Congrès de l’American Society of Hematology – déc. 2014

Comunications Orales au congrès de l’American Society of Hematology à San Fransisco, Déc. 2014 par les équipes du CARPEM

Co-Activation of AMPK and mTORC1 Is Synthetically Lethal in Acute Myeloid Leukemia Session: 604. Molecular Pharmacology and Drug Resistance in Myeloid Diseases

Pierre Sujobert¹, Etienne Paubelle^2,3, Adrien Grenier¹, Florence Zylbersztejn⁴, Mireille Lambert¹, Laury Poulain¹, Elizabeth Townsend⁵, Jean-Marie Brusq⁶, Edwige Nicodeme⁶, Justine Decrooqc⁷, Ina Nepstad⁸, Alexa Green¹, Marie-Anne Hospital¹, Nathalie Jacque¹, Olivier Hermine^2,3,7, Marc Foretz¹, Benoit Viollet¹, Catherine Lacombe¹, David M. Weinstock⁹, Patrick Mayeux¹, Ivan C Moura⁷, Didier Bouscary¹⁰ and Jerome Tamburini^1,10

¹ Cochin Institute, Développement, Reproduction, Cancer Department, Paris Descartes – Sorbonne Paris Cité University, Paris, France ² Hématologie, Hôpital Necker, Paris, France ³ Faculté de Médecine, Hôpital Necker, CNRS UMR 8147, Université Paris Descartes, Paris, France ⁴INSERM U699, Paris, France, ⁵ Harvard University, Boston, ⁶ GlaxoSmithKline Research Center, Les Ulis, France ⁷INSERM UMR 1163, Laboratory of cellular and molecular mechanisms of hematological disorders and therapeutic implications, Paris Descartes – Sorbonne Paris Cité University, Imagine Institute, ⁸Bergen University, Bergen, Norway, ⁹Dana-Farber Cancer medical Oncology and Institute, Boston, MA ¹⁰ Service d’Hématologie Clinique, APHP – Hôpital Cochin, Paris, France

Virtually all recurrent molecular alterations in acute myeloid leukemia (AML) functionally converge to cause signal transduction pathway dysregulation that drives cellular proliferation and survival. The mammalian target of rapamycin complex 1 (mTORC1) is a rapamycin-sensitive signaling node defined by the interaction between mTOR and raptor. Constitutive mTORC1 activity is nearly universal in AML. However, pharmacologic inhibition with rapamycin or second-generation mTOR kinase inhibitors has shown limited anti-leukemic activity in both preclinical models as well as patients, suggesting that addiction to this oncogene is not a recurrent event in AML. Here we report that sustained mTORC1 activity is nonetheless essential for the cytotoxicity induced by pharmacologic activation of AMP-activated protein kinase (AMPK) in AML.

Our studies employed a novel AMPK activator called GSK621. Using CRISPR and shRNA-mediated silencing of the AMPKa1 catalytic subunit, we showed that AMPK activity was necessary for the anti-leukemic response induced by this agent. GSK621-induced AMPK activation precipitated autophagy, as demonstrated by western blotting, immunofluorescence, flow cytometry and electron microscopy. Blocking autophagy via shRNA-mediated knockdown of ATG5 and ATG7 protected AML cells from cytotoxicity resulting from treatment with GSK621, suggesting that autophagy promotes cell death in the context of active AMPK. GSK621 cytotoxicity was consistently observed across twenty different AML cell lines, primary AML patient samples and AML xenografts in vivo. GSK621-induced AMPK activation also impaired the self-renewal capacity of MLL-ENL- and FLT3-ITD-induced murine leukemias as measured by serial methylcellulose replating assays.

Strikingly, GSK621 did not induce cytotoxicity in normal CD34+ hematopoietic progenitor cells. We hypothesized that the differential sensitivity to GSK621 could be due to the difference in amplitude of mTORC1 activation in AML and normal CD34+ cells. In contrast to most reported cellular models in which AMPK inhibits mTORC1 both directly (through raptor phosphorylation) and indirectly (through TSC2 phosphorylation), sustained mTORC1 activity was seen following GSK621-induced AMPK activation in AML. Inhibition of mTORC1 either pharmacologically (using rapamycin) or genetically (using shRNAs targeting raptor and mTOR) abrogated AMPK-induced cytotoxicity in AML cells, including primary AML patient samples. This protective effect was mediated by mTORC1-dependent modulation of the ATF4/CHOP stress response pathway. The ultimate functional consequence was that, rather than diminishing GSK621-induced cytotoxicity, persistent mTORC1 activity was in fact synthetically lethal with AMPK activity in AML cells. This synthetic lethality could be recapitulated in normal CD34+ progenitors by constitutive activation of mTORC1 using a lentivirally-transduced myrAKT construct. It could also be enhanced in AML cells by mTORC1 overactivation induced by CRISPR-mediated deletion of TSC2.

