseases and the possible role of myosin IIA in their different manifestations. Homozygous mice with complete inactivation of the Myh9 gene (knockout) are not viable and die at the E7.5 embryonic stage despite the presence of myosin IIB, confirming the specific functions of the different isoforms of myosin II.13,14 Selective inactivation of the gene in the murine platelet lineage reproduces the macrothrombocytopenia observed in patients.15 Platelet shape changes and clot retractions are abolished, resulting in a decreased stability of platelet thrombi in vivo. This model has also helped to answer a number of questions concerning platelet abnormalities in MYH9 patients. Although based on in vitro experiments with human MYH9-RD or Myh9-/- progenitors, a possible ectopic platelet release within the extravascular bone marrow compartment was proposed, which could account for thrombocytopenia;16,17 no such observation has been reported in vivo using mouse models. On the contrary, a lower viability of megakaryocytes in the marrow and defective DMS formation prevent in situ normal proplatelet extension,18 which accounts for the thrombocytopenia. Less expectedly, these mice have helped to identify the role of myosin in the distribution of platelet granules. Finally, a preclinical study of the long-term effects of romiplostim, a second-generation thrombopoietic agent, showed increased myelofibrosis secondary to the treatment of myosin IIA-deficient mice.19 The most recent development by two independent teams of "knock-in" mice recreating mutations present in patients now allows us to study the human disease more closely.20,21 In both cases, the heterozygous mice display not only macrothrombocytopenia but also non-platelet events, namely intracytoplasmic aggregates of myosin in leukocytes, kidney failure, hearing loss and cataracts. 899

C. Léon et al.

However, the genotype-phenotype correlation observed in humans is not found in these murine models. Despite this difference, these mouse strains are still excellent models which can be used to study the mechanisms altered in the patients' platelets and the pathophysiology of the renal, auditory and visual defects.

Bernard-Soulier syndrome Bernard-Soulier syndrome (BSS) is in its classical form an autosomal recessive disease characterized by thrombocytopenia and the presence of giant platelets. There is also a dominant monoallelic form, usually more moderate, where the patients display mild thrombocytopenia and large platelets, but little or no bleeding. The disease results from mutations in genes encoding one of the subunits of the platelet GPIb-V-IX complex (GP1BA, GP1BB or GP9). The GPIb-V-IX complex is the receptor for von Willebrand factor (vWF), which plays a key role in the early phases of platelet adhesion to injured subendothelium. In the classical biallelic forms, these mutations lead to a substantial absence of surface expression of the complex. The resultant concomitance of macrothrombocytopenia and a functional defect in platelets explains the severity of the disease. The monoallelic form leads to a reduction of about 40% in expression of the GPIb-V-IX complex, and platelet function is little or not affected.22 Whereas the absence of the GPIb-V-IX complex explains the functional defect in platelets, the mechanisms underlying the macrothrombocytopenia are not fully understood. Various BSS mouse models have been developed and have helped to explore potential mechanisms involving

cytoskeletal proteins. Among these models, complete inactivation of the gene encoding GPIbα23 or GPIbβ24,25 reproduces the bleeding phenotype, macrothrombocytopenia and failure to extend proplatelets. The decrease in proplatelet formation may partly result from a lack of maturation of megakaryocytes, as shown especially by abnormalities in the development of the DMS observed in situ.23,24,26,27 A third murine model has been generated by inserting a transgene corresponding to a chimeric receptor comprising an extracellular portion of the interleukin-4 receptor fused to the transmembrane and intracellular portions of GPIbα.28 After crossing with GPIbα-deficient mice, the complex is expressed on the plasma membrane. The number of circulating platelets is increased and platelet size is reduced as compared to GpIbα-deficient animals. These findings point to an important role of the intracellular domain of GPIbα in regulating the number and size of platelets, presumably through interaction with the actin cytoskeleton via filamin A binding. Defects in structuration or dynamics of the microtubule cytoskeleton could also account for the macrothrombocytopenia. Thus, absence of the GPIb-V-IX complex results in an increase in the number of microtubule coils in proplatelets and in the marginal band of circulating platelets.27 Finally, knockout of the Gp5 gene has shown that this subunit is not necessary for the expression of other subunits of the GPIb-V-IX complex or for its main functions, including the production of platelets. This probably explains the lack of BSS patients related to a defect in GPV.29 BSS mouse models have also proved useful in preclinical studies. GPIbα-deficient mice display resistance to throm-

Figure 1. Main stages of megakaryocytopoiesis. After differentiation from the hematopoietic stem cells, megakaryocytes undergo a series of endomitoses leading to polyploid giant cells. The cytoplasm also undergoes a maturation phase during which the granules are synthesized together with a complex network of internal membranes. In the final stages, the megakaryocytes extend cytoplasmic projections in the sinusoid vessel to release platelets. (Drawing by Fabien Pertuy).

900

haematologica | 2016; 101(8)

Mouse models of constitutional thrombocytopenia

bosis in several in vitro and in vivo models, while their platelet procoagulant activity is greatly reduced.27,30,31 Moreover, GPIbÎą-deficient mice and transgenic animals expressing human GPIbÎą have provided proof of the concept that a genetic platelet disease can be cured by gene therapy.32,33

Wiskott-Aldrich syndrome (WAS) and X-linked thrombocytopenia (XLT) Wiskott-Aldrich syndrome (WAS) and moderate socalled "X-linked thrombocytopenia" (XLT) result from mutations in the WAS gene, which is located on the X chromosome and codes for the protein WASp (WiskottAldrich syndrome protein). Nonsense mutations, deletions, insertions or complex mutations lead to a truncated, absent or non-functional protein and cause severe WAS, a condition characterized by thrombocytopenia associated with the progressive onset of severe immune deficiency, eczema and an increase in the incidence of autoimmune and malignant diseases. Missense mutations generally result in XLT syndrome. WASp is expressed exclusively in hematopoietic cells and plays a key role in regulating the intracellular cytoskeleton. Its absence leads to pronounced

microcytic thrombocytopenia (typically less than 10% of the normal platelet count) accompanied by severe bleeding events, the most serious being intracranial, gastrointestinal or oral, which are major causes of morbidity and mortality.34 It was long thought that thrombocytopenia resulted essentially from a decreased platelet life span, this being less than 24 hours in patients, and comparable when WAS platelets are transfused into normal control recipients. The number of megakaryocytes in the bone marrow of patients is normal or even increased, ruling out the possibility of a platelet production defect. Finally, WAS megakaryocytes differentiated in vitro from CD34+ cells exhibit normal ultrastructure and proplatelet formation capacity. Since the platelets produced in vitro are of normal size, the microcytosis observed in patients could result from abnormal platelet cytoskeleton reorganization in the circulation. Mouse models shed new light on the role of WASp and the effects of its absence on platelet formation. Total inactivation reproduces the thrombocytopenia, which is moderate to severe depending on the genetic background and the individual, although microcytosis is absent.35-37

Figure 2. Hematopoiesis and Megakaryocytic commitment. Key cytokines (red) and transcription factors (blue) controlling megakaryocytopoiesis, some of which are the targets of mutations responsible for thrombocytopenia in humans. MPP: multipotent progenitor; CMP: common myeloid progenitor; MEP: megakaryocyteerythrocyte progenitor. (Drawing by Fabien Pertuy).

haematologica | 2016; 101(8)

901

C. Léon et al.

Absence of WASp resulted in ectopic platelet release.37 By relaying the signalling induced by collagen I and favoring the migration of megakaryocytes to sinusoids, WASp plays a key role in preventing the premature release of platelets in the bone marrow compartment.37 In addition, this model proved to be useful in showing that the increased platelet consumption resulted from both intrinsic platelet factors and extrinsic factors, due to an increased susceptibility of these mice to develop platelet antibodies with subsequent opsonization and phagocytosis of platelets.35,36 The development of “knock-in” mouse models has allowed one to explore the mechanisms governing the activation and stability of WASp, including the important role of phosphorylation at the Y293 single site (the murine equivalent of the Y291 site in human WASp). This phosphorylation site is critically involved in the activation of WASp in numerous cellular responses including proliferation, phagocytosis and especially the assembly of adhesion and chemotaxis structures.38 Finally, these murine models have been successfully used for ten years to conduct preclinical trials evaluating somatic gene therapy as an alternative to transplantation. In fact, despite the advances in their diagnosis, the prognosis of WAS patients remains poor. While transfusion therapy in the event of severe hemorrhage together with splenectomy may suffice in moderate cases, allogeneic HSC grafts remain the treatment of choice for WAS/XLT syndromes. HSC transduction trials have been carried out using murine retroviral or lentiviral vectors encoding WASp, and followed by transplantation into WAS-deficient mouse models. The results show a very high transduction efficiency, long-term engraftment of the cells and transgene expression in the lymphoid and myeloid lineages. Transplantation corrects the immune deficiency, while reducing the production of autoantibodies.39-44 These advances, together with improvements in the transcription vectors and promoters, have permitted the evaluation of gene therapy trials to treat some of these patients.45,46

FNLA-related thrombocytopenia Filamin A, encoded by the gene FLNA, is an actin-binding protein belonging to the family of filamins historically called ABP-280 (actin-binding protein 280). The filamins are large dimeric molecules ensuring cross-linking and stabilization of the filamentous actin network and its anchorage to transmembrane glycoproteins, including the GPIbV-IX complex and integrin αIIbβ3, and also serve as a protein scaffold for different signalling intermediates. Monoallelic mutations in the FLNA gene lead to various defects including brain, bone and cardiovascular abnormalities. At the platelet level, macrocytosis is observed with or without thrombocytopenia.8,9 These observations are consistent with data obtained in a mouse model lacking filamin A in the platelet lineage, which is characterized by severe macrothrombocytopenia and decreased expression of the GPIb-IX complex on the platelet surface.47 Thrombocytopenia results from both rapid clearance of the platelets appearing most vulnerable and ineffective platelet production. The proplatelets are formed prematurely and release large cell fragments. These are then further fragmented into platelets, which are subject to microvesiculation and rapidly cleared from the circulation. The microvesiculation is due to a reduced stability of the cell membranes, possibly caused by instability of the submembranous actin network.47 902

Thrombocytopenia related to β1 tubulin

The β1 tubulin isoform is predominantly expressed in blood platelets. Mutations in the TUBB1 gene have been described in two families but have not yet been reproduced in animals. These patients exhibit macrothrombocytopenia associated with a decrease in the amount of β1 tubulin in platelets. The mutations make the protein unstable and unable to be incorporated into microtubules, leading to a decreased number of megakaryocytes forming proplatelets. The importance of this particular tubulin isoform has been demonstrated by inactivation of the Tubb1 gene in mice.48 As in patients, the mice display macrothrombocytopenia resulting from defective proplatelet formation, despite compensatory overexpression of Tubb2 and Tubb5. The lack of phenotype observed in heterozygous mice suggests that the phenotype of patients could be related to a dominant negative effect. A dominant negative effect due to another mutation causing instability of microtubules has been reported in dogs presenting β1 tubulin mutations.49

Changes affecting the cMPL-TPO pathway Congenital amegakaryocytic thrombocytopenia (CAMT) Congenital amegakaryocytic thrombocytopenia (CAMT) is the best known form of amegakaryocytic thrombocytopenia. It is an autosomal recessive disease caused by mutations in the MPL gene encoding the TPO receptor. This receptor is present on HSCs, megakaryocytes and platelets and activates different signalling pathways, in particular Jak2/STAT, Ras/MPK and PI3K. The thrombocytopenia is severe at birth and progresses to aplastic anemia during the first year of life, inevitably requiring HSC transplantation.50 There is a clear correlation between genotype and phenotype as a non-functional protein causes profound thrombocytopenia, while patients with residual activity of the receptor are less severely affected. Conversely, MPL mutations resulting in constitutive activation of the receptor or reduced clearance of TPO, as likewise mutations in the THPO gene leading to increased gene translation, have been identified in some familial forms of essential thrombocythemia. Mice deficient in Mpl partially reproduce the clinical picture. As in humans, the thrombocytopenia is severe, although pancytopenia is not observed despite a decrease in the numbers of hematopoietic progenitor cells of different lineages.51,52 Subsequent studies of this model have demonstrated the role of TPO-MPL signalling in the regulation of HSCs in adult mice53 and in the establishment of definitive hematopoiesis during embryonic development.54,55 On the other hand, the presence of residual functional platelets in MPL-deficient animals point to the existence of other factors involved in platelet formation, and these mice have proved an extremely useful model to identify such factors. Various studies have shown that the interleukins IL-3, IL-5, IL-6, IL-11 and IL-17 do not participate in the residual thrombopoiesis.56-59 In contrast, the chemokine SDF-1 and FGF-4 promote thrombopoiesis independently of TPO, allowing the interaction of megakaryocytic progenitors with the vascular niche.60 Recently, the inactivation of MPL specifically in the murine megakaryocytic lineage has revealed that the TPO-MPL axis plays its key role at the level of bipotent megakaryocytic precursors. Thus, the MPL receptor is not essential for megakaryocyte prolifhaematologica | 2016; 101(8)

Mouse models of constitutional thrombocytopenia

eration, maturation or platelet production, but is crucial to control the local concentrations and therefore the availability of TPO in the bone marrow microenvironment, and hence to avoid myeloproliferation through over-stimulation of progenitors.61 Finally, as in other pathologies, mice invalidated for the MPL gene have proved a tool of choice for the development of gene therapy and the validation of novel vectors.62,63

Changes affecting transcription factors GATA 1-related thrombocytopenia There are two types of GATA1-related thrombocytopenia, dyserythropoietic anemia with thrombocytopenia (DAT) and X-linked thrombocytopenia with beta-thalassemia (XLTT). In both cases GATA1 mutations are responsible for dysmegakaryopoiesis and dyserythropoiesis with varying degrees of anemia and macrothrombocytopenia, the phenotype being less severe in XLTT. GATA1 is a transcription factor controlling the expression of numerous genes involved in megakaryocytopoiesis (GPIBB, PF4, MPL, NFE2 etc.) and genes of the erythroid lineage (HBB, ALAS1, BCL2L1).64 GATA1 contains two zinc finger domains, one N-terminal (Nf) and the other Cterminal (Cf). The Nf domain interacts with both FOG1 and DNA. Mutations responsible for XLTT affect DNA binding but not the interaction with FOG1, while mutations causing DAT affect FOG1 binding. Complete gene inactivation is lethal at the E10.5-11.5 embryonic stage. Studies of chimeric mice derived from Gata1-deficient ES cells have shown an increased number of megakaryocytes in the fetal liver.65 Similar observations have been made in a mouse model with a point mutation induced by ENU at the initiation codon (Gata1Plt13), mimicking the mutations present in some patients.66 The further development of transgenic mice carrying the Gata1 promoter sequence and reporter gene has allowed for the identification of the promoter regions and the specificity of GATA1 expression during embryonic development.67 Targeted replacement of a long sequence upstream of the Gata1 locus including the distal promoter has been used to generate a viable murine lineage (Gata1low), with a specific lack of GATA1 expression in megakaryocytes.68 These mice display severe thrombocytopenia resulting from a defect in megakaryocyte maturation. The cells exhibit low ploidy and an underdeveloped DMS, leading to a failure to extend proplatelets and defective granule formation.69 These defects are associated with a reduction in the transcription of genes coding for key proteins (GP1Bα, GP1Bβ, PF4, MPL, NFE2) during megakaryocyte maturation. The absence of GATA1 also causes deregulation of the cell cycle with hyperproliferation of the mutant megakaryocytes. With age, Gata1low mice then develop a hematological clinical picture resembling idiopathic myelofibrosis in several respects,70 and therefore represent an excellent model to study this disease.71

Radioulnar Synostosis with amegakaryocytosis thrombocytopenia (RUSAT) This rare congenital disorder is characterized by severe amegakaryocytic thrombocytopenia which is present at birth and associated with proximal fusion of the radius and ulna. The thrombocytopenia does not improve with age and the children can also develop myeloid failure requiring bone marrow transplantation.72 All these symphaematologica | 2016; 101(8)

toms allow for the distinction of the disease from CAMT and TAR syndrome (thrombocytopenia with absent radii), and in most cases mutations in the HOXA11 gene have been identified. HOXA11 is a member of the HOX family of homeobox genes which code for transcription factors. These proteins play important roles during embryonic development, where their expression is strongly regulated in space and time. At the hematopoietic level, HOXA11 is expressed only in the earliest precursors. Mouse models with Hoxa11 inactivation have been developed but do not permit the study of the mechanisms involved in platelet formation. Heterozygous and homozygous Hoxa11-deficient mice have malformations in the lower and hind limbs, consistent with the observations in patients, but thrombocytopenia has never been reported.73,74 At present, it is not known whether the megakaryocytopoietic defects present in patients result from a dominant negative effect of the point mutations or not.75

Familial platelet disorder with a predisposition for acute myeloid leukemia (FPD/AML) Familial platelet disorder with a predisposition for acute myeloid leukemia (FPD/AML) is the consequence of mutations in the RUNX1 gene (also known as AML1 or CBFA2). Somatic mutations in RUNX1 are also responsible for a non-inherited form of AML and for chronic myelomonocytic leukemia.76 RUNX1 encodes the α subunit of the CBF (core binding factor) transcription complex.77 The heterodimerization of CBFα with CBFβ increases its affinity for DNA and protects it from proteolytic degradation. The transcription factor CBF regulates the expression of many specific hematopoietic genes and is essential for the establishment of definitive hematopoiesis. Most of the mutations identified in patients lead to haploinsufficiency, although some variants may act through a dominant negative effect. The patients exhibit thrombocytopenia, sometimes accompanied by a functional defect in platelet aggregation in response to collagen, together with an increased risk of developing acute myeloid leukemia or myelodysplastic syndromes (40% of cases). Complete inactivation of Runx1 in mice is lethal at the E11.5-12.5 embryonic stage due to intracranial hemorrhages. These mice have nevertheless been useful to show that Runx1 is required for the establishment of definitive hematopoiesis during embryonic development.78,79 Conditional gene ablation by crossing with mice expressing cre recombinase under the control of the Mx1 promoter allows inactivation of the gene in adult mice. In this case, the loss of RUNX1 in adulthood does not result in a total loss of hematopoiesis.80-83 However, megakaryocyte and lymphocyte maturation are inhibited, consistent with a role of Runx1 in the maturation of these cells. Furthermore, the fraction of immature progenitors is increased since Runx1 negatively regulates HSC quiescence.84 It has been reported that hematopoietic progenitors from FPD/AML patients have an increased clonogenic potential, and in some cases abnormal self-renewal capacity.85 The expansion of these immature cells observed in RUNX1-deficient mice could thus be relevant to the pathogenesis of human hematological tumors associated with a lack of RUNX1.86 More recently, inactivating Runx1 during megakaryocytic maturation by crossing floxed mice with Pf4-cre mice has demonstrated the importance of this factor in 903

C. Léon et al.

megakaryocyte maturation and allowed a genomic analysis of RUNX1-controlled gene expression during the late stages of megakaryocytopoiesis, again providing useful information for our understanding of FPD/AML.87

Paris-Trousseau and Jacobsen syndromes (FLI1) Paris-Trousseau syndrome and Jacobsen syndrome, with dominant transmission, result from a partial deletion of the long arm of chromosome 11 (11q23-ter). About 200 cases have been described with twice as many women affected as men. The distinction between the two syndromes is based on the severity of the disease. In all cases thrombocytopenia due to a defect in megakaryocyte maturation is present, with many immature, microcytic, hypolobulated and dystrophic cells.88 The patients also suffer from facial dysmorphia. Additional symptoms are observed in Jacobsen syndrome including cardiac, renal, gastrointestinal and genital abnormalities and an intellectual deficit. The variability of the symptoms is probably correlated with the type of deletion, which is indeed highly variable in size, up to 20 Mb. The cleavage site is generally located in or near the sub-band 11q23.3 and the deletion usually extends to the telomere. This region covers in particular two genes encoding the ETS family of transcription factors, ETS1 and FLI1. The study of a mouse model with inactivation of the Ets1 gene has demonstrated that this transcription factor is not critical for megakaryopoiesis, whereas differentiation of the lymphoid lineage is strongly disrupted.89 In contrast, studies of mice with Fli1 inactivation have shown that this protein plays a key role in the human syndromes through its involvement in megakaryocytic gene transactivation. Complete inactivation of Fli1 results in embryonic lethality at the E11.5 stage, following intracranial hemorrhage.90 The absence of FLI1 affects both vasculogenesis and megakaryopoiesis. As in patients, the megakaryocytes display morphological abnormalities with characteristics of immature cells. In this study heterozygous Fli1+/- mice showed no abnormal phenotype, which raised the question of the mechanisms through which the haploinsufficiency in patients might lead to the various symptoms. Another recent study compared FLI1+/- and [Ets1+/+; Fli1+/-] animals. While mild thrombocytopenia and craniofacial abnormalities were found in Fli1+/- mice, some symptoms were more severe in [Ets1+/-; Fli1+/-]animals, suggesting partial functional redundancy of the two transcription factors. Hence, at least some of the patients’ symptoms may result from hemizygosity.91 Finally, mutant mice expressing a truncated form of the molecule lacking the C-terminal regulatory domain have been generated to study the role of functional FLI1 domains.92 Homozygous mice have a decreased viability (30% survive into adulthood) and present thrombocytopenia associated with functional defects due in particular to a decrease in megakaryocytic gene expression (Mpl, Itga2b, Gp5, Gp9, Pf4, Nfe2, Mafg and Rab27b).

Thrombocytopenia associated with GFI1B A nonsense mutation was very recently discovered in the GFI1B gene in a family with an autosomal dominant form of Gray platelet syndrome.93 GFI1B (growth factor independent 1B) is a transcription factor containing a SNAG repressor domain (Snail GFI1) in the N-terminal conserved domains and 6 zinc fingers. The SNAG domain plays the role of a transcriptional repressor. GFI1B is 904

expressed in hematopoietic cells and especially in erythroid cells, megakaryocytes and their common progenitors (MEPs). The generation of a reporter mouse line expressing GFP (green fluorescent protein) at the Gfi1b locus (Gfi1b+/GFP animals) showed that the expression of this protein during hematopoiesis is inversely proportional to the expression of another protein of the same family, GFI.94 In humans, the mutation results in a truncated protein lacking the last 44 amino acids, which has a dominant negative effect with respect to the normal protein. Similarly as in autosomal recessive Gray platelet syndrome (see below), thrombocytopenia is associated with large platelets displaying partial or total absence of the αgranules, depending on the patient. Moderate to severe bleeding has been reported, which may be attributed in part to a decrease in the expression of GPIbα. Within the bone marrow, the development of stage I myelofibrosis has been observed with an increased number of pleomorphic megakaryocytes containing only a few granules. Megakaryocytes and platelets retain strong expression of the CD34+ stem cell marker. Studies in mice have demonstrated the role of GFI1B as a transcriptional repressor acting during embryonic and adult erythropoiesis and megakaryopoiesis. Total gene deletion causes the arrest of embryonic development at around E14.5 due to the delayed maturation of primitive erythrocytes and a failure to produce definitive anucleate erythrocytes, accompanied by a lack of differentiation in the megakaryocytic lineage.95 Heterozygous animals develop normally, indicating that haploinsufficiency is not responsible for the human disease, in agreement with a potential dominant negative effect of the human deletion. Moreover, the generation of conditional knockout mice in which cre recombinase is under the control of the inducible Mx1 promoter has demonstrated that this protein is critically involved in regulating the dormancy and proliferation of adult HSCs.96 Specific inactivation of Gfi1b in erythrocytes [by crossing with mice expressing cre recombinase under the control of the Epor (erythropoietin receptor) promoter] has revealed the involvement of GFI1B in the differentiation of pro-erythroblasts into mature erythrocytes, and the extinction of globin gene expression in embryonic and adult cells.97 Finally, by crossing with mice expressing cre recombinase under the control of an inducible doxycycline promoter, it was possible to confirm the role of GFI1B as a transcriptional repressor of adult erythropoiesis and thrombopoiesis.98 During megakaryopoiesis in adults, GFI1B acts as a repressor mainly after the completion of endomitoses but before the onset of cytoplasmic maturation.

Thrombocytopenia associated with ETV6 Very recently, a few families with variable thrombocytopenia have been found to harbor germline mutations in ETV6 (Ets-variant 6, also known as TEL). ETV6 encodes a transcriptional repressor of the ETS family, and was initially identified as a tumor suppressor through its role in childhood leukemia resulting from somatic translocations. In addition to autosomal dominant thrombocytopenia, the patients present a high erythrocyte mean corpuscular volume, pointing to a defect affecting megakaryocyticerythroid precursors. Mild to moderate bleeding has been reported, with an increased susceptibility to develop acute lymphoblastic leukemia. It has been proposed that the haematologica | 2016; 101(8)

Mouse models of constitutional thrombocytopenia

mutations affect megakaryocyte development with an increased presence of small and immature cells, possibly due to abnormal cytoskeleton organization.99-101 Total inactivation of mouse Etv6 has long been known to be embryonically lethal on account of yolk sac angiogenic defects.102 While ETV6 appears to be dispensable for fetal liver hematopoiesis, it is required for bone marrow hematopoiesis and hematopoietic stem cell maintenance. Heterozygous loss of one Etv6 allele does not affect hematopoiesis, suggesting a dominant negative effect in patients, which has now been confirmed through in vitro studies.

Other mutations Gray platelet syndrome The name Gray platelet syndrome derives from the gray appearance of platelets on a blood smear stained with May-Grünwald-Giemsa, which is due to the absence of αgranules. This syndrome is characterized by macrothrombocytopenia associated with myelofibrosis and splenomegaly.103 The autosomal recessive disease is due to a biallelic mutation in the NBEAL2 (neurobeachin-like 2) gene, the patients being either homozygous or compound heterozygous with a resultant loss of function of the protein. Studies in patients have shown that the number of megakaryocytes is normal while thrombocytopenia increases with age. A decreased platelet survival, together with the progressive development of myelofibrosis, probably due to the premature release of growth factors, may explain these observations. Three mouse models with knockout of Nbeal2 have been generated recently.104-107 While all three reproduce the macrothrombocytopenia and lack of α-granules in platelets, they display differences which are worth noting. The study from Kahr et al.,107 (Model 1), reports splenomegaly, like the human pathology, but no myelofibrosis. Ultrastructural examination of megakaryocytes from these mice show abnormalities in the development of the DMS and in the maturation of α-granules, with an abundance of immature cells. Cultured megakaryocytes are hypopolyploid and extend fewer proplatelets, while von Willebrand factor is abnormally located at the cell periphery and can be secreted. All these features suggest that the normal development of α-granules contributes to the maturation of megakaryocytes. The mouse model reported by Deppermann et al.,105 (Model 2), exhibits little or no splenomegaly, even in aged mice, and the demarcation membranes develop normally in megakaryocytes. αgranules are absent, but there is an increase in the numbers of mitochondria and vacuoles. Since proplatelet formation and platelet survival are normal, the authors propose that the thrombocytopenia results from defects in the terminal stages of platelet release. Finally, the mouse line developed by Guerrero et al.,106 (Model 3), displays both splenomegaly and myelofibrosis. In situ, the megakaryocytes are smaller and generally less polylobed, while ultrastructural images reveal a slightly more rudimentary DMS. The proplatelet formation visualized by the adhesion of megakaryocytes cultured in vitro to a fibrinogen surface is similar between these animals and those of Model 2. Surprisingly, mature α-granules are present in megakaryocytes in situ and in the proplatelet buds of megakaryocytes differentiated in vitro. The authors suggest that the lack of α-granules in platelets haematologica | 2016; 101(8)

results from an abnormal retention of the granules, rather than from their abnormal production. In all three genotypes the deletion is total and Models 1 and 2 share the same origin. Hence the discrepancies, at least between the first two models, could arise from a difference in strain or from heterogeneity between mice, especially as the numbers of animals used in these studies were relatively low. Moreover, we cannot exclude that the differences with respect to the third model result from the way in which the inactivation was performed. Although these discrepancies do not allow one to conclude the exact role of NBEAL2 in the formation of platelet α-granules and thrombocytopenia, on the basis of the present data it has been possible to formulate a number of hypotheses. The studies need to be continued to determine the actual mechanisms responsible for the human diseases, and how they depend on the type of mutation.

Thrombocytopenia associated with ANKRD26 / familial thrombocytopenia 2 Familial thrombocytopenia 2 (THC2) with a predisposition for leukemia is a rare form of autosomal dominant thrombocytopenia. Whereas the platelets are morphologically and functionally normal, the thrombocytopenia is moderate to severe and leads to moderate bleeding. A platelet production defect has been proposed in view of the fact that the bone marrow of patients displays dysmegakaryopoiesis and micromegakaryocytosis. This disease has recently been associated with point mutations in the non-coding 5' part of the ANKRD26 (ankyrin repeat domain 26) gene.108 ANKRD26 encodes a 192 kDa protein abundantly expressed in various tissues such as the brain, liver and adipose tissue, skeletal muscle and hematopoietic tissue. Interestingly, Ankrd26-deficient mice develop obesity and gigantism associated with hyperphagia but do not exhibit thrombocytopenia, suggesting that haploinsufficiency is not responsible for the thrombocytopenia observed in patients.109,110 Consistent with these observations, recent data have shown that the 5 'UTR mutations affect binding of the RUNX1 and FLI1 transcription factors, resulting in the absence of the repression of the ANKRD26 gene which normally occurs in the late stages of megakaryopoiesis.111 This abnormal overexpression increases ANKRD26 signalling via the TPO/cMPL axis, causing abnormal signalling through the ERK/MAPK pathway and consequently defective proplatelet formation. Hence, mouse overexpression of ANKRD26 is anticipated to better recapitulate the pathology.

Conclusions Recent years have seen major advances in our understanding of the normal and pathological mechanisms of thrombopoiesis. The contribution of animal models in this area has been important to date, not so much to identify genes implicated in thrombocytopenia, but rather to elucidate the mechanisms involved through their use in structure-function studies in mutated genotypes, and as preclinical models to test new drugs or develop gene therapies. At present, the cause of thrombocytopenia is unknown in 40% to 50% of cases. The new high-speed sequencing techniques, in particular the sequencing of complete genomes or exomes, now allow us to identify candidate genes. These new candidates will be the targets of many 905

C. Léon et al.

future murine models. Complementarily, programs of random mutagenesis in mice should enable us to identify other candidate genes involved in thrombocytopenia, prior to their investigation in patients. The ideal animal model remains the "knock-in" mouse, with identical reproduction of the human mutation when this is known, provided the gene and its function are conserved between mouse and man. The recent possibility of generating induced pluripotent stem cells (iPS) now permits the production of mutated hematopoietic stem cells from fibroblasts of patients. Xenotransplantation of these stem cells into immunodeficient (NOD/SCID/γc) mice may be expected to mimic or reproduce the human pathology, even when the gene responsible is unknown.

References 13. 1. Gresele P, Harrison P, Bury L, et al. Diagnosis of suspected inherited platelet function disorders: results of a worldwide survey. J Thromb Haemost. 2014;12(9):1562-1569. 2. Pertuy F, Aguilar A, Strassel C, et al. Broader expression of the mouse platelet factor 4-cre transgene beyond the megakaryocyte lineage. J Thromb Haemost. 2015;13(1):115-125. 3. Tiedt R, Schomber T, Hao-Shen H, Skoda RC. Pf4-Cre transgenic mice allow the generation of lineage-restricted gene knockouts for studying megakaryocyte and platelet function in vivo. Blood. 2007;109(4):15031506. 4. Alexander WS, Viney EM, Zhang JG, et al. Thrombocytopenia and kidney disease in mice with a mutation in the C1galt1 gene. Proc Natl Acad Sci USA. 2006;103(44): 16442-16447. 5. Anderson NM, Javadi M, Berndl E, et al. Enu mutagenesis identifies a novel platelet phenotype in a loss-of-function Jak2 allele. PLoS One. 2013;8(9):e75472. 6. Carpinelli MR, Hilton DJ, Metcalf D, et al. Suppressor screen in Mpl-/- mice: c-Myb mutation causes supraphysiological production of platelets in the absence of thrombopoietin signaling. Proc Natl Acad Sci USA. 2004;101(17):6553-6558. 7. Balduini CL, Savoia A, Seri M. Inherited thrombocytopenias frequently diagnosed in adults. J Thromb Haemost. 2013;11(6):10061019. 8. Kumar R, Kahr WH. Congenital thrombocytopenia: clinical manifestations, laboratory abnormalities, and molecular defects of a heterogeneous group of conditions. Hematol Oncol Clin North Am. 2013;27(3): 465-494. 9. Pecci A, Balduini CL. Lessons in platelet production from inherited thrombocytopenias. Br J Haematol. 2014;165(2):179-192. 10. Bury L, Falcinelli E, Chiasserini D, Springer TA, Italiano JE, Jr., Gresele P. Cytoskeletal perturbation leads to platelet dysfunction and thrombocytopenia in variant forms of Glanzmann thrombasthenia. Haematologica. 2016;101(1):46-56. 11. Woolthuis CM, Park CY. Hematopoietic stem/progenitor cell commitment to the megakaryocyte lineage. Blood. 2016;127(10): 1242-1248. 12. Balduini CL, Pecci A, Savoia A. Recent advances in the understanding and management of MYH9-related inherited thrombo-

906

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

This strategy, already employed for hematological malignancies, will allow us to study the abnormal mechanisms in the cells of interest (megakaryocytes and platelets) and evaluate new therapeutic approaches. Sometimes, however, the mouse model does not reproduce the entire human phenotype and can even raise new questions, as in the case of Nbeal2-deficient mice. Hence the in vitro culture and differentiation of human megakaryocytes remains absolutely essential and fully complementary to studies in mice. Indeed, it is the confrontation of data obtained in animals and humans, in vivo and in vitro, which has and will enable major advances in our understanding of thrombocytopenia, and likewise in our knowledge of the normal mechanisms of platelet production.

cytopenias. Br J Haematol. 2011;154(2):161174. Conti MA, Even-Ram S, Liu C, Yamada KM, Adelstein RS. Defects in cell adhesion and the visceral endoderm following ablation of nonmuscle myosin heavy chain II-A in mice. J Biol Chem. 2004;279(40):41263-41266. Matsushita T, Hayashi H, Kunishima S, et al. Targeted disruption of mouse ortholog of the human MYH9 responsible for macrothrombocytopenia with different organ involvement: hematological, nephrological, and otological studies of heterozygous KO mice. Biochem Biophys Res Commun. 2004;325(4):1163-1171. Leon C, Eckly A, Hechler B, et al. Megakaryocyte-restricted MYH9 inactivation dramatically affects hemostasis while preserving platelet aggregation and secretion. Blood. 2007;110(9):3183-3191. Pecci A, Malara A, Badalucco S, et al. Megakaryocytes of patients with MYH9related thrombocytopenia present an altered proplatelet formation. Thromb Haemost. 2009;102(1):90-96. Eckly A, Strassel C, Freund M, et al. Abnormal megakaryocyte morphology and proplatelet formation in mice with megakaryocyte-restricted MYH9 inactivation. Blood. 2009;113(14):3182-3189. Eckly A, Rinckel JY, Laeuffer P, et al. Proplatelet formation deficit and megakaryocyte death contribute to thrombocytopenia in Myh9 knockout mice. J Thromb Haemost. 2010;8(10):2243-2251. Leon C, Evert K, Dombrowski F, et al. Romiplostim administration shows reduced megakaryocyte response-capacity and increased myelofibrosis in a mouse model of MYH9-RD. Blood. 2012;119(14):3333-3341. Suzuki N, Kunishima S, Ikejiri M, et al. Establishment of mouse model of MYH9 disorders: heterozygous R702C mutation provokes macrothrombocytopenia with leukocyte inclusion bodies, renal glomerulosclerosis and hearing disability. PLoS One. 2013;8(8):e71187. Zhang Y, Conti MA, Malide D, et al. Mouse models of MYH9-related disease: mutations in nonmuscle myosin II-A. Blood. 2012;119(1):238-250. Savoia A, Kunishima S, De Rocco D, et al. Spectrum of the mutations in BernardSoulier syndrome. Hum Mutat. 2014;35(9): 1033-1045. Ware J, Russell S, Ruggeri ZM. Generation and rescue of a murine model of platelet dys-

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

function: the Bernard-Soulier syndrome. Proc Natl Acad Sci USA. 2000;97(6):28032808. Kato K, Martinez C, Russell S, et al. Genetic deletion of mouse platelet glycoprotein Ibbeta produces a Bernard-Soulier phenotype with increased alpha-granule size. Blood. 2004;104(8):2339-2344. Strassel C, Nonne C, Eckly A, et al. Decreased thrombotic tendency in mouse models of the Bernard-Soulier syndrome. Arterioscler Thromb Vasc Biol. 2007;27(1): 241-247. Poujol C, Ware J, Nieswandt B, Nurden AT, Nurden P. Absence of GPIbalpha is responsible for aberrant membrane development during megakaryocyte maturation: ultrastructural study using a transgenic model. Exp Hematol. 2002;30(4):352-360. Strassel C, Eckly A, Leon C, et al. Intrinsic impaired proplatelet formation and microtubule coil assembly of megakaryocytes in a mouse model of Bernard-Soulier syndrome. Haematologica. 2009;94(6):800-810. Kanaji T, Russell S, Ware J. Amelioration of the macrothrombocytopenia associated with the murine Bernard-Soulier syndrome. Blood. 2002;100(6):2102-2107. Lanza F. Platelet glycoprotein V: a thrombin substrate marker of thrombosis and of megakaryocyte differentiation. Hématologie. 2000;6(6):424-431. Joglekar MV, Ware J, Xu J, Fitzgerald ME, Gartner TK. Platelets, glycoprotein Ib-IX, and von Willebrand factor are required for FeCl(3)-induced occlusive thrombus formation in the inferior vena cava of mice. Platelets. 2013;24(3):205-212. Ravanat C, Strassel C, Hechler B, et al. A central role of GPIb-IX in the procoagulant function of platelets that is independent of the 45-kDa GPIbalpha N-terminal extracellular domain. Blood. 2010;116(7):1157-1164. Kanaji S, Fahs SA, Ware J, Montgomery RR, Shi Q. Non-myeloablative conditioning with busulfan before hematopoietic stem cell transplantation leads to phenotypic correction of murine Bernard-Soulier syndrome. J Thromb Haemost. 2014;12(10): 1726-1732. Kanaji S, Kuether EL, Fahs SA, et al. Correction of murine Bernard-Soulier syndrome by lentivirus-mediated gene therapy. Mol Ther. 2012;20(3):625-632. Buchbinder D, Nugent DJ, Fillipovich AH. Wiskott-Aldrich syndrome: diagnosis, current management, and emerging treatments. Appl Clin Genet. 2014;7:55-66.

haematologica | 2016; 101(8)

Mouse models of constitutional thrombocytopenia 35. Marathe BM, Prislovsky A, Astrakhan A, Rawlings DJ, Wan JY, Strom TS. Antiplatelet antibodies in WASP(-) mice correlate with evidence of increased in vivo platelet consumption. Exp Hematol. 2009;37(11):13531363. 36. Prislovsky A, Marathe B, Hosni A, et al. Rapid platelet turnover in WASP(-) mice correlates with increased ex vivo phagocytosis of opsonized WASP(-) platelets. Exp Hematol. 2008;36(5):609-623. 37. Sabri S, Foudi A, Boukour S, et al. Deficiency in the Wiskott-Aldrich protein induces premature proplatelet formation and platelet production in the bone marrow compartment. Blood. 2006;108(1):134-140. 38. Blundell MP, Bouma G, Metelo J, et al. Phosphorylation of WASp is a key regulator of activity and stability in vivo. Proc Natl Acad Sci USA. 2009;106(37):15738-15743. 39. Astrakhan A, Sather BD, Ryu BY, et al. Ubiquitous high-level gene expression in hematopoietic lineages provides effective lentiviral gene therapy of murine WiskottAldrich syndrome. Blood. 2012;119(19): 4395-4407. 40. Blundell MP, Bouma G, Calle Y, Jones GE, Kinnon C, Thrasher AJ. Improvement of migratory defects in a murine model of Wiskott-Aldrich syndrome gene therapy. Mol Ther. 2008;16(5):836-844. 41. Bosticardo M, Draghici E, Schena F, et al. Lentiviral-mediated gene therapy leads to improvement of B-cell functionality in a murine model of Wiskott-Aldrich syndrome. J Allergy Clin Immunol. 2011;127(6):1376-1384 e1375. 42. Catucci M, Prete F, Bosticardo M, et al. Dendritic cell functional improvement in a preclinical model of lentiviral-mediated gene therapy for Wiskott-Aldrich syndrome. Gene Ther. 2012;19(12):1150-1158. 43. Charrier S, Stockholm D, Seye K, et al. A lentiviral vector encoding the human Wiskott-Aldrich syndrome protein corrects immune and cytoskeletal defects in WASP knockout mice. Gene Ther. 2005;12(7):597606. 44. Dupre L, Marangoni F, Scaramuzza S, et al. Efficacy of gene therapy for Wiskott-Aldrich syndrome using a WAS promoter/cDNAcontaining lentiviral vector and nonlethal irradiation. Hum Gene Ther. 2006;17(3):303313. 45. Bosticardo M, Ferrua F, Cavazzana M, Aiuti A. Gene therapy for Wiskott-Aldrich Syndrome. Curr Gene Ther. 2014;14(6):413421. 46. Hacein-Bey Abina S, Gaspar HB, Blondeau J, et al. Outcomes following gene therapy in patients with severe Wiskott-Aldrich syndrome. JAMA. 2015;313(15):1550-1563. 47. Jurak Begonja A, Hoffmeister KM, Hartwig JH, Falet H. FlnA-null megakaryocytes prematurely release large and fragile platelets that circulate poorly. Blood. 2011;118(8): 2285-2295. 48. Schwer HD, Lecine P, Tiwari S, Italiano JE, Jr., Hartwig JH, Shivdasani RA. A lineagerestricted and divergent beta-tubulin isoform is essential for the biogenesis, structure and function of blood platelets. Curr Biol. 2001;11(8):579-586. 49. Davis B, Toivio-Kinnucan M, Schuller S, Boudreaux MK. Mutation in beta1-tubulin correlates with macrothrombocytopenia in Cavalier King Charles Spaniels. J Vet Intern Med. 2008;22(3):540-545. 50. Ballmaier M, Germeshausen M. Congenital amegakaryocytic thrombocytopenia: clinical presentation, diagnosis, and treatment.

haematologica | 2016; 101(8)

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

Semin Thromb Hemost. 2011;37(6):673681. Alexander WS, Roberts AW, Maurer AB, Nicola NA, Dunn AR, Metcalf D. Studies of the c-Mpl thrombopoietin receptor through gene disruption and activation. Stem Cells. 1996;14 Suppl 1:124-132. Alexander WS, Roberts AW, Nicola NA, Li R, Metcalf D. Deficiencies in progenitor cells of multiple hematopoietic lineages and defective megakaryocytopoiesis in mice lacking the thrombopoietic receptor c-Mpl. Blood. 1996;87(6):2162-2170. Kimura S, Roberts AW, Metcalf D, Alexander WS. Hematopoietic stem cell deficiencies in mice lacking c-Mpl, the receptor for thrombopoietin. Proc Natl Acad Sci USA. 1998;95(3):1195-1200. Fleury M, Petit-Cocault L, Clay D, Souyri M. Mpl receptor defect leads to earlier appearance of hematopoietic cells/hematopoietic stem cells in the Aorta-Gonad-Mesonephros region, with increased apoptosis. Int J Dev Biol. 2010;54(6-7):1067-1074. Petit-Cocault L, Volle-Challier C, Fleury M, Peault B, Souyri M. Dual role of Mpl receptor during the establishment of definitive hematopoiesis. Development. 2007;134(16): 3031-3040. Chen Q, Solar G, Eaton DL, de Sauvage FJ. IL-3 does not contribute to platelet production in c-Mpl-deficient mice. Stem Cells. 1998;16 Suppl 2:31-36. Gainsford T, Nandurkar H, Metcalf D, Robb L, Begley CG, Alexander WS. The residual megakaryocyte and platelet production in cmpl-deficient mice is not dependent on the actions of interleukin-6, interleukin-11, or leukemia inhibitory factor. Blood. 2000;95(2):528-534. Scott CL, Robb L, Mansfield R, Alexander WS, Begley CG. Granulocyte-macrophage colony-stimulating factor is not responsible for residual thrombopoiesis in mpl null mice. Exp Hematol. 2000;28(9):1001-1007. Tan W, Liu B, Barsoum A, Huang W, Kolls JK, Schwarzenberger P. Requirement of TPO/c-mpl for IL-17A-induced granulopoiesis and megakaryopoiesis. J Leukoc Biol. 2013;94(6):1303-1308. Avecilla ST, Hattori K, Heissig B, et al. Chemokine-mediated interaction of hematopoietic progenitors with the bone marrow vascular niche is required for thrombopoiesis. Nat Med. 2004;10(1):64-71. Ng AP, Kauppi M, Metcalf D, et al. Mpl expression on megakaryocytes and platelets is dispensable for thrombopoiesis but essential to prevent myeloproliferation. Proc Natl Acad Sci USA. 2014;111(16):5884-5889. Heckl D, Wicke DC, Brugman MH, et al. Lentiviral gene transfer regenerates hematopoietic stem cells in a mouse model for Mpl-deficient aplastic anemia. Blood. 2011;117(14):3737-3747. Wicke DC, Meyer J, Buesche G, et al. Gene therapy of MPL deficiency: challenging balance between leukemia and pancytopenia. Mol Ther. 2010;18(2):343-352. Millikan PD, Balamohan SM, Raskind WH, Kacena MA. Inherited thrombocytopenia due to GATA-1 mutations. Semin Thromb Hemost. 2011;37(6):682-689. Pevny L, Lin CS, D'Agati V, Simon MC, Orkin SH, Costantini F. Development of hematopoietic cells lacking transcription factor GATA-1. Development. 1995;121(1):163172. Majewski IJ, Metcalf D, Mielke LA, et al. A mutation in the translation initiation codon of Gata-1 disrupts megakaryocyte matura-

67.

68.

69.

70.

71.

72.

73.

74. 75.

76.

77. 78.

79.

80.

81.

82.

tion and causes thrombocytopenia. Proc Natl Acad Sci USA. 2006;103(38):1414614151. McDevitt MA, Shivdasani RA, Fujiwara Y, Yang H, Orkin SH. A "knockdown" mutation created by cis-element gene targeting reveals the dependence of erythroid cell maturation on the level of transcription factor GATA-1. Proc Natl Acad Sci USA. 1997;94(13):6781-6785. Shivdasani RA, Fujiwara Y, McDevitt MA, Orkin SH. A lineage-selective knockout establishes the critical role of transcription factor GATA-1 in megakaryocyte growth and platelet development. Embo J. 1997;16(13):3965-3973. Vyas P, Ault K, Jackson CW, Orkin SH, Shivdasani RA. Consequences of GATA-1 deficiency in megakaryocytes and platelets. Blood. 1999;93(9):2867-2875. Vannucchi AM, Bianchi L, Cellai C, et al. Development of myelofibrosis in mice genetically impaired for GATA-1 expression (GATA-1(low) mice). Blood. 2002;100(4): 1123-1132. Spangrude GJ, Lewandowski D, Martelli F, et al. P-Selectin Sustains Extramedullary Hematopoiesis in the Gata1(low) Model of Myelofibrosis. Stem Cells. 2016;34(1):67-82. Castillo-Caro P, Dhanraj S, Haut P, Robertson K, Dror Y, Sharathkumar AA. Proximal radio-ulnar synostosis with bone marrow failure syndrome in an infant without a HOXA11 mutation. J Pediatr Hematol Oncol. 2010;32(6):479-485. Connell KA, Guess MK, Chen H, Andikyan V, Bercik R, Taylor HS. HOXA11 is critical for development and maintenance of uterosacral ligaments and deficient in pelvic prolapse. J Clin Invest. 2008;118(3):10501055. Small KM, Potter SS. Homeotic transformations and limb defects in Hox A11 mutant mice. Genes Dev. 1993;7(12A):2318-2328. Thompson AA, Nguyen LT. Amegakaryocytic thrombocytopenia and radio-ulnar synostosis are associated with HOXA11 mutation. Nat Genet. 2000;26(4): 397-398. Antony-Debre I, Duployez N, Bucci M, et al. Somatic mutations associated with leukemic progression of familial platelet disorder with predisposition to acute myeloid leukemia. Leukemia. 2016;30(4):999-1002. Owen C. Insights into familial platelet disorder with propensity to myeloid malignancy (FPD/AML). Leuk Res. 2010;34(2):141-142. Okuda T, van Deursen J, Hiebert SW, Grosveld G, Downing JR. AML1, the target of multiple chromosomal translocations in human leukemia, is essential for normal fetal liver hematopoiesis. Cell. 1996;84(2):321330. Wang Q, Stacy T, Binder M, Marin-Padilla M, Sharpe AH, Speck NA. Disruption of the Cbfa2 gene causes necrosis and hemorrhaging in the central nervous system and blocks definitive hematopoiesis. Proc Natl Acad Sci USA. 1996;93(8):3444-3449. Growney JD, Shigematsu H, Li Z, et al. Loss of Runx1 perturbs adult hematopoiesis and is associated with a myeloproliferative phenotype. Blood. 2005;106(2):494-504. Ichikawa M, Asai T, Chiba S, Kurokawa M, Ogawa S. Runx1/AML-1 ranks as a master regulator of adult hematopoiesis. Cell Cycle. 2004;3(6):722-724. Ichikawa M, Asai T, Saito T, et al. AML-1 is required for megakaryocytic maturation and lymphocytic differentiation, but not for maintenance of hematopoietic stem cells in

907

C. LĂŠon et al.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

908

adult hematopoiesis. Nat Med. 2004;10(3): 299-304. Putz G, Rosner A, Nuesslein I, Schmitz N, Buchholz F. AML1 deletion in adult mice causes splenomegaly and lymphomas. Oncogene. 2006;25(6):929-939. Ichikawa M, Goyama S, Asai T, et al. AML1/Runx1 negatively regulates quiescent hematopoietic stem cells in adult hematopoiesis. J Immunol. 2008;180(7): 4402-4408. Bluteau D, Gilles L, Hilpert M, et al. Downregulation of the RUNX1-target gene NR4A3 contributes to hematopoiesis deregulation in familial platelet disorder/acute myelogenous leukemia. Blood. 2011;118 (24):6310-6320. Nakagawa M, Shimabe M, WatanabeOkochi N, et al. AML1/RUNX1 functions as a cytoplasmic attenuator of NF-kappaB signaling in the repression of myeloid tumors. Blood. 2011;118(25):6626-6637. Pencovich N, Jaschek R, Dicken J, et al. Cellautonomous function of Runx1 transcriptionally regulates mouse megakaryocytic maturation. PLoS One. 2013;8(5):e64248. Favier R, Jondeau K, Boutard P, et al. ParisTrousseau syndrome : clinical, hematological, molecular data of ten new cases. Thromb Haemost. 2003;90(5):893-897. Barton K, Muthusamy N, Fischer C, et al. The Ets-1 transcription factor is required for the development of natural killer cells in mice. Immunity. 1998;9(4):555-563. Hart A, Melet F, Grossfeld P, et al. Fli-1 is required for murine vascular and megakaryocytic development and is hemizygously deleted in patients with thrombocytopenia. Immunity. 2000;13(2):167-177. Carpinelli MR, Kruse EA, Arhatari BD, et al. Mice Haploinsufficient for Ets1 and Fli1 Display Middle Ear Abnormalities and Model Aspects of Jacobsen Syndrome. Am J Pathol. 2015;185(7):1867-1876. Moussa O, LaRue AC, Abangan RS, Jr., et al. Thrombocytopenia in mice lacking the carboxy-terminal regulatory domain of the Ets transcription factor Fli1. Mol Cell Biol. 2010;30(21):5194-5206. Monteferrario D, Bolar NA, Marneth AE, et al. A dominant-negative GFI1B mutation in the gray platelet syndrome. N Engl J Med. 2014;370(3):245-253. Vassen L, Okayama T, Moroy T. Gfi1b:green fluorescent protein knock-in mice reveal a dynamic expression pattern of Gfi1b during hematopoiesis that is largely complementary to Gfi1. Blood. 2007;109(6):2356-2364. Saleque S, Cameron S, Orkin SH. The zincfinger proto-oncogene Gfi-1b is essential for development of the erythroid and megakaryocytic lineages. Genes Dev.

2002;16(3):301-306. 96. Khandanpour C, Sharif-Askari E, Vassen L, et al. Evidence that growth factor independence 1b regulates dormancy and peripheral blood mobilization of hematopoietic stem cells. Blood. 2010;116(24):5149-5161. 97. Vassen L, Beauchemin H, Lemsaddek W, Krongold J, Trudel M, Moroy T. Growth factor independence 1b (gfi1b) is important for the maturation of erythroid cells and the regulation of embryonic globin expression. PLoS One. 2014;9(5):e96636. 98. Foudi A, Kramer DJ, Qin J, et al. Distinct, strict requirements for Gfi-1b in adult bone marrow red cell and platelet generation. J Exp Med. 2014;211(5):909-927. 99. Zhang MY, Churpek JE, Keel SB, et al. Germline ETV6 mutations in familial thrombocytopenia and hematologic malignancy. Nat Genet. 2015;47(2):180-185. 100. Noetzli L, Lo RW, Lee-Sherick AB, et al. Germline mutations in ETV6 are associated with thrombocytopenia, red cell macrocytosis and predisposition to lymphoblastic leukemia. Nat Genet. 2015;47(5):535-538. 101. Poggi M, Baccini V, Favier M, et al. Mutation in Ets Variant Gene 6 associates with autosomal dominant thrombocytopenia and raised levels of circulating CD34(+) cells. J Thromb Haemost. 2015;13 Suppl S2:5. AS014 abstr. 102. Wang LC, Swat W, Fujiwara Y, et al. The TEL/ETV6 gene is required specifically for hematopoiesis in the bone marrow. Genes Dev. 1998;12(15):2392-2402. 103. Nurden AT, Nurden P. The gray platelet syndrome: clinical spectrum of the disease. Blood Rev. 2007;21(1):21-36. 104. Deppermann C, Cherpokova D, Nurden P, et al. Gray platelet syndrome and defective thrombo-inflammation in Nbeal2-deficient mice. J Clin Invest. 2013 Jul 1. pii: 69210. [Epub ahead of print] 105. Deppermann C, Nurden P, Nurden AT, Nieswandt B, Stegner D. The Nbeal2(-/-) mouse as a model for the gray platelet syndrome. Rare Dis. 2013;1:e26561. 106. Guerrero JA, Bennett C, van der Weyden L, et al. Gray platelet syndrome: proinflammatory megakaryocytes and alpha-granule loss cause myelofibrosis and confer metastasis resistance in mice. Blood. 2014;124(24): 3624-3635. 107. Kahr WH, Lo RW, Li L, et al. Abnormal megakaryocyte development and platelet function in Nbeal2(-/-) mice. Blood. 2013;122(19):3349-3358. 108. Noris P, Perrotta S, Seri M, et al. Mutations in ANKRD26 are responsible for a frequent form of inherited thrombocytopenia: analysis of 78 patients from 21 families. Blood. 2011;117(24):6673-6680. 109. Bera TK, Liu XF, Yamada M, et al. A model for obesity and gigantism due to disruption

of the Ankrd26 gene. Proc Natl Acad Sci USA. 2008;105(1):270-275. 110. Pippucci T, Savoia A, Perrotta S, et al. Mutations in the 5' UTR of ANKRD26, the ankirin repeat domain 26 gene, cause an autosomal-dominant form of inherited thrombocytopenia, THC2. Am J Hum Genet. 2011;88(1):115-120. 111. Bluteau D, Balduini A, Balayn N, et al. Thrombocytopenia-associated mutations in the ANKRD26 regulatory region induce MAPK hyperactivation. J Clin Invest. 2014;124(2):580-591. 112. Zhang Y, Conti MA, Malide D, et al. Mouse models of MYH9-related disease: mutations in nonmuscle myosin II-A. Blood. 2012;119(1):238-250. 113. Massaad MJ, Ramesh N, Geha RS. WiskottAldrich syndrome: a comprehensive review. Ann N Y Acad Sci. 2013;1285:26-43. 114. Stritt S, Nurden P, Turro E, et al. A gain-offunction variant in DIAPH1 causes dominant macrothrombocytopenia and hearing loss. Blood. 2016 Feb 24. pii: blood-2015-10675629. [Epub ahead of print]. 115. Thomas SG, Calaminus SD, Machesky LM, Alberts AS, Watson SP. G-protein coupled and ITAM receptor regulation of the formin FHOD1 through Rho kinase in platelets. J Thromb Haemost. 2011;9(8):1648-1651. 116. Manchev VT, Hilpert M, Berrou E, et al. A new form of macrothrombocytopenia induced by a germ-line mutation in the PRKACG gene. Blood. 2014;124(16):25542563. 117. Dombrowski F, Stieger B, Beuers U. Tauroursodeoxycholic acid inserts the bile salt export pump into canalicular membranes of cholestatic rat liver. Lab Invest. 2006;86(2):166-174. 118. Morison IM, Cramer Borde EM, Cheesman EJ, et al. A mutation of human cytochrome c enhances the intrinsic apoptotic pathway but causes only thrombocytopenia. Nat Genet. 2008;40(4):387-389. 119. De Rocco D, Cerqua C, Goffrini P, et al. Mutations of cytochrome c identified in patients with thrombocytopenia THC4 affect both apoptosis and cellular bioenergetics. Biochim Biophys Acta. 2014;1842(2): 269-274. 120. Hao Z, Duncan GS, Chang CC, et al. Specific ablation of the apoptotic functions of cytochrome C reveals a differential requirement for cytochrome C and Apaf-1 in apoptosis. Cell. 2005;121(4):579-591. 121. Nurden AT, Pillois X, Fiore M, Heilig R, Nurden P. Glanzmann thrombasthenia-like syndromes associated with Macrothrombocytopenias and mutations in the genes encoding the alphaIIbbeta3 integrin. Semin Thromb Hemost. 2011;37(6): 698-706.

haematologica | 2016; 101(8)

ARTICLE

Red Cell Biology & Its Disorders

Functional characterization of novel ABCB6 mutations and their clinical implications in familial pseudohyperkalemia

EUROPEAN HEMATOLOGY ASSOCIATION

Ferrata Storti Foundation

Immacolata Andolfo,1,2 Roberta Russo,1,2 Francesco Manna,1,2 Gianluca De Rosa,1,2 Antonella Gambale,1,2 Soha Zouwail,3 Nicola Detta,2 Catia Lo Pardo,4 Seth L. Alper,5 Carlo Brugnara,6 Alok K. Sharma,5 Lucia De Franceschi,7 and Achille Iolascon1,2

Department of Molecular Medicine and Medical Biotechnologies, “Federico II” University of Naples, Italy; 2CEINGE, Biotecnologie Avanzate, Naples, Italy; 3Department of Biochemistry and Immunology, Cardiff and Vale University Health Board, University Hospital of Wales, Cardiff, UK and Department of Medical Biochemistry, School of Medicine, Alexandria University, Egypt; 4Servizio Immunotrasfusionale, “A. Cardarelli” Hospital, Naples, Italy; 5Division of Nephrology and Center for Vascular Biology Research, Beth Israel Deaconess Medical Center and Department of Medicine, Harvard Medical School, Boston, MA, USA; 6Department of Laboratory Medicine, Boston Children’s Hospital and Department of Pathology, Harvard Medical School, Boston, MA, USA; and 7Department of Medicine, University of Verona, Italy 1

Haematologica 2016 Volume 101(8):909-917

ABSTRACT

solated familial pseudohyperkalemia is a dominant red cell trait characterized by cold-induced ‘passive leak’ of red cell potassium ions into plasma. The causative gene of this condition is ABCB6, which encodes an erythrocyte membrane ABC transporter protein bearing the Langereis blood group antigen system. In this study analyzing three new families, we report the first functional characterization of ABCB6 mutants, including the homozygous mutation V454A, heterozygous mutation R276W, and compound heterozygous mutations R276W and R723Q (in trans). All these mutations are annotated in public databases, suggesting that familial pseudohyperkalemia could be common in the general population. Indeed, we identified variant R276W in one of 327 random blood donors (0.3%). Four weeks' storage of heterozygous R276W blood cells resulted in massive loss of potassium compared to that from healthy control red blood cells. Moreover, measurement of cation flux demonstrated greater loss of potassium or rubidium ions from HEK-293 cells expressing ABCB6 mutants than from cells expressing wild-type ABCB6. The R276W/R723Q mutations elicited greater cellular potassium ion efflux than did the other mutants tested. In conclusion, ABCB6 missense mutations in red blood cells from subjects with familial pseudohyperkalemia show elevated potassium ion efflux. The prevalence of such individuals in the blood donor population is moderate. The fact that storage of blood from these subjects leads to significantly increased levels of potassium in the plasma could have serious clinical implications for neonates and infants receiving large-volume transfusions of whole blood. Genetic tests for familial pseudohyperkalemia could be added to blood donor pre-screening. Further study of ABCB6 function and trafficking could be informative for the study of other pathologies of red blood cell hydration.

