Search

Background: Herpes simplex virus (HSV) oral excretions are common after bone marrow transplantation (BMT). Objective: We report a 4-year systematic survey of HSV excretions in an adult population who underwent BMT (289 transplantations). Study design: Patients received either intravenous ACV treatment when mucositis occurred or systematic intravenous ACV prophylaxis from initiation of the BMT conditioning until the end of aplasia. All patients were followed up for HSV excretions. Results: Twenty-eight patients (9.7%) excreted HSV. The occurrence of HSV excretions was similar in both allogeneic and autologous transplant patients. The incidence was significantly lower when ACV was systematically used after transplantation (2.5%) compared to when ACV was implemented for mucositis (12%). ACV-resistant HSV was detected in three patients who received allogeneic tranplantation, representing 27% of allogeneic recipients excreting HSV. Conclusion: HSV infection prophylaxis with high dose of intravenous ACV resulted in a decreased incidence of HSV excretion. Nevertheless, the risk of emergence of ACV resistance, especially among allogeneic transplant patients, appears to be identical whatever the route and dose of ACV prophylaxis.

Over the past decade, great progress has been made in the treatment of adult acute lymphoblastic leukemia (ALL). Improvements in the quality and availability of broad-spectrum antibacterial antibiotics, antifungal agents, and platelet transfusions, as well as greater understanding of the optimal use of a variety of antineoplastic agents, have resulted in gradual improvements in CR rates and the identification of a significant fraction of patients achieving long-term disease free survival [6, 9, 11, 12, 13, 16, 17]. Stem cell transplantation (SCT) is one of the postremission strategies for the eradication of residual disease in ALL. Long-term disease-free survival for patients with ALL after management with chemotherapy plus SCT is now widely accepted [7, 8, 20]. However, little data is still available on long-term survival beyond 3 years.

Background Hematopoietic stem cell transplantation (HSCT) has contributed to improve outcome for pediatric patients with acute leukemia (AL). However, post-transplantation relapse is still associated with a dismal prognosis and its optimal treatment remains unclear. Therapeutic strategies may vary according to the relapse delay, as well as centers and child's specific requirements. Aim We aimed to compare patients' related factors and treatment strategy, in case of relapse or progression post-allogeneic HSCT in children with acute leukemia in a recent ten-year period. Methods We analysed all consecutive 334 children ( Results 279 patients (110 females and 169 males) with a median age of 8.8 years at transplantation (range: 0.52-17.99) were included. Diseases were 149 ALL (B-cell phenotype, n=120; T-cell phenotype, n=29), 122 AML and 8 biphenotypic AL. At transplantation, 233 patients were in complete remission (CR) (CR1 n=121, CR2 n=94, CR3 n=18) while 46 had a cytologically detectable disease. Donors were matched related siblings (36%), mismatched related (4%), matched unrelated (18%), mismatched unrelated (24%), unrelated without precision (16%) or syngenic (2%). Stem cell source was bone marrow, peripheral stem cells, umbilical cord blood in 65%, 12%, 23% of cases, respectively. Median time from diagnosis to first HSCT was 263 days. 91% (n=254) children received myeloablative conditioning and 9% (n=25) received reduced intensity conditioning regimens. Total body irradiation (TBI) was performed in 153 children. Acute GVHD after first HSCT occurred in 157 patients (grade I n= 62 grade II to IV n= 94). The median delay from first HSCT to relapse/progression was 182 days. The treatments for relapse after first HSCT consisted in chemotherapy alone (n=103), chemotherapy followed by second HSCT (n=70), supportive and palliative care (n=66), combination of chemotherapy and DLI (n=27), or isolated reinfusion of donor lymphocytes (DLI) (n=13). At the time of analysis 40 patients (12%) were alive with a median follow up of 1315 days (range 58;4182). The median OS duration after relapse was 149 days among the 334 patients. The median survival duration in days according to therapy was as follows [confidence interval]: DLI after chemotherapy = 407 [156;658], Second allograft: 391 [264;518], chemotherapy: 180 [121;239], DLI alone: 140 [10;270], Palliative care: 43 [33;53] (see Figure 1). A Cox model analysis on 260 patients with complete files showed the impact of the following factors on survival, when comparing to the best line of treatment, ie chemotherapy+DLI; treatment by chemotherapy alone (HR 2.05, p=.008), palliative care (HR=5,7, p270 days yielded hazard ratios of respectively .549, .479 and .293 (p There was no impact on outcome from the following variables: gender, age at transplant, leukemia subtype, type of conditioning, conditioning with TBI , GVHD occurrence after the first transplant, stem cell source. Conclusion Despite evident limitations due to the retrospective setting of our analysis, best treatment consisted in second transplantation or the combination of chemotherapy plus DLI, both yielded similar long-term results. Those results favor the use of immune based strategies when possible in this particular situation of relapse post-HSCT, which should be evaluated in prospective trials. Figure 1 Figure 1. Disclosures Mohty: Gentium S.p.A./Jazz Pharmaceuticals Inc.: Honoraria, Research Funding.

Introduction: De novo Ph+ acute myeloid leukemia (AML) has recently been described as a distinct AML subtype with a specific genome signature compared to chronic myeloid leukemia (CML). Although Allo-SCT is often proposed as the curative option, the outcome of Ph+ AML treated with Allo-SCT remains unknown in the literature. Methods: Between 2000 and 2012, 19 patients with Ph+ AML who received Allo-SCT were identified in France and compared to 21 patients with Ph+ biphenotypic acute leukemia (BAL) and 52 patients with myeloid blast crisis CML (MBC-CML). Clinical data were prospectively collected though ProMISe (Project Manager Internet Server), an internet-based system shared by all centers of The French Society of Bone Marrow Transplantation and Cell Therapies. The study was approved by local ethical committee. Results: For AML, BAL and MBC-CML patients, the median age per group was respectively 46 (18-67), 36 (18-52) and 36 (18-63) years. Median follow-up was 24 months. Additional cytogenetic abnormality was not different between the 3 groups. The MBC-CML group had a lower rate of induction therapy before Allo-SCT compared to the 2 other groups and a lower rate of remission status before transplant (p Conclusion: Limited data are available on de novo Ph+ AML. This study showed a promising 65% two-years overall survival in Ph+ AML subtype in comparison with others Ph+ acute leukemia. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.

Background: Multiple myeloma (MM) in relapse after allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a challenging situation because of particular aggressive disease, frail patients and a specific immune context. Our previous data suggested that lenalidomide salvage therapy could lead to an overall response rate (ORR) of 83%, and a progression free survival (PFS) of 18 months. However, this was achieved at the cost of 25% incidence of de novo acute graft versus host disease (aGVHD). On the other hand, bortezomib has both anti- myeloma activity and a potential ability to modulate GVHD. With this background, this study aimed to evaluate the efficacy, tolerance and immunomodulatory effect of bortezomib salvage therapy post allo-HSCT. Patients: This retrospective study included 97 patients (15 SFGM-TC centers) with MM who relapsed after allo-HSCT and received bortezomib as salvage therapy any time post allo-HSCT. Median age at allo-HSCT was 50 years (range: 29-64). Most of the patients had high-risk MM (72% with poor-risk cytogenetics) and received a median of 2 lines (range: 1-5) before allo-HSCT. Previous autologous HSCT was performed in 95% of cases (2 autologous HSCT in 25% of cases). Allo-HSCT was mainly performed after a reduced intensity conditioning (94%) from a sibling donor (67%). The median time of relapse/disease progression after allo-HSCT was 15 months (range, 1-84). Bortezomib treatment started at a median of 24 months (range: 1-97) after allo-HSCT (first line salvage therapy in 53% of cases). Results: Patients received a median of 4 cycles of bortezomib (range, 0.5-13), combined with dexamethasone in 66% of cases. ORR was 58% (16% complete response, 20% very good partial response, 22% partial response), with the best response being achieved after a median of 4 cycles (range: 1-7). ORR was independent of cytogenetic status at diagnosis (p=0.19), dexamethasone use (p= 0.39) and bortezomib introduction (first line therapy post relapse after allo-HSCT versus beyond first line) (p=0.68). With a median follow-up of 34 months (range: 1-112) after bortezomib administration, the median PFS was 7 months (range: 1-86) and the median overall survival (OS) was 27 months (range: 1-112). Dexamethasone association, cytogenetic status and timing of bortezomib introduction did not impact PFS or OS. Excluding patients who relapsed before the 4th cycle of treatment (n=25), we show that prolonging bortezomib administration > 4 cycles (n=62), independently of being in response at the 4th cycle, improved PFS from 6.8 to 13.5 months (Hazard ratio (HR)=0.4, 95%CI: 0.2-0.7, p=0.01) while the statistical significance wasn't reached for OS (increase from 26.2 to 39.1 months; HR=0.6, 95%CI: 0.3-1.2, p=0.15) (Figure 1). During bortezomib therapy, no occurrence of acute GVHD was observed, whereas 7 patients (7%) developed chronic GVHD (cGVHD) (4 de novo cases and 3 worsening of pre-existing cGVHD). On the other hand, 7 cases of GVHD (6 chronic and 1 acute) improved during bortezomib therapy. Bortezomib was administered intravenously for all patients. Usual grade 3/4 hematological (21% thrombocytopenia), neurological (16%), fatigue (16%) and infectious (9%) side effects were observed. Toxicity led to treatment discontinuation in 27% of patients. Most of the side effects (96%) were reversible without any death due to treatment toxicity. Conclusion: Bortezomib is safe and effective in case of MM patients in relapse post allo-HSCT. ORR was about 58% with a median PFS of 7 months. A prolonged administration (> 4 cycles) may improve PFS. The efficacy of bortezomib in this situation does not seem to be linked to an enhanced graft versus myeloma effect since we observed both improvement and worsening of GVHD after the introduction of this treatment. These data contrast to our previous report on lenalidomide and our results will be discussed in order to identify the best treatment option in this particular difficult medical situation and to lead to further prospective studies. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.

