ALBERT

Description: Outcomes of patients with active AML/MDS remain poor after HSCT. Treatment failure is in part a result of poor disease control with the preparative regimen. Here we report toxicity and efficacy of the immunoconjugate GO incorporated to a reduced intensity regimen. Methods: Objective: determine which dose of GO between doses 2 and 4mg/m2 has the lowest toxicity(tox) probability with the highest response probability when added to FM (statistical method of Thall, Estey, Sung (1999)). Tox was defined as grade 3–4 organ tox, engraftment failure or early death (first 30 days post HSCT). Response was defined as engraftment, no tox and remission (CR) on day +30. Trial was planned with dose levels 4, 6 and 9 mg/m2 but given tox observed at 4mg/m2, the other dose levels were not explored and dose 2mg/m2 was added. Eligibility: patients (pts) age 12–75 years, ineligible for high-dose regimens or with high-risk CD33+ AML/MDS. Donors: related or unrelated (MUD). Treatment plan: GO 2 or 4mg/m2 day -12, fludarabine 30mg/m2 (days -5 to -2), melphalan 140mg/m2 (day -2); HSCT day zero. ATG was added in MUD HSCT. Graft-versus-host disease (GVHD) prophylaxis: tacrolimus and mini-methotrexate (5mg/m2 day+1,+3,+6,+11). GO was given on day -12 to avoid delay in engraftment and to decrease the number of committed myeloid progenitors prior to FM. Here we also included pts. treated under a compassionate IND and performed a non-randomized comparison with historic pts. treated with FM140 mg/m2 without GO. Results: 29 pts. have been treated. Table shows baseline characteristics. GO dose 0 (n=49) 2 mg/m2 (n=21) 4 mg/m2 (n=8) P value Remission at BMT 26% 9% 12% 0.25 MUD 41% 28% 25% 0.52 Median age (years-range) 54 (21–75) 53 (13–72) 30 (23–67) 0.08 Peripheral blood stem cell transplant 47% 86% 100% 0.0005 High-risk cytogenetics 43% 33% 37% 0.83 Median follow-up is 10 months (range 2.5–36) for GOFM alive patients (n=18). CR rate: GO 2mg/m2 =90%; GO 4 mg/m2= 71%; control=71%. There was only 1 case of moderate, reversible hepatic VOD (GO 2mg/m2). Grade III-IV tox and early death rates were: GO 2 mg/m2, 19% and 5% (n=1 due to pulmonary hemorrhage); GO 4mg/m2, 25% and 25% (n=2, due to pneumonia and multi-organ failure); control, 21% and 12% (n=6). Acute gd. II-IV GVHD rate was 44% for GO treated patients. All evaluable patients had engrafted by day +30; 8 pts have relapsed. Comparison with historic controls: active disease and high-risk cytogenetics were significantly associated with worse event-free (EFS) and overall survival (unadjusted log-rank test). Cox proportional hazards model: risk of death or relapse was increased with older age (RR 1.02; P=0.06), MUD vs related donor (RR 1.91; P=0.03), active disease vs CR (RR 0.23; P=0.001), and having GO 4 mg/m2 or zero vs 2 mg/m2 (RR 0.46, P=0.04).Figure shows survival. Figure Figure Conclusions: The addition of GO at the 2 mg/m2 dose level is promising and merits evaluation in a larger, controlled trial.

