ALBERT

Description: Abstract 1697 Background Abnormal and clonal hematopoiesis resulting in peripheral blood cytopenias, and risk of progression to acute myeloid leukemia (AML) are the main characteristics of myelodysplastic syndromes (MDS) which present a high diversity of somatically mutated genes. Recently, mutations targeting genes whose products participate to the early steps of RNA splicing (SF3B1, SRSF2, ZRSR2, and U2AF35) have been reported. Mutations in ASXL1, DNMT3, EZH2, IDH1/2 and TET2, suggest deregulation of the epigenetic control of transcription. Additional genes known to be mutated in MDS include RUNX1, TEL/ETV6, TP53 and NRAS The alteration of some of these genes may carry prognostic value. Next-generation sequencing analyses of AML and related disorders recently identified mutations in two chromosome X genes, BCOR and BCORL1, that code for related transcriptional co-repressors interacting with histone deacetylases and presenting specific properties. BCOR interacts with BCL6 and constitutional inactivating mutations have been described in the Oculo-Facial-Cardio-Dental syndrome. BCOR is also affected by mutations or translocations in retinoblastoma and sarcoma. BCORL1 has been implicated in chromosomal rearrangements in hepatocellular carcinoma. In this study, we have investigated BCOR and BCORL1 gene by Sanger sequencing in a cohort of 221 MDS samples. Methods The 221 MDS samples were collected at diagnosis in multicenter clinical trials in France between 1999 and 2011. The coding sequences of BCOR (ENST00000378444) and BCORL1 (ENST00000218147) were analyzed by Sanger sequencing. Mutational analyses of ASXL1, CBL, DNMT3A, ETV6, EZH2 IDH1/2, JAK2, NRAS, RUNX1, SF3B1, SRSF2, TET2, TP53, U2AF35, and ZRSR2 were previously reported. Non tumoral material (buccal swab or CD3+ cells) was analyzed for the presence of the identified variations of BCOR or BCORL1 when available (n=9). The prognostic impact of BCOR mutations was evaluated in MDS patients with available follow-up information (n=203). Results BCOR and BCORL1 coding sequences were analyzed in the cohort of 221 MDS patients and we found mutations in 8.6% (n=19) and 2.3% (n=5) respectively. Alterations were distributed all over the coding regions. Strikingly, two patients presented concomitant inactivating mutations of BCOR and BCORL1. Among the 19 BCOR alterations, 9 were missense, 3 nonsense, 5 frameshift and 2 splice site mutations. No significant difference in age, sex, karyotype, blood counts or bone marrow blasts between BCORmut and BCORwt patients was observed. Mutations were found in patients of all IPSS risk-groups and WHO subtypes. Comparison of cytological bone marrow reports revealed a trend for a higher rate of dysgranulopoiesis in BCORmut patients (P=0.06). Because only truncating mutations (splice, frameshift and nonsense mutations) are unambiguously expected to affect BCOR function, statistical analyses were restricted to truncating BCOR mutations (n=10/221, 4.5%). Truncating BCOR mutations were frequent in RUNX1mut patients (19% versus 3% in RUNX1wt patients; P=0.027). They were by trend associated with SRSF2 and DNMT3A mutations (P=0.09) and were exclusive with IDH1/2, JAK2, NRAS, TP53, and ZRSR2 mutations. In univariate analysis, Overall Survival (OS) and AML transformation rate did not differ between patients with either missense or truncating BCOR mutations and BCORwt patients (Figure A&B). However, inferior OS (P=0.034) and higher AML transformation rate by trend (P=0.051) were observed for patients with truncating BCOR mutations (Figure C&D). Multivariate analysis demonstrated that a truncating BCOR mutation was an independent unfavourable prognostic factor for OS (HR 3.3; 95%CI 1.4 – 8.1; P=.008). The low number of BCORL1 mutated patients precludes any statistical analyses but clinical follow-up was available for 4 out of 5 patients. Three patients died 7, 17 and 27 months after MDS diagnosis. The fourth patient received allogeneic bone marrow transplantation after AML transformation and is in complete remission 5 years after transplantation. Conclusion In summary, truncating mutations of BCOR were independently associated with a worse OS in MDS. These data support the idea that BCOR mutations appear as associated with RUNX1 and DNMT3A mutations as reported in AML. Moreover, and despite a relatively low mutation frequency in MDS, BCOR might be considered as a key gene in risk stratification. Disclosures: No relevant conflicts of interest to declare.