Taken together, these data show that the magnitude of mTORC1 activity determines the degree of cytotoxicity triggered by AMPK activation. This finding may have important implications for AMPK and mTORC1 signaling pathways in cancer biology more broadly. Context-dependent permissiveness towards mTORC1 activation may amplify the response to cytotoxic stress, such as that resulting from AMPK activation by GSK621. Our results therefore support AMPK activation as a promising therapeutic strategy in AML and other mTORC1-active malignancies which warrants further investigations in clinical trials.

Rituximab Maintenance Versus Wait and Watch after Four Courses of R-DHAP Followed By Autologous Stem Cell transplantation in Previously Untreated Young Patients with Mantle Cell Lymphoma: First Interim Analysis of the Phase III Prospective Lyma Trial, a Lysa Study Session: 623. Lymphoma: Chemotherapy, excluding Pre-Clinical Models: Indolent Lymphoma and Mantle Cell Lymphoma

Steven Le Gouill, MD, PhD^1*, Catherine Thieblemont, MD, PhD², Lucie Oberic^3*, Krimo Bouabdallah, MD^4*, Emmanuel Gyan, MD, PhD⁵, Gandhi Damaj, MD⁶, Vincent Ribrag, MD⁷, Serge Bologna, MD^8*, Remy Gressin, MD⁹, Olivier Casasnovas, MD^10*, Corinne Haioun, MD, PhD¹¹, Philippe Solal-Celigny, MD¹², Herve Maisonneuve, MD^13*, Eric Van Den Neste, MD, PhD^14*, Anne Moreau, MD^15*, Marie C Bene^16*, Gilles Salles, MD PhD¹⁷, Hervé Tilly, MD, PhD¹⁸, Thierry Lamy, MD, PhD¹⁹ and Olivier Hermine, MD, PhD²⁰

¹Hematology, CHU Nantes, Nantes, France ²Service d’Hématologie Clinique, Hôpital Saint-Louis, Paris, France ³Hematology, Purpan University Hospital, Toulouse, France ⁴Service d’Hématologie et de Thérapie Cellulaire, University Hospital of Bordeaux, Bordeaux, France ⁵Service d’Hématologie et de Thérapie cellulaire, CHRU de Tours, Tours, France ⁶Hematology Department, CHU d’Amiens, Amiens, France ⁷Département de Médecine, Institut Gustave Roussy, Villejuif, France ⁸Service d’Hématologie, CHU de Nancy, Hôpital de Brabois, Vandoeuvre-lès-Nancy, France ⁹Dept. of Onc.-Hem., CHU Michallon, Grenoble, France ¹⁰Hematology Clinic, CHU Le Bocage, Dijon, France ¹¹Hematology, Hôpital Henri Mondor, Créteil Cedex, France ¹²Centre Jean Bernard, Le Mans, France ¹³Service D’OncoHematologie, Centre Hospitalier Departemental, La Roche-sur-Yon, France ¹⁴Hematology Department, Cliniques universitaires UCL Saint-Luc, Brussels, Belgium ¹⁵CHU de Nantes, Nantes, France ¹⁶Laboratoire d’hématologie, Centre Hospitalier Universitaire, Nantes, France ¹⁷Centre Hospitalier Lyon-Sud, Pierre-Benite, France ¹⁸Service d’Hématologie Clinique, Centre Henri Becquerel, Rouen, France ¹⁹Hematology Department, INSERM U917 / University Hospital of Rennes, Rennes, France ²⁰Hematology department and Institut Imagine, Hôpital Necker-Enfants Malades, Paris, France

The LyMa (ClinicalTrials.gov, NCT00921414) is a prospective randomised phase III trial conducted by the LYSA group (GOELAMS and GELA groups) and that assessed the potential benefit of Rituximab maintenance after autologous stem cell transplantation (ASCT) in young previously untreated Mantle Cell Lymphoma (MCL) patients (<66y).