Description: Acute (a)GVHD is a major cause of morbidity and mortality after UD or mismatched related (MRD) donor stem cell transplantation (HSCT). The nucleoside analog pentostatin (pentos) is a potent inhibitor of the enzyme adenosine deaminase that induces lymphocyte depletion, with low risk of myelosuppression. We have investigated the addition of pentos to tacrolimus (tacro) and MTX as GVHD prophylaxis. Methods: This was an adaptive randomized, dose finding study that took into account toxicity and efficacy in a Bayesian adaptive design. Probability of assignment to the control group was fixed at 20%. Probability of assignment to the 4 dose groups was based on the probability that an arm’s success rate was greater than the control’s. Probabilities were recalculated after each observed patient outcome. Recipients of UD and MRD were eligible; all analysis is done by intention to treat. Success was patient being alive, engrafted, in complete remission (CR), without GVHD at study completion(100 days post HSCT). Grade III-IV aGVHD defined failure at any time, while gd. I-II did not constitute failure if absent on day 100. The study was designed to have a false positive rate of 0.05 and a power of 0.7 to detect a dose that had a 60% success rate for low-risk patients (HLA matched, in CR) and 45% for high-risk pts (mismatched, not in CR). High-resolution typing was prospectively available for 95% of donor-recipient pairs (HLA-A, B, C, DRB1 and DQB1 loci). Treatment plan: tacro from day-2; MTX 5 mg/m2 on days +1, +3, +6; day+11 was given only to controls. Pentos was given on days +8, +15, +22 and +30, in 4 ‘arms’: 0.5, 1, 1.5 and 2 mg/m2. Results: 147 pts, median age 49 yrs (18–72) were enrolled from 2001 to 2007. Diagnosis were AML/MDS (n=105), ALL(n=17), CML (n=18) and NHL (n=7); 48% of the pts were not in CR. Conditioning regimens (with ATG in 92% of the cases) were busulfan based (n=104), melphalan based(n=26), BEAM(n= 2), and CyTBI (n=15); 75% were ablative and 25%, reduced intensity. Stem cell source: bone marrow(n=120) and peripheral blood(n=27). Donors: UD(n=140) and MRD (n=7). Proportion of pts with donor-recipient HLA mismatches (HLA A, B, C, DRB1, DQB1) was 27%, 60%, 38%, 21% and 30%, respectively for the 5 study arms; median age was similar. 90% of the intended pentos doses were delivered. Time to engraftment was unafected. Incidence of toxicities (control vs study arms): renal(gd I-III) 34% vs 36%; TTP/HUS 6% vs 6%; early relapse 9% vs 5%; engraftment failure 3% vs 5.5%; delayed engraftment (〉21 days): 6% vs 5%. Extensive cGVHD rate (control vs 1.5 mg/m2) was 52% vs 38%. Relapse rates are comparable. Conclusions: aGVHD rate may be reduced with pentostatin 1.5 mg/m2. A multicenter controlled study is planned in order to confirm these findings. Pentostatin dose control group n=37 0.5 mg/m² n=10 1 mg/m² n=29 1.5 mg/m² n=61 2 mg/m² n=10 gd II-IV aGVHD 57% 44% 40% 33% 50% gd III-IV aGVHD 20% 33% 18% 10% 10% Skin only aGVHD 45% 25% 50% 63% 67% Hepatic aGVHD 17% 50% 14% 2% 0 gd III-IV aGVHD n pts with

Description: Cord blood (CB) is used to restore hematopoiesis in transplant patients lacking marrow donors. CB is associated with higher rates of delayed/failed engraftment. Peled et al developed an expansion technology using the copper chelator tetraethylenepentamine (TEPA), which enhanced the expansion of primitive CB populations when combined with early acting cytokines. A phase I clinical trial employing this technology was initiated. 10 patients with high-risk, heavily pre-treated hematologic malignancies (AML-2, ALL-5, HD-2, and NHL-1) have been enrolled with CB units that were cryopreserved in 2 fractions [20:80% (n=2), 40:60% (n=5) or 50:50% (n=3)]. 