Description: Background: Lenalidomide (LEN) is a reference treatment in IPSS low and in1 (lower) risk MDS patients with isolated de(5q) (MDS-del(5q)) and RBC - TD. Most low risk MDS-del(5q) patients with anemia and independent of transfusions develop TD or need of treatment for symptomatic anemia very early after diagnosis (median time to transfusion/treatment of 20 months, López Cadenas et al abstract 3180 ASH, 2016). LEN directly targets the del(5q) clone improving anemia, quality of life and survival in this subset of patients. Limited data also suggest a role of LEN in non-TD patients with del(5q) (Oliva et al Cancer Med 2015). However no prospective randomized study of LEN has been performed in this group of patients. Material: The Sintra-Rev clinical trial is a phase III European multicenter study, in low-risk MDS-del(5q) patients, with anemia (Hb

Description: Abstract 2916 Background: Increased serum ferritin (SF) level has been associated with worse overall survival (OS) in lower risk MDS, presumably though iron overload in vital organs including the heart and liver (Cazzola, NEJM 2005, Malcovati, JCO 2005, Garcia-Manero G et al. Leukemia 2008). However, those high levels are generally associated with RBC transfusion requirement (Malcovati, JCO 2007, Sanz, Blood (ASH 2008 : Abstract 640). High SF level has also been reported in non transfused lower risk MDS, and attributed mainly to ineffective erythropoiesis, but its prognostic value in this context is unclear. We evaluated the prognostic value of SF level at diagnosis in lower risk non transfusion dependent (TD) MDS patients included in the GFM registry of MDS. Methods: We selected in the GFM Registry (1900 pts included between mid 2003 and 2010), 485 newly diagnosed IPSS low and int 1 (lower) risk MDS patients (pts), not transfused during the first 6 months of follow-up, who had a baseline SF level (normal values 20–300 ng/ml), and no other cause of increased SF, including infection and liver disease (due in particular to alcohol abuse). Pts with SF 〈 20ng/mL were excluded. The prognostic value of SF for AML progression and OS was analyzed. Results: Median age of the 485 pts was 77 years (range 29–103), with 53.8% males. WHO classification was RA 21.5%, RCMD 23.5%, RARS 23%, RCMD-RS 0.5%, RAEB-1 22% and 5q- syndrome 5%. Karyotype according to IPSS was fav (61%), int (10%), unfav (2%), not available (17%). IPSS was 0 (44%), 0.5 (31%), 1 (8%), ND (17%). The median level of SF was 276ng/ml (range 20–5558) with 225 (46%), 145 (30%) and 47(10%) pts having SF〉300ng/ml, 〉500ng/ml and 〉1000ng/ml respectively (resp.). In univariate analysis, male gender (P=0.0005), Hb level100μm3 (P=0.02), erythroblasts 〉30% in bone marrow (P=0.008), serum (s) EPO level〉100IU/l (P=0.04), RARS (P300ng/ml) while karyotype, IPSS, % marrow blasts, reticulocytes, platelets showed no correlation. In multivariate analysis, sEPO level〉100IU/l and RARS were significantly associated with higher SF level. As the SF threshold of 1000ng/mL is often proposed to start chelation therapy, prognostic analyses were also made for that level. Hb level 100 were associated with SF 〉1000 ng/ml in univariate analysis, while only Hb level

Description: Abstract 2841 Introduction: Myelofibrosis (MF) is a myeloproliferative neoplasm (MPN) characterized by bone marrow fibrosis, splenomegaly, cytopenias and constitutional symptoms. Ruxolitinib was recently approved by the FDA for the treatment of MF in the USA; its approval in Europe is still pending. However, EU patients may access to ruxolitinib through compassionate programs. In France, health authorities opened a compassionate patient-named program (Authorization for Temporary Utilization [ATU] program. Methods: 241 French patients (pts) with MF, including primary (PMF), post-polycythemia vera (PPV) and post-essential thrombocythemia (PET) MF were granted ruxolitinib therapy through ATU program, independently of their JAK2 mutational status, between April 15, 2011 and May 31, 2012. Physicians were asked to provide information on disease characteristics, treatment history, constitutional symptoms, spleen size, platelet and neutrophils count, as well as the ruxolitinib dose prescribed and adverse events (AE). Request forms had to be submitted at the time of initial application and every 3 months upon drug resupply or in case of treatment discontinuation. This analysis has been performed based on data available at baseline (n= 241), after 3 months (n= 101), 6 months (n= 57), 9 months (n= 21) and 12 months (n= 4). Results: In the entire cohort, 138 pts were men and 103 women. Median age was 68.3 years. 