Introduction Isolated familial pseudohyperkalemia (FP) is a dominant red cell trait characterized by an increase of plasma potassium ion concentration upon exposure of whole blood to temperatures below 37°C. Red blood cells from individuals with FP have an increased mean corpuscular volume and shape abnormalities. Cation leak in FP haematologica | 2016; 101(8)

Correspondence: andolfo@ceinge.unina.it

Received: January 12, 2016. Accepted: April 29, 2016. Pre-published: May 5, 2016. doi:10.3324/haematol.2016.142372

Check the online version for the most updated information on this article, online supplements, and information on authorship & disclosures: www.haematologica.org/content/101/8/909

©2016 Ferrata Storti Foundation Material published in Haematologica is covered by copyright. All rights reserved to the Ferrata Storti Foundation. Copies of articles are allowed for personal or internal use. Permission in writing from the publisher is required for any other use.

909

I. Andolfo et al.

families described to date has shown several patterns of temperature-dependence.1-4 The causative gene in FP was previously mapped to 2q35-36.5 In three multigenerational FP families functional gene mapping and sequencing analysis of the candidate genes within the 2q35-q36 critical interval identified two novel heterozygous missense mutations in the ABCB6 gene which co-segregated with disease phenotype.6 The two genomic substitutions altered two adjacent nucleotides within codon 375 of ABCB6, a previously identified porphyrin transporter7 that in erythrocyte membranes8 bears the Langereis (Lan) blood group antigen system.9 We previously showed in three separate models of erythropoiesis that ABCB6 expression increased during erythroid differentiation and localized to the plasma membrane.6 The ABCB6 R375Q mutation did not alter levels of mRNA or protein, or subcellular localization in mature erythrocytes or erythroid precursor cells, but was predicted to have pathogenic consequences for protein function. Recently, the ABCB6 substitution R723Q was found in healthy subjects from a family affected by FP, two of whom had been evaluated as regular blood donors.10 The blood of both exhibited increased potassium leak upon storage at temperatures below 37°C. This interesting finding encouraged further study on the implications for neonates and infants of transfusion of whole blood from unknown FP subjects. In this report, we describe three novel missense mutations in ABCB6 identified in FP families with a non-dominant pattern of inheritance of the condition. The presence of these mutations in human variation databases confirms that the prevalence of asymptomatic FP is likely underestimated and, moreover, frequently undetected in blood donor populations. We also report the results of ABCB6 screening in a blood donor population, and present the first functional study of the effects of ABCB6 FP mutations on that component of red cell potassium cation (K+) efflux characterized by resistance to ouabain plus bumetanide.

Methods Patients Three new patients from three independent pedigrees were enrolled in this study (Table 1), and blood samples were obtained from them. Whenever possible, relatives were also investigated. The diagnosis of FP was based on the patients’ history, clinical findings, routine laboratory data, peripheral blood smear, and

genetic testing. Information about every clinical characteristic was not available for all cases. A cohort of 327 blood donors from the blood transfusion center of the Cardarelli Hospital in Naples was enrolled to undergo genetic screening for the ABCB6 mutations found in this study. Patients’ data and samples were collected by the clinicians responsible for the patients’ care, with informed consent according to the Declaration of Helsinki, and with approval by local university ethical committees.

Exome capture and sequencing Blood was obtained for genetic analysis from affected and unaffected family members of the Irish family and from healthy controls, with signed informed consent according to the Declaration of Helsinki. Reads were aligned to the most recent version of the human genome (GRCh37/hg19) using the BWA software package v0.5.9 as previously described11 (see the Online Supplementary Material). Direct sequencing analysis of the additional families was performed (see the Online Supplementary Material).

ABCB6 screening in blood donors Amplification-refractory mutation system (ARMS) analysis, using allele-specific tetra-primer ARMS-polymerase chain reaction primers designed by PRIMER1 (http://primer1.soton.ac.uk/primer1.html), was applied to screen for the ABCB6 variants R276W, V454A, and R723Q in a population of 327 blood donors.12

Bioinformatic modeling of ABCB6 protein structure To assess the potential effects of the identified mutations on protein structure, we generated 3D structural models of dimeric human wild-type (WT) ABCB6 residues 231–827 and the corresponding regions of FP mutant ABCB6 polypeptides V454A, R276W, and R723Q, as described in the legends to Figure 1, and Online Supplementary Figures S1 and S2, and as previously described.6 Sequences were aligned in ClustalW2. MODELLER v9.913 was used for homology modeling in both inward- and outward-facing conformations. The best five structural models with lowest objective function values (as implemented in MODELLER) were subjected to energy minimization in GROMACSv4.5.4.14 Structural models were converged using steepest descent energy minimization with 1,000 steps of step size 0.01 nm. The stereochemical quality of each energy-minimized structure was assessed by PROCHECK.15 The average of three models of highest stereochemical quality was chosen for the ABCB6 structural models. Three dimensional structural models were visualized and aligned using MolMol16 and PyMOL 1.5.0.4 (Schrödinger, LLC).

Table 1. Clinical data of patients with familial pseudohyperkalemia.

Family code Normal range Bolivian Irish Cardiff-2 Lille§ Falkirk§ East London§

Ethnicity

Hb (g/dL)

MCV (fL)

MCH (pg)

K+ (mmol/L) (< 37°C)

South America Ireland United Kingdom France Pakistani Bangladesh

12-18 13.0 14.8-15.8 11.9-12.3 13.7 11.3-14-7 12.0-14.2

80-95 95-98 95-98 116 96 81-106 97

25-31 29.3 33.7 32.3 30.7 34.3 32.6

3.5-4.5 8.3-8.7 6.7-7.6 8.0-8.6 4.0-5.5 7.0-8.2 7.0-13.7

These families were previously described.6 Hb: hemoglobin; MCV: mean corpuscular volume; MCH: mean corpuscular hemoglobin.

910

haematologica | 2016; 101(8)

Functional characterization of ABCB6 mutations in familial pseudohyperkalemia

Figure 1. Three-dimensional structural model of human ABCB6 mutations. (A) Three-dimensional structural model of a portion of homodimeric WT human ABCB6 in an inward-facing conformation, as modeled on the aligned structure of M. musculus ABCB1A (PDB ID 3G5U). Monomer "a" (blue) of the homodimer represents ABCB6 amino acid (aa) residues 246 (N-) to 826 (-C), modeled on transmembrane helices 1-6 and NBD-1 of ABCB1A. Monomer "b" (pink) of the homodimer represents ABCB6 aa residues 237 (N’-) to 826 (-C’) modeled on ABCB1A transmembrane helices 7-12 and its NBD-2. A surface model is superposed on the modeled polypeptide backbone ribbon structure. The dehydrated hereditary stomatocytosis (DHSt) homozygous mutation site V454 (red spheres) is located between the membrane-spanning helices and the NBD, extending into the cytoplasmic vestibule of the dimer. Locations of the compound heterozygous DHSt mutation sites R276 (magenta spheres)/R723 (olive spheres) are also shown. Arrows mark R276 of monomer A (located within the lipid bilayer) and R723 of monomer B (located within the NBD region in the cytoplasmic vestibule) of the dimer. The cavity (cyan spheres) at the intermonomeric interface outlines a postulated intra-membrane binding site for inhibitors of ABCB6-mediated porphyrin transport,32 corresponding to the ABCB1 binding site of inhibitor QZ59.33 In this and subsequent figures, each modeled ABCB6 monomer lacks its ectofacial N-terminal tail and putative transmembrane spans 1-5, but includes putative transmembrane spans 6-11 followed by the single NBD. (B) Transverse intra-membrane profile of the modeled inward-facing conformation of dimeric WT ABCB6 (as in panel A), with transmembrane helices rotated 90° around the axis shown. The view (lacking NBD) looks outward from the ICL region, near the site of separated mutation site V454 (red) and further from mutation site R276. The colored M1 domain helices are numbered 6-11 for ABCB6 monomer "a", and 6'-11' for monomer "b" of the ABCB6 dimer. The arrows between helices 9 and 11 on one side, and helices 9' and 11' on the other side of the dimer mark the locations of side apertures proposed in mouse ABCB1 to mediate hydrophobic drug uptake from the inner leaflet of the lipid bilayer for subsequent efflux from the cell, or for flippase-like transfer to the outer leaflet.33 (C) Three-dimensional structural model of homodimeric WT human ABCB6 in an outward-facing conformation, as modeled on the aligned structure of S. aureus Sav1866 (PDB ID 2HYD). The black oval encloses a central cavity at the inter-monomeric interface, hypothesized to be an intra-membrane substrate binding site (as predicted for homodimeric Sav1866 of S. aureus34). Sites of homozygous and compound heterozygous mutations are shown using similar colored spheres as in panel A. (D) Transverse intramembranous profile of the modeled outward-facing conformation of dimeric ABCB6 (as in panel C), with the transmembrane helices rotated 90° around the axis shown. The view (lacking NBD) looks inward from the extracellular edge of the outer leaflet of the membrane bilayer towards the approximated mutation sites V454 (at the level of the ICL region) and R276 (further from the ICL region); color scheme as in panel C. Helices are labeled at ends closest to reader. The figure was prepared in PyMOL. NBD: nucleotide-binding domain; TMD: transmembrane domain; ICL: intracytoplasmic loop.

haematologica | 2016; 101(8)

911

I. Andolfo et al. Table 2. Mutations found in patients with familial pseudohyperkalemia.

Family code Bolivian Cardiff-2 Irish Lille Falkirk East London

ABCB6 mutations

SNP ID

MAFa

PolyPhen2/ SIFT scores

References

c.1361T>C; p.V454A* c.826C>T; p. R276W§ c.2168G>A; p.R723Q§ c.826C>T; p. R276W c.1123 C>T; p. R375Q c.1124 G>A; p.R375W c.1124 G>A; p.R375W

rs61733629 rs57467915 rs148211042 rs61733625 Not annotated Not annotated Not annotated

0.43% (C) 1.5% (T) 0.08% (A) 1.5% (T) -

0.996/0 1/0 0.997/0 1/0 1/0 1/0 1/0

unpublished data unpublished data unpublished data Andolfo et al. 2013 Andolfo et al. 2013 Andolfo et al. 2013

*Mutations in homozygous state. §The two mutations are in trans (see results section for details). aOverall minor allele frequencies (MAF) estimated from public databases 1000 Genomes (URL: http://browser.1000genomes.org);20 NHLBI Exome Sequencing Project (URL: http://evs.gs.washington.edu/EVS); and Exome Aggregation Consortium, Cambridge, MA (URL: http://exac.broadinstitute.org); for details see Online Supplementary Table 2S.

Cell culture and transfection assay Human HEK-293 cells were maintained in DMEM (Sigma) supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 mg/mL streptomycin (all from Life Technologies), in a humidified 5% CO2 atmosphere at 37°C. pcDNA3.1-ABCB6-WT and pcDNA3.1-ABCB6 mutant constructs (5 μg) were transfected into HEK-293 cells using XtremeGENE HP DNA Transfection Reagent (Roche, Indianapolis, IN, USA). To phenocopy the heterozygous genotypes, cells were transfected with 2.5 μg of pcDNA3.1-ABCB6-WT plus 2.5 μg of pcDNA3.1-ABCB6 mutants R375Q, p.R276W or R375W. For the compound heterozygous genotypes, cells were transfected with 2.5 μg pcDNA3.1-ABCB6-R276W plus 2.5 μg pcDNA3.1-ABCB6R723Q. For the homozygous genotype, cells were transfected with 5 μg pcDNA3.1-ABCB6-V454A. After 72 h, cells were harvested for analysis.

Immunofluorescence analysis HEK-293 cells (2x106) on coverslips were transfected with ABCB6 cDNA as previously described6 (see the Online Supplementary Methods).

Table 3. Root-mean-square deviations (Å) of superposed homologymodeled structures of the indicated patient-derived mutant ABCB6 dimers with the modeled wild-type ABCB6 homodimer.

Mutant ABCB6

Inward-facing

Outward-facing

V454A/V454A R276Wa/R723Qb R276Wb/R723Qa R276W/R276W R723Q/R723Q

1.48 1.88 1.57 0.98 1.17

0.81 0.58 0.74 0.29 0.79

150 mM choline chloride, 1 mM MgCl2, and 10 mM Tris MOPS, then lysed in order to determined intracellular Rb concentration. Intracellular Rb content and extracellular K+ concentration were measured in triplicate by atomic absorption spectroscopy (ANALYST 2000, Perkin-Elmer) as previously described.17

Results Measurements of potassium ion fluxes in red blood cells from the blood donor carrying the ABCB6 R276W variant Blood samples from the donor carrying the ABCB6 R276W mutation and from two controls obtained from the transfusion center of the Cardarelli Hospital (Naples) were stored for 4 weeks at 4°C in citrate-phosphate-dextrose solution as anticoagulant, under blood bank conditions. During the 4 weeks of storage, plasma potassium levels were measured in triplicate by atomic absorption spectroscopy (ANALYST 2000, Perkin-Elmer) as previously described.17 The red blood cells were washed gently with a buffer containing 150 mM choline chloride, 1 mM MgCl2, and 10 mM Tris MOPS, then lysed to measure intracellular potassium by atomic absorption spectroscopy.17 The free hemoglobin levels were measured to evaluate the degree of hemolysis as for potassium.

Measurements of ouabain-plus-bumetanide-resistant rubidium and potassium ion fluxes in transfected HEK-293 cells At 72 h after transfection, HEK-293 cells were maintained for 8 h under shear stress (rotary shaking at 0.12g at 30°C) in a K+-free medium containing 140 mM NaCl, 5 mM RbCl, 2 mM CaCl2, 1 mM MgCl2, 10 mM HEPES, 10 mM glucose, to which 10 μM ouabain and 10 μM bumetanide were added. At the end of the incubation, cell viability was determined by trypan blue staining and medium was removed for determination of the extracellular K+ concentration. Cells were washed gently in buffer containing 912

Case reports The Bolivian patient is a 41-year old female from a consanguineous family. Multiple outpatient blood samples indicated elevated plasma K+ (8.3 mmol/L). The patient’s renal and adrenal function and electrocardiogram were normal, as were hematologic indices except for macrocytosis (mean corpuscular volume 95-98 fL; Table 1). The patient’s past medical history included premature menopause and migraines. Physical examination was unremarkable. Patient Cardiff-2 is a 35-year old female with a prior tentative diagnosis of hereditary spherocytosis with a positive family history, despite a normal red cell eosin maleimide binding test. Outpatient values of plasma K+ concentration ranged between 8.0 and 8.6 mmol/L, in contrast to hospital clinical values which were between 5.4 and 5.9 mmol/L (Table 1). The patient had been under outpatient care for diabetes and other conditions since 1999. Her peripheral blood smear revealed polychromasia, target cells, and a few spherocytes. Her past medical history included irritable bowel syndrome, oophorectomy in 2007, splenectomy secondary to trauma in 2008, and a diagnosis of depression in 2009. Medications included omeprazole, mebevirine and citalopram. The patient’s blood pressure was normal, as was the remainder of the physical examination. Her absolute reticulocyte count was haematologica | 2016; 101(8)

Functional characterization of ABCB6 mutations in familial pseudohyperkalemia 176.0 x109 (4.10 %) with a mean reticulocyte volume of 118 fL (reference range, 90-110 fL). The large Irish family was originally described by Stewart and colleagues18 as having autosomal dominant dehydrated hereditary stomatocytosis with FP. The propositus had several thrombotic episodes following splenectomy at the age of 40 years. An increased passive K+ leak was noted (Table 1). The families Lille, Falkirk and East London have already been described.6 Affected individuals from the FP Lille family (of Flemish descent) had normal hematologic indices except for a slightly elevated mean corpuscular volume. Affected individuals from the FP Falkirk family (of Pakistani origin) also had macrocytosis. Affected individuals from the FP East London family (of Bangladeshi origin) were anemic and had hyperkalemia, but in the absence of reticulocytosis and jaundice were considered nonhemolytic. None of the carriers had colobomatous abnormalities of iris or retina, also associated with missense ABCB6 mutations. Carriers were not tested for Lan-/- status, as this phenotype is caused by nonsense mutations that cause complete absence of ABCB6 polypeptide in circulating red cells.

patient Cardiff-2 we found compound heterozygosity for the two mutations c.826G>T; p.R276W and c.2168G>A; p.R723Q (Table 2). The parents of both patients were unavailable for genetic analysis. To analyze the allelic pattern of the two mutations in patient Cardiff-2, we cloned the genomic region encompassing both ABCB6 variants (about 6 Kb) into a plasmid vector. DNA sequencing of this cloned region demonstrated that the two mutations are in trans in patient Cardiff-2. In the Bolivian patient, the presence of heterozygous single nucleotide polymorphisms excluded the possibility of a deletion within the region of the gene harboring the mutation. The abnormality in the Irish family originally diagnosed as having dominant dehydrated hereditary stomatocytosis plus FP was at first mapped to chromosome 1619 but subsequent analysis of the PIEZO1 gene was negative. We, therefore, subjected the Irish family to whole-exome analysis and identified ABCB6 variant c.826G>T; p.R276W, with subsequent confirmation by direct sequencing. Although there was probably a mis-assignment in the Irish pedigree linkage analysis, it cannot be ruled out that there are intronic mutations that could explain epistasis between PIEZO1 and ABCB6. ABCB6 amino acid residues R276, R723 and V454 are conserved in all species analyzed and each have Polyphen2 scores of 1 (damaging) and SIFT scores of 0 (damaging). Each of the three FP mutations is annotated in public databases: 1000 Genomes (URL:

ABCB6 mutational analysis in families with familial pseudohyperkalemia and blood donor screening We sequenced the ABCB6 gene in two patients and one family with FP. In the Bolivian patient we found the homozygous mutation c.1361T>C; p.V454A, while in

Figure 2. Expression and localization of ABCB6 mutants. (A) ABCB6 mRNA levels in HEK-293 cells transfected with ABCB6 WT and mutants and empty vector as control. Values are means ÂąSEM of three independent experiments. *P<0.001 WT, WT/R375Q, WT/R375W, WT/R276W, V454A/V454A, WT/R723Q, R276W/R723Q vs empty vector. (B) Immunoblot showing ABCB6 Flag protein expression in HEK-293 cells transfected with FLAG-tagged WT or mutant ABCB6 variants, or with empty vector as control and GAPDH as loading control. One of two similar experiments. (C) Laser-scanning confocal microscopy images of HEK-293 cells transfected with WT or mutant ABCB6 variants, or with empty vector as control, analyzed by immunofluorescence with rabbit polyclonal anti-ABCB6 antibody (green) and WGA (membrane marker, staining both the nuclear envelope and the plasma membrane, red), with the merged signal showing regions of co-localization in yellow (white arrows indicate the yellow regions in the merge). Cells were imaged with a Zeiss LSM 510 meta confocal microscope equipped with a 1.4 NA oil immersion plan Apochromat 100Ă&#x2014; objective. Intensity and contrast were adjusted with Axiovision software. Representative of three independent experiments.

haematologica | 2016; 101(8)

913

I. Andolfo et al.

http://browser.1000genomes.org);20 NHLBI Exome Sequencing Project (URL: http://evs.gs.washington.edu/EVS); and Exome Aggregation Consortium, Cambridge, MA (URL: http://exac.broadinstitute.org). The minor allele frequency was 0.43% for the V454A variant, 0.08% for R723Q, and 1.5% for R276W (Table 2 and Online Supplementary Table S2). The high frequency of the variants found in our patients prompted our genetic screening of a cohort of 327 blood donors. Of note, our analysis demonstrated the presence of variant R276W in 0.3% of this cohort (1/327) and the absence of the other two mutations V454A and R723Q, consistent with the minor allele frequencies reported to date.

ABCB6 mutations produce conformational changes in model structures To analyze potential consequences of the identified mutations on protein structure, we generated threedimensional structural models of the (putatively dimeric) human WT ABCB6 residues 231-827 and FP mutant polypeptides V454A, R276W, and R723Q (see Methods). Figure 1 shows three-dimensional structural models of homodimeric WT ABCB6 in inward- and outward-facing conformations, highlighting sites of the homozygous and compound heterozygous FP missense mutations investi-

gated in this study. The structural models of ABCB6 homodimeric FP mutant V454A and heterodimeric mutant R276W (chain a)/R723Q (chain b) in both inward- and outward-facing conformations are presented in Online Supplementary Figures S1 and S2, respectively. Transverse views of intra-membrane bilayer regions are also presented. Comparison of WT with mutant models revealed that these mutations cause detectable conformational changes in regions on or near the missense substitution sites and at several more remote locations. In inward-facing models of homodimeric FP mutant V454A, the presence of Ala decreased by 2.3 Å the WT Cα-Cα interatomic distance between chain a residue 454 and chain b residue 454. In contrast, this change was minimal in the outward-facing conformation. The WT loop structure at amino acids 362367 (packed adjacent to chain b residue 454 in the inwardfacing conformation) underwent a partial loop-to-helix transition in the homodimeric FP mutant V454A. Furthermore, the WT Cα-Cα interatomic distance between chain a residue 276 and remote chain b residue 723 decreased by 3.3 Å in the heterodimeric FP mutant R276W (chain a)/R723Q (chain b). The inward-facing conformation of this heterodimeric mutant also induced a loop-tohelix transition of chain a residues 408-409 at the ecto-end of a transmembrane helix and spatially adjacent to chain a missense substitution R276W. These amino acid substitutions also modestly altered interhelical distances near the

C Figure 3. ABCB6 protein expression and potassium efflux in red blood cells of a blood donor carrying the ABCB6 R276W variant. (A) Plasma K+ content (expressed as mmol/L of whole blood) of blood from a donor heterozygous for the ABCB6 mutation R276W and from two healthy controls after 0, 7, 14, 21 and 28 days of cold storage under blood banking conditions. *P<0.01 donor vs. two healthy controls. (B) Intracellular K+ content (expressed as mmol*1013 cells) of blood from a donor heterozygous for R276W and from two healthy controls after 0, 7, 14, 21 and 28 days of cold storage as in (A). Ion contents in (A) and (B) were measured by atomic absorption spectrometry, and represent means ± SEM of three independent experiments. *P<0.01 donor vs. two healthy controls. (C) Immunoblot showing ABCB6 protein expression in RBC membranes from a blood donor heterozygous for mutation R276W and pooled membranes from two healthy controls. β-actin was used as the loading control. One of three similar experiments.

914

haematologica | 2016; 101(8)

Functional characterization of ABCB6 mutations in familial pseudohyperkalemia mutation sites. Structural superposition of modeled WT polypeptide with each modeled mutant polypeptide revealed larger global structural deviations of mutant polypeptides in inward-facing conformations than in outward-facing ones (Table 3 and Online Supplementary Table S1). Modeled heterodimeric R276W/R723Q and homodimeric V454A mutant polypeptides exhibited greater structural deviation from WT than did homodimeric mutants R276W or R723Q (Table 3, Online Supplementary Table S1).

ABCB6 mutations cause no alteration of expression or cellular localization We modeled our patients’ genotypes in vitro by transient transfection of WT and mutant ABCB6 expression plasmids into HEK-293 cells. No significant differences between mutant and WT mRNA accumulation were evident 72 h after transfection (Figure 2A). Similarly, immunoblot analysis of heterologous FLAG tag confirmed equivalent accumulation of WT and mutant heterologous ABCB6 polypeptides (Figure 2B). We also tested effects of the mutations on ABCB6 membrane localization. Confocal microscopy analysis showed that all mutant polypeptides were expressed predominantly at the peripheral membrane of HEK-293 cells, as demonstrated by co-localization of ABCB6-FLAG with the lectin membrane marker, wheat germ agglutinin (Figure 2C).

ABCB6 mutation R276W increases potassium efflux from red blood cells of a blood donor A blood sample from the blood donor heterozygous for ABCB6 variant and samples from two control donors were obtained and stored for 4 weeks at 4°C under blood banking conditions. Extracellular and intracellular potassium levels were measured throughout the storage period. As shown in Figure 3A, the potassium efflux of donor’s blood after 28 days of storage was about 3.5-fold higher than that of the controls. Correspondingly, the intracellu-

lar red blood cell potassium content was about 2.5-fold lower than that of the controls (Figure 3B). The degree of hemolysis over time for the blood samples was evaluated during storage by measurement of free hemoglobin levels and was the same for three samples (data not shown). The data demonstrated that the physiological consequences of the blood donor's mutation is greater potassium efflux than in controls, and similar potassium efflux to that observed in FP patients. Moreover, immunoblot analysis of red blood cells from the blood donor carrying the R276W mutation demonstrated that the expression of ABCB6 did not differ between this donor with a mutation and healthy controls (Figure 3C).

ABCB6 mutations cause cation flux alterations We next evaluated cell potassium content in HEK-293 cells over-expressing WT ABCB6 and different ABCB6 mutant variants. Preliminary experiments comparing cells maintained for 8 h at 37°C or 30°C revealed no differences (data not shown). To mimic shipping conditions (critical for the alteration in serum K+ concentrations observed in the FP patients, see the Case report section) we exposed HEK293 cells over-expressing WT or mutant ABCB6 variants to 0.12 g rotary shaking at 30°C for 8 h. As shown in Figure 4A, levels of extracellular potassium in media from HEK-293 cells over-expressing mutant ABCB6 variants were significantly higher than the levels for cells expressing WT ABCB6. Correspondingly, residual intracellular Rb content was significantly reduced in cells expressing three of the ABCB6 mutant genotypes, WT/R375Q, WT/R375W, V454A/V454A, as well as in the doublemutant R276W/R723Q, compared to either WT ABCB6 or the other ABCB6 variants (WT/R273Q, WT/R276W) (Figure 4B). These data show different impacts of individual ABCB6 mutations on cellular K+ efflux insensitive to ouabain plus bumetanide, and an incrementally increased effect on cell K+ efflux of co-expression of the compound heterozygous ABCB6 mutations R276W/R723Q.

Figure 4. Analysis of potassium efflux of ABCB6 mutants. (A) K+ content of extracellular medium sampled from cultures of cells overexpressing ABCB6 WT or ABCB6 FP mutants. Ion contents measured by atomic absorption spectrometry are expressed as mmol/mg protein. **P<0.001 R276W/R723Q vs. empty vector and WT; *P<0.05 for WT/R375Q, WT/R375W, WT/R276W, V454A/V454A, WT/R723Q vs. WT. (B) Rb content of cells overexpressing ABCB6 WT and ABCB6 FP mutants, expressed as mmol/mg protein. Values in (A) and (B) are means ±SEM of four independent experiments. haematologica | 2016; 101(8)

915

I. Andolfo et al.

Discussion Here we report three new mutations in the FP-disease gene ABCB6. FP had been described previously as a dominant condition, but, for the first time, we report two FP patients with homozygous or compound heterozygous mutations, both novel patterns of inheritance for FP. Of note, those patients homozygous and compound heterozygous for ABCB6 mutations showed higher plasma K+ concentrations than heterozygous patients. Moreover, the compound heterozygote also had a greater mean corpuscular volume than that of other patients. FP inheritance patterns thus constitute a crucial part of the diagnostic evaluation of patients. ABCB6 variations are more common than previously predicted, as also reported for Lan- blood group carriers with ABCB6 nonsense mutations causing the ABCB6-null red cell phenotype. Through screening erythroid ABCB6 expression, Koszarska et al. found an unexpectedly high frequency of Lan mutations in healthy individuals.21 Indeed, in public databases (1000 Genomes, NHLBI Exome Sequencing Project, Exome Aggregation Consortium) the allele frequencies of V454A, R723Q and R276W are 0.43%, 0.08% and 1.5%, respectively. Moreover, the high frequency of ABCB6 variations in FP, including two FP patients found in a Cardiff blood donor cohort recently described by Bawazir et al.,22 has clinical implications for blood transfusion screening and practice. Our own screening of 327 blood donors of different geographical and ethnic origin corroborates this observation, since we found the R276W mutation in 0.3% of our cohort. Our analysis of potassium efflux from the blood donor red blood cells under blood banking storage conditions confirmed the cation leak shown by FP patients. Refrigerated storage of blood from FP patients causes rapid loss of potassium, while the extracellular potassium content of bags of stored cells increases during storage. This is of little consequence for the majority of transfusions, since the total amount of potassium transfused is relatively small compared to the total blood volume of the recipient. In contrast, this extracellular potassium can have serious or fatal consequences in neonates and infants given whole blood transfusions of large volume proportionate to body size. Several cases of whole blood transfusion leading to cardiac arrest and death in infants have been described.23-27 The ABCB6 FP mutants overexpressed in HEK-293 cells showed no difference in accumulation of mRNA or protein, or in peripheral membrane immunolocalization as compared to WT ABCB6, and as previously demonstrated for the ABCB6 FP variant R375Q. Consistent with these findings, in silico modeled threedimensional structures of these mutant ABCB6 polypeptide dimers predicted modest structural alterations of transmembrane and cytosolic ATP binding domains in both inward- and outward-facing conformations. Prediction of the consequences of these structural alterations to the cation leak process remains uncertain, since the relationship between the mutant cation leak and the (proposed but still debated) WT transport of porphyrins is still poorly understood. Future simulation studies of molecular dynamics in a model lipid bilayer across

916

microsecond time-scales will expand our understanding of the impact of these FP mutants on the structure, and possibly the function, of ABCB6. To further characterize the role of ABCB6 mutants, we tested the hypothesis that their expression could modify K+ transport in HEK-293 cells in a manner similar to the altered K+ efflux in FP red blood cells. We found that cells expressing each mutant variant tested exhibited increased K+ efflux compared to that of the WT cells. Co-expression in HEK-293 cells of the two mutant variants expressed by patient Cardiff in a compound heterozygous form produced the highest value of K+ efflux among all tested mutants. These data demonstrate that the new mutations, whether homozygous or compound heterozygous, act at the cellular level as gain-of-function mutations. Among ABC proteins, only the cystic fibrosis transmembrane regulator CFTR/ABCC7 is known to mediate ion channel function. However, several ABC proteins, in addition to CFTR, function as ion channel regulators,28-30 including the Kir6 KATP channel regulatory subunits, sulphonylurea receptors SUR1, SUR2A, and SUR2B.31 It is unclear whether ABCB6 FP mutant polypeptides generate intrinsic cation leak pathways in membranes of red cells (or experimentally in HEK-293 cells), or might secondarily dysregulate one or more endogenous membrane cation permeability pathways in red cells (or HEK-293 cells). The negative results obtained to date in our electrophysiological studies conducted in HEK-293 cells and Xenopus laevis oocytes expressing WT or mutant ABCB6 variants (data not shown) encourage further consideration of dysregulated endogenous electroneutral (or low-level electrogenic) transporters as cation leak mediators in FP red cells or cell models of heterologous expression. Our findings demonstrate that both heterozygous and homozygous missense mutations in ABCB6 lead to increased efflux of cellular K+ from HEK-293 cells, a property shared with red blood cells from FP patients. Screening for the most frequently found ABCB6 variant, R276W, confirmed that patients with FP are relatively common in the blood donor population. Storage of FP blood can cause a significant increase in whole blood K+ levels, with serious clinical implications for neonates and infants receiving large-volume transfusions of whole blood. For these reasons, we endorse the proposal to conduct genetic screening for ABCB6 FP mutations among potential blood donors, especially when whole blood is needed. Finally, investigation of ABCB6 may contribute to our understanding of other pathologies of red blood cell hydration, such as sickle cell anemia. Acknowledgments: This work was supported by grants from the Italian Ministero dellâ&#x20AC;&#x2122;UniversitĂ e della Ricerca, by grants MUR-PS 35-126/Ind, by grants from Regione Campania (DGRC2362/07), by EU Contract LSHM-CT-2006-037296, by PRIN to AI and LDF (20128PNX83), Italy. The authors thank the CEINGE Service Facility platforms including the Sequencing Core, the Rheology Facility, the FACS Core Laboratory and the Dynamic Imaging Facility (particularly Dr Daniela Sarnataro for providing helpful technical support).

haematologica | 2016; 101(8)

Functional characterization of ABCB6 mutations in familial pseudohyperkalemia

References 1. Stewart GW, Corrall RJ, Fyffe JA, et al. Familial pseudohyperkalemia. A new syndrome. Lancet. 1979;2(8135):175-177. 2. Dagher G, Vantyghem MC, Doise B, et al. Altered erythrocyte cation permeability in familial pseudohyperkalemia. Clin Sci (Lond). 1989;77(2):213-216. 3. Vantyghem MC, Dagher G, Doise B, et al. Pseudo-hyperkalemia. Apropos of a familial case. Ann Endocrinol (Paris). 1991;52 (2):104-108. 4. Haines PG, Crawley C, Chetty MC, et al. Familial pseudohyperkalemia Chiswick: a novel congenital thermotropic variant of K and Na transport across the human red cell membrane. Br J Haematol. 2001;112(2): 469-474. 5. Carella M, d’Adamo AP, GrootenboerMignot S, et al. A second locus mapping to 2q35-36 for familial pseudohyperkalemia. Eur J Hum Genet. 2004;12(12):1073–1076. 6. Andolfo I, Alper SL, Delaunay J, et al. Missense mutations in the ABCB6 transporter cause dominant familial pseudohyperkalemia. Am J Hematol. 2013;88(1): 66-72. 7. Krishnamurthy PC, Du G, Fukuda Y, et al. Identification of a mammalian mitochondrial porphyrin transporter. Nature. 2006;443 (7111):586-589. 8. Kiss K, Brozik A, Kucsma N, et al. Shifting the paradigm: the putative mitochondrial protein ABCB6 resides in the lysosomes of cells and in the plasma membrane of erythrocytes. PLoS One. 2012;7(5):e37378. 9. Helias V, Saison C, Ballif BA, et al. ABCB6 is dispensable for erythropoiesis and specifies the new blood group system Langereis. Nat Genet. 2012;44(2):170–173. 10. Bawazir WM, Flatt JF, Wallis JP, et al. Familial pseudohyperkalemia in blood donors: a novel mutation with implications for transfusion practice. Transfusion. 2014; 54(12):3043-3050. 11. Andolfo I, Alper SL, De Franceschi L, et al. Multiple clinical forms of dehydrated hereditary stomatocytosis arise from mutations in PIEZO1. Blood. 2013;121(19):3925-3935. 12. Ye S, Dhillon S, Ke X, Collins AR, Day IN.

haematologica | 2016; 101(8)

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

An efficient procedure for genotyping single nucleotide polymorphisms. Nucleic Acids Res. 2001;29(17):E88-8. Sali A, Potterton L, Yuan F, van Vlijmen H, Karplus M. Evaluation of comparative protein modeling by MODELLER. Proteins. 1995;23(3):318-326. Van Der Spoel D, Lindahl E, Hess B, Groenhof G, Mark AE, Berendsen HJ. GROMACS: fast, flexible, and free. J Comput Chem. 2005;26(16):1701-1718. Vaguine AA, Richelle J, Wodak SJ. SFCHECK: a unified set of procedures for evaluating the quality of macromolecular structure-factor data and their agreement with the atomic model. Acta Crystallogr D Biol Crystallogr. 1999;55(Pt 1):191-205. Koradi R1, Billeter M, Wüthrich K. MOLMOL: a program for display and analysis of macromolecular structures. J Mol Graph. 1996;14(1):51-55. De Franceschi L, Ronzoni L, Cappellini MD, et al. K-CL co-transport plays an important role in normal and beta thalassemic erythropoiesis. Haematologica. 2007;92(10):1319-1326. Stewart GW, Amess JA, Eber SW, et al. Thrombo-embolic disease after splenectomy for hereditary stomatocytosis. Br J Haematol. 1996;93(2):303-310. Carella M, Stewart G, Ajetunmobi JF, et al. Genomewide search for dehydrated hereditary stomatocytosis (hereditary xerocytosis): mapping of locus to chromosome 16 (16q23-qter). Am J Hum Genet. 1998;63 (3):810-816. 1000 Genomes Project Consortium, Auton A, Brooks LD, et al. A global reference for human genetic variation. Nature. 2015; 526(7571):68-74. Koszarska M, Kucsma N, Kiss K, et al. Screening the expression of ABCB6 in erythrocytes reveals an unexpectedly high frequency of Lan mutations in healthy individuals. PLoS One. 2014;9(10):e111590. Bawazir WM, Flatt JF, Wallis JP, et al. Familial pseudohyperkalemia in blood donors: a novel mutation with implications for transfusion practice. Transfusion. 2014;54(12):3043-3050. Hall TL, Barnes A, Miller JR, et al. Neonatal

24.

25.

26.

27.

28. 29.

30.

31.

32.

33.

34.

mortality following transfusion of red cells with high plasma potassium levels. Transfusion. 1993;33(7):606-609. Chen CH, Hong CL, Kau YC, et al. Fatal hyperkalemia during rapid and massive blood transfusion in a child undergoing hip surgery: a case report. Acta Anaesthesiol Sin. 1999;37(3):163-166. Baz EM, Kanazi GE, Mahfouz RA, et al. An unusual case of hyperkalaemia-induced cardiac arrest in a paediatric patient during transfusion of a “fresh” 6-day-old blood unit. Transfus Med. 2002;12(6):383-386. Smith HM, Farrow SJ, Ackerman JD, et al. Cardiac arrests associated with hyperkalemia during red blood cell transfusion: a case series. Anesth Analg. 2008; 106(4):1062-1069. Lee AC, Reduque LL, Luban NL, et al. Transfusion associated hyperkalemic cardiac arrest in pediatric patients receiving massive transfusion. Transfusion. 2014;54 (1):244-254. Higgins CF. The ABC of channel regulation. Cell. 1995;82(5):693-669. Welsh MJ, Anderson MP, Rich DP, et al. Cystic fibrosis transmembrane conductance regulator: a chloride channel with novel regulation. Neuron. 1992;8(5):821829. Inagaki N, Gonoi T, Clement JP, et al. Reconstitution of IKATP: an inward rectifier subunit plus the sulfonylurea receptor. Science. 1995;270(5239):1166-1170. Cui Y, Giblin JP, Clapp LH, Tinker A. A mechanism for ATP-sensitive potassium channel diversity: functional coassembly of two pore-forming subunits. Proc Natl Acad Sci USA. 2001;98(2):729-734. Polireddy K, Khan MM, Chavan H, et al. A novel flow cytometric HTS assay reveals functional modulators of ATP binding cassette transporter ABCB6. PLoS One. 2012; 7(7):e40005. Aller SG, Yu J, Ward A, et al. Structure of Pglycoprotein reveals a molecular basis for poly-specific drug binding. Science. 2009;323(5922):1718-1722. Dawson RJ, Locher KP. Structure of a bacterial multidrug ABC transporter. Nature. 2006;443(7108):180-185.

917

ARTICLE EUROPEAN HEMATOLOGY ASSOCIATION

Myelodysplastic Syndromes

Ferrata Storti Foundation

Haematologica 2016 Volume 101(8):918-925

A randomized phase II trial of azacitidine +/- epoetin-β in lower-risk myelodysplastic syndromes resistant to erythropoietic stimulating agents

Sylvain Thépot,1* Raouf Ben Abdelali,2* Sylvie Chevret,3 Aline Renneville,2 Odile Beyne-Rauzy,4 Thomas Prébet,5 Sophie Park,6 Aspasia Stamatoullas,7 Agnes Guerci-Bresler,8 Stéphane Cheze,9 Gérard Tertian,10 Bachra Choufi,11 Laurence Legros,12 Jean Noel Bastié,13 Jacques Delaunay,14 Marie Pierre Chaury,15 Laurence Sanhes,16 Eric Wattel,17 Francois Dreyfus,6 Norbert Vey,5 Fatiha Chermat,18 Claude Preudhomme,2 Pierre Fenaux19 and Claude Gardin1 on behalf of the Groupe Francophone des Myélodysplasies (GFM)

Service d'Hématologie Clinique, Hôpital Avicenne, Assistance Publique–Hôpitaux de Paris (AP-HP), and Université Paris 13, Bobigny; 2Laboratoire d’hématologie,CHRU de Lille; 3Service de biostatistique et information médicale, Hôpital Saint-Louis, AP-HP and Université Paris 7; 4Service d'Hématologie Clinique, Centre Hospitalier Universitaire, Toulouse; 5Département d'Hématologie, Institut Paoli-Calmettes, Marseille; 6Service d'Hématologie Clinique, Hôpital Cochin, AP-HP and Université Paris 5; 7Service d'Hématologie, Centre Henri Becquerel, Rouen; 8Service d'Hématologie Clinique, Centre Hospitalier Universitaire, Nancy; 9Service d'Hématologie Clinique, Centre Hospitalier Universitaire, Caen; 10Service de médecine interne, Hôpital du kremlin-Bicetre, AP-HP and Université Paris; 11Service d'Hématologie Clinique, Hôpital de Boulogne sur mer; 12 Service d'Hématologie, Centre Hospitalier Universitaire, Nice; 13Service d'Hématologie Clinique, Centre Hospitalier Universitaire, Dijon; 14Service d'Hématologie, Centre Hospitalier Universitaire, Nantes; 15Service d'Hématologie Clinique, Centre Hospitalier Universitaire, Limoges; 16Service d'Hématologie, Centre Hospitalier, Perpignan; 17Service d'Hématologie, Hôpital Universitaire Lyon Sud, Lyon; 18Groupe francophone des Myélodysplasies, Hôpital Saint-Louis, AP-HP; and 19Service d'Hématologie Clinique senior, Hôpital Saint-Louis, AP-HP and Université Paris 7, France 1

*ST and RBA contributed equally to this work.

Correspondence: claude.gardin@avc.aphp.fr

Received: December 21, 2015. Accepted: May 19, 2016. Pre-published: May 26, 2016. doi:10.3324/haematol.2015.140988

Check the online version for the most updated information on this article, online supplements, and information on authorship & disclosures: www.haematologica.org/content/101/8/918

918

ABSTRACT

he efficacy of azacitidine in patients with anemia and with lowerrisk myelodysplastic syndromes, if relapsing after or resistant to erythropoietic stimulating agents, and the benefit of combining these agents to azacitidine in this setting are not well known. We prospectively compared the outcomes of patients, all of them having the characteristics of this subset of lower-risk myelodysplastic syndrome, if randomly treated with azacitidine alone or azacitidine combined with epoetin-β. High-resolution cytogenetics and gene mutation analysis were performed at entry. The primary study endpoint was the achievement of red blood cell transfusion independence after six cycles. Ninety-eight patients were randomised (49 in each arm). Median age was 72 years. In an intention to treat analysis, transfusion independence was obtained after 6 cycles in 16.3% versus 14.3% of patients in the azacitidine and azacitidine plus epoetin-β arms, respectively (P=1.00). Overall erythroid response rate (minor and major responses according to IWG 2000 criteria) was 34.7% vs. 24.5% in the azacitidine and azacitidine plus epoetin-β arms, respectively (P=0.38). Mutations of the SF3B1 gene were the only ones associated with a significant erythroid response, 29/59 (49%) versus 6/27 (22%) in SF3B1 mutated and unmutated patients, respectively, P=0.02. Detection of at least one “epigenetic mutation" and of an abnormal single nucleotide polymorphism array profile were the only factors associated with significantly poorer overall survival by multivariate analysis. The transfusion independence rate observed with azacitidine in this lower-risk population, but resistant to erythropoietic stimulating agents, was lower than expected, with no observed benefit of added epoetin, (clinicaltrials.gov identifier: 01015352). haematologica | 2016; 101(8)

Azacitidine in lower risk MDS

Introduction Anemia is the most common cytopenia in lower-risk [i.e. low- or intermediate-1 risk by the International Prognostic Scoring System (IPSS)] myelodysplastic syndrome (MDS). First-line treatment of anemia in lower risk MDS (LR-MDS), with the exception of cases with chromosomal deletion 5q (del 5q) that respond better to lenalidomide, generally consists of erythropoietic stimulating agents (ESA), which yield an erythroid response in about 40% to 50% of patients, with a median response duration of about 24 months.1-3 Second-line treatments to avoid anemia recurrence and regular RBC transfusions include: (i) hypomethylating agents (HMA),4-6 approved in this context in several countries, (ii) lenalidomide, yielding about 25% of erythroid responses,7 but not approved in the absence of del 5q, and (iii) investigational agents. Among HMA, azacitidine (AZA) at a standard daily dose of 75mg/m2 for 7 days, every 28 days, significantly reduces transfusion dependence, decreases the risk of transformation to acute myeloid leukemia (AML), and improves quality of life (QOL) in higher-risk MDS (i.e. intermediate-2 or high IPSS risk MDS).5,6 In lower-risk MDS, AZA using 5 or 7 day regimens has been reported to yield RBC transfusion independence (RBC-TI) in 30– 40% of LR-MDS,8 although lower response rates to AZA were recently reported in Europe in compassionate use programs9,10 and in a prospective trial.11 In a recent report of preliminary results of a 3-day only administration of either AZA or decitabine in 83 evaluable LR-MDS patients, overall response was 61%, but RBC-TI was only 24% in the 38 transfusion dependent patients.12 In those studies, the resistance of anemia to ESA was not always documented, and the use of AZA has not been prospectively tested in LR-MDS patients selected for their resistance to ESA, an important subset of LR-MDS in daily practice. In addition, the impact of recently described acquired genetic abnormalities in response to AZA in LRMDS with anemia has not been studied prospectively. Finally, whether the addition of ESA to AZA is beneficial to those patients, as it is in our experience with lenalidomide,13 has only recently been studied in a small cohort.11 In a previous retrospective work, our group had indeed suggested a possible benefit (higher transfusion independence rate and better overall survival) of adding ESA to AZA in higher-risk MDS.14 To address those issues, the GFM designed a phase II randomized clinical trial which compared AZA (using a 5day monthly schedule) and AZA plus an ESA, in RBC transfusion dependent LR-MDS patients resistant to highdose ESA.

Methods Patient Eligibility Inclusion criteria were: (i) MDS according to WHO criteria or chronic myelomonocytic leukemia (CMML) with WBC < 13 G/L, (ii) age >= 18 years, (iii) low- or intermediate-1 risk MDS according to IPSS, (iv) resistance to or relapse after at least 12 weeks of high dose ESA (>= 60 000 U/w for epoetin-α or epoetin-β and >= 300 μg/w for darbepoetin), (v) transfusion dependence of at least 4 RBC units/8 weeks, calculated over the previous 16 weeks, and (vi) ECOG-PS score ≤ 2. haematologica | 2016; 101(8)

Exclusion criteria are listed in the Online Supplementary Information. This trial was registered in both the EudraCT (2008004541-29) and the Clinical trial databases (GFMAzaEpo-2008-1 trial, clinicaltrials.gov identifier: 01015352) and was approved by an ethics committee (CPP Ile de France, Aulnay sous Bois) and L'Agence Nationale de Sécurité des Médicaments (ANSM), according to French regulations and in accordance with the Declaration of Helsinki. Each patient provided written informed consent.

Treatment regimen Patients were randomly assigned to receive either AZA alone (75 mg/m2/d) injected subcutaneously (sc) for 5 days every 28 days, or AZA plus epoetin-β 60 000 U/w sc. Response was assessed after the 4th and 6th cycles of treatment, according to IWG 2000 and IWG 2006 erythroid response (HI-E) criteria. Responders could receive additional cycles, using the same treatment schedule of AZA+/-epoetin-β for a maximum of 18 cycles or until relapse.

Conventional cytogenetic and SNP- A karyotyping Cytogenetic R-banding analysis was performed on diagnostic bone marrow samples using standard methods. Patient genomic DNA extracted from bone marrow or blood mononuclear cells was processed and hybridized to Genome-Wide Human SNP 6.0 arrays (Affymetrix, Santa Clara, CA, USA) according to the manufacturers' instructions15 (See Online Supplementary Information for further details).

Mutational analysis At study entry, known or putative mutational gene targets in MDS were examined for mutations using massively parallel sequencing. See Online Supplementary Information and Online Supplementary Table S1 and S3 for further details.

Response criteria Erythroid response was evaluated after 4 and 6 cycles of AZA, according to IWG 200016 and IWG 2006 criteria.17 RBC-TI was defined as no need for red blood cell transfusions (performed at a Hb level of less than 9 g/dl), with a stable hemoglobin level >= 9 g/dl lasting for at least 8 weeks. Safety was evaluated by monitoring and recording of adverse events.

Endpoints The primary endpoint was the achievement of RBC-TI after 6 cycles (major erythroid response HI-E according to IWG 2000). Secondary endpoints were minor and major response according to IWG 2000 criteria and response according to IWG 2006 criteria after 4 and 6 cycles, response duration, overall survival, IPSS progression-free survival, and toxicity.

Sample size justification and statistical analysis Sample size computation was based on the primary endpoint, assuming a response rate of 40% and 70% in the AZA and AZA plus EPO arms, respectively, based on previously published findings with AZA alone in lower-risk MDS. With type I and type II error rates fixed at 0.05 and 0.20, respectively, a minimum of 49 patients had to be enrolled in each randomized arm, based on a two-sided c2 test with Yates continuity correction. (See Online Supplementary Information for statistical analysis).

Results Baseline Patient characteristics (Table 1) Ninety-eight patients were enrolled between February 2009 and November 2010 in 22 centers (listed in the Online 919

S. Thépot et al.

Supplementary Information), including 68 males and 30 females. Forty-nine were randomized in the AZA arm and 49 in the AZA+EPO arm. Median age was 72 years [Interquartile range (IQR) : 65-78]. Diagnosis according to the WHO 2008 classification was RA in 6 patients (6%), RARS in 41 (42%), RCMD in 14 (14%), RCMD-RS in 17 (17%), RAEB-1 in 12 (12%), CMML in 7 (7 %) and MDSU in 1 (1%). The median interval from MDS diagnosis to inclusion was 37.3 months [IQR : 22.8-59]. IPSS was low in 38 and int-1 in 59 patients (not available in 1 patient with previously normal cytogenetics, due to failed cytogenetics at study entry). Cytogenetics, according to IPSS, were favorable in 77 patients, intermediate in 18, unfavorable in 2 patients, and a failure at study entry in 1. According to the revised IPSS, 1 patient had very low-risk MDS, 77 patients had low-risk and 14 patients had intermediate-risk MDS (6 patients were missing data). The median number of RBC units received in the 8 weeks preceding inclusion were 6 [6-8] and 6 [4-8] in the AZA and AZA+EPO arms, respectively, and the median serum ferritin levels were 1432 [1017-1928] and 1512 [1003-2130] μg/L, in the AZA and AZA+EPO arms, respectively, as expected in such a transfusion dependent lower-risk population, with a median time from MDS diagnosis to study entry of 37 months. As shown in Table 1, no imbalance for baseline patient characteristics was observed between the 2 arms. Apart from ESA, no patient had received any disease-related treatment other than RBC transfusions.

Conventional cytogenetics and SNP array karyotyping A SNP array karyotype was available in 79 of the 98 enrolled patients (Table 1). Overall, 33 (43%) of these 79 patients had at least one genomic abnormality detected by SNP-A, including 14 patients with favorable karyotype, 16 with intermediate karyotype, 2 with unfavorable karyotype and 1 failure. SNP array karyotype detected 79 CNA (49 losses/30 gains) and 9 UPD. Details of SNP-A lesions are provided in the Online Supplementary Table S2.

Mutation analysis Sequencing was performed in 90/98 patients, of whom 75 (83%) had one or more mutations (Figure 1). Among 17 genes with detected mutations, only 6 were found mutated in more than 3 patients, namely SF3B1, TET2, DNMT3A, ASXL1, JAK2 and U2AF1 with mutations detected in 59/86, 29/87, 12/86, 5/89, 3/87 and 4/90 patients, respectively. The median number of gene mutations was 1 (range 0-3). In a single patient in this cohort, a TP53 mutation was detected and associated to complex cytogenetics.

Treatment received The median number of cycles administered was 6 [6-10] in the AZA arm, and 6.5 [5-9] in the AZA+EPO arm. Fortysix patients (93.9%) in the AZA arm and 41 patients (83.7%) in the AZA+EPO arm received at least 4 cycles. Seven and 17 patients did not receive the planned 6 cycles in the AZA and AZA+EPO arm, respectively. Four patients did not receive any treatment due to: sudden death (n=1), screening failure (n=1), the patient’s decision (n=1), or the diagnosis of a solid tumor just after screening (n=1). The reasons for treatment interruption before 6 cycles were: the patient’s decision (n=6), the investigator’s decision (n=1), an absence of response after 4 cycles (n=2), AML 920

Table 1. Baseline patient characteristics.

AZA arm, n=49 Median age, [IQR] (years) 71.6 [67.2-78.1] Sex M 34 F 15 Median ferritin level, [IQR] (mg/L) 1432 Missing data [1017-1928] 1 Median RBC Units needs*, [IQR] 6 [6-8] Missing data 4 WHO classification RA 2 RARS 24 RCMD 5 RCMD- RS 9 RAEB1 5 CMML 3 Unclassified 1 IPSS Low 19 Int-1 29 Missing data 1 R-IPSS Very low 1 Low 41 Intermediate 5 Missing data 2 IPSS Cytogenetics Favorable 35 Intermediate 13 Unfavorable 0 Failure 1 SNP array Normal 21 Abnormal 21 Missing data 7 Gene mutations SF3B1 mutated/N 33/44 TET2 mutated/N 15/43 DNMT3A mutated/N 7/47 ASXL1 mutated/N 3/45 JAK2 mutated/N 0/43 U2AF1 mutated/N 2/47

AZA+EPO arm, n=49 73.3 [62.2-77.6] 34 15 1512 [1003-2130] 4 6 [4-8] 4 4 17 9 8 7 4 0 19 30 0 0 36 9 4 42 5 2 0 25 12 12 26/42 14/44 5/43 2/44 3/43 2/43

*in the 8 weeks preceding entry.

transformation (n=2), persistent cytopenia (n=1), documented infection (n=2), febrile neutropenia (n=1), other azacitidine side effects (n=3), and the discovery of pancreatic cancer (n=1). One allogeneic stem cell transplantation was performed in an early responder (Figure 2). With a median follow-up of 47.3 months, 9 (18.4%) and 7 (14.3%) patients had received at least 18 cycles in the AZA and AZA+EPO arms, respectively.

Primary Endpoint (Table 2) In an intention to treat analysis, RBC-TI after 6 cycles (major HI-E according to IWG 2000) was achieved in 8 patients (16.3%, 95%CI: [7.3-29.7]) in the AZA arm and in 7 patients (14.3%, 95%CI: [5.9-27.2]) in the AZA+EPO arm (P=1.0). No predefined factors (including treatment arm, WHO subtype, cytogenetics, IPSS, presence of at least one SNP array abnormality or of one of the 6 most frequent gene mutations) were significantly associated with RBC-TI. haematologica | 2016; 101(8)

Azacitidine in lower risk MDS

Figure 1. Distribution of mutations according to WHO diagnosis.

Secondary Endpoints (Table 2) RBC-TI rate after 4 cycles was achieved in 5 patients (10.2%, 95%CI: [3.4-22.2]) in the AZA arm and in 7 patients (14.3%, 95%CI: [5.9-27.2]) in the AZA+EPO arm (P=0.76). Overall response rate (minor and major response) according to IWG 2000 criteria after 4 cycles was achieved in 18 patients (36.7%, 95%CI: [23.4-51.7]) in the AZA arm and in 15 patients (30.6%, 95%CI: [18.345.4]) in the AZA+EPO arm (P=0.67). Overall response rate (minor and major response) according to IWG 2000 after 6 cycles was achieved in 17 patients (34.7%, 95%CI: [21.7-49.6]) in the AZA arm and in 12 patients (24.5%, 95%CI: [13.3-38.9]) in the AZA+EPO arm (P=0.38). According to IWG 2006 criteria, HI-E was achieved after 4 cycles in 14 patients (28.6%, 95%CI: [16.6-43.3]) in the AZA arm and in 14 patients (28.6%, 95%CI: [16.6-43.3]) in the AZA+EPO arm (P=1.0). After 6 cycles, HI-E was achieved in 15 patients (30.6%, 95%CI: [18.3-45.4]) in the AZA arm and in 13 patients (26.5%, 95%CI: [14.9-41.1]) in the AZA+EPO arm (P=0.82). An overall response according to IWG 2006 criteria was achieved in 21 patients (42.9% with 95%CI [28.8-57.8]) in the AZA arm versus 17 patients (34.7% with 95%CI: [21.7-49.6]) in the AZA+EPO arm (P=0.53). After 4 cycles, in the whole cohort, HI-P according to IWG 2006 criteria was achieved in 6/20 (30%) patients with thrombocytopenia. HI-N was achieved in 9/15 (60%) patients with neutropenia. SF3B1 mutation was associated with the erythroid response according to IWG 2006 criteria, with 29/59 (49%) responses observed in SF3B1 mutated versus 6/27 (22%) in SF3B1 unmutated patients (P=0.02). The median number of RBC units received in the 8 weeks preceding study entry was significantly higher in responders (7 [69.8]) than in non-responders (6 [4-8]) (P=0.017), while no other prognostic factors of overall response were observed (Table 3). In the multivariate analysis, RBC transfusion burden, in the 8 weeks preceding inclusion (P=0.041) and SF3B1 (P=0.022), were associated with the erythroid response. In the 38 responders according to IWG 2006 criteria, median response duration was 7.6 months (95%CI: [4.416.8]) in the AZA arm and 9.7 months (95%CI: [5.0-21.2]) haematologica | 2016; 101(8)

in the AZA plus EPO arm (P=0.53). (Figure 3). The median duration of disease, using our primary endpoint, i.e. RBCTI (or major IWG 2000 criteria), was 10.5 months (95%CI: [5.9-NA]) in the AZA arm and 16.6 months (95%CI: [13.8NA]) in the AZA plus EPO arm (P=0.15).