Patients were enrolled at times of diagnosis. All patients received 4 courses of R-DHAP followed by ASCT (patients who did not reach at least a PR after these 4 courses could receive 4 additional courses of R-CHOP). The conditioning regimen of ASCT was Rituximab (500mg/m2) plus BEAM. Patients achieving a complete or partial response after ASCT were then randomly assigned to receive 3 years of Rituximab maintenance therapy (375mg/m2, one injection every two months) or wait and watch (WW) (1:1).

The primary endpoint was EFS at 4 years after randomization, EFS being defined as death of any cause, disease progression, severe allergic reaction to Rituximab or severe infection. PFS and OS were secondary objectives. Herein, we present the first planned interim analysis. Analysis was performed by intention to treat.

From September 2008 to August 2012, 299 patients were included (one patient withdrawn his consent, data of one patient with incomplete data at time of the present analysis). Median age at registration was 57y (27-65) and 236 (78,9%) patients were male. MIPI score was low in 53,2% (n=159), intermediate in 27,4% (n=82) and high in 19,4% (n=58). In all, 257 (86%) patients proceeded to ASCT. The CR/CRu rates before and after ASCT were 81,4% and 92%, respectively. At the time of the present interim analysis, 58 patients died. With a median follow-up calculated from time of inclusion of 35.8 months, median PFS and OS are not reached. The estimates 3y-PFS and -OS are 73.7% (95%CI ; 67.8-78.7) and 82.6% (95%CI ; 77.3-86.8), respectively. Last randomization was done in February 2013. Two hundred and thirty eight patients were randomised: 119 patients were assigned to rituximab maintenance and 119 to WW. The mFU (n=238) calculated from date of randomization is 29.7 months. Median EFS and PFS are not reached : the 2y-EFS is 87.5% (95%CI ; 82.4-91.2) and 2y-PFS is 87.5% (95%CI ; 82.4-91.2). The EFS and PFS are statistically different between the treatment arms (p=0.015 for both) : the 2y-EFS is 93.2% (95%CI, 86.9-96.6) in the Rituximab arm versus 81.5% (95%CI, 72.7- 87.7) in the WW arm (HR=2.1). OS does not differ between the two groups. The 2y-OS is 93.4% (95%CI, 86.6-96.9) in the Rituximab arm versus 93.9% (95%CI, 86.7-97.3) in the WW arm.

This planned interim analysis of the LyMa trial shows that a 3 years of rituximab maintenance after R-DHAP plus ASCT as first-line treatment for young patients with MCL significantly improves both EFS and PFS. Thus, as reported in elderly MCL, the Lyma trial demonstrates that Rituximab should be used in maintenance therapy after ASCT and provides the rational for a new standard of care in MCL.

MRD Eradication Should be the Therapeutic Goal in Mantle Cell Lymphoma and May Enable Tailored Treatment Approaches: Results of the Intergroup Trials of the European MCL Network

Session: 623. Lymphoma: Chemotherapy, excluding Pre-Clinical Models

Christiane Pott, MD, PhD^1*, Elizabeth Macintyre, MD, PhD², Marie-Helene Delfau-Larue, MD, PhD^3*, Vincent Ribrag, MD⁴, Michael Unterhalt, MD^5*, Michael Kneba, MD, PhD⁶, Wolfgang Hiddemann, MD⁷, Martin Dreyling, MD, PhD⁸, Olivier Hermine, MD, PhD⁹ and Eva Hoster, PhD^5*

¹ Cochin Institute, Développement, Reproduction, Cancer Department, Paris Descartes – Sorbonne Paris Cité University, Paris, France ² Hématologie, Hôpital Necker, Paris, France ³ Faculté de Médecine, Hôpital Necker, CNRS UMR 8147, Université Paris Descartes, Paris, France ⁴INSERM U699, Paris, France, ⁵ Harvard University, Boston, ⁶ GlaxoSmithKline Research Center, Les Ulis, France ⁷INSERM UMR 1163, Laboratory of cellular and molecular mechanisms of hematological disorders and therapeutic implications, Paris Descartes – Sorbonne Paris Cité University, Imagine Institute, ⁸Bergen University, Bergen, Norway, ⁹Dana-Farber Cancer medical Oncology and Institute, Boston, MA ¹⁰ Service d’Hématologie Clinique, APHP – Hôpital Cochin, Paris, France

 Background: Minimal residual disease (MRD) measurement is an increasingly recognized tool for response assessment in B-cell malignancies and represents a useful surrogate marker for prognosis, allowing prediction of the clinical course prior to clinical study endpoints. Based on the randomized intergroup trials of the European MCL Network for patients with stage II-IV mantle cell lymphoma (MCL) investigating immuno-chemotherapy followed by autologous stem cell transplantation (ASCT) for patients 65 yrs, we prospectively monitored MRD.