51.5% of pts had PMF, 22.8% PPV-MF and 23.8% PET-MF. 99.2% of pts had received ≥1 lines of therapy for MF prior to ruxolitinib (hydroxyurea: 56%; pipobroman: 15.4%; iMIDs: 13.7%; interferons: 13.7%; erythropoietins: 6.6%; spleen irradiation: 6.6%; anagrelide: 5.8%; corticosteroids: 4.9%). Despite these therapies, 93.7% had constitutional symptoms and 94.2% of patients presented a palpable splenomegaly (median 15 cm below costal margin) at inclusion. Efficacy: According to the baseline platelet count, ruxolitinib therapy was initiated at 15 mg BID in 132 pts (54.8%) or 20 mg BID in 103 pts (42.7%), or other doses in a minority of pts (n=6). Among the pts who were evaluable after 3 and 6 months of therapy, 96.5% and 90% presented a mean reduction in the spleen size (by palpation) by 47.2% and 46% from baseline, respectively. Constitutional symptoms resolved in 65.3% and 70.2% of pts at the aforementioned time points, respectively. In pts who completed 9 months follow-up (n=21), benefits in spleen size reduction and symptoms resolution were durable (95% and 71.4%). Safety: Since the beginning of the ruxolitinib ATU program, 83 pts presented at least one AE, including 27 pts with serious AE (SAE), with or without causal relationship to ruxolitinib therapy. AE, all grades, were essentially hematologic abnormalities (51.6%), gastro-intestinal 6.3%, cardiac 3.1%, musculoskeletal 3.1%, hepatic 2.3%, infection 2.3%. Dose adjustments were reported in 60 pts, mainly due to thrombocytopenia (n=36) and anemia (n=13). However, no patient discontinued ruxolitinib therapy because of cytopenia. Treatment was discontinued in 11 patients after a median duration of 2.6 months (range 0.3–7.9 months). Reasons for discontinuation were death (n=6), AE (n=2), inefficacy (n=2), and patient decision (n= 1). Conclusion: In this large, unselected population of heavily pretreated MF pts, ruxolitinib therapy appeared to be effective in treating both constitutional symptoms and splenomegaly, in line with previously reported efficacy in clinical trials. The safety profile seems also comparable. Comprehensive data and updated follow-up of the cohort will be presented. Disclosures: Off Label Use: compassionate use of ruxolitinib for myélofibrosis in France (indication not yet approved by EMEA). Rey:Novartis: Consultancy. Nicolini:novartis, Bristol myers Squibb, Pfizer, Ariad and Teva: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Recher:Celgène, Genzyme, Sunesis, Jansen-Cilag: Membership on an entity's Board of Directors or advisory committees, Research Funding, Travel to ASH Other. Ranta:Novartis: Membership on an entity's Board of Directors or advisory committees. Legros:Novartis, Bristol Myers-Squibb: Research Funding, Speakers Bureau, Travel to meeting Other. Viallard:Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees. Dupriez:novartis: Membership on an entity's Board of Directors or advisory committees. Coiteux:Novartis, Bristol Myers-Squibb: Speakers Bureau. Demory:Novartis: Honoraria. Giraudier:Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Ugo:Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees, Travel to ASH Other. Kiladjian:Incyte: Membership on an entity's Board of Directors or advisory committees; Shire: Membership on an entity's Board of Directors or advisory committees; Novartis: Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Research Funding. Roy:Novartis: Membership on an entity's Board of Directors or advisory committees, Research Funding, Travel to ASH Other.