Survival With a median follow-up of 47.3 months [IQR: 24.4-65], the 3-year overall survival was 72.1% (95%CI: [60.3-86.2]) in the AZA arm vs. 66.8 % (95%CI: [54.3-82.]) in the AZA+ EPO arm, respectively (P=0.93) (Figure 4). In univariate analyses, factors significantly associated with poorer overall survival were: any SNP array abnormality (P=0.013), ASXL1 mutation (P=0.01), and the presence of at least one “epigenetic mutation”, (defined as any mutation observed in the TET2, DNMT3A, ASXL1, IDH2, KDM6A, and EZH2 genes) (P=0.022). In the 8 weeks preceding inclusion, WHO diagnosis, IPSS, IPSS cytogenetics, RBC transfusion burden and serum ferritin levels had no significant impact on overall survival. Early ESA failure, defined by primary ESA resistance or relapse within 6 months of response,18 was not associated with survival in this cohort (P=0.63). In the multivariate analysis, only the detection of a SNP array abnormality (P=0.01) or of an “epigenetic mutation” (P=0.02) were associated with a significantly poorer overall survival.

IPSS progression-free survival During their time in the study, disease evolution according to IPSS was documented in 18 patients, 6 and 12 of them in the AZA and AZA+EPO arms, respectively. Three-year IPSS progression-free survival was 91.1% (95%CI: [83.6-99.8]) and 72.4% (95%CI: [65.7-90.8]) in the AZA and AZA+EPO arms, respectively (P=0.12).

Discussion In this phase II trial of azacitidine in lower risk-MDS patients, selected for their resistance to an ESA, the overall 921

S. Thépot et al. Table 2. Erythroid response evaluated at different timepoints by treatment arms.

AZA arm N=49

AZA+EPO arm N=49

P (Fisher's exact test)

RBC-TI IWG 2000 major response criteria After 4 cycles • No response • Response After 6 cycles • No response • Response

44 (89.8%) 5 (10.2%)

42 (85.7%) 7 (14.3%)

0.76

41 (83.7%) 8 (16.3%)

42 (85.7%) 7 (14.3%)

1.0

Overall response IWG 2000 major+minor response criteria After 4 cycles • No response • Response After 6 cycles • No response • Response

31 (63.3%) 18 (36.7%)

34 (69.4%) 15 (30.6%)

0.67

32 (65.3%) 17 (34.7%)

37 (75.5%) 12 (24.5%)

0.38

35 (71.4%) 14 (28.6%)

1.0

34 (69.4%) 15 (30.6%)

36 (73.5%) 13 (26.5%)

0.82

28 (57.1%) 21 (42.9%)

32 (65.3%) 17 (34.7%)

0.53

HI-E IWG 2006 criteria After 4 cycles • No response • Response After 6 cycles • No response • Response Overall HI-E • No response • Response

response rate after 6 cycles of azacitidine, according to IWG 2000 criteria, was estimated at 34.7% in the AZA arm versus 24.5% in the AZA+EPO arm. However, the RBC transfusion independence rate was only 16.3% in the AZA arm and 14.3% in the AZA+ EPO arm, i.e. lower than the 45% and 44% RBC-TI rates previously reported by other groups in unselected lower-risk MDS.6,8,12 Explanations for those differences possibly include the fact that our patients had been selected for their resistance to ESA (which was not a prerequisite in most prior studies), and had a minimal RBC transfusion dependency of 4 units in the 8 weeks prior to the study (the median number of RBC units in the 8 weeks prior to study entry was 6 [414]). By comparison, in the Lyons et al. trial8 only 47% of patients were RBC transfusion dependent, and a lower RBC transfusion requirement (less than 2 units/8 weeks) was predictive of RBC-TI achievement with AZA (Online Supplementary Table S4). In other studies,9-11 which included a higher proportion of transfusion dependent patients, a lower erythroid response rate, ranging from 30% to 40% was also observed. Another difference with several other series was that most of our patients had anemia as cytopenia. Only 22% of the patients also had thrombocytopenia, compared to 56% in the trial by Lyons et al. Therefore, hematologic improvement in other lineages, frequently taken into account in the evaluation of the overall response to azacitidine in other series, was by definition lower in our patients.6,8,12 The addition of ESA in our randomised trial did not significantly improve response rate, contrary to what we 922

Figure 2. Consort flow diagram.

observed using lenalidomide in lower-risk MDS resistant anemia.13 Our group had published that the addition of haematologica | 2016; 101(8)

Azacitidine in lower risk MDS

EPO to AZA in higher-risk MDS patients improved response rate, but this difference was not associated with better OS.14 The detection of a SF3B1 mutation and the median number of RBC transfusions were significant prognostic factors of the response according to IWG 2006 criteria in our study. In the trial by Lyons et al.,8 the absence of neutropenia and thrombocytopenia, and a baseline transfusion requirement of <=2 RBC units every 8 weeks were predictive of higher RBC TI. In higher-risk

MDS, our group had published that RBC transfusion dependence was a prognostic factor for poorer OS, but not for response to AZA.19 With prolonged follow-up in all our patients, the median duration of response was relatively short (7.6 months and 9.7 months in the AZA arm and in the AZA plus EPO arm, respectively), but 13.6% and 18.8% of responses were longer than 2 years in the 2 arms, respectively. This result is similar to previous series, where median response

Table 3. Predictive factors of overall HI-E response according to the IWG 2006 criteria.

AZA arm AZA+EPO arm Median age, [IQR] (years) Sex M/F Median serum ferritin level, [IQR] (mg/L) Median RBC Units needs*, [IQR] IPSS (%) Low Int-1 Missing data R-IPSS (%) Very low Low Intermediate Missing data Cytogenetics Favorable Intermediate Unfavorable Failure SNP array Normal Abnormal Missing data Gene mutations SF3B1 Absent Present Missing data TET2 Absent Present Missing data DNMT3A Absent Present Missing data ASXL1 Absent Present Missing data JAK2 Absent Present, Missing data U2AF1 Absent Present Missing data

Response n, (%) n=38

No response n,(%) n=60

P (Fisher's exact test or Wilcoxon test )

21 (55) 17 (45) 73.74 [67.17-78.32] 29/9 1250 [942-1955] 7 [6-9.8]

28 (47) 32 (53) 71.3 [64.77-77.43] 39/21 1530 [1065-2148] 6[4-8]

0.53

19 (51) 18 (49) 1

19 (32) 41 (68)

0.086

0 (0) 33 (89) 4 (11) 1

1 (2) 44 (80) 10 (18) 5

0.45

33 (87) 3 (8) 1 (3) 1 (3)

44 (73) 15 (25) 1 (2) 0 (0)

0.051

20 (62) 12 (38) 6

26 (55) 21 (45) 13

0.64

6 (17) 29 (83) 3

21 (41) 30 (59) 9

0.020

21 (60) 14 (40) 3

37 (71) 15 (29) 8

0.35

33 (87) 5 (13) 0

45 (87) 7 (13) 8

1.00

35 (97) 1 (3) 2

49 (92) 4 (8) 7

0.64

35 (100) 0 (0) 3

49 (94) 3 (6) 8

0.27

33 (100) 0 (0) 5

48 (92) 4 (8) 8

0.13

0.40 0.27 0.28 0.017

* In the 8 weeks preceding entry.

haematologica | 2016; 101(8)

923

S. ThĂŠpot et al. durations ranging from 511 to 10 months20 were reported (Online Supplementary Table S4). Overall survival at 3 years in the present trial was 72.1% and 66.8% in the AZA arm and in the AZA plus EPO arm, respectively. It was similar, but with longer follow-up, than previously reported in a phase II trial with decitabine21 and with azacitidine in the Nordic trial,11 where median OS survival was not reached after 14.6 and 30 months, respectively. Recent USA retrospective data reported a median OS of 16 months after HMA failure in lower-risk MDS.22 Our study was also the first to prospectively study the impact of SNP array and mutational analysis in lower-risk

MDS treated with a HMA. The frequency of mutations in this lower-risk MDS patient cohort was different from that previously published by Bejar et al.,23 with a higher frequency of SF3B1 mutations (68.6% compared to 22% for Bejar et al.), explained by the high proportion of sideroblastic anemias included in our series. TET2 and ASXL1 mutations were present in 33.3% and 5.6% of our patients, compared to 23% and 15%, respectively, in Bejar et al.â&#x20AC;&#x2122;s cohort. The mutation of ASXL1, previously associated with an adverse outcome, was found in 5.6% of our patients. The mutation of TP53 was detected in a single patient in the present study cohort of de novo MDS, in con-

AZA arm AZA+EPO arm

Time (Months)

Figure 3. Duration of erythroid response.

AZA arm AZA+EPO arm

Time post random (Months) 924

Figure 4. Overall survival from randomization.

haematologica | 2016; 101(8)

Azacitidine in lower risk MDS

trast to other studies12,23 in which patients with therapyrelated MDS were also analyzed. In univariate analysis, the presence of a SNPa abnormality, of an ASXL1 mutation and of any “epigenetic mutation" were significantly associated with poorer survival, whereas only a trend was observed for longer interval from diagnosis of MDS. In multivariate analysis, only the detection of a SNP array abnormality and of at least one “epigenetic mutation" were associated with overall survival. In the lower-risk MDS series of Bejar et al., analyzed irrespective of treatment and adjusted on a lower-risk prognostic system (LR-PSS), only the presence of EZH2 mutations was predictive of a shorter OS. Our study also confirmed that SNP array analysis can be of interest in lower-risk MDS, as genomic abnormalities were detected by this technique in 14 patients with normal karyotype. SNP analysis, rarely performed in large series of MDS, when used in a previous series, had allowed for the detection of cytogenetic abnormalities in 74 % of the patients versus 44% by conventional banding

References 1. Mannone L, Gardin C, Quarre MC, et al. High-dose darbepoetin alpha in the treatment of anaemia of lower risk myelodysplastic syndrome results of a phase II study. Br J Haematol. 2006;133(5):513-519. 2. Jadersten M, Malcovati L, Dybedal, et al. Long-term outcome of treatment of anemia in MDS with erythropoietin and G-CSF. Blood. 2005;106(3):803-811. 3. Park S, Grabar S, Kelaidi C, et al. Predictive factors of response and survival in myelodysplastic syndrome treated with erythropoietin and G-CSF: the GFM experience. Blood. 2008;111(2):574-582. 4. Silverman LR, Holland JF, Weinberg RS, et al. Effects of treatment with 5-azacytidine on the in vivo and in vitro hematopoiesis in patients with myelodysplastic syndromes. Leukemia. 1993;7(1):21-29. 5. Silverman LR, Demakos EP, Peterson BL, et al. Randomized controlled trial of azacitidine in patients with the myelodysplastic syndrome: a study of the cancer and leukemia group B. J Clin Oncol. 2002;20(10):2429-2440. 6. Silverman LR, McKenzie DR, Peterson BL, et al. Further analysis of trials with azacitidine in patients with myelodysplastic syndrome: studies 8421, 8921, and 9221 by the Cancer and Leukemia Group B. J Clin Oncol. 2006;24(24):3895-3903. 7. Raza A, Reeves JA, Feldman EJ, et al. Phase 2 study of lenalidomide in transfusiondependent, low-risk, and intermediate-1 risk myelodysplastic syndromes with karyotypes other than deletion 5q. Blood. 2008;111(1):86-93. 8. Lyons RM, Cosgriff TM, Modi SS, et al. Hematologic response to three alternative dosing schedules of azacitidine in patients with myelodysplastic syndromes. J Clin Oncol. 2009;27(11):1850-1856. 9. Falantes J, Delgado RG, Calderon-Cabrera C, et al. Multivariable time-dependent analysis of the impact of azacitidine in

haematologica | 2016; 101(8)

10.

11.

12.

13.

14.

15.

16.

17.

studies, and also had prognostic significance.24,25 None of those studies, however, had focused on homogeneously treated lower-risk MDS. In the present study, patients with at least one SNPa abnormality had a trend for poorer survival, which reached significance in multivariate analysis, along with the presence of at least one “epigenetic mutation”. In conclusion, a lower than expected overall response rate to azacitidine was observed in this cohort of lowerrisk MDS patients, selected for their resistance to ESA. No benefit of the addition of an ESA could be demonstrated in this population. As responders were significantly more likely to be mutated for the SF3B1 gene, the use of azacitidine remains an available therapeutic option in these patients, often resistant or refractory to ESA alone, until new treatments clearly emerge for this population, such as sotatercept and luspatercept, currently under development. A direct comparison with lenalidomide7,13 would be needed to more clearly assess each drug’s benefit in ESA resistant lower-risk MDS.

patients with lower-risk myelodysplastic syndrome and unfavorable specific lowerrisk score. Leuk Res. 2015;39(1):52-57. Musto P, Maurillo L, Spagnoli, et al. Azacitidine for the treatment of lower risk myelodysplastic syndromes : a retrospective study of 74 patients enrolled in an Italian named patient program. Cancer. 2010;116(6):1485-1494. Tobiasson M, Dybedahl I, Holm MS, et al. Limited clinical efficacy of azacitidine in transfusion-dependent, growth factor-resistant, low- and Int-1-risk MDS: Results from the nordic NMDSG08A phase II trial. Blood Cancer J. 2014;7(4):2014-2018. Short NJ, Garcia-Manero G, Montalban Bravo G, et al. Low-Dose Hypomethylating Agents (HMAs) Are Effective in Patients (Pts) with Lowor Intermediate-1-Risk Myelodysplastic Syndrome (MDS): A Report on Behalf of the MDS Clinical Research Consortium. Blood. 2015; 126(23):94a-94a. Toma A, Kosmider O, Chevret S, et al. Lenalidomide with or without erythropoietin in transfusion dependent erythropoiesis-stimulating agent-refractory lower risk MDS without 5q deletion. Leukemia. 2016;30(4):897-905. Itzykson R, Thépot S, Beyne-Rauzy O, et al. Does addition of erythropoiesis stimulating agents improve the outcome of higher-risk myelodysplastic syndromes treated with azacitidine? Leuk Res. 2012;36(4):397-400. Renneville A, Abdelali RB, Chevret S, et al. Clinical impact of gene mutations and lesions detected by SNP-array karyotyping in acute myeloid leukemia patients in the context of gemtuzumab ozogamicin treatment: results of the ALFA-0701 trial. Oncotarget. 2014;5(4):916-932. Cheson BD, Bennett JM, Kantarjian H, et al. Report of an international working group to standardize response criteria for myelodysplastic syndromes. Blood. 2000;96(12):36713674. Cheson BD, Greenberg PL, Bennett JM, et al. Clinical application and proposal for modifi-

18.

19.

20.

21.

22.

23.

24.

25.

cation of the International Working Group (IWG) response criteria in myelodysplasia. Blood. 2006;108(2):419-425. Kelaidi C, Park S, Sapena R, et al. Long-term outcome of anemic lower-risk myelodysplastic syndromes without 5q deletion refractory to or relapsing after erythropoiesis-stimulating agents. Leukemia. 2013;27(6):1283-1290. Itzykson R, Thépot S, Quesnel B, et al. Prognostic factors for response and overall survival in 282 patients with higher-risk myelodysplastic syndromes treated with azacitidine. Blood. 2011;117(2):403-411. Fili C, Malagola M, Follo MY, et al. Prospective phase II Study on 5-days azacitidine for treatment of symptomatic and/or erythropoietin unresponsive patients with low/INT-1-risk myelodysplastic syndromes. Clin Cancer Res. 2013; 19(12):3297-3308. Garcia-manero G, Jabbour E, Borthakur G, et al. Randomized Open-Label Phase II Study of Decitabine in Patients With Lowor Intermediate-Risk Myelodysplastic Syndromes. J Clin Oncol. 2013; 31(20):25482553. Jabbour EJ, Garcia-Manero G, Strati P, et al. Outcome of Patients With Low-Risk and Intermediate-1-Risk Myelodysplastic Syndrome After Hypomethylating Agent Failure. Cancer. 2015;121(6):876-882. Bejar R, Stevenson KE, Caughey BA, et al. Validation of a prognostic model and the impact of mutations in patients with lowerrisk myelodysplastic syndromes. J Clin Oncol. 2012;30(27):3376-3382. Tiu RV, Gondek LP, O'Keefe CL, et al. Prognostic impact of SNP array karyotyping in myelodysplastic syndromes and related myeloid malignancies. Blood. 2011; 117(17):4552-4560 Mohamedali A, Gäken J, Twine NA, et al. Prevalence and prognostic significance of allelic imbalance by single-nucleotide polymorphism analysis in low-risk myelodysplastic syndromes. Blood. 2007; 110(9):33653373.

925

ARTICLE EUROPEAN HEMATOLOGY ASSOCIATION

Myeloproliferative Disorders

Ferrata Storti Foundation

Haematologica 2016 Volume 101(8):926-931

Antiplatelet therapy versus observation in low-risk essential thrombocythemia with a CALR mutation

Alberto Alvarez-Larrán,1 Arturo Pereira,2 Paola Guglielmelli,3 Juan Carlos Hernández-Boluda,4 Eduardo Arellano-Rodrigo,2 Francisca Ferrer-Marín,5 Alimam Samah,6 Martin Griesshammer,7 Ana Kerguelen,8 Bjorn Andreasson,9 Carmen Burgaleta,10 Jiri Schwarz,11 Valentín García-Gutiérrez,12 Rosa Ayala,13 Pere Barba,14 María Teresa Gómez-Casares,15 Chiara Paoli,3 Beatrice Drexler,16 Sonja Zweegman,17 Mary F. McMullin,18 Jan Samuelsson,19 Claire Harrison,6 Francisco Cervantes,20 Alessandro M. Vannucchi,3 and Carlos Besses1

Hematology Department, Hospital del Mar, IMIM, UAB, Barcelona, Spain; Hematotherapy and Hemostasis Department, Hospital Clínic, Barcelona, Spain; 3Center for Research and Innovation of MPN (CRIMM); AOU Careggi, and Department of Experimental and Clinical Medicine, University of Florence, Italy; 4Hematology Department, Hospital Clínico, Valencia, Spain; 5Hematology and Medical Oncology Department, Hospital Morales Messeguer, IMIB-Arrixaca, UCAM, Murcia, Spain; 6 Haematology Department, Guys' and St Thomas' NHS Foundation Trust, London, UK; 7 Hematology, Oncology & Palliative Care, Johannes Wesling Academic Medical Center, University of Hannover Teaching Hospital, Germany; 8Hematology Department, Hospital La Paz, Madrid, Spain; 9Hematology Section, Uddevalla Hospital, NU Hospital Group, Sweden; 10Hematology Department, Hospital Príncipe de Asturias, Alcalá de Henares, Spain; 11Institute of Hematology & Blood Transfusion, Prague, Czech Republic; 12 Hematology Department, Hospital Ramón y Cajal, Madrid, Spain; 13Hematology Department, Hospital Universitario 12 de Octubre, Madrid, Spain; 14Hematology Department, Hospital Vall d’Hebron, Barcelona, Spain; 15Hematology Department, Hospital Dr. Negrín, Las Palmas de Gran Canaria, Spain; 16Division of Hematology, University Hospital Basel, Switzerland; 17Department of Hematology, VU University Medical Center, Amsterdam, The Netherlands; 18Centre for Cancer Research and Cell Biology, Queen’s University, Belfast, UK; 19Department of Clinical Science and Education, Karolinska Institute, South Hospital, Stockholm, Sweden; and 20Hematology Department, Hospital Clínic, IDIBAPS, Barcelona, Spain. 1 2

Correspondence: 95967@parcdesalutmar.cat ABSTRACT Received: March 23, 2016 Accepted: May 3, 2016 Pre-published: May 12, 2016. doi:10.3324/haematol.2016.146654

Check the online version for the most updated information on this article, online supplements, and information on authorship & disclosures: www.haematologica.org/content/101/8/926

926

he role of antiplatelet therapy as primary prophylaxis of thrombosis in low-risk essential thrombocythemia has not been studied in randomized clinical trials. We assessed the benefit/risk of lowdose aspirin in 433 patients with low-risk essential thrombocythemia (271 with a CALR mutation, 162 with a JAK2V617F mutation) who were on antiplatelet therapy or observation only. After a follow up of 2215 person-years free from cytoreduction, 25 thrombotic and 17 bleeding episodes were recorded. In CALR-mutated patients, antiplatelet therapy did not affect the risk of thrombosis but was associated with a higher incidence of bleeding (12.9 versus 1.8 episodes per 1000 patient-years, P=0.03). In JAK2V617F-mutated patients, low-dose aspirin was associated with a reduced incidence of venous thrombosis with no effect on the risk of bleeding. Coexistence of JAK2V617F-mutation and cardiovascular risk factors increased the risk of thrombosis, even after adjusting for treatment with low-dose aspirin (incidence rate ratio: 9.8; 95% confidence interval: 2.3-42.3; P=0.02). Time free from cytoreduction was significantly shorter in CALR-mutated patients with essential thrombocythemia than in JAK2V617F-mutated ones (median time 5 years and 9.8 years, respectively; P=0.0002) and cytoreduction was usually necessary to control extreme thrombocytosis. In conclusion, in patients with low-risk, CALR-mutated essential thrombocythemia, low-dose aspirin does not reduce the risk of thrombosis and may increase the risk of bleeding.

haematologica | 2016; 101(8)

Antiplatelet therapy in low-risk ET

Introduction Essential thrombocythemia (ET) is a myeloproliferative neoplasm characterized by an increased risk of thrombosis and hemorrhage. Patients older than 60 years or with a previous history of thrombotic complications are considered at high risk of thrombosis and managed with cytoreduction and antiplatelet agents. In contrast, watchful waiting, with or without antiplatelet therapy, is the recommended approach for younger patients in the absence of a history of thrombosis.1,2 The role of antiplatelet therapy as primary prophylaxis of thrombosis in this group of low-risk ET patients has not been studied in randomized clinical trials. Recommendations for its use are based on the extrapolation to ET of results from the ECLAP study3 conducted in patients with polycythemia vera, and some observational studies in ET.4 Nevertheless, recent studies in low-risk ET have found that antiplatelet therapy is effective in the prevention of thrombosis only in patients with cardiovascular risk factors or those who bear the JAK2V617F mutation, while it might increase the risk of bleeding in those with marked thrombocytosis.5 JAK2V617F and CALR exon 9 mutations are the most frequent, mutually exclusive, genetic alterations in ET, being present in 60% and 20% of such patients, respectively.6,7 Recent studies have suggested that CALR-positive ET can be considered a distinct clinical entity from JAK2V617F-positive ET because of higher platelet counts and a lower incidence of thrombosis in the former.8-12 This observation is especially relevant with regard to the use of antiplatelet therapy in CALR-mutated low-risk ET patients, in whom the increased risk of bleeding associated with higher platelet counts might offset the benefits of reducing the relatively low risk of thrombosis. The aim of the present study was to assess the benefitrisk balance of using antiplatelet therapy in the primary prevention of thrombosis in patients with low-risk ET according to CALR and JAK2V617F mutational status.

Methods Study design The medical records of patients diagnosed with CALR-mutated ET at several European institutions were reviewed. Patients were eligible for inclusion in the study if they were younger than 60 years at diagnosis of ET, had no prior history of thrombosis or major bleeding, and had not received cytoreductive therapy as the initial treatment for ET. An additional number of JAK2V617Fmutated cases with similar clinical characteristics and follow-up were used as a control group for comparisons. Since time free from cytoreduction was longer in JAK2V617F-mutated cases, patients with CALR and JAK2V617F genotypes were included in a 2:1 proportion, with this allowing a similar time at risk in both groups of patients. The decision between starting treatment with low-dose aspirin or keeping the patient on careful observation without antiplatelet therapy was taken by the attending hematologist. The diagnosis of ET was established at the hospital of origin using the updated criteria of the World Health Organization.13 Informed consent for the scientific use of the patients’ clinico-hematologic data and biological samples was obtained as required by the local ethics committees. In all patients the main clinico-hematologic data at presentation of ET were collected, including age, sex, cardiovascular risk haematologica | 2016; 101(8)

factors, hemoglobin values, leukocyte and platelet counts, as well as JAK2V617F and CALR mutational status. Further data included the therapeutic approach (starting and stopping dates, reasons for initiation and withdrawal), occurrence of thrombosis and bleeding, transformation into polycythemia vera, myelofibrosis or acute leukemia, and cause of death.

Outcomes The primary outcome of the study was the occurrence of thrombosis, either arterial or venous. The safety outcome was major hemorrhage. Thrombosis was defined according to the International Classification of Diseases (ninth revision). Arterial thrombosis included stroke, transient ischemic attacks, retinal artery occlusion, coronary arterial disease, and peripheral arterial disease. Venous thrombosis included cerebral venous sinus thrombosis, deep-vein thrombosis, pulmonary thromboembolism, Budd-Chiari syndrome, and spleno-portal vein thrombosis. Minor occlusive events, such as erythromelalgia and superficial thrombophlebitis of the extremities, were not considered. Severe hemorrhage was defined as symptomatic bleeding in a critical organ or an overt hemorrhage requiring transfusion or associated with a decrease in hemoglobin ≥20 g/L without transfusion.

Statistical methods For the purpose of the present study, the time at risk of thrombosis or hemorrhage was computed from the date of diagnosis of ET to the date of death, last follow-up, first thrombotic or bleeding event, start of cytoreductive therapy or initiation of oral anticoagulation, whichever occurred first. The incidence rate of thrombosis or bleeding while the patients were on low-dose aspirin or on observation alone was calculated as the number of events per 1000 patient-years of follow-up. The incidence rate method allows periods without low-dose aspirin to be accounted for in those patients who were started on this therapy after an initial period of careful observation or withdrew from it at some time during follow-up. A multivariate analysis of factors influencing the incidence rate of thrombosis or bleeding was performed using Poisson regression. In Poisson regression, the exponentiated coefficients of covariates can be regarded as incidence rate ratios (IRR) and are comparable to the hazard ratios in Cox models. Variables analyzed for their independent association with the incidence rate of thrombosis or hemorrhage included age, sex, cardiovascular risk factors, hematologic values at diagnosis, presence of the JAK2V617F or CALR mutation and whether the patient was on low-dose aspirin or not. Marked thrombocytosis was defined as a platelet count at diagnosis of >1000x109/L. Leukocytosis was defined as a leukocyte count at diagnosis of >10x109/L. Further models were fitted with interactions between low-dose aspirin and selected clinical or biological features of the patients, such as the presence of cardiovascular risk factors or the JAK2V617F or CALR mutation. Since a physician’s decision to initiate lowdose aspirin was not random but influenced by the patient’s characteristics, a propensity score was calculated from the binary logistic regression of the initial clinical and laboratory features predicting antiplatelet therapy. The propensity score assigns every patient a probability of being in the low-dose aspirin group instead of the observation group, conditional on that patient’s clinico-biological features at the diagnosis of ET, and it was forced into the Poisson models in order to control for confounding. All the statistical analyses were performed with Stata, version 11 (www.stata.com). The time-span splitting method14 was used to calculate the incidence rates and fit the Poisson models. 927

A. Alvarez-Larrรกn et al.

Results Characteristics of the patients and follow-up data A total of 433 low-risk ET patients were included in the study, 271 were CALR-mutated and 162 carried the JAK2V617F mutation. The type of mutation was known for 211 (78%) of the 271 CALR-positive patients (type 1 in 111; type 2 in 80; other types in 20). The main characteristics of the patients according to JAK2V617F or CALR mutational status are shown in Table 1. Cytoreductive therapy was started in 231 patients. The projected time from diagnosis to cytoreduction was significantly shorter in patients with CALR-mutated ET than in JAK2V617F-mutated patients (median 5.0 years and 9.8 years, respectively; log-rank test P=0.002. Figure 1). Reasons for starting cytoreductive therapy are shown in Table 2. As can be seen, extreme thrombocytosis was the commonest reason for starting cytoreductive therapy in CALR-positive patients. First-line cytoreductive therapy included hydroxyurea (143 patients), anagrelide (66 patients), interferon (18 patients), and busulfan (4 patients). Some of these patients received second- and third-line cytoreductive therapy (data not shown). Time at risk of thrombosis and bleeding, free of cytoreductive therapy, was 2215 person-years. Antiplatelet therapy with low-dose aspirin (81-100 mg/day) was started in 353 patients, either at diagnosis of ET or later during followup. Low-dose aspirin was withdrawn in 50 out of these 353 patients, either permanently (46 patients) or temporarily (4 patients). Taking into consideration the periods of time on or off antiplatelet therapy in every patient, time at risk of thrombosis or major bleeding was 1307 and 908 personyears for low-dose aspirin and observation only, respectively. With regard to mutational status, time at risk was 1192 and 1023 person-years for CALR-mutated and JAK2V617Fmutated ET, respectively.

Incidence of thrombosis and major bleeding A total of 25 thrombotic events (arterial or venous) were recorded over the 2215 person-years of follow-up time in which patients remained in the low-risk status free from cytoreductive therapy. Fourteen out of these 25 thrombotic events occurred while patients were receiving low-dose aspirin whereas 11 thromboses appeared while patients were on observation only, resulting in an incidence rate of 10.7 and 12.1 thrombotic events per 1000 person-years, respectively (Figure 2, P=0.7). CALR-positive patients had an incidence rate of thrombosis of 9.7 and 6.9 events per

1000 person-years while on antiplatelet therapy and careful observation, respectively (Figure 2, P=0.6). In the JAK2V617Fmutated patients, the incidence rate of thrombosis was higher while they were on observation only without aspirin but the difference did not attain statistical significance (21.1 versus 11.6 events per 1000 person-years for observation and aspirin, respectively, P=0.3, Figure 2). Fourteen arterial thrombotic events were recorded during the period at risk. The incidence rate of arterial thrombosis was similar for periods on and off antiplatelet therapy, both in the whole cohort and the subgroups of patients with CALR or JAK2V617F mutations (Figure 3). Venous thrombosis was observed in 11 patients, four of them were on antiplatelet therapy and seven on careful observation (incidence rate: 3.1 and 7.7 episodes per 1000 person-years, respectively; P=0.1, Figure 4). In the JAK2V617F-mutated patients, there was a significantly higher rate of venous thrombosis when patients were on careful observation than when they were on antiplatelet therapy (15 versus 2.9 events per 1000 person-years, respectively, P=0.045, Figure 4). In contrast, in the CALR-mutated patients, no significant association was found between low-dose aspirin and the incidence rate of venous thrombosis (Figure 4). A multivariate model including age, sex, presence of cardiovascular risk factors leukocyte count at diagnosis, type of mutation (CALR or JAK2V617F) and whether the patient was on or off antiplatelet therapy did not identify any risk factor for thrombosis (arterial and venous together). On interaction analyses, patients with both JAK2V617F mutation

Table 2. Reason for starting cytoreductive therapy in 433 patients with low-risk essential thrombocythemia.

Age >60 years, n (%) Thrombosis, n (%) Bleeding, n (%) Microvascular symptoms, n (%) Extreme thrombocytosis, n (%) Other, n (%)

JAK2V617F n= 74

CALR n=157

11 (15) 13 (18) 5 (7) 19 (26) 18 (24) 8 (11)

11 (7) 8 (5) 5 (3) 23 (15) 98 (62) 12 (8)

JAK2V617F-mutated ET n=162, CALR-mutated ET n=271

Table 1. Main clinico-hematologic characteristics at diagnosis of 433 patients with low-risk essential thrombocythemia.

JAK2V617F n=162 Age, years* 42 (10-59) Sex M/F, % 35/65 Cardiovascular risk factors, n (%) 79 (49) Microvascular symptoms, n (%) 38 (24) Hemoglobin, g/L* 145 (112-171) Leukocyte count, x109/L 8.6 (3.9-17.1) Platelet count, x109/L* 702 (457-1451) >1000x109/L, n (%) 22 (14) *Median (range). M: male. F: female.

928

CALR n=271

42 (12-59) 44/56 74 (27) 66 (25) 136 (94-166) 8.0 (3.1-16) 891 (480-2409) 93 (35)

0.5 0.055 0.0001 0.6 0.0001 0.0002 0.0001 <0.0001

Figure 1. Time to cytoreductive therapy (95% confidence interval) according to genotype in low-risk ET. P value = 0.0002.

haematologica | 2016; 101(8)

Antiplatelet therapy in low-risk ET

and cardiovascular risk factors had a significantly higher risk of thrombosis (incidence rate: 22 versus 8.9 events per 1000 patient-years in the remaining patients; P=0.04), which persisted after adjustment for treatment with low-dose aspirin (IRR: 2.5, 95% CI: 1.1-5.7, P=0.03). Coexistence of the JAK2V617F mutation and cardiovascular risk factors increased the risk of arterial thrombosis (IRR: 3.2, 95% CI: 1.1-9.4, P=0.03) but had no effect on the incidence of venous thrombosis (IRR: 1.7, 95% CI: 0.5-6.7, P=0.4). Patients with concomitant cardiovascular risk factors and leukocytosis showed a higher rate of arterial thrombosis than the remainder (incidence rate: 26.3 versus 4.5 events per 1000 patient-years, P=0.005). Adjusting by low-dose aspirin treatment showed a higher risk of arterial thrombosis in patients with both cardiovascular risk factors and leukocytosis (IRR: 6.2, 95% CI: 1.9-19.5, P=0.02) whereas treatment with aspirin did not result in a reduction of thrombotic risk (IRR 1.4, 95% CI: 0.4-4.6, P=0.5). When genotype was included in the multivariate model the increased risk associated with concomitant cardiovascular risk factors and leukocytosis was only observed in JAK2V617Fmutated patients (IRR: 5.8, 95% CI: 1.3-26.9, P=0.023) but

not in those with the CALR mutation (IRR: 4.9, 95% CI: 0.6-37.3, P=0.1) Overall, 17 major bleeding episodes were registered, 13 while patients were on antiplatelet therapy and four while on careful observation (incidence rate: 9.9 and 4.6 events per 1000 person-years, respectively; P=0.2, Figure 5). No significant differences in the rate of major bleeding were observed in JAK2V617F-positive patients according to type of therapy (Figure 5), whereas CALR-mutated patients experienced a higher rate of major bleeding while on antiplatelet therapy than on careful observation (12.9 versus 1.8 bleeding events per 1000 person-years, respectively, P=0.03, Figure 5). Interaction analyses were performed to assess whether the risk of hemorrhage associated with the coexistence of antiplatelet therapy and marked thrombocytosis (>1000x109/L) varied with the CALR/JAK2V617F genotype. In CALR-mutated patients, antiplatelet therapy was associated with a tendency to an increased risk of bleeding (IRR 6.9, 95% CI: 0.9-54.1, P=0.06) whereas extreme thrombocytosis was not (IRR: 2.7, 95% CI: 0.7-9.5, P=0.1). In contrast, in JAK2V617F-mutated patients, extreme thrombocytosis was

Figure 2. Incidence rate of thrombosis (arterial or venous) in ET patients treated with low-dose aspirin or careful observation. Rates according to therapy are provided for the whole cohort of patients (P=0.7), JAK2V617F-mutated patients (P=0.2) and CALR-mutated patients (P=0.6).

Figure 3. Incidence rate of arterial thrombosis in ET patients treated with low-dose aspirin or careful observation. Rates according to therapy are provided for the whole cohort of patients (P=0.7), JAK2V617F-mutated patients (P=0.9) and CALR-mutated patients (P=0.7).

haematologica | 2016; 101(8)

929

A. Alvarez-Larrรกn et al.

associated with an increased risk of bleeding (IRR: 9.8, 95% CI: 2.3-42.3, P=0.002) whereas antiplatelet therapy was not (IRR: 0.9, 95% CI: 0.2-3.3, P=0.9). Additional interaction analyses failed to demonstrate a significant association between leukocytosis and major bleeding even when antiplatelet therapy or genotype were included in the multivariate model.

Discussion In this study, we retrospectively assessed the benefit/risk balance of antiplatelet therapy in low-risk ET during the period of time in which patients were free from cytoreduction. To the best of our knowledge, it represents the largest series of low-risk patients and the first one evaluating antiplatelet therapy in CALR-mutated ET. The main finding was the lack of reduction of thrombosis combined with a higher incidence of major bleeding in CALR-mutated patients when receiving low-dose aspirin. These results question the indication for monotherapy with antiplatelet agents in CALR-positive low-risk ET. Recently, a new score including cardiovascular risk factors and JAK2V617F mutational status, in addition to age and

history of thrombosis, has been proposed to assess the risk of thrombosis in ET.15 Application of this new score to patients with low-risk ET implies that both the CALR/JAK2 genotype and cardiovascular risk factors should be taken into account in guiding therapeutic decisions.16 In this regard, a previous study of 300 patients with low-risk ET showed that antiplatelet therapy was effective in reducing the incidence of venous thrombosis in patients with the JAK2V617F mutation and arterial thrombosis in patients with cardiovascular risk factors.5 The present study expands on these previous findings by revealing a synergistic effect between cardiovascular risk factors with both JAK2V617F mutation and leukocytosis, resulting in an increased risk of arterial thrombosis and, more importantly, that antiplatelet therapy is not able to offset such increased risk. Interaction analyses showed a higher risk of bleeding in CALR-mutated patients treated with low-dose aspirin which was independent of the platelet count. In contrast, in JAK2V617F-mutated ET, the risk of bleeding was mostly associated with marked thrombocytosis, and was not influenced by antiplatelet therapy. This observation could be related to differences in platelet function according to the type of mutation. In this regard, recent studies showed

Figure 4. Incidence rate of venous thrombosis in ET patients treated with low-dose aspirin or careful observation. Rates according to therapy are provided for the whole cohort of patients (P=0.1), JAK2V617F-mutated patients (P=0.045) and CALR-mutated patients (P=0.9).

Figure 5. incidence rate of major bleeding in ET patients treated with low-dose aspirin or careful observation. Rates according to therapy are provided for the whole cohort of patients (P=0.17), JAK2V617F-mutated patients (P=0.8) and CALR-mutated patients (P=0.03).

930

haematologica | 2016; 101(8)

Antiplatelet therapy in low-risk ET

more pronounced platelet dysfunction and lesser platelet activation in patients carrying CALR mutations than in those with JAK2V617F-positive ET, a feature that might result in a higher aspirin-induced bleeding diathesis.17,18 Interestingly, patients with CALR-mutated ET were treated with cytoreductive therapy sooner after diagnosis (median 5 years compared to 9.8 years in JAK2V617F-mutated cases), usually to control extreme thrombocytosis. This finding suggests that, in routine clinical practice, treatment needs in low-risk ET patients differ according to whether they have the JAK2V617F or CALR mutation. Thus, while JAK2V617Fmutated patients are usually managed in the long-term with antiplatelet therapy, this is not the case with their CALRmutated counterparts, who appear to have a higher requirement for cytoreduction. Current recommendations from experts adopt the same treatment algorithm for all low-risk ET patients, regardless of the JAK2V617F and CALR genotype.1,2 However, the results from the present study call for a distinctive, genotypebased therapeutic approach. Indeed, in CALR-mutated patients both the failure of antiplatelet agents to prevent thrombosis and the increased need for cytoreductive therapy using current algorithms suggest that such patients could benefit from an individualized therapeutic approach, different from that used in JAK2V617F-mutated patients. Thus, in low-risk CALR-mutated ET, in which the incidence of thrombosis is very low, careful observation would be a reasonable option for asymptomatic patients, while in patients with symptoms or marked thrombocytosis, cytoreductive therapy would be preferable because of its efficacy and low associated risk of bleeding. On the contrary, in patients with JAK2V617F-positive ET, antiplatelet therapy would be superior to abstention, providing an adequate antithrombotic effect without a definite increase in the risk of bleeding. Nevertheless, in patients with JAK2V617F mutation with concomitant cardiovascular risk factors and/or leukocytosis, the antithrom-

References 1. Barbui T, Barosi G, Birgegard G, et al. European LeukemiaNet. Philadelphia-negative classical myeloproliferative neoplasms: critical concepts and management recommendations from European LeukemiaNet. J Clin Oncol. 2011;29(6):761-770. 2. Harrison CN, Bareford D, Butt N, et al. Guideline for investigation and management of adults and children presenting with a thrombocytosis. Br J Haematol. 2010;149(3):352-375. 3. Landolfi R, Marchioli R, Kutti J, et al. Efficacy and safety of low-dose aspirin in polycythemia vera. N Engl J Med. 2004;350(2):114-124. 4. van Genderen PJ, Mulder PG, Waleboer M, van de Moesdijk D, Michiels JJ. Prevention and treatment of thrombotic complications in essential thrombocythaemia: efficacy and safety of aspirin. Br J Haematol. 1997;97(1):179-184. 5. Alvarez-Larrán A, Cervantes F, Pereira A, et al. Observation versus antiplatelet therapy as primary prophylaxis for thrombosis in low-risk essential thrombocythemia. Blood. 2010;116(8):1205-1210. 6. Klampfl T, Gisslinger H, Harutyunyan AS, et al. Somatic mutations of calreticulin in myeloproliferative neoplasms. N Engl J Med. 2013;369(25):2379-2390.

haematologica | 2016; 101(8)

botic effect of antiplatelet therapy may be insufficient. This subset of patients might be candidates for cytoreduction, especially in the presence of marked thrombocytosis. The retrospective design of the present study is a limitation when drawing conclusions to guide clinical practice, because of the possibility of biases in patient selection and therapeutic decisions. Such drawbacks can be avoided with randomized clinical trials. Of particular interest would be trials comparing therapeutic abstention to antiplatelet therapy in CALR-mutated low-risk ET as well as those comparing the effects of aspirin versus cytoreduction in low-risk patients with JAK2V617F mutation and cardiovascular risk factors. However, these studies require the inclusion of a large number of patients and a very long follow-up, which is highly costly and logistically complicated. Meanwhile, observational studies may help in defining a reasonable treatment strategy. In conclusion, in patients with low-risk CALR-mutated ET, antiplatelet therapy with low-dose aspirin does not reduce the frequency of thrombosis and may increase the incidence of bleeding. Acknowledgments We are indebted to all members of the Grupo Español de Enfermedades Mieloproliferativas Filadelfia Negativas (GEMFIN) and to members of the Scientific Working Group of Myeloproliferative Neoplasms of the European Hematology Association for participating in the study. Funding This work was supported by grants from the Instituto de Salud Carlos III, Spanish Health Ministry, PI13/00557, PI1300393, and RD012/0036/0004. The Florence team was supported by AIRC project number 1005 "Special program Molecular Clinical Oncology 5x1000 to Associazione Italiana per la Ricerca sul Cancro Gruppo Italiano Malattie Mieloproliferative (AGIMM). PG was supported by AIRC IG-2014-15967."

7. Nangalia J, Massie CE, Baxter EJ, et al. Somatic CALR mutations in myeloproliferative neoplasms with nonmutated JAK2. N Engl J Med. 2013;369(25):2391-2405. 8. Rumi E, Pietra D, Ferretti V, et al. JAK2 or CALR mutation status defines subtypes of essential thrombocythemia with substantially different clinical course and outcomes. Blood. 2014;123(10):1544-1551. 9. Rotunno G, Mannarelli C, Guglielmelli P, et al. Impact of calreticulin mutations on clinical and hematological phenotype and outcome in essential thrombocythemia. Blood. 2014;123(10):1552-1555. 10. Gangat N, Wassie EA, Lasho TL, et al. Mutations and thrombosis in essential thrombocythemia: prognostic interaction with age and thrombosis history. Eur J Haematol. 2015;94(1):31-36. 11. Palandri F, Latagliata R, Polverelli N, et al. Mutations and long-term outcome of 217 young patients with essential thrombocythemia or early primary myelofibrosis. Leukemia. 2015;29(6):1344-1349. 12. Al Assaf C, Van Obbergh F, Billiet J, et al. Analysis of phenotype and outcome in essential thrombocythemia with CALR or JAK2 mutations. Haematologica. 2015;100 (7):893-897. 13. Tefferi A, Thiele J, Orazi A, et al. Proposals and rationale for revision of the World Health Organization diagnostic criteria for

14. 15.

16.

17.

18.

polycythemia vera, essential thrombocythemia, and primary myelofibrosis: recommendations from an ad hoc international expert panel. Blood. 2007;110(4):1092-1097. Jann B. Splitting time-span records with categorical time-varying covariates. Stata J. 2004;4(2):221-222. 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. Barbui T, Vannucchi AM, Buxhofer-Ausch V, et al. Practice-relevant revision of IPSETthrombosis based on 1019 patients with WHO-defined essential thrombocythemia. Blood Cancer J. 2015;5:e369. Torregrosa JM, Ferrer-Marín F, Lozano ML, et al. Impaired leucocyte activation is underlining the lower thrombotic risk of essential thrombocythaemia patients with CALR mutations as compared with those with the JAK2 mutation. Br J Haematol. 2016;172(5):813-815. Arellano-Rodrigo E, Alvarez-Larran A, Reverter JC, Colomer D, Bellosillo B, Cervantes F. CALR mutations are associated with lower platelet and endothelial activation than the JAK2 mutation in essential thrombocythemia. J Thromb Haemost. 2015;13 (Suppl 2):S2(abstract 435).

931

ARTICLE EUROPEAN HEMATOLOGY ASSOCIATION

Acute Myeloid Leukemia

Ferrata Storti Foundation

Targeted positron emission tomography imaging of CXCR4 expression in patients with acute myeloid leukemia Peter Herhaus,1 Stefan Habringer,1,2,# Kathrin Philipp-Abbrederis,1 Tibor Vag,3 Carlos Gerngross,3 Margret Schottelius,4 Julia Slotta-Huspenina,5 Katja Steiger,5 Torben Altmann,6 Tanja Weißer,1 Sabine Steidle,1 Markus Schick,1 Laura Jacobs,3 Jolanta Slawska,3 Catharina Müller-Thomas,1 Mareike Verbeek,1 Marion Subklewe,2,6 Christian Peschel,1,2 Hans-Jürgen Wester,5 Markus Schwaiger,2,3 Katharina Götze,1,2 and Ulrich Keller1,2

III Medical Department, Technische Universität München; 2German Cancer Consortium (DKTK), German Cancer Research Center (DKFZ), Heidelberg; 3Nuclear Medicine Department, Technische Universität München; 4Pharmaceutical Radiochemistry, Technische Universität München; 5Department of Pathology, Technische Universität München; and 6III Medical Department, Ludwig-Maximilians-Universität, Munich, Germany 1

Haematologica 2016 Volume 101(8):932-940

PH and SH contributed equally to this work.

ABSTRACT

Correspondence: ulrich.keller@tum.de

Received: January 18, 2016. Accepted: May 4, 2016. Pre-published: May 12, 2016. doi:10.3324/haematol.2016.142976

Check the online version for the most updated information on this article, online supplements, and information on authorship & disclosures: www.haematologica.org/content/101/8/932

932

cute myeloid leukemia originates from leukemia-initiating cells that reside in the protective bone marrow niche. CXCR4/CXCL12 interaction is crucially involved in recruitment and retention of leukemia-initiating cells within this niche. Various drugs targeting this pathway have entered clinical trials. To evaluate CXCR4 imaging in acute myeloid leukemia, we first tested CXCR4 expression in patient-derived primary blasts. Flow cytometry revealed that high blast counts in patients with acute myeloid leukemia correlate with high CXCR4 expression. The wide range of CXCR4 surface expression in patients was reflected in cell lines of acute myeloid leukemia. Next, we evaluated the CXCR4-specific peptide Pentixafor by positron emission tomography imaging in mice harboring CXCR4 positive and CXCR4 negative leukemia xenografts, and in 10 patients with active disease. [68Ga]Pentixafor-positron emission tomography showed specific measurable disease in murine CXCR4 positive xenografts, but not when CXCR4 was knocked out with CRISPR/Cas9 gene editing. Five of 10 patients showed tracer uptake correlating well with leukemia infiltration assessed by magnetic resonance imaging. The mean maximal standard uptake value was significantly higher in visually CXCR4 positive patients compared to CXCR4 negative patients. In summary, in vivo molecular CXCR4 imaging by means of positron emission tomography is feasible in acute myeloid leukemia. These data provide a framework for future diagnostic and theranostic approaches targeting the CXCR4/CXCL12-defined leukemia-initiating cell niche. Introduction Acute myeloid leukemia (AML) is an aggressive hematologic neoplasm originating from a myeloid hematopoietic stem/precursor cell (HSPC). AML is rapidly fatal if untreated. Although rates of complete remission after initial induction chemotherapy approach 70%, many patients relapse. Prognosis remains particularly dismal for those patients with adverse prognostic disease features (i.e. poor risk cytogenetics and/or poor risk molecular genetics), as well as for elderly patients unable to undergo intensive therapy, highlighting the clinical need for effective novel therapeutic strategies.1-3 Acute myeloid leukemia relapses are thought to arise from quiescent leukemiahaematologica | 2016; 101(8)

In vivo CXCR4 imaging of AML

initiating cells (LIC) harbored by the specialized bone marrow (BM) microenvironment, termed the stem cell niche. Several pre-clinical studies have shown that LICs are resistant to conventional chemotherapy as well as targeted therapy, and are selectively protected by interaction with the stem cell niche. Cross-talk between LICs and niche cells has also been demonstrated to be important for disease maintenance and progression.4-6 Thus, targeting the BM niche is an emerging and attractive therapeutic concept in AML. CXC-motif chemokine receptor 4 (CXCR4) functions together with its sole known chemokine ligand CXCL12 (also named Stromal cell-derived factor-1, SDF-1) as a master regulator of leukocyte migration and homing, and of HSPC retention in BM niches.7-11 CXCR4 is physiologically expressed on myeloid and lymphoid cells as well as on subtypes of epithelial cells. The activation of the CXCR4/CXCL12 pathway has been identified in several hematologic and solid malignancies.12 In this context, the CXCR4/CXCL12 axis is a key regulator of proliferation, chemotaxis to organs that secrete CXCL12, and aberrant angiogenesis, all of which are pivotal mechanisms of tumor progression and metastasis.13 The interaction between CXCR4 on malignant cells and secreted CXCL12 from the microenvironment is a fundamental component of the crosstalk between LIC and their niche.14 The CXCR4/CXCL12 axis is essential for both normal and leukemic HSPC migration in vivo.15,16 In NOD/SCID mice, homing and subsequent engraftment of normal human or AML HSPC are dependent on the expression of cell surface CXCR4, and CXCL12 produced within the murine BM niche.9,14 As shown for several other cancers, CXCR4 expression negatively impacts prognosis in AML.17 Recent data in acute lymphoblastic leukemia (ALL) further substantiate the crucial role of this interaction in acute leukemia.18,19 Therefore, targeting CXCR4 and the CXCR4/CXCL12defined LIC niche is an obvious and highly promising approach for long-term cure of hematopoietic stem cell malignancies, and CXCR4 is clearly a druggable target. Consequently, several novel therapies involving antibodies or small molecule drugs directed against CXCR4 or CXCL12 are currently being evaluated in clinical trials, with encouraging results.20-22 Our previous work identified the high affinity/specificity CXCR4-binding peptide Pentixafor as a suitable tracer for molecular in vivo CXCR4 positron emission tomography (PET) imaging in lymphoid malignancies.23,24 Beyond imaging, however, and in particular in systemic malignancies like lymphoma and leukemia, the real impact of such a peptide would be its therapeutic application. Pentixafor labeled to therapeutic radionuclides is feasible and has already been applied in individual patients with multiple myeloma,25 and a phase I/II clinical trial is currently under investigation (EudraCT: 2015-001817-28). The data presented here identify CXCR4 as a suitable target for imaging in AML, implying the potential for CXCR4-directed peptide-receptor radiotherapy (PRRT) in acute leukemia.

ary AML (sAML) were investigated for CXCR4 surface expression by flow cytometry. Ten patients with active myeloid disease underwent PET imaging for CXCR4. Five patients with non-hematologic malignancies examined through different analytical approaches served as controls. As previously reported for other [68Ga]-labeled peptides,26 [68Ga]Pentixafor was administered under the conditions of pharmaceutical law (The German Medicinal Products Act, AMG, Section 13, 2b) according to the German law and in accordance with the regulatory agencies responsible (Regierung von Oberbayern). All patients gave written informed consent prior to the investigation. The Ethics Committee of the Technische Universität München approved data analysis. Detailed information on patients' characteristics are provided in the Online Supplementary Appendix.

Cell lines The following human AML cell lines were used: Molm-13, MV411, NOMO-1, NB4, KG1a, OCI-AML2, OCI-AML3, Mono-Mac-1, Mono-Mac-6, OCI-AML5, GF-D8. The human Burkitt lymphoma line Daudi served as a positive control for CXCR4 expression. For details see the Online Supplementary Appendix.

RNA isolation and real-time PCR Assessment of CXCR4 mRNA of AML cell lines was performed as described in the Online Supplementary Appendix.

CRISPR-Cas9 mediated knock-out of CXCR4 OCI-AML3 cells were stably transduced with lentiCRISPRv2 (Addgene plasmid #52961), coding for Cas9 and a CXCR4-specific sgRNA. Indel formation was assessed as described previously.27 Additional information is provided in the Online Supplementary Appendix.

Migration assay Cell migration towards CXCL12 (R&D Systems, Minneapolis, MN, USA) was performed in transwell plates with 5 μm pore size (Corning Inc., Corning, NY, USA) and was quantified with CountBright beads (Thermo Fisher, Waltham, MA, USA). For details see the Online Supplementary Appendix.

Mice and tumor xenograft experiments Animal studies were performed in agreement with the Guide for Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication n. 85-23, revised 1996), in compliance with the German law on the protection of animals, and with the approval of the regional authorities responsible (Regierung von Oberbayern). PET scans of xenotransplanted AML cell lines in SCID mice were performed as previously described24 and are described briefly in the Online Supplementary Appendix.

Flow cytometry and immunohistochemistry The following antibodies were used for flow cytometry: Beckman Coulter: CD45-ECD (clone J33), CD34-FITC (clone 581), CD117-PE (clone 104D2D1); BD Biosciences (Franklin Lakes, NJ, USA): CXCR4-PE (clone 12G5), PE mouse IgG2a (Clone G155-178); for immunohistochemistry: ab12482 (clone UMB-2, abcam, Cambridge, UK), CD34 (QBEnd/10, Cell Marque), CD117 (c-kit, Dako), CD43 (Novocastra). Further details are provided in the Online Supplementary Appendix.

Methods Patients

PET/MR and PET/CT imaging studies in patients and animals

Samples from 67 unselected patients with active myeloid disease (myelodysplastic syndrome (MDS), de novo AML or second-

[68Ga]Pentixafor was synthesized and PET/MRI analysis was performed as previously described.28-33 Detailed descriptions of

haematologica | 2016; 101(8)

933

P. Herhaus et al.

imaging protocols are provided in the Online Supplementary Appendix.

Statistical analysis All statistical tests were performed using GraphPad Prism (GraphPad Software, La Jolla, CA, USA). P<0.05 was considered statistically significant. Quantitative values were expressed as mean±standard deviation (SD) or standard error of the mean (SEM) as indicated. Additional information is given in the Online Supplementary Appendix.

Results CXCR4 is highly expressed on leukemic blasts in a subset of AML patients To address CXCR4 abundance in myeloid malignancies, we first assessed CXCR4 expression in an unselected cohort of 67 consecutive patients with active disease (AML, MDS) by flow cytometry of bone marrow (BM) and/or peripheral blood (PB). For details of patients' characteristics see Online Supplementary Table S1. Myeloid blasts were gated as CD45dim cells, and CD117 was used as a marker for myeloid blasts (gating strategy depicted in Online Supplementary Figure S1A). Lymphocytes with known CXCR4 positivity served as an intraindividual control (Online Supplementary Figure S1B). These analyses revealed a wide range of surface CXCR4 expression on myeloid blasts, from virtually absent expression to high levels in a distinct subset of AML patients. Representative flow cytometry data from AML patients are shown in Figure 1A. Quantification of CXCR4 surface expression showed significantly higher CXCR4 expression in patients with a blast percentage exceeding 30%. There was a trend towards higher CXCR4 expression in blasts derived from AML samples compared to MDS samples (Figure 1B and C). No significant correlation between high CXCR4

expression on blasts and disease stage (first diagnosis vs. refractory/relapsed disease), de novo vs. sAML, age (<65 vs. ≥65 years), prognostic risk group according to the modified ELN classification34 or existing genetic aberrations was found (Online Supplementary Figure S2A-F). No significantly different CXCR4 surface expression in paired PB and BM samples was observed (Online Supplementary Figure S2G).

[68Ga]Pentixafor-PET enables in vivo CXCR4 imaging of AML xenografts Since CXCR4 is an attractive target for novel therapeutic approaches directed against the leukemic microenvironment, we sought to evaluate the clinical applicability of the novel CXCR4-binding PET tracer Pentixafor labeled with a Gallium isotope (68Ga), [68Ga]Pentixafor, in myeloid malignancies. To select appropriate AML cell lines to model AML with detectable CXCR4 expression, transcript levels and surface expression of CXCR4 was evaluated in ten established AML cell lines. As expected from flow cytometry data in AML patients (Figure 1), CXCR4 expression in cell lines ranged from low (KG1a) to high (NOMO-1, OCI-AML3) (Figure 2A and B). CXCR4 surface expression assessed by flow cytometry correlated with transcript levels (Figure 2C and D). Of all cell lines analyzed, OCI-AML3 showed the highest expression and was, therefore, chosen as a cell line for modeling CXCR4high AML in further imaging experiments. To test if PET imaging of AML cells with [68Ga]Pentixafor was feasible in vivo, we chose OCI-AML3 and NOMO-1 as CXCR4high and KG1a as CXCR4low cell line to generate subcutaneous xenograft mouse models. After tumor engraftment was apparent in all mice, [68Ga]Pentixafor and PET imaging was performed. NOMO-1 and OCI-AML3 xenografts were clearly visible, whereas KG1a xenografts were not (Figure 3A), demonstrating that CXCR4-high AML cells can be visualized with [68Ga]Pentixafor in vivo.

Figure 1. CXCR4 expression in patients with acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS). (A) Flow cytometric evaluation of CXCR4 surface expression using an anti-CXCR4 antibody. Blasts were gated as CD45low cell population. Anti-CD117 antibody was used for back-gating. Representative data of CXCR4 positive (upper panels) and CXCR4 negative (lower panels) patients are shown. (B and C) Median fluorescence intensity of surface CXCR4 expression relative to isotype control (n=67 patients). Horizontal bars indicate the mean of all individual patient values±SEM; Student’s t-test was used to compare mean relative blast CXCR4 expression. *Statistically significant differences between the groups. (B) MDS versus AML; P=0.062. (C) CXCR4 expression in patients with less than 30% blasts versus CXCR4 expression in patients with at least 30% blasts; P=0.004.