21days prior to infusion, AC133+ cells were isolated from the smaller (if unequal) or 50% CB fractions using the CliniMACS device and cultured for 21 days in media containing 10% FBS and SCF, FLT-3, IL6, TPO plus the copper chelator TEPA (Gamida). Patients then received myeloablative therapy with ATG and either fludara and busulfan (AML), or fludara, melphalan, thiotepa (ALL, HD, NHL) with infusion of the unmanipulated CB fraction on day 0, and the expanded fraction on day +1. GVHD prophylaxis was methotrex 5 mg/m2 days 2, 4,7, and tacrolimus for 6 months. The median age was 21 (range 7–51) and weight 69 (range 31–156) kg. The CB units were matched at 4/6 (n=8) or 5/6 (n=2) HLA antigens. The pre-thaw total nucleated cell (TNC) dose of the CB units was a median of 2.5x107/kg with post-thaw TNC of 2.4 x107/kg. Following AC133-selection, the manipulated CB fractions were a median of 73 (range 38–95)% AC133+ with a median of 0.650 (range 0.16–2.7) x106 TNCs, which were placed in culture. After 21 days of culture the expanded fraction had 69 (range 2–1638) x106 TNCs representing a 207 (range 2–616) fold TNC expansion. Patients received a total (expanded plus unmanipulated) median of 1.8 (range 1.1–6.1) x107 TNC/kg and 1.6x105 (range 0.4–49.9) CD34+ cells/kg. Two patients have CB cultures in progress. Of the 8 patients transplanted, 1 had autologous recovery with relapse of AML on day 30 and death. Of the remaining patients, 7 were evaluable for neutrophil engraftment and 4 of them for platelet engraftment (2 too early for platelet evaluation and 1 early death). The median time to engraftment was 27 days for neutrophils (range 16–46) and 48 days for platelets (range 27–96). Preliminary analysis suggest a correlation between a shorter time to neutrophil engraftment and total TNC/kg infused (p=0.02), and a trend for CD34+ cells/kg infused (p=0.09). Three patients have developed grade ≤2 acute skin GVHD and one had chronic extensive GVHD of the skin and GI tract; all resolved with steroids. One patient (without GVHD) died of a systemic viral infection on day 56, despite adequate neutrophil recovery (not platelets). All of the remaining patients are all alive and free of malignancy at a median follow-up of 4 (range 1–16) months. Conclusion: There was no toxicity associated with infusion of the TEPA-expanded CB cells. Additional data is necessary to determine the efficacy of this approach. Future directions include the expansion of the entire CB unit and removal of methotrex from the GVHD regimen to improve time to neutrophil engraftment, as well as comprehensive assessment of immune reconstitution.

Description: GVHD remains a major obstacle to a successful unrelated (UD) or mismatched related (MRD) donor hematopoietic stem cell transplantation (HSCT). Pentostatin is a purine nucleoside analog that targets adenosine deaminase and leads to lymphocyte depletion, with low potential for myelosuppression. We are investigating the incorporation of pentostatin to our standard GVHD prophylaxis regimen with tacrolimus (tacro) and methotrexate (MTX). Methods: This is an adaptive randomized, dose finding study that takes into account toxicity and efficacy in a Baysean “play the winner” design. The “winner” dose moves to the phase II portion of the study. Probability of assignment to the control group was fixed at 20%. Recipients of UD and MRD are eligible; all analysis is done by intention to treat. Success was defined as being alive, engrafted, in complete remission(CR), without GVHD at study completion (100 days post HSCT). Development of grade III-IV acute (a) GVHD defined failure at any time, while grade I–II did not constitute failure if absent by day 100. This design has power 0.7 to detect a dose that has a success rate of 60% for low-risk patients (HLA matched, in CR) and 45% for high-risk patients (mismatched, not in CR). High-resolution typing was available for all donor-recipient pairs at HLA-DRB1 and -DQB1 loci, and to 83% of the pairs at HLA-A and -B