Methods: According to the protocols MRD analyses were performed after completed induction and subsequently in 3-monthly intervals after ASCT or 2-monthly during maintenance until progression. MRD results were evaluated according to ESG criteria (van der Velden, Leukemia 2007) and compared with clinical outcome. RQ-PCR was designed to reach a sensitivity of 10E-5, MRD negativity was defined as a negative RQ-PCR result with a sensitivity of at least 10E-4. Patients in complete or partial remission 6 months after ASCT/end of induction (MCL Younger/MCL Elderly) were included in this analysis. During follow-up MRD status in peripheral blood (PB) was assessed and defined as negative, if there were only negative MRD values, or positive in case of at least one positive MRD value during a 6-months period. Landmark analyses were performed for progression-free survival (PFS) according to MRD status for each 6-months period until 4 years of follow-up and PFS was compared between PB positive and PB negative patients.

Results: Among patients in remission 6 months after ASCT/end of induction, MRD results during follow-up were available in 406 patients (255 MCL Younger,151 MCL Elderly). The distribution into low, intermediate, and high risk MIPI of 44%, 34% and 22%, respectively, was similar to the total study population. During follow-up period, percentage of patients with evaluable MRD status dropped from about 50%-55% during the first two years to 40%-45% during years 3-4, and less than 40% afterwards. The rate of MRD positive samples was about 20-25% during the first 3 years, and less than 20% in the subsequent follow-up.

In each landmark analysis, a positive MRD status in PB was highly associated with a shorter PFS. This association was independent of baseline MIPI score, treatment arm or protocol (Figure 1). Thus, the prognostic value of MRD positivity was maintained in both trials MCL Younger and MCL Elderly. Remarkably, landmark analyses at all time points indicated a strong association also of low-level MRD with a shortened PFS. Finally, MRD analyses in 76 patients at relapse confirmed the close association of clinical relapse to MRD positivity, with only 10 patients (13%) being PB-negative.

Conclusion: Our data demonstrate that achievement and preservation of MRD negativity is the strongest independent predictor of prognosis in patients with MCL. Applying MRD status as tool for tailored treatment, therapeutic approaches should focus on a maximum MRD response to improve long term outcome.

Pediatric ALL-like Therapy in Adults with T-Cell Lymphoblastic Lymphoma: Results of the Graall-Lysa LL03 Study Session: 612. Acute Lymphoblastic Leukemia: Clinical Studies II

Stephane Lepretre, MD^1*, Aurore Touzart, MD^2*, Thomas Vermeulin^3*, Jean-Michel Picquenot, MD^4*, Aline Tanguy-Schmidt, MD^5*, Gilles Salles, MD PhD⁶, Thierry Lamy⁷, Marie C Bene^8*, Emmanuel Raffoux, MD, PhD^9*, Françoise Huguet, MD^10*, Patrice Chevallier, MD, PhD^11*, Serge Bologna, MD^12*, Reda Bouabdallah, MD^13*, Jacques Benichou, MD^3*, Josette Briere, MD^14*, Anne Moreau, MD^15*, Valerie Simon^4*, Stephanie Seris^4*, Carlos Graux, MD, PhD^16*, Vahid Asnafi, MD, PhD^17*, Norbert Ifrah¹⁸, Elizabeth Macintyre, MD, PhD^17* and Herve Dombret¹⁹