934

haematologica | 2016; 101(8)

In vivo CXCR4 imaging of AML

To further test the specificity of [68Ga]Pentixafor binding to CXCR4, OCI-AML3 cells were selected for a CRISPRCas9 based stable knock-out of CXCR4 using a modified lentiCRISPRv2 to co-express Streptococcus pyogenes Cas9 and sgRNAs directed against human CXCR4.35 This approach resulted in effective indel formation in the CXCR4 gene (Online Supplementary Figure S3A), reduction of CXCR4 surface expression (Figure 3B) and CXCL12dependent migration (Figure 3C), while the growth kinetics remained unaffected in vivo (Figure 3D) and in vitro (Online Supplementary Figure S3C). For in vivo experiments, sg2 (sequence in Online Supplementary Figure S3A), targeting exon 2 of CXCR4, was chosen. OCI-AML3 stably transduced with lentiCRISPRv2-sg2 and non-targeting lentiCRISPRv2 as control were subcutaneously injected into SCID mice. [68Ga]Pentixafor-PET imaging of these AML xenografts showed that OCI-AML3 control cells could be detected, and knock-out of CXCR4 in the same cell line abolished binding and PET positivity of AML xenografts. Binding of the imaging probe to mouse tissues was low, owing to the known specificity of [68Ga]Pentixafor to human CXCR4 (Figure 3B and E). Thus, in vivo PET imaging of AML xenografts with [68Ga]Pentixafor is feasible and enables visualizing AML cells in a CXCR4-dependent manner.

CXCR4 directed PET/MR imaging in patients with myeloid malignancies Our findings in the AML xenograft model (Figure 3), the specific binding characteristics of [68Ga]Pentixafor to human CXCR4,23,24 as well as the expression data generated in the flow cytometry patient cohort (Figure 1) encouraged us to test if CXCR4 imaging was also feasible in patients with myeloid malignancies. For this purpose, CXCR4-directed PET was combined with MR imaging, a method that is suitable for evaluating replacement of normal BM by malignant processes, including AML.36 Ten patients underwent [68Ga]Pentixafor-PET imaging after signing informed consent. In 9 of the 10 patients, PET was combined with a whole body magnetic resonance (MR) imaging approach. In one patient, a PET/CT was conducted. One patient with extramedullary relapse and absence of BM infiltration as shown by biopsy received [68Ga]Pentixafor-PET/MR and standard [18F]FDG-PET/CT. Eight of 10 patients who underwent PET/MR imaging had BM involvement of AML, and one had an MDS-RAEB2. For details of patients' characteristics see Online Supplementary Table S3. Four out of 9 patients with BM involvement were visually positive as assessed by [68Ga]Pentixafor-PET. The PET positive areas correlated well with the expected signal

Figure 2. Surface CXCR4 expression of acute myeloid leukemia (AML) cell lines correlates with transcript levels. (A) Flow cytometric evaluation of CXCR4 surface expression of the indicated AML cell lines using an anti-CXCR4 antibody. An isotype control antibody was used as a control. (B) Mean fluorescence intensity of surface CXCR4 expression relative to isotype control. Three replicates for each cell line were used. (C) CXCR4 transcript levels measured by qRT-PCR. Mean relative expression±SEM is shown (n=3 independent experiments). ΔΔCt values relative to ubiquitin (Ub) were normalized to those of peripheral blood mononuclear cells (PBMC) of 3 healthy individuals. (D) Correlation analysis between relative CXCR4 transcript and relative CXCR4 surface expression levels.

haematologica | 2016; 101(8)

935

P. Herhaus et al.

To correlate in vivo imaging of CXCR4 with its expression level within the AML compartment, immunohistochemistry for CXCR4 was performed in 3 patients where BM biopsies in close time proximity to PET imaging were available. The high CXCR4 expression determined by IHC in Patient #1 and Patient #4 correlated well with tracer uptake detected by [68Ga]Pentixafor-PET. Patient #10, who was visually negative in [68Ga]Pentixafor-PET, revealed an undetectable to low CXCR4 expression as assessed by IHC (Figure 5). In summary, these results reveal that in vivo imaging of myeloid malignancies, especially AML, is feasible with the new PET-tracer [68Ga]Pentixafor. The variability in PET positivity for CXCR4 reflects the wide range of CXCR4 surface expression obtained with flow cytometry. Due to the limited number of patients, and the missing data on CXCR4 surface expression at the time of imaging in several patients, a statistically significant correlation between Pentixafor uptake and CXCR4 surface expression analyzed by flow cytometry and/or IHC cannot be made at this time; this will be investigated in a large planned prospective study (EudraCT 2014-003411-12).

alterations as determined by MR imaging (n=4, representative images shown in Figure 4A-F). Five of the 9 patients were visually graded as PET negative (representative images shown in Figure 4G-I). To clearly depict those differences between PET positive and negative AML and control patients, the vertebra are the best examples. Whereas all AML patients show decreased BM signal in the T1w MR sequences (Figure 4B, E, H), those BM areas only show elevated tracer uptake in the PET positive patients (Figure 4C and F). The tracer uptake within the infiltrated BM areas of the PET negative AML patient (Figure 4I) resembles those of the control patient without BM signal alterations in T1w MR sequences (Figure 4K and L). In order to allow for standardized evaluation of SUV, 5 anatomic locations with active hematopoiesis in adults were chosen for the quantification of the PET signal (Figure 4M). Compared to visually PET negative AML patients and patients with non-hematologic malignancies, the SUVmax of the five pre-defined areas of measurement was significantly higher in PET positive patients (Figure 4M). The calculated meanSUVmax was significantly higher in patients with PET positive AML compared to PET negative AML (Figure 4N). One of the 10 patients imaged had biopsy-proven with [68Ga]Pentixafor-PET extramedullary relapse of AML after allogeneic stem cell transplantation (SCT) in the absence of BM involvement. [68Ga]Pentixafor-PET/CT imaging in this patient revealed visually positive extramedullary disease and normal background BM signal. The extramedullary lesion showed a SUVmax of 5.2, comparable to the meanSUVmax measured in the BM of [68Ga]Pentixafor-PET positive patients. Moreover, this CXCR4 positive lesion displayed high [18F]FDG uptake (SUVmax 9.51) in the routine diagnostic [18F]FDG-PET (Online Supplementary Figure S4).

Discussion There are compelling data to show that the BM microenvironment contributes to treatment resistance and relapse in AML. CXCR4 and its ligand CXCL12 are essential for retention of normal HSPC and LICs within their protective niche and are, therefore, considered attractive targets for overcoming microenvironment-mediated resistance and inevitable subsequent clinical leukemia relapse.

Figure 3. In vivo Pentixafor PET imaging in acute myeloid leukemia (AML) correlates with CXCR4 surface expression and migration towards a CXCL12 gradient. (A) [68Ga]Pentixafor-PET imaging of AML xenografts. The indicated cell lines were injected into immunodeficient mice to generate xenograft tumors. CXCR4 expression was then analyzed using in vivo [68Ga]Pentixafor-PET (upper panels). CXCR4 surface expression was analyzed by flow cytometry (lower panels). N=2 tumors/cell line; n=1 mouse/cell line. (B) [68Ga]Pentixafor-PET imaging of control and CXCR4 knock-out (sg2) OCI-AML3 xenografts (upper panel). The lower panel shows CXCR4 surface expression as assessed by CD184 flow cytometry. A representative image and histogram is shown. (C) CRISPR/Cas9-mediated CXCR4 knock-out results in significantly reduced migration towards a CXCL12 gradient. OCI-AML3 cells were assessed using a transwell chamber migration assay. N=3 independent experiments. Mean±SEM is shown. *P=0.002 (Student’s t-test). (D) Images of the explanted tumor shown in (B) and (C) (left panel). Tumor weight (right panel). Mean±SEM, no significant difference. (E) Quantification of [68Ga]Pentixafor uptake. Xenograft tumors were analyzed by means of voxel intensity measurement. Mean±SEM is shown, n=3 tumors for control and sg2, n=3 mice; *P=0.049 (Student’s t-test).

936

haematologica | 2016; 101(8)

In vivo CXCR4 imaging of AML

The clinical significance of CXCR4 in AML is underscored by data showing that high CXCR4 expression on AML blasts correlates with poor prognosis.17,37-39 In a pediatric AML cohort, blast CXCR4 surface expression was increased by chemotherapy and contributed to resistance.40 There was no significant difference in CXCR4 surface expression between prognostic groups according to the modified ELN prognostic system34 in our cohort, possibly due to sample size. In agreement with previous stud-

ies, CXCR4 surface expression in our cohort was highly variable. High CXCR4 expression correlated with high blast counts in our cohort, which might account for the poor prognosis seen in other studies. In addition to aberrant expression of CXCR4 in a substantial proportion of AML patients, ligand-mediated phosphorylation of serine 339 of CXCR4 appears to drive resistance to chemotherapy, and to increase retention of AML cells within the BM.41 Such augmented interaction with the BM niche, in partic-

Figure 4. [68Ga]Pentixafor-PET/magnetic resonance (MR) imaging in acute myeloid leukemia (AML) patients. (A-F) Shown are 2 AML patients (#2 and #1) with visually positive [68Ga]Pentixafor-PET/MR imaging. (G-I)[68Ga]Pentixafor-PET/MR images of a visually negative AML patient. (J-L) Control patient without BM malignancy who underwent [68Ga]Pentixafor-PET/MR imaging. (A, D, G, J) Maximum intensity projections of [68Ga]Pentixafor uptake. (B, E, H, K) T1w MR imaging coronal sections. (C, F, I, L) Coronal PET/MR imaging fusion. (M) (Left) Schematic graph of locations assessed for SUV quantification. 1: cervical vertebra (7); 2: thoracic vertebra (12); 3: right os ilium; 4: lumbal vertebra (5); 5: left os ilium. (Right) Heatmap of SUV values in the 5 visually positive (AML+), 5 visually negative (AMLâ&#x20AC;&#x201C;), and 5 control patients with non-hematologic disease (control). *Patient #5 was scored positive because of a [68Ga]Pentixafor-PET positive extramedullary lesion. (N) Quantification of SUV values from (m). *P=0.036 for AML+ versus AMLâ&#x20AC;&#x201C; and P=0.040 for AML+ versus control. Error Bars represent the SEM. Patient #5 was excluded due to the lack of bone marrow involvement (extramedullary AML).

haematologica | 2016; 101(8)

937

P. Herhaus et al. A

Figure 5. CXCR4 expression in bone marrow of acute myeloid leukemia (AML) patients undergoing [68Ga]Pentixafor imaging. (A-C) Representative H&E stains of 3 AML patients show hypercellular bone marrow (BM) with blast infiltration; embedded are the PET images of the corresponding patients; (A) and (B) are visually positive for CXCR4-directed PET and (C) is negative. (D-F) IHC for patient specific myeloid/blast markers; stained markers are shown in white. (G-I) IHC for CXCR4 in the corresponding BM samples. (A, D, G) Patient #1. (B, E, H) Patient #4. (C, F, I) Patient #10.

ular differentiating osteoblasts, has recently been shown to counteract the induction of apoptosis within the leukemic compartment which can be triggered by CXCL12 ligation to CXCR4.42,43 Against this background, it is currently unclear what impact CXCR4 targeting by small molecule CXCR4 antagonists or monoclonal antibodies will have in the clinic, and, in particular, on eliminating the LICs that fundamentally contribute to relapse. Despite this mechanistic uncertainty, the first-in-class CXCR4 inhibitor AMD3100 (Plerixafor) has been tested as a chemosensitizing agent in relapsed or refractory AML in a phase I/II trial with encouraging preliminary results.44 Further trials involving monoclonal antibodies and novel CXCR4-targeting small molecule inhibitors such as BL8040 are under way (EudraCT 2014-002702-21). Disrupting ligand-mediated CXCR4 downstream activity by antagonists is one approach currently being tested. Physically targeting the BM niche characterized by the CXCR4-CXCL12 interaction could be an attractive alternative. One highly interesting method that provides such physical targeting is peptide receptor radionuclide therapy (PRRT). PRRT has been successfully integrated into the therapeutic algorithm of neuroendocrine tumors (NETs).45 It usually involves the diagnostic imaging of the receptor 938

to ensure target expression, followed by the application of a therapeutically labeled peptide (e.g. Lutetium-177 octreotate), thus constituting a theranostic procedure. In patients with AML, an endoradiotherapeutic approach with CD45 as target has been successfully tested in a phase I/II trial in the conditioning regimen prior to allogeneic SCT.46 For such a purpose, the data presented within our CXCR4 examinations represent an important step, as they show that, at least in a subgroup of patients, there is a substantial expression of CXCR4, and that AML can even be imaged using the novel CXCR4-specific molecular PET probe Pentixafor. Pentixafor has already been labeled with therapeutic radionuclides such as 99Yttrium and 177Lutetium, and compassionate use therapies have been applied to patients with very advanced multiple myeloma.25 A phase I/II study in myeloma using the CXCR4-directed theranostic approach is currently under investigation (EudraCT 2015-001817-28). With regard to AML, however, it is still not at all clear whether measurable high CXCR4 expression is a prerequisite for such a therapy, since it can be assumed that targeting the niche via CXCR4 could have an effect on all hematopoietic cells harbored there. The imaging data presented in our study reveals crucial information on in vivo CXCR4 expression in haematologica | 2016; 101(8)

In vivo CXCR4 imaging of AML

myeloid malignancy. Although we still have no data on ALL, very recent work defines the CXCR4/CXCL12 interaction as crucial for disease maintenance and progression in ALL.18,19 We are continuing to learn more about both the molecular and the genetic characterization of AML and ALL.47 Thus, markers for detecting MRD are available that provide high sensitivity,48 avoiding the need for additional imaging. We foresee the major application of CXCR4 targeting using the herein described CXCR4-binding peptide within a theranostic approach, i.e. as a conditioning regimen within an allogeneic SCT. The importance of the CXCR4/CXCL12 axis as a label of the LIC niche, as well as the observation that relapsed leukemias frequently

References 1. Byrd JC, Mrozek K, Dodge RK, et al. Pretreatment cytogenetic abnormalities are predictive of induction success, cumulative incidence of relapse, and overall survival in adult patients with de novo acute myeloid leukemia: results from Cancer and Leukemia Group B (CALGB 8461). Blood. 2002;100(13):4325-4336. 2. Mayer RJ, Davis RB, Schiffer CA, et al. Intensive postremission chemotherapy in adults with acute myeloid leukemia. Cancer and Leukemia Group B. N Engl J Med. 1994;331(14):896-903. 3. Sekeres MA. Treatment of older adults with acute myeloid leukemia: state of the art and current perspectives. Haematologica. 2008;93(12):1769-1772. 4. Schepers K, Campbell TB, Passegue E. Normal and leukemic stem cell niches: insights and therapeutic opportunities. Cell Stem Cell. 2015;16(3):254-267. 5. Sipkins DA, Wei X, Wu JW, et al. In vivo imaging of specialized bone marrow endothelial microdomains for tumour engraftment. Nature. 2005;435(7044):969973. 6. Matsunaga T, Takemoto N, Sato T, et al. Interaction between leukemic-cell VLA-4 and stromal fibronectin is a decisive factor for minimal residual disease of acute myelogenous leukemia. Nat Med. 2003;9(9): 1158-1165. 7. Sugiyama T, Kohara H, Noda M, Nagasawa T. Maintenance of the hematopoietic stem cell pool by CXCL12-CXCR4 chemokine signaling in bone marrow stromal cell niches. Immunity. 2006;25(6):977-988. 8. Shen H, Cheng T, Olszak I, et al. CXCR-4 desensitization is associated with tissue localization of hemopoietic progenitor cells. J Immunol. 2001;166(8):5027-5033. 9. Peled A, Petit I, Kollet O, et al. Dependence of human stem cell engraftment and repopulation of NOD/SCID mice on CXCR4. Science. 1999;283(5403):845-848. 10. Lapidot T. Mechanism of human stem cell migration and repopulation of NOD/SCID and B2mnull NOD/SCID mice. The role of SDF-1/CXCR4 interactions. Ann N Y Acad Sci. 2001;938:83-95. 11. Zou YR, Kottmann AH, Kuroda M, Taniuchi I, Littman DR. Function of the chemokine receptor CXCR4 in haematopoiesis and in cerebellar develop-

haematologica | 2016; 101(8)

express high levels of CXCR4, makes radiolabeled CXCR4 targeting an attractive novel therapeutic approach. Acknowledgments The authors would like to thank the staff of the diagnostic laboratory of the III Medical Department and the staff of the small animal PET facility of the Nuclear Medicine Department, TU MĂźnchen, Germany, for their assistance. Funding UK, HJW, and MS received support from the Deutsche Forschungsgemeinschaft (DFG, SFB824). UK was further supported by DFG (grant KE 222/7-1). This work received support from the German Cancer Consortium (DKTK).

ment. Nature. 1998;393(6685):595-599. 12. Zhao H, Guo L, Zhao H, et al. CXCR4 overexpression and survival in cancer: a system review and meta-analysis. Oncotarget. 2015;6(7):5022-5040. 13. Teicher BA, Fricker SP. CXCL12 (SDF1)/CXCR4 pathway in cancer. Clin Cancer Res. 2010;16(11):2927-2931. 14. Tavor S, Petit I, Porozov S, et al. CXCR4 regulates migration and development of human acute myelogenous leukemia stem cells in transplanted NOD/SCID mice. Cancer Res. 2004;64(8):2817-2824. 15. Lapidot T, Sirard C, Vormoor J, et al. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature. 1994;367(6464):645-648. 16. Kollet O, Spiegel A, Peled A, et al. Rapid and efficient homing of human CD34(+)CD38(-/low)CXCR4(+) stem and progenitor cells to the bone marrow and spleen of NOD/SCID and NOD/SCID/B2m(null) mice. Blood. 2001;97(10):3283-3291. 17. Spoo AC, Lubbert M, Wierda WG, Burger JA. CXCR4 is a prognostic marker in acute myelogenous leukemia. Blood. 2007;109(2):786-791. 18. Pitt LA, Tikhonova AN, Hu H, et al. CXCL12-Producing Vascular Endothelial Niches Control Acute T Cell Leukemia Maintenance. Cancer Cell. 2015;27(6):755768. 19. Passaro D, Irigoyen M, Catherinet C, et al. CXCR4 Is Required for Leukemia-Initiating Cell Activity in T Cell Acute Lymphoblastic Leukemia. Cancer Cell. 2015;27(6):769-779. 20. Cho B-S, Zeng Z, Mu H, et al. Antileukemia activity of the novel peptidic CXCR4 antagonist LY2510924 as monotherapy and in combination with chemotherapy. Blood. 2015;126(2):222-232. 21. Zeng Z, Shi YX, Samudio IJ, et al. Targeting the leukemia microenvironment by CXCR4 inhibition overcomes resistance to kinase inhibitors and chemotherapy in AML. Blood. 2009;113(24):6215-6224. 22. Nervi B, Ramirez P, Rettig MP, et al. Chemosensitization of acute myeloid leukemia (AML) following mobilization by the CXCR4 antagonist AMD3100. Blood. 2009;113(24):6206-6214. 23. Philipp-Abbrederis K, Herrmann K, Knop S, et al. In vivo molecular imaging of chemokine receptor CXCR4 expression in patients with advanced multiple myeloma.

EMBO Mol Med. 2015;7(4):477-487. 24. Wester HJ, Keller U, Schottelius M, et al. Disclosing the CXCR4 expression in lymphoproliferative diseases by targeted molecular imaging. Theranostics. 2015;5(6): 618-630. 25. Herrmann K, Schottelius M, Lapa C, et al. First-in-man experience of CXCR4-directed endoradiotherapy with 177Lu- and 90Ylabelled pentixather in advanced stage multiple myeloma with extensive intra- and extramedullary disease. J Nucl Med. 2016;57(2):248-251. 26. Haug AR, Cindea-Drimus R, Auernhammer CJ, et al. Neuroendocrine tumor recurrence: diagnosis with 68GaDOTATATE PET/CT. Radiology. 2014;270 (2):517-525. 27. Brinkman EK, Chen T, Amendola M, van Steensel B. Easy quantitative assessment of genome editing by sequence trace decomposition. Nucleic Acids Res. 2014;42(22): e168. 28. Demmer O, Gourni E, Schumacher U, Kessler H, Wester HJ. PET Imaging of CXCR4 receptors in cancer by a new optimized ligand. Chem Med Chem. 2011; 6(10):1789-1791. 29. Gourni E, Demmer O, Schottelius M, et al. PET of CXCR4 expression by a (68)Galabeled highly specific targeted contrast agent. J Nucl Med. 2011;52(11):1803-1810. 30. Martin R, Juttler S, Muller M, Wester HJ. Cationic eluate pretreatment for automated synthesis of [(6)(8)Ga]CPCR4.2. Nucl Med Biol. 2014;41(1):84-89. 31. Drzezga A, Souvatzoglou M, Eiber M, et al. First clinical experience with integrated whole-body PET/MR: comparison to PET/CT in patients with oncologic diagnoses. J Nucl Med. 2012;53(6):845-855. 32. Silva JR Jr, Hayashi D, Yonenaga T, et al. MRI of bone marrow abnormalities in hematological malignancies. Diagn Interv Radiol. 2013;19(5):393-399. 33. Zamagni E, Patriarca F, Nanni C, et al. Prognostic relevance of 18-F FDG PET/CT in newly diagnosed multiple myeloma patients treated with up-front autologous transplantation. Blood. 2011;118(23):5989-5995. 34. Estey EH. Acute myeloid leukemia: 2014 update on risk-stratification and management. Am J Hematol. 2014;89(11):10631081. 35. Sanjana NE, Shalem O, Zhang F. Improved vectors and genome-wide libraries for

939

P. Herhaus et al.

36.

37.

38.

39.

40.

940

CRISPR screening. Nat Methods. 2014;11 (8):783-784. De Silva RA, Peyre K, Pullambhatla M, et al. Imaging CXCR4 Expression in Human Cancer Xenografts: Evaluation of Monocyclam Cu-64-AMD3465. J Nucl Med. 2011;52(6):986-993. Rombouts EJ, Pavic B, Lowenberg B, Ploemacher RE. Relation between CXCR-4 expression, Flt3 mutations, and unfavorable prognosis of adult acute myeloid leukemia. Blood. 2004;104(2):550-557. Tavernier-Tardy E, Cornillon J, Campos L, et al. Prognostic value of CXCR4 and FAK expression in acute myelogenous leukemia. Leuk Res. 2009;33(6):764-768. Bae MH, Oh SH, Park CJ, et al. VLA-4 and CXCR4 expression levels show contrasting prognostic impact (favorable and unfavorable, respectively) in acute myeloid leukemia. Ann Hematol. 2015;94(10):16311638. Sison EA, McIntyre E, Magoon D, Brown P.

41.

42.

43.

44.

Dynamic chemotherapy-induced upregulation of CXCR4 expression: a mechanism of therapeutic resistance in pediatric AML. Mol Cancer Res. 2013;11(9):1004-1016. Brault L, Rovo A, Decker S, et al. CXCR4SERINE339 regulates cellular adhesion, retention and mobilization, and is a marker for poor prognosis in acute myeloid leukemia. Leukemia. 2014;28(3):566-576. Kremer KN, Dudakovic A, McGeeLawrence ME, et al. Osteoblasts protect AML cells from SDF-1-induced apoptosis. J Cell Biochem. 2014;115(6):1128-1137. Kremer KN, Peterson KL, Schneider PA, et al. CXCR4 chemokine receptor signaling induces apoptosis in acute myeloid leukemia cells via regulation of the Bcl-2 family members Bcl-XL, Noxa, and Bak. J Biol Chem. 2013;288(32):22899-22914. Uy GL, Rettig MP, Motabi IH, et al. A phase 1/2 study of chemosensitization with the CXCR4 antagonist plerixafor in relapsed or refractory acute myeloid leukemia. Blood.

2012;119(17):3917-3924. 45. van Essen M, Krenning EP, Bakker WH, et al. Peptide receptor radionuclide therapy with 177Lu-octreotate in patients with foregut carcinoid tumours of bronchial, gastric and thymic origin. Eur J Nucl Med Mol Imaging. 2007;34(8):1219-1227. 46. Pagel JM, Gooley TA, Rajendran J, et al. Allogeneic hematopoietic cell transplantation after conditioning with 131I-antiCD45 antibody plus fludarabine and lowdose total body irradiation for elderly patients with advanced acute myeloid leukemia or high-risk myelodysplastic syndrome. Blood. 2009;114(27):5444-5453. 47. Swerdlow SH, Campo E, Harris NL, et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. IARC. 2008. 48. Ossenkoppele GJ, Schuurhuis GJ. MRD in AML: it is time to change the definition of remission. Best Pract Res Clin Haematol. 2014;27(3-4):265-271.

haematologica | 2016; 101(8)

ARTICLE

Acute Lymphoblastic Leukemia

RNA sequencing unravels the genetics of refractory/relapsed T-cell acute lymphoblastic leukemia. Prognostic and therapeutic implications

EUROPEAN HEMATOLOGY ASSOCIATION

Ferrata Storti Foundation

Valentina Gianfelici,1* Sabina Chiaretti,1* Sofie Demeyer,2,3* Filomena Di Giacomo,1,4 Monica Messina,1 Roberta La Starza,5 Nadia Peragine,1 Francesca Paoloni,6 Ellen Geerdens,2,3 Valentina Pierini,5 Loredana Elia,1 Marco Mancini,1 Maria Stefania De Propris,1 Valerio Apicella,1 Gianluca Gaidano,7 Anna Maria Testi,1 Antonella Vitale,1 Marco Vignetti,1,6 Cristina Mecucci,5 Anna Guarini,1 Jan Cools,2,3 and Robin FoĂ 1 Hematology, Department of Cellular Biotechnologies and Hematology, Sapienza University, Rome, Italy; 2Center for Human Genetics, KU Leuven, Belgium; 3Center for the Biology of Disease, VIB, Leuven, Belgium; 4Department of Molecular Biotechnology and Health Science and Center for Experimental Research and Medical Studies (CeRMS), University of Turin, Italy; 5Hematology and Bone Marrow Transplantation Unit, Department of Medicine, University of Perugia, Italy; 6GIMEMA Data Center, Rome, Italy; and 7Division of Hematology, Department of Translational Medicine, Amedeo Avogadro University of Eastern Piedmont, Novara, Italy 1

* VG, SC and SD contributed equally to this work.

Haematologica 2016 Volume 101(8):941-950

ABSTRACT

espite therapeutic improvements, a sizable number of patients with T-cell acute lymphoblastic leukemia still have a poor outcome. To unravel the genomic background associated with refractoriness, we evaluated the transcriptome of 19 cases of refractory/early relapsed T-cell acute lymphoblastic leukemia (discovery cohort) by performing RNA-sequencing on diagnostic material. The incidence and prognostic impact of the most frequently mutated pathways were validated by Sanger sequencing on genomic DNA from diagnostic samples of an independent cohort of 49 cases (validation cohort), including refractory, relapsed and responsive cases. Combined gene expression and fusion transcript analyses in the discovery cohort revealed the presence of known oncogenes and identified novel rearrangements inducing overexpression, as well as inactivation of tumor suppressor genes. Mutation analysis identified JAK/STAT and RAS/PTEN as the most commonly disrupted pathways in patients with chemorefractory disease or early relapse, frequently in association with NOTCH1/FBXW7 mutations. The analysis on the validation cohort documented a significantly higher risk of relapse, inferior overall survival, disease-free survival and event-free survival in patients with JAK/STAT or RAS/PTEN alterations. Conversely, a significantly better survival was observed in patients harboring only NOTCH1/FBXW7 mutations: this favorable prognostic effect was abrogated by the presence of concomitant mutations. Preliminary in vitro assays on primary cells demonstrated sensitivity to specific inhibitors. These data document the negative prognostic impact of JAK/STAT and RAS/PTEN mutations in T-cell acute lymphoblastic leukemia and suggest the potential clinical application of JAK and PI3K/mTOR inhibitors in patients harboring mutations in these pathways.

haematologica | 2016; 101(8)

Correspondence: rfoa@bce.uniroma1.it

Received: November 12, 2015. Accepted: April 29, 2016. Pre-published: May 5, 2016. doi:10.3324/haematol.2015.139410

Check the online version for the most updated information on this article, online supplements, and information on authorship & disclosures: www.haematologica.org/content/101/8/941

ÂŠ2016 Ferrata Storti Foundation Material published in Haematologica is covered by copyright. All rights reserved to the Ferrata Storti Foundation. Copies of articles are allowed for personal or internal use. Permission in writing from the publisher is required for any other use.

941

V. Gianfelici et al.

Introduction T-cell acute lymphoblastic leukemia (T-ALL) is a genetically heterogeneous disease caused by the accumulation of molecular lesions acting in a multistep pathogenic process.1,2 While more than 80% of children can expect to be cured nowadays, among adults younger than 60 years managed with conventional treatment the survival rates are in the range of 40-50% and older patients have a much worse outcome.3-6 Although the use of intensified strategies results in a survival advantage, many patients still relapse and eventually experience refractory leukemia associated with a poor likelihood of cure.4,5,7,8 Over the last years much effort has been put into understanding the molecular background of relapsed and chemotherapy-resistant ALL.9-12 In the pediatric setting, Tzoneva et al. identified mutations affecting the NT5C2 gene13, which appeared to be acquired at relapse and, overall, to be more frequent in T-ALL than in B-ALL; similar results were recently reported in T-ALL also by Kunz et al.14 A comprehensive description of the genetic alterations in patients with chemorefractory T-ALL has thus far been lacking. We previously reported that whole transcriptome sequencing (RNAseq) is a powerful approach for the detection of translocations, single nucleotide variants, small insertions/deletions (INDEL) and gene expression deregulation in T-ALL.15 In this study, we applied RNAseq to 19 diagnostic T-ALL samples from patients who were refractory to treatment or experienced an early relapse. Our aim was to identify key oncogenic pathways that may predict treatment failure and that could be targets for molecularly tailored therapies. The recurrence and the prognostic value of the lesions identified were validated in an additional set of 49 newly diagnosed T-ALL samples.

Methods Discovery cohort This study was carried out on T-ALL samples collected at diagnosis from 19 patients who proved refractory to first-line treatment (n=11), experienced an early relapse after first complete remission (n=7: median time to relapse 4 months; range, 1-9), and a single patient who relapsed after 16 months. Fourteen were male and five were female and they had a median age of 36 years (range, 11-55). The diagnosis of T-ALL was based on the World Health Organization (WHO) classification.16 Cytogenetic and molecular analyses were performed as reported elsewhere.17-19 Molecular screening included searches for recurrent fusion genes, e.g. BCR/ABL1, MLL and MLLT10 fusions, STIL/TAL1, NUP98/RAP1GDS1, SET/NUP214, NUP214/ABL1, ETV6/RUNX1 and E2A/PBX1. SET-NUP214 was identified in three cases and STIL-TAL1 in one, while the remaining 15 patients were negative for recurrent fusion genes (Online Supplementary Table S1). For all 19 patients, RNA from peripheral blood or bone marrow samples was analyzed by RNAseq on the initial diagnostic material collected before starting treatment at the Hematology Center, â&#x20AC;&#x2DC;Sapienzaâ&#x20AC;&#x2122; University of Rome. In the three relapsed cases with the longest duration of first complete remission, RNAseq was also performed on the matched relapse samples. All patients or guardians gave their informed consent to blood/marrow collection and biological analyses, in agreement with the Declaration of Helsinki. The study was approved by the Institutional Review Board. 942

RNA sequencing analysis on the discovery cohort RNAseq and bioinformatic analyses were performed at the Laboratory for the Molecular Biology of Leukemia, VIB-KU Leuven, Belgium. Paired-end sequencing was performed on an Illumina HiSeq2000 instrument. Sequence reads were processed to identify fusion transcripts, single nucleotide variants, INDEL and gene expression levels, as reported previously (Online Supplementary Figure S1).15

Validation of RNA sequencing data on the discovery cohort Candidate fusion transcripts were validated by reverse-transcription polymerase-chain reaction (RT-PCR) and Sanger sequencing (Online Supplementary Table S2). Fluorescence in situ hybridization (FISH) was applied to confirm the presence of fusion transcripts in four samples (Online Supplementary Methods). Candidate variants (single nucleotide variants and INDEL) were confirmed by PCR amplification and Sanger sequencing on genomic DNA (gDNA) (Online Supplementary Table S3). In six cases with available material, germline gDNA (extracted from saliva) was analyzed to establish the somatic nature of the mutations. In vitro assays were also performed to test the sensitivity of primary cells carrying identified molecular alterations to specific inhibitors, as detailed in the Online Supplemental Methods.

Validation cohort Candidate lesions validated in the discovery cohort were screened by Sanger sequencing of diagnostic gDNA in an independent cohort of 49 T-ALL patients. The validation cohort included refractory, relapsed and responsive cases enrolled in the multicenter GIMEMA LAL 2000 and LAL 0904 protocols (NCT00537550 and NCT00458848, respectively). Two age-cohorts were considered: adolescents and young adults (15-35 years, n=22) and adults (36-60 years, n=27). Overall, there were 36 males and 13 females with a median age of 37 years (range, 15-59). Molecular analysis detected the STILTAL1 fusion in six cases, MLLT10 fusions in three, NUP98-PSIP1 and NUP214-ABL1 in single cases, while the remaining 38 cases were negative (Online Supplementary Table S4). The experimental strategy used to study the discovery and validation cohorts is detailed in Online Supplementary Figure S2.

Statistical analyses The prognostic impact of the mutated genes was assessed in the validation cohort. The statistical methods are detailed in the Online Supplementary Methods.

Results Fusion transcript and gene expression findings in the discovery cohort RNAseq enabled an average 123x106 reads per sample, leading to an average coverage of 80X (Online Supplementary Table S5). We identified 183 predicted fusion transcripts (median 6/sample; range, 0-36) (Online Supplementary Table S6A) predominantly involving genes localized next to each other on the same chromosome, and likely representing read-through of transcription.15,20 For example, two fusion transcripts involving adjacent genes, namely SMG5-PAQR6 (n=2) and TTTY15-USP9Y (n=2), previously described in prostatic cancer21-23 and also identified in two cases of our validation cohort, were detected in normal thymus cells from healthy donors. haematologica | 2016; 101(8)

RNAseq in refractory/relapsed T-ALL

After ‘noise’ removal, several known and novel fusion transcripts were identified and further validated by Sanger sequencing of RT-PCR products (Table 1). Fusion transcripts were identified also in cases with normal or failed cytogenetics. In addition to the STIL-TAL1 (R24) and SETNUP214 fusions (R20, R21, R28), we identified and validated four fusions involving T-cell receptor genes (TCR). In three cases, the TCR was fused to known oncogenes TAL1 (R11), LMO2 (R19) or HOXA10-AS (R23) - and induced overexpression of the partner genes (Figure 1A and Online Supplementary Figure S3A). A novel rearrangement joining the TRAC to SOX8 on chromosome 16p13 was identified in the remaining case, also harboring STILTAL1 (R24). This was associated with the transcriptional activation of SOX8 (Figure 1B). The fusion was confirmed by FISH (Online Supplementary Figure S3B) and was also detected at relapse. Furthermore, a novel fusion juxtaposing HOXA11-AS to MIR181A1HG was documented in a patient showing overexpression of HOXA13, HOXA11 and HOXA10 (R27). Indeed, FISH confirmed a rearrangement between the HOXA cluster and chromosome region 1q31-1q32 (Online Supplementary Figure S3C).

We also identified out-of-frame fusions generated by deletions or inversions, predicted to cause inactivation of transcriptional regulators, i.e. ETV6-SLC15A5 (R27), GATA3-GS1-756B1.2 (R11) and WT1-THEM7P (R28), or inactivation of PTEN in two cases harboring the PTENFAS (R13) or MAST3-C19orf10 (R19) transcripts. In the patient harboring the MAST3-C19orf10 fusion, RNAseq also documented the GLT25D1-AC020911.1 out-offrame fusion generated by an inversion on chromosome 19p13. FISH analysis documented an amplification of the 19p13 region, and RNAseq data further indicated increased expression of NOTCH3 and JAK3, both localized in this region (Figure 1C and Online Supplementary Figure S4A-C). Paired diagnosis and relapse RNAseq analysis was performed in three cases. These cases retained the same fusions at the later stage of the disease, namely the SETNUP214 (R20 and R21) and the TRBC2-HOXA10-AS fusion (R23). Importantly, in all three cases the number of reads carrying the fusion transcripts was higher at relapse than in the diagnostic samples, suggesting clonal maintenance and expansion (Online Supplementary Figure S5).

Table 1. Summary of the lesions detected by RNAseq. The lesions listed below were validated by RT-PCR and Sanger sequencing (*), FISH (†) or Sanger sequencing on gDNA (§).

Karyotype

Fusion transcript

JAK/STAT lesion(§)

RAS/PTEN lesion(§)

NOTCH1/FBXW7 lesion(§)

R11

t(1;14)(p32;q13)

PTEN S226fsX23

46, xy [20]

TAL1-TRDC (*) GATA3-GS1-756B1.2 (*) -

R12

R13 R14

NA 46, xy [20]

PTEN-FAS(*) -

NRAS G12D KRAS G12V -

R15

47-48,XY,+8,9,der(11)t(9;11)(q13;p15), 19,-19,-22,+4mar[cp] 46, xy [20] 46, xy [20]

MAST3-C19orf10 (*;†) GLT25D1- AC020911.1(*;†)

LMO2-TRBC2 (*)

R20

46+,XY,t(7;11)(q33;p15)[8]/ 46,idem, del(6)(q13q21)[5] NA

R21 R22 R23

NA 46, xy [20] 46, xx [20]

R24

R25

R26 R27

48,XY,+2mar[3]/46,XY[22] 46, xx [15]

R28

R17 R18 R19

haematologica | 2016; 101(8)

JAK1 R724H PTPRC 1721-2A>G JAK3 R657Q STAT6 E185K

KRAS G12D

IL7R 243_245 LT>MICTL JAK1 604_606 delDYKinsRNDYN PTEN R233fsX7 JAK3 R657W PIK3R1 R642X -

NOTCH1 1693_1695 FQS>LG FBXW7 R689W NOTCH1 V1604E

FBXW7 R505C NOTCH1 F1606LinsIQ FBXW7 R465H -

SET-NUP214(*)

IL7R T244>ILCYPP

SET-NUP214(*) TRBC2-HOXA11-AS(*;†)

TYK2 R527Q JAK3 V674A STAT5A T628S -

ITPR1 A1205V PTEN M239fsX14

JAK1 R724H JAK3 R657Q JAK3 S1000L JAK3 M511I STAT5A G472S -

NOTCH1 1578del V NOTCH1 V2443fsX35

NOTCH1 L1574P NOTCH1 L1709P NOTCH2 F1167V NOTCH1 V1722M NOTCH1 Y2490X

TRAC-SOX8(*;†) STIL-TAL1(*) -

MIR181A1HG-HOXA11-AS(*;†) ETV6-SLC15A5(*) WT1-THEM7P(*) SET-NUP214(*)

ITPKB R590Q

NOTCH1 L1585P NOTCH1 V2473X NOTCH3 G438R - -

943

V. Gianfelici et al.

JAK/STAT and RAS/PTEN pathway mutations in the discovery cohort RNAseq analysis identified 1,527 protein-altering single nucleotide variants (median 78/sample; range, 17-157) and 1,115 INDEL (median 59/sample; range, 30-82) across the 19 patients with refractory/relapsed T-ALL (Online Supplementary Tables S6A and S7). Two hundred and twenty-nine genes were recurrently affected in at least two samples and involved specific pathways (Table 1, Online Supplementary Tables S8 and S9, Figure 2A). Overall, mutations in the JAK/STAT pathway were identified in nine of 19 cases (47%). Mutations were found in all the subtypes, including two out of five cases of early T-cell precursor leukemia. The most frequently mutated gene was JAK3 (n=5), followed by JAK1 (n=3), IL7R (n=2), STAT5A (n=2), STAT6 (n=1) and TYK2 (n=1). Interestingly, JAK1 and JAK3 mutant cases concomitantly harbored another lesion in the same pathway, while no additional lesions were detected in the IL7R and TYK2 mutant cases. Furthermore, we identified one case with abnormal splicing of PTPRC, encoding the tyrosine phosphatase CD45, a negative regulator of the JAK/STAT pathway.24 RT-PCR and Sanger sequencing confirmed that mRNA retained PTPRC introns 15 and 16, whereas sequencing of the gDNA documented the presence of a splice site mutation, as previously reported.25 The RAS/PTEN pathway was also recurrently affected

(8/19 cases, 42%). Cases carrying mutations in RAS or PI3K/AKT signaling were grouped together because of the biological convergence of the two pathways in mTOR activation.26 PTEN was affected by out-of-frame fusions leading to potential inactivation (n=2, see above) and frameshift mutations resulting in downstream premature stop codons (n=3). Furthermore, we identified mutations in K-RAS (n=2) or N-RAS (n=1) as well as in driver genes involved in other cancers but so far not reported in T-ALL, namely ITPKB (n=1), ITPR1 (n=1) and PIK3R1 (n=1). A double hit within the pathway was found in three cases. Concomitant JAK/STAT and RAS/PTEN mutations were observed in three cases (2 of them with a double hit in both pathways). In addition, altered expression of PTEN, PTPN11 and FLT3 was documented by gene expression analysis (Figure 2B). NOTCH1/FBXW7 was also recurrently mutated (9/19 cases, 47%). Six NOTCH1-mutated cases harbored either a double mutation (n=3, one of them also had a mutation in NOTCH3), or a mutation in another gene of the same pathway, namely FBXW7 (n=2) or NOTCH2 (n=1). Eight of the nine NOTCH1/FBXW7-positive cases harbored concomitant alterations in JAK/STAT or RAS/PTEN. RNAseq analysis on matched diagnostic and relapse samples revealed that NOTCH1 and FBXW7 status could differ between diagnosis and relapse, with the NOTCH1 mutation being lost (n=1) and the FBXW7 mutation being

944

Figure 1. Heatmap of selected genes: expression levels. (A) Expression levels of transcription factors known to be overexpressed in T-ALL. TAL1-overexpressing cases included sample R24, harboring the STIL-TAL1 fusion, and sample R11, carrying the TAL1-TCR rearrangement, whereas within LMO2-overexpressing cases we identified sample R19 harboring the LMO2-TCR fusion. HOXA-overexpressing cases included samples R20, R21, R28 harboring the SET-NUP214 fusion, sample R23 with the TRBC2-HOXA10-AS rearrangement and sample R27 carrying the MIR181A1HG-HOXA11-AS fusion. The majority of samples showed overexpression of MEF2C reflecting the immature phenotype. (B) The heatmap illustrates the expression patterns of SOX8, together with its immediately upstream and downstream flanking genes in the genome. It shows strong overexpression (red) of SOX8 in the R24 case harboring the TRAC-SOX8 fusion. (C) The heatmap shows overexpression of NOTCH3 and JAK3 in the R15 case harboring rearrangements and amplifications on 19p13. The heatmaps are plotted with the normalized log2 (count) values.

haematologica | 2016; 101(8)

RNAseq in refractory/relapsed T-ALL

acquired (n=1) at relapse. In contrast, mutations in JAK/STAT and RAS/PTEN pathways were retained at the later stage of the disease. Mutations in chromatin modifier and transcription regulators - including PHF6 (n=5), KDM3A (n=2), EZH2 (n=2), H3F3A (n=1), HDAC6 (n=1), MLL1 (n=1), MLL5 (n=1), CNOT3 (n=1), SF3B1 (n=1) and NCOR1 (n=1) were also identified. Finally, mutations in genes involved in purine synthesis or glucose metabolism (NT5DC1, GMPS, GMPR and DOK2) were found in single cases.

Incidence of JAK/STAT, RAS/PTEN and NOTCH1/FBXW7 mutations in the validation cohort To assess the incidence of mutations in the JAK/STAT, RAS/PTEN and NOTCH1/FBXW7 pathways, Sanger sequencing was performed in a total of 49 gDNA samples from additional, newly diagnosed T-ALL patients (i.e. the validation cohort) older than 15 years, enrolled in two consecutive, multicenter GIMEMA protocols and including responsive, refractory and relapsed cases. The results are summarized in Figure 3 and Online Supplementary Table S10. Eight patients (16%) harbored JAK/STAT pathway mutations. We identified missense point mutations affecting a residue of the pseudokinase and kinase domains of JAK1 (n=2), JAK3 (n=2) or both (n=1), the DNA binding and SH2 domain of STAT5B (n=2), and in-frame INDEL in exon 6 of IL7R, coding for the transmembrane domain (n=3). All STAT5B mutants harbored concomitant mutations in JAK3 (1 of them also had a JAK1 mutation), whereas the IL7R-positive patients had no additional mutations of the same pathway. These lesions were localized in hotspot residues already reported as drivers or pre-

dicted to have a damaging effect. In addition, the JAK3 P151R and JAK3 V722I were observed in three patients; however, given their distribution also in the normal population, they were not taken into account. No statistically significant differences were observed in white blood cell count, gender and age distribution between cases harboring the above-mentioned mutations and those not carrying them. The PICALM-MLLT10 fusion was identified in a single case, while the remaining cases were negative for recurrent fusion genes. Alterations in the RAS/PTEN pathway were found in ten patients (20%) including: five PTEN INDEL, four K/NRAS mutations affecting hotspot residues and one missense point mutation affecting the tyrosine kinase domain of FLT3, this last found in a case of early T-cell precursor leukemia. PTEN mutations were associated with younger age (P=0.04), whereas no significant differences were observed in gender and white blood cell count. Molecular analysis identified a STIL-TAL1 fusion in three cases harboring PTEN mutations and proved negative in the remaining seven cases. NOTCH1/FBXW7 mutations were found in 28 cases (57%), in 16 without additional alterations, while 12 carried a concomitant mutation in JAK/STAT (n=8) or K/NRAS (n=4). ITPKB, ITPR1, PIK3R1, STAT5A and STAT6 hotspots were screened: no additional positive cases were identified in this cohort.

Correlation between genetics, response to chemotherapy and outcome in the validation cohort Patients included in the validation cohort were grouped

Figure 2. Heatmap of recurrently targeted pathways in refractory/early relapsed T-ALL cases in the discovery cohort. (A) Heatmap of selected genes belonging to cellular pathways that are recurrently affected in refractory/early relapsed T-ALL cases. Validated protein-altering mutations and INDEL, alternative splicing events and fusions are shown. Blue boxes indicate missense mutations; red boxes INDEL; green boxes nonsense mutations; purple boxes fusion events resulting in inactivation of tumor suppressors; and, finally, the orange box indicates an alternative transcript event (ATE). (B) Heatmap of FLT3, PTEN and PTPN11 expression.

haematologica | 2016; 101(8)

945

V. Gianfelici et al.

into three subtypes according to their mutational profile JAK/STAT-positive, RAS/PTEN-positive and NOTCH1/ FBXW7-positive only (i.e. without JAK/STAT and/or RAS/PTEN mutations) - and evaluated for response to therapy and survival. Of the eight JAK/STAT-positive patients, five obtained a complete remission after induction whereas three did not because of refractoriness (n=2) or death (n=1). Notably, all patients but one relapsed shortly after achieving complete remission (median time: 10 months; range, 1-11). Eight of the ten RAS/PTEN-positive patients obtained a complete remission after induction, whereas two died during induction treatment. Seven of the eight responsive patients (87.5%, i.e. 4 K/N-RAS- and 3 PTEN-positive patients) experienced an early relapse (median time: 4 months; range, 1-11). Fifteen of the 16 patients with NOTCH1/FBXW7 mutations only were evaluable for response to therapy: 11 obtained a complete remission and four did not because of refractoriness (n=3) or death (n=1). Of the 11 patients who achieved complete remission, five (45.5%) have relapsed (median time from complete remission: 31 months; range, 6-41) and six (54.5%)

are in long-term complete remission at a median followup of 73 months (range, 51-96). While no significant difference in achievement of complete remission was found between the three subtypes, a significantly higher cumulative incidence of relapse was observed in JAK/STAT- and RAS/PTEN-positive cases when compared to negative patients or to cases harboring NOTCH1/FBXW7 mutations only (P=0.002, P=0.0001 and P=0.0015, respectively). Along this line, significantly shorter overall survival, disease-free survival and event-free survival were observed in patients harboring JAK/STAT mutations compared to patients without alterations in the pathways (Figure 4A-C). In detail, the overall, disease-free and eventfree survival probabilities were 0% at 20 months in patients carrying JAK/STAT mutations (median overall survival 15.7 months; median disease-free survival 11 months; median event-free survival 3.3 months) compared to 69.1% (overall survival, 95% CI: 88.2%-54.1%; median, not reached, P=0.0045), 70% (disease-free survival, 95% CI: 93.3%52.5%; median not reached, P=0.002) and 48.3% (eventfree survival, 95% CI: 70.4%-33.1%; 17.3 months, P=0.027) in wild-type patients.

Figure 3. Heatmap of selected targeted pathways in the validation cohort. Heatmap of selected genes belonging to cellular pathways in the validation cohort. Patients are divided according to response to therapy. Blue boxes indicate the presence of a lesion. ID: induction death; CCR: continuous complete remission; *Patient lost from follow-up.

946

haematologica | 2016; 101(8)

RNAseq in refractory/relapsed T-ALL

Similarly, a significantly shorter overall survival (15% at 20 months, 95% CI: 80.4%-2.8%, median 7.9 months, P=0.0032) and poorer disease-free survival (12.5% at 20 months, 95% CI: 78.2%-2%; median 4.3 months; P=0.0001) and event-free survival (10% at 20 months, 95% CI: 64.2%-1.6%; median 3.3 months; P=0.027) were observed in patients harboring RAS/PTEN alterations compared to wild-type patients (Figures 5A-C). Conversely, patients harboring NOTCH1/FBXW7 mutations only had significantly better overall, disease-free and event-free survivals compared to those of wild-type patients or to patients with concomitant mutations in JAK/STAT or RAS/PTEN (Figure 6A-C). In particular, the overall survival probability was 73.3% at 20 months in patients carrying NOTCH1/FBXW7 mutations alone (95% CI: 99.5%-54%; median not reached; P=0.0325) compared

to 50.1% (95% CI: 78.3%-32.1%; median 27.3 months) in wild-type patients and 0% (median 15.7 months) in patients with concomitant mutations in JAK/STAT or RAS/PTEN. Likewise, the disease-free and event-free survival probabilities were 81.8% (95% CI: 100%-61.9%; median not reached; P=0.0015) and 60% (95% CI: 90.7%39.7%; median 36.8 months; P=0.03724) at 20 months in patients carrying NOTCH1/FBXW7 mutations alone compared to 46.2% (95% CI: 83%-25.7%; median 14.5 months) and 30% (95% CI: 58.6%-15.4%; median 7.3 months) in wild-type patients and 0% (median 9.8 months) and 0% (median 6.3 months) in patients with concomitant mutations in JAK/STAT or RAS/PTEN. Thus, these latter findings indicate that the favorable impact of NOTCH1/FBXW7 was overuled by the concomitant presence of JAK/STAT or RAS/PTEN mutations.

A A

B B

Figure 4. Clinical relevance of JAK/STAT mutations in the T-ALL validation cohort. Kaplan-Meier estimates of (A) overall survival (OS), (B) disease-free survival (DFS) and (C) event-free survival (EFS) in the validation cohort, according to JAK/STAT mutation status. Significantly shorter OS, DFS and EFS were observed in JAK/STAT-positive patients than in JAK/STAT-negative patients.

haematologica | 2016; 101(8)

Figure 5. Clinical relevance of RAS/PTEN mutations in the T-ALL validation cohort. Kaplan-Meier estimates of (A) overall survival (OS), (B) disease-free survival (DFS) and (C) event-free survival (EFS) in the validation cohort, according to the RAS/PTEN mutational status. Significantly shorter OS, DFS and EFS were observed in RAS/PTEN-positive patients than in RAS/PTEN-negative patients.

947

V. Gianfelici et al.

Efficacy of target specific inhibition in refractory/relapsed cases The in vitro effects of specific inhibitors were evaluated on primary cells from T-ALL patients with targetable genetic lesions, as detailed in the Online Supplementary Results. These experiments documented a specific effect on cell proliferation and viability of ruxolitinib in three cases with JAK1 mutations; this effect was less pronounced in cases harboring more than one mutation in the pathway. We also tested the sensitivity of primary cells to crenolanib and quizartinib (FLT3/PDGFR inhibitors) and observed a decreased cell survival in the case overexpressing FLT3. Finally, decreases in proliferation and viability rate were observed in the single case with a PTEN frameshift mutation upon exposure to the PI3K/mTOR inhibitor BEZ235 or rapamycin.

that different lesions might cooperate in the acquisition of an aggressive phenotype. We observed a higher frequency of mutations resulting in aberrant JAK/STAT and RAS/PTEN signaling (47% and 42%, respectively). In contrast, the incidence of NOTCH1/FBXW7 mutations (47%) was lower in refractory/relapsed T-ALL than in the general unselected T-ALL population and these mutations were mostly found in combination with a second lesion in the pathway or different pathways, in line with their association with a favorable prognosis.30-33 Moreover, we observed that NOTCH1/FBXW7 status could differ between matched diagnostic and relapse samples, indicating that NOTCH1/FBXW7 mutations might be secondary events in T-ALL.

Discussion A more refined genetic characterization at diagnosis of adult T-ALL may optimize the prognostic stratification and may enable the design of more targeted anti-leukemic strategies. To shed light onto the genome of chemorefractoriness and to identify biological pathways responsible for and/or predictive of drug resistance, we used RNAseq to analyze a series of refractory/early relapsed cases of TALL, sampled at diagnosis. RNAseq is a powerful technique since it provides information on fusion genes, gene expression levels and point mutations, and it has been successfully applied to study T-ALL samples.15 In the present cohort of refractory/relapsed cases we identified a high number of lesions, strengthening the previous observation that the genetic complexity is correlated with an increased likelihood of drug resistance.9,27,28 Besides SET-NUP214 fusions, out-of-frame fusions resulting in deregulated expression or in inactivation of transcriptional regulators or tumor suppressor genes were detected, including four TCR fusions. A novel rearrangement joining TRAC to SOX8 on 16p13 was identified. This fusion was associated with transcriptional activation of SOX8 and points to SOX8 being a novel driver in T-cell leukemogenesis. RNAseq analysis also revealed other mechanisms involved in gene expression deregulation. In fact, a novel non-TCR translocation joining HOXA11-AS to MIR181A1HG on chromosome 1q32 was associated with overexpression of HOXA genes; interestingly, the 1q32 region was recently described to be rearranged with MYC in a case of T-ALL29 and might represent a novel region of chromosomal rearrangement holding actively transcribed sequences. In addition, RNAseq allowed the identification of complex intrachromosomal 19p13 rearrangements and amplifications producing an out-of-frame MAST3C19orf10 fusion, probably causing PTEN inactivation, and amplification and overexpression of NOTCH3 and JAK3, both located on 19p13. Importantly, most of the validated fusions were not detected by conventional cytogenetics but were later confirmed by FISH, thus corroborating the power of RNAseq to identify fusion transcripts even in cases with uninformative or normal kariotyping. Mutational analysis of refractory/relapsed T-ALL cases revealed the frequent co-occurrence of lesions affecting the same pathway and/or different pathways, suggesting 948

Figure 6. Clinical relevance of NOTCH1/FBXW7 mutation in the T-ALL validation cohort. (A) Kaplan-Meier estimates of overall survival (OS), (B) disease-free survival (DFS) and (C) event-free survival (EFS) in the validation cohort, according to the NOTCH1/FBXW7 mutation status. Significantly better OS, DFS and EFS were observed in patients harboring NOTCH1/FBXW7 mutations alone than in NOTCH1/FBXW7 negative patients or patients with concomitant mutations in K/N-RAS or JAK/STAT.

haematologica | 2016; 101(8)

RNAseq in refractory/relapsed T-ALL

Activating mutations of JAK1, JAK3 or IL7R have been reported in both B- and T-ALL, in up to 25% of cases.34-40 It has been shown that JAK3 and IL7R mutants promote cell transformation and tumor formation, and that the use of selective JAK inhibitors can reduce cell viability and tumor burden.36,41,42 Several studies have described an enrichment of mutations in genes mediating JAK/STAT signaling in early T-cell precursor leukemia, a subgroup of ALL associated with a poor response to standard chemotherapy.39,42 However, it is unclear whether JAK/STAT mutations affect T-ALL outcome. Contradictory results have been reported for JAK1 mutations in adults, probably due to different chemotherapeutic approaches.34,43 Zenatti et al. found no association between IL7R mutational status and clinical outcome in childhood.36 Conversely, Bandapalli et al. identified a high rate of STAT5B mutations in relapsed pediatric patients.44 In the present study, RNAseq analysis revealed a high incidence of mutations in the JAK/STAT pathway in refractory/relapsed cases, indicating that these lesions are a hallmark of very poor prognosis in T-ALL. The majority of mutations detected involve JAK3, in line with the observation that JAK3 mutations are drivers of T-ALL.41 All cases carrying JAK1 and JAK3 mutations harbored at least one other lesion in the same pathway or another pathway. Although conducted in a small number of cases, in vitro experiments showed that primary JAK1-mutated TALL cells may be susceptible to the anti-JAK1-2 inhibitor ruxolitinib. Moreover, cells with a JAK1 mutation alone were more sensitive than cases harboring additional mutations in the pathway, as observed in cell lines.45 As already noted for JAK/STAT, RAS/PTEN alterations are a common feature of T-ALL; their prognostic impact is still controversial in childood.46-49 In particular, in their study of children treated with the Berlin-FrankfurtMunster protocol, Bandapalli et al. reported that patients with PTEN and NOTCH1 mutations had a marked sensitivity to induction treatment and excellent long-term outcome, similar to that of patients with NOTCH1 mutations only and more favorable than that of patients with PTEN mutations only.48 The study of Jenkinson et al. on pediatric patients treated on the Medical Research Council UKALL2003 trial showed that neither PTEN nor RAS genotypes have a significant impact on response to therapy or long-term outcome; it was also shown that neither PTEN nor RAS genotypes affect the highly favorable outcome of patients with concomitant NOTCH1/FBXW7 mutations.49 Conversely, the GRAALL group reported that the favorable prognostic significance of NOTCH1/FBXW7 mutations was restricted to adult patients without RAS/PTEN abnormalities. In fact, K/N-RAS mutations and/or PTEN gene alterations have been associated with poor prognosis.33,50 In line with this, we identified a high rate of alterations involving the RAS/PTEN pathway in the discovery cohort of highly unfavorable cases. To evaluate the prognostic impact of these lesions, the

References 1. De Keersmaecker K, Marynen P, Cools J. Genetic insights in the pathogenesis of Tcell acute lymphoblastic leukemia. Haematologica. 2005;90(8):1116-1127.

haematologica | 2016; 101(8)

mutational analysis was extended to a validation cohort of 49 adolescent, young adult and adult patients enrolled in two consecutive GIMEMA protocols, in which the induction treatment was similar. The series included refractory, relapsed and responsive cases. The overall incidence of JAK/STAT and RAS/PTEN alterations within the validation cohort was 16% and 20%, respectively. In the validation cohort, JAK/STAT and RAS/PTEN alterations were again associated with a significantly higher risk of relapse (P=0.002 and P=0.0001, respectively) and inferior 20-month overall survival (P=0.0045 and P=0.0032, respectively), disease-free survival (P=0.002 and P=0.0001, respectively) and event-free survival (P=0.027 and P=0.027, respectively). In contrast, the presence of NOTCH1/FBXW7 mutations was associated with a better response to treatment and a reduced risk of relapse. Notably, the favorable impact of NOTCH1/FBXW7 mutations was lost in the presence of concomitant JAK/STAT or RAS/PTEN mutations, suggesting that mutations activating these pathways are important for T-ALL progression and prognosis, in agreement with the GRAAL report.50 Thus, the screening confirmed that the genetic profile is a predictor of outcome in T-ALL, at least when using standard chemotherapy regimens. Recently, no adverse effect on prognosis was observed in the UKALL2003 trial for T-ALL patients with JAK1, JAK3 or IL7R mutations:40 it must be underlined however, that the patients were treated with very intensive chemotherapy and very few relapses were observed in general. In conclusion, we provide a comprehensive overview of recurrent lesions present in cases of refractory/early relapsed T-ALL and have identified pathways with a prognostic role, which could be targeted by novel therapeutic agents. We also describe a high rate of JAK3 mutations in refractory/relapsed T-ALL cases, as well as a frequent, often concomitant, deregulation of the JAK/STAT, RAS/PTEN and NOTCH1/FBXW7 pathways; the presence of additional JAK/STAT/RAS/PTEN mutations appears to be noxious. Thus, accurate genetic characterization of T-ALL at diagnosis is recommended in order to define individual patients’ risk optimally and to identify patients who might benefit from more intensive treatment or combined targeted approaches. Preliminary in vitro experiments suggest that the use of specific inhibitors might be clinically valuable depending on the underlying lesions and that a suboptimal response might be sustained by the presence of a second hit. Finally, the marked overlap of mutations suggests that combinations of specific inhibitors targeting different pathways might prove useful for these patients and prospective studies are currently underway to address this issue. Acknowledgments This work was supported by Associazione Italiana per la Ricerca sul Cancro (AIRC) Special Program Molecular Clinical Oncology, 5 x 1000, Milan, Italy (MCO-10007); Fondo per gli Investimenti della Ricerca di Base (FIRB), Rome, Italy.