Introduction Hematopoietic stem cell transplantation (HSCT) is a therapeutic option for the treatment of relapsed anaplastic large cell lymphoma (ALCL) in children. To date, few pediatric reports have assessed its efficiency and tolerance. While processing the evaluation of new targeted drugs, the impact of HSCT have to be precised in this population. Patients and methods We retrospectively analyzed the data all patients (n=34, median age 7.4y [1.1-17]) registered in the SFGM-TC database, who underwent a HSCT for the treatment of an ALCL ALK+ between 1993 and 2011. Central histology review was carried out for 29/34 patients. Most of patients were treated in specific french protocols (HM, n=4 and ALCL99, n=27). Indications of HSCT were based on the european group EICNHL recommandations for 14 patients : on-therapy relapse (n=1) and CD3 positivity (n=13). For 14 other patients, the decision of HSCT relied on one or several of the following criteria : leukemic form (n=6), post autologous transplantation relapse (n=5) and medullar positivity of the NPM1-ALK fusion transcript at relapse (n=7). In the 6 last patients, none risk factor was identifed. At transplant, 26 patients (65%) were in complete remission (CR) (CR1 n=1, CR2 n=18, CR3 n=4, CR4 n=3) whereas 8 (23,5%) had a detectable disease. Five patients had an history of previous autologous transplantation while 1 patient received 2 subsequent HSCT. Of the 35 HSCT performed, donors were HLA-matched related in 11 cases, mismatched related n=1, matched unrelated n=9 and mismatch unrelated n=14 (including 10 cord blood units). Conditionning regimens were mostly myelo-abaltive (n=30) with the use of a total body irradiation of 12 Gy in 26 patients. Results Engraftment was observed in 32 patients (median time for neutrophil recovery : 22d [9-44]).With a median follow up of 2.9y [0,1-11,8], 2-year (2y) overall and event-free survivals were respectively 69,6%+-8% and 58,1%+-9%. Cumulative incidences of relapse and non-relapse mortality were 17,8%+-6% and 24,2%+-7% respectively. Six patients (including 3 without CR at transplant) relapsed (median time 108 days [35-235]). Prolounged CR was obtained in 4/6 patients after donor lymphocytes injection (n=1) or Vinblastine-corticosteroid treatment (n=3, with a follow-up of 6y, 2y and 8 months). Twenty-five patients (73.5%) presented an acute GVHD and 5 (14.7%) a chronic GVHD. Eventually, 10 patients died, including 8 of transplant toxicity (5 out of the 8 had been allografted before 2003). Conclusion This report shows that HSCT is an efficient treatment in pediatric patients with high risk relapse of ALK+ ALCL. However, the high level of treatment-related mortality rises the question of the place of alternative treatments as Vinblastine, ALK or CD30 inhibitors in place of HSCT. When allograft is required, the use of reduced-intensity conditionning could be help reduce toxicity in these heavily pre-treated patients. Disclosures: No relevant conflicts of interest to declare.

Introduction Outcomes after RIC allo-SCT in children are rarely reported. The largest series published thus far included 47 children with hematologic malignancies (Pulsipher et al, Blood, 2009). There is thus currently no large series describing the results of RIC allo-SCT in children. Patients and Methods 163 pediatric patients ( Results Engraftment was observed in 85.5% of cases with a median time for neutrophils recovery of 20 days (range: 1-79). With a median follow-up of 23 months (range: 1-169), 2-year (2-y) OS, DFS, RI and NRM were 46+-5%, 37.5+-4%, 40+-3% and 22+-3%, respectively. Cumulative incidence (CI) of acute GVHD at day+100 was 36+-4% while 2-y CI of chronic GVHD was 24.5+-3%. In univariate analysis, factors such as disease status (CR vs active), donor type, source of stem cells, TBI or ATG administration were not associated with outcomes. There were a trend for better 2-y DFS in patients allotransplanted after 15.5 months from the diagnosis (42+-6% vs 33+-6%, p=0.09), for higher 2-y RI but a lower 2-y NRM in patients allografted before 10 years old (52+-7% vs 35+-5%, p=0.07; 12+-5% vs 35+-5%, p=0.06, respectively) and finally a trend for better 2-y OS in patients with lymphoid malignancies (54+-7 vs 41+-6%, p=0.09). Patients allografted after 2004 (n=122; median age: 13.2 years (range: 1-17.9); male 60%; myeloid malignancies 59%, CR: 63%) showed significantly higher 2-y OS (52+-5% vs 31+-7%, p=0.006) and DFS (45+-5% vs 18+-6%) due in part to a significantly lower 2-y NRM (17+-4% vs 36+-8%, p=0.01) reflecting probably a better patient’s selection before transplant. In patients allografted since 2004 multivariate analysis revealed that lymphoid malignancy was the only favourable and independent factor for LFS and OS (HR: 0.55, 95%CI: 0.32-0.97, p=0,04; HR: 0.47, 95%CI: 0.25-0.88, p=0.02). Conclusion This large series shows that RIC allo-SCT is feasible with acceptable toxicity in pediatric patients, especially in cases of lymphoid malignancy. Prospective studies are needed to confirm the potential role of RIC allo-SCT in the pediatric setting, with a special focus on long-term side effects, which are a matter of concern in children. Disclosures: Mohty: Novartis: Honoraria, Research Support Other.

Up to now allo-HSCT is the only curative treatment for patients with myelofibrosis (MF) (Ballen, et al, Guardiola, et al 1999, Kroger, et al 2007, Robin, et al 2011). The engraftment rate ranges from 70% to 95% and is influenced by spleen size (delayed in patients with splenomegaly and faster in patients splenectomized prior to allo-HSCT). JAK1/JAK2 inhibitor Ruxolitinib is reported to decrease spleen size and constitutional symptoms. These effects could also be benefit for patients eligible for transplantation, by reducing their splenomegaly and improving their performance status, factors known to impact the results of allo-HSCT. The present study analyzes outcome of patients with MF who have been reported to the SFGM-TC registry. From 2008 to 2013, 11 patients with primary MF (n=7), MF secondary to PV or ET (n=3) or acute myeloid leukemia secondary to MF (n=1) were treated with Ruxolitinib before allo-HSCT. Median age at allo-HSCT was 54 years (range: 44-66); there were 9 men and 2 women. Nine patients had the JAK2V617F mutation. At Ruxolitinib initiation, median white blood cells were 4.4G/L (3.5–13.7), hemoglobin was 10g/dL (7.5-13.3) and platelets were 99G/L (41-290); all patients had palpable splenomegaly and 8 had constitutional symptoms. Lille score was low in 4, intermediate in 2 and high in 4 patients, respectively. Cytogenetics were favorable in 2, unfavorable in 3 (complex karyotype) and not done or failure in 6 patients, respectively. The median time between diagnosis and treatment with Ruxolitinib was 353 days (16-2687) and median time from diagnosis to allo-HSCT was 433 days (199–2793). Ruxolitinib was progressively tapered in 3 patients (with steroids for 2 of them) and discontinued without tapering in 4 (missing data for 3). Median time between Ruxolitinib withdrawal and allo-HSCT was 10.5 days (4-66). All patients received reduced intensity conditioning regimen based on Fludarabine and a graft-versus-host disease (GVHD) prophylaxis including ciclosporine. Donors were: HLA matched sibling in 4, matched unrelated donor in 3 and mismatched unrelated donor in 4 patients. All but one patients received peripheral stem cell transplantation (one patients received bone marrow). Before allo-HSCT, 7 patients were in partial response and 4 were stable. After Ruxolitinib treatment, 8 patients had an estimated reduction of spleen size > 25% and 2 underwent splenectomy. There was no progressive disease on therapy and only one patient experienced a grade III-IV hematological toxicity. All patients had neutrophil engraftment (>0.5G/L) in median on day +17 and 10/11 patients had platelet engraftment (> 20G/L) in median on day +21. Chimerism was full donor in 8, mixed in 1 (who relapsed at day 178), and not available in 2 patients, respectively. One patient who was 100% donor chimerism remains pancytopenic and transfused 90 days after allo-HSCT. Three patients experienced grade II and 2 grade IV acute GVHD after a median time of 18 days after allo-HSCT. Acute GVHD was refractory to corticosteroids in 2 patients and was the cause of death in 1 patient. Only 2 patients had chronic GVHD. Median time from Ruxolitinib initiation to last follow-up was 339 days (166 – 1928), 6 (54%) patients were assumed in complete remission (blood cell count normal and 100% donor chimerism) and 9 (80%) were alive. Two patients relapsed on days 77 and 148 and one of them received a second allo-HSCT. In this retrospective study, Ruxolitinib before allo-HSCT seems to be well tolerated with excellent engraftment rate. A prospective study (JAK-ALLO) is ongoing in France on behalf of SFGM-TC to precisely analyze the role of Ruxolitinib before allo-HSCT. Disclosures: Kiladjian: Novartis, Celgene, AOP Orphan: Research Funding; Novartis, Sanofi, AOP Orphan: Honoraria; Novartis, Sanofi, AOP Orphan: Membership on an entity’s Board of Directors or advisory committees. Mohty:Novartis: Honoraria, Research Support Other. Robin:Novartis: Research Funding.