¹Hematology, Henri Becquerel Center, Rouen, France ²Paris Descartes – Sorbonne Paris Cité University, Hopital Necker-Enfants Malades AP-HP Laboratoire d’Hematologie, Paris, France ³CHU C Nicolle, Rouen, France; ⁴CLCC H Becquerel, Rouen, France ⁵CHU d’Angers, Angers, France ⁶Centre Hospitalier Lyon-Sud, Pierre-Benite, France ⁷Department of Hematology, Hôpital Pontchaillou, Rennes, France ⁸Immunology, CHU, Nantes, France ⁹Hematology, Hôpital Saint-Louis, APHP, Paris, France ¹⁰Hematology, Hôpital Purpan, Université Toulouse, Toulouse, France ¹¹Hematology Department, Nantes University Hospital, Nantes, France ¹²Service d’Hématologie, CHU de Nancy, Hôpital de Brabois, Vandoeuvre-lès-Nancy, France ¹³Institut Paoli-Calmettes, Marseille, France ¹⁴Hematology, Hopital Saint-Louis, APHP, Paris, France ¹⁵CHU de Nantes, Nantes, France ¹⁶Department of Hematology, Mont-Godinne University Hospital, Yvoir, Belgium ¹⁷Hôpital Necker, AP-HP, Université Paris 5, Paris, France ¹⁸Hematology, CHU, Angers, France ¹⁹ALFA coordination, Saint-Louis Hospital, Paris, France

Background: It has been suggested that using an acute lymphoblastic leukemia (ALL) rather than non-Hodgkin lymphoma protocol to treat patients with lymphoblastic lymphoma (LL) might be associated with better results (Hoelzer, Best Pract Res Clin Hematol 2002). To address this issue, the GRAALL and LYSA groups have conducted the Phase 2 LL03 trial in adult patients with LL, using the GRAALL-2003 protocol, which yielded good results in adult patients with ALL (Huguet, JCO 2009).

Patients and Methods: Between 2004 and 2012, 155 patients aged 18-59 years were enrolled, including 131 evaluable patients with T-cell LL (T-LL). The pediatric-inspired ALL treatment included a corticosteroid prephase, a 5-drug induction with sequential cyclophosphamide, high dose consolidation, late intensification, CNS prophylaxis with IT injections and cranial irradiation, and a 2-year maintenance. Response, including complete remission (CR) and unconfirmed CR (CRu), were assessed using Cheson criteria (Cheson, JCO 1999). Allogeneic stem cell transplantation (SCT) was offered to CR/CRu patients with high-risk disease (defined as need for a second-induction salvage course and/or CNS disease) and a donor.

Results: Of 131 T-LL patients (median age, 33 years; M/F ratio 4.0; mediastinal enlargement, 95%; CNS involvement, 5%), 119 patients (91%) reached CR/CRu (30 patients needing a salvage course) and 34 relapsed. Response evaluation was based on CT scan, as PET scan was performed in only 73/131 and 20/30 patients after first induction and salvage, respectively. At 5 years, estimated DFS, EFS and overall survival were 71%, 61% and 66%, respectively. The lymphoma IPI-score had no prognostic value, but increased serum LDH level (observed in 71% of the patients) was associated with a significant decrease in EFS (HR = 2.8 [1.3 – 6.1]) and OS (HR = 3.5 [1.4 – 9.1]) in multivariable analysis. Of note, need for a salvage course was not associated with shorter DFS in CR/CRu patients.

In a subset of 49 patients studied for oncogenetic markers, the 4-gene risk classifier (based on NOTCH1, FBXW7, N/K-RAS and PTEN status) we have recently reported to be a powerful predictor in T-ALL patients (Trinquand, JCO 2013) also demonstrated strong prognostic value in T-LL. Among these patients, 29 (60%) had a high-risk genetic profile (defined as no NOTCH1/FBXW7 mutation and/or N/K-RAS mutation and/or PTEN deletion). At 3 years, the high-risk genetic profile was predictive of shorter EFS (HR = 14.3 [1.9 – 107.8]), DFS (HR = 9.5 [1.2 – 74.3]) and OS (HR = 11.5 [1.5 – 87.5]) in univariable analysis, as well as in multivariable analysis after adjustment on age, ECOG-PS and LDH level (HR = 20.5 [2.6 – 164.1], 12.6 [1.5 – 104.8] and 17.0 [2.1 – 136.8], respectively. A total of 30 CR/CRu patients were eligible for allogeneic SCT (25 for late CR/CRu, 4 for CNS involvement, and 1 for both criteria) and 17 of them were actually transplanted in first CR/CRu. When analysed as a time-dependent event, allogeneic SCT was not associated with prolonged DFS in these high-risk patients. Finally, Grade III/IV adverse events were those commonly observed with the GRAALL regimen. Overall, 46 patients died during the study (37 after relapse or progression; 5 during induction; 3 from allograft toxicity and 1 after a highway accident).