2. Chiaretti S, Gianfelici V, Ceglie G, Foà R. Genomic characterization of acute leukemias. Med Princ Pract. 2014;23(6):487506. 3. Silverman LB, Gelber RD, Dalton VK, et al. Improved outcome for children with acute lymphoblastic leukemia: results of Dana-

Farber Consortium Protocol 91-01. Blood. 2001;97(5):1211-1218. 4. Pui CH, Sandlund JT, Pei D, et al. Improved outcome for children with acute lymphoblastic leukemia: results of Total Therapy Study XIIIB at St Jude Children’s Research Hospital. Blood. 2004;104(9):2690-2696.

949

V. Gianfelici et al. 5. Rowe JM, Buck G, Burnett AK, et al. Induction therapy for adults with acute lymphoblastic leukemia: results of more than 1500 patients from the international ALL trial: MRC UKALLXII/ECOG E2993. Blood. 2005;106(12):3760-3767. 6. Vitale A, Guarini A, Ariola C, et al. Adult Tcell acute lymphoblastic leukemia: biologic profile at presentation and correlation with response to induction treatment in patients enrolled in the GIMEMALAL 0496 protocol. Blood. 2006;107(2):473-479. 7. Bassan R, Spinelli O, Oldani E, et al. Improved risk classification for risk-specific therapy based on the molecular study of minimal residual disease (MRD) in adult acute lymphoblastic leukemia (ALL). Blood. 2009;113(18):4153-4162. 8. Fielding AK, Richards SM, Chopra R, et al. Medical Research Council of the United Kingdom Adult ALL Working Party; Eastern Cooperative Oncology Group. Outcome of 609 adults after relapse of acute lymphoblastic leukemia (ALL); an MRC UKALL12/ECOG 2993 study. Blood. 2007;109(3):944-950. 9. Mullighan CG, Phillips LA, Su X, et al. Genomic analysis of the clonal origins of relapsed acute lymphoblastic leukemia. Science. 2008;322(5906):1377-1380. 10. Chiaretti S, Brugnoletti F, Tavolaro S, et al. TP53 mutations are frequent in adult acute lymphoblastic leukemia cases negative for recurrent fusion genes and correlate with poor response to induction therapy. Haematologica. 2013;98(5):e59-61. 11. Ma X, Edmonson M, Yergeau D, et al. Rise and fall of subclones from diagnosis to relapse in pediatric B-acute lymphoblastic leukaemia. Nat Commun. 2015;6:6604. 12. Messina M, Chiaretti S, Wang J, et al. Prognostic and therapeutic role of targetable lesions in B-lineage acute lymphoblastic leukemia without recurrent fusion genes. Oncotarget. 2016;7(12):13886-13901. 13. Tzoneva G, Perez-Garcia A, Carpenter Z, et al. Activating mutations in the NT5C2 nucleotidase gene drive chemotherapy resistance in relapsed ALL. Nat Med. 2013;19(3):368-371. 14. Kunz JB, Rausch T, Bandapalli OR, et al. Pediatric T-cell lymphoblastic leukemia evolves into relapse by clonal selection, acquisition of mutations and promoter hypomethylation. Haematologica. 2015;100 (11):1442-1450. 15. Atak ZK, Gianfelici V, Hulselmans G, et al. Comprehensive analysis of transcriptome variation uncovers known and novel driver events in T-cell acute lymphoblastic leukemia. PLoS Genet. 2013;9(12):e1003997. 16. Vardiman JW, Thiele J, Arber DA, et al. The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia: rationale and important changes. Blood. 2009;114(5):937-951. 17. 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. 18. 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 lym-

950

19.

20.

21.

22.

23.

24.

25.

26.

27. 28.

29.

30.

31.

32.

33.

34.

35.

phoblastic leukemia. Haematologica. 2003; 88(3):275-279. Brandimarte L, La Starza R, Gianfelici V, et al. DDX3X-MLLT10 fusion in adults with NOTCH1 positive T-cell acute lymphoblastic leukemia. Haematologica. 2014;99(5):64-66. Halvardson J, Zaghlool A, Feuk L. Exome RNA sequencing reveals rare and novel alternative transcripts. Nucleic Acids Res. 2013;41(1):e6. Kannan K, Wang L, Wang J, et al. Recurrent chimeric RNAs enriched in human prostate cancer identified by deep sequencing. Proc Natl Acad Sci USA. 2011;108(22):9172-9177. Ren S, Peng Z, Mao JH, et al. RNA-seq analysis of prostate cancer in the Chinese population identifies recurrent gene fusions, cancer-associated long noncoding RNAs and aberrant alternative splicing. Cell Res. 2012;22(5):806-821. Ren G, Zhang Y, Mao X, et al. Transcriptionmediated chimeric RNAs in prostate cancer: time to revisit old hypothesis? OMICS. 2014;18(10):615-624. Porcu M, Kleppe M, Gianfelici V, et al. Mutation of the receptor tyrosine phosphatase PTPRC (CD45) in T-cell acute lymphoblastic leukemia. Blood. 2012;119(19): 4476-4479. Raponi S, Gianfelici V, Chiaretti S, et al. CD45 antigen negativity in T-lineage ALL correlates with PTPRC mutation and sensitivity to a selective JAK inhibitor. Br J Haematol. 2015;171(5):884-887. Chappell WH, Steelman LS, Long JM, et al. Ras/Raf/MEK/ERK and PI3K/PTEN/ Akt/mTOR inhibitors: rationale and importance to inhibiting these pathways in human health. Oncotarget. 2011;2(3):135-164. Almendro V, Marusyk A, Polyak K. Cellular heterogeneity and molecular evolution in cancer. Annu Rev Pathol. 2013;8:277-302. Messina M, Del Giudice I, Khiabanian H, et al. Genetic lesions associated with chronic lymphocytic leukemia chemo-refractoriness. Blood. 2014;123(15):2378-2388. La Starza R, Borga C, Barba G, et al. Genetic profile of T-cell acute lymphoblastic leukemias with MYC-translocations. Blood. 2014;124(24):3577-3582. Weng AP, Ferrando AA, Lee W, et al. Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia. Science. 2004;306(5694):269-271. Asnafi V, Buzyn A, Le Noir S, et al. NOTCH1/FBXW7 mutation identifies a large subgroup with favorable outcome in adult T-cell acute lymphoblastic leukemia (T-ALL): a Group for Research on Adult Acute Lymphoblastic Leukemia (GRAALL) study. Blood. 2009;113(17):3918-3924. van Vlierberghe P, Ambesi-Impiombato A, De Keersmaecker K, et al. Prognostic relevance of integrated genetic profiling in adult T-cell acute lymphoblastic leukemia. Blood. 2013;122(1):74-82. Beldjord K, Chevret S, Asnafi V, et al. Oncogenetics and minimal residual disease are independent outcome predictors in adult patients with acute lymphoblastic leukemia. Blood. 2014;123(24):3739-3749. Flex E, Petrangeli V, Stella L, et al. Somatically acquired JAK1 mutations in adult acute lymphoblastic leukemia. J Exp Med. 2008;205(4):751-758. Shochat C, Tal N, Bandapalli OR, et al. Gainof function mutations in interleukin-7 recep-

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

tor-a (IL7R) in childhood acute lymphoblastic leukemias. J Exp Med. 2011;208(5):901-908. Zenatti PP, Ribeiro D, Li W, et al. Oncogenic IL7R gain-of-function mutations in childhood T-cell acute lymphoblastic leukaemia. Nat Genet. 2011;43(10):932-939. Bains T, Heinrich MC, Loriaux MM, et al. Newly described activating JAK3 mutations in T-cell acute lymphoblastic leukemia. Leukemia. 2012;26(9):2144-2146. Kalender Atak Z, de Keersmaecker K, Gianfelici V, et al. High accuracy mutation detection in leukemia on a selected panel of cancer genes. PLoS ONE. 2012;7(6): e38463. Zhang J, Ding L, Holmfeldt L, et al. The genetic basis of early T-cell precursor acute lymphoblastic leukaemia. Nature. 2012;481 (7380):157-163. Vicente C, Schwab C, Broux M, et al. Targeted sequencing identifies association between IL7R-JAK mutations and epigenetic modulators in T-cell acute lymphoblastic leukemia. Haematologica. 2015;100(10): 1301-10. Degryse S, de Bock CE, Cox L, et al. JAK3 mutants transform hematopoietic cells through JAK1 activation causing T-cell acute lymphoblastic leukemia in a bone marrow transplant mouse model. Blood. 2014;124 (20):3092-3100. Maude SL, Dolai S, Delgado-Martin C, et al. Efficacy of JAK/STAT pathway inhibition in murine xenograft models of early T-cell precursor (ETP) acute lymphoblastic leukemia. Blood. 2015;125(11):1759-1767. Asnafi V, Le Noir S, Lhermitte L, et al. JAK1 mutations are not frequent events in adult TALL: a GRAALL study. Br J Haematol. 2010;148(1):178-179. Bandapalli OR, Schuessele S, Kunz JB, et al. The activating STAT5B N642H mutation is a common abnormality in pediatric T-cell acute lymphoblastic leukemia and confers a higher risk of relapse. Haematologica. 2014;99(10):e188-192. Springuel L, Hornakova T, Losdyck E, et al. Cooperating JAK1 and JAK3 mutants increase resistance to JAK inhibitors. Blood. 2014;124(26):3924-3931. Gutierrez A, Sanda T, Grebliunaite R, et al. High frequency of PTEN, PI3K, and AKT abnormalities in T-cell acute lymphoblasticleukemia. Blood. 2009;114(3):647-65. Zuurbier L, Petricoin EF 3rd, Vuerhard MJ, et al. The significance of PTEN and AKT aberrations in pediatric T-cell acute lymphoblastic leukemia. Haematologica. 2012;97(9): 1405-1413. Bandapalli OR, Zimmermann M, Kox C, et al. NOTCH1 activation clinically antagonizes the unfavorable effect of PTEN inactivation in BFM-treated children with precursor T-cell acute lymphoblastic leukemia. Haematologica. 2013;98(6):928-936. Jenkinson S, Kirkwood AA, Goulden N, Vora A, Linch DC, Gale RE. Impact of PTEN abnormalities on outcome in pediatric patients with T-cell acute lymphoblastic leukemia treated on the MRC UKALL2003 trial. Leukemia. 2016;30(1):39-47. Trinquand A, Tanguy-Schmidt A, Ben Abdelali R, et al. Toward a NOTCH1/ FBXW7/RAS/PTEN-based oncogenetic risk classification of adult T-cell acute lymphoblastic leukemia: a Group for Research in Adult Acute Lymphoblastic Leukemia Study. J Clin Oncol. 2013;31(34): 4333-4342.

haematologica | 2016; 101(8)

ARTICLE

Acute Lymphoblastic Leukemia

Deletions of the long arm of chromosome 5 define subgroups of T-cell acute lymphoblastic leukemia

EUROPEAN HEMATOLOGY ASSOCIATION

Ferrata Storti Foundation

Roberta La Starza,1 Gianluca Barba,1 Sofie Demeyer,2,3 Valentina Pierini,1 Danika Di Giacomo,1 Valentina Gianfelici,4 Claire Schwab,5 Caterina Matteucci,1 Carmen Vicente,2,3 Jan Cools,2,3 Monica Messina,4 Barbara Crescenzi,1 Sabina Chiaretti,4 Robin Foà,4 Giuseppe Basso,6 Christine J. Harrison,5 and Cristina Mecucci1

Molecular Medicine Laboratory, Center for Hemato-Oncology Research, University of Perugia, Italy; 2Center for Human Genetics, KU Leuven, Belgium; 3Center for the Biology of Disease, VIB, Leuven, Belgium; 4Hematology, Department of Cellular Biotechnologies and Hematology, “Sapienza” University, Rome, Italy; 5Leukaemia Research Cytogenetic Group, Northern Institute for Cancer Research, Newcastle University, Newcastle-uponTyne, UK; and 6Pediatric Hemato-Oncology, Department of Pediatrics “Salus Pueri”, University of Padova, Italy 1

Haematologica 2016 Volume 101(8):951-958

ABSTRACT

ecurrent deletions of the long arm of chromosome 5 were detected in 23/200 cases of T-cell acute lymphoblastic leukemia. Genomic studies identified two types of deletions: interstitial and terminal. Interstitial 5q deletions, found in five cases, were present in both adults and children with a female predominance (chi-square, P=0.012). Interestingly, these cases resembled immature/early T-cell precursor acute lymphoblastic leukemia showing significant down-regulation of five out of the ten top differentially expressed genes in this leukemia group, including TCF7 which maps within the 5q31 common deleted region. Mutations of genes known to be associated with immature/early T-cell precursor acute lymphoblastic leukemia, i.e. WT1, ETV6, JAK1, JAK3, and RUNX1, were present, while CDKN2A/B deletions/mutations were never detected. All patients had relapsed/resistant disease and blasts showed an early differentiation arrest with expression of myeloid markers. Terminal 5q deletions, found in 18 of patients, were more prevalent in adults (chi-square, P=0.010) and defined a subgroup of HOXA-positive T-cell acute lymphoblastic leukemia characterized by 130 up- and 197 down-regulated genes. Down-regulated genes included TRIM41, ZFP62, MAPK9, MGAT1, and CNOT6, all mapping within the 1.4 Mb common deleted region at 5q35.3. Of interest, besides CNOT6 down-regulation, these cases also showed low BTG1 expression and a high incidence of CNOT3 mutations, suggesting that the CCR4-NOT complex plays a crucial role in the pathogenesis of HOXA-positive T-cell acute lymphoblastic leukemia with terminal 5q deletions. In conclusion, interstitial and terminal 5q deletions are recurrent genomic losses identifying distinct subtypes of T-cell acute lymphoblastic leukemia.

Introduction Deletion of the long arm of chromosome 5, del(5q), is the most frequent genomic loss in myeloid diseases.1 Two distinct common deleted regions (CDR) were identified in myelodysplastic syndromes and acute myeloid leukemia. Del(5q), as the sole cytogenetic abnormality, occurs in 10-15% of myelodysplastic syndromes and is known as “the 5q- syndrome”.1 It is characterized by a 1.5 Mb CDR (5q32q33) where the putative oncosuppressor RPS14 is mapped.2 Del(5q) associated with other cytogenetic changes, often within a complex karyotype, is prevalent in haematologica | 2016; 101(8)

Correspondence: cristina.mecucci@unipg.it

Received: February 5, 2016 Accepted: April 29, 2016 Pre-published: May 5, 2016. doi:10.3324/haematol.2016.143875

Check the online version for the most updated information on this article, online supplements, and information on authorship & disclosures: www.haematologica.org/content/101/8/951

951

R. La Starza et al.

acute myeloid leukemia and in high-risk myelodysplastic syndromes.3 Here it is characterized by a ~1 Mb CDR (5q31) and haploinsufficiency of the oncosuppressor EGR1.4 Conversely, del(5q) has been rarely reported in Band T- acute lymphoblastic leukemia (ALL).5-8 T-lineage ALL (T-ALL), a heterogeneous group of leukemias, is characterized by co-occurrence of multiple genetic lesions.9 Interestingly, cooperative genetic defects have been described in T-ALL, suggesting that perturbation of specific cell processes are needed for the development of overt leukemia.10,11 Translocations causing TAL/LMO, TLX1, TLX3, MYB, MEF2C, NKX2-1/2, or HOXA over-expression define distinct gene expression signatures and are known as “type A” abnormalities.6,12-14 These rearrangements co-occur with multiple mutations and imbalances, named “type B” abnormalities, which activate oncogenic signaling cascades, including JAK/STAT, PI3K/AKT, and RAS/MEK/ERK. The most prevalent type B abnormalities are loss-of-function mutations and genomic losses, indicating that tumor suppressor genes, such as CDKN2A/B, PHF6, LEF1, PTEN, WT1, ETV6, and PTPN2, play pivotal roles in the initiation of TALL.9,15 To provide insights into the prevalence and specific features of del(5q), in T-ALL, we screened a series of 200 TALL cases by fluorescence in situ hybridization (FISH). We found two distinct types of recurrent 5q deletions: inter-

stitial (I-5q) and terminal (T-5q). I-5q was identified in 2.5% of cases with a genomic profile closely resembling immature/early T-cell precursor (ETP) ALL. These findings indicate that I-5q is a recurrent cytogenetic event in T-ALL with early differentiation arrest of blasts. T-5q, identified in 9% of cases, clustered within the HOXA category and was characterized by a high incidence of abnormalities involving the CCR4-NOT complex and its closely related transcription factor BTG1. These observations indicate that HOXA over-expression, haplo-insufficiency of genes at T-5q, and deregulation of the CCR4NOT complex are characteristic of this subgroup of leukemia.

Methods We investigated the incidence of del(5q) in a cohort of 66 adult (≥18 years) and 134 pediatric T-ALL patients from the Italian (GIMEMA LAL-0496 and LAL-0904 and AIEOP LLA-2000) and UK clinical trials (MRC, ALL2003 and ALL97) (Online Supplementary Table S1, Online Supplementary Information).10,16,17 All patients or their parents/guardians gave informed consent to sample collection and molecular analyses, in agreement with the Declaration of Helsinki. The study was approved by the local bio-ethical committee (research project 2014-025). Combined interphase FISH identified T-ALL-associated

Table 1. Clinical, hematologic and molecular-cytogenetic features of 5 cases of T-ALL with interstitial 5q deletion.

Sex/Age Immunophenotype

Karyotype

ETP-ALL Genetic related mutations category

CI-FISH

CNV

SET-NUP214[80%] del(5)(q31q33) [87%] del(6)(q15q21) [82%] del WT1 [85%] HOXA-translocation [60%] del(5)(q14q31) [55%] del ETV6/CDKN1B [59%]

LOSS: CREBBP/16p13.3, JAKMIP2/5q32, SYNCRIP/6q14.3

F/9

undefined: CD34+, CD33+

n.a.

DNM2, WT1, LPHN2

HOXA

F/16

ETP

46,XX[15]

RUNX1, JAK3, ETV6, PHF6

HOXA

F/27

pro-T

n.a.

WT1, PHF6

HOXA

SET-NUP214[93%] del(5)(q31) [99%] del(6)(q21) [98%] del WT1 [85%] del RB1 [96%] del ATM [99%] del TP53 [99%]

F/49

pro-T

46,XX[15]

JAK3,JAK1, PHF6

MEF2C

M/37

ETP

RUNX1-translocation [82%] del(5)(q31) [84%] gain DDX3X [46%] del(5)(q31)[93%] del(6)(q16q21)[93%] del(9)(q32/TAL2[97%] del RB1[99%] del(20)(p11)[92%] trisomy 21[95%]

n.a. 48,XY,del(5) (q31q33),+8, del(9q),der(12),del(13q), +21[10]

unclassified

Follow-up (months) 64†

LOSS: 5q14.3-q32,12p13.2-p12.3, 63† 14q32.11-q32.12,16q22.2-q24.3 GAIN: 1q32.1-q44,19q13.2-q13.4 CN-LOH: 6q12-q27 LOSS: 2p14-p13.3,3q26.31-q26.33,5q11.112† q12.2,5q21.1-q22.1,5q23.2,5q31.1q31.3,5q31.3-q32,6q21,6q22.31,6q22.33,7p12.2p11.2,9q34.11-q34.13,11p14.1-p13,11p11.2p11.12,11q14.1-q23.3,12q14.1-q23.3,13q13.3q14.3,13q21.33-q31.1,14q12,16q21-q22.1,17p13.2p13.1,19q13.32 CN-LOH: 1p36.33-p36.11 LOSS: 5q31.1 7† GAIN: Xp22.33-p11.23 CN-LOH: 9q33.2-q34.3 LOSS: 1q25.1, 2p13.2, 3q13.1-q13.3, 5q21.3-33.2, 6q15-q22.3, 9q22.3-q33.2, 11q14.2-q23.2, 12q13.1, 13q13.3-q21.3, 16p13.3, 20p11.2 GAIN: 21q11.2-q22.3

1†

N.: patient number; F: female; M: male; CI-FISH: combined interphase fluorescence in situ hybridization (the percentage of cells with abnormal CI-FISH is indicated between brackets); CNV: copy number variations were determined by haloplex analysis (case n.1) or single nucleotide polymorphism array (cases n. 2-5); n.a.: not available; ETP: early T-cell precursor; †dead. Only mutations typically associated with immature/ETP-ALL were reported. In patient n. 1 the panel of antibodies tested did not allow to stage of differentiation of leukemic blasts to be determined. Patients n. 2 and 3 were studied by transcriptome sequencing and gene expression profiling.

952

haematologica | 2016; 101(8)

Deletion of chromosome 5 long arm in T-ALL

genomic rearrangements which classified cases into defined subgroups (Online Supplementary Table S2).10,16,17 Deletions of 5q were detected by FISH with LSI EGR1/D5S23, D5S721 Dual Colour probe, RP11-182E4/RP11-453D13 and CTB-31E20/RP11266N12 for rearrangements of TLX3, RP11-117L6 for NPM1 abnormalities, and RP1-240G13 for deletions of the subtelomeric 5q region. Cases with del(5q) were further characterized with clones for 5p13-qter (Online Supplementary Table S3). Single nucleotide polymorphism array, denaturing high performance liquid chromatography, Sanger sequencing, haloplex polymerase chain reaction,11 transcriptome sequencing,18 and gene expression profiling were performed in cases with available material (Online Supplementary Information).

Results and Discussion Incidence and distribution of del(5q) in T-cell acute lymphoblastic leukemia FISH and single nucleotide polymorphism array revealed two types of recurrent del(5q) in 23/200 T-ALL patients: I-5q (5 cases) (Table 1) and T-5q (18 cases) (Table 2). In 11/18 cases with T-5q, the deletion was very large and included the CDR of I-5q cases (Figure 1A). Overall, del(5q) was mainly associated with the HOXA category (16/23; 69.5%). It was found at diagnosis in 22 cases while in the remaining case it was detected only during

Table 2. Clinical, hematologic, and molecular-cytogenetic features of 18 cases of T-ALL with terminal 5q deletion

Sex/Age Immunophenotype

Karyotype

Gene mutations

Genetic category

n.a.

NOTCH1, CNOT3

HOXA

CI-FISH

CNV

Follow-up (months)

TCRB-HOXA[65%] del(5)(q35)[96%] del CDKN2A/B[95%] HOXA-translocation[80%] del(5)(q31)[94%]

LOSS: 5q34-q35.3, 9p21.3 GAIN: 13q21-q34 CN-LOH: 9p24.3-p21.3, 9p21.3-p21.1 LOSS: 5q14.3-q35.3 GAIN: 1q32.1-q42.13, 1q42.3-q44

CHILDREN 1

F/11

pre-T

F/9

pre-T

M/10

n.a.

M/18

ETP

n.a.

F/4

n.a.

46,XX[20]

M/17

cortical

F/7

cortical

46,XX,del(5) NOTCH1 (q31q35), inv(7)(p15q?)[8] 46,XX[6] n.a. NOTCH1, ETV6

46,XY,del(6)(q25)[5] NOTCH1, 46,XY[19] NRAS

Sex/Age Immunophenotype

n.a.

NOTCH1

Karyotype

Gene mutations

HOXA

NUP98-RAP1GDS1[90%] del(5)(q31)[98%] TCRB-rearrangement[80%] del HOXA[99%] del EZH2[99%] HOXA SET-NUP214[96%] del(5)(q31)[96%] UNCLASSIFIED del(5)(q35)[50%] del FBXW7[13%] del CDKN2AB[50%] UNCLASSIFIED del(5)(q31)[90%] gain 11p13-15[79%] del NF1[79%] UNCLASSIFIED del(5)(q31)[42%] del(6q15q16)[20%] del HOXA[26%] gain MYB[23%] gain MYC[46%] del CDKN2AB[45%] gain RB1[12%]

Genetic category

LOSS: 5q23.2-q35.3, 7q22.2-q31.1, 7q34-q36.3 GAIN: 7p14.1, 7q22.1-q22.2, 7q31.1-q34

LOSS: 5q31-q35.3, 16q11.2-q24.2, 9q34.11-q34.13 24† n.a.

LOSS: 1p36.33-p36.22, 1q43, 5q31.1-q35.3, 17q11.2, 17q11.2 GAIN: 7q21.11, 11p15.5-p11.2 LOSS: 2q34-q37.3, 5q15-q35.3, 6q12-q16.3, 7p22.2-p11.2, 8p23.2-p12, 9p21.3 GAIN: Xp22.33-q28, 2q32.3-q34, 6q21-q27, 7q11.21-q11.22, 8q21.3-q24.3, 15q11.2-q26.3

9†

CI-FISH

CNV

TCRB-HOXA[24%] del(5)(q31)[27%] del(6)(q15-21)[27%] del ETV6[53%] del TP53[20%] del NF1[32%] TCRB-HOXA[80%] del(5)(q14)[83%] del(6)(q15q16)[85%] del CDKN2AB[56%] trisomy 8[67%]

LOSS: 4q32.3, 5q21.3-q35.3, 6q13-q22.1, 7q34, 12p13.31-p11.22, 13q21.33, 17p13.3-q12, 17q22 GAIN: 18p11.32

10†

Follow-up (months)

ADULTS 8

M/22

ETP

M/32

pre-T

46,XY,del(6)(q15q16), NOTCH1, del(12)(p13)[1] NRAS, 46,XY[10] PIK3R1

46,XY[15]

NOTCH1

HOXA

21†

LOSS: 1p36.3-36.2, 4q26,5q11.1-q35.3, 10† 6p22.3-22.2,6p12.1, 6q14.1-16.1, 9p21.3,9q21.3-21.3, 9q32-q33.2,9q33.3-q34.1,19p13.2 GAIN: 5p15.3-q11.1,7p21.1-p15.2,7q11.2-q21.1, 7q22.1-22.3,7q34,7q36.1,8p23.3-q24.3 CN-LOH: 7p22.3-p21.1,7p15.2-p11.1,7q22.1-q22.3,7q22.2-q34 continued on the next page

haematologica | 2016; 101(8)

953

R. La Starza et al.

disease progression (13 months after diagnosis); it belonged to the major abnormal clone in all cases but one. Cytogenetically, both types of del(5q) were always associated with additional chromosome abnormalities and had significantly more DNA copy number abnormalities than cases without del(5q) [median 7 (range, 3-28) versus 3.5 (1-14), P=0.003] (Online Supplementary Table S4).

Deletions of 5q define two independent subgroups of Tcell acute lymphoblastic leukemia We have data to show that I-5q and T-5q are cytogenetic markers of two subgroups of T-ALL, with different age and gender distributions and distinct genomic backgrounds.

Interstitial 5q T-cell acute lymphoblastic leukemia I-5q was detected in five patients (2.5%) (Table 1). Four of these five patients were females (Pearson c2 test, P=0.012) but there was no distinctive age distribution. Cytogenetically, three cases carried HOXA-activating rearrangements (2 carried SET-NUP214 and 1 carried TCRB-HOXA); patient n. 4 had a translocation involving RUNX1, suggesting likely membership of the MEF2C category;14 patient n. 5 was unclassified (Online Supplementary Information and Online Supplementary Figure S1). The cases of I-5q T-ALL had a higher incidence of genomic losses than cases of T-ALL without del(5q) [mean 9.5 (range, 1-22) versus 3.1 (1-11)], a difference which did not, however, reach statistical significance likely due to the small sample size (Online Supplementary

continued from the previous page

M/51

pre-T

46,XY[15]

NOTCH1

HOXA

TCRB-HOXA[91%] del(5)(q31)[97%] del CDKN2AB[90%] gain BCL11B[80%]

M/37

pre-T

n.a.

PTEN

HOXA

CALM-MLLT10[98%] del(5)(q31)[96%] del(6)(q15q16)[92%] del CDKN2AB[92%] gain MYC[90%]

M/26

cortical

HOXA

DDX3X-MLLT10[85%] del(5)(q35)[90%] del(6)(q15-21)[77%] del CDKN2AB[87%]

M/25

HOXA

NUP98-RAP1GDS1[90%] del(5)(q31)[80%] MYB tandem dup[26%]

LOSS: 5q23.2-q35.3, 16p13.3-p13.1 GAIN: Xp22.33-p11.3, 5p15.3-p13.3, 6q23.3

NUP98-translocation [16%] del(5)(q13) [17%] del TP53 [20%] SQSTM1-NUP214[85%] del CDKN2AB[81%] del WT[85%] del NF1[85%]

LOSS: 1p36.33-p36.23, 5q11.1-q35.3, 17p13.3-p11.1 GAIN: 1q32.3-q44 LOSS: 5q35.3, 9p23-p22.3, 9p21.3, 9q32-q33.3, 10p12.1-p11.2, 11p13, 13q13.1, 13q13.2-13.3, 13q32.2-q32.3, 13q32.3-q33.1, 13q33.1-q33.2, 16q24.1-q24.3, 17q11.2, 21q22.2 GAIN: 7q31.3, 9q34.1-q34.3 LOSS: 5q14.3-q35.3, 6q11.1-q22.33, 10q23.2-q23.31 GAIN: 5p15.33, 6q23.3, 13q31.1-q34

F/19

M/20

46,Y,t(X;10)(p12;p13), NOTCH1, add(1)(p36), CNOT3 del(9)(p11p24)[10] 46,idem,del(6q15)[1] 46,XY[11] cortical 46,XY,del(1)(q42), NOTCH1 t(4;11)(q21;p15), del(5)(q31q35),del(7)(p21), add(16)(p13)[10] pre-T 45,XX,t(4;11) TP53 (q13;p15),der(5;17) (p10;q10)[15] pre-T 46,XY[15]

F/39

mature

46,XX[10]

PTEN, CNOT3

HOXA

M/22

cortical

n.a.

NOTCH1, KRAS

TLX3

M/25

n.a.

46,XY,add(2)(q36) NOTCH1 [5]/46,XY[18]

HOXA

TLX3

MLL-translocation[60%] del(5)(q35)[17%] del(6)(q16q21)[60%] MYB tandem dup[59%] del PTEN[50%] TLX3-translocation[92%] del(5)(q35)[92%] del CDKN2AB[96%] del LEF1[38%] gain ETV6[97%] TLX3-translocation[98%] del(5)(q35)[98%] del BCL11B[98%]

LOSS: 5q21.1, 5q21.3-q35.3, 9p24.3-p21.1, 26† 15q25.2-26.3 GAIN: 9p21.1-p13.2, 14q31.3-q32.3 CN-LOH: 1q31.2-q32.2, 2q22.1-q31.3, 3q23-q26.3, 6q22.3-q24.3, 10q24.3-q26.2 LOSS: 2p16.1, 2q22.1, 2q24.2-q24.3, 2q31.1, 1† 2q31.3, 2q32.1-q32.2, 2q33.3-q34, 2q34-q35, 2q35-q37.3, 3p24.3, 3p12.2-p12.1, 5q22.3-q35.3, p25.1-p24.3, 6p24.3-p24.2, 6p22.3, 6p22.2-p21.33, 6 6p21.2-p21.1, 6p21.1-q21,9p24.1-p23, 9p21.3, 9q31.1, 11q14.1-q23.3, 21q11.2-q22.3 GAIN: 5p15.33-q22.3, 6q21, 8q21.12-q24.3, 19p13.3-q13.43 CN-LOH: 6q21-q27 n.a. 84

16†

n.a.

LOSS: 5q35.1-q35.3 GAIN: 14q32.2-q32.3 CN-LOH: 4q11-q35.1

38†

N.: patient number; CI-FISH: combined interphase fluorescence in situ hybridization (the percentage of cells with abnormal CI-FISH is indicated between brackets); CNV: copy number variation (detected by single nucleotide polymorphism array); n.a.: not available; nl: normal; del: deletion; †: dead. Patient n.3 was studied by haloplex; n. 12, 16, and 17 by transcriptome sequencing; n. 5 was not studied for gene mutations. Patients n. 4 and 15 were wild type for hotspot mutations of NOTCH1, FBXW7, FLT3, TP53, CNOT3, and K/NRAS. Gene expression profiling was performed in patients n. 4, 6, 12, 15-18. In patient n. 8 the T-5q was detected 13 months after diagnosis, while the patient was still receiving treatment.

954

haematologica | 2016; 101(8)

Deletion of chromosome 5 long arm in T-ALL

Figure 1. Genomic characteristics of I-5q and T-5q T-ALL. (A) Schematic representation of 5q deletions in our 23 T-ALL cases. Patientsâ&#x20AC;&#x2122; numbers refer to Table 1 for cases with I-5q, and to Table 2 for cases with T-5q. Black boxes indicate monoallelic deletion; gray boxes, diploidy; white boxes, not tested; gain, presence of three copies; (B) TCF7 and NR3C1 expression in two T-ALL with I5q vs. control [7 cases without del(5q) and 3 cases with T-5q]; (C) Distribution of terminal CDR (tCDR) (black) (3 cases) and interstitial CDR (iCDR) (gray) (3 cases with I-5q and 9 cases with large T5q including the I-5q CDR) within the HOXA-category: all five cases with an immature/ETP phenotype as well as five of seven cases of early T-ALL had I-5q or large T-5q including the iCDR; (D) Expression of CD34, MEF2C, and LMO2 in two cases with I-5q vs. ten controls [7 cases without del(5q) and 3 cases with T-5q]; (E) Supervised gene expression profiling analysis of four HOXA positive cases with T-5q (t) and six without (n), identified 327 differentially expressed genes.

haematologica | 2016; 101(8)

955

R. La Starza et al.

Table S4). They were significantly associated with WT1 deletions (n=2) (c2, p=0.002) and del(6q) (n=3) (c2, P=0.002); these del(6q) shared a CDR at band 6q21, encompassing the putative onco-suppressor genes SEC63 and FOXO3 (Online Supplementary Table S5).7,19 All I-5q had a CDR at band 5q31 (Figure 1A), encompassing the 1 Mb CDR of high-risk myelodysplastic syndrome/acute myeloid leukemia del(5q).3,4,20 When we compared I-5q- positive cases with T-5q and T-ALL without del(5q), we found that the expression levels of IRF1, EGR1, CTNNA1, HNRNPAO, TIFAB, and CXXC5, known putative onco-suppressors in in vitro and/or in vivo models20-24 mapping within the CDR, did not differ between the three groups; while the presence of I-5q was associated with significant down-regulation of NR3C1 and TCF7 genes (Figure 1B, Online Supplementary Information and Online Supplementary Figures S2 and S3). NR3C1 belongs to the nuclear hormone receptor superfamily that includes the mineral-corticoid and estrogen receptors. Once NR3C1 binds to steroid hormones, it acts as a direct transcriptional regulator.25 NR3C1 deletion has been associated with relapse in pediatric and adult B-cell ALL and predicted corticosteroid resistance in T-ALL, thus influencing response to treatment.26,27 TCF7 is an essential transcriptional regulator of T-cell specification, commitment, and lineage determination.28 In mouse models, Tcf7-/- induced a T-cell malignancy that was similar to human ETP-ALL as 117 deregulated genes were common to both.29 In vivo studies suggested that TCF7 haplo-insufficiency rendered pre-malignant thymocytes susceptible to later lesions which subsequently transformed them into leukemic blasts. In fact, low Tcf7 expression predisposed murine thymocytes to acquire Notch1-activating mutations which were invariably found in Tcf7-/- lymphomas.29,30 Interestingly, NOTCH1 mutations were present in four out of five cases of T-ALL with I-5q. Together these data suggest strong similarities between mouse Tcf7-/- T-cell malignancies and human T-ALL with I-5q assigning a role to TCF7 haplo-insufficiency in this subgroup of leukemia. Finally, among HOXA-positive T-ALL, ten of 12 cases with immature/ETP or early phenotype lost the I-5q CDR (Online Supplementary Table S6 and Figure 1C). Among them five cases had the typical features of the high-risk subgroup recently named HOXA-positive ETP ALL.31,32 As far as we know, loss of genes at the I-5q CDR is the first recurrent cytogenetic change so far described in this subgroup.

Terminal 5q T-cell acute lymphoblastic leukemia T-5q T-ALL was detected in 18 patients (9%), who were mainly adults (c2, P=0.010) but there was no association with gender (12 males; 6 females), stage of blast differentiation (7 pre-T, 5 cortical, 2 ETP, 1 mature, and 3 undefined), or white blood cell count (Table 2). Cytogenetically, 13 cases belonged to the HOXA category [with rearrangements of: HOXA (n=5), NUP98 (n=3), NUP214 (n=2), MLLT10 (n=2), and MLL (n=1)], two cases to the TLX3 category (Online Supplementary Information and Online Supplementary Figure S4), while three cases remained unclassified. T-5q ALL were significantly associated with del(6)(q14q15) (c2, P=0.002), and genomic gains (P=0.0038) (Online Supplementary Table S4), of which the most frequent was gain of chromosome 5p arm, found in four out of 17 (23.5%) fully characterized cases. 956

We also found a high incidence of NF1 deletions (4 cases)(c2, P=0.002) and recurrent N/KRAS mutations (3 cases), consistent with RAS/MEK pathway involvement in ~23% of cases (Online Supplementary Information).33-35 Although T-5q deletions varied greatly in size, they all had a common 1.4 Mb CDR (chr5:179257527-180719789, GRCh37) telomeric of SQSTM1 (Figure 1A). The T-5q CDR contained one LIN gene (long intergenic non-protein coding), five olfactory receptor genes, three microRNA, eight LOC non-coding RNA, and 37 genes. Supervised gene expression profiling analysis showed that only eight out of the 26 genes with probe-sets available at the CDR, i.e. MAPK9, TBC1D9B, RFP130, TRIM52, TRIM52-AS1, HEIH, ZFP62, and CNOT6, were significantly down-regulated in six cases with T-5q T-ALL compared to 22 TALL without.

Genetic profile links interstitial 5q with immature/early T-cell precursor acute lymphoblastic leukemia ETP-ALL, a distinct subgroup of T-ALL, is defined by a typical immunophenotype which is negative for CD1a and CD8, negative or weakly positive for CD5, and positive for at least one of the following markers: CD34, CD117, HLA-DR, CD13, CD33, CD11b, and CD65.36 ETP-ALL also shows a distinct genetic profile with high expression of two bHLH transcription factors, LYL1 and LMO2, a high incidence of mutations typically associated with the pathogenesis of acute myeloid leukemia, and a low frequency of typical T-ALL lesions, such as CDKN2A/B deletions and NOTCH1 mutations.37 ETPALL has been associated with a dismal outcome due to poor response to chemotherapy and a high rate of resistance/early relapse, namely in cases with genomic rearrangements associated with HOXA deregulation.31,32,38 On the other hand, a recent clinical trial demonstrated a high rate of continuous complete remission in children.39 In our study, all I-5q T-ALL were characterized by early differentiation arrest of leukemic blasts with expression of at least one stem cell/myeloid antigen (Table 1). In fact patients n. 2 and n. 5 satisfied all diagnostic criteria for ETP-ALL.36 RB1/13q14 deletions co-occurred in two of five cases. A critical analysis of previous studies showed that del(5q) had already been found in immature/ETP ALL.6,34 Indeed, del(5q) together with del(13q) were the two cytogenetic changes most frequently associated with immature/ETP ALL as they were both detected in 23% of cases (4/17), and co-occurred in 11% (2/17).6,36 Additional evidence that I-5q T-ALL are closely related to the immature/ETP subtype of T-ALL came from identification of PHF6, JAK3, JAK1, DNM2, WT1, ETV6, and/or RUNX1 mutations and lack of CDKN2AB deletion/mutation in all analyzed cases.37,40,41 It is noteworthy that in addition to TCF7, other significantly down-regulated genes in I-5q T-ALL were TDRKH, PCGF5, HDAC4, and MTA3, so that our I-5q patients carried five out ten of the most differentially expressed (down-regulated) genes seen in human ETP-ALL (Online Supplementary Table S7 and Online Supplementary Figures S2, S3, S5-8).29,36 Moreover, among the 22 genes which have been reported to be significantly over-expressed in ETP-ALL,6,14,36,37,40-42 MEF2C, LMO2, and CD34 were significantly up-regulated in I-5q (Figure 1D). Overall, our data highlight two informative aspects of I-5q. First, genomic profiles link I-5q and immature/ETP ALL; furthermore all haematologica | 2016; 101(8)

Deletion of chromosome 5 long arm in T-ALL

five cases with I-5q were poor responders to standard therapy. In fact, patient n. 1 was a late remitter despite being assigned to the most intensive arm of the ALL2003 MRC trial. She received a bone marrow transplant from an unrelated donor in second remission, relapsed shortly afterwards and died. Patient n. 2 had an early relapse and also received a bone marrow transplant from an unrelated donor while she was in second remission. She had a second relapse after the transplant and died of her disease. The other three patients had resistant disease and died within 1 year. Although there were too few patients in the present series to draw any definitive conclusions, I-5q marks a particularly high-risk subgroup of immature/ETP ALL for which alternative targeted therapies should be developed. Among them, JAK/STAT inhibitors, which are highly effective in ETP-ALL xenograft models,43 and the BCL2 inhibitor ABT-737, which restores the sensitivity to steroids in cell lines with high MEF2C expression,44,45 might be considered in the treatment of these refractory leukemias.

Terminal 5q is a HOXA-positive cytogenetic subgroup T-5q was found in 27% of patients belonging to the HOXA group (13/48 cases) (c2; P<0.001) in which it behaved as a type B event (Table 2). Supervised gene expression profiling analysis compared four HOXA-positive cases with T-5q and six without to determine whether T-5q defined specific pathways within the HOXA category: t-test analysis (Pâ&#x2030;¤0.005) identified 327 genes (Figure 1E). Of the 130 over-expressed genes in T5q cases, functional annotation analysis revealed enrichment of genes involved in nuclear lumen, DNA replication and mRNA metabolic processes, such as CDC45, CDC5L, CHEK1, E2F3, and FANCD2, suggesting specific deregulation of these pathways. Among the down-regulated genes, FYN and LCK tyrosine kinases, IL7R, ZAP70, ADD3, and the adaptor protein, BTG1, are known oncogenes/tumor suppressors in ALL. Within the T-5q CDR we observed down-regulation of TRIM41, ZFP62, MAPK9, MGAT1, and interestingly, CNOT6. This is the first report of CNOT6 involvement in human cancer and its down-regulation is consistent with it being a putative onco-suppressor gene as indicated by in vitro data.46 Besides CNOT6 down-regulation, our T-5q TALL cases were associated with a high rate of CNOT3 mutations (18%). Notably, CNOT3 loss-of-function muta-

References 1. Van Den Berghe H, Michaux L. 5q-, twentyfive years later: a synopsis. Cancer Genet Cytogenet. 1997;94(1):1-7. 2. Boultwood J, Pellagatti A, McKenzie ANJ, Wainscoat JS. Advances in the 5q- syndrome. Blood. 2010;116(26):5803-5811. 3. Lai F, Godley LA, Joslin J, et al. Transcript map and comparative analysis of the 1.5Mb commonly deleted segment of human 5q31 in malignant myeloid diseases with a del(5q). Genomics. 2001;71(2):235-245. 4. Joslin JM, Fernald AA, Tennant TR, et al. Haploinsufficiency of EGR1, a candidate gene in the del(5q), leads to the development

haematologica | 2016; 101(8)

tions were found in ~7% of adult T-ALL but no specific association with any major molecular subgroups has been reported.47 Both CNOT6 and CNOT3 encode for members of the CCR4-NOT complex, which consists of two major modules: the deadenylase module composed of two subunits with deadenylation enzymatic activity (CNOT6 or CNOT6L and CNOT7 or CNOT8) and the NOT module (CNOT1, CNOT2, and CNOT3). The CCR4-NOT complex is involved in chromatin modification, cellular response to DNA-damage, transcription elongation, RNA export, nuclear RNA surveillance, and miRNA-mediated deadenylation of mRNA.48-50 Although, CCR4-NOT involvement in human tumors has been rarely reported, a recurrent hotspot P131L mutation of RQCD1 (formerly known as CNOT9) has recently been identified in ~4% of cutaneous melanomas.51 Our findings of a high rate of deletion/loss-of-function mutations and down-regulation of members of the CCR4-NOT complex as well as low expression of BTG1, a directly interacting adaptor protein of CCR4-NOT, suggest that this complex plays a role in the subset of HOXA-positive T-ALL with T-5q. In conclusion, the present study has identified distinct recurrent 5q deletions in T-ALL, corresponding to different genomic landscapes and defining two cytogenetic subgroups. I-5q identified a subgroup of immature T-ALL, found predominantly in females, with an ETP-like genetic profile and poor response to current treatments. T-5q designated a subgroup of HOXA-positive T-ALL, mainly found in adults and associated with a high rate of CCR4NOT complex abnormalities. In I-5q, two putative oncosuppressors, NR3C1 and TCF7, mapping to the 5q31 CDR, were down-regulated. In T-5q CNOT6, a member of the CCR4-NOT complex, mapping to the 5q35 CDR, was significantly down-regulated. Due to the heterogeneity of treatment and age of our 18 patients, the clinical impact of T-5q could not be established, while the unresponsive T-ALL with I-5q should be considered for new experimental therapies. Acknowledgments Associazione Italiana per la Ricerca sul Cancro (AIRC, IG15525), Fondo per gli Investimenti della Ricerca di Base (FIRB 2011 RBAP11TF7Z_005), Programmi di Ricerca scientifica di rilevante Interesse Nazionale (PRIN Cod. 2010NYKNS7_003), Associazione Sergio Luciani, Fabriano, Italy, and AIRC 5x1000; Bloodwise, formerly Leukaemia and Lymphoma Research.

of myeloid disorders. Blood. 2007;110(2): 719-726. 5. Faderl S, Gidel C, Kantarjian HM, Manshouri T, Keating M, Albitar M. Loss of heterozygosity on chromosome 5 in adults with acute lymphoblastic leukemia. Leuk Res. 2001;25(1):39-43. 6. Ferrando AA, Neuberg DS, Staunton J, et al. Gene expression signatures define novel oncogenic pathways in T cell acute lymphoblastic leukemia. Cancer Cell. 2002;1(1):75-87. 7. Grossmann V, Haferlach C, Weissmann S, et al. The molecular profile of adult T-cell acute lymphoblastic leukemia: mutations in RUNX1 and DNMT3A are associated with poor prognosis in T-ALL. Genes

Chromosomes Cancer. 2013;52(4):410-422. 8. Van Vlierberghe P, Ambesi-Impiombato A, De Keersmaecker K, et al. Prognostic relevance of integrated genetic profiling in adult T-cell acute lymphoblastic leukemia. Blood. 2013;122(1):74-82. 9. Durinck K, Goossens S, Peirs S, et al. Novel biological insights in T-cell acute lymphoblastic leukemia. Exp Hematol. 2015;43(8):625-639. 10. La Starza R, Borga C, Barba G, et al. Genetic profile of T-cell acute lymphoblastic leukemias with MYC translocation. Blood. 2014;124(24):3577-3582. 11. Vicente C, Schwab C, Broux M, et al. Targeted sequencing identifies association between IL7R/JAK mutations and epigenetic

957

R. La Starza et al.

12.

13.

14.

15. 16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

958

modulators in T-cell acute lymphoblastic leukemia. Haematologica. 2015;100(11): 1373-1375. Soulier J, Clappier E, Cayuela JM, et al. HOXA genes are included in genetic and biologic networks defining human acute Tcell leukemia (T-ALL). Blood. 2005;106(1): 274-286. Clappier E, Cuccuini W, Kalota A, et al. The C-MYB locus is involved in chromosomal translocation and genomic duplications in human T-cell acute leukemia (T-ALL), the translocation defining a new T-ALL subtype in very young children. Blood. 2007;110 (4):1251-1261. 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. Van Vlieberghe P, Ferrando A. The molecular basis of T cell acute lymphoblastic leukemia. J Clin Invest. 2012;122(10):3398-3406. Gorello P, La Starza R, Varasano E, et al. Combined interphase fluorescence in situ hybridization elucidates T-cell acute lymphoblastic leukemia genetic heterogeneity in adults. Haematologica. 2010;95(1):79-86. La Starza R, Lettieri A, Pierini V, et al. Linking genomic lesions with minimal residual disease improves prognostic stratification in children with T-cell acute lymphoblastic leukaemia. Leuk Res. 2013;37 (8):928-935. Atak ZK,Gianfelici V, Hulselmans G, et al. Comprehensive analisis of transcriptome variation uncovers known and novel driver events in T-cell acute lymphoblastic leukemia. PLoS Genet. 2013;9(12):e1003997. Karube K, Nacagawa M, Tsuzuki S, et al. Identification of FOXO3 and PRDM1 as tumor-suppressor gene candidates in NKcell neoplasms by genomic and functional analyses. Blood. 2011;118(12):3195-3204. Varney ME, Niederkorn M, Konno H, et al. Loss of Tifab, a del(5q) MDS gene, alters hematopoiesis through derepression of Tolllike receptor-TRAF6 signaling. J Exp Med. 2015;212(11):1967-1985. Testa U Stellacci E, Pelosi E, et al. Impaired myelopiesis in mice devoid of interferon regulatory factor 1. Leukemia. 2004; 18(11): 1864-1871. Liu TX, Becker MW, Jelinek J, et al. Chromosome 5q deletion and epigenetic suppression of the gene encoding alphacatenin (CTNNA1) in myeloid cell transformation. Nat Med. 2007;13(1):78-83. Young DJ, Stoddart A, Nakitandwe J, et al. Knockdown of Hnrnpa0, a del(5q) gene, alters myeloid cell fate in murine cells through regulation of AU-rich transcripts. Haematologica. 2014;99(6):1032-1040. Kühnl A, Valk PJ, Sanders MA, et al. Downregulation of the Wnt inhibitor CXXC5 predicts a better prognosis in acute myeloid leukemia. Blood. 2015;125(19): 2985-2994. Bray PJ, Cotton RG. Variations of the human

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

glucocorticoid receptor gene (NR3C1): pathological and in vitro mutations and polymorphisms. Hum Mutat. 2003;21(6): 557-568. Mullighan CG, Phillips LA, Su X, et al. Genomic analysis of the clonal origins of relapsed acute lymphoblastic leukemia. Science. 2008;322(5906):1377-1380. Kuster L, Grausenburger R, Fuka G, et al. ETV6/RUNX1-positive relapses evolve from an ancestral clone and frequently acquire deletions of genes implicated in glucocorticoid signalling. Blood. 2011;117(9):26582667. Weber BN, Chi AW, Chavez A, et a. A critical role for TCF-1 in T-lineage specification and differentiation. Science. 2011;476(7358): 63-68. Yu S, Zhou X, Steinke FC, et al. The TCF-1 and LEF-1 Transcription factors have cooperative and opposing roles in T cell development and malignancy. Immunity. 2012;13 (5):813-826. Tiemessen MM, Baert MR, Schonewille T, et al. The nuclear effector of Wnt-signaling, Tcf1, functions as a T-cell-specific tumor suppressor for development of lymphomas. PLoS Biol. 2012;10(11):e1001430. Matlawska-Wasowska K, Kang H, Devidas M, et al. Mixed lineage leukemia rearrangements (MLL-R) are determinants of high risk disease in homeobox A (HOXA)-deregulated T-lineage acute lymphoblastic leukemia: a Children’s Oncology Group Study. Blood (ASH Annual Meeting Abstracts). 2015; 126(23):694. Bond J, Marchand T, Touzart A, et al. An early thymic precursor phenotype predicts outcome exclusively in HOXA-overexpressing adult T-ALL: a GRAALL study. Blood (ASH Annual Meeting Abstracts). 2015;126 (23):808. Jenkinson S, Kirkwood AA, Goulden N, Vora A, Linch DC, Gale RE. Impact of PTEN abnormalities on outcome in pediatric patients with T-cell acute lymphoblastic leukemia treated on the MRC UKALL2003 trial. Leukemia. 2016;30(1):39-47. Beldjord K, Chevret S, Asnafi V, et al. Oncogenetics and minimal residual disease are independent outcome predictors in adult patients with acute lymphoblastic leukemia Blood. 2014;123(24):3739-3749. Trinquand A, Tanguy-Schmidt A, Ben Abdelali R, et al. Toward a NOTCH1/FBXW7/RAS/PTEN-based oncogenetic risk classification of adult T-cell acute lymphoblastic leukemia: a Group for Research in Adult Acute Lymphoblastic Leukemia study. J Clin Oncol. 2013;31 (34):4333-4342. Coustan-Smith E, Mullighan CG, Onciu MM, et al. Early T-cell precursor leukemia: a subtype of very risk acute lymphoblastic leukemia identified in two independent cohorts. Lancet Oncol. 2009;10(2):147-156. Zhang J, Ding L, Holmfeldt L, et al. The genetic basis of early T-cell precursor acute lymphoblastic leukemia. Nature. 2012;481

(7380):157-163. 38. Jain N, Lamb AE, O'Brien S, et al. Early T-cell precursor acute lymphoblastic leukemia/ lymphoma (ETP-ALL/LBL) in adolescents and adults: a high-risk subtype. Blood. 2016;127(15):1863-1869. 39. Conter V, Valsecchi MG, Buldini B, et al. Favourable outcome for children with early T-cell precursor acute lymphoblastic leukemia treated in AIEOP centers on the AIEOP-BFM ALL 2009 contemporary protocol. Lancet Hematol. 2016 Jan 25 [Epub ahead of print]. 40. Neumann M, Heesch S, Gökbuget N, et al. Clinical and molecular characterization of early T-cell precursor leukemia: a high-risk subgroup in adult T-ALL with a high frequency of FLT3 mutations. Blood Cancer J. 2012;2(1):e55. 41. Neumann M, Heesch S, Schlee C, et al. Whole exome sequencing in adult ETP-ALL reveals a high rate of DNMT3A mutations. Blood. 2013;121(23):4749-4752. 42. Goossens S, Radaelli E, Blanchet O, et al. ZEB2 drives immature T-cell lymphoblastic leukaemia development via enhanced tumour-initiating potential and IL-7 receptor signalling. Nat Commun. 2015;6:5794. 43. Maude SL, Dolai S, Delgado-Martin C, et al. Efficacy of JAK/STAT pathway inhibition in murine xenograft models of early T-cell precursor (ETP) acute lymphoblastic leukemia. Blood. 2015;125(11):1759-1767. 44. Chonghaile TN, Roderick JE, Glenfield C, et al. Maturation stage of T-cell acute lymphoblastic leukemia determines BCL-2 versus BCL-XL dependence and sensitivity to ABT-199. Cancer Discov. 2014;4(9): 1074-1087. 45. Kawashima-Goto S, Imamura T, Tomoyasu C, et al. BCL2 Inhibitor (ABT-737): a restorer of prednisolone sensitivity in early T-cell precursor-acute lymphoblastic leukemia with high MEF2C expression? PLoS One. 2015;10(7):e0132926. 46. Sanchez-Perez I, Manguan-Garcia C, Menacho-Marquez M, Murguía JR, Perona R. hCCR4/cNOT6 targets DNA-damage response proteins. Cancer Lett. 2009;273(2): 281-291. 47. De Keersmaeker K, Atak ZK, Li N, et al. Exome sequencing identifies mutation in CNOT3 and ribosomal genes RPL5 and RPL10 in T-cell acute lymphoblastic leukemia. Nat Genet. 2013;45(2):186-190. 48. Collart M, Panasenko OO. The Ccr4-Not complex. Gene. 2012;492(1):42-53. 49. Miller JE, Reese JC. Ccr4-Not complex: the control freak of eukaryotic cells. Crit Rev Biochem Mol Biol. 2012;47(4):315-333. 50. Wahle E, Winkler GS. RNA decay machines: deadenylation by the Ccr4-not and Pan2Pan3 complexes. Biochim Biophys Acta. 2013;1829(6-7):561-570. 51. Wong SQ, Behren A, Mar VJ, et al. Whole exome sequencing identifies a recurrent RQCD1 P131L mutation in cutaneous melanoma. Oncotarget. 2015;6(2):11151127.

haematologica | 2016; 101(8)

ARTICLE

Chronic Lymphocytic Leukemia

Different spectra of recurrent gene mutations in subsets of chronic lymphocytic leukemia harboring stereotyped B-cell receptors

Lesley-Ann Sutton,1 Emma Young, 1 Panagiotis Baliakas, 1 Anastasia Hadzidimitriou,2 Theodoros Moysiadis,2 Karla Plevova,3 Davide Rossi,4 Jana Kminkova,3 Evangelia Stalika,2 Lone Bredo Pedersen,5 Jitka Malcikova,3 Andreas Agathangelidis,6,7 Zadie Davis,8 Larry Mansouri,1 Lydia Scarfò,6,7 Myriam Boudjoghra,9 Alba Navarro,10 Alice F. Muggen,11 Xiao-Jie Yan,12 Florence Nguyen-Khac,9 Marta Larrayoz,13 Panagiotis Panagiotidis,14 Nicholas Chiorazzi,12 Carsten Utoft Niemann,5 Chrysoula Belessi,15 Elias Campo,10 Jonathan C. Strefford,13 Anton W. Langerak,11 David Oscier,8 Gianluca Gaidano,4 Sarka Pospisilova,3 Frederic Davi,9 Paolo Ghia,6,7 Kostas Stamatopoulos,1,2,16* Richard Rosenquist,1* and on behalf of ERIC, the European Research Initiative on CLL

Department of Immunology, Genetics and Pathology, Science for Life Laboratory, Uppsala University, Uppsala, Sweden; 2Institute of Applied Biosciences, CERTH, Thessaloniki, Greece; 3Central European Institute of Technology, Masaryk University and University Hospital Brno, Czech Republic; 4Division of Haematology, Department of Translational Medicine, Amedeo Avogadro University of Eastern Piedmont, Novara, Italy; 5 Department of Hematology, Rigshospitalet, Copenhagen, Denmark; 6Università VitaSalute San Raffaele, Milan, Italy; 7Division of Experimental Oncology and Department of Onco-Hematology, IRCCS, San Raffaele Scientific Institute, Milan, Italy; 8Department of Haematology, Royal Bournemouth Hospital, Bournemouth, UK; 9Hematology Department and University Pierre et Marie Curie, Hopital Pitie-Salpetriere, Paris, France; 10 Hematopathology Unit and Department of Hematology, Hospital Clinic, University of Barcelona, Institut d’Investigacions Biomèdiques August Pi iSunyer (IDIBAPS), Barcelona, Spain; 11Department of Immunology, Erasmus MC, University Medical Center Rotterdam, The Netherlands; 12The Feinstein Institute for Medical Research, North Shore-Long Island Jewish Health System, Manhasset, New York, NY, USA; 13Cancer Sciences, Faculty of Medicine, University of Southampton, UK; 14First Department of Propaedeutic Medicine, University of Athens, Greece; 15Hematology Department, Nikea General Hospital, Piraeus, Greece; and 16Hematology Department and HCT Unit, G. Papanicolaou Hospital, Thessaloniki, Greece

EUROPEAN HEMATOLOGY ASSOCIATION

Ferrata Storti Foundation

Haematologica 2016 Volume 101(8):959-967

*KS and RR contributed equally to this work.