Purpose The pediatric-inspired intensified GRAALL protocol yielded a marked improvement in the outcome of adults with Ph-negative acute lymphoblastic leukemia (ALL) (Huguet et al,JCO 2009), raising the issue of the place of allogeneic HSCT in this new context. We report here the results associated with HSCT in first complete remission (CR) in younger adults treated in the GRAALL-2003/2005 trials. Patients and methods In these trials, HSCT was offered in first CR to patients aged 15-55 years with ALL at higher risk of relapse. High-risk factors included white blood count >30 x 109/L for B-lineage ALL, central nervous system (CNS) involvement, t(4;11)/MLL anomalies, t(1;19), low hypodiploidy/near triploidy, complex karyotype, early resistance to the steroid prephase (CsR), early resistance to chemotherapy (ChR) as assessed by poor bone marrow blast clearance at day 8, and late CR. Per protocol, HSC donor should be a matched sibling or an unrelated donor (10/10 or eventually 9/10 HLA matched) and conditioning regimen should include single fraction 10-Gy or fractionated 12-Gy total body irradiation (TBI) and high-dose cyclophosphamide. The role of HSCT was evaluated by HSCT versus no-HSCT cohort comparison (Mantel-Byar time-dependent analysis). Results Among the 522 high-risk patients eligible, 283 (54%) actually received HSCT in first CR (185 B-lineage and 98 T-lineage ALL; median age, 31 years). There were no significant differences in baseline characteristics and early response between HSCT and no-HSCT cohorts, except more HSCT patients with t(4;11)/MLL anomalies. Among the 283 HSCT patients, 46 had t(4;11)/MLL anomalies, 13 had CNS disease, 116 had CsR and 176 had ChR ALL. Origin of HSC was 140 sibling and 143 unrelated donors, including 47 unrelated donors with 1 HLA mismatch. The following HSCT protocol violations were observed: a) conditioning did not include TBI in 17 patients; b) 10 patients received reduced-intensity conditioning (RIC); and c) 13 patients received cord blood HSCT. These patients remained in the HSCT cohort for the analysis. With a median post-transplant follow-up of 3.5 years, post-HSCT cumulative incidence of relapse (CIR), non-relapse mortality (NRM) and relapse-free survival (RFS) were estimated respectively at 19% (95% CI, 15-25), 16% (95% CI, 12-21) and 64% (95% CI, 58-70) at 3 years, without any RFS difference between sibling and unrelated donors. A marked trend for a higher NRM was observed in patients aged 45 years or more (26% versus 13% at 3 years; HR, 1.6; p=0.059), while post-transplant CIR was similar in older and younger patients. Mantel-Byar RFS estimations showed no significant difference between both HSCT and no-HSCT cohorts (HR, 0.80; p=0.13), meaning that the risk factors used in these protocols failed to identify patients who could significantly benefit from HSCT in first CR. As minimal residual disease (MRD) had a major prognostic impact, we retrospectively analyzed the effect of HSCT in the 278/522 patients evaluated for 6-week Ig/TCR MRD level (154 HSCT and 124 no-HSCT patients, including 92 and 62 patients with MRD >=10-4, respectively). Mantel-Byar RFS estimations showed that HSCT did not benefit to patients with MRD =10-4 (HR, 0.6; p=0.02), with a positive HSCT-by-MRD interaction (p=0.04) (Figure 1). Conclusion In adult patients with high-risk ALL in first CR treated in the GRAALL-2003/2005 trials, myeloablative allogeneic HSCT from sibling and unrelated donors was associated with a 64% RFS at 3 years. The 26% NRM observed at 3 years in patients aged 45 years or more suggests considering RIC-HSCT in these patients. MRD response appears to be a powerful tool to select patients who might benefit from HSCT. This selection tool will be prospectively evaluated in the next GRAALL-2013 trial. ND and AH contributed equally to the work. Disclosures: No relevant conflicts of interest to declare.

1947 Introduction Sclerotic cGvHD is one of the most severe complications after allogeneic hematopoietic stem cell transplantation (HSCT). The sclerotic skin manifestations of cGvHD result from inflammation and fibrosis of the dermis, subcutaneous tissue, or muscular fascia, leading to functional disability. Risk factors associated with sclerotic cGvHD remain not well studied. TBI, PBSC, female recipient, and prior severe skin acute GvHD were identified as factors associated with an increased risk of sclerotic cGvHD in a large previous retrospective study from Seattle (n=986 patients) which was presented at the ASH 2011 meeting. Objective In the aim to validate the previous risk factors in France, we have conducted this retrospective multicentric analysis. We present here the preliminary results from four centers, seven are expected in the next months. Between 2005 and 2010, 437 consecutive patients who received systemic therapy for chronic GvHD after a first allogeneic HSCT were included. Chronic GvHD was diagnosed by the NIH consensus criteria. Sclerotic GvHD was defined by manifestations of cutaneous sclerosis, fasciitis or joint contracture in the medical record. Analyses to determine risk factors included as variable patient and donor age, kind of donor, HLA matching, diagnosis, stem cell source, patient and donor gender, intensity of conditioning regimen, use of total body irradiation (TBI) or antithymocyte globulin (ATG) in the conditioning regimen, and prior acute GvHD. Results All patients had hematological malignancies, the median age is 45 (17–65), 227 patients (52%) had HLA-matched related donors, 125 (29%) had HLA-matched unrelated donors, 48 (11%) had HLA- 9/10 unrelated donors and 37 (8%) had one or two cord blood unit (HLA-4/6 or 5/6 matched). The median time from allogeneic HSCT to chronic GvHD was 6.5 (1.8–41.5) months; 69 patients (15.7%) presented sclerotic features at initial diagnosis or after the onset of chronic GvHD. The cumulative incidence of sclerotic cGvHD was 12% (95%CI, 11–13.5) at 5 years after onset of chronic GvHD ([Figure][1]). From the 69 patients who experienced sclerotic cGvHD, 49 patients (71%) had HLA-matched related donors, 19 patients (27.5%) had HLA-matched unrelated donors, and 1 patient (1.5%) had HLA-mismatched unrelated donor. In univariate analysis, we found more sclerotic cGvHD in patients who had multiple myeloma (p=0.018); this effect seems to be associated with the prophylaxis for GvHD with CsA alone. There was no effect of TBI or intensity of conditioning. There was more sclerotic GvHD in female recipient (p=0.032) and for patients who had an unrelated donor (p=0.0028). In multivariate analysis ([Table][2]), factors associated with an increased risk of sclerotic cGvHD were female recipient, patients with unrelated donor, and patients with previous aGvHD. Cord blood as stem cell source was associated with a lower risk of sclerotic GvHD. Prophylaxis of GvHD with CsA alone showed a trend to a higher risk. At the last follow-up, of the 69 patients, 51 (74%) were alive and 10 (14.5%) died from cGvHD. Conclusion In this intermediate analysis, we found that female recipient, unrelated donor, previous aGvHD are risk factors for developing sclerotic cGvHD. Moreover, we found that cord blood as stem cell source is a protector factor. Study of GvHD prophylaxis, ABO matching, type of HLA-mismatching, acute skin GvHD is ongoing on the whole cohort (n= 1587 patients from 7 centers) and results will be communicated later. After the final analysis on the whole cohort we would like to establish a score risk of sclerotic cGvHD in the goal to conduct a prospective study in France with a prophylactic treatment for high risk patients. | | HR | p | |:-------------------------------------------- | ------------------ | ------ | | Stem cells source and conditioning intensity | | | | MAC + Bone marrow | 1.0 | | | RIC + Bone marrow | 2.03 (0.39–10.6) | .4 | | MAC + Cord blood | 0.0 | .00001 | | RIC + Cord blood | 0.68 (0.02–18.01) | .82 | | MAC + PBSC | 0.32 (0.081–1.28) | .11 | | RIC + PBSC | 1.15 (0.5–2.5) | .73 | | HLA and Donor | | | | Matched related | 1.0 | | | Matched unrelated | 0.04 (0.011–0.181) | .00001 | | Mismatched unrelated | 0.62 (0.13–2.7) | .53 | | Patient age | | NS | | Donor age | | NS | | Sexmatching | | | | Male to male | 1.0 | | | Female to female | 2.7 (1.2–6.06) | .016 | | Female to male | 1.8 (0.78–4.33) | .16 | | Male to female | 1.44 (0.67–3.13) | .34 | | Disease diagnosis | | NS | | ATG in conditioning | | NS | | TBI in conditioning | 1.35 (0.69–2.64) | .38 | | Prior acute GvHD | 1.8 (0.9–3.36) | .064 | * NS= not statistically significant. Table. Multivariate analysis ![Figure][3] Disclosures: No relevant conflicts of interest to declare. [1]: #F1 [2]: #T1 [3]: pending:yes

Abstract 2353 Background: The key point in allogeneic hematopoietic stem cell transplantation (allo-HSCT) is to find the optimal balance between the graft-versus-host disease (GVHD) and the graft-versus-leukemia (GVL) effect particularly in the unrelated HSCT settings, without forgetting the important role of conditioning. Objective: To evaluate the toxicity and efficacy of a new reduced intensity conditioning regimen including Treosulfan, followed by allo-HSCT from a 10/10 HLA identical unrelated donor. Materials and Methods: This prospective study included adult patients presenting a hematological malignancy in need of allo-HSCT using a 10/10 HLA identical unrelated donor. Accepted diagnoses were: CML in 1st chronic phase (CP) resistant to imatinib, in 2nd CP or in complete remission (CR) after blast crisis. Multiple myeloma (MM) or chronic lymphocytic leukemia (CLL) in stage B or C in response after relapse post-auto-HSCT. Acute lymphocytic or myeloid leukemia (ALL, AML) patients in CR1 at high risk or The conditioning included: Treosulfan 12g/m2/day intra-venous (day-6 to day-4), fludarabine 30mg/m2/day intra-venous (day-6 to day -2) and ATG 2.5 mg/kg/day (day-2 to day-1). GVHD prophylaxis used cyclosporine A oral (5mg/kg/day) or iv. (3mg/kg/day) from day-1. Peripheral stem cells after mobilization by G-CSF were used as HSC source. Results: Between February 2005 and July 2009, 56 patients were included in 5 different French centers, 30 (54%) males and 26 females with a median age of 57 years (18-65.5). There were 38 (68%) myeloproliferative disorders [29 AML (14 in CR1, 14 CR2 & 1 2 lines. For sex-matching, 49% of patients were sex-mismatched (half of them were F>M). For CMV, 43% were -/-, 25% +/+, 28% +/− and 1% -/+. For ABO matching, 52% were compatible, 24% major incomp. & 24% minor incopm. The median time between diagnosis and allograft was 15 months (4-168). The median number of infused CD34+ cells was 6.5 × 106/Kg(1-17.2) Fifty-four (96%) patients engrafted with a median time to neutrophils (>0.5G/l), and platelets (>50G/l) recovery of 16 days (4-86), 11 (4-82) respectively. Seventeen patients developed aGVHD grade ≥II (8 grade II, 2 grade III & 7 grade IV) with a cumulative incidence at 3 months of 31% (25-38). The cumulative incidence of cGVHD was, at 12 months: 32% (25-39) limited and 6% (2-10) extensive; at 18 months: 34% (27-47) limited and 8% (5-12) extensive. After a median follow-up of 13 months (1-57), the median overall survival (OS) was not reached with a 3 years probability of 52% (38-71). The median time of event free survival (EFS) was 15 months (8 - 57) with a 3-years probability of 47% (35-64). The cumulative incidence of relapse at 3 years was 25% (19-31) and the cumulative incidence of transplant-related mortality (TRM) at 12, 18 and 36 months was 20% (16-27), 23% (16-29) (same at 18 & 36 months) respectively. At the last follow-up, 22 patients died, 7 due to relapse and 15 due to TRM (5 pneumonia, 4 GVHD, 1 secondary malignancy and 6 other causes). We showed in a univariate analysis a higher 3-years OS of patients in CR1 (65%) compared to those Conclusion: We showed very promising results in terms of OS, relapse and TRM with an impressive GVL effect, in heavily pre-treated patients and transplanted using unrelated donors. Treosulfan appears to be a good choice for conditioning especially with a better outcome in AML patients in CR1. Disclosures: No relevant conflicts of interest to declare.