Description: Background Induction therapy followed by autologous stem cell transplantation (ASCT) is the standard of care for patients with symptomatic multiple myeloma less than 66 years old. The quality of response to the induction treatment is an important prognostic factor and is predictive of progression free survival (PFS) following ASCT. The triplet combinations bortezomib-thalidomide-dexamethasone (VTD) and bortezomib-cyclophosphamide-dexamethasone (VCD) have demonstrated high response rates in prospective phase 2 and phase 3 clinical trials, they are 2 of the most commonly used induction regimens prior to ASCT, and are both recommended in the International guidelines. To date, no comparative data from prospective randomized trials of the safety and efficacy of VTD vs VCD are available. This provided the rationale for the phase 3 investigation of VTD vs VCD prior to ASCT in patients with de novo MM in this randomized multicenter study (NCT01564537). Methods Patients with de novo symptomatic MM less than 66 years old were prospectively randomized to receive either 4 cycles of VTD (arm A) or 4 cycles of VCD (arm B) followed by ASCT. The VTD regimen consisted of four 21-day cycles of bortezomib 1.3 mg/m²/d, subcutaneously (SC) D1, 4, 8 and 11, Dexamethasone 40 mg/d, PO D1 to 4, D9 to 12 and Thalidomide 100 mg/d, PO D1 to D21. The VCD regimen consisted of four 21-day cycles of bortezomib 1.3 mg/m²/d, SC D1, 4, 8 and 11, Dexamethasone 40 mg/d, PO D1 to 4, D9 to 12 and Cyclophosphamide 500 mg/m²/d, PO D1, 8, 15. Patients were stratified according to ISS (1-2 versus 3) and cytogenetics (high-risk defined by 17p deletion and t(4;14) versus other).The primary endpoint was very good partial response (VGPR) rate following 4 cycles. Response was assessed in a central lab according to the IMWG criteria. Assuming a VGPR rate of 60% in the VTD arm versus 45% in the VCD arm (15% difference), the possibility to detect a statistically significant difference required the enrolment of 340 patients overall (170 per arm). Adverse events were graded using the NCI CTCAE catalogue, version 4.0. Results From 11/2013 to 02/2015, 358 patients were enrolled into the study. 18 were screening failures, and 170 were randomized each to arm A (VTD) and arm B (VCD). The median age was 60 years (range, 26-65), 62% of the patients were male, and overall, the patient characteristics were well-balanced across the 2 arms of the study. The median number of induction cycles administered in both arms was 4 (1-4). On an intent-to-treat basis, the overall response rate (〉 partial response [PR]) was 92.3% in arm A, including a 10.7% complete response (CR) rate and a VGPR rate of 66.7%, while in arm B the overall response rate was 84%, with a 9.5% CR and a 56.2% VGPR rate. VGPR and PR rates were significantly higher in the VTD arm with p-values of 0.04 and 0.02, respectively. Seven patients died during induction therapy (2%), 3 in arm A from infections (2) and pulmonary embolism (1), and 4 in arm B from progression to extramedullary myeloma (2) and infections (2). Grades 3 / 4 peripheral neuropathy occurred in 4% and 2.2% in Arm A and B, respectively. Grade 3/4 neutropenia was seen in 11.9% vs 22.5% in Arm A and B, respectively. Conclusion This trial is the first prospective randomized comparison of 4 cycles of VTD versus 4 cycles of VCD administered as induction therapy prior to ASCT. VTD was shown to be significantly superior to VCD in terms of VGPR and PR rates. Neuropathy rates low in both arms, while neutropenia was more frequent with VCD. Disclosures Moreau: Millennium: Honoraria, Membership on an entity's Board of Directors or advisory committees; Bristol-Myers Squibb: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees; Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen-Cilag: Honoraria, Membership on an entity's Board of Directors or advisory committees. Hulin:Celgene Corporation: Honoraria; Janssen: Honoraria; Bristol Myers Squibb: Honoraria; Amgen: Honoraria. MACRO:celgene: Membership on an entity's Board of Directors or advisory committees; jansen: Membership on an entity's Board of Directors or advisory committees; millenium: Membership on an entity's Board of Directors or advisory committees. Garderet:Bristol-Myers Squibb: Consultancy. Stoppa:Celgene: Consultancy, Honoraria, Research Funding; Amgen: Consultancy, Honoraria; Novartis: Consultancy, Honoraria; Janssen: Consultancy, Honoraria. Facon:Janssen: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Millenium: Membership on an entity's Board of Directors or advisory committees; Onyx: Membership on an entity's Board of Directors or advisory committees; BMS: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Novartis: Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees; Pierre Fabre: Membership on an entity's Board of Directors or advisory committees. Laribi:Hospira SAS: Research Funding. Avet-Loiseau:jansen: Membership on an entity's Board of Directors or advisory committees; onyx: Membership on an entity's Board of Directors or advisory committees; BMS: Membership on an entity's Board of Directors or advisory committees; millenium: Membership on an entity's Board of Directors or advisory committees; millenium: Membership on an entity's Board of Directors or advisory committees; celgene: Membership on an entity's Board of Directors or advisory committees; jansen: Membership on an entity's Board of Directors or advisory committees; onyx: Membership on an entity's Board of Directors or advisory committees; BMS: Membership on an entity's Board of Directors or advisory committees. Attal:jansen: Honoraria; celgene: Membership on an entity's Board of Directors or advisory committees.