Correspondence: richard.rosenquist@igp.uu.se

ABSTRACT

e report on markedly different frequencies of genetic lesions within subsets of chronic lymphocytic leukemia patients carrying mutated or unmutated stereotyped B-cell receptor immunoglobulins in the largest cohort (n=565) studied for this purpose. By combining data on recurrent gene mutations (BIRC3, MYD88, NOTCH1, SF3B1 and TP53) and cytogenetic aberrations, we reveal a subset-biased acquisition of gene mutations. More specifically, the frequency of NOTCH1 mutations was found to be enriched in subsets expressing unmutated immunoglobulin genes, i.e. #1, #6, #8 and #59 (22-34%), often in association with trisomy 12, and was significantly different (P<0.001) to the frequency observed in subset #2 (4%, aggressive disease, variable somatic hypermutation status) and subset #4 (1%, indolent disease, mutated immunoglobulin genes). Interestingly, subsets harboring a high frequency of NOTCH1 mutations were found to carry few (if any) SF3B1 mutations. This starkly contrasts with subsets #2 and #3 where, despite their immunogenetic differences, SF3B1 mutations occurred in 45% and 46% of cases, respectively. In addition, mutations within TP53, whilst enriched in subset #1 (16%), were rare in subsets #2 and #8 (both 2%), despite all being clinically aggressive. All subsets were negative for MYD88 mutations, whereas BIRC3 mutations were infrequent. Collectively, this striking bias and skewed distribution of mutations and cytogenetic aberrations within specific chronic lymphocytic leukemia subsets implies that the mechanisms underlying clinical aggressiveness are not uniform, but rather support the existence of distinct genetic pathways of clonal evolution governed by a particular stereotyped B-cell receptor selecting a certain molecular lesion(s). haematologica | 2016; 101(8)

Received: January 2, 2016. Accepted: May 12, 2016. Pre-published: May 19, 2016. doi:10.3324/haematol.2016.141812

Check the online version for the most updated information on this article, online supplements, and information on authorship & disclosures: www.haematologica.org/content/101/8/959

959

L.-A. Sutton et al.

Introduction Immunogenetic studies have been instrumental in revealing that the ontogeny of chronic lymphocytic leukemia (CLL) is not stochastic, but rather antigen-driven, through the discovery that: (i) the immunoglobulin (IG) gene repertoire of the clonotypic B-cell receptor (BcR) displays restriction and, (ii) the level of somatic hypermutations (SHM) present in rearranged IG heavy chain genes defines two disease subtypes, each associated with a different clinical course.1-5 Such studies led to the discovery of quasi-identical or stereotyped BcR IGs in more than 30% of CLL patients who can be assigned to distinct subsets, each defined by a particular BcR immunogenetic motif.6-14 Importantly, from both a biological and clinical perspective, evidence suggests that this classification of CLL based on BcR stereotypy is highly relevant and extends well beyond the SHM status of the BcR IG, thereby enabling the identification of homogeneous disease subgroups and, hence, overcoming the heterogeneity characteristic of CLL. Indeed, studies indicate that patients with similar SHM status but assigned to different stereotyped subsets can exhibit distinct, subset-biased biological profiles and clinical behavior.10,15-25 In addition, preliminary observations in CLL, in relatively small patient series, suggest that the frequency and patterns of mutations within several genes, namely, NOTCH1, SF3B1 and TP53, may differ amongst subsets of patients carrying stereotyped BcRs, the paradigmatic example being the recently observed enrichment of SF3B1 mutations in the clinically aggressive subset #2.26-28 With this in mind, we sought to systematically evaluate the mutational status of BIRC3, MYD88, NOTCH1, SF3B1 and TP53 in 565 CLL patients assigned to one of 10 major stereotyped subsets, and representing cases with varying SHM status, i.e. cases harboring either unmutated IGHV genes (U-CLL) or mutated IGHV genes (M-CLL). We demonstrate markedly different frequencies and spectra of genomic defects amongst the various subsets. On these grounds, we speculate that common genetic aberra-

tions, acquired and/or selected in the context of shared immune pathways originating from highly similar BcR IGs could shape the evolutionary pathway of individual CLL subsets.

Methods Patients A total of 565 CLL patients, selected based on the expression of stereotyped BcR IGs leading to their assignment to a major subset,10,14 were included in this study (Table 1). A minimum requirement was that data be available for at least 10 cases/subsets to enable meaningful comparisons; this criterion resulted in 10 major subsets being evaluated. All cases were diagnosed according to the 2008 IWCLL criteria.29 Informed consent was collected according to the Declaration of Helsinki, and ethical approval was granted by local review committees.

Cytogenetic and SNP-array studies Interphase fluorescence in situ hybridization (FISH) for the 13q14, 13q34, 11q22, 17p13 chromosomal regions and the centromere of chromosome 12 was performed as previously described.30 For 30 cases recurrent genomic aberration data was obtained using the Affymetrix 250K SNP Array.31

Sequence analysis of IGHV–IGHD–IGHJ rearrangements PCR amplification, sequence analysis and interpretation of IGHV-IGHD-IGHJ rearrangements were performed following established international guidelines and using the IMGT® database and the IMGT/V-QUEST tool, as previously reported.2,7,8,10 Clonotypic IGHV gene sequences were defined as either mutated or unmutated based on the clinically relevant 98% cutoff value for identity to the closest germline gene.4,5 Assignment of cases to specific stereotyped subsets was performed following established guidelines and based on the following stringent criteria: the IG sequences must: (i) have ≥ 50% amino acid identity and 70% similarity within the variable heavy complementarity-determining region 3 (VH CDR3); (ii) have the same VH CDR3 length and the shared amino acid patterns must occur at identical codon positions; and (iii) utilize IGHV genes belonging

Table 1. Immunogenetic characteristics of the major stereotyped subsets analyzed in the present study.

Subset #1 #2 #3 #4 #5 #6 #7 #8 #59 #99 Total

IG Mutational Subset size Cases status* (%)† analyzed (N) unmutated variable unmutated mutated unmutated unmutated unmutated unmutated unmutated unmutated

2.4 2.8 0.6 1.0 0.7 0.9 0.3 0.5 0.3 0.3 9.8

137 162 26 78 25 46 12 43 18 18 565

IGHV gene(s)

IGHJ gene(s)

Clan I genes IGHV3-21 IGHV1-69 IGHV4-34 IGHV1-69 IGHV1-69 IGHV1-69 IGHV4-39 Clan I genes Clan I genes

IGHJ4 IGHJ6 IGHJ6 IGHJ6 IGHJ6 IGHJ3 IGHJ6 IGHJ5 variable IGHJ4

VH CDR3 AA length

VH CDR3 pattern

13 ARx[NQ]W[AVLI]xxxxFDx 9 [AVLI]x[DE]xxxM[DE]x 22 Axxxxx[AVLI][AVLI]VxxAxxxxYYGMDx 20 [AVLI]RGxxxxxxx[KRH]RYYYYGx[DE]x 20 ARxxxxxx[AVLI]xxxYYYYxMDx 21 ARGGxYDY[AVLI]WGSYRxx[DE][AVLI]FDx 24 AxxxxxxDFW[ST]GYxxxxYYYxxDx 18-19 AxxxxYSSSWxxxxNWFDP 12 AxxxDFWSGxxx 14 ARxQWLxxxxxFDx

IG: immunoglobulin; IGHV: immunoglobulin heavy variable; IGHJ: immunoglobulin heavy joining; VH CDR3: heavy variable complementarity determining region 3; AA: amino acid. *Cases with ≥98% identity to the germline were classed as unmutated (U-CLL), whereas cases with <98% germline identity were considered mutated (M-CLL). Subset #2 is comprised of both U-CLL and M-CLL. †Subset size refers to their approximate proportion within all CLL.14

960

haematologica | 2016; 101(8)

Recurrent gene mutations in stereotyped CLL subsets

to the same phylogenetic clan.13,14 The sole exception to these rules concerned subset #8, where the specific combination of IGHV439, IGHD6-13 and IGHJ5 genes resulted in a VH CDR3 motif that was shared by two subgroups of cases bearing VH CDR3s that differed in length by a single amino acid residue (18 and 19 amino acids) (Online Supplementary Table S1).10,14 For the present study, these two ‘sub-subsets’ were considered as a single entity. The immunogenetic characteristics of the subsets analyzed are provided in Table 1.

Gene mutation analysis The date of the sample used for mutational analysis as well as the date of the first treatment was available for >80% of patients, and >80% of patients were tested prior to treatment (Online Supplementary Table S2). Mutational screening was performed for the following genes: NOTCH1, the entire exon 34 or targeted analysis for del7544_7545/p.P2514Rfs*4; TP53, exons 4-10 (depending on the medical center); SF3B1, exons 14-16; BIRC3, exons 6-9; and MYD88, exons 3 and 5 or targeted analysis for the p.L265P substitution (exon 5). Specific methodologies are detailed in the Online Supplementary Table S3.

Statistical analysis Due to the complete absence or rarity of mutations in MYD88 and BIRC3, respectively, statistical analysis was only performed for NOTCH1, SF3B1 and TP53 gene mutations. Pearson’s Chisquared test was used to evaluate the null hypothesis that the frequency of mutations within each of the aforementioned genes is equal among all subsets analyzed; the P value was computed by Monte Carlo simulation with 10 000 replicates. Comparisons between subsets were performed using the Fisher’s exact test and all tests were two-sided. P values were corrected for multiple comparisons using the Bonferroni method and the level of significance was set at P<0.001 (Online Supplementary Table S4). All calculations were performed using R (version 3.1.2). Overall survival (OS) and time to first treatment (TTFT) were

measured from the date of diagnosis until last follow-up/death, or date of initial treatment, respectively. Survival curves were constructed according to the Kaplan-Meier method, using Statistica Software 10.0 (Stat Soft Inc., Tulsa, OK, USA), and the log-rank test was used to determine the differences between survival proportions.

Results Markedly different frequencies of genetic lesions amongst stereotyped CLL subsets To investigate the associations between recurrent genetic lesions in CLL and BcR IG stereotypy, we profiled 565 patients assigned to one of 10 major stereotyped subsets (Table 1).10,14 Subset #4 cases (n=78) carried uniformly mutated BcR IGs (M-CLL), whereas subset #2 cases (n=162) exhibited significant heterogeneity with regard to SHM load, leading to 98 cases being considered as M-CLL and the remainder (n=64) as U-CLL (range: 93.3-99.7%; median: 97.8%). All remaining subsets concerned cases with unmutated or minimally mutated BcR IGs, hence constituting U-CLL. Mutational analysis for all 5 genes, BIRC3, MYD88, NOTCH1, SF3B1 and TP53, was performed for 520/565 (92%) cases; mutation data for one or two genes was lacking for 37 (6.5%) and 8 (1.5%) cases, respectively (Online Supplementary Table S5). For comparison purposes, data concerning the frequency of recurrent gene mutations and genetic lesions within non-stereotyped CLL is also provided in Table 2 and Online Supplementary Table S6.

MYD88 mutations All cases analyzed (n=557) were devoid of MYD88 mutations, which, bearing in mind that our cohort was predominantly composed of U-CLL, was not surprising,

Figure 1. Recurrent gene mutations in stereotyped CLL subsets. The distribution of mutations within NOTCH1, SF3B1 and TP53 varied considerably between the major stereotyped subsets included in the present study. All cases were devoid of MYD88 mutations and BIRC3 mutations were rare, with no clear bias to any subset (this data is not shown in the graph). M-CLL: mutated IGHV gene; U-CLL: unmutated IGHV gene.

haematologica | 2016; 101(8)

961

L.-A. Sutton et al. Table 2. Frequency of recurrent mutations and cytogenetic aberrations within each subset included in the present study.

Immunogenetics Mutations NOTCH1mut TP53mut SF3B1mut BIRC3mut MYD88mut Genetic aberrations del(17p) del(11q) trisomy 12 del(13q)* no RCAs

#1 (n=137) #99 (n=18) #59 (n=18) #3 (n=26) #5 (n=25) #6 (n=46) #7 (n=12) #8 (n=43) #2 (n=162) #4 (n=78) n=289 UM Clan I genes UM IGHV1-69 UM M & UM M Heterogeneous IGHV4-39 IGHV3-21 IGHV4-34 CLL 37/137 (27%) 21/135 (16%) 9/137 (7%) 2/131 (2%) 0/135 (0%)

4/18 (22%) 6/18 (33%) 0/18 (0%) 0/17 (0%) 0/18 (0%)

6/18 (33%) 1/26 (4%) 2/24 (8%) 10/45 (22%) 1/12 (8%) 0/18 (0%) 2/25 (8%) 0/25 (0%) 2/45 (4%) 1/10 (10%) 2/18 (11%) 12/26 (46%) 2/25 (8%) 6/46 (13%) 3/12 (25%) 0/18 (0%) 0/25 (0%) 1/21 (5%) 0/45 (0%) 1/12 (8%) 0/18 (0%) 0/25 (0%) 0/24 (0%) 0/45 (0%) 0/12 (0%)

13/43 (30%) 7/162 (4%) 1/43 (2%) 3/150 (2%) 0/43 (0%) 72/161 (45%) 3/42 (7%) 0/153 (0%) 0/42 (0%) 0/159 (0%)

1/78 (1%) 3/78 (4%) 0/78 (0%) 0/77 (0%) 0/78 (0%)

10/280 (3.6%) 11/237 (4.6%) 10/280 (3.6%) 0/189 (0%) 5/206 (2.4%)†

12/100 (12%) 22/98 (22%) 14/90 (16%) 18/89 (20%) 30/89 (34%)

2/13 (15%) 2/13 (15%) 3/12 (25%) 3/12 (25%) 2/12 (17%)

0/12 (0%) 1/12 (8%) 8/10 (80%) 1/10 (10%) 0/10 (0%)

3/23 (13%) 0/128 (0%) 3/22 (14%) 30/126 (24%) 13/20 (65%) 5/124 (4%) 1/20 (5%) 67/123 (54%) 3/20 (15%) 26/123 (21%)

0/59 (0%) 3/59 (5.1%) 1/52 (2%) 28/52 (54%) 20/52 (38%)

10/269 (3.7%) 29/269 (11%) 29/269 (11%) 135/269 (50%) 72/269 (27%)

2/18 (11%) 7/19 (37%) 1/17 (6%) 3/17 (18%) 6/17 (35%)

0/20 (0%) 2/34 (6%) 3/9 (33%) 8/21 (38%) 5/34 (15%) 1/9 (11%) 3/18 (17%) 3/34 (9%) 2/8 (25%) 5/17 (29%) 9/34 (26%) 2/8 (25%) 5/17 (29%)16/34 (47%) 1/8 (13%)

*refers to del(13q) as the sole aberration. RCAs: recurrent cytogenetic aberrations; †one MYD88-mutant case also carried a mutation within TP53. NOTCH1mut: mutation in NOTCH1; TP53mut: mutation in TP53; SF3B1mut: mutation in SF3B1; BIRC3mut: mutation in BIRC3; MYD88mut: mutation in MYD88; UM: cases with unmutated IGHV genes; M: cases with mutated IGHV genes. The ‘Heterogeneous CLL’ cases described in the last column of this table refers to newly diagnosed CLL patients from a population-based cohort called SCALE (Scandinavian Lymphoma Etiology). Within this study, 330 CLL cases had immunogenetic and mutation data available, resulting in 41 cases (12%) carrying stereotyped BcR IGs and therefore being assigned to a major subset. For comparison purposes, the frequency of recurrent mutations and cytogenetic aberrations in the remaining cases carrying heterogeneous BcR IGs (n=289) are provided.32

since existing evidence indicates that MYD88 mutations exclusively occur in M-CLL.32-37 That said, 32% (176/557) of cases analyzed concerned M-CLL, subset #2 (n=98) (mixed SHM profile) and subset #4 (n=78), and the complete absence of MYD88 mutations amongst these cases implies that mutations within MYD88 are absent from M-CLL assigned to major stereotyped subsets.

BIRC3 mutations Mutations within BIRC3 were infrequent (7/541 cases, 1.3%) and primarily concerned truncating mutations i.e. small frameshift deletions, duplications or insertions, as opposed to single nucleotide variants (Table 2; Online Supplementary Tables S6 and S7). BIRC3-mutated cases lacked del(17p), however a single subset #1 case did coexist with a TP53 mutation within the DNA-binding domain (Online Supplementary Table S7); the remaining cases harbored del(11q) (n=1), trisomy 12 (n=3) or both del(11q) and trisomy 12 (n=2). Five out of 7 (71%) BIRC3-mutant cases carried mutations within one of the other genes analyzed; 4/5 cases had concurrent mutations within NOTCH1. Irrespective of subset assignment, BIRC3 mutations never coincided with SF3B1 mutations (Online Supplementary Table S7).

NOTCH1 mutations Mutations within exon 34 of the NOTCH1 gene were detected in 15% (82/563) of cases. The frequency of NOTCH1 mutations varied considerably among subsets and can be summarized as follows: (i) subsets #1, #59 and #99, all concerning U-CLL, exhibited high frequencies of NOTCH1 mutations (22%, 30% and 33%, respectively); (ii) NOTCH1 mutations were also enriched in subset #6 (22%), but relatively infrequent in subsets #3, #5 and #7 (4%, 8% and 8%, respectively), despite the fact that all these subsets utilize the IGHV1-69 gene and concern UCLL; (iii) NOTCH1 mutations were prevalent in the clinically aggressive subset #8 (IGHV4-39/IGKV1(D)-39), ranging from 21-34% depending on the VH CDR3 length and 18 or 19 amino acids, respectively; (iv) aberrations within NOTCH1 were uncommon in subset #2, which comprises 962

both U- and M-CLL, being present in only 4% of analyzed cases (7/162; 5/7 (71%) NOTCH1-mutant subset #2 cases concerned M-CLL while the remaining 2 cases were UCLL); and, finally, (v) defects within NOTCH1 were rare in subset #4 (1/78; 1.3%) (Figure 1; Table 2; Online Supplementary Tables S1 and S6). To account for multiple testing, Bonferroni correction was performed and pairwise comparisons between subsets was checked at a level of significance of P<0.001, indicating that the frequencies of NOTCH1 mutations in subsets #2 and #4 were statistically significant compared to the frequencies observed in subsets #1, #6, #8 and #59 (Online Supplementary Table S4A). Although significance was not reached for other subset comparisons, this could in part be explained by the small number of cases included within each of these subsets, thus underscoring an inherent limitation when analyzing stereotyped subsets due to the fact that even the largest subsets account for less than 5% of all CLL cases. Although NOTCH1 mutations tended to coincide with trisomy 12, their co-occurrence differed among subsets (Table 2; Online Supplementary Table S6). Concurrent mutations were uncommon in NOTCH1-mutated cases, with the vast majority (79%; 65/82) devoid of mutations in the other four genes included in the study. The combinatorial patterns of mutations occurring in the remaining 21% of cases (17/82) were as follows: (i) NOTCH1 and TP53 mutations coexisted in 7/17 (41%) cases comprising subset #1 (n=4), subset #99 (n=2) and subset #6 (n=1); (ii) mutations in NOTCH1 and either SF3B1 or BIRC3 occurred at similar frequencies, 5/17 (29%) and 4/17 (24%), respectively, with no bias to any particular subset; and (iii) a single subset #1 case carried mutations in NOTCH1, TP53 and SF3B1. Notably, subsets harboring NOTCH1 mutations at high frequencies were either absent for (subsets #8 and #99) or concerned a single case with a co-occurring SF3B1 mutation (subsets #1, #6 and #59).

SF3B1 mutations Mutations within the SF3B1 gene (hotspot exons 14-16) were detected in 19% (106/564) of cases. A very high frehaematologica | 2016; 101(8)

Recurrent gene mutations in stereotyped CLL subsets

quency of SF3B1 mutations was not only observed in subset #2 (45%; 72/161) but also in subset #3 (46%; 12/26), which collectively accounted for 79% (84/106) of all SF3B1 mutated cases. The frequency of SF3B1 mutations in both subsets #2 and #3 sharply contrasted the frequency observed in all other subsets, and reached statistical significance when compared to subsets #1, #4, #8 and #99 (P<0.001, the frequency of SF3B1 mutations in subset #2 also reached significance when compared to subsets #5 and #6) (Figure 1; Table 2; Online Supplementary Tables S4B, S6 and S8). No significant difference in the distribution of SF3B1 mutations was observed between subset #2 M-CLL cases (47/98; 48%) versus subset #2 U-CLL cases (25/64; 39%) (P=0.27).

The SF3B1 mutation distribution within subset #2 was remarkably skewed, with 77% (58/75; 3 subset #2 cases carried 2 SF3B1 mutations) of mutations localized to two codons (p.K700E: n=43/75, 57%; p.G742D: n=15/75, 20%; Figure 2A; Online Supplementary Table S8). Similar to subset #2, the p.K700E substitution accounted for a high proportion of SF3B1 mutations in subset #3 (33%; 4/12), while the p.G742D mutation was absent (Figure 2A; Online Supplementary Table S8). When considering cytogenetic aberrations, subset #2 was enriched for del(11q) (30/126; 24%) and isolated del(13q) (67/123; 54%), in line with previous reports.16,23,38 SF3B1-mutated subset #2 cases showed a negative association with isolated del(13q) (25/57 vs. 42/66; P=0.028).

Figure 2. SF3B1 mutations in subsets #2 and #3. (A) Distribution of SF3B1 mutations in subsets #2 and #3. Overall, 45% (72/161) of subset #2 and 46% (12/26) of subset #3 cases were found to carry mutations within SF3B1. While the majority of subset #2 cases (69/72; 96%) carried a single SF3B1 mutation, 3 cases had 2 mutations within SF3B1 (2/3 cases carried the p.K700E change, with one case carrying both p.K700E and p.G742D). Although the most frequent amino acid change in both subsets involved a lysine to glutamic acid substitution at codon 700 (exon 15), representing 57% (43/75) and 33% (4/12) of all SF3B1 mutations in subsets #2 and #3, respectively, the overall distribution of mutations varied. To elaborate, in subset #3, all remaining SF3B1 mutations (excluding p.K700E) occur within exon 14. This is in contrast to the SF3B1 mutational profile in subset #2 where only 17% of mutations are found within exon 14 and where two particular substitutions account for 96% and 88% of the alterations observed in exon 15 and exon 16 (p.K700E and p.G742D, respectively). *indicates that more than one amino acid change occurred at this codon (detailed in Online Supplementary Table S8). (B) Prognostic implications of SF3B1 mutations within subset #2 on overall survival (OS) and time to first treatment (TTFT). (Binet A cases).

haematologica | 2016; 101(8)

963

L.-A. Sutton et al.

Within subset #2 cases, del(11q) was associated with shorter TTFT (Online Supplementary Figure S1A), whereas SF3B1 mutations had no significant impact on either OS or TTFT (for Binet A patients) (Figure 2B; Online Supplementary Figure S1B-S1F). Despite their immunogenetic differences, similar to subset #2, subset #3 was also enriched for del(11q) (7/19; 37%), with both SF3B1-mutant and wild-type subset #3 cases carrying del(11q) in similar proportions (Table 2). Finally, across all subsets, SF3B1 mutations tended to occur in isolation, e.g. for SF3B1-mutant cases, 94% (68/72) of subset #2, 75% (9/12) of subset #3 and 83% (5/6) of subset #6 had no coincident mutations within any other gene analyzed. The sole exception concerned subset #1, within which 56% (5/9) of SF3B1-mutant cases coexisted with mutations in TP53 (one case also carried a mutation within NOTCH1).

TP53 mutations Overall, the majority of cases analyzed had no mutations within TP53 (93%; 508/547). That said, we noted differences in the frequency of TP53 mutations amongst specific subsets, ranging from 0% to 33% (Figure 1; Table 2; Online Supplementary Tables S4C, S6 and S9). In particular, we found that TP53 mutations were: (i) absent in subset #59; (ii) rare in subsets #2 (3/150; 2%), #4 (3/78; 4%) and #8 (1/43; 2%); (iii) had varying, albeit low, frequencies in U-CLL subsets utilizing the IGHV1-69 gene (Subsets #3: 2/25 (8%); #5: 0/25 (0%); #6: 2/45 (4%) and #7: 1/10 (10%)); and (iv) were enriched in subsets #1 (21/135; 16%) and #99 (6/18; 33%); the latter representing a less populated subset that immunogenetically bears a high similarity to subset #1. The high frequency of TP53 mutations in these latter two subsets reached statistical significance compared to the frequency observed in subset #2 (Online Supplementary Table S4C). Of note, when incorporating del(17p) into our analysis, the frequency of TP53 aberrant cases in subsets #1 and #99 increased to 22% and 40%, respectively (Online Supplementary Table S9). Subsets #5 and #59 remained unaffected by any TP53 lesions, while the number of cases inflicted with TP53 defects increased to 3/8 (37.5%) for subset #7 and 4/19 (21%) for subset #3 (Online Supplementary Table S9). As mentioned above, a significant proportion of TP53-mutated cases carried concurrent mutations in other genes analyzed. More specifically, almost 50% of TP53-mutant subset #1 cases were positive for either SF3B1 or NOTCH1 mutations (both 4/21, 19%), whereas one case was positive for all three mutations; an additional case carried a mutation within BIRC3. Regarding subset #99, 2/6 (33%) of TP53-mutant cases carried double mutations, with both cases carrying the recurrent 2 base pair deletion in NOTCH1.

Discussion Within this study, by combining data concerning both recurrent gene mutations and cytogenetic aberrations in the largest series of stereotyped subset CLL cases studied to date for this purpose, we were able to offer important novel insights into the molecular mechanisms driving the pathogenesis and evolution of each subset. Overall, we document that the genetic makeup of individual stereotyped subsets is remarkably distinct, which alludes to the 964

subset-biased acquisition and/or selection of genetic aberrations likely in the context of particular BcR signaling initiated by the subset-specific IG. Beginning with U-CLL subsets, it is noteworthy that even when comparing subsets with BcR IGs utilizing the same IGHV gene, their genomic profiles are distinct, as exemplified by the subsets utilizing IGHV1-69 (subsets #3, #5, #6 and #7), which displayed markedly different spectra of genomic aberrations. For instance, SF3B1 mutations were detected in an impressive 46% of subset #3 cases versus only 8% of subset #5 cases. Similar observations regarding the skewed distribution of particular recurrent gene mutations were also apparent for NOTCH1 mutations, which were detected at a much higher frequency in subset #6 (22%) versus all other IGHV1-69 subsets (frequencies 4-8%); interestingly, within subset #6, NOTCH1 mutations rarely co-occurred with trisomy 12. Finally, TP53 defects due to mutations within TP53 and/or del(17p) ranged from 37.5% in subset #7 to 0% in subset #5. Switching our focus to U-CLL subsets #8 and #59, two clinically aggressive subsets, with the former exhibiting the highest risk for Richter's transformation amongst all CLL,15 revealed a very high frequency of trisomy 12 (65% and 80%, respectively) and an enrichment in NOTCH1 mutations (30% and 33%, respectively), with a relatively low frequency or complete absence of all other genomic aberrations evaluated.23 A similarly skewed distribution of genomic aberrations was identified for another very aggressive CLL subset, subset #2, which is the largest subset overall.14,23 Indeed, we herein confirm and significantly extend previous observations that subset #2 is remarkably enriched for mutations within the SF3B1 gene.26,27 Of note, the targeting profile of mutations in the hotspot regions of SF3B1 (exons 14-16) clearly differed in subset #2 from that in subset #3. Although the actual significance of these observations is currently unknown, the high frequency and striking bias of SF3B1 mutations to subsets #2 and #3 bodes strongly for their critical role in the pathobiology of these subsets. Furthermore, it supports the argument that the mechanisms underlying clinical aggressiveness in CLL are not uniform, but rather differ among the various disease subgroups. Along these lines, the relative paucity of TP53 defects in subset #2 (2%) implies that the poor prognosis of subset #2 is attributable to other factors, with SF3B1 seemingly appearing as a top candidate. That said, when comparing subset #2 cases with/without SF3B1 mutations, no difference in OS or TTFT (Binet A cases) were observed, suggesting that SF3B1 dysregulation alone does not explain the clinical aggressiveness of subset #2. Since SF3B1 gene mutations are also found in other malignancies, most notably myelodysplastic syndrome, within which their presence appears to confer a more favorable prognosis,39 it is conceivable that the functional impact of SF3B1 mutations may be influenced by the specific microenvironment and/or differ in distinct blood cell lineages, hence producing context-dependent effects. The discussion about aggressive CLL subsets culminates with subset #1, the biggest within U-CLL and second largest overall after subset #2.14 In contrast to the subsets mentioned above, the pattern of cytogenetic aberrations and recurrent gene mutations in subset #1 is quite heterogeneous (Figure 3A and 3B; Table 2). Of note, TP53 disruphaematologica | 2016; 101(8)

Recurrent gene mutations in stereotyped CLL subsets

tion (due to del(17p) and/or TP53 mutations), NOTCH1 mutations and trisomy 12 were all frequent in subset #1 (22%, 27% and 16% of cases, respectively); although lesions within both TP53 and NOTCH1 co-occurred in only 5 cases, all were negative for trisomy 12. Altogether, these findings support the existence of distinct genetic pathways of clonal evolution in subset #1, one influenced and dictated by TP53 disruption, while the other is dependent on trisomy 12 and/or NOTCH1 mutations. Regarding the latter, given the very recent discovery of recurrent mutations in the 3â&#x20AC;&#x2122; UTR of NOTCH1 in approximately 3% of CLL patients,40 it is plausible that their frequency in both subset #1 and CLL at large has been underestimated to date, nevertheless, this does not detract from the apparent non-random association of the p.P2514Rfs*4 NOTCH1 mutation reported herein with certain stereotyped CLL subsets. We next integrated these findings with the specific immunogenetic nature of subset #1, taking into consideration the fact that cases assigned to this subset differ from

most major CLL subsets, in that they do not all express the same IGHV gene but rather carry IGHV genes that share common ancestry and thus, belong to the same IGHV phylogenetic clan, clan I (comprising IGHV1, IGHV5 and IGHV7 genes).10,14 Focusing on the IGHV1-2*02 and IGHV1-3*01 genes, which accounted for 64% of all subset #1 cases (28.7% and 35.3%, respectively), it was interesting to note that 17/39 (43.6%) IGHV1-2*02 expressing cases carried mutations within NOTCH1 compared to only 7/48 (14.6%) IGHV1-3*01 expressing subset #1 cases (P=0.003) (Figure 3C). This finding further exemplifies how the expression of a particular stereotyped immunoglobulin may be linked to a distinct evolutionary pathway through the acquisition of specific genomic aberrations. Whether these genomic differences may translate into different clinical outcomes remains to be elucidated. Another important observation relates to the finding that approximately 20% of subset #1 cases were negative for any of the genetic lesions tested for (mutations or aberrations). Thus, for at least a proportion of subset #1 cases,

Figure 3. Main biological associations within subset #1. (A) Concurrent mutations within subset #1. Only cases for which the mutational status of all 3 genes (NOTCH1, SF3B1 and TP53) was available were included in the figure (n=135). Fifty-seven cases had a mutation in at least one of the gene hotspots whereas 78 cases were wild-type for these genes. (B) Spectrum of mutations and genomic aberrations within subset #1. Despite a high frequency of NOTCH1 mutations, a large proportion (58%) of subset #1 cases carried no mutations within the 5 genes analyzed. Specifically, considering the subset #1 cases lacking any recurrent gene mutations, 35% also lacked any recurrent genetic aberrations. Collectively, this resulted in the absence of any recurrent gene mutation or cytogenetic aberration in approximately 20% of subset #1 cases, thereby implying that additional mechanisms must account for the clinically aggressive nature of this subset. Only cases for which the mutational status of all 3 genes (NOTCH1, SF3B1 and TP53) was available were included in the figure (n=135). *indicates that none of the known recurrent genomic aberrations were present; NOTCH1mut: mutation in NOTCH1 only; TP53mut: mutation in TP53 only; SF3B1mut: mutation in SF3B1 only. Concurrentmut refers to the presence of mutations in more than one of the genes analyzed. Absolute numbers and percentages are provided in brackets. For del(17p), 2/53 correspond to the 4% indicated in the figure. (C) The frequency of NOTCH1 mutations in subset #1 cases varies depending on specific IGHV gene usage. Only the top 3 utilized IGHV genes within subset #1 patients in our cohort were included in the graph, collectively accounting for 73% (99/136) subset #1 cases. Mutations within NOTCH1 were found to be particularly frequent in subset #1 cases expressing IGHV1-2*02 (17/39; 44%).

haematologica | 2016; 101(8)

965

L.-A. Sutton et al.

it appears that the full extent of their genomic complexity has yet to be revealed, and additional mechanisms must underlie their clinical aggressiveness. This claim is supported by recent high-throughput studies which indicate that mutations within the NFKBIE or RPS15 gene may serve as novel pathogenic mechanisms linked to a more aggressive disease.41,42 While highlighting an inherent limitation of the present study, which was restricted to the analysis of 5 recurrently mutated genes, these results, together with the recent finding of non-coding recurrent mutations in CLL, emphasize the need for more comprehensive approaches, such as whole-genome sequencing, in order to obtain a complete picture of the genomic landscape of CLL subsets.40 Finally, we looked to subset #4, the largest subset within M-CLL and now considered as a prototype for indolent disease.14,23 Previous studies have reported that subset #4 is virtually devoid of cytogenetic aberrations associated with adverse prognosis [del(11q) and, especially, del(17p)].16 Herein, we not only confirm and significantly extend these observations, but also take a decisive step further by showing that subset #4 essentially lacks recurrent gene mutations, at least amongst the five genes analyzed. These results provide insights into the ‘mild’ genomic background of subset #4 CLL, which is reflected in the indolent clinical course experienced by these patients. Although our understanding of the genetic basis, clonal architecture and evolution in CLL pathogenesis is rapidly advancing, the CLL cell of origin and the precise timeline of events leading to leukemogenesis remain elusive. It has recently been proposed that hematopoietic stem cells (HSCs) may be involved in CLL pathogenesis, since they have been shown to harbor mutations within potential CLL oncogenes.43,44 These cells could undergo antigen-driven clonal selection and progressively acquire additional genetic abnormalities giving rise to a clonal B cell expansion, termed monoclonal B-cell lymphocytosis (MBL), which we know invariably precedes CLL, and ultimately may progress to CLL. However, in trying to reconcile the above scenario with the biology of CLL, this model fails to account for the fact that none of the recurrent CLL chromosomal abnormalities were found in the HSC compartment; how could such clones be selected and subsequently expand into CLL? In addition, CLL patients who achieve remissions do not relapse with polyclonal disease or distinct clones. Further studies are therefore required to both confirm and clarify the precise role of HSCs in CLL pathogenesis. A minute proportion of MBL transform to overt CLL requiring treatment (1-2%/year), thus significant somatic changes coupled with microenvironmental factors must occur.45 The existence of stereotyped BcRs together with the tendency of CLL cells to express poly- and auto-reac-

References 1. Hashimoto S, Dono M, Wakai M, et al. Somatic diversification and selection of immunoglobulin heavy and light chain variable region genes in IgG+ CD5+ chronic lymphocytic leukemia B cells. J Exp Med. 1995;181(4):1507-1517. 2. Fais F, Ghiotto F, Hashimoto S, et al. Chronic lymphocytic leukemia B cells

966

tive BcRs are indicative of selective pressures, such as autoantigens or microbial pathogens, that favor specific IG gene rearrangements. Such persistent stimulation through the surface IG may be the central event that drives the evolution from a preleukemic state to overt leukemia, thereafter chronic BcR engagement may favor the selection of a monoclonal population which subsequently acquires genetic alterations, which in some instances may provide a survival and growth advantage.45 In conclusion, our present findings imply that distinctive modes of microenvironmental interactions, mediated by certain stereotyped BcRs, may be associated with the selection or occurrence of particular genetic aberrations, with the combined effect determining both clonal and clinical evolution, and ultimately disease outcome. Our study further serves to highlight the fact that even findings within smaller subsets help to dissect the pathophysiology of CLL, and may eventually have clinical usability akin to the known prognostication value of TP53 gene mutations, despite being present in only 5-10% of cases at diagnosis or the association between subset #8 (0.6% of all CLL) and Richter’s transformation.15,46 Overall, since the distinct genomic profiles evidenced amongst stereotyped CLL subsets are potentially linked to varying mechanisms of clinical aggressiveness based on a reliance on specific intracellular signaling pathways, our findings may have important implications for patient monitoring and therapeutic management in the era of targeted pathway inhibition. Funding The authors would like to thank the Swedish Cancer Society, the Swedish Research Council, The Lion’s Cancer Research Foundation and Selander’s Foundation, Uppsala; the ENosAI project (code 09SYN-13-880) co-funded by the EU and the General Secretariat for Research and Technology of Greece; the KRIPIS action, funded by the General Secretariat for Research and Technology of Greece; the AIRC Special Program Molecular Clinical Oncology, 5 x 1000, #9965 and 10007 and Investigator grant, Milano, Italy); Ricerca Finalizzata 2010 - Ministero della Salute, Roma, Italy; Leukaemia and Lymphoma Research, Kay Kendall Leukaemia Fund, UK; the Spanish ICGC-CLL Genome Project funded by the Instituto de Salud Carlos III (ISCIII); The Danish Cancer Society; the research grants IGA MZ CR NT13493-4/2012 and MSMT CR VaVPI project CZ.1.05/1.1.00/02.0068 (CEITEC); H2020 “AEGLE, An analytics framework for integrated and personalized healthcare services in Europe”, and H2020 “MEDGENET, Medical Genomics and Epigenomics Network” (No.692298), both funded by the European Commission, for their financial support; and also Barbara Kantorova and Veronika Navrkalova for their technical help with NOTCH1 and SF3B1 sequencing analysis of the Czech cohort.

express restricted sets of mutated and unmutated antigen receptors. J Clin Invest. 1998;102(8):1515-1525. 3. Dighiero G, Hamblin TJ. Chronic lymphocytic leukaemia. Lancet. 2008; 371(9617): 1017-1029. 4. Damle RN, Wasil T, Fais F, et al. Ig V gene mutation status and CD38 expression as novel prognostic indicators in chronic lymphocytic leukemia. Blood. 1999;94(6):18401847.

5. Hamblin TJ, Davis Z, Gardiner A, Oscier DG, Stevenson FK. Unmutated Ig V(H) genes are associated with a more aggressive form of chronic lymphocytic leukemia. Blood. 1999;94(6):1848-1854. 6. Tobin G, Thunberg U, Johnson A, et al. Chronic lymphocytic leukemias utilizing the VH3-21 gene display highly restricted Vlambda2-14 gene use and homologous CDR3s: implicating recognition of a common antigen epitope. Blood. 2003;101(12):

haematologica | 2016; 101(8)

Recurrent gene mutations in stereotyped CLL subsets

4952-4957. 7. Ghiotto F, Fais F, Valetto A, et al. Remarkably similar antigen receptors among a subset of patients with chronic lymphocytic leukemia. J Clin Invest. 2004;113(7):1008-1016. 8. Messmer BT, Albesiano E, Efremov DG, et al. Multiple distinct sets of stereotyped antigen receptors indicate a role for antigen in promoting chronic lymphocytic leukemia. J Exp Med. 2004;200(4):519-525. 9. Tobin G, Thunberg U, Karlsson K, et al. Subsets with restricted immunoglobulin gene rearrangement features indicate a role for antigen selection in the development of chronic lymphocytic leukemia. Blood. 2004;104(9):2879-2885. 10. Stamatopoulos K, Belessi C, Moreno C, et al. Over 20% of patients with chronic lymphocytic leukemia carry stereotyped receptors: Pathogenetic implications and clinical correlations. Blood. 2007;109(1):259-270. 11. Murray F, Darzentas N, Hadzidimitriou A, et al. Stereotyped patterns of somatic hypermutation in subsets of patients with chronic lymphocytic leukemia: implications for the role of antigen selection in leukemogenesis. Blood. 2008;111(3):15241533. 12. Bomben R, Dal Bo M, Capello D, et al. Molecular and clinical features of chronic lymphocytic leukaemia with stereotyped B cell receptors: results from an Italian multicentre study. Br J Haematol. 2009;144(4): 492-506. 13. Darzentas N, Hadzidimitriou A, Murray F, et al. A different ontogenesis for chronic lymphocytic leukemia cases carrying stereotyped antigen receptors: molecular and computational evidence. Leukemia. 2010;24(1):125-132. 14. Agathangelidis A, Darzentas N, Hadzidimitriou A, et al. Stereotyped B-cell receptors in one-third of chronic lymphocytic leukemia: a molecular classification with implications for targeted therapies. Blood. 2012;119(19):4467-4475. 15. Rossi D, Spina V, Cerri M, et al. Stereotyped B-cell receptor is an independent risk factor of chronic lymphocytic leukemia transformation to Richter syndrome. Clin Cancer Res. 2009;15(13):44154422. 16. Marincevic M, Cahill N, Gunnarsson R, et al. High-density screening reveals a different spectrum of genomic aberrations in chronic lymphocytic leukemia patients with 'stereotyped' IGHV3-21 and IGHV434 B-cell receptors. Haematologica. 2010;95(9):1519-1525. 17. Marincevi3c M, Mansouri M, Kanduri M, et al. Distinct gene expression profiles in subsets of chronic lymphocytic leukemia expressing stereotyped IGHV4-34 B-cell receptors. Haematologica. 2010;95(12): 2072-2079. 18. Arvaniti E, Ntoufa S, Papakonstantinou N, et al. Toll-like receptor signaling pathway in chronic lymphocytic leukemia: distinct gene expression profiles of potential pathogenic significance in specific subsets of patients. Haematologica. 2011;96(11):16441652.

haematologica | 2016; 101(8)

19. Maura F, Cutrona G, Fabris S, et al. Relevance of stereotyped B-cell receptors in the context of the molecular, cytogenetic and clinical features of chronic lymphocytic leukemia. PloS one. 2011;6(8):e24313. 20. Ntoufa S, Vardi A, Papakonstantinou N, et al. Distinct innate immunity pathways to activation and tolerance in subgroups of chronic lymphocytic leukemia with distinct immunoglobulin receptors. Mol Med. 2012;18:1281-1291. 21. Kanduri M, Marincevic M, Halldorsdottir AM, et al. Distinct transcriptional control in major immunogenetic subsets of chronic lymphocytic leukemia exhibiting subsetbiased global DNA methylation profiles. Epigenetics. 2012;7(12):1435-1442. 22. Papakonstantinou N, Ntoufa S, Chartomatsidou E, et al. Differential microRNA profiles and their functional implications in different immunogenetic subsets of chronic lymphocytic leukemia. Mol Med. 2013;19:115-123. 23. Baliakas P, Hadzidimitriou A, Sutton L, et al. Clinical effect of stereotyped B-cell receptor immunoglobulins in chronic lymphocytic leukaemia: a retrospective multicentre study. Lancet Haematol. 2014;1(2): 74-84. 24. Del Giudice I, Chiaretti S, Santangelo S, et al. Stereotyped subset #1 chronic lymphocytic leukemia: a direct link between B-cell receptor structure, function, and patients' prognosis. Am J Hematol. 2014;89(1):74-82. 25. Gounari M, Ntoufa S, Apollonio B, et al. Excessive antigen reactivity may underlie the clinical aggressiveness of chronic lymphocytic leukemia stereotyped subset #8. Blood. 2015;125(23):3580-3587. 26. Strefford JC, Sutton LA, Baliakas P, et al. Distinct patterns of novel gene mutations in poor-prognostic stereotyped subsets of chronic lymphocytic leukemia: the case of SF3B1 and subset #2. Leukemia. 2013;27(11):2196-2199. 27. Rossi D, Spina V, Bomben R, et al. Association between molecular lesions and specific B-cell receptor subsets in chronic lymphocytic leukemia. Blood. 2013;121 (24):4902-4905. 28. Malcikova J, Stalika E, Davis Z, et al. The frequency of TP53 gene defects differs between chronic lymphocytic leukaemia subgroups harbouring distinct antigen receptors. Br J Haematol. 2014;166(4):621625. 29. 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. 30. Baliakas P, Iskas M, Gardiner A, et al. Chromosomal translocations and karyotype complexity in chronic lymphocytic leukemia: a systematic reappraisal of classic cytogenetic data. Am J Hematol. 2014; 89(3):249-255. 31. Gunnarsson R, Isaksson A, Mansouri M, et al. Large but not small copy-number alterations correlate to high-risk genomic aber-

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

rations and survival in chronic lymphocytic leukemia: a high-resolution genomic screening of newly diagnosed patients. Leukemia. 2010;24(1):211-215. Cortese D, Sutton LA, Cahill N, et al. On the way towards a 'CLL prognostic index': focus on TP53, BIRC3, SF3B1, NOTCH1 and MYD88 in a population-based cohort. Leukemia. 2014;28(3):710-713. Baliakas P, Hadzidimitriou A, Sutton LA, et al. Recurrent mutations refine prognosis in chronic lymphocytic leukemia. Leukemia. 2015;29(2):329-336. Puente XS, Pinyol M, Quesada V, et al. Whole-genome sequencing identifies recurrent mutations in chronic lymphocytic leukaemia. Nature. 2011;475(7354):101105. Wang L, Lawrence MS, Wan Y, et al. SF3B1 and other novel cancer genes in chronic lymphocytic leukemia. N Engl J Med. 2011;365(26):2497-2506. Martinez-Trillos A, Pinyol M, Navarro A, et al. Mutations in TLR/MYD88 pathway identify a subset of young chronic lymphocytic leukemia patients with favorable outcome. Blood. 2014;123(24):3790-3796. Baliakas P, Hadzidimitriou A, Agathangelidis A, et al. Prognostic relevance of MYD88 mutations in CLL: the jury is still out. Blood. 2015;126(8):10431044. Dohner H, Stilgenbauer S, Benner A, et al. Genomic aberrations and survival in chronic lymphocytic leukemia. N Engl J Med. 2000;343(26):1910-1916. Malcovati L, Papaemmanuil E, Bowen DT, et al. Clinical significance of SF3B1 mutations in myelodysplastic syndromes and myelodysplastic/myeloproliferative neoplasms. Blood. 2011;118(24):6239-6246. Puente XS, Bea S, Valdes-Mas R, et al. Noncoding recurrent mutations in chronic lymphocytic leukaemia. Nature. 2015;526 (7574):519-524. Mansouri L, Sutton LA, Ljungstrom V, et al. Functional loss of IkappaBepsilon leads to NF-kappaB deregulation in aggressive chronic lymphocytic leukemia. J Exp Med. 2015;212(6):833-843. Ljungstrom V, Cortese D, Young E, et al. Whole-exome sequencing in relapsing chronic lymphocytic leukemia: clinical impact of recurrent RPS15 mutations. Blood. 2016;127(8):1007-1016. Kikushige Y, Ishikawa F, Miyamoto T, et al. Self-renewing hematopoietic stem cell is the primary target in pathogenesis of human chronic lymphocytic leukemia. Cancer Cell. 2011;20(2):246-259. Damm F, Mylonas E, Cosson A, et al. Acquired initiating mutations in early hematopoietic cells of CLL patients. Cancer Discov. 2014;4(9):1088-1101. Sutton LA, Rosenquist R. Deciphering the molecular landscape in chronic lymphocytic leukemia: time frame of disease evolution. Haematologica. 2015;100(1):7-16. Pospisilova S, Sutton LA, Malcikova J, et al. Innovation in the prognostication of chronic lymphocytic leukemia: how far beyond TP53 gene analysis can we go? Haematologica. 2016;101(3):263-265.

967

ARTICLE EUROPEAN HEMATOLOGY ASSOCIATION

Hodgkin Lymphoma

Ferrata Storti Foundation

ENGAGE- 501: phase II study of entinostat (SNDX-275) in relapsed and refractory Hodgkin lymphoma Connie Lee Batlevi,1 Yvette Kasamon,2 R. Gregory Bociek,3 Peter Lee,4 Lia Gore,5 Amanda Copeland,1 Rachel Sorensen,6 Peter Ordentlich,6 Scott Cruickshank,6 Lori Kunkel,6 Daniela Buglio,7 Francisco Hernandez-Ilizaliturri,8 and Anas Younes1,7

Haematologica 2016 Volume 101(8):968-975

Memorial Sloan Kettering Cancer Center, New York, NY; 2John Hopkins University, Baltimore, MD; 3University of Nebraska Medical Center, Omaha, NE; 4Tower Cancer Research Foundation, Beverly Hills, CA; 5University of Colorado Cancer Center, Aurora, CO; 6Syndax Pharmaceuticals, Inc., Waltham, MA; 7MD Anderson Cancer Center, Houston, TX; and 8Roswell Park Cancer Institute, Buffalo, NY, USA 1

ABSTRACT

C Correspondence: younesa@mskcc.org

Received: January 13, 2016. Accepted: April 29, 2016. Pre-published: May 5, 2016. doi:10.3324/haematol.2016.142406

Check the online version for the most updated information on this article, online supplements, and information on authorship & disclosures: www.haematologica.org/content/101/8/968

lassical Hodgkin lymphoma treatment is evolving rapidly with high response rates from antibody-drug conjugates targeting CD30 and immune checkpoint antibodies. However, most patients do not achieve a complete response, therefore development of novel therapies is warranted to improve patient outcomes. In this phase II study, patients with relapsed or refractory Hodgkin lymphoma were treated with entinostat, an isoform selective histone deacetylase inhibitor. Forty-nine patients were enrolled: 33 patients on Schedule A (10 or 15 mg oral entinostat once every other week); 16 patients on Schedule B (15 mg oral entinostat once weekly in 3 of 4 weeks). Patients received a median of 3 prior treatments (range 1-10), with 80% of the patients receiving a prior stem cell transplant and 8% of patients receiving prior brentuximab vedotin. In the intention-to-treat analysis, the overall response rate was 12% while the disease control rate (complete response, partial response, and stable disease beyond 6 months) was 24%. Seven patients did not complete the first cycle due to progression of disease. Tumor reduction was observed in 24 of 38 (58%) evaluable patients. Median progression-free survival and overall survival was 5.5 and 25.1 months, respectively. The most frequent grade 3 or 4 adverse events were thrombocytopenia (63%), anemia (47%), neutropenia (41%), leukopenia (10%), hypokalemia (8%), and hypophosphatemia (6%). Twenty-five (51%) patients required dose reductions or delays. Pericarditis/pericardial effusion occurred in one patient after 12 cycles of therapy. Future studies are warranted to identify predictive biomarkers for treatment response and to develop mechanism-based combination strategies. (clinicaltrials.gov identifier: 00866333)

Introduction ÂŠ2016 Ferrata Storti Foundation Material published in Haematologica is covered by copyright. All rights reserved to the Ferrata Storti Foundation. Copies of articles are allowed for personal or internal use. Permission in writing from the publisher is required for any other use.

968

While Hodgkin lymphoma (HL) is highly curable through multi-agent chemotherapy and radiation therapy, approximately 20% of patients require second-line therapy which generally includes high-dose therapy with autologous stem cell transplantation (ASCT).1,2 Only 50% of patients undergoing ASCT are cured and the remaining patients treated with a palliative intent have a median overall survival (OS) of 2.4 years and median post progression-free survival (PFS) of 1.3 years.2,3 In recent years, two highly effective treatments have been identified for HL. These include the antibody drug conjugate, brentuximab vedotin, and the immune checkpoint inhibitors, nivolumab and pembrolizumab.4-6 Despite high response rates, the majority of the observed responses are partial.4-6 In the case of brentuxhaematologica | 2016; 101(8)

Entinostat in relapsed/refractory Hodgkin lymphoma

imab vedotin, patients treated in the relapsed setting have a median post progression-free survival of less than six months. For patients treated with brentuximab vedotin who achieved a complete response, approximately 50% of these patients relapse despite some receiving an allogeneic stem cell transplant for consolidation.6,7 Re-treatment with brentuximab vedotin is feasible, offering 60% overall response rate (ORR) and 30% complete response (CR), with a potential for improvement using combination strategies.8 Long-term results with immune checkpoint inhibitors are lacking, but to date, most responses are partial and some patients have progressed on therapy. This again highlights the need for additional drug development and identification of targeted therapy with single agent activity for combination therapy. Furthermore, the high activity of these novel targeted agents may drive the development of mechanism-based chemotherapy-free regimens with potentially less toxicity with respect to secondary cancers and cardiovascular disease. Accordingly, future treatment strategies will be aimed at developing effective new regimens that maintain a high cure rate while reducing treatment-related toxicities. The success of brentuximab vedotin and immune checkpoint inhibitors was based on taking advantage of the unique biology of HL, restricted CD30 expression, high PD-L1 expression and large numbers of T cells in the microenvironment. In this regard, histone deacetylase (HDAC) inhibitors are ideal candidates to exploit the biology of HL by modulating tumor cell death and the tumor microenvironment via non-overlapping mechanisms. HDAC inhibitors directly affect proliferation by increasing expression of p21 and down-regulating STAT6, culminating in caspase-induced cell death.9 Effects of HDAC inhibitors on the microenvironment include upregulation of OX40L (that is involved in generation of antigen specific memory T cells) and inhibition of thymus and activation regulated chemokine (TARC) (which attracts activated T-helper cells).9-11 The effects of HDAC inhibitors on the microenvironment include downregulation of tumor suppressor T cells to aid immune-mediated response.12 More recently, HDAC inhibitors have been shown to modulate PD-1 expression on peripheral blood T cells, suggesting synergism with immune checkpoint therapy.13 Several trials have studied HDAC inhibitors in HL with response rate of approximately 20%.14-16 HDAC inhibitors vary according to their specificity for HDAC isoforms, route of administration, and schedule. Entinostat is an oral pyridylcarbamate, class I isoform selective HDACi that targets HDAC 1, 2, 3, and HDAC 11.17,18 Compared to other HDAC inhibitors, entinostat has a unique pharmacokinetic (PK) signature with a prolonged half-life of approximately 140 hours, allowing for once or twice weekly dosing.19 In vitro experiments with entinostat suggest a strong anti-proliferative and immunomodulatory signal through upregulation of p21, downregulation of anti-apoptotic proteins, and modulation of chemokines including TARC.13,20,21 This phase II study evaluates the efficacy and safety of entinostat in patients with relapsed or refractory HL.

Methods Patient selection Patients with relapsed or refractory HL after an ASCT or those ineligible for ASCT were enrolled on this study. The eligibility crihaematologica | 2016; 101(8)

teria included age 18 years or over, Eastern Cooperative Oncology Group (ECOG) performance status of 0-1, and at least one site of measurable disease (≥1.5 cm). Adequate renal [serum creatinine ≤1.5xupper limit of normal (ULN)], bone marrow (absolute neutrophil count ≥1x109/L, and a platelet count ≥25x109/L in Schedule A and 50x109/L in Schedule B], and hepatic function (serum total bilirubin ≤1.5xULN, alanine aminotransferase and aspartate aminotransferase ≤2.5xULN) was required. Previous chemotherapy must have been completed three weeks prior to the first dose of entinostat. Exclusion criteria included known positivity for human immunodeficiency virus, active hepatitis B or C virus, central nervous system lymphoma, pregnancy or lactation, or a history of allogeneic stem cell transplantation within three months and active immunosuppressive therapy or graft-versus-host disease requiring treatment. Patients with a history of pericarditis or pericardial effusion requiring medical intervention within six months were also excluded from this study. Prior HDAC inhibitor treatment was not permitted. Enrollment began in 2009, prior to approval of brentuximab vedotin for relapsed or refractory HL. The study was conducted in accordance with the Declaration of Helsinki and the International Conference on Harmonization. All patients provided written informed consent. The Institutional Review Board and Ethical Committee at each site approved the study.

Study design and treatment plan This study was an open-label non-randomized, multicenter phase II trial of oral entinostat administered to patients with relapsed or refractory HL (clinicaltrials.gov identifier: 00866333) with the primary objective of assessing ORR (CR and PR). The secondary objectives included assessments of duration of response (DOR), OS, PFS, and safety and tolerability of entinostat. Patients were enrolled on two dosing schedules, both with 1 cycle defined as 28 days. Patients on Schedule A received 10 mg entinostat administered orally (PO) once every other week on a 28-day cycle. Upon determining tolerability of the 10 mg dose, entinostat was increased to 15 mg once every other week starting in week 2. Schedule B was initiated with 15 mg entinostat administered once weekly for three weeks on a 28-day cycle to determine whether greater frequency of entinostat led to increased control of disease. Entinostat was dose-reduced or held for grade 2 or greater nonhematologic toxicity or hematologic toxicity defined by absolute neutrophil count (ANC) less than 1x109/L or platelets less than 25x109/L. In the case of drug-associated grade 3 or 4 toxicities experienced by the patient in spite of optimal supportive care (including growth factor support and transfusion), treatment was withheld until symptoms improved to grade 1 or lower. Recurrence of grade 3 or 4 toxicities despite 2 levels of dose reduction to 10 mg and 7 mg of entinostat required treatment discontinuation. If symptoms did not resolve after four weeks of treatment interruption, the patients were removed from the study. Therapy was discontinued if there was evidence of progressive disease (PD), unacceptable toxicity, or withdrawal of consent.

Study assessments Computed tomography (CT) of the chest, abdomen and pelvis was performed at baseline, every 2 cycles for the first 9 cycles, and every 3-4 cycles thereafter. Disease assessment by fluorodeoxyglucose (FDG) positron emission tomography (PET) was performed at the investigator’s discretion. Tumor responses were based on the 2007 revised response criteria for malignant lymphoma.22 Progression of disease was defined as the appearance of new lesions or a greater than 50% increase in the sum of the products of perpendicular lesion diameters. A patient was classified as 969

C.L. Batlevi et al. Table 1. Patients’ characteristics (n=49, intent-to-treat population).