Abstract 2266 We conducted a national retrospective analysis on 63 patients (pts) with either myeloid or lymphoid BC CML who underwent a first allogeneic stem cell transplantation as consolidation therapy for BC, since 2000 in France and look at whether 1st and 2nd generation TKIs ±chemotherapy prior to transplant may impact on outcome. There were 46 males and 17 females, and the median age at transplant was 34 (range 3–63) years and the median delay between BC and transplant was 19.6 (2.3-113) months. Thirty six (57%) pts were in chronic phase (CP), 3 (5%) in accelerated and 24 (38%) in BC at CML diagnosis. Twenty three pts received 1 line of treatment, 26 pts 2 lines, and 14 pts 3 or more lines for CML treatment before BC. For BC treatment, 25/63 (39%) pts had chemotherapy without TKI and 29/63 (46%) had TKIs combined or not to chemotherapy (6 chemo+dasatinib, 4 chemo+imatinib, 8 dasatinib alone, 1 dasatinib and imatinib, 9 imatinib, 9 unknown). At transplant 35/63 (55%) were considered in 2nd CP and 28 (45%) still in BC. The majority of the pts had an identical sibling donor (42/63: 66%), 1 mismatched related, 7 matched unrelated, and 13 mismatched unrelated. Fifty-six (89%) pts had a standard conditioning regimen, 7 (11%) pts a RIC regimen and 39 (62%) had a TBI. The stem cell source was bone marrow in 33/63 (52%) pts, PBSC in 24/63 (38%) and cord blood for 6/63 (10%). Fourteen transplants were sex-mismatched (female donor for male recipient). The EBMT-Gratwohl score was low (1+2+3) for 35/63 (55%) pts, intermediate (4) for 14 pts and high (5) for 9 pts, (and 6) for 3 pts and undetermined for 2 pts. The median follow-up since transplant was 25 (range 0.43–109) months. The overall median time for neutrophil recovery (neutrophils > 0.5 109/l) was 20 days. Fourty-one (65%) pts had an aGVHD (29 grade i-II and 12 grade III-IV) with no statistical differences between stem cell sources (p=0.4 for cord vs BM and 0.58 for PBSC vs BM). The cumulative incidence of chronic GVHD (14 extensive, 10 limited) was 43% at latest follow-up, more frequent in RIC transplant. The cumulative incidence of CML relapse was 38% at 2 yrs and multivariate analysis demonstrated the significant impact on relapse of the status at transplant in favour of disease control prior to transplant [HR 2.22 (1-4.93, p=0.05] and unexpectedly, no impact of BC treatment with or without TKIs prior to transplant and of other variables. The median overall survival (OS) was 22 months (see Figure 1). The TRM was 33% at latest follow-up, and it was significantly negatively influenced by the EBMT score (p=0.0077 for scores ≥ 4), BC status at diagnosis (p=0.04). The OS was not improved when using standard conditioning regimens (p=ns), but was improved by disease status at transplant (2nd CP better than persistent BC, p=0.01) and transplant with cord blood and BM stem cell sources over PBSC (p=0.011). Multivariate analysis on OS detected as significant adverse prognosis factors EBMT score ≥5 (HR=4.63 EBMT 5, and 4.77 EBMT 6). The types of treatment of BC CML had no significant impact (with or without stratification on chemotherapy). The median progression-free survival (PFS) was 5.8 months (see Figure 2). The PFS was not significantly improved by standard conditioning regimens but again was improved by disease control prior to transplant (2nd CP better than persistent BC, p=0.001) and with cord blood and BM stem cell sources over PBSC (p=0.057). Multivariate analysis on PFS detected as significant adverse prognosis factors EBMT score ≥5 (HR=5.82 EBMT 5, and 3.82 EBMT 6), but again, the types of treatment of BC CML prior to transplant had no significant impact. In conclusion, this analysis suggests that OS and PFS rates are significantly improved when compared to figures from the pre-TKI era. The use of 1st or 2nd generation TKIs alone or combined to chemotherapy does not seem to influence transplant results. CML status and the EBMT score are the major factors that can significantly influence transplant outcome. Disclosures: No relevant conflicts of interest to declare.

Abstract 3456 Patients with active refractory/relapsed hematological diseases have a very poor outcome. Best supportive care or investigational therapies in phase 1 trials are usually proposed to these patients. However, some previous data suggested that allo-SCT might be an efficient therapy even in the setting of chemorefractory disease, because long-term immune-mediated disease control can be achieved in some patients after allo-SCT. The aim of this study was to evaluate on a large series the outcome of adult patients with active refractory/relapsed hematological diseases at time of allo-SCT and to determine which sub-group would most benefit from such approach. Between 2005 and 2009, 861 patients with various hematological diseases (AML, n=323; ALL, n=43; MDS, n=129, CMML, n=12; MPS, n=110; CML, n=28; NHL, n=100; HL, n=40; myeloma, n=36; CLL, n=24; and other, n=16) were treated with allo-SCT, and reported to the SFGM-TC Registry. Per study criteria, all patients presented with active refractory or relapsed disease at time of transplant. This series included 517 males (60%) and 344 females (40%). The median age at transplant was 50 (range, 16–71) years. The median interval between diagnosis and transplant was 17 (range, 1–99) months. 32% of patients failed at least one prior SCT (Autologous or allogeneic prior to allo-SCT). 350 (41%) patients received allo-SCT from an HLA-matched sibling donor, while the remaining 59% received an allogeneic graft from a matched unrelated or mismatched donor. The stem cell source was mainly peripheral blood stem cells (n=617; 72%). Bone marrow was used in 139 patients (16%), and cord blood in 107 patients (12%). Myeloablative conditioning regimen was used in 328 patients (38%), and various reduced-intensity regimens were used in other cases (62%). With a median follow-up of 290 (range, 1–1854) days after allo-SCT, engraftment was observed in 88% of cases. Grade II-IV and grade III-IV acute GVHD occurred in 35% (n=301) and 17% (n=144) of patients, respectively. Chronic GVHD was observed in 185 patients (21%; limited: n=77; extensive: n=82; missing data: n=24). At last follow-up, 347 patients (40%) were still alive (of whom 297 were in CR; 86%). 246 patients (28.5%) died of disease progression, and 232 patients died of transplant-related causes (NRM: 27%). The Kaplan-Meier (KM) estimates of overall survival (OS) at one and 2 years were 39% (95%CI, 36–43%) and 31% (95%CI, 28–35%), respectively. Of note, in patients with lymphoma (n=140), OS at 1 and 2 years were 57% (95%CI, 48–66%) and 49% (95%CI, 40–58%) versus 36% (95%CI, 32–40%) and 27% (95%CI, 23–31%), respectively, in all other diagnoses (P=0.00004). In a Cox multivariate analysis accounting for relevant factors, a diagnosis of lymphoma (NHL or Hodgkin) was the most significant factor associated with improved survival (RR=1.68; 95%CI, 1.3–2.2; P=0.0001). Despite its retrospective nature and the inherent selection biases, in case of availability of suitable donor, this data support the use of allo-SCT in adult patients with active refractory/relapsed hematological diseases, especially in patients with lymphomas. Results are expected to be further improved with the advent of novel conditioning regimens and maintenance therapies after transplant that are currently tested as part of prospective studies. Disclosures: No relevant conflicts of interest to declare.