Description: Background: BIA-ALCL is a rare subtype of T cell lymphomas associated with textured breast implants which has recently been recognized. The pathogenesis of this entity remains elusive even if mutations in the JAK/STAT pathway have been identified. Little is known about the causes, prognostic factors of this disease, and treatment outcome. Methods: since 2016, a WebEx national multidisciplinary meeting has been implemented by the French Cancer Agency in order to better define therapeutic strategies for newly diagnosed cases after histologic confirmation. In the same time, BIA-ALCL registry was funded by LYSA in order to collect ambispectively, in France and Belgium, patient clinical data including reasons for breast implantation (breast augmentation, reconstruction), implant manufacturer, treatments and outcome. A biological program aiming molecular characterization of this T-cell lymphoma subtype has been set in coordination with the registry. Results: Fifty-eight patients (pts) have been analyzed so far among the 88 (67 in France and 21 in Belgium) identified from 2009 to 2019. Median age was 58 years (range 29-82) at diagnosis. In 29 out of the 58 pts (50%) the first implant followed a mastectomy for breast cancer. In this analysis, only implants in the breast(s) where the lymphoma occurred have been considered. Four pts (6.9%) had bilateral lymphoma and 54 pts had unilateral lymphoma (50% left side and 50% right side), 25 pts were implanted once (43.1%), 24 twice (41.4%) and 9 pts (15.5%) 3 times or more. The median delay between first implant and BIA-ALCL diagnosis was 11.9 years (range 4.1-37), and median delay from last implant to diagnosis was 6.5 years (range 0.2-25.4). The two clinical presentations i.e. seroma (n = 43, 74.1%) and breast tumor mass with or without seroma (n = 12, 20.7%) were most often correlated with the two distinct histological subtypes (in situ /mixed (n=41) or infiltrative (n=17). Three pts were diagnosed without any mass or seroma (1 lymph node involvement, 2 in the context of systematic implant removal). The majority of pts were stage I-II (n=45, 77.6%), and 13 (22.4%) pts were stage IV. One hundred and five implants have been used on lymphoma associated breast for these 58 patients. Considering available information regarding the type of implants, almost all patients had at least one silicone-filled (n=51) and at least one textured implant (n=49) with Biocell texturation (n=40, 69%). No patient had only smooth implant. Implant removal with total capsulectomy was performed in 49 patients and 17 underwent chemotherapy based mostly on CHOP or CHOP-like chemotherapy regimens (n=12) and brentuximab vedotin CHP (n=3). After 21 months of median follow-up, 52 pts are alive and free of evolutive disease and one was lost to follow up. Five pts have died, either from lymphoma progression alone (n=2), or associated with concomitant active breast cancer (n=2) and one due to another disease. All had an infiltrative histology, and the 2 patients who died from lymphoma were stage IV. All but one received systemic chemotherapy and one received palliative care only due to concomitant active breast cancer. One of these patients early relapsed after a first complete remission. After the BIA-ALCL diagnosis, breast reconstruction was performed in in 23 pts (39.7%), 17 with a new implant, lipofilling in 4 pts, with a flap in 4 pts, and one benefit from combined approaches. Whole exome sequencing and/or targeted deep sequencing was performed in 29 of these patients. Recurrent mutation of epigenetic modifiers were seen in 22 pts (76%) involving notably KMT2C (28%), CM2D (14%) and CREBBP (14%). Eighteen pts (62%) showed mutations in at least one member of JAK STAT signaling pathway including STAT3 (38%) and JAK1(21%). Conclusions: We here confirm that in situ BIA-ALCLs have an indolent clinical course and remain in complete remission mainly after implant removal. Infiltrative histological subtype which have a more aggressive clinical course should be precisely identified at baseline. In our series, most BIA-ALCL cases were associated with macrotextured implants with