Median age (years) Sex Female Male ECOG performance status 0 1 Number of previous chemotherapy treatments Median < 4 lines of therapy ≥4 prior regimens Prior therapy with brentuximab vedotin Prior bone marrow or stem cell transplant Prior autologous transplant Prior allogeneic transplant Prior autologous & allogeneic transplant Transplant ineligible Response to last treatment Refractory <50% response to last treatment PD within 3 months of most recent therapy Relapsed disease PD following therapy(ies) with curative intent Bulky disease (1 or more baseline lesions ≥ 5 cm) Prior radiotherapy

Schedule A (n=33)

Schedule B (n=16)

Total (n=49)

33 (19-73)

33 (20-55)

33 (19-73)

14 (42%) 19 (58%)

10 (63%) 6 (38%)

24 (49%) 25 (51%)

19 (58%) 14 (42%)

11 (69%) 5 (31%)

30 (61%) 19 (39%)

3 (1-10) 19 (58%) 14 (42%) 3 (9%) 28 (85%) 21 (64%) 4 (12%) 3 (9%) 4 (12%)

3 (2-7) 8 (50%) 8 (50%) 1 (6%) 11 (69%) 9 (56%) 2 (13%) 4 (12%)

3 (1-10) 27 (55%) 22 (45%) 4 (8%) 39 (80%) 30 (61%) 4 (8%) 5(10%) 8 (16%)

4 (12%) 9 (27%)

4 (25%) 4 (25%)

8 (16%) 13 (27%)

20 (61%) 28 (85%) 26 (79%)

8 (50%) 11 (69%) 12 (75%)

28 (57%) 39 (80%) 38 (78%)

ECOG: Eastern Cooperative Oncology Group; PD: progressive disease.

SD if they did not meet criteria for complete response, partial response or progressive disease. Durable stable disease was classified as stable disease lasting six months or more. Safety assessments including vital signs, complete blood cell count analysis serum chemistry analysis, and physical examination were performed every two weeks of a 4-week cycle. Adverse events (AEs) and laboratory variables were assessed using the Common Terminology Criteria for Adverse Events version 3.0.

Correlative studies Serum cytokines including TARC were measured at baseline, after one week, and two weeks after initiation of entinostat. Sera were analyzed with commercially available enzyme-linked immunosorbent assays (ELISA) (R&D Systems, Minneapolis, MN, USA), and Multiplex human cytokine 30-plex kits (Invitrogen Corporation, Carlsbad, CA, USA) according to the manufacturers' protocols. Entinostat plasma concentrations were measured for a subset of patients. Samples were collected on days 1, 8 and 15 of cycle 1 for PK studies using a validated liquid chromatographytandem mass spectrometry (LC-MS/MS) assay performed in the Analytical Pharmacology Core laboratory at the Sidney Kimmel Comprehensive Center at Johns Hopkins Medical Center.23 Samples were collected at 0 hours (pre-dose), 0.25, 1, and 2 hours post dose as well as pre-dose on days 8 and 15.

Statistical analysis Sample size was based on Simon’s optimal 2-stage design and an ORR end point for both Schedule A and B. Response evaluable patients as defined by the per-protocol population were required to complete 2 cycles of entinostat and undergo response evaluation at screening and end of cycle 2. Patients who discontinued due to intolerable treatment-related AE prior to cycle 3 day 1 were 970

included in the intent-to-treat population. Efficacy was assessed in both the intent-to-treat and per-protocol population. Patients were included in the safety analysis if they received at least one dose of entinostat. Tumor response rate was estimated on the basis of the proportion of patients whose best overall response was CR or PR. Rate of disease control is defined by patients with a CR, PR, or SD lasting longer than or equal to six months. SD was measured from the start date of entinostat until the criteria for disease progression was first met. DOR was calculated for patients who achieved PR or better. For such patients, DOR represents the number of days from the start date of response to the date recurrent or progressive disease was first documented. Progressionfree survival was measured from the date of the first dose of entinostat to the earlier of documented disease progression or death due to any cause. The duration of DOR and PFS was right-censored at the last disease assessment for patients alive and without documentation of PD. Patients who started a non-protocol defined anticancer therapy prior to documentation of PD were censored at the last disease assessment prior to the initiation of such therapy. OS was measured from date of first dose of entinostat to the date of death from any cause and right-censored for patients reported alive as of the date of last contact. Time to event end points (DOR, PFS and OS) were summarized descriptively using the Kaplan-Meier method. Biomarker values were summarized in a descriptive manner at each sample collection time point [day 1 (baseline), Day 8, and Day 15]. Changes were evaluated by calculating differences within patients from baseline to each post-baseline time point. The Wilcoxon signed rank test was used to detect statistically significant (P<0.05) within-patient changes from baseline. The entinostat maximum plasma concentration (Cmax) and time to maximum haematologica | 2016; 101(8)

Entinostat in relapsed/refractory Hodgkin lymphoma

Figure 1. Waterfall plot of 38 evaluable patients (per protocol population) treated with entinostat. Schedule A (green) is oral entinostat at 10 or 15 mg given once every other week in a 4-week cycle. Schedule B (blue) is oral entinostat at 15 mg given once weekly three out of four weeks. Patients who have undergone autologous stem cell transplant (ASCT) and prior therapy (n=23) with brentuximab vedotin (n=4) are indicated. Approximately 40% of patients with bulky disease demonstrated tumor decrease but did not meet partial response (PR) criteria. SD: stable disease; PD: progressive disease.

Table 2. Treatment-related grade 3/4 adverse events occurring in more than 4% of patients.

Thrombocytopenia Anemia Neutropenia Leukopenia Hypokalemia Hypophosphatemia

Schedule A (n=33)

Schedule B (n=16)

Total (n=49)

19 (58%) 15 (45%) 12 (36%) 5 (15%) 1 (3%) 2 (6%)

12 (75%) 8 (50%) 8 (50%) 0 (0%) 3 (19%) 1 (6%)

31 (63%) 23 (47%) 20 (41%) 5 (10%) 4 (8%) 3 (6%)

Results

vedotin. Eight (16%) patients were refractory to prior therapy and were never eligible for transplant. Twenty-eight (57%) patients were refractory on entry, with 8 (16%) having no response to last treatment and 13 (27%) experiencing relapse within three months of last treatment. Thirty-nine (80%) had bulky disease. Baseline characteristics were similar between patients treated in Schedule A and Schedule B.

Patients' characteristics

Safety and treatment administration

Forty-nine patients were enrolled between April 2009 and March 2011, 33 patients in Schedule A (10 or 15 mg on days 1 and 15) and 16 patients in Schedule B (15 mg on days 1, 8, and 15) (Table 1). Median age of patients was 33 years (range 19-73). Twenty-two (45%) patients had had 4 or more previous treatment regimens, and 39 (80%) patients had previously undergone one or more allogeneic or autologous hematopoietic stem cell transplants. Four patients had been previously treated with brentuximab

All 49 patients received at least one dose of entinostat and were monitored for toxicity. Mean number of cycles of entinostat therapy was 5.7 (range 2-55). All patients have discontinued therapy. In Schedule A, 24 (73%) patients discontinued due to PD and 3 (9%) due to AEs. In Schedule B, 6 (38%) patients discontinued due to PD and 2 (13%) due to AEs. Seven patients had disease progression prior to completing one cycle of therapy. Other reasons for discontinuation were: 8 (16%) due to an adminis-

plasma concentration (Tmax) were obtained from the entinostat concentration data in the subset of patients who participated in the PK portion of the study; Cmax and Tmax were analyzed using descriptive statistics. Syndax Pharmaceuticals, Inc. analyzed all data and provided access to primary clinical data to all authors.

haematologica | 2016; 101(8)

971

C.L. Batlevi et al.

Figure 2. Swim plot of patients treated with entinostat in the per-protocol population who have achieved a partial response (PR) or stable disease (SD) (n=24). Median duration of response for 6 patients achieving PR is 28.5 months (range >1 day to 39.9 months; note that >1 day represents a patient who achieved PR but did not have any subsequent disease assessments that could be used for the analysis. In this case the patient was censored having undergone transplantation shortly after the response assessment that showed PR). Median duration of response for 18 patients with stable disease was 6.1 months (range >1 month to 15 months).

trative decision (most commonly for intervening therapy), 3 (6%) due to withdrawal of consent, 1 (2%) due to protocol deviation, and 2 (4%) for other reasons. Of the 49 patients, 5 (10%) experienced a treatment-related AE that required entinostat to be permanently stopped. Two patients treated in Schedule A discontinued therapy; one patient was diagnosed with pericarditis and pericardial effusion, while another patient had thrombocytopenia. Three patients treated in Schedule B discontinued therapy, one each for pulmonary embolism, spinal cord compression, and respiratory failure. The most common grade 3 or 4 adverse events (AEs) were thrombocytopenia (31 patients, 63%), anemia (23 patients, 47%), neutropenia (20 patients, 41%), leukopenia (5 patients, 10%), hypokalemia (4 patients, 8%), hypophosphatemia (3 patients, 6%) (Table 2). Twentyfive (51%) patients had a dose decrease or dose delay. The majority (31%) of dose modifications were for hematologic toxicities, primarily neutropenia and thrombocytopenia. Fatigue was associated with dose modifications in 2 patients, one each in Schedule A and B. Overall, grade 3 or 4 non-hematologic toxicity did not exceed 10% in any system. Twelve patients developed 26 serious adverse events (SAEs), 9 of 33 (27%) patients in Schedule A and 3 of 16 (19%) in Schedule B; multiple events were reported in these 12 patients. Treatment-related SAEs occurred in 6 patients, one each with: fever; pericarditis/pericardial effusion; renal calculi and subdural hemorrhage; dehydration; thrombocytopenia, anemia, neutropenia; and pulmonary embolism. The patient who developed pericarditis and pericardial effusion had been heavily pre-treated with 5 972

prior regimens, including radiation to the mediastinum; the event occurred after 12 cycles and is considered to be possibly related to entinostat. One patient on Schedule B developed a fatal respiratory failure unrelated to entinostat.

Efficacy In the intention-to-treat analysis of efficacy, 38 of 49 patients, 27 in Schedule A and 11 in Schedule B, completed two cycles of therapy and completed radiological restaging prior to initiation of cycle 3. Eleven of 49 patients did not complete more than 1 cycle or did not undergo restaging at the required time point (7 PD, 1 AE, 2 withdrew consent, and 1 protocol violation). Six of the 49 patients (4 in Schedule A, 2 in Schedule B) treated with entinostat obtained a PR; therefore, the overall response was 12% (Table 3). Nineteen patients achieved SD with 6 patients having durable SD (defined as stable disease lasting >6 months). Disease control (CR, PR, and durable SD) was noted for 12 of 49 patients (24%): 9 of 33 patients (27%) in Schedule A and 3 of 16 patients (19%) in Schedule B. In 38 evaluable patients who completed at least two cycles of therapy, disease was controlled in 12 of 38 patients (32%) and overall response was seen in 6 of 38 patients (16%) (Table 3). Tumor reduction, ranging between 3% to 92% as measured from baseline, was observed in 22 of 38 (58%) patients in the per-protocol population and 49% of intent-to-treat population (Table 3 and Figure 1). All patients treated on Schedule B demonstrated tumor reduction. In the 24 patients with clinical benefit (CR, PR and SD), 19 patients (79%) demonstrated reduction in tumor size by two months with a maximum haematologica | 2016; 101(8)

Entinostat in relapsed/refractory Hodgkin lymphoma

Table 3. Best overall response in intent-to-treat and per-protocol populations.

Response Regimen A (n=33) CR PR SD (â&#x2030;Ľ6 mo) SD (<6 mo) Disease control (CR+PR+SDâ&#x2030;Ľ6 mo) Clinical benefit (CR+PR+SD) Tumor reduction (CR+PR +SD with tumor reduction) PD Not assessable PD < cycle 1 Other reasons

4 (12%) 5 (15%) 7 (21%) 9 (27%)

16 (48%)

11 (33%) 6 (18%) 5 (15%) 1 (3%)

Intent-to-treat population Regimen B Total (n=16) (n=49)

Regimen A (n=27)

Per-protocol population Regimen B (n=11) 2 (18%) 1 (9%) 5 (45%) 3 (27%)

6 (16%) 6 (16%) 12 (32%)* 12 (32%)*

2 (13%) 1 (6%) 6 (38%) 3 (19%)

6 (12%) 6 (12%) 13 (27%) 12 (24%)

9 (56%)

25 (51%)

16 (59%)

8 (73%)

24 (63%)

22 (49%)

14 (52%)

8 (73%)

22 (58%)

14 (29%) 11 (22%) 7 (14) 4 (8%)

11 (41%)

3 (27%)

14 (37%)

3 (19%) 5 (31%) 2 (12.5%) 3 (19%)

4 (15%) 5 (19%) 7 (26%) 9 (33%)

Total (n=38)

CR: complete response; PR: partial response; SD: stable disease; mo: months.*Seven patients with SD (per Cheson response criteria) at end of treatment: 5 discontinued due to physician's discretion, one due to serious adverse events, and one patient decision.

Figure 3. Kaplan-Meier estimates of progression-free survival (PFS) and overall survival (OS) in 38 evaluable patients (per-protocol population). (A) The median PFS is 5.5 months and median OS is 25.1 months. (B) PFS and OS for Schedule A and B. PFS is 3.8 months for Schedule A and 5.5 months for Schedule B.

response achieved after four months (Figures 1 and 2). Of 18 patients with SD, 6 patients (33%) are experiencing duration of responses lasting over six months. Two patients, both on Schedule A, have durable responses lasting longer than 32 months. Of the 6 patients with a PR, 2 patients proceeded to an allogeneic transplant and radiation therapy, respectively, for consolidation therapy, while the other patients continued on study until disease progression. With a median follow up of 27.9 months for Schedule A and 19.9 months for Schedule B, median PFS from among the 38 evaluable patients was 3.8 months [95% confidence interval (95%CI) 1.9, 6.2 months] for Schedule A and 12 months (95%CI: 2.2, 12.8 months) for Schedule B. Median OS was 24.6 months (95%CI: 22.1, not reached) for Schedule A and was not reached for Schedule B (Figure 3). Of the 30 patients who had previously undergone ASCT, 26 patients were evaluable for efficacy analysis haematologica | 2016; 101(8)

with a median OS of 62.5 months when measured from the date of ASCT (Online Supplementary Figure S1A). Overall response was observed in 4 of 26 (15.4%) with stable disease observed in an additional 13 (50%) patients (Online Supplementary Figure S1B).

Entinostat induced chemokine/cytokine variations and pharmacokinetics Changes in cytokine/chemokine levels were measured in 18 patients and TARC levels were measurable in 20 patients. Changes in TARC levels between day 1 to day 8 were measured in 20 patients and changes in TARC levels between day 1 and day 15 were measured in 18 patients (Online Supplementary Figure S2). The median TARC level on day 1 was 1647 pg/mL (range 259-8176 pg/mL) and reduced to 1312 pg/mL (range 275-5155 pg/mL) on day 8, 802 pg/mL (range 107-1830 pg/mL) on day 15. Comparison of within-patient changes from day 1 to day 8 and from day 1 to day 15 showed a significant reduction 973

C.L. Batlevi et al. Table 4. Summary of results of clinical trials with HDAC inhibitors in Hodgkin lymphoma.

Study

Phase

Drug

Isotype

Route

Schedule

ORR N (%)

Tumor reduction (evaluable patients)

Median PFS

Morschhauser et al., 201524

Abexinostat

Pan

Various BID dosing

3 (27%)

0 (0%)

3 (27%)

54% (6/11)

Not reported

Younes et al., 201116

Mocetinostat 1,2,3,11

85 or 110 mg 3x per week

14 (27%)

2 (4%)

12 (23%)

17 (33%)

81% (34/42)

10 months

Younes et al., 201215

Panobinostat

Pan

40 mg 3x per week

129

35 (27%)

5 (4%)

30 (23%)

71 (55%)

74% (89/120)

6.1 months

Kirschbaum et al., 201214

Vorinostat

Pan

1 (4%)

0 (0%)

1 (4%)

12 (48%)

Not reported

4.8 months

Current study

Entinostat 1,2,3,11

200 mg BID, 25 day 1-14 every 21 days 10 or 15 mg 49 once every other week, or 15 mg once weekly in 3 of 4 weeks

6 (12%)

0 (0%)

6 (12%)

58% (24/38)

5.5 months

N: number; ORR: overall response rate; CR: complete response; PR: partial response; SD: stable disease; PFS: progression-free survival; PO: oral administration; BID: twice daily.

in TARC levels in patients on entinostat therapy supporting an on target effect. The multiplex cytokine panel of 30 cytokines (including IL-2, IL-4-8, G-CSF, GM-CSF, RANTES, Eotaxin, EGF, HGF, VEGF, interferon alpha, interferon gamma and TNF-alpha) showed wide variability between days 1, 8 and 15, with a general reduction in cytokine levels that was not statistically significant and not associated with clinical outcome. Entinostat systemic exposure increased with increased dose of entinostat as measured in 12 of 13 patients assessed for entinostat concentrations. Nine patients treated with 10 mg entinostat on day 1 demonstrated a mean Cmax of 85.7 ng/mL (SD±78.2 ng/mL, range 3.47-222.4 ng/mL), and a mean Tmax of 0. 44 hours (SD±0.24 h, range 0.25-1 h. Treatment with 15 mg of entinostat resulted in high serum concentrations but clinically insignificant mean maximum time of systemic exposure. Three patients who received entinostat 15 mg on day 1 demonstrated mean Cmax of 173.6 ng/mL (SD± 221.0 ng/mL, range 32.55-428.2 ng/mL), and a mean Tmax of 0.33 hours (SD±0.14 h, range 0.25-0.5 h).

Discussion Despite high response rates seen with brentuximab vedotin and immune checkpoint antibodies, the observed responses are usually partial indicating the need for combination therapies to improve efficacy.4-6 Immune checkpoint therapies expand cytotoxic effector T cells which are only fully functional in the context of reduced immune suppressor cells. In a murine model system, entinostat enhanced the activity of immune checkpoint blockade through potent inhibition of growth and function of myeloid-derived suppressor cells (MDSCs).12 Combination of anti-PD-1, anti-CTLA-4 and epigenetic modifiers with DNA methyltransferase and HDAC inhibitors triggered complete regression of large orthotopic tumors. Addition of entinostat greatly reduced MDSCs directly improving cytotoxic effector T-cell 974

activity. The ability of HDAC inhibitors to modulate PD1 expression is of particular interest suggesting the full clinical potential of HDAC inhibitors has yet to be fully explored.12,13 The dependence of Hodgkin Reed Sternberg cells on the microenvironment for survival suggests that combination of immune checkpoint and HDAC inhibitors may be an effective independent strategy to modulate the tumor microenvironment. HDAC inhibitors have been studied in HL with overall response rates of 4%-27%.14-16 While the ORR with entinostat was modest (12%), entinostat provided clinical benefit in 51% of this heavily pre-treated population. Of the 19 patients with SD, the duration of stability was over six months in 32% of these patients. The longest duration of response was greater than four years (50 months). Several HDAC inhibitors have been studied in HL and the majority demonstrate modest overall response rates ranging from 4% to 27% (Table 414-16,24). Despite modest response rates, 49% of patients by intent-to-treat and 58% of evaluable patients had a reduction in tumor size (CR, PR and SD with negative tumor volume decreased from baseline) (Table 3). The median PFS of 5.5 months observed in this study is similar to HDAC inhibitors in HL (Table 4). While the median overall survival in this study was two years, the duration of clinical benefit in patients who previously underwent an ASCT is more pronounced, with a median OS of 5.2 years. This suggests that entinostat should be considered for mechanism-based combination therapy with agents such as immune checkpoint antibodies or brentuximab vedotin. In particular, recent publications show synergism between PD-1 blockade and HDAC inhibitor, possibly through the enhanced modulation of myeloid-derived suppressor cells, increased expression of PD-1 mediated by HDAC inhibitors, and alteration of the tumor immune microenvironment.12,13,25 Consistent with the growing experience of HDAC inhibitors in HL, entinostat demonstrates clinical activity with a well-tolerated clinical profile suitable for combination therapy. Entinostat appears well tolerated, with only haematologica | 2016; 101(8)

Entinostat in relapsed/refractory Hodgkin lymphoma

10% of patients discontinuing therapy compared to 16%24% observed with other HDAC inhibitors.14-16 However, similar to other HDAC inhibitors, dose reductions were necessary to mitigate hematologic toxicities. Dosing of entinostat every other week or weekly in 3 out of 4 weekly cycles showed similar toxicity. Analysis of PK data in the small number of patients for whom entinostat plasma concentrations were available was consistent with previously reported results, with a highly variable C1D1 Cmax and a rapid Tmax. The range of Cmax concentrations was consistent with biologically active concentrations. Biomarker analyses in this study were designed to confirm pre-clinical data demonstrating that entinostat could down-regulate T-helper 2-associated cytokines and growth factors while up-regulating T-helper 1-associated factors. The general trends observed in changes of cytokine levels, along with the reduction of TARC, suggest that the mechanism of action of entinostat may involve immunomodulatory effects that contribute to its anti-tumor effects. A previous study in advanced HL patients demonstrated that reduction in TARC with an HDAC inhibitor was associated with tumor response.16 Although the majority of patients experienced reduced

References 1. Diehl V, Franklin J, Pfreundschuh M, et al. Standard and increased-dose BEACOPP chemotherapy compared with COPPABVD for advanced Hodgkin's disease. N Engl J Med. 2003;348(24):2386-2395. 2. Schmitz N, Pfistner B, Sextro M, et al. Aggressive conventional chemotherapy compared with high-dose chemotherapy with autologous haemopoietic stem-cell transplantation for relapsed chemosensitive Hodgkin's disease: a randomised trial. Lancet. 2002;359(9323):2065-2071. 3. Arai S, Fanale M, DeVos S, et al. Defining a Hodgkin lymphoma population for novel therapeutics after relapse from autologous hematopoietic cell transplant. Leuk Lymphoma. 2013;54(11):2531-2533. 4. Moskowitz CH, Ribrag V, Michot J-M, et al. PD-1 Blockade with the Monoclonal Antibody Pembrolizumab (MK-3475) in Patients with Classical Hodgkin Lymphoma after Brentuximab Vedotin Failure: Preliminary Results from a Phase 1b Study (KEYNOTE-013), Study. KEYNOTE-013; ASH Meeting. 2014. 5. Ansell SM, Lesokhin AM, Borrello I, et al. PD-1 blockade with nivolumab in relapsed or refractory Hodgkin's lymphoma. N Engl J Med. 2015;372(4):311-319. 6. Younes A, Gopal AK, Smith SE, et al. Results of a pivotal phase II study of brentuximab vedotin for patients with relapsed or refractory Hodgkin's lymphoma. J Clin Oncol. 2012;30(18):2183-2189. 7. Gopal AK, Chen R, Smith SE, et al. Durable remissions in a pivotal phase 2 study of brentuximab vedotin in relapsed or refractory Hodgkin lymphoma. Blood. 2015; 125(8):1236-1243. 8. Bartlett NL, Chen R, Fanale MA, et al. Retreatment with brentuximab vedotin in patients with CD30-positive hematologic malignancies. J Hematol Oncol. 2014;7:24.

haematologica | 2016; 101(8)

TARC levels on addition of entinostat, no association with tumor response was observed in this study. We will continue to evaluate TARC and additional putative biomarkers in future studies. In conclusion, entinostat is well tolerated with demonstrable clinical activity in heavily pre-treated HL patients. The mild toxicity profile, mechanism of action and the potential synergism with immune checkpoint therapies support the further development of this therapy in combination with other novel agents, including PD-1 and PD-L1 targeted antibodies. Acknowledgments The drug was provided by Syndax Pharmaceuticals. CB, YK, GB, PL, AC, LG, DB, FH have no conflict of interest. RS, PO, SC, LK are employed at Syndax Pharmaceuticals. Funding AY with research funding from Johnson and Johnson, Novartis, Seattle Genetics, Curis; honoraria from Celgene, Bayer, Bristol Meyer Squibb, Sanofi-Aventis, Janssen, Takeda, Incyte. CB is funded by the Mortimer J. Lacher Foundation and the Lymphoma Research Foundation.

9. Buglio D, Georgakis GV, Hanabuchi S, et al. Vorinostat inhibits STAT6-mediated TH2 cytokine and TARC production and induces cell death in Hodgkin lymphoma cell lines. Blood. 2008;112(4):1424-1433. 10. Buglio D, Khaskhely NM, Voo KS, Martinez-Valdez H, Liu YJ, Younes A. HDAC11 plays an essential role in regulating OX40 ligand expression in Hodgkin lymphoma. Blood. 2011;117(10):29102917. 11. Sharpe AH, Wherry EJ, Ahmed R, Freeman GJ. The function of programmed cell death 1 and its ligands in regulating autoimmunity and infection. Nat Immunol. 2007; 8(3):239-245. 12. Kim K, Skora AD, Li Z, et al. Eradication of metastatic mouse cancers resistant to immune checkpoint blockade by suppression of myeloid-derived cells. Proc Natl Acad Sci USA. 2014;111(32):11774-11779. 13. Oki Y, Buglio D, Zhang J, et al. Immune regulatory effects of panobinostat in patients with Hodgkin lymphoma through modulation of serum cytokine levels and Tcell PD1 expression. Blood Cancer J. 2014;4:e236. 14. Kirschbaum MH, Goldman BH, Zain JM, et al. A phase 2 study of vorinostat for treatment of relapsed or refractory Hodgkin lymphoma: Southwest Oncology Group Study S0517. Leuk Lymphoma. 2012; 53(2):259-262. 15. Younes A, Sureda A, Ben-Yehuda D, et al. Panobinostat in patients with relapsed/refractory Hodgkin's lymphoma after autologous stem-cell transplantation: results of a phase II study. J Clin Oncol. 2012;30(18):2197-2203. 16. Younes A, Oki Y, Bociek RG, et al. Mocetinostat for relapsed classical Hodgkin's lymphoma: an open-label, single-arm, phase 2 trial. Lancet Oncol. 2011; 12(13):1222-1228. 17. Suzuki T, Ando T, Tsuchiya K, et al. Synthesis and histone deacetylase inhibito-

18.

19.

20.

21.

22.

23.

24.

25.

ry activity of new benzamide derivatives. J Med Chem. 1999;42(15):3001-3003. Saito A, Yamashita T, Mariko Y, et al. A synthetic inhibitor of histone deacetylase, MS-27-275, with marked in vivo antitumor activity against human tumors. Proc Natl Acad Sci USA. 1999;96(8):4592-4597. Gore L, Rothenberg ML, O'Bryant CL, et al. A phase I and pharmacokinetic study of the oral histone deacetylase inhibitor, MS-275, in patients with refractory solid tumors and lymphomas. Clin Cancer Res. 2008; 14(14):4517-4525. Jona A, Khaskhely N, Buglio D, et al. The histone deacetylase inhibitor entinostat (SNDX-275) induces apoptosis in Hodgkin lymphoma cells and synergizes with Bcl-2 family inhibitors. Exp Hematol. 2011; 39(10):1007-1017.e1. Knipstein J, Gore L. Entinostat for treatment of solid tumors and hematologic malignancies. Expert Opin Investig Drugs. 2011;20(10):1455-1467. Cheson BD, Pfistner B, Juweid ME, et al. Revised response criteria for malignant lymphoma. J Clin Oncol. 2007;25(5):579586. Zhao M, Rudek MA, Mnasakanyan A, Hartke C, Pili R, Baker SD. A liquid chromatography/tandem mass spectrometry assay to quantitate MS-275 in human plasma. J Pharm Biomed Anal. 2007;43(2):784787. Morschhauser F, Terriou L, Coiffier B, et al. Phase 1 study of the oral histone deacetylase inhibitor abexinostat in patients with Hodgkin lymphoma, nonHodgkin lymphoma, or chronic lymphocytic leukaemia. Invest New Drugs. 2015;33(2):423-431. Shen L, Ciesielski M, Ramakrishnan S, et al. Class I histone deacetylase inhibitor entinostat suppresses regulatory T cells and enhances immunotherapies in renal and prostate cancer models. PLoS One. 2012; 7(1):e30815.

975

ARTICLE EUROPEAN HEMATOLOGY ASSOCIATION

Non-Hodgkin Lymphoma

Ferrata Storti Foundation

Immune-checkpoint expression in Epstein-Barr virus positive and negative plasmablastic lymphoma: a clinical and pathological study in 82 patients

Camille Laurent,1,2 Bettina Fabiani,3 Catherine Do,4 Emmanuelle Tchernonog,5 Guillaume Cartron,5 Pauline Gravelle,1,2 Nadia Amara,1 Sandrine Malot,6,7 Maryknoll Mawanay Palisoc,8 Christiane Copie-Bergman,9 Alexandra Traverse Glehen,10 Marie-Christine Copin,11 Pierre Brousset,1,2 Stefania Pittaluga,8 Elaine S. Jaffe, and 8 Paul Coppo6,7,12,13

Département de Pathologie, Institut Universitaire du Cancer-Oncopole, Toulouse, France; 2INSERM, U.1037, Centre de Recherche en Cancérologie de Toulouse-Purpan, Toulouse, France; 3Département de Pathologie, AP-HP, Hôpital Saint-Antoine, Paris, France; 4Institute for Cancer Genetics, Columbia University, New York, NY, USA; 5Service d’Hematologie, Hôpital Gui de Chauliac-Saint Eloi, Montpellier, France; 6Service d’Hématologie, AP-HP, Hôpital Saint-Antoine, Paris, France; 7Centre de Référence des Microangiopathies Thrombotiques, AP-HP, Paris, France; 8Hematopathology Section, Laboratory of Pathology, National Cancer Institute, Bethesda, MD, USA; 9Département de Pathologie, AP-HP, Groupe Hospitalier Henri Mondor - Albert Chenevier, Créteil, France; 10Département de Pathologie, Centre Hospitalier Lyon-Sud, Lyon, France; 11 Département de Pathologie, Centre Hospitalier Lille, France; 12UPMC, Université Paris VI, France; and 13Inserm U1170, Institut Gustave Roussy, Villejuif, France 1

Haematologica 2016 Volume 101(8):976-984

ABSTRACT

Correspondence: laurent.c@chu-toulouse.fr

Received: January 27, 2016. Accepted: May 3, 2016. Pre-published: May 12, 2016. doi:10.3324/haematol.2016.141978

Check the online version for the most updated information on this article, online supplements, and information on authorship & disclosures: www.haematologica.org/content/101/8/976

976

lasmablastic lymphoma is a rare and aggressive diffuse large B-cell lymphoma commonly associated with Epstein-Barr virus co-infection that most often occurs in the context of human immunodeficiency virus infection. Therefore, its immune escape strategy may involve the upregulation of immune-checkpoint proteins allowing the tumor immune evasion. However, the expression of these molecules was poorly studied in this lymphoma. We have investigated 82 plasmablastic lymphoma cases of whom half were Epstein-Barr virus positive. Although they harbored similar pathological features, Epstein-Barr virus positive plasmablastic lymphomas showed a significant increase in MYC gene rearrangement and had a better 2-year event-free survival than Epstein-Barr virus negative cases (P=0.049). Immunostains for programmed cell death-1, programmed cell death-ligand 1, indole 2,3dioxygenase and dendritic cell specific C-type lectin showed a high or moderate expression by the microenvironment cells in 60%-72% of cases, whereas CD163 was expressed in almost all cases. Tumor cells also expressed programmed cell death-1 and its ligand in 22.5% and 5% of cases, respectively. Both Epstein-Barr virus positive and negative plasmablastic lymphomas exhibited a high immune-checkpoint score showing that it involves several pathways of immune escape. However, Epstein-Barr virus positive lymphomas exhibited a higher expression of programmed cell death-1 and its ligand in both malignant cells and microenvironment as compared to Epstein-Barr virus negative cases. In conclusion, plasmablastic lymphoma expresses immune-checkpoint proteins through both malignant cells and the tumor microenvironment. The expression of programmed cell death-1 and its ligand constitutes a strong rationale for testing monoclonal antibodies in this often chemoresistant disease. haematologica | 2016; 101(8)

PD1/PD-L1 expression in plasmablastic lymphoma

Introduction Plasmablastic lymphoma (PL) is a distinct entity of diffuse large B-cell lymphoma first described in 1997.1 It commonly occurs in the context of human immunodeficiency virus (HIV) infection or in association with other contexts of immunodeficiency such as autoimmune diseases, organ transplantation and in the elderly;2-5 few studies have identified PL in patients without immunodeficiency.6,7 Neoplastic cells resemble immunoblasts or plasmablasts with a B-cell terminal differentiation phenotype and they mostly have a non-germinal center (GC)-B subtype profile.2-8 These cells constitutively express the plasma cell antigen CD138 (syndecan-1) and often harbor immunoglobulin (Ig) chain restriction with no or weak expression of B-cell mature markers such as CD20, CD79a and PAX5.2,9 PL cells frequently express Epstein Barr virus (EBV) genome with type I latency, especially in HIV patients. On the other hand, EBV negative (EBV–) PL still remains poorly characterized, due in part to its rarity as this limits the assessment of distinctive clinicopathological features. In a recent study, Loghavi et al.6 analyzed 61 cases with treatment information available in 42. Among them, 17 were EBV– and had a worse event-free survival as compared to EBV+PL patients.6 Despite recent advances in lymphoma therapeutic strategies, EBV+PL, and to an even greater extent EBV–PL, still represent an aggressive lymphoma with adverse prognosis; therefore, novel therapy strategies are urgently needed. Cancer cells, including lymphoma cells, are able to escape surveillance from the immune system by co-opting physiological mechanisms such as the programmed cell death-1 (PD-1) receptor pathway. By expressing PD-L1 on the tumor cell surface and engaging PD-1-positive infiltrating lymphocytes, tumors utilizing the PD-1 pathway can, therefore, evade an immune response by providing critical inhibitory signals that down-regulate T-cell function in the context of antigen recognition.10-13 Besides the PD-1/PD-L1 axis, the recruitment of M2 monocytes characterized by the expression of CD163 antigen has been involved in the tumor immune escape mechanism, especially through the production of immunosuppressive molecules such as the dendritic cell (DC)-specific C-type lectin DC-SIGN and/or the indole 2,3-dioxygenase (IDO).10,14-19 Accordingly, a blockade of these immune checkpoints alone or in combination with chemotherapy could be envisaged as an attractive therapeutic strategy.20-24 To the best of our knowledge, immune checkpoint profile molecule distribution has not previously been evaluated in PL. Given that EBV infection of malignant lymphocytes could be associated with PD-L1 overexpression,21,22,25,26 we hypothesized that PL cells could express PD-1/PD-L1 proteins and could constitute a prime target for PD-1 blockade. We, therefore, investigated the PD-1/PD-L1 axis and immune checkpoint profiles in PL and correlated these features with clinical presentation, histological findings and EBV status.

lymphomas in France.27,28 The diagnosis of PL was based on histological criteria and the expression of plasma cell differentiation markers as described by the WHO classification.2 Patients with a prior diagnosis of plasma cell myeloma or with multiple bone lesions or other laboratory criteria supporting the diagnosis of myeloma were excluded from the study. The following data were collected: age, sex, disease location at presentation, an associated context of immunodeficiency (HIV infection, transplantation, autoimmune disease, immunosuppressive drugs), EBV serology, Ann Arbor staging, treatment and outcome, current status and date of last follow up.

Histological and immunohistochemical analysis All cases were centrally reviewed. Institutional ethical approval was obtained in compliance with the Declaration of Helsinki. Paraffin tissue sections were processed for routine histopathological examination. For immunohistochemical examination, 3 µmthick sections were tested using a Ventana Benchmark XT (Ventana, Tucson, AZ, USA).12,28 The antibodies used are detailed in Online Supplementary Table S1. Each case was scored as positive or negative for CD10, BCL6, MUM1, MYC protein, GCET1 and FOXP1 according to the cut-off points previously defined.29-31 The immune checkpoint score was based on the percentage of PD-1+ and PD-L1+ tumor cells (ranged from 0= <5% to 2= ≥30%) or the percentage of PD-1+, PD-L1+, IDO+, DC-SIGN+ or CD163+ immune cells (ranged from 0= <10% to 2= ≥30%) combined to the value corresponding to the staining intensity (ranged from 0= negative or low staining at + to 2= intense staining at +++ ); these were calculated by two pathologists (CL and BF), as recently published32 (Online Supplementary Table S2).The percentage threshold for positive PD-L1 staining in malignant cells used here is comparable to that already published.25,33,34

In situ hybridization for Epstein–Barr virus and FISH study for MYC, BCL2 and BCL6 rearrangement Epstein-Barr virus detection was performed by in situ hybridization using EBV-encoded RNA (EBER) probes (Ventana Medical Systems). Fluorescence in situ hybridization (FISH) studies were performed using break-apart FISH DNA probes for cMYC/8q24, BCL2/18q21, and BCL6/3q27 (probes Y5410, Y5407 and Y5408; Dako) and were analyzed using Pannoramic 250 Flash digital microscopes (3DHISTECH, Hungary).35

Statistical analysis Comparison of clinicopathological, immunological, and genetic features between EBV– and EBV+ patients was carried out using c2 test (or Fisher exact test when required). Event-free survival was determined from time of diagnosis until time of death, progression or last follow up. Survival curves were constructed by the Kaplan-Meier method. Survival distributions were compared with the log rank test. For co-variates with less than 20% of missing values and with a P-value<0.05 in the log rank test, Cox proportional hazards model was performed. Using a backward stepwise removal method, only significant covariates were kept in the final Cox model. Statistical significance was set at P=0.05. Analyses were performed using GraphPrism and STATA v.13 software.

Methods Results Patients We reviewed 82 cases of PL diagnosed between 2008 and 2014, collected from 42,145 samples registered in Lymphopath, a government-supported network of expert hematopathologists which conducts a systematic review of all newly diagnosed or suspected haematologica | 2016; 101(8)

Clinical features of EBV+ and EBV– plasmablastic lymphoma patients Clinical features of EBV+ and EBV– plasmablastic lymphoma patients are shown in Table 1. Mean age at diagno977

C. Laurent et al.

sis was 62 years (range 22-88 years) with a male:female (M:F) ratio of 62:20. Diagnosis was made on the basis of a biopsy of a lymph node (n=17) or from an extranodal site (n=65): oral cavity (n=26), digestive tract (n=16), bone (n=5), soft tissue (n=7), respiratory tract (n=3), a kidney graft (n=2), testis (n=2), pleura (n=1), breast (n=1), orbital cavity (n=1), ureter (n=1). Half of the patients had underlying immunosuppression: HIV infection (41% of cases), and immunosuppressive (IS) treatment for kidney transplantation (3% of cases) or for autoimmune disease (5% of cases). Forty-three percent of patients (n=25) had no recorded history of immune suppression. We excluded from these categories patients for whom HIV status was not available (n=13) and patients who were HIV-negative and for whom a possible context of IS treatment could not be assessed (n=10). Half of the patients had localized disease, i.e. Ann Arbor stage I (29%) or II (21%); the remaining patients had disseminated disease, i.e. stage III (7%) or IV (43%). Extra-nodal disease was particularly common; the oral cavity was the most frequently observed site (46% of EBV+PL patients). Conversely, only 18% of EBV–PL patients had oral cavity involvement (P=0.016); moreover, a higher proportion of males was observed in

EBV+PL cases versus EBV–PL cases (M:F ratio=32:7 vs. 26:12, respectively). EBV+PL patients tended to be more often HIV+ than EBV–PL patients (53% vs. 29%, respectively; P=0.05). No significant difference was observed with regard to age, bone marrow involvement and Ann Arbor stage between EBV+ and EBV–PL patients. Nine patients were not treated for PL. Treatment details were available for 44 patients. Forty-two of 53 patients (79%) were treated with a sequential polychemotherapy, in association with autologous hematopoietic stem cell transplantation in 2 cases. Two patients were treated with radiotherapy alone. Most patients were treated with CHOP (cyclophosphamide, adriamycin, vincristine and prednisone) or CHOP-like regimens (n=18), in association with rituximab in 7 cases (Table 1). According to Cheson 1999 criteria,36 complete response (CR) was observed in 54%, 22% of cases had a partial response (PR), and 24% had persistent disease or died early.

Histological, immunophenotypic, and FISH analysis in EBV+ and EBV– plasmablastic lymphoma patients All biopsies were diffusely infiltrated by a proliferation of large monomorphic predominantly plasmablast-like

Table 1. Clinical data in 82 plasmablastic lymphoma patients.

Age

<60 years > 60 years Female Male Nodal extranodal - oral - digestive tract - others

Sex Site of involvement

LDH increased Stage

1-2 3-4

Bone marrow or marrow aspiration infiltration HIV seropositivity Post-transplantation Other ID context No ID context known Therapy

Therapy response

None CHOP or CHOP like RCHOP or RCHOP like EPOCH Other chemotherapies* Autologous hematopoietic stem cell transplantation Radiotherapy alone None/failure Partial response Complete response

All PL patients (n=82)

EBV+ PL patients (n=39)

EBV- PL patients (n=38)

45% (37/82) 55% (45/82) 24% (20/82) 76% (62/82) 21% (17/82) 79% (65/82) 32% (26/82) 20% (16/82) 27% (23/82) 37% (13/35) 50% (22/44) 50% (22/44) 19% (7/37) 41% (28/69) 3% (2/59) 7% (4/59) 42% (25/59) 17% (9/53) 34% (18/53) 13% (7/53) 7 % (4/53) 21% (11/53)

54% (21/39) 46% (18/39) 18% (7/39) 82% (32/39) 10% (4/39) 90% (35/39) 46% (18/39) 21% (8/39) 23% (8/39) 44% (8/18) 54% (13/24) 46% (11/24) 12% (2/17) 53% (20/38) 6% (2/34) 3% (1/34) 32% (11/34) 19% (5/27) 22% (6/27) 26% (7/27) 7% (2/27) 18% (5/27)

37% (14/38) 63% (24/38) 32% (12/38) 68% (26/38) 34% (13/38) 66% (25/38) 18% (7/38) 16% (6/38) 32% (12/38) 29% (5/17) 45% (9/20) 55% (11/20) 25% (5/20) 29% (8/28) 0% (0/25) 12% (3/25) 56% (14/25) 15% (4/26) 46% (12/26) 0% (0/26) 8% (2/26) 23% (6/26)

4% (2/53)

4% (1/27)

4% (1/26)

4% (2/53) 24% (10/41) 22% (9/41) 54% (22/41)

4% (1/27) 9% (2/22) 27% (6/22) 64% (14/22)

4% (1/26) 42% (8/19) 16% (3/19) 42% (8/19)

P 0.134 0.165 0.016

0.358 0.545 0.416** 0.051 0.503** 0.302** 0.069 0.083**

0.049

PL: plasmablastic lymphoma; EBV: Epstein-Barr virus; LDH: lactate dehydrogenase; ID: immune disease; CHOP or CHOP-like: cyclophosphamide, adriamycin, vincristine and prednisone; R-CHOP or CHOP-like: rituximab, cyclophosphamide, adriamycin, vincristine and prednisone; EPOCH: etoposide, prednisone, vincristine, cyclophosphamide and doxorubicine. *Other chemotherapies: velcade, dexamethasone and methotrexate followed by rituximab (n=1); velcade, dexamethasone and liposomal doxorubicin (n=1); velcade and dexamethasone (n=4); rituximab associated with bendamustine (n=1); velcade, thalidomide and dexamethasone (n=2);VP16, mitoxantrone and ifosfamide (n=1); vincristine, adriamycine and dexamethasone (n=1). **P-values were calculated using c2 test or Fisher exact test when required.

978

haematologica | 2016; 101(8)

PD1/PD-L1 expression in plasmablastic lymphoma

cells, and sometimes immunoblasts with round to oval nuclei usually containing a single central prominent nucleolus and eccentrically located in an abundant basophilic cytoplasm. Mitotic figures and areas of necrosis were frequent. Mature plasma cells in the background were uncommon and were not intermingled with the large tumor cells. Tumor cells were strongly and uniformly positive for CD138/VS38 with Ig light chain restriction detected in nearly 90% of cases. Neoplastic cells were usually EMA positive, and were mostly negative or focally and weakly positive for the B-cell marker CD79a, CD20 and PAX5 (Table 2). Tumor cells strongly expressed MUM1 (96%) and 28% of cases expressed FOXP1; they were negative for GCET1 and showed weak positivity for BCL6 (20%), while only 10% of cases expressed CD10. In addition, one-third of cases were positive for CD30 (36%) and BCL2 (32%). All cases were negative for ALK and HHV8 ruling out the diagnosis of ALK+ large B-cell lymphoma and diffuse large B-cell lymphoma arising in multicentric Castleman disease.37 Only 10% of cases stained weakly for CD56. Cyclin D1 was negative in all 42 cases tested. Most cases (73%) were positive with anti-MYC protein antibody. FISH with the MYC break-apart probe was positive in 28% of cases tested (10 of 36). One case (1 of 31) showed a BCL6 rearrangement. No case was found rearranged for BCL2 (0 of 32). Notably, all cases that showed MYC rearrangement also had a strong expression of MYC protein in more than 80% of tumor cells. Moreover, 50% of cases with MYC rearrangement had BCL2 protein expression.

Half of the cases tested (39 of 77) expressed EBER in more than 90% of tumor cells. The morphological analysis of EBV+PL and EBV–PL cases showed similar features and harbored a similar phenotype. However, 43% of EBV+ PL tested (9 of 21) displayed an MYC rearrangement versus 6% in EBV– PL (1 of 15) (P=0.017).

Immune checkpoint expression in EBV-positive and EBV-negative plasmablastic lymphoma patients Plasmablastic lymphoma biopsies showed a distinct pattern of expression for PD-1, PD-L1, CD163, IDO and DCSIGN. Indeed, their microenvironment comprised a variable proportion of PD-1, PD-L1, CD163, IDO and DCSIGN immune cells (Figure 1A-F). Among the immune cells, PD-1 and PD-L1 were primarily expressed by lymphoid cells and macrophages, respectively. However, some PL cases contained tumor cells, which expressed PD-L1 and/or PD-1 (Figure 1G and H). Immune checkpoint (ICP) score of PD-1, PD-L1, CD163 IDO and DC-SIGN was visually inspected on both tumor cells and immune cells from 42 available PLs and were quantified in terms of percentage of stained cells and intensity of staining, as described in the Methods section (Online Supplementary Table S2). As shown in Table 3, PD-L1 staining in immune cells ranged from low PD-L1 score with negative or weak/focal expression of PDL1 (40%; n=15 of 40) to moderate or high PD-L1 score (60%; n=25 of 40). PD-1 staining ranged from low PD-1 score with negative or weak/focal expression (40%; n=13 of 32) to moderate or high PD-1 score (60%; n=19 of 32). Most of the PL cases were associated with a moderate or

Table 2. Immunophenotype and FISH rearrangement studies in Epstein-Barr virus (EBV)+ and EBV– plasmablastic lymphomas.

B-cell markers CD20 CD79a PAX5 CD138/VS38 Light chain restriction Kappa Lambda GC/non-GC markers CD10 BCL6 GCET1 MUM1 FOXP1 MYC protein CD30 EMA BCL2 CD56 Cyclin D1 HHV8 ALK Gene rearrangement MYC BCL2 BCL6

% of positive cases in all PL (number of positive cases among total cases tested)

% of positive cases in EBV+ PL (number of positive cases among total cases tested)

% of positive cases in EBV– PL (number of positive cases among total cases tested)

9% (7/80)* 62% (47/76)* 26% (14/53)* 100% (80/80) 88% (63/72) 44% (32/72) 43% (31/72)

8% (3/39) 59% (22/37) 35% (9/26) 100% (39/39) 86% (30/35) 49% (17/35) 37% (13/35)

8% (3/37) 69% (24/35) 15% (4/26) 100% (37/37) 88% (29/33) 39% (13/33) 48% (16/33)

1.000** 0.421 0.109 NC 0.792 0.446 0.345

10% (6/58) 20% (9/44) 0% (0/37) 96% (53/55) 28% (11/40) 73% (30/41) 36% (23/64)* 60% (28/47) 32% (9/28) 10% (5/51)* 0% (0/42) 0% (0/24) 0% (0/24)

17% (5/30) 14% (3/22) 0% (0/22) 96% (26/27) 27% (6/22) 82% (18/22) 34% (11/32) 48% (11/23) 21.5% (3/14) 7% (2/27) 0% (0/24) 0% (0/12) 0% (0/12)

4% (1/26) 29% (6/21) 0% (0/15) 96% (26/27) 29% (5/17) 63% (12/19) 37% (11/30) 74% (17/23) 43% (6/14) 13% (3/23) 0% (0/18) 0% (0/12) 0% (0/12)

0.200** 0.281** NC 1.000** 1.000** 0.179 0.851 0.070 0.225 0.651** NC NC NC

28% (10/36) 0% (0/32) 3% (1/31)

43% (9/21) 0% (0/19) 5% (1/19)

7% (1/15) 0% (0/13) 0% (0/12)

0.017 NC 1.000**

*Less than 30% of cells express the marker. **P-values were calculated using c2 test or Fisher exact test when required. NC: incalculable.

haematologica | 2016; 101(8)

979

C. Laurent et al.

high score of CD163+ histiocytic cell infiltrates (97%; n=34 of 35). Histiocytic/dendritic cells of most PL showed an expression of IDO and DC-SIGN molecules with moderate or high IDO and DC-SIGN scores in 69% (n= 24 of 35) and in 72% (n=26 of 36) of PL cases, respectively. In addition, only a few CD8+ T cells expressed cytotoxic markers such as Granzyme B (data not shown). Interestingly, PD-1 was also expressed by tumor cells in 5% of PL cases (n=2 of 40), which exhibited a high PD-1 score, whereas PD-L1 was positive in tumor cells in 22.5% of PL cases (n=9 of 40) showing a high PD-L1 score in 77% of cases (n=7 of 9). In one case, both PD-1 and PD-L1 were expressed in tumor cells (Patient #54). We next compared the distribution and the expression of PD-1, PD-L1, IDO, DC-SIGN and CD163 between EBV+ PL and EBV–PL samples (Figure 2). Mean rates of

980

PD-L1 and PD-1 expression in each group were significantly higher in the microenvironment of EBV+ PL than in EBV–PL (P=0.02 and P=0.03, respectively) (Figure 2A and B). The percentage of PD-L1+ immune cells per sample was nearly 2-fold higher in EBV+ PL than in EBV– PL. In contrast, EBV+PL and EBV–PL samples showed a similar rates of CD163+ cell staining (Figure 2C) and were similar for IDO and DC-SIGN expression in the PL microenvironment (Figure 2D and E). Interestingly, strong expression of PD-L1 in tumor cells was observed in the majority of EBV+PL cases (n=7 of 9) (P=0.01). Moreover, the majority of PL cases with high PD-L1 score in tumor cells were EBV+PL, whereas tumor cells in 2 EBV–PL cases showed only low (n=1) or moderate (n=1) PD-L1 score. Furthermore, the single case co-expressing PD-1 and PD-L1 was an EBV+PL (Patient #54).

Figure 1. Immunohistochemical analysis of immune checkpoints in plasmablastic lymphomas (PL). (A and B) Representative cases of PL stained with anti-PD-L1 showing distinct membranous staining in intra-tumoral macrophages. (C) Representative case of PL stained with anti-PD-1 showing predominantly membranous staining in immune lymphoid cells. Example of (D) CD163, (E) IDO and (F) DC-SIGN stainings high-lighting macrophages in the PL microenvironment (x200). Representative cases of PL showing tumor cells that are positive for (G) PD-L1 or (H) for PD-1 (x200).

haematologica | 2016; 101(8)

PD1/PD-L1 expression in plasmablastic lymphoma

Prognostic impact of EBV status in plasmablastic lymphoma patients Clinical outcome was available in 47 patients. After a median follow up of 10.5 months (range 1 week-80 months), 51% of patients died, 6% were alive with stable or progressive disease, and 43% were alive and in complete remission. In the whole cohort, 2-year event-free survival was 40.8% (95%CI: 24%-57%) (Figure 3A). The 2year event-free survival was significantly shorter for EBV–PL patients than for EBV+ PL patients (22% vs. 58%, respectively; P=0.049) (Figure 3B). Multivariate analysis confirmed EBV status as an independent prognostic factor

(Online Supplementary Table S3). In contrast, MYC rearrangement status and PD-1/PD-L1 overexpression (with cut off: ICP score ≥3) were not associated with survival (Online Supplementary Table S3).

Discussion We report here the largest known clinico-pathological analysis of PL cases arising in HIV+ and HIV– patients. In accordance with previous reports, most PL occurring in the context of immune deficiency (mainly an HIV infec-

Figure 2. Rates of infiltrating immune cells expressing immune escape markers in Epstein-Barr virus (EBV)+ versus EBV– plasmablastic lymphomas (PL). Immune checkpoint scores of (A) PD-L1, (B) PD-1, (C) CD163, (D) IDO and (E) DC-SIGN stainings in immune cell infiltrates of EBV+ and EBV– PL. *Significant differences between groups (P=0.02, P=0.03, P=0.46, P=0.24, respectively).

Table 3. Immune checkpoint score in Epstein-Barr virus (EBV)+ and EBV– plasmablastic lymphoma immune cells.

ICP: immune checkpoint.

haematologica | 2016; 101(8)

981

C. Laurent et al. A

Figure 3. Event-free survival in plasmablastic lymphoma (PL) patients. Kaplan-Meier curve showing event-free survival in all (A) PL patients and (B) Epstein-Barr virus (EBV)+ and EBV– PL patients. Event was defined as death or disease progression.

tion or immunosuppressive treatment for transplantation) were associated with EBV.5,38-40 Our series shows a lower frequency of EBV+PL than previously reported,5,6,41,42 which may be due, at least in part, to the smaller number of patients with an underlying immune deficiency. No differences in histological and immunophenotypic features were observed between EBV+ and EBV– PL. The distinction between EBV– PL and plasmacytoma could be challenging, but unlike PL, plasmacytoma usually arises in immunocompetent patients and is composed of mature plasma cells without cytological atypia. However, MYC rearrangement was observed significantly more often in EBV+PL than in EBV–PL, in agreement with previous reports.6,41,42 Notably, all cases harboring MYC rearrangement had strong expression of MYC protein, which was also observed in 66% of PL cases without MYC rearrangement. MYC rearrangement has been reported to be the commonest chromosomic alteration in PL and was initially proposed as an aggressive factor in PL behavior.9 However, consistent with other reports,38,42 we found that MYC rearrangement did not impact survival. In addition, all but one PL case did not have BCL6 rearrangement at the major breakpoint region; BCL2 rearrangement was also negative in all cases.9,41 Our study suggests that PL develops several patterns of immune escape by expressing a number of immune checkpoint markers. Indeed, we found that nearly all PL express PD-L1 and PD-1 in the immune infiltrate, and that one-quarter of them strongly express PD-L1 in tumor cells and in immune cells. We also show that the PD-1/PD-L1 axis is more over-expressed in the microenvironment in EBV+PL, which is typically associated with situations of immunodeficiency. These findings suggest that an antiviral response against EBV may favor the recruitment of immune cells PD-L1. In this regard, it has been shown that cytokines, such as interferon γ, can also potentially up-regulate PD-L1 on macrophages via the ISRE/IRF1 motif in the PD-L1 (CD274) promoter, and thus favor PD-L1 expression in immune cell infiltrates during inflammatory responses.10,43,44 Interestingly, all but 2 cases expressing PDL1 in tumor cells were associated with EBV infection, which could be consistent with the view that PD-L1 expression is promoted by LMP1 of EBV which increases 982

PD-L1 promoter expression.25,43 However, the majority of EBV+PL tumor cells expressing PD-L1 had an EBV latency type 1 and did not express LMP1 (n=7 of 9) (data not shown), suggesting that PD-L1 can be up-regulated by other mechanisms. In this regard, it has been shown that PD-L1 expression can be driven by intrinsic genetic aberrations of 9p24.1 (a genomic region including CD274, PDCD1LG2 and JAK2) and by the dysregulation of JAK/STAT pathway.43,45,46 As suggested for EBV+ CD20+ diffuse large B-cell lymphoma (DLBCL),25,47 further genetic analyses of PD-L1 (CD274) locus are needed to identify the intrinsic mechanism of PD-L1 upregulation by malignant cells in PL. In this study we found that PL patients (either associated or not with EBV) showed a high content of CD163+ macrophages, which are known to be associated with immunoregulatory function by promoting Th2 immune responses to the detriment of Th1 anti-tumor immune responses. Furthermore, PLs express IDO and DC-SIGN molecules in a very similar manner in the histiocytic/dendritic cell microenvironment. These immune evasion proteins have been known to be involved in negative immunoregulatory response. For instance, IDO attenuates T-cell clonal expansion, and induces anergy and apoptosis on effector T cells. Nevertheless, reported data differ as to the prognostic impact of PD-1/PD-L1 and IDO expression in lymphoma.16,17,48,49 In spite of this, in our cohort, we did not find any correlation between immune checkpoint expression and PL patients’ survival, and further biological studies on larger cohorts are needed to really evaluate the impact of the immune escape checkpoint on PL prognosis. Altogether, given their association with oncogenic viruses and immunodeficiency, PL should provide attractive targets for immune-based therapy. Finally, this study confirms the prognostic value of EBV status in PL.6 Indeed, EBV+PL had a better event-free survival, which nevertheless contrasts with the aggressive clinical course of EBV+DLBCL patients.50 In conclusion, PL expresses immune checkpoint proteins as PD-1/PD-L1 in both the microenvironment and in the malignant cells, particularly in EBV+PL. Anti-PD-1 monoclonal antibodies recently received FDA approval for advanced melanoma or non-small cell lung carcinoma and haematologica | 2016; 101(8)

PD1/PD-L1 expression in plasmablastic lymphoma

promising data on therapeutic response were also seen in Hodgkin lymphoma.20-24 Therefore, our findings constitute a strong rationale for testing anti-PD-1 monoclonal antibodies in the treatment of PL, a severe and often chemoresistant form of lymphoid malignancy. Acknowledgments We thank Laurence Jalabert, Audray Benest and Gabrielle Perez for immunohistochemistry staining and FISH studies (IUCT, Toulouse, France) and Marina Bousquet for reviewing the paper. We thank François-Xavier Frenois for whole slide imaging (IUCT Toulouse, France). We thank the Lymphopath consortium for sending their samples: Pr Gaulard (Hôpital Henri Mondor, Créteil, France); Pr J.F. Fléjou (Hôpital Saint Antoine, Paris, France); Pr V. Costes Martineau (CHU de Montpellier, France); Pr A. de Mascarel et Dr M. Parrens (CHU de Bordeaux, France); Dr I. Soubeyran (Institut Bergonié, Bordeaux, France); Dr C Chassagne-Clement (Centre Léon

References 1. Delecluse HJ, Anagnostopoulos I, Dallenbach F, et al. Plasmablastic lymphomas of the oral cavity: a new entity associated with the human immunodeficiency virus infection. Blood. 1997;89(4):1413-1420. 2. Swerdlow SH, Campo E, Harris NL, et al. World Health Organization Classification of Tumours of Haematopoietic and Lymphoid Tissues (4th ed). Lyon, France, IARC Press, 2008. 3. Rafaniello Raviele P, Pruneri G, Maiorano E. Plasmablastic lymphoma: a review. Oral Dis. 2009;15(1):38-45. 4. Colomo L, Loong F, Rives S, et al. Diffuse large B-cell lymphomas with plasmablastic differentiation represent a heterogeneous group of disease entities. Am J Surg Pathol. 2004;28(6):736-747. 5. Morscio J, Dierickx D, Nijs J, et al. Clinicopathologic comparison of plasmablastic lymphoma in HIV-positive, immunocompetent, and posttransplant patients: single-center series of 25 cases and meta-analysis of 277 reported cases. Am J Surg Pathol. 2014;38(7):875-886. 6. Loghavi S, Alayed K, Aladily TN, et al. Stage, age, and EBV status impact outcomes of plasmablastic lymphoma patients: a clinicopathologic analysis of 61 patients. J Hematol Oncol. 2015;8:65. 7. Carbone A, Gloghini A. Plasmablastic lymphoma: one or more entities? Am J Hematol. 2008;83(10):763-764. 8. Covens K, Verbinnen B, Geukens N, et al. Characterization of proposed human B-1 cells reveals pre-plasmablast phenotype. Blood. 2013;121(26):5176-5183. 9. Bogusz AM, Seegmiller AC, Garcia R, Shang P, Ashfaq R, Chen W. Plasmablastic lymphomas with MYC/IgH rearrangement: report of three cases and review of the literature. Am J Clin Pathol. 2009; 132(4):597-605. 10. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer. 2012;12(4):252-264. 11. Andorsky DJ, Yamada RE, Said J, Pinkus GS, Betting DJ, Timmerman JM. Programmed death ligand 1 is expressed by

haematologica | 2016; 101(8)

12.