Abstract 4335 Background Primary induction failure or relapse in AML patients maintains a very poor prognosis despite the cytogenetic stratification and the introduction of new molecules. Nevertheless, allo-HSCT can improve the overall survival of some of these patients, although new strategies are needed especially in conditioning to allow a better outcome in this high risk population of patients. Aims The aim of this analysis was to investigate the safety and efficacy of FLAMSA sequential regimen in patients with high-risk acute leukaemia and MDS, in a multicenter retrospective French study. Methods We analyzed 79 patients, 44 males & 35 females with a median age of 55 years (19-69), there were 50 de novo acute leukaemias, 3 secondary leukaemias and 26 MDS, who underwent allo-HSCT after FLAMSA conditioning [fludarabine (30 mg/m2), high-dose AraC(2 g/m2), and amsacrine (100 mg/m2) from day -12 to -9 + cyclophosphamide (40 mg/kg with related donors, 60 mg/kg with unrelated or mismatched donors) on days -4 and -3, ATG 2.5mg/kg on days -4, -3, and -2 + TBI 4 Gy on day -5 (n=42) or Busilvex 3.2 mg/kg on days -4 & -5(n=37)] who were reported to the Societe Francaise Greffe de Moelle et de Therapie Cellulaire (SFGM-TC) registry. Thirteen patients had undergone previous HSCT (10 allogeneic and 3 autologous). The disease status at transplantation were: 19 (24%) of the patients in primary induction failure (PIF) or refractory disease, 14 (17,7%) in second relapse, 25 (32%) in early relapse, 8 (10%) in untreated disease and 13 (16.3%) in progressive disease. Thirty-four patients were transplanted from HLA identical siblings and 45 from unrelated donors. Sixty four grafts were PBSC, 5 bone marrow and 8 cord blood cells. Results After HSCT, 73 patients engrafted (6 non valuable early deaths) with a median time to neutrophil recovery of 16 days (8-41). There were 31 aGVHD (grade I: n=10, grade II: n=12, grade III: n=4 and grade IV: n=5) and 18 chronic GvHD (10 limited, 5 extensive and 3 unclassified). At the last follow-up, 32 patients are alive, after a median follow-up of 4.2 months, the probability of OS for the whole population at 2 years was 22.8% with 43.8% for untreated patients, 13.8% for patient in relapse, it was 47% for progressive and 23% for PIF patients. The univariate analysis showed a better significant OS according to 4 factors: disease status (untreated patients) (p= 0.04), kind of disease (p Conclusion Our study showed a total OS of 23% at 2 years with a better survival in the untreated patients (44%) and progressive patients (47%); a worse survival for the PIF and relapsed patients group (23% & 12% respectively). When comparing to the German study, we observed similar results regarding TRM and relapse incidence, the only difference was the leukemia mortality with a higher rate in our patients which could be explained with the different relapse treatment. Disclosures: No relevant conflicts of interest to declare.

Abstract 3225 Poster Board III-162 Background Autologous Stem Cell Transplantation (ASCT) with Peripheral Blood Stem Cells (PBSC) is widely used in first line treatment for Multiple Myeloma (MM) patients. The rapidity and stability of cell engraftment may correlate with the number of CD34+ cells in the autograft. However, the impact of this number on hematologic recovery and long term outcomes is not very well defined yet. Having a higher number of collected CD34+ cells seems to be a very important and accessible objective in 2009 (availability of mobilization agents), also it has in the same time cost and efficacy results. Materials and methods This study concerned 130 MM patients who underwent ASCT in our center between years 2000 and 2007. There were 79 males and 51 females with a median age of 56.8 years (34-72). At diagnosis there were 71 IgG (49κ, 22λ 26 IgA (15κ,11λ 2 IgD (1κ, 1λ 27 light chain (18κ,9λ 2 plasma cell leukemia and 2 nonsecretory. According to the IMWG classification, there were 11 patients in stage I (10A and 1B), 12 IIA, 96 III(75A and 21B) and 11 not classified. At diagnosis, 24 patients had a del(13), and 65 had high levels of β2microglobulin. The median interval between diagnosis and ASCT was 7.8 months (3.5-131). Before transplantation, all patients received granulocyte-colony stimulating factor (G-CSF) 5μg/kg/day, PBSC were mobilized in steady state in 135 cases, 62 after G-CSF + cyclophosphamide. As conditioning, all pts received melphalan alone with a median total dose of 304mg (130-440). Sixty six patients received a single ASCT and 64 patients received 2 ASCT in a double ASCT program. After transplantation, there were 2 graft failure, 40% of patients received red blood cell (RBC) transfusions (median number: 0 [0-23]), and 64% received platelet transfusions (median number: 1 [0-20]). The median number of days with neutrophils 3×106/kg. Results We found a high significant impact of the high number of infused CD34+(group 2) on platelets recovery (p=0.002). The univariate analysis using Cox model, showed a trend for the high number of infused CD34+ cells (group2) on leukocyte recovery O.R= 0.748 [0.5-1.0] (p=0.0568) and a high significant impact of the same group on neutrophils recovery O.R= 0.670 [0.5-0.9] (p=0.009). These results were not changed even after adjustment on age also on the sequential number of the ASCT in the double auto ASCT program. The multivariate analysis using Cox model, studying the impact of CD34+ group, age, gender, poor prognostic factors [high level of β2microglobulin and del(13)], mobilization (G-CSF alone or G-CSF + cyclophsphamide), showed a significant impact only of poor prognostic factors on overall survival O.R.= 7.94 [1.0-59.2] (p=0.04) and also on progression free survival (PFS) O.R.= 2.55 [1.1-5.7] (p=0.024). Conclusion High level of infused CD34+ appeared to be very optimal for hematological recovery after ATSC in MM, without any significant impact on O.S and PFS. An economical study is running on this population to assess the impact of this level on hospitalization and treatment costs, results will communicated in the future. Disclosures No relevant conflicts of interest to declare.

The short-term immunological effects of granulocyte-colony stimulating factor (G-CSF) administration have been extensively studied, but concerns regarding its long-term safety have been raised by recent reports such as G-CSF-induced genetic perturbations (Nagler et al, Exp Hematol, 2004) as analyzed in a limited number of donors. In response to such concerns, we conducted a prospective study to further evaluate both short and long-term immunologic and cytogenetic alterations potentially induced by G-CSF mobilization in a cohort of 24 consecutive healthy donors. Blood samples were taken before and at time of G-CSF administration and of cytapheresis as well as 1, 3, 6 and 12 months after. G-CSF treatment followed by cytaphereris induced a significant blood platelet count decrease while granulocyte, monocyte, B as well as dendritic cell (monocyte-derived DC and even more so plasmocytoid DC) blood counts increased. Such perturbations were found to return to baseline values at one month. T-cell (both CD4 and CD8) as well as NK-cell counts decreased significantly after cytapheresis and did not recover normal values before 3 months. In vitro immunoglobulin G and M production by pokeweed mitogen-stimulated PBMC was found to be significantly (P < 10−5) increased of 63+/−7% and 84+/−7% over baseline values; increases that persisted up to 6 months after mobilization. Such an observation is in line with our previous findings pertaining to enhanced humoral immune reconstitution after allo- PBSCT (Lapierre et al, Blood 2001, Blood 2002; Tayebi et al., Br J Haematol 2001). Quite strikingly, G-CSF mobilization was also associated with significant (P < 10−6) increased PBMC aneuploidy, analyzed by in situ hybridization of chromosomes 8 and 17 centromere-specific fluorescent probes. Such increase in the frequency of aneuploïd cells was not observed in CD34+ selected cells, suggesting that G-CSF mobilization did not affect CD34+ cells with regard to cytogenetic abnormalities and that the increase in aneuploidy was induced in more mature cells. The frequency of chromosomes 8 and 17 aneuploidy increased an average of 4.5 and 4.3-fold, respectively, after G-CSF mobilization (before mobilization: 1.2+/−0.1% and 2.4+/−0.2%; after mobilization: 4.2+/− 0.4% and 8.8+/−0.4% for chromosome 8 and 17, respectively). Cells with multisomy, rare events before mobilization (recorded in 5/24 donors and at very low levels (< 0.2%)), were observed at significantly higher frequencies in all 24 donors after mobilization. On the other hand, cells with monosomy and cells with trisomy were detectable at baseline in all donors although at low frequencies which increased significantly after mobilization. The level of aneuploidy subsequently decreased after G-CSF mobilization, but, importantly remained significantly higher than baseline values at three and six months after mobilization. The frequency of cells with monosomy was higher, and returned to baseline values more gradually than the frequency of cells with trisomy or multisomy. In conclusion, our data show that immunological and cytogenetic perturbations induced by G-CSF administration in normal donors may be more pronounced and persistent than previously described. Such perturbations remain however transient with a return to baseline values within one year for most perturbations. Consequences, if any, of such alterations for the donor are presently unknown. Although EBMT and NMDP surveys have no shown any evidence in favor of an increased incidence of malignant hemopathies or solid tumors in mobilized normal donors, assessment of the potential long-term risk of G-CSF administration will require prolonged observation of large G-CSF-mobilized donor cohorts.