Biocell texturation observed in 69% of the cases. The molecular characterization of these cases highlights the key role of the JAK/STAT pathway, and the importance of epigenomics. Such observation provide basis to develop novel targeted therapies for patients with aggressive disease. Disclosures Le Bras: Takeda: Research Funding; Pfizer: Other: Travel grant; Jansen: Other: Travel grant. Haioun:novartis: Honoraria; celgene: Honoraria; roche: Consultancy; celgene: Consultancy; gilead: Consultancy; takeda: Consultancy; janssen cilag: Consultancy; amgen: Honoraria; servier: Honoraria. Bachy:Janssen Cilag: Honoraria; Janssen Cilag: Other: Travel, accomodation, Expense; Roche: Honoraria; Amgen: Honoraria; Roche: Consultancy; Gilead Science: Honoraria. Oberic:Roche: Membership on an entity's Board of Directors or advisory committees; Takeda: Membership on an entity's Board of Directors or advisory committees; Janssen: Honoraria. Tilly:Roche: Consultancy; roche: Membership on an entity's Board of Directors or advisory committees; servier: Honoraria; merck: Honoraria; Gilead: Honoraria; Janssen: Honoraria; BMS: Honoraria; Karyopharm: Consultancy; Astra-Zeneca: Consultancy; Celgene: Consultancy, Research Funding.

Description: Abstract 1880 Background: Red blood cell (RBC) cell transfusion dependency (TD) is an indicator of poor prognosis in IPSS low and int-1 (lower risk) MDS. In addition, median response duration to ESAs is only about 2 years (Park, Blood, 2008). AZA can lead to RBC transfusion independence (RBC-TI) in 30–40% of lower risk MDS (Lyons, JCO, 2009), but it has not been systematically evaluated in ESA-resistant lower risk MDS and it remains unknown if the combination of AZA and ESA would be useful in such patients (pts). Methods: In this randomized phase-II trial (GFMAzaEpo-2008-1 trial, NCT01015352), we compared AZA 75mg/m2/d for 5 days every 28 days for 6 cycles (AZA arm) to the same treatment plus epoetin beta 60000 U/week (AZA+EPO arm) in lower risk MDS. Inclusion criteria were: IPSS low or int-1 MDS resistant to ESA (i.e having received at least 12 weeks of EPO ≥60000 U/w or darbepoetin ≥250 μg/w or having relapsed after response to ESA), and with RBC-TD of at least 4 RBC units in the 8 weeks prior to enrollment. Responders in both arms were eligible for maintenance up to 12 monthly cycles, unless progression or loss of erythroid response occurred. The primary endpoint was major erythroid responses (HI-E major) after 6 courses, according to IWG 2000 criteria. Secondary endpoints included overall IWG 2000 HI-E, including major and minor, after 4 and 6 courses, response duration, IPSS progression, survival and toxicity. An interim analysis was planned after 49 of 98 planned patients were evaluable for response after 6 courses. Results: From Feb 09 to first of Jul 10, 96 pts were included (M/F=2:1); median age 72y (45-85); 3 pts did not receive any study drug and were excluded from the analysis (one consent withdrawal, one pancreatic cancer and one fatal cardiac event); 93 pts are the subject of this analysis (RARS=40, RCMD-RS=16, RCMD=12, RA=5, RAEB1=12, CMML=7, Unclassified=1). IPSS cytogenetics was favorable in 73, intermediate in 18, adverse in 1 (this pt was not excluded from the present analysis) and failed in 1 cases, with no imbalance for all these characteristics between arms. Overall, 68% of the pts were resistant to ESAs and 32% had lost response to ESA (after a median response duration of 32 weeks, range: 4–120). Median RBC-TD was 6 units (range: 4–16) in the 8 weeks prior to enrollment. Fifteen pts were too early for evaluation of response, 78 were evaluable for toxicity and 72 and 52 pts were evaluable for response after 4 and 6 courses, respectively, as 6 and 22 patients went off-study before 4 and 6 courses, respectively, due to toxicity or progression. Although overall HI-E rates were similar in the AZA and AZA+EPO arms, (40 and 36.5 % respectively, P=0.51), there was a trend for more frequent HI-E major after 6 courses (ie the primary endpoint of the study), in the AZA+EPO arm (7/22,32%), compared to the AZA arm (4/30,13%, P=0.17). Furthermore, in responding patients, the