Conclusion: As compared to historical studies, we report here a relatively good outcome in T-LL patients treated with a pediatric-inspired ALL strategy. Very interestingly, the NOTCH1/FBXW7/RAS/PTEN T-ALL risk classification was also a strong prognostic factor in these T-LL patients. Allogeneic SCT did not appear to significantly influence the outcome of selected T-LL patients.

TP53 Mutation Status Divides MDS Patients with Complex Karyotypes into Distinct Prognostic Risk Groups: Analysis of Combined Datasets from the International Working Group for MDS-Molecular Prognosis Committee

Session: 637. Myelodysplastic Syndromes – Clinical Studies I

Rafael Bejar, MD, PhD¹, Elli Papaemmanuil, PhD^2*, Torsten Haferlach, MD³, Guillermo Garcia-Manero, MD⁴, Jaroslaw P. Maciejewski, MD, PhD⁵, Mikkael A. Sekeres, MD, MS⁵, Matthew J. Walter, MD⁶, Timothy A. Graubert, MD⁷, Mario Cazzola, MD⁸, Luca Malcovati, MD^9*, Peter J Campbell, PhD^2*, Seishi Ogawa, MD, PhD¹⁰, Jacqueline Boultwood, PhD¹¹, David Bowen, MD, PhD¹², Sudhir Tauro, PhD^13*, Michael J Groves, PhD^13*, Michaela Fontenay, MD, PhD^14*, Lee-Yung Shih, MD¹⁵, Heinz Tüchler^16*, Kristen E. Stevenson, MS^17*, Donna Neuberg, ScD¹⁸, Peter L. Greenberg, MD, PhD¹⁹ and Benjamin L. Ebert, MD PhD²⁰

¹UCSD Moores Cancer Center, La Jolla, CA ²Cancer Genome Project, Wellcome Trust Sanger Institute, Hinxton, United Kingdom ³MLL Munich Leukemia Laboratory, Munich, Germany ⁴Leukemia, The University of Texas MD Anderson Cancer Center, Houston, TX ⁵Leukemia Program, Department of Hematology and Medical Oncology, Cleveland Clinic Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH ⁶Department of Medicine, Division of Oncology, Washington University School of Medicine, St. Louis, MO ⁷Massachusetts General Hospital / Harvard Medical School, Boston, MA ⁸Department of Hematology Oncology, Fondazione IRCCS Policlinico San Matteo & University of Pavia, Pavia, Italy ⁹Departments of Molecular Medicine & Hematology Oncology, University of Pavia & Fondazione IRCCS Policlinico San Matteo, Pavia, Italy ¹⁰Department of Pathology and Tumor Biology, Graduate School of Medicine, Kyoto University, Japan ¹¹LRF Molecular Hematology Unit, NDCLS, John Radcliffe Hospital, Oxford, United Kingdom ¹²St. James’s Institute of Oncology, Leeds Teaching Hospitals, Leeds, United Kingdom ¹³Dept of Hematology, Ninewells Hospital, Dundee, United Kingdom ¹⁴Assistance Publique-Hopitaux De Paris, Paris, France ¹⁵Chang Gung Memorial Hospital at Taipei, Taipei, Taiwan ¹⁶Ludwig-Boltzmann Institute for Leukemia Research, Vienna, Austria ¹⁷Dana Farber Cancer Institute, Boston, MA ¹⁸Biostatistics and Computational Biology, Dana-Farber Cancer Institute, Boston, MA ¹⁹Stanford University Cancer Center, Stanford University, Stanford, CA ²⁰Division of Hematology, Brigham and Women’s Hospital, Boston, MA

Background: Complex karyotypes in patients with myelodysplastic syndromes (MDS) are defined by the presence of 3 or more cytogenetic abnormalities and are considered indicators of a poor prognosis by nearly all models in clinical use. There is evidence that the prognostic significance of complex karyotypes in MDS can be refined by considering the number of cytogenetic abnormalities or the presence of specific lesions, the presence of monosomy, and the mutation status of the TP53 gene. Mutations of TP53 are strongly associated with adverse clinical features including complex karyotypes, yet still have independent prognostic significance. Whether these mutations or other elements associated with complex karyotypes carry the greatest prognostic significance is unknown and has been the subject of recent contradictory findings. Here we examine the relationship between TP53 mutations, the number and type of chromosomal abnormalities, and clinical features including overall survival in a combined analysis of MDS patients with complex karyotypes.