This analysis concerned 164 allogeneic HSCT after standard (Std) or reduced intensity conditioning (RIC) using double cord blood cells (CBC), reported to the SFGM-TC registry. There were 57 females (F) and 107 males (M) with a median age of 39.5 years (18–66). The diagnosis pretransplant were acute leukaemia: 93 (56 AML and 37 ALL), MDS: 9, MPS: 9 (CML: 4), CLL: 5, NHL: 18, HD: 7, MM: 13, AA: 7, other diseases: 3 (2 inborn errors and 1 solid tumour). The median interval between diagnosis and HSCT was 20.6 months (2.6–385.5). Among 154 evaluated patients (pts), disease status prior conditioning were 92 CR (CR1: 45, CR 2: 34, >CR2: 12 and 1 non classified), 21 PR, 21 stable diseases (SD) (SD: 6, AA: 6, inborn errors: 2, CML in CP: 3, myelofibrosis: 2 and MDS: 2), 22 progressive diseases (PD) (PD: 4, relapses: 10 and refractory diseases: 8), 8 pts were not documented. Among 158 pts documented, 49 (31%) were completely sex matched and 109 (69%) sex mismatched [72 CBC1+2 (F+F or F+M) for a M recipient and 37 CBC1+2 (M+M or M+F) for a F recipient]. Among 158 documented, 42% of recipients, 37% of CBC1 and 38% of CBC2 were CMV+; there was a complete ABO compatibility between the 2 CBC and the recipient in 38 cases, 1 or 2 minor incompatibilities in 40 cases and almost 1 major incompatibility in 80 cases. For HLA matching, we distinguished 3 groups: group1 [n=78 (HLA compatibility at least 4/6 between recipient and CBC1+2) and between CB1 and CB2 including total HLA DRB1 matching)], group 2 [n=55 (HLA compatibility at least 4/6 between recipient and CBC1+2) considering neither HLA compatibility between CBC1-CBC2 nor HLA DRB1 matching)] and group3 [n=25 (all others)]. At harvesting, the median number of total nucleated cells (TNC) (x107/kg) was 4.67(1.6–12.2), the CD34+ (x105/kg) 1.8 (0.3–10.8), the CFU-GM (x104/kg) 3.67 (0.94–25), and after thawing 3.24(0.58–12), 1.4(0.2–9.2) and 2(0.27–16.4) respectively. The TNC threshold number was set to 5x107/kg because more than 96% of pts received more than 3x107/kg. Among 135 documented, 26 (19%) received Std conditioning and 109 (81%) RIC. After transplantation, 129 pts (96%) engrafted with 86% (80–92) of neutrophil recovery at day 60 with no significant difference according to TNC (5.107/kg: 46%(32–60). Twenty-one pts presented a chronic GVHD (9 limited and 12 extensive) and the cumulative incidence at 1 year was 13.7%(4–24) for limited and 20%(9–21) for extensive. With a median follow-up of 7.3 months, the probability of 1-year and 2-year overall survival and disease-free survival were 49.6% (40–61.5) and 38% (27–54), 43% (33.5–54.5) and 36% (25–51) respectively. The probabilities of OS, NRM and RM according to TNC, disease status pre-transplant, HLA matching and sex matching are shown in Figure 1 and Table1.The multivariate analysis showed a significant impact of 2 factors on OS: disease status PD vs CR: HR=6.16 (1.87–20.25) (p=0.002); HLA matching group2 vs group1: HR=0.29 (0.11–0.82) (p=0.01), and 3 significant factors on DFS: sex-matched HR=0.29 (0.09–0.94) (p=0.03), sex-mismatched (F recipient) HR=0.15 (0.04–0.61) (p=0.008) and HLA matching (group 2 vs group1) HR=0.32(0.11–0.88) (p=0.02).A refined chimerism analysis is ongoing and will be presented. In conclusion, this large retrospective analysis showed that the quantitative objective of double cord blood use for allogeneic HSCT is achieved (only 4% had received < 3x107/kg TNC) with no further significant impact of TNC number on OS, NRM and RM. As usual in other types of allogeneic HSCT, we demonstrated the significant impact of disease status before transplantation on transplant outcome. Finally, the most interesting point was the better results observed in group 2 but which needs more precise analysis in the future. Table 1. Probability of OS, NRM, RM according to different variables. Probability of OS Probability of NRM Probability of RM Whole population (cummulative 1 year) 49.6%(40–61.5) 49.7%(39–60) 7.5%(2–17) TNC TNC < 5 × 107/kg 50%(36.6–68.5)) 47%(32–62) 8%(0–16) TNC ≥ 5 × 107/kg 44.5%(29.6–67) 56.5%(38.5–74.5) 9.5%(0–20) Disease status pre-transplant CR 51%(38–68) PR 54%(31–92.5) SD 75%(43–100) PD relapse 35.5%(18–70) HLA compatibility Group 1: 4/6 or more for all(2/2 for DRBI) 34%(21–54) 62%(46–78) 10%(3–20) Group 2: 4/6 or more for R-CB1 and R-CB2 61%(45–81) 39%(21–57) 2%(0–6) Group 3: Others 61.6%(42–90) 42.5%(19–66) 10%(0–24.5) Sex matching M recipient with sexmismatch 35%(22–57) 63%(46–79) 8%(2–14) Sexmatch 61%(45–82) 43%(23.5–63) 4.5%(0–13.5) F recipient with sexmismatch 54%(36.5–80)) 42%(24–62) 9%(0–22) Figure 1: Probability of OS according to different HLA groups Figure 1:. Probability of OS according to different HLA groups

This analysis studied 197 autologous HSCT performed in 132 patients treated for multiple myeloma (MM) in our center between 2000 and 2007. There were 53 females and 79 males with a median age of 56.8 years (34–72). At diagnosis there were 71 IgG (49κ, 22λ), 26 IgA (15κ,11λ), 2 IgD (1κ, 1λ), 27 light chain (18κ,9λ), 2 plasma cell leukemia, 3 non secretory, 1 non secretory/non excretory. There were 11 stage I (10A and 1B), 12 IIA, 96 III(75A and 21B) and 13 not classified. At diagnosis, 24/98pts had a del(13), and 65/179 had high levels of β2microglobulin. Before 2004, 50% of transplants were done and the median interval between diagnosis and HSCT was 7.8 months (3.5–131). We divided the population into 2 groups (I = 65: double-auto, II = 67 simple auto). The disease status pre-transplant according to the number of apheresis were: 1 apheresis (group I: 3CR, 29PR, 3SD, 1PD; group II: 2CR, 24PR, 3PD), 2 apheresis (group I: 16PR, 1SD, 1PD; group II: 3CR, 9PR, 1PD), 3 apheresis (group I: 1CR, 4PR; group II: 10PR, 1unknown), 4 apheresis (group I: 4PR, 1PD; group II: 1CR, 10PR, 2SD), 6 apheresis (group II: 1PR) and 8 apheresis (group: 1PR). PBSC were mobilized in steady state in 135 cases, 53 after cyclophosphamide alone, 9 cyclophosphamide with other drugs. During mobilization, we used GCSF in 179cases, GM-CSF in 5cases, SCF in 4 cases and associations of GM-CSF+ GCSF in 2(1%) cases and SCF+GCSF in 7(3.5%) cases. The median number of infused cells were: TNC 5×107/kg (1–59), CFU-GM 70.5×104/kg (0–2616) and CD34+cells 3×106/kg (0–27). Of these, 115 (58%) had a number of CD34+cells a conditional logistic-regression model to analyze associations between these variables and length of hospitalization, number of RBC and Pt transfusions; a multivariate analysis using Cox model to analyze the impact of these variables on length of aplasia ( We observed no significant impact of all studied variables on length of hospitalization and RBC transfusions and a significant negative impact of long interval diagnosis-T (p=0.05) and of the period > 2004 on Pt transfusion number (p=0.03). We showed a significant positive impact of CFU-GM number [HR=1 (1000–1.002) (p=0.03)] and growth factor use after transplantation [HR=0.55 (0.36–0.85) (p=0.005)] on days

This report updates a retrospective study from SFGM-TC registry concerning 1108 patients who underwent allogeneic hematopoeitic stem cell transplantation (HSCT) after reduced intensity conditioning (RIC) from HLA identical siblings (84%) and unrelated donors (16%) for hematological malignancies. At time of conditioning, 442 patients were in CR, 337 in PR, 107 in stable disease (SD) and 222 in progressive disease (PD). As conditioning, 255 patients received fludarabine and TBI (2 grays), 465 patients fludarabine, busulfan and ATG and 388 patients an other regimen. After transplant, 336 patients (30%) developed an acute GVHD ≥ grade II (grade II: 178, III: 80 and IV: 78). A chronic GVHD was present in 388 patients (35%) (185 limited and 203 extensive). With a median follow-up of 30 months, the 3 and 5-year probability of overall survival (OS) were 43.5% (40–47) and 32%(29–35) respectively and the 3 and 5-year probability of event-free survival (EFS) were 35%(31–39) and 28% (24.5–31) respectively. The TRM at 1 year, 2 years and 3 years was 15% (13–17), 18% (15.5–21) and 20% (17–23). A mixture model, gfcure with Splus statistical package determined the percentages of long-term survivors and its adequacy was verified graphically. The probability to be a long-survivor was 24% (17.5–32.5) (Fig.1) and to be a long event-free survivor was 23% (19–28) (Fig. 2). The multivariate analysis has tested recipient and donor age, disease status pre-transplant, number of transplants before RICT, HSC source, sex matching, HLA matching, CMV status and ABO compatibility. The only factor which had a significant impact on long-term survival after RICT was the disease status just prior conditioning: PR versus CR: HR: 3.63 [1.14–9.18] p Figure 1 Figure 1. Figure 2 Figure 2.

This prospective multi-center trial (MiniMud) concerns 27 patients (13 females, 14 males - 56 years[18–65]) who underwent allogeneic hematopoietic stem cell transplantation (HSCT) from unrelated donors. The diagnosis and disease status pre-transplant were AML (n=12; 5 CR1, 7 CR2), ALL (n=1; 1 CR2), MM (n=5; 3 PR2,1 PR3, 1 PR4), NHL (n=4; 1 CR1, 2 PR3, 1 PR4), MPS (n=3; 2 stable disease (SD), 1 Relapse), acute RAEB (n=1; 1 progressive disease (PD) and CLL (n=1; 1 PR2). In total, 14 patients were in CR, 9 in PR, 2 in SD, 2 in PD. The median interval between diagnosis and transplant was 27 months [5;171], 24 patients received PBSC and 3 received bone marrow from unrelated 10/10 antigen HLA-identical (high-resolution level). All patients received Fludarabine 30 mg/m2/dx5d, Treosulfan 12 g/m2/dx3d and ATG 2,5 mg/kg/dx3d. After transplant, we observed 2 graft failures, 7 patients developed an AGVHD ≥ grade II (30%). Only 23 patients were eligibile to be analysed after 3 months of follow-up post transplant, 7 patients developed cGVHD (30%) (5 limited and 2 extensive), 12 patients died (7 from TRM causes and 5 from relapse) and 12 patients are alive in CR with a median follow-up of 14 months. Among the global population we defined a low risk subgroup: 17 patients [12 AML (5 CR1, 7 CR2), 1 ALL (CR2), 3 MM (3 PR2) and 1 NHL (CR1)], 2 patients presented an AGVHD ≥ grade II (13%). With a median follow-up of 18 months, 2 patients had a follow-up of less than 3 months, 3 patients developed a cGVHD (20%), 6 patients died (40%) and 9 patients are alive in CR. At one year, the probability of overall survival (OS) and event-free survival (EFS) were 50.4% [32;79], 40% [23;70] for the total population and 60% [37;96], 51.5% [29;90] for the low risk subgroup respectively. To try to demonstrate if Treosulfan allowed a better transplant outcome we performed a paired match-analysis [center and 4 out of 5 other parameters (age, gender, HSC, pre-transplant status and diagnosis] comparing our Treosulfan series and RIC transplants receiving Fludarabine, Busulfan and ATG (2.5–12.5mg/kg) from the SFGM-TC registry. This paired match-analysis found only RICT from related donors in the SFGM-TC database and showed no difference in term of OS and EFS between our unrelated donor-Treosulfan and related donor-SFGM-TC series. Download : Download high-res image (64KB) Download : Download full-size image Figure Download : Download high-res image (66KB) Download : Download full-size image Figure In conclusion, these results demonstrate that Treosulfan appears to be a very promising drug that could be included in the conditioning regimen before HSCT from either related or unrelated donors.