13. 14.

15. 16.

17.

18.

19.

20.

21. 22.

23. 24.

Bérard, Lyon, France); Dr L. Xerri (Institut Paoli Calmettes, Marseille, France); Dr A Moreau (CHU de Nantes, France); Dr F. Arbion (CHU de Tours, France); Dr I. Quintin-Roué (CHU de Brest, France); Dr J.M Picquenot (Centre Henri Becquerel, Rouen France); Pr L. Martin (CHU de Besançon-Dijon, France); Pr J.M. Vignaud and Dr C. Bastien (CHU de Nancy, France); Pr F. Labrousse (CHU de Limoges, France). Funding This work was supported in part by institutional grants from the Ligue Regionale de Lutte Contre le Cancer (Canceropole Grand Sud Ouest), the Institut National du Cancer (INCA), the Association de Recherche Contre le Cancer (ARC) (subvention PJA 20131200091), the Laboratoire d'Excellence Toulouse Cancer (TOUCAN) (contract ANR11-LABX), the Programme HospitaloUniversitaire en Cancérologie CAPTOR (contract ANR11PHUC0001), and the Institut Carnot Lymphome (CALYM). PB is supported by the Institut Universitaire de France.

non-hodgkin lymphomas and inhibits the activity of tumor-associated T cells. Clin Cancer Res. 2011;17(13):4232-4244. Laurent C, Charmpi K, Gravelle P, et al. Several immune escape patterns in nonHodgkin's lymphomas. Oncoimmunology. 2015;4(8):e1026530. Armand P. Checkpoint blockade in lymphoma. Hematology Am Soc Hematol Educ Program. 2015;2015(1):69-73. Mantovani A, Sica A. Macrophages, innate immunity and cancer: balance, tolerance, and diversity. Curr Opin Immunol. 2010; 22(2):231-237. Sica A, Mantovani A. Macrophage plasticity and polarization: in vivo veritas. J Clin Invest. 2012;122(3):787-795. Choe JY, Yun JY, Jeon YK, et al. Indoleamine 2,3-dioxygenase (IDO) is frequently expressed in stromal cells of Hodgkin lymphoma and is associated with adverse clinical features: a retrospective cohort study. BMC Cancer. 2014;14:335. Taylor MW, Feng GS. Relationship between interferon-gamma, indoleamine 2,3-dioxygenase, and tryptophan catabolism. FASEB J. 1991;5(11):2516-2522. Kim R, Emi M, Tanabe K. Cancer immunoediting from immune surveillance to immune escape. Immunology. 2007; 121(1):1-14. Gajewski TF, Schreiber H, Fu YX. Innate and adaptive immune cells in the tumor microenvironment. Nat Immunol. 2013; 14(10):1014-1022. Westin JR, Chu F, Zhang M, et al. Safety and activity of PD1 blockade by pidilizumab in combination with rituximab in patients with relapsed follicular lymphoma: a single group, open-label, phase 2 trial. Lancet Oncol. 2014;15(1):69-77. Ansell SM. Targeting immune checkpoints in lymphoma. Curr Opin Hematol. 2015; 22(4):337-342. Ansell SM, Lesokhin AM, Borrello I, et al. PD-1 blockade with nivolumab in relapsed or refractory Hodgkin's lymphoma. N Engl J Med. 2015;372(4):311-319. Novakovic BJ. Checkpoint inhibitors in Hodgkin lymphoma. Eur J Haematol. 2016; 96(4):335-343. Cheah CY, Fowler NH, Neelapu SS.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

Targeting the programmed death-1/programmed death-ligand 1 axis in lymphoma. Curr Opin Oncol. 2015;27(5):384-391. Chen BJ, Chapuy B, Ouyang J, et al. PD-L1 expression is characteristic of a subset of aggressive B-cell lymphomas and virusassociated malignancies. Clin Cancer Res. 2013;19(13):3462-3473. Twa DD, Chan FC, Ben-Neriah S, et al. Genomic rearrangements involving programmed death ligands are recurrent in primary mediastinal large B-cell lymphoma. Blood. 2014;123(13):2062-2065. de Leval L, Parrens M, Le Bras F, et al. Angioimmunoblastic T-cell lymphoma is the most common T-cell lymphoma in two distinct French information data sets. Haematologica. 2015;100(9):e361-364. Laurent C, Delas A, Gaulard P, et al. Breast implant-associated anaplastic large cell lymphoma: two distinct clinicopathological variants with different outcomes. Ann Oncol. 2016;27(2):306-314. Hans CP, Weisenburger DD, Greiner TC, et al. Confirmation of the molecular classification of diffuse large B-cell lymphoma by immunohistochemistry using a tissue microarray. Blood. 2004;103(1):275-282. Green TM, Young KH, Visco C, et al. Immunohistochemical double-hit score is a strong predictor of outcome in patients with diffuse large B-cell lymphoma treated with rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone. J Clin Oncol. 2012;30(28):3460-3467. Meyer PN, Fu K, Greiner TC, et al. Immunohistochemical methods for predicting cell of origin and survival in patients with diffuse large B-cell lymphoma treated with rituximab. J Clin Oncol. 2011;29(2): 200-207. Laurent C, Muller S, Do C, et al. Distribution, function, and prognostic value of cytotoxic T lymphocytes in follicular lymphoma: a 3-D tissue-imaging study. Blood. 2011;118(20):5371-5379. Topalian SL, Hodi FS, Brahmer JR, et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med. 2012;366(26):2443-2454. Taube JM, Anders RA, Young GD, et al. Colocalization of inflammatory response

983

C. Laurent et al.

35.

36.

37.

38.

39.

984

with B7-h1 expression in human melanocytic lesions supports an adaptive resistance mechanism of immune escape. Sci Transl Med. 2012;4(127):127ra37. Laurent C, Guerin M, Frenois FX, et al. Whole-slide imaging is a robust alternative to traditional fluorescent microscopy for fluorescence in situ hybridization imaging using break-apart DNA probes. Hum Pathol. 2013;44(8):1544-1555. Cheson BD, Horning SJ, Coiffier B, et al. Report of an international workshop to standardize response criteria for nonHodgkin's lymphomas. NCI Sponsored International Working Group. J Clin Oncol. 1999;17(4):1244. Hsi ED, Lorsbach RB, Fend F, Dogan A. Plasmablastic lymphoma and related disorders. Am J Clin Pathol. 2011;136(2):183194. Castillo JJ, Furman M, Beltran BE, et al. Human immunodeficiency virus-associated plasmablastic lymphoma: poor prognosis in the era of highly active antiretroviral therapy. Cancer. 2012;118(21):5270-5277. Cattaneo C, Re A, Ungari M, et al. Plasmablastic lymphoma among human immunodeficiency virus-positive patients: results of a single center's experience. Leuk Lymphoma. 2015;56(1):267-269.

40. Guan B, Zhang X, Hu W, et al. Plasmablastic lymphoma of the oral cavity in an HIV-negative patient. Ann Diagn Pathol. 2011;15(6):436-440. 41. Boy SC, van Heerden MB, Babb C, van Heerden WF, Willem P. Dominant genetic aberrations and coexistent EBV infection in HIV-related oral plasmablastic lymphomas. Oral Oncol. 2011;47(9):883-887. 42. Valera A, Balague O, Colomo L, et al. IG/MYC rearrangements are the main cytogenetic alteration in plasmablastic lymphomas. Am J Surg Pathol. 2010;34(11):1686-1694. 43. Green MR, Monti S, Rodig SJ, et al. Integrative analysis reveals selective 9p24.1 amplification, increased PD-1 ligand expression, and further induction via JAK2 in nodular sclerosing Hodgkin lymphoma and primary mediastinal large Bcell lymphoma. Blood. 2010;116(17):32683277. 44. Zou W. Immunosuppressive networks in the tumour environment and their therapeutic relevance. Nat Rev Cancer. 2005;5(4):263-274. 45. Green MR, Rodig S, Juszczynski P, et al. Constitutive AP-1 activity and EBV infection induce PD-L1 in Hodgkin lymphomas and posttransplant lymphoproliferative dis-

46.

47.

48.

49.

50.

orders: implications for targeted therapy. Clin Cancer Res. 2012;18(6):1611-1618. Marzec M, Zhang Q, Goradia A, et al. Oncogenic kinase NPM/ALK induces through STAT3 expression of immunosuppressive protein CD274 (PD-L1, B7-H1). Proc Natl Acad Sci USA. 2008; 105(52): 20852-20857. Nicolae A, Pittaluga S, Abdullah S, et al. EBV-positive large B-cell lymphomas in young patients: a nodal lymphoma with evidence for a tolerogenic immune environment. Blood. 2015;126(7):863-872. Godin-Ethier J, Hanafi LA, Piccirillo CA, Lapointe R. Indoleamine 2,3-dioxygenase expression in human cancers: clinical and immunologic perspectives. Clin Cancer Res. 2011;17(22):6985-6991. Kiyasu J, Miyoshi H, Hirata A, et al. Expression of programmed cell death ligand 1 is associated with poor overall survival in patients with diffuse large B-cell lymphoma. Blood. 2015;126(19):21932201. Oyama T, Yamamoto K, Asano N, et al. Age-related EBV-associated B-cell lymphoproliferative disorders constitute a distinct clinicopathologic group: a study of 96 patients. Clin Cancer Res. 2007;13(17): 5124-5132.

haematologica | 2016; 101(8)

ARTICLE

Cell Therapy & Immunotherapy

Mesenchymal stromal cells from pooled mononuclear cells of multiple bone marrow donors as rescue therapy in pediatric severe steroid-refractory graft-versus-host disease: a multicenter survey

EUROPEAN HEMATOLOGY ASSOCIATION

Ferrata Storti Foundation

Zyrafete Kuçi,1* Halvard Bönig,2* Hermann Kreyenberg,1 Milica Bunos,2 Anna Jauch,3 Johannes W.G. Janssen,3 Marijana Škifić,1,4 Kristina Michel,1 Ben Eising,1 Giovanna Lucchini,1,5 Shahrzad Bakhtiar,1 Johann Greil,6 Peter Lang,7 Oliver Basu,8 Irene von Luettichau,9 Ansgar Schulz,10 Karl-Walter Sykora,11 Andrea Jarisch,1 Jan Soerensen,1 Emilia Salzmann-Manrique,1 Erhard Seifried,2 Thomas Klingebiel,1 Peter Bader1* and Selim Kuçi1*

University Hospital Frankfurt, Department for Children and Adolescents, Division for Stem Cell Transplantation and Immunology, Frankfurt am Main, Germany; 2Institute of Transfusion Medicine and German Red Cross Blood Center Frankfurt, Frankfurt am Main, Germany; 3Institute of Human Genetics, University of Heidelberg, Germany; 4 University Hospital Centre Zagreb, Clinical Department of Transfusion and Transplantation Biology, Division of Cellular Therapy, Zagreb, Croatia; 5Great Ormond Street Hospital, Department of Hematology/Oncology, London, United Kingdom; 6 University Children’s Hospital Heidelberg, Germany; 7University Children’s Hospital Tübingen, Germany; 8University Children’s Hospital Essen, Germany; 9University Children’s Hospital München-Schwabing, Germany; 10University Children’s Hospital Ulm, Germany; and 11Children’s Hospital, Medizinische Hochschule Hannover, Germany 1

*ZK, HB, PB and SK contributed equally to this work

Haematologica 2016 Volume 101(8):985-994

ABSTRACT

o circumvent donor-to-donor heterogeneity which may lead to inconsistent results after treatment of acute graft-versus-host disease with mesenchymal stromal cells generated from single donors we developed a novel approach by generating these cells from pooled bone marrow mononuclear cells of 8 healthy “3rd-party” donors. Generated cells were frozen in 209 vials and designated as mesenchymal stromal cell bank. These vials served as a source for generation of clinical grade mesenchymal stromal cell end-products, which exhibited typical mesenchymal stromal cell phenotype, trilineage differentiation potential and at later passages expressed replicative senescence-related markers (p21 and p16). Genetic analysis demonstrated their genomic stability (normal karyotype and a diploid pattern). Importantly, clinical end-products exerted a significantly higher allosuppressive potential than the mean allosuppressive potential of mesenchymal stromal cells generated from the same donors individually. Administration of 81 mesenchymal stromal cell end-products to 26 patients with severe steroidresistant acute graft-versus-host disease in 7 stem cell transplant centers who were refractory to many lines of treatment, induced a 77% overall response at the primary end point (day 28). Remarkably, although the cohort of patients was highly challenging (96% grade III/IV and only 4% grade II graft-versus-host disease), after treatment with mesenchymal stromal cell end-products the overall survival rate at two years follow up was 71±11% for the entire patient cohort, compared to 51.4±9.0% in graft-versus-host disease clinical studies, in which mesenchymal stromal cells were derived from single donors. Mesenchymal stromal cell end-products may, therefore, provide a novel therapeutic tool for the effective treatment of severe acute graft-versus-host disease. haematologica | 2016; 101(8)

Correspondence: selim.kuci@kgu.de/peter.bader@kgu.de

Received: December 2, 2015. Accepted: May 4, 2016. Pre-published: May 12, 2016. doi:10.3324/haematol.2015.140368

Check the online version for the most updated information on this article, online supplements, and information on authorship & disclosures: www.haematologica.org/content/101/8/985

985

Kuç

Z. Kuçi et al.

Introduction Since the first clinical trial of mesenchymal stromal cells (MSC) in 1995,1 their use has expanded rapidly. To date, 561 registered clinical trials (retrieved from www.clinicaltrials.gov, 2nd December 2015) have been performed to examine an extremely wide spectrum of therapeutic MSC applications. Despite the general consensus that MSCs appear to be welltolerated, safe and effective for the treatment of various diseases, there has been limited progress in this field due to inconsistencies in the outcome of clinical trials. These inconsistencies may be attributed to the lack of a standardized methodology for MSC generation2 and MSC dosing, the heterogeneity in MSC potency between donors3 and tissue sources,4 and the variable number of MSC progenitor cells between tissue samples.5 MSCs exhibit donor-specific variations in their immunosuppressive properties not only at the donor level,3,6 but also at the clonal level.7 A recent study demonstrated passage effect on the immunosuppressive effect of MSCs by obtaining an optimal immunosuppressive effect in patients with steroid-resistant acute graft-versus-host disease (aGvHD) after administration of MSCs at passage 1 or 2.8 In contrast, several other reports demonstrated that the immunosuppressive effect of MSCs remains unchanged for up to 7 or 8 passages in culture.9,10 Another important issue regarding the clinical application of MSCs is their culture under serum-free conditions. The majority of clinical studies have used MSCs that were expanded in media supplemented with fetal bovine serum (FBS).1,11-15 To avoid the risks associated with the use of FBS,16 platelet lysate (PL) was proposed as a supplement to tissue culture media for MSCs.17 Recently, several studies showed that MSCs that were expanded in PL exhibited the same efficacy as MSCs cultured in serum-containing media for the treatment of GvHD.18-22 To date, clinical studies have used MSCs that have been generated from several individual donors. Considering the aforementioned inter-donor heterogeneity and the need for a large number of “off-the-shelf” MSCs, the establishment of MSC banks appears to be an indispensable strategy for providing a continuous supply of MSCs with predictable potency. To our knowledge, there are few established MSC banks worldwide, and these MSC banks were generated by separately isolating, expanding, and freezing MSCs from up to 10 donors in FBS-containing media.23-26 In the current study, we report for the first time the establishment of a serum-free and GMP-compliant MSC bank generated from pooled bone marrow mononuclear cells (BM-MNCs) of multiple donors as a novel strategy to circumvent donor-to-donor variability. Clinical-grade MSC endproducts (MEPs) derived from the MSC bank were thoroughly assessed for their proliferation, differentiation, and, in particular, for the allosuppressive potential in vitro. Importantly, 81 MEPs were administered as a rescue therapy to 26 pediatric patients with severe steroid-refractory aGvHD in seven transplantation centers. Safety and efficacy of MEPs was compared to MSCs generated from a single or several individual donors that have been used in the GvHDclinical studies reported thus far.

Methods Generation of MSC bank and clinical-grade MEPs Bone marrow was collected from 8 healthy volunteers (age 2145 years old) after written informed consent and after the approval 986

of the local Ethics Committee (n. 275/09). BM-MNCs were enriched from the bone marrow aspirate by using the Sepax II NeatCell process (Biosafe, Eysins, Switzerland) and frozen individually. After thawing and washing these BM-MNCs were pooled. This pool of BM-MNCs from 8 donors was used to generate MSCs over 14 days in culture. After their detachment, passage 1 mesenchymal stromal cells (MSC-P1) were washed and aliquoted into 209 cryovials (each containing 1.5x106 MSC-P1). Cryopreserved vials with MSC-P1 were referred to as the MSC bank. To generate clinical-grade MEPs, MSC-P1 aliquots from the MSC bank were thawed and after washing they were expanded in medium containing 10% PL till the end of passage 2. These MSCs were re-suspended in cryomedium (0.9% NaCl containing 5% HSA and 10% DMSO), distributed in cryobags (each containing 1-3x106 MSCs/mL in 45 mL of cryomedium) and frozen in liquid nitrogen until use. Further details on methods as to generation of MSC bank from BM-MNC pool, including collection and testing of platelet lysates, validation of MEPs regarding their phenotype (flow cytometry), differentiation and allosuppressive potential, genetic analysis (cytogenetics, FISH, RT-PCR, STR-PCR), determination of senescence and its typical markers, characteristics of patients with severe aGvHD, assessment of the disease and its treatment with MEPs and statistics, are presented in the Online Supplementary Appendix.

Results Collection of bone marrow from 8 healthy 3rd-party donors and isolation of BM-MNCs After obtaining written informed consent, healthy donors donated 152-184 mL of bone marrow (Online Supplementary Table S1). Following isolation of the BM-MNCs using the Sepax method, a total of 10.82x109 BM-MNCs were collected from 8 donors. The absolute number of BM-MNCs per 1 mL of bone marrow after isolation was 4.1x106±7.8x105. All donors were equally represented in the BM-MNC pool, i.e. the relative contribution of BM-MNCs by each donor was 12.5±2.7% (Online Supplementary Table S1). The cells from each donor were re-suspended in cryomedium and frozen individually in bags. The total number of frozen BM-MNCs was 9.86x109.

Testing of the concentration of PL for the generation and expansion of MSC Before the establishment of the MSC bank, we determined the optimal concentration of PL to support the adherence of MSC progenitors and to assess whether PL filtration was needed. We observed that unfiltered PL at a concentration of 10% (Online Supplementary Figure S1A) or 5% (Online Supplementary Figure S1B) was optimal for MSC generation compared with media supplemented with filtered PL at the same concentrations (P<0.002 and P<0.01, respectively). In addition, 10% PL was significantly more efficient for MSC generation by plastic adherence (P<0.02) (Online Supplementary Figure S1C). Evaluation of the capacity of PL to expand MSC revealed that unfiltered PL at a concentration of 10% (Online Supplementary Figure S1D) or 5% (Online Supplementary Figure S1E) induced significantly greater expansion of MSC than filtered PL at the corresponding concentration (P<0.0008 and P<0.003, respectively). Furthermore, unfilhaematologica | 2016; 101(8)

MSCs from pooled BM-MNCs of 8 donors for aGvHD

tered PL at 10% was significantly more effective for expansion of MSC than 5% unfiltered PL (P<0.0001) (Online Supplementary Figure S1F).

Establishment of the MSC bank and generation of clinical-scale MEP After thawing, the number of pooled BM-MNCs was 2.8x109. Culture of these cells for 14 days resulted in 320x106 primary MSC which exhibited a viability of 98.9% and expressed consensus markers for MSC. These cells were designated as MSC-P1 (passage 1) and were aliquoted and frozen in 209 cryovials, representing the MSC bank (Online Supplementary Figure S2Bi-iii). To generate clinical-scale MEP and validate their proliferative, differentiation and allosuppressive potential, three randomly selected MSC aliquots from the MSC bank were thawed after cryopreservation for 6-8 weeks. The mean cell recovery was 1.39x106 (range 1.23-1.48x106) viable cells/vial, and the viability of these cells was 95.25±1.73% (range 93.45%-96.9%). On average, the expansion of these MSC over two weeks until the end of passage 2 resulted in the generation of 4.7x108 viable MSC

(range 4.2-5.48x108), in which the number of cumulative population doublings (CPD) was 8.5±0.04. These samples (units), referred to as clinical-scale MEP, were frozen in 4-7 bags containing 50-129x106 MSC until use. These units served as clinical doses for recipients of various body weights (Online Supplementary Figure S2Ci-iii). Before freezing, the MEP phenotypes were analyzed to assess whether the samples fulfilled the International Society for Cellular Therapy (ISCT) MSC criteria. They were sterile, were mycoplasma-negative and contained endotoxin levels below the limit of detection (<0.2 EU/mL). For quality control and release purposes, all bags generated from one aliquot were considered as one batch.

Validation of the MEP: phenotype, allosuppressive and differentiation potential For quality assessment, the MEP from three different batches were thawed after 4-6 weeks of cryopreservation. After thawing, 87.9±3.6% of the cells were viable, representing a mean recovery of 78.5±9.8%. MEP were negative for HLA-DR and the hematopoietic markers CD45, CD14, and CD34. However, they expressed high levels of

Figure 1. Proliferation potential and senescence of mesenchymal stromal cells (MSC) and MSC end-products (MEP). (A) MSC proliferated at a rate of approximately four population doublings (PD) per passage. The number of cumulative PD (CPD) was 8.5±0.4. Data presented as mean±SEM (n=3). (B) Ex vivo expansion of nineteen MEP through 2 passages. (C) Expansion of MEP for 11 passages and estimation of the number of PD (n=3). (D) RT-PCR analysis of genes related to cell senescence in three clinical-scale MEP. Data presented as mean±SEM (n=3). **P<0.003. Statistical analysis was performed using Student’s t-test.

haematologica | 2016; 101(8)

987

Z. Kuçi et al.

the consensus MSC markers CD73, CD90 and CD105 and HLA class I molecules (Online Supplementary Figure 3B). The mean inhibitory effect of thawed MEP on the proliferation of HLA-mismatched peripheral blood MNC (PBMNC) was 36.7±3.2% (Online Supplementary Figure S3B). All thawed MEP differentiated into adipocytes, osteoblasts or chondrocytes (Online Supplementary Figure S3C-E).

Proliferation potential and senescence of MEP Three MSC bank cryovials were expanded to end-products, and these samples demonstrated a similar number of population doublings (PD) in passages 1 and 2 (4.3 PD/passage). The number of cumulative PD (CPD) did not exceed 8.5±0.024 (Figure 1A). In the interim, we have expanded 19 MSC-aliquots up to the end-product. The mean viability of these thawed aliquots was 95.1±2.1% (range 88.8%-98.2%). As shown in Figure 1B, the mean cell number of all expanded MEP at the end of passage 2 was 5.64x108±0.78×108 MSC, indicating a predictable proliferation potential. To test whether the MSC were immortalized during expansion, we expanded three MEP for 13 passages. As shown in Figure 1C, at some point between passages 5 and 11, the MSC underwent replicative senescence, and the number of PD diminished rapidly. The three MEP that were expanded for 11 passages underwent 30.2 CPDs in 68 days. MEP expansion until passage 13 exhibited no post-senescence proliferation (data not shown). Moreover, analysis of senescence marker expression in three clinical-scale MEP demonstrated no significant elevation in the levels of p16 and p53 gene expression. In contrast, p21 gene expression was significantly increased at passages 11 and 12 compared with passage 4 (Figure 1D). Consistent with the senescent behavior of the MSC from our bank, none of the three examined MEP expressed the oncogene c-myc or hTERT (data not shown).

Allosuppressive potential of MSC isolated from individual donors and MEP Based on our preliminary data, we hypothesized that MSC generated from pooled BM-MNC of 8 donors may exhibit a higher allosuppressive potential than MSC generated from individual donors or pooled MSC from different donors. To test our hypothesis, we expanded MSC from the 8 individual donors from the start of passage 2, and pooled MSCs from the 8 donors (pooled-MSCs) or one MEP (MSC-140) until the end of passage 2. As expected, the allosuppressive potential of the MSC from individual donors in mixed leukocyte reaction (MLR) was highly heterogeneous, ranging from 20% (donors 1 and 8) to approximately 80% (donors 2 and 3) (Figure 2A). The allosuppressive potential of the pooled-MSC was equal to the mean allosuppressive potential of the MSC from the 8 individual donors. In contrast, the allosuppressive potential of the expanded MSC-140 end-product from the MSC bank was significantly greater than that of the pooledMSC or the mean allosuppressive potential of the MSC from the 8 individual donors (P<0.001 and P<0.01, respectively). These results show the advantage of pooling BMMNC for MSC generation. In addition, the allosuppressive potential of six freeze-thawed MEP (as usually administered to patients) demonstrated a consistent allosuppressive effect in vitro (Figure 2B), indicating the equipotency of MSC batches (mean 52±8.7%). 988

Table 1. Patients’ characteristics. Patients enrolled Sex Female Male Age Median [range] years Diagnosis ALL AML MDS RMS SCN/SAA/CGD/DBA SCT source BM PBSC CB Donor MSD FD UD Conditioning regimen TBI-based Chemotherapy-based Serotherapy ATG Campath Without serotherapy GvHD prophylaxis CSA CSA+MTX CSA+MMF MMF Without prophylaxis

100

10 16

38 62

6.5 [1 - 19] 8 5 6 2 2/1/1/1

31 19 23 8 8/4/4/4

13 12 1

50 46 4

4 8 14

15 31 54

5 21

19 81

14 7 5

54 27 19

6 10 3 1 6

23 38 12 4 23

MDS: myelodysplastic syndrome; RMS: rhabdomyosarcoma; SCN: severe congenital neutropenia; SAA: severe aplastic anemia; CGD: chronic granulomatous disease; DBA: Diamond-Blackfan anemia; SCT: stem cell transplantation; BM: bone marrow; PBSC: peripheral blood stem cells; CB: cord blood; MSD: matched sibling donor; FD: family donor; UD: unrelated donor; TBI: total body irradiation; CSA: cyclosporine A; MTX: methotrexate; MMF: mycophenolate mofetil.

Genetic characterization of MEP Because in vitro culture may cause chromosomal aberrations in cells, we performed chromosomal analysis of 25 MSC undergoing mitosis with a resolution of approximately 350-400 bands; 21 of the 25 analyzed metaphases demonstrated a normal karyotype (Figure 3A). Four out of the 25 analyzed metaphases displayed balanced translocation between the short arms of chromosomes 5 and 19. Breakpoints were identified in bands 5p13 and 19p13.3. The karyotype was mos 46,XY[21]/46,XY,t(5;19)(p13;p13.3)[4]. FISH analysis using a 2-color probe for chromosome 5p15 (hTERT) and 5q35 (NSD1) and a 3-color break-apart probe for the MYC gene at chromosomal locus 8q24 demonstrated that the majority of MEP possessed a normal diploid pattern (Figure 3B and C). Interphase nuclei after 2-color hybridization of probe sets 5p15 (green) and 5q35 (red) revealed that 97.2% of the cells showed a normal diploid pattern for chromosome 5, and that only 2.8% of the cells showed a tetraploid hybridization pattern (Figure 3D). Similarly, visualization of interphase nuclei after 3-color hybridization of the MYC break-apart probe (Figure 3C) showed that 97% of the MSCs carried two normal fusion signals haematologica | 2016; 101(8)

MSCs from pooled BM-MNCs of 8 donors for aGvHD

Table 2. Acute graft-versus-host disease and mesenchymal stromal cell treatment.

Patient aGvHD aGvHD aGVHD at initial MSC treatment onset duration aGVHD involved (days) prior to MSC overall organs (days) grade and stages

N. Mean cells/kg Cumulative of BW per dose of infusions infusion cells/kg BW (x106) (x106)

1 2 3 4 5 6 7 8 9 10 11 12 13

77 70 12 22 29 13 21 12 29 23 81 270 10

32 13 34 8 29 22 19 56 21 23 14 19 42

IV III III IV IV III IV II III III III III III

2 5 4 2 2 4 4 1 4 5 4 2 1

1.7 3.5 3.2 1.5 2.0 2.6 0.9 6.0 1.5 1.8 2.1 2.6 1.4

3.5 17.0 18.1 3.0 4.0 10.4 3.5 6.0 6.0 9.2 8.4 5.3 1.4

14 15 16 17 18

22 32 21 30 13

9 112 7 10 96

III IV IV IV IV

GI IV GI IV/Skin IV GI IV/Skin II/Liver IV GI IV/Skin IV GI IV/Skin IV

2 2 3 2 4

3.3 2.5 2.5 1.0 7.7

6.6 5.0 7.6 2.0 31.0

19 20 21 22 23 24 25 26

15 83 30 85 140 21 9 14

9 53 54 14 26 73 54 21

IV III III IV IV IV III IV

GI IV/Skin IV/Liver IV GI III GI IV GI IV/Liver IV GI IV/Liver IV GI IV/Skin IV/Liver IV GI IV Skin IV

4 1 4 4 9 3 2 1

1.6 2.0 2.6 1.4 2.3 1.5 4.4 4.0

6.4 2.0 10.4 5.5 21.5 3.5 8.8 4.0

Prior and concurrent immunosuppression

Steroids, MMF, CSA Steroids, MMF, CSA, ECP Steroids, MMF, CSA, Etanercept Steroids, MMF, CSA Steroids, Budesonide, Tacrolimus Steroids, MMF, CSA Steroids, MMF, ECP Steroids, MMF, CSA Steroids, MMF, CSA, Tacrolimus Steroids, MMF, CSA, ECP Steroids, MMF, CSA Steroids, MMF Steroids, MMF, CSA, Etanercept, Pentostatin, ECP Steroids, MMF, CSA, Basiliximab, Sirolimus Steroids, MMF, MTX, Etanercept, Infliximab Steroids, MMF, CSA, Everolimus, ECP Steroids, MMF, CSA Steroids, MMF, CSA, Infliximab, Basiliximab, ECP Steroids, MMF, CSA Steroids, Infliximab, Tacrolimus Steroids, MMF, CSA, ECP Steroids, MMF, CSA, Everolimus Steroids, MMF, CSA, Everolimus, ECP Steroids, MMF, CSA, ECP Steroids, MMF, Infliximab, Tacrolimus Steroids, MMF, CSA

MSC: mesenchymal stromal cells; aGvHD: acute graft-versus-host disease; GI: gastrointestinal; CSA: cyclosporine A; MTX: methotrexate; MMF: mycophenolate mofetil; ECP: extracorporeal photopheresis; BW: body weight.

for chromosome 8q24 and that 3% of the MSCs displayed a tetraploid signal pattern (Figure 3E).

Comparison of the proliferation potential of MSC from individual donors, pooled MSCs from the 8 donors and MEP Before MSC bank generation, we tested the capacity of BM-MNC from each donor to generate MSC. The number of generated MSCs per 1x106 BM-MNCs after 13 days in culture varied by more than one order of magnitude, ranging from 0.5x105 to 5.4x105 MSC (Figure 4A). Moreover, to validate the rationale of pooling BM-MNC from 8 donors to establish the MSC bank, we compared the in vitro proliferation capacity of the MSC from the 8 individual donors, the pooled MSC of the 8 individual donors, and the four MEP (Figure 4B). The MSC from each bone marrow donor showed different proliferation rates; these varied from 3x105 MSC (donor 7) to 1.7x106 MSC (donor 5). The mean of proliferation of the MSC from the 8 donors was 1x106±5x105 MSC, which correlated well with the number of expanded MSC generated from the haematologica | 2016; 101(8)

pooled-MSC from the 8 donors (1.06x106 MSC). Interestingly, both values correlated very well with the mean number of MSC obtained from the expansion of four MSC bank aliquots within a passage (1.09x106±1×105 MSC). These results confirmed our hypothesis that pooling BM-MNC enables the generation of an “arithmetic mean” of high- and low-proliferating MSC. Because the MSC in our bank were generated from a pool of BM-MNC from 8 “3rd-party” donors, we were interested in the contribution of the BM-MNC from each donor to the MEP. Chimeric analysis via STR-PCR using a series of genetic markers demonstrated the distinct proportions of the MEP derived from the 8 donor samples (Figure 4C). In principle, the relative contribution of each donor sample to the MEP did not strictly correlate with the proliferation potential of the MSC generated from the individual donors (Figure 4A). In addition, donor proportion in the MEP did not correlate with the relative donor proportion in the initially pooled BM-MNCs, which were used as a source for generation of our MSC bank (Online Supplementary Table S1). 989

Z. Kuçi et al.

Safety of MEP in severe aGvHD

Twenty-six patients with severe aGvHD were enrolled in this compassionate use study (Table 1). They received a median of 2.2x106 MSCs per kg BW (range 0.9-4.4x106 MSC per kg BW in 24 patients). One patient received 6x106 MSCs per kg BW, and another patient received 7.7x106 MSCs per kg BW (Table 2). Overall, a median of 3 (range 1-9) MSC infusions were administered to each patient. Only 2 patients exhibited adverse effects to MSC infusion (one incident each of headache and nausea), which presumably may be attributed to the cryoprotectant.

Response to treatment with MSC and overall survival Based on the assessment criteria on day 28 after the initial MSC infusion, 5 of 26 patients (19%) showed a complete response (CR), 15 of 26 patients (58%) a partial response (PR), 4 of 26 patients (15%) did not respond, and 2 of 26 patients (8%) died before day 28 and thus their response could not be evaluated. Overall response rate, defined as patients with CR or PR, was 20 of 26 patients (77%). Follow up for 15 months demonstrated an increase in the CR rate to 73.1% (19 of 26) and a decrease in the percentage of patients experiencing a PR to 11.5% (3 of 26). This treatment resulted in a 2-year overall survival (OS) estimate of 71±11% for the entire patient cohort (n=26) (Figure 5A), indicating the safety of this treatment and suggesting its efficacy in vivo. In addition, cumulative incidence (CI) of non-relapse mortality (NRM) estimate at two years in our patients was 15±7% (Figure 5B). All details concerning the responses to MSC treatment are presented in Table 3. A total of 4 of 26 patients died due to non-relapse mortality (NRM). Two of these 4 patients died due to multiorgan failure based on progressive aGvHD. One of these 4 patients died because of cerebral thromboembolism of unknown origin, whereas the other patient died due to uncontrollable infection. Another 2 patients died due to relapse of their underlying leukemia.

Prognostic factors Using Fisher’s exact test or the Kruskal-Wallis test, we found that none of the clinical factors (Online Supplementary Table S4), including sex, age, diagnosis, donor, conditioning regimen, graft source, GvHD prophylaxis, type and number of drugs used in the initial treatment, severity of aGvHD, and time of aGvHD onset, correlated with the response to MSC treatment.

Discussion In the GvHD-related clinical studies reported so far, patients were treated with MSC generated from individual donors after their culture in either serum-containing13 or serum-free media.18-21 However, donor-to-donor heterogeneity27-29 and the lack of standardized manufacturing protocols may lead to inconsistent clinical results that cannot be compared. In addition, we have recently demonstrated for the first time at the clonal level the intra-donor heterogeneity of the allosuppressive potential of MSC. Interestingly, the net allosuppressive effect of MSC represented an “arithmetic mean” of the high and low allosuppressive clones composing the MSC population.7 Given these findings, the selection of an appropriate donor with 990

Figure 2. Allosuppressive potential of mesenchymal stromal cells (MSC) generated from individual donors and of MSC end-products (MEP). (A) MSC from the 8 individual donors and pooled MSC from these 8 donors (Pooled MSC); and one MEP (MSC-140) were expanded from the start of passage 2 to the end of passage 2. The MSC were evaluated for their allosuppressive effect in a mixed leukocyte reaction (MLR). (B) Six MEP (clinical doses) were thawed, washed, and directly tested in an MLR. Results presented as mean±SEM. Statistical analysis was performed using Student’s t-test.

potent MSC is challenging and, so far, those in the scientific community who are studying MSC have not offered any solution. In an attempt to resolve this issue, in this study, we developed a novel strategy to circumvent or at least to minimize donor-to-donor heterogeneity by establishing an MSC bank from pooled BM-MNCs of multiple “3rd-party” donors (in our case, 8 donors) for the generation of clinical-scale MSC. To validate the rationale of this approach, we tested the allosuppressive effect of MEP from the MSC bank and of MSC derived from 8 donors individually. As expected, the allosuppressive potential of individual MSC was highly heterogeneous. MSC derived from the individuals displayed effects on the alloantigeninduced proliferation of PB-MNC ranging from 20% (donors 1 and 8) to approximately 80% inhibition (donors 2 and 3), with a mean allosuppressive effect of 48%. These results correlated very well with the strikingly high inter-donor differences in the immunosuppressive effects of MSC (range 0-90%).30 In MLR, the allosuppressive haematologica | 2016; 101(8)

MSCs from pooled BM-MNCs of 8 donors for aGvHD Table 3. Response to the mesenchymal stromal cell treatment.

At initial MSC treatment aGvHD Patient aGvHD overall Involved organs grade and stage 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

IV III III IV IV III IV II IV III III III III III IV IV IV IV IV III III IV IV IV III IV

GI IV/Skin IV/Liver IV GI III/Skin III GI IV Skin IV GI IV/Skin IV/Liver II GI III/Skin III/Liver III GI IV/Skin IV/Eyes IV GI I-II GI III/Liver III GI III GI III/Skin III/Eyes III GI II/Skin II GI III/Skin III GI IV GI IV/Skin IV GI IV/Skin II/Liver IV GI IV/Skin IV GI IV/Skin IV GI IV/Skin IV/Liver IV GI III GI IV GI IV/Liver IV GI IV/Liver IV GI IV/Skin IV/Liver IV GI IV Skin IV

Response at 28 days treatment Response at after the initial MSC last follow up aGvHD aGvHD aGvHD overall Involved organs Response aGvHD overall Response grade and stage grade III II III * I II III I 0 0 II 0 0 III II III * II I II III III III II III 0

GI II/Liver II GI I/Skin I GI IV * Skin I GI II GI II/Skin II Skin I no no GI II/Skin II no no GI IV GI I/Skin II GI II/Liver I * GI I/Skin II Skin II GI I GI IV GI II/Liver II GI II/Liver II GI III GI IV no

PR PR NR death PR PR PR PR CR CR PR CR CR NR PR PR death PR PR PR NR PR PR PR NR CR

0 0 IV I 0 0 0 0 III 0 0 0 0 0 0 0 IV II 0 0 0 0 IV 0 IV 0

Outcome

Follow up (months)

CR Alive, cGvHD CR Alive, cGvHD NR Alive PR** Death (Thromboembolism) CR Alive CR Alive CR Alive CR** Death (Relapse) PR** Death (Relapse) CR Alive, cGvHD CR Alive CR** Death (Infection) CR Alive, cGvHD CR Alive CR Alive CR Alive NR** Death (GvHD) PR Alive CR Alive CR Alive CR Alive CR Alive NR** Death (GvHD) CR Alive NR Alive CR Alive

9 10 22 1 39 13 10 12 15 16 6 2 24 42 28 4 0.4 11 13 5 24 6 2 20 18 25

aGvHD: acute graft-versus-host disease; MSC: mesenchymal stromal cells; GI: gastrointestinal; CR: complete response; PR: partial response; NR: non-response; cGvHD: chronic graft-versus-host disease. *Death prior to day 28. **Last response status at follow up before the death of the patient.

potential of pooled MSC from the 8 donors was similar to the mean allosuppressive potential of the MSC from each individual donor. Remarkably, the allosuppressive potential of the expanded MEP, generated from pooled BMMNC, was significantly higher than that of the pooled MSC and the mean allosuppressive potential of the MSC from each individual donor. Therefore, generation of MSC from pooled BM-MNC of multiple donors appears to be more efficient than pooling MSC from several donors, which was reported to generate greater and more stable suppression in vitro and in vivo.10,31,32 Importantly, all tested MEP demonstrated an equivalent allosuppressive effect in vitro after thawing (equipotent MSC doses) (mean 52±8.7%). Although MSC banks provide a large number of “offthe-shelf” products, a few reports have cautioned that freeze-thawed MSC display lower therapeutic efficacy than fresh MSC.3,33,34 In contrast, other studies10,35-37 have demonstrated that cryopreserved MSC exhibit equivalent viability and immunosuppressive potential to freshly isolated MSC from cell culture. Consistent with these findhaematologica | 2016; 101(8)

ings, we found that freeze-thawed MEP displayed a viability of 95% and retained the ability to effectively suppress lymphocyte proliferation in vitro. One of the major criteria required for the clinical application of MSC is that the MSC should enter senescence without undergoing oncogenic transformation. Tarte et al.38 demonstrated that all bone marrow-derived MSC exhibit complete growth arrest at a PD between 35 and 52 and lack post-senescence proliferation even after longterm culture. Three of the thawed MEP, which were expanded for 13 passages, entered replicative senescence between passage 10 or 11 (30.2 CPD) after 68 days in culture. Their senescence was followed by increased levels of the cell cycle regulators p21 and p16 but no change in the TP53 gene expression. This finding is consistent with the data reported for lethally irradiated MSC.39 Importantly, none of the three MEP expressed the proto-oncogene cmyc or hTERT, and no post-senescence proliferation was observed, as previously demonstrated in other studies.28,40 After thorough phenotypic, genetic and functional characterization of our MSC bank, we administered a total of 991

Z. Kuรงi et al. A

Figure 3. Genetic characterization of the clinical-grade mesenchymal stromal cells end-products (MEP). (A) Normal karyogram of MEP. (B) Interphase nuclei after 2-color hybridization of probe sets 5p15 (green) and 5q35 (red). (C) Interphase nuclei after 3-color hybridization of an MYC break-apart probe showed that almost all cells exhibited two normal fusion signals. (D) The number of MSC displaying a normal diploid or aneuploid pattern after 2-color hybridization of probe sets 5p15 and 5q35. (E) The number of MSC displaying a normal diploid or aneuploid pattern after 3-color hybridization of an MYC break-apart probe for chromosome 8q24.

81 MEP to 26 patients with severe refractory aGvHD on a compassionate use basis after individual approval by the regulatory authorities. All patients who received MSC infusions had exhibited failed responses to several other lines of treatment. It is known that the more drugs that fail in the treatment of aGvHD patients, the higher the risk that the patients succumb to GvHD. We did not observe any MSC-related side-effects during transfusion. However, 2 patients died due to progressive GvHD, 2 patients developed a relapse of their underlying leukemia, and 2 others died due to treatment complications. One of these died from infectious complications; as this patient was heavily immunosuppressed at the time of MSC infusion, it is impossible to attribute this event to any single treatment. As the relapse rate in our patients with malignant disease was only 9%, we found no evidence that MSC might hamper a graft-versus-leukemia effect. In our cohort, 20 of 26 patients (77%) responded to the MSC treatment by day 28 (overall response), which is comparable to the results obtained in a randomized placebo-controlled study by using MSC product from the Osiris company (Prochymal) for the treatment of aGvHD (63%). Although the primary end point in that study was not achieved for the whole group of patients, there was a significant benefit over placebo group in the liver and GI tract.41 Response rate in our cohort of patients is also similar to that reported by Introna et al.20 in a cohort of 12 pediatric patients; that study demonstrated an overall response 992

of 66.7%. However, in their patient cohort, only 25% of the patients exhibited aGvHD over grade III, whereas in our series, 96% of the patients exhibited aGvHD grade III or IV. Lucchini et al.19 observed a 62.5% overall response among 8 patients with aGvHD (50% grade I/II and 50% grade III/IV). Similar findings were reported by Prasad et al.14 in a compassionate use study (overall response of 66.7% at day 32 in 12 pediatric patients) and by Kurtzberg et al.15 (61.3% overall response in a large cohort of 75 pediatric patients after treatment of aGvHD with Prochymal). Our results are comparable to the results of the latter group considering the composition of our patient cohort (96% grade III/IV and only 4% grade II), which represents a very challenging patient population. The CR rate increased from 19% at day 28 to 73.1% at the last follow up, whereas the primary PR rate decreased from 58% to 11.5%, indicating that tissue recovery requires time. The inversion of PR by the time in CR is similar to the data obtained by Le Blanc46 in their pediatric cohort. The increased CR rate in our cohort translated into a very favorable 2-year OS estimate of 71%. Kurtzberg et al.15 observed an OS of 57.3% (43 out of 75 patients) at day 100, and this value was similar to the OS rate at two years (40%) in a study by Prasad et al.14 and Le Blanc et al.42 who observed an OS rate at two years of 45% in their cohort of pediatric patients. Although our study was not randomized, considering that our patients were treated in very advanced stages of aGvHD the survival rate is very haematologica | 2016; 101(8)

MSCs from pooled BM-MNCs of 8 donors for aGvHD

Figure 4. Capacity of bone marrow mononuclear cells (BM-MNC) from the 8 donors to generate mesenchymal stromal cells (MSC), their proliferation potential and chimeric analysis of the MSC end-products (MEP). (A) Data are expressed as the number of generated MSC per 1x106 BM-MNC from each donor cultured in 5% human platelet lysis (PL) for 13 days. Age of each donor (in years) is shown in brackets. (B) Comparison of the proliferation potential of MSC from the 8 individual donors with that of pooled MSC from the 8 individual donors and four MEP. (C) Determination of the relative proportion of each donor in MEP via STR-PCR; ns: not significant. Statistical analysis was performed using Student’s t-test.

Figure 5. Overall survival (OS) and nonrelapse mortality (NRM). (A) For all patients (n=26), OS estimate at two years was 71%±11%. (B) Cumulative incidence (CI) of NRM estimate at two years was 15%±7%. Estimates presented as mean±SE.

encouraging (Online Supplementary Table S5). In summary, to our knowledge this is the first serumfree MSC bank generated from pooled BM-MNC of multiple donors as a source for bulk production of clinicalgrade MSC with a predictable potency. Importantly, clinical data presented in this study demonstrated the in vivo safety and efficacy of MEP. Although the results of this single patient treatment are encouraging, a prospective randomized study is required to evaluate the beneficial effect of MEP as a novel cell-based therapy in the treatment of severe aGvHD.

References 1. Lazarus HM, Haynesworth SE, Gerson SL, Rosenthal NS, Caplan AI. Ex vivo expansion and subsequent infusion of human bone marrow-derived stromal progenitor cells (mesenchymal progenitor cells): implications for therapeutic use. Bone Marrow Transplant. 1995;16(4):557-564. 2. Menard C, Pacelli L, Bassi G, et al. Clinicalgrade mesenchymal stromal cells produced under various good manufacturing practice

haematologica | 2016; 101(8)

Funding The authors would like to thank the Robert Pfleger Stiftung, DKMS and Else Kröner-Fresenius-Stiftung (2011_A186) for funding this study. HB, TK and PB are supported by the LOEWE Center for Cell and Gene Therapy Frankfurt/Main funded by Hessisches Ministerium für Wissenschaft und Kunst (HMWK) (funding reference number: III L 4- 518/17.004, 2010). The authors also express their gratitude to Frankfurter Stiftung für krebskranke Kinder (Frankfurt, Germany) for the kind financial support of SK and Dr. Andrea Jochheim-Richter for the expert help in preparation of regulatory issues..

processes differ in their immunomodulatory properties: standardization of immune quality controls. Stem Cells Dev. 2013;22(12): 1789-1801. 3. Galipeau J. The mesenchymal stromal cells dilemma--does a negative phase III trial of random donor mesenchymal stromal cells in steroid-resistant graft-versus-host disease represent a death knell or a bump in the road? Cytotherapy. 2013;15(1):2-8. 4. Wegmeyer H, Broske AM, Leddin M, et al. Mesenchymal stromal cell characteristics vary depending on their origin. Stem Cells

Dev. 2013;22(19):2606-2618. 5. Castro-Malaspina H, Gay RE, Resnick G, et al. Characterization of human bone marrow fibroblast colony-forming cells (CFU-F) and their progeny. Blood 1980;56(2):289-301. 6. Francois M, Romieu-Mourez R, Li M, Galipeau J. Human MSC suppression correlates with cytokine induction of indoleamine 2,3-dioxygenase and bystander M2 macrophage differentiation. Mol Ther. 2012;20(1):187-195. 7. Kuçi Z, Seiberth J, Latifi-Pupovci H, et al. Clonal analysis of multipotent stromal cells

993

Z. Kuรงi et al.

10.

11.

12.

13.

14.

15.

16.

17.

18.

994

derived from CD271+ bone marrow mononuclear cells: functional heterogeneity and different mechanisms of allosuppression. Haematologica. 2013;98(10):16091616. von Bahr L, Sundberg B, Lonnies L, et al. Long-term complications, immunologic effects, and role of passage for outcome in mesenchymal stromal cell therapy. Biol Blood Marrow Transplant. 2012;18(4):557564. Binato R, de Souza FT, Lazzarotto-Silva C, et al. Stability of human mesenchymal stem cells during in vitro culture: considerations for cell therapy. Cell Prolif. 2013; 46(1):1022. Samuelsson H, Ringden O, Lonnies H, Le BK. Optimizing in vitro conditions for immunomodulation and expansion of mesenchymal stromal cells. Cytotherapy. 2009; 11(2):129-136. Koc ON, Gerson SL, Cooper BW, et al. Rapid hematopoietic recovery after coinfusion of autologous-blood stem cells and culture-expanded marrow mesenchymal stem cells in advanced breast cancer patients receiving high-dose chemotherapy. J Clin Oncol. 2000;18(2):307-316. Kebriaei P, Isola L, Bahceci E, et al. Adult human mesenchymal stem cells added to corticosteroid therapy for the treatment of acute graft-versus-host disease. Biol Blood Marrow Transplant. 2009;15(7):804-811. Le Blanc K, Rasmusson I, Sundberg B, et al. Treatment of severe acute graft-versus-host disease with third party haploidentical mesenchymal stem cells. Lancet. 2004;363 (9419):1439-1441. Prasad VK, Lucas KG, Kleiner GI, et al. Efficacy and safety of ex vivo cultured adult human mesenchymal stem cells (Prochymal) in pediatric patients with severe refractory acute graft-versus-host disease in a compassionate use study. Biol Blood Marrow Transplant. 2011;17(4):534-541. Kurtzberg J, Prockop S, Teira P, et al. Allogeneic human mesenchymal stem cell therapy (remestemcel-L, Prochymal) as a rescue agent for severe refractory acute graft-versus-host disease in pediatric patients. Biol Blood Marrow Transplant. 2014;20(2):229-235. Spees JL, Gregory CA, Singh H, et al. Internalized antigens must be removed to prepare hypoimmunogenic mesenchymal stem cells for cell and gene therapy. Mol Ther. 2004;9(5):747-756. Doucet C, Ernou I, Zhang Y, et al. Platelet lysates promote mesenchymal stem cell expansion: a safety substitute for animal serum in cell-based therapy applications. J Cell Physiol 2005;205(2):228-236. Muller I, Kordowich S, Holzwarth C, et al. Animal serum-free culture conditions for isolation and expansion of multipotent mes-

19.

20.

21.

22.

23. 24.

25.

26.

27.

28.

29.

30.

31.

enchymal stromal cells from human BM. Cytotherapy. 2006;8(5):437-444. Lucchini G, Introna M, Dander E, et al. Platelet-lysate-expanded mesenchymal stromal cells as a salvage therapy for severe resistant graft-versus-host disease in a pediatric population. Biol Blood Marrow Transplant. 2010;16(9):1293-1301. Introna M, Lucchini G, Dander E, et al. Treatment of graft versus host disease with mesenchymal stromal cells: a phase I study on 40 adult and pediatric patients. Biol Blood Marrow Transplant. 2014;20(3):375-381. von Bonin M, Stolzel F, Goedecke A, et al. Treatment of refractory acute GVHD with third-party MSC expanded in platelet lysatecontaining medium. Bone Marrow Transplant. 2009;43(3):245-251. Te Boome LC, Mansilla C, van der Wagen LE, et al. Biomarker profiling of steroid-resistant acute GVHD in patients after infusion of mesenchymal stromal cells. Leukemia. 2015;29(9):1839-1846. Cooper K, Viswanathan C. Establishment of a mesenchymal stem cell bank. Stem Cells Int. 2011;2011:905621. Sabatino M, Ren J, David-Ocampo V, et al. The establishment of a bank of stored clinical bone marrow stromal cell products. J Transl Med. 2012;10:23. Gong W, Han Z, Zhao H, et al. Banking human umbilical cord-derived mesenchymal stromal cells for clinical use. Cell Transplant. 2012;21(1):207-216. Mamidi MK, Nathan KG, Singh G, et al. Comparative cellular and molecular analyses of pooled bone marrow multipotent mesenchymal stromal cells during continuous passaging and after successive cryopreservation. J Cell Biochem. 2012; 113(10):31533164. Phinney DG, Kopen G, Righter W, Webster S, Tremain N, Prockop DJ. Donor variation in the growth properties and osteogenic potential of human marrow stromal cells. J Cell Biochem. 1999;75(3):424-436. Russell KC, Phinney DG, Lacey MR, Barrilleaux BL, Meyertholen KE, O'Connor KC. In vitro high-capacity assay to quantify the clonal heterogeneity in trilineage potential of mesenchymal stem cells reveals a complex hierarchy of lineage commitment. Stem Cells 2010;28(4):788-798. Muraglia A, Cancedda R, Quarto R. Clonal mesenchymal progenitors from human bone marrow differentiate in vitro according to a hierarchical model. J Cell Sci. 2000; 113:1161-1166. Moll G, Jitschin R, von Bahr L, et al. Mesenchymal stromal cells engage complement and complement receptor bearing innate effector cells to modulate immune responses. PLoS One. 2011;6(7):e21703. Ringden O, Le Blanc K. Mesenchymal stem cells for treatment of acute and chronic graft-

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

versus-host disease, tissue toxicity and hemorrhages. Best Pract Res Clin Haematol. 2011;24(1):65-72. Ringden O, Keating A. Mesenchymal stromal cells as treatment for chronic GVHD. Bone Marrow Transplant. 2011;46(2):163164. Francois M, Copland IB, Yuan S, RomieuMourez R, Waller EK, Galipeau J. Cryopreserved mesenchymal stromal cells display impaired immunosuppressive properties as a result of heat-shock response and impaired interferon-gamma licensing. Cytotherapy. 2012;14(2):147-152. Moll G, Alm JJ, Davies LC, et al. Do cryopreserved mesenchymal stromal cells display impaired immunomodulatory and therapeutic properties? Stem Cells. 2014; 32(9):24302442. de Lima PK, de Santis GC, Orellana MD et al. Cryopreservation of umbilical cord mesenchymal cells in xenofree conditions. Cytotherapy. 2012;14(6):694-700. Al-Saqi SH, Saliem M, Quezada HC, et al. Defined serum- and xeno-free cryopreservation of mesenchymal stem cells. Cell Tissue Bank. 2015;16:181-93. Luetzkendorf J, Nerger K, Hering J, et al. Cryopreservation does not alter main characteristics of Good Manufacturing Process-grade human multipotent mesenchymal stromal cells including immunomodulating potential and lack of malignant transformation. Cytotherapy. 2015;17(2):186-198. Tarte K, Gaillard J, Lataillade JJ, et al. Clinical-grade production of human mesenchymal stromal cells: occurrence of aneuploidy without transformation. Blood. 2010;115(8):1549-1553. Fekete N, Erle A, Amann EM, et al. Effect of High-Dose Irradiation on Human BoneMarrow-Derived Mesenchymal Stromal Cells. Tissue Eng Part C Methods. 2015; 21:112-22. Shibata KR, Aoyama T, Shima Y, et al. Expression of the p16INK4A gene is associated closely with senescence of human mesenchymal stem cells and is potentially silenced by DNA methylation during in vitro expansion. Stem Cells. 2007;25(9): 2371-2382. Martin PJ, Uberti JP, Soiffer RJ, et al. Prochymal improves response rates in patients with steroid-refractory acute graft versus host disease (SR-GVHD) involving the liver and gut: results of a randomized, placebo-controlled, multicenter phase III trial in GVHD. Biol Blood Marrow Transplant. 2010;16(2):169-170. Le Blanc K, Frassoni F, Ball L, et al. Mesenchymal stem cells for treatment of steroid-resistant, severe, acute graft-versushost disease: a phase II study. Lancet. 2008; 371(9624):1579-1586.

haematologica | 2016; 101(8)

ERRATA CORRIGE Treatment of relapsed and refractory multiple myeloma

The HDAC inhibitors panobinostat and vorinostat are epigenetic drugs that can be combined with other agents. Panobinostat had a significant advantage when combined with bortezomib and dexamethasone over the same drugs with placebo; at the cost, however, of gastrointestinal symptoms and fatigue. Panobinostat combined with bortezomib and dexamethasone has been approved by the EMA for use in RRMM patients. Pieter Sonneveld1 and Annemiek Broijl2 University Hospital Rotterdam; 2ErasmusMC Cancer Institute, The Netherlands 1

Correspondence: p.sonneveld@erasmusmc.nl doi:10.3324/haematol.2016.148882 Key words: multiple myeloma, chemotherapy, immunotherapy, relapse. Information on authorship, contributions, and financial & other disclosures was provided by the authors and is available with the online version of this article at www.haematologica.org.

haematologica | 2016; 101(8)

995

haematologica Journal of the European Hematology Association Published by the Ferrata Storti Foundation

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

Ancient Greek

aÂma [haima] = blood a·matow [haimatos] = of blood lÒgow [logos]= reasoning

Scientific Latin

haematologicus (adjective) = related to blood

Scientific Latin

haematologica (adjective, plural and neuter, used as a noun) = hematological subjects

Modern English

The oldest hematology journal, publishing the newest research results. 2015 JCR impact factor = 6.671

Haematologica, as the journal of the European Hematology Association (EHA), aims not only to serve the scientific community, but also to promote European cultural identify.