proportion of HI-E major was significantly greater in the AZA+EPO arm (87.5%) compared to the AZA arm (30%, P=0.03). Finally, a significant increase in HI-E major between 4 and 6 cycles was noted only in the AZA+EPO arm (P=0.016). Seventeen responding patients entered the maintenance phase. Of the 78 pts evaluable for toxicity, 22 (28%) had to be hospitalized at least once for an anemia-related event (N=6) and/or a clinical infection/febrile neutropenia (N= 16). Interestingly, only 6 pts had to be hospitalized in the AZA+EPO arm, compared to 16 in the AZA arm, (P=0.o4). Conclusions: In this first randomized study comparing AZA and AZA+ EPO in highly transfusion-dependent lower-risk MDS, this planned interim analysis shows a promising trend for more major HI-E in the AZA+EPO arm, suggesting that addition of EPO to AZA increases the frequency of major erythroid responses in those patients, and may also significantly decrease the hospitalization rate due to infectious and non-infectious complications, allowing more patients to receive prolonged treatment with azacitidine. Updated results will be presented at the meeting. Disclosures: Récher: Celgene: Consultancy, Honoraria, Research Funding. Vey:Celgene: Consultancy, Honoraria, Research Funding. Fenaux:Celgene: Consultancy, Honoraria, Research Funding. Gardin:Celgene: Consultancy, Honoraria, Research Funding.

Description: Abstract 3519 Poster Board III-456 Introduction Romiplostim is a thrombopoietic agent that has been unequivocaly demonstrated as highly effective in adult's ITP in prospective studies and has recently been licensed for adults with chronic ITP in USA, Europe, Canada, and Australia. France has been the only country where romiplostim could be given for a compassionate use outside clinical studies from January 2008. The official indication for obtaining romiplostim in this setting was: chronic ITP according to the international criteria and failure or relapse after at least one previous line of therapy regardless the status towards splenectomy. We report here the data on safety and efficacy of the first consecutive 80 ITP patients who have been registered by the French health authorities as receiving the treatment “off-label” and with at least one-year of follow up. Patients and methods The data were retrospectively reviewed using a standardized form. The protocol was approved by local ethical committee. Patients who did not fulfil official indication criteria of compassionate use were excluded (i.e. secondary-ITP). Platelet count was monitored at least monthly during the follow-up. Platelet response was defined as platelet count of 50×109/L or more and a doubling of the pre-treatment count in the absence of any rescue medication within the last 8 weeks. One-year sustained response was defined as a platelet response on at least 2 of the last 3 platelet determinations at month-10-11 and 12. Patients who received rescue medication at any time during the study could not be counted as having a one-year sustained response. Report of adverse events was captured using a standardized form. Results Among the 80 patients, 8 were actually excluded from the analysis (6 with secondary ITP and 2 have been misdiagnosed as having ITP). The analysis was then conducted on 72 patients (43 females) with a median age of 60 years (20 to 91). The median duration of ITP prior to romiplostim administration was 8.7 years (0.1 to 49) and the patients had received a median of 5 (2 to 12) treatment-lines before romiplostim including: corticosteroids (100%), rituximab (65/72, 90%) and splenectomy (39/72, 54%). Among the 33 non-splenectomized patients, 13 patients were reluctant to undergo splenectomy whereas splenectomy was considered as contra-indicated in 20 of them. At time of romiplostim first administration, median platelet count was 16×109/L (1 to 60) and 48 patients (66%) were receiving a concurrent treatment for ITP, including mainly steroids (n=29) ± immunosuppressive drugs (n=8). A platelet response was observed at least once in 76% (55/72) of the patients. On average, patients who responded at any point during the study had a platelet response