This study concerns the correlation between chimerism and minimal residual disease (MRD) documentation of 12 patients (9 males and 3 females − 51 years) who underwent allogeneic hematopoietic stem cell transplantation (HSCT) for chronic lymphocytic leukemia (CLL). Chimerism was assessed from total peripheral blood (PB), bone marrow (BM) and CD3+ cells by STR-PCR (sensitivity [S]:5%) followed by SNP-PCR (S:0.2%) and MRD from PB and BM by an international standardized multicolor flow cytometric approach (S:0.01%). At diagnosis, there were 4 Binet stage A, 7 stage B and 1 stage C. Four patients were in CR, 7 in PR and 1 in progressive disease pre-transplant. Eleven patients underwent 1 HSCT and one patient 2 HSCT from HLA identical sibling donors (10 PBSC and 3 BM); 2 patients received a myeloablative regimen and 10 reduced intensity conditioning. After transplant, 6 patients developed an acute GVHD≥ grade 2 and 8 a chronic GVHD. The chimerism kinetics were followed in all patients and showed a conversion from mixed chimerism (MC) into full donor chimerism (FDC) at day 30 post transplant in 6 patients, between day 30 and day 120 in 4 patients, 2 patients never converted to FDC and remained either in stable MC or progressed to recipient profile. MRD documentation was introduced more recently and we have MRD kinetics only for 5 patients showing a slower MRD decrease when compared to chimerism kinetics. The median follow-up was 39 months and at the last follow-up, among the 11 patients who received 1 HSCT, 2 patients died from infection and they showed FDC and MRD negativity, 9 patients are alive (7 in CR according to NCI criteria and 2 were too early). Among the 7 CR patients, 6 were in FDC and MRD negative and 1 in MC and MRD negative. The patient who received 2 allogeneic transplants had in fact relapsed 24 months after the first allogeneic transplant and the second transplant allowed a CR associated to a MRD negative status and FDC, followed by a recent increase of MRD level in spite of FDC maintenance at last follow-up. We studied 246 samples for chimerism and 88 for MRD. To correlate quantitative MRD levels with chimerism data only 56 samples were comparable. Considering kinetics, we observed a concordance in 50/56 (89%) with 16/50 (32%) both positive early after transplant and 34/50 (68%) both negative later. Among the 6 non concordant samples which belonged to 2 patients, 5/6 (83%) were MC and MRD negative explained by the mixed chimerism profile of CD3+ cells and 1/6 (relapsed patient) was MRD positive and FDC explained by the better sensitivity of MRD. Relation between MRD and chimerism for PB and BM was tested using a mixed effect linear regression to account for repeated measures for each patient. This analysis showed a significant correlation between PB MRD and chimerism (p

Conventional chemotherapy in NHL does not substantially modify the natural course of disease especially for indolent lymphomas. A major advantage of allogeneic HSCT is the immunological recognition of lymphoma cells by the allogeneic donors cells and the low relapse risk in long-term survivors is highly suggestive of a graft-versus-lymphoma effect. However the still high mortality rate of this procedure mainly due to GVHD limits its indications. To determine the place of allogeneic HSCT in the NHL therapeutic strategies, we performed a retrospective study from 1987 to 2004. Twenty-six patients [17 males, 9 females, median age of 37 years (18–61)] underwent allogeneic HSCT: 18 conventional bone marrow transplantations and 8 allogeneic HSCT after reduced intensity conditioning (RICT) from HLA identical sibling donors. The diagnosis pre transplant were: 11 high grade lymphomas (9 phenotype B and 2 T), 8 patients with a transformation of an indolent form (2 CLL, 2 follicular and 3 marginal zone lymphomas, 1 mycosis fungoides) [73% aggressive forms of NHL] and 7 indolent forms (3 lymphocytic lymphomas, 3 follicular lymphomas and 1 mantle cell). All patients have received at least 3 courses of chemotherapy (3–7). Eighty-eight percent had a stage IV at diagnosis, 7% a stage III and 7% a stage II. At transplant, 10 (38.5%) were in complete remission, 2 in partial response, 9 in stable disease and 5 (19%) in relapse or progressive disease. Twenty-two patients received bone marrow as HSC source and 4 PBSC. Total body irradiation was used as part of the conditioning regimen in 78% of patients. With a median follow-up of 85 months (41–225) from diagnosis and 66 months (9–185) from transplantation, 13 (50%) are alive without evidence of lymphoma at last follow-up and 13 died [9 (35%) from transplant-related causes (1 acute GVHD grade IV, 3 pneumonitis, 1 septicemia, 1 veino-occlusive disease, 1 leucoencephalitis and 2 others causes) and 4 (15%) from relapse. Eight (30%) did not develop any acute GVHD, 9 a grade I, 4 a grade II, 3 a grade III and 2 a grade IV. Nineteen patients did not develop chronic GVHD, 6 a limited form and only one an extensive form. Overall survival at 5 years was 61.5% [95%CI (48%–83%)] from diagnosis and 57% [95% CI (40%–80%)] from transplantation. Transplant-related mortality was estimated at 35%. Multivariate analysis did not show any significant influence on overall survival (OS) of age, disease status, type of transplantation (conventional vs RICT), HSC source, TBI, acute or chronic GVHD but this lack of significance is probably in relation with the small number of patients. The long-term follow-up in this study demonstrated the effectiveness of allogeneic HSCT in achieving long-term disease control even in heavily pretreated aggressive NHL patients.

Autologous stem cell transplantation as consolidation therapy for de novo AL is able to improve long-term survival especially after second complete remission (CR). In this study, we retrospectively analyzed the karyotyping results performed in 79 out of 101 harvested autologous PBSC from 38 de novo AML patients and 8 de novo ALL in CR. We secondly, analyzed their impact on overall survival (OS) and event-free survival (EFS) post-transplant. All patients had abnormal medullar karyotype at diagnosis. We determined two groups of patients: (1) group 1 with chromosomal abnormalities in harvested PBSC included 10 AML [4 males and 6 females, median age: 53 years (45–66.5)] with at diagnosis 3 favorable, 4 intermediate and 3 unfavorable-risk karyotype ; 2 M0, 3 M2, 1 M3, 2 M4, 2 M5 according to Fab classification ; and 9 autotransplanted patients in CR1 and 1 in CR2 and (2) group 2 without any chromosomal abnormalities in harvested PBSC including 27 AML+ 9 ALL [22 males and 14 females, median age : 44 years (19–70)] ; 1 M0, 1 M1, 4 M2, 5 M3, 11 M4, 5 M5 with at diagnosis 18 favorable prognosis, 1 intermediate and 8 unfavorable-risk karyotype ; among ALL patients, 7 B ALL and 2 T ALL and at transplant, 29 autotransplant in CR1, 6 in CR2 and 1 in CR3]. Cytapheresis products collected contained a median number of 7.7x 108/Kg (5.2–17.3) and 6.79x108/Kg (1.26–16.4) mononuclear cells and 19x106/Kg (3–43) and 20.5x106/Kg (1–4O) CD34+ cells in group 1 and in group 2 respectively. All PBSC harvested were reinfused. Conditionning regimen was quite similar in the two groups and most of the patients of group 1 (60%) and group 2 (70%) received cyclophosphamide and TBI. The hematopoietic reconstitution was similar in the 2 groups [neutrophiles > 0.5 G/l: 26 days in group1 versus 19 days in group 2 (p=0,86)]. At 3 years after transplant, there was a trend for a better OS (54.7% CI 95% 40–74.8) (p=0.06) and EFS (52.5% 95%CI 39.6–74) (p=0.05) in patients who received PBSC without any chromosomal abnormalities compared to patients receiving PBSC with abnormal karyotype (OS : 30% 95% CI 11.6–77.3 ; EFS : 30% 95% CI 11.6–77.3). In conclusion, results observed from autologous harvested PBSC karyotyping are promising but a larger number of patients and a longer follow-up could enable to reach a more important significance.

Background & aims The only curative treatment of CML to date, remains allogeneic stem cell transplantation (Allo-SCT) despite some observations of non-detectable disease recurrence after tyrosine kinase inhibitor (TKI) cessation. The scope of allogeneic stem cell transplant for chronic phase (CP-) CML remains debatable and it seems interesting to retrospectively analyse the settings of this procedure in such patients since the introduction of TKI within the therapeutic arsenal of this disease. Methods We retrospectively analysed the registry of the Francophone society of stem cell transplantation and cellular therapy (SFGM-TC) from 2002 to 2014, for patients being in CP at diagnosis and at transplant. All data were captured according to thefrench regulations and were collected after signed up informed consent for each patient. All patients were transplanted for different degrees of resistance or severe recurrent intolerance to TKI(s). We segmented the observation period into two parts: 2002-2006 (Imatinib era) and 2006-2014, once second generation TKI were available in our country (TKI2 era). All patients were in CP-CML at diagnosis and first chronic phase at transplant. Second transplants for the same patient were excluded from this analysis. Results From 2002 to 2014 the proportion of transplants for CML dramatically decreased form 7.1% to Conclusion Allo-SCT still remains a curative treatment of CP-CML despite significant toxicities over time and the picture of this procedure in the therapeutic arsenal has dramatically changed over the last decade due to multiple therapeutic options offered now. Despiteallo-SCT of patients with longer diseases histories, probably more co-morbidities, there is an encouraging trend in the improvement of OS and RFS. Figure 1 OS and RFS for CP-CML according to the engraftment period. Figure 1. OS and RFS for CP-CML according to the engraftment period. Figure 2 Figure 2. Disclosures Nicolini: BMS: Consultancy, Honoraria; Ariad: 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, Research Funding. Coiteux:Novartis, BMS, ARIAD: Speakers Bureau. Michallet:Bristol-Myers Squibb: Consultancy, Honoraria, Research Funding; Pfizer: Consultancy, Honoraria; Novartis: Consultancy, Honoraria; Pfizer: Consultancy, Honoraria; Astellas Pharma: Consultancy, Honoraria; MSD: Consultancy, Honoraria; Genzyme: Consultancy, Honoraria. Peffault De Latour:Novartis: Consultancy, Honoraria, Research Funding; Alexion: Consultancy, Honoraria, Research Funding; Amgen: Research Funding; Pfizer: Consultancy, Honoraria, Research Funding.