Methods: Data on 258 complex karyotype MDS patients from the US and Europe gathered by members of the International Working Group for Prognosis in MDS-Molecular Committee were combined for this analysis. Patients gave informed consent to have their data and clinical samples collected at their respective institutions in accordance with the Declaration of Helsinki. Tumor samples were examined for TP53 mutations primarily with next generation sequencing techniques. Categorical variables were compared using a Fisher exact test or Kruskal-Wallis test as appropriate, while continuous variables were compared using a Wilcoxon rank-sum test. All p-values reported are two-sided, unadjusted for multiple testing, and considered significant at the 0.05 level.

Results: Of the 258 complex karyotype MDS samples examined, TP53 mutation status was determined in 223, of which 133 (60%) carried a TP53 mutation. Patients with TP53 mutations were slightly older than unmutated patients (median age 70.8 vs. 65.5 years, p=0.055), were more thrombocytopenic (median platelet count 42 vs. 78 x109/L, p=0.002), and had a greater percentage of bone marrow blasts (7.75% vs 4%, p=0.03). Mutated patients were also more likely to have a monosomal karyotype (85% vs. 59%, p<0.0001), have 5 or more chromosomal abnormalities (85% vs. 51%, p<0.0001), and have abnormalities of chromosome 17 (38% vs. 23%, p=0.03). Patients with TP53 mutation had a significantly shorter median overall survival of 7.7 months compared to 23.4 months for patients with unmutated TP53 (p<0.0001, see figure). Multivariate analysis that considered potential confounders such as monosomal karyotypes, number of abnormalities, blast proportions and platelet count identified TP53 mutation status as the most significant prognostic marker for overall survival (hazard ration [HR] 2.12, p<0.0001). The presence of only 3 or only 4 karyotype abnormalities was protective compared with 5 or more such lesions (HR 0.385, p=0.0005 for 3 lesions and HR 0.505, p=0.0083 for 4 lesions).

Discussion: Despite strong associations with adverse clinical and cytogenetic abnormalities incorporated into existing prognostic scoring systems, TP53 mutations carry significant independent prognostic value for patients with MDS. In our multivariable analyses, TP53 mutation appears to dominate over monosomal karyotype as an adverse prognostic marker in MDS patients with complex cytogenetics. Many of the TP53 mutations identified had a small variant allele fraction suggesting that they were present in a disease subclone. The small number of mutant cells in these cases may be insufficient to alter clinical measures, but could still indicate a poor prognosis if they represent a clone capable of treatment resistance and rapid disease progression. This combined analysis is a work in progress that will include data from additional groups and explore mutation status of other MDS-related genes, TP53 variant allele fraction, clinical risk scores, and data from patients without complex karyotypes.

Conclusions: Somatic TP53 mutations are highly enriched in MDS patients with complex karyotypes and identify a group with significantly shortened overall survival. Sequencing of TP53 can improve the prediction of prognosis in MDS patients with 3 or more chromosomal abnormalities.

NPM1 Expression Level and a CRBN Polymorphism Are Able to Predict the Rate of Response to Lenalidomide in Non Del(5q) Lower Risk MDS Patients Resistant to Erythropoiesis-Stimulating Agents: The GFM Experience

Session: 637. Myelodysplastic Syndromes – Clinical Studies I

Virginie Chesnais^1*, Aline Renneville^2,3*, Andrea Toma^4*, Veronique Sardnal^1*, Marie Passet^1*, Audrey Gauthier^1*, Jacques Delaunay, MD^5*, Aspasia Stamatoulas, MD^6*, Odile Beyne-Rauzy, MD-PhD^7*, Pierre Fenaux, MD, PhD⁸, Francois Dreyfus, MD^9*, Claude Preudhomme, PharmD, PhD¹⁰, Olivier Kosmider, MD, PhD^11* and Michaela Fontenay^12*

¹Institut Cochin, Paris, France ²Hematology Laboratory, Biology and Pathology Center, Lille University Hospital, Lille, France 3U837, Team ³, INSERM, Lille, France ⁴Hopital Henri Mondor, Creteil, France ⁵CHU de Nantes – Hôtel Dieu, Nantes, France ⁶Département d’Hématologie, Centre Henri BECQUEREL, Rouen, France ⁷Hematology department, Purpan universitary hospital, Toulouse, France ⁸Service d’Hématologie Séniors, Hôpital St Louis, Université Paris 7, Paris, France ⁹Hematology, APHP, Cochin Hospital, University Paris Descartes, Paris, France ¹⁰Institut d’Hématologie Transfusion, CHU Lille, Lille, France ¹¹Laboratory of Hematology, AP-HP, Hôpital Cochin, INSERM U1016 CNRS UMR 8104, Cochin Institute, Université Paris Descartes, Faculty of Medicine, Paris, France ¹²Groupe Francophone des Myélodysplasies (GFM), Paris, France