during 64% of the time (range: 37 to 100%). Romiplostim was stopped in 28 % (20/72) of patients for either lack of efficacy (n=16), for intolerance (n=1) or because it was administered only transiently in preparation for surgery (n=3). Two patients had died respectively from a septic shock and from an ITP-related intracranial hemorrhage. At one-year of follow up, 52 patients were still receiving romiplostim at a median dose of 6.5 μg/kg per week. This dose remained relatively stable as after the first 12 weeks of treatment, romiplostim could be pursued at a stable dose ± 2 μg/kg in 82% of the cases. Twenty % (14/72) of patients received a rescue medication during the study and 50% (36/72) of the patients had a one-year sustained response. The percentage of one-year sustained response was similar in splenectomized and non splenectomized groups [respectively 54% (21/39) and 45% (15/33), NS)]. Among the 29 patients who responded to romiplostim and who were receiving a treatment at time of romiplostim initiation, 86 % (25/29) had discontinued this medication and a further 7% (2/29) had reduced the dose by at least 25%. Only one patient stopped romiplostim because of an adverse event (headache). The most frequent otherwise reported adverse events were: arthralgia (26%), fatigue (13%) and nausea (7%). A transient thrombocytosis 〉 400×109/L and 〉 1000×109/L has been observed in respectively 19% (14/72) and 4% (3/72) of patients. A transient stroke occurred in 2 elderly patients (age 〉 70 yrs). No deep vein thrombosis occurred. Myelofibrosis was not observed. Conclusion Our study confirms in “real-life” that romiplostim is definitely an effective and safe treatment for severe chronic ITP in both non-splenectomized and splenectomized adults. Disclosures: Godeau: AMGEN: Consultancy.

Description: Introduction: Myelodysplastic syndromes (MDS) are characterized by ineffective hematopoiesis and peripheral cytopenia. In about half of patients with lower-risk (LR) MDS, thrombocytopenia is present at the time of diagnosis and associated with shortened survival and an increased risk of progression to acute myeloid leukemia (AML). The thrombopoietin receptor agonist (TPO-RA) romiplostim has shown safety and marked efficacy in a still poorly-defined subset of LR-MDS patients with thrombocytopenia. Methods: The EUROPE multicenter phase 2 trial within the EMSCO network investigated the impact of biomarkers like endogenous thrombopoietin (TPO) level and platelet transfusion events (PTE) on the efficacy of romiplostim (750µg SC qw) treatment in patients with LR-MDS (IPSS low/int-1). Patients were eligible if baseline bone marrow blast count was

Description: Romiplostim, a thrombopoietic agent with demonstrated efficacy against immune thrombocytopenia (ITP) in prospective controlled studies, was recently licensed for adults with chronic ITP. Only France has allowed romiplostim compassionate use since January 2008. ITP patients could receive romiplostim when they failed to respond to successive corticosteroids, intravenous immunoglobulins, rituximab, and splenectomy, or when splenectomy was not indicated. We included the first 80 patients enrolled in this program with at least 2 years of follow-up. Primary platelet response (platelet count ≥ 50 × 109/L and double baseline) was observed in 74% of all patients. Long-term responses (2 years) were observed in 47 (65%) patients, 37 (79%) had sustained platelet responses with a median platelet count of 106 × 109/L (interquartile range, 75-167 × 109/L), and 10 (21%) were still taking romiplostim, despite a median platelet count of 38 × 109/L (interquartile range, 35-44 × 109/L), but with clinical benefit (lower dose and/or fewer concomitant treatment(s) and/or diminished bleeding signs). A high bleeding score and use of concomitant ITP therapy were baseline factors predicting romiplostim failure. The most frequently reported adverse events were: arthralgias (26%), fatigue (13%), and nausea (7%). Our results confirmed that romiplostim use in clinical practice is effective and safe for severe chronic ITP. This trial was registered at www.clinicaltrials.gov as #NCT01013181.