Abstract 4113 Purpose: Despite the raise of major new drugs in the management of multiple myeloma (MM), it is still an incurable disease and allogeneic stem cell transplantation (alloSCT) is currently the sole potential curative therapy mediated by a graft-versus-myeloma (GVM) effect. For over ten years, reduced intensity conditioning regimens (RIC) in alloSCT has been developed to maintain the GVM effect and to decrease the high transplanted related mortality (TRM) associated with myeloablative conditioning (MAC) in myeloma patients. Yet, as numerous studies have assessed RIC alloSCT versus autologous SCT, only few data provides a comparison between RIC and MAC for MM in large series. Our study aims to compare RIC and MAC alloSCT for MM on a large retrospective French cohort. Methods: We report a retrospective multicenter study based on the Société Française de Greffe de Moelle et de Thérapie Cellulaire (SFGM-TC) registry, evaluating the outcome of patients undergoing alloSCT for MM. Between 1995 and 2009, 811 patients allografted for MM were registered. Data concerning disease characteristics and the allograft procedure were screened using the French Promise® database; computerized discrepancy errors and data verification of every single file ensure data quality, important missing data were directly retrieved in each centre. We assign the type of conditioning regimen for each patient according to the registry's data reported by the local physician and check this information according to the following rules. We consider as myeloablative (MAC) all regimens containing > 8 mg/kg oral busulfan, > 600 cGY TBI, ≥ 200 mg/kg cyclophosphamide or combinations as BEAM like regimens. All other chemotherapy combinations with or without TBI are defined as RIC regimens generally including fludarabine. Patients were excluded when major data or conditioning information were not available; 566 patients (70% of the entire cohort) were finally eligible for analysis, 478 RIC and 88 MAC. Chi-squared test and T-test were used for comparisons. Survival analyses (Overall survival, OS; progression-free survival, PFS) are made using the Kaplan-Meier curve and log rank test. Multivariate analysis is made using the Cox proportional hazards model. Results: The RIC and MAC populations were statistically different regarding age (respectively 53 versus 46 years, p 1 prior transplant in 94% RIC versus 77% MAC, p The 2-year OS is 69.5% after RIC and 79.9% after MAC (p Factors associated with better OS and PFS in multivariate analysis are the following in the whole cohort: < 12 months between diagnosis and alloSCT (p=.0078 and p=.0037, respectively); disease status at transplantation (at least partial response, p=.0409 and p=.0026); no or limited acute graft-versus-host disease (GvHD) (p Conclusions: In the French practice, MAC regimens have had more limited indications due to their higher toxicity. This study suggests that the outcome after MAC-alloSCT is comparable to that after RIC-alloSCT with a trend to a better OS in the late follow-up (after 5 years) after MAC. The aspect of flattening in the survival curves suggests that the anti-tumoral action of the conditioning regimen remains essential on the anti-myeloma effect after alloSCT and that should be considered in the transplantation procedure. Disclosures: No relevant conflicts of interest to declare.

Introduction: Acute myeloid leukemia (AML) patients failing to achieve complete remission or relapsing after intensive chemotherapy, have a dismal prognosis. In this setting, allogeneic hematopoietic cell transplantation (HCT) is a curative salvage therapy, but myeloablative conditioning (MAC) transplants are usually associated with high non-relapse mortality (NRM). Promising results have been reported using a sequential treatment approach (SEQ) consisting of cytoreductive chemotherapy, followed by reduced-intensity conditioning (RIC) and prophylactic transfusion of donor lymphocytes. Nevertheless, for young patients with refractory or relapsed AML, there is no definitive evidence to prefer SEQ rather than MAC HCT. The aim of this study, analysing the data from the SFGM-TC, was to compare the results of these two approaches in patients transplanted in AML not in complete remission. Methods: Inclusion criteria: a) HCT performed from January 2006 to December 2013 b) Patient age up to 50 years (y), c) Transplant for refractory or relapsed AML, c) SEQ or MAC regimen d) Matched sibling donor, matched or mismatched unrelated donor. PatientsÕ characteristics: 108 patients (median age: 38 y (range 1-50)) were analysed. Eighty-eight percent of patients were diagnosed with de novo AML and 12% with secondary AML. At diagnosis, 4% of patients were classified favorable, 33% Intermediate-1, 25% intermediate-2, and 38% adverse, according to the European LeukemiaNet (ELN) score. Before transplant, patients received a median of 2 lines of chemotherapy (range 0-3). At transplant, 53% of patients were in primary induction failure (PIF), 43% in relapse, and 4% were never treated and the median of bone marrow (BM) and circulating blast percentages were 19% (range 0-96) and 2% (range 0-93) respectively. Transplant modalities: 60 patients received a SEQ approach (cytoreductive chemotherapy with FLAMSA (N=40) or with clofarabine (N=20) + high-dose cytarabine, followed by RIC combining cyclophosphamide (CY) with 4 Gy total body irradiation (TBI) or with busulfan (BU)) while 48 patients received a MAC regimen (CY combined with 10-12 Gy TBI (N=20) or with BU (N=28)). Forty-one percent of patients were transplanted from a sibling and 59% from an unrelated donor. Stem cell source was peripheral blood in 76% and BM in 24%. Results: MAC and SEQ groups had similar ages, ELN scores, numbers of pre-transplant chemotherapy lines and status at transplant. They only differed from the percentage of circulating blasts at transplant (median 1%, range (0-66) in MAC group vs 8%, range (0-93) in SEQ group, p=0.004) and from the time between diagnosis and transplant (median 164 days, range (62-610) and 239 days range (61-3702) respectively, p=0.036). In univariate analysis, type of conditioning regimen (MAC versus SEQ) did not impact post-transplant outcomes: 2-year overall survival (OS) was 37.2% (95% confidence interval (CI): 23.8-50.7%) in MAC patients vs 32.9% (95%CI: 21.4-44.9%) in SEQ patients (p=0.43); 2-year cumulative incidence of relapse was 56.7% (95%CI: 42.2-71.1%) vs 50.1% (95%CI: 37.2-63%) respectively (p=0.89) and 2-year non-relapse mortality was 14.7% (95%CI: 4.5-25%) vs 16.7% (95%CI: 7.1-26.2%) respectively (p=0.46). Grade II-IV acute graft versus host disease (GVHD) occurred more frequently in patients who received MAC: 66.7% vs 36.7% in the SEQ group (p=0.002), there was also a trend for more chronic-GVHD (41.7% vs 25%, p=0.066). Multivariate analysis identified high number of chemotherapy lines (hazard ratio (HR) 1.66, 95%CI 1.11-2, p=0.013) and high percentage of circulating blasts at transplant (HR 1.01, 95%CI 1.003-1, p=0.011), as factors independently associated with poor OS. Heterogeneity analyses performed for OS identified significant interactions between the type of HCT conditioning and BM blast percentage (p=0.01), or CD34+ cell dose (p=0,04), analysed as categorical variables with cut-offs at medians (20% and 6.106/kg, respectively). Though limited by low patients numbers, subgroup analyses suggested that SEQ HCT conferred a higher risk of death in patients with BM blasts > or = to 20% (HR=1.42), but a lower risk (HR=0.59) with BM blasts In conclusion, these data confirm that HCT is a valid salvage therapy for relapsed or refractory AML. In patients up to 50 years, both MAC and SEQ conditioning regimens offer similar outcomes, without increased NRM after MAC transplant. Figure 1. Figure 1. Disclosures Mohty: Janssen: Honoraria; Celgene: Honoraria.

International audience; Graft-versus-host disease (GVHD) and cytomegalovirus (CMV)-related complications are leading causes of mortality after unrelated-donor hematopoietic cell transplantation (UD-HCT). The non-conventional MHC class I gene MICB, alike MICA, encodes a stress-induced polymorphic NKG2D ligand. However, unlike MICA, MICB interacts with the CMV-encoded UL16, which sequestrates MICB intracellularly, leading to immune evasion. Here, we retrospectively analyzed the impact of mismatches in MICB amino acid position 98 (MICB98), a key polymorphic residue involved in UL16 binding, in 943 UD-HCT pairs who were allele-matched at HLA-A, -B, -C, -DRB1, -DQB1 and MICA loci. HLA-DP typing was further available. MICB98 mismatches were significantly associated with an increased incidence of acute (grade II-IV: HR, 1.20; 95% CI, 1.15 to 1.24; P < 0.001; grade III-IV: HR, 2.28; 95% CI, 1.56 to 3.34; P < 0.001) and chronic GVHD (HR, 1.21; 95% CI, 1.10 to 1.33; P < 0.001). MICB98 matching significantly reduced the effect of CMV status on overall mortality from a hazard ratio of 1.77 to 1.16. MICB98 mismatches showed a GVHD-independent association with a higher incidence of CMV infection/reactivation (HR, 1.84; 95% CI, 1.34 to 2.51; P < 0.001). Hence selecting a MICB98-matched donor significantly reduces the GVHD incidence and lowers the impact of CMV status on overall survival.