Lenalidomide (Len) is the reference treatment of anemia in patients with del(5q) MDS with half of them being responders. An erythroid response is also achieved in ~25% of non del(5q) low-risk or int-1 MDS patients. The mechanisms of resistance are still unclear although the molecular target of Len, Cereblon, a receptor for several substrates including Ikaros, Aiolos and the casein kinase 1 in the E3 ubiquitin ligase Cul4A-DDB1-Roc1, had been identified. Biomarkers are needed to avoid inappropriate exposure to the risk of severe neutropenia or thrombocytopenia.

In this study, we investigated biomarkers of response to Len treatment in a cohort of 132 non del(5q) MDS patients, non-responders to erythropoiesis-stimulating agent (ESA) and enrolled in the GFM-LenEpo 08 clinical trial (NCT01718379). Patients were randomized to Len 10mg/day (L-arm) or Len 10 mg/day plus Epoetin beta (60,000 units/w) (LE-arm) and evaluated after 4 cycles. The biological studies were conducted in 99/132 patients including 41 responders and 58 non responders. According to IWG2006, HI-E was significantly higher in LE-arm (52%) vs. L-arm (31%) (p=0.031).

Extensive genotyping study of 29 genes (ASXL1, CBL, CSNK1A, DNMT3A, ETV6, EZH2, FLT3, IDH1, IDH2, IKZF1, IKZF3, JAK2, KIT, KRAS, MPL, NPM1, NRAS, PHF6, PTPN11, RIT1, RUNX1, SETBP1, SF3B1, SRSF2, TET2, TP53, U2AF1, WT1, ZRSR2) was conducted in the 99 patients by NGS or Sanger approaches. Found all along the coding sequence of the genes and with variable VAF, mutations in SF3B1 (73%), TET2 (46%), ASXL1 (20%) and DNMT3A (20%) genes did not influence the response in the L-arm or LE-arm. Previously identified in myeloma cells resistant to Len treatment, an IKZF1 mutation located in the exon 5 of the gene and affecting the binding domain of the protein has been reported during Len treatment in one patient with del(5q) MDS and was associated with a loss of efficacy of the Len treatment in vivo. In this cohort, we did not find any mutation in the exon 5 of IKZF1 or IKZF3 or in the exons 3 and 4 of CSNK1A which are described as affecting their domain of interaction with Cereblon. A A>G polymorphism in the 5’UTR region of CRBN gene (rs1672753) was significantly associated with HI-E in the whole cohort (41.5% in responders vs. 22.4% in non-responders; p=0.048). A gene expression profiling (GEP) study was conducted on BMMC of 50 at inclusion and 24 paired samples before and after 4 cycles of treatment. Using a Gene Set Enrichment Analysis (GSEA), the comparison of GEP in 24 paired samples linked the response in L-arm or LE-arm to a signature of 32 up-regulated genes exclusively involved in the immune response. A supervised GSEA analysis combined with a Pam R strategy of GEP before treatment identified a predictive signature of 36 up-regulated genes mainly involved in translation, epigenetic regulation of transcription, cell division and DNA repair. Slight variations of CRBN gene expression level were not correlated to the response. However, the resistance in L-arm and LE-arm was predicted by a low expression level of NPM1 (P<0.001) before treatment with a sensitivity of 86.7 % and a specificity of 92.8%. A Kaplan-Meier analysis confirmed that the probability of treatment failure was significantly higher in patients with low NPM1 expression (P=0.002). Futhermore, in patients with high NPM1 expression level, the mean delay for treatment failure was 5 months in patients with A/A CRBN polymorphism versus 17 months in patients with A/G or G/G polymorphisms (P=0.026).

In conclusion, resistance to Lenalidomide or Lenalidomide plus EPO seems not be associated with a particular mutational profile nor with mutations in recently identified Cereblon targets. By contrast, a low expression level of NPM1 associated with a A/A polymorphism of CRBN is highly predictive of a treatment failure by Lenalidomide or Lenalidomide plus EPO non del(5q) MDS patients.