loci; all patients had low-resolution -C typing. Treatment plan: tacro from day -2 (target level of 5–15 ng/ml) and MTX 5 mg/m2 on days +1, +3, +6 to all patients; day +11 was given only to the control group. Pentostatin was given on days +8, +15, +22 and +30, in treatment arms: 0.5 mg/m2, 1 mg/m2, 1.5 mg/m2, and 2 mg/m2. Results: 73 patients, median age 45 yrs (range 18–72) have been enrolled. Diagnosis were AML/MDS(n=48), ALL(n=8), CML (n=10) and NHL(n=7); 58% of the patients were not in CR at HSCT. Conditioning regimens were busulfan based(n=52), melphalan based(n=10), BEAM(n= 2), and CyTBI(n =9); 71% were ablative and 29%, reduced intensity. ATG was used in the regimen in 86% of the cases. Stem cell source: bone marrow (n=67) and peripheral blood(n=6). Donors: UD(n=67) and MRD(n=6). Proportion of patients with donor-recipient HLA mismatches was 24%, 20%, 33%, 21% and 40%, respectively for the 5 study arms; median age was similar. 85% of the intended pentostatin doses were delivered. Pentostatin did not delay engraftment. Incidence of toxicities (control vs. study arms): renal (all grade I/II)=47% vs 36%; TTP/HUS= 12% vs 9% (more severe among pentostatin patients); early relapse= 12% vs 5%; engraftment failure=6% vs 3%; delayed engraftment (〉21 days): 0 vs. 5%. Probability that dose 1.5 mg/m2 is better than control is 0.9341. Pentostatin dose control group (n=17) 0.5 mg/m2 (n=10) 1 mg/m2 (n=12) 1.5 mg/m2 (n=24) 2 mg/m2 (n=10) gd II-IV aGVHD 47% 44% 63% 29% 50% gd III-IV aGVHD 20% 33% 27% 10% 10% CMV reactivation 41% 30% 33% 50% 50% bact/fungal infection 59% /12% 60% /10% 50% /18% 55% /17% 70% /10% Not evaluable n=2 n=1 n=1 n=4 n=0 Failure rate 47% 70% 33% 29% 40% Conclusions: This preliminary analysis indicates that aGVHD rate may be reduced with pentostatin 1.5 mg/m2, without interference with engraftment. Longer follow-up and larger number of patients will be needed to assess impact on survival.

Description: Most patients (pts) transplanted with active myeloid leukemias will relapse after HSCT. Intensity of the conditioning regimen is an important component of disease control. We hypothesized that GO could safely increase the anti-disease activity of FM, and investigated this hypothesis in the trial reported here. Patients and Methods: Objective: to determine the dose of GO with the lowest toxicity (tox), and the highest response probabilities. Tox: grade 3–4 organ tox, engraftment failure or early death (ED; first 30 days post HSCT). Response was defined as no tox, engraftment, and remission (CR) on day +30. Trial was designed with GO doses of 4, 6 and 9 mg/m2 but given tox observed at 4mg/m2, doses 6 and 9 mg/m2 were not used and dose 2mg/m2 was added. Eligible were pts aged 12–75 years, not candidates for high-dose regimens or with high-risk CD33+ disease. Treatment: GO 2 or 4mg/m2 day -12, F 30mg/m2 (days -5 to -2), M 140mg/m2 (day -2); HSCT day zero. ATG was added in unrelated (UD) HSCT. Graft-versus-host disease (GVHD) prophylaxis: tacrolimus and mini-methotrexate (5mg/m2 day+1,+3,+6,+11). GO was given on day -12 to minimize probability of delaying engraftment. We treated 52 pts, median age 53 yrs (13–72), with AML (n=47), MDS (n=4) or blast crisis CML(n=1). Disease status at HSCT: CR (n=3),induction failure (n=15), first/second relapse (n=33), untreated MDS (n=1). 18 pts had a previous HSCT (allogeneic, n=11 or autologous, n=7). At study entry, median number of bone marrow blasts was 17% (0–95%); 33% of the pts had circulating blasts, 4 pts were FLT3 positive and 38% had poor prognosis cytogenetics. Median blast CD33 expression: 89% (22.5–99.6). Donors: related (n=33) or UD (n=19). Stem cell source was peripheral blood (n=43) or bone marrow (n=9). Results: Median day 30 donor cell chimerism was 100%. Median time to ANC〉 500/mm3 was 13 days. Eight pts received GO at 4 mg/m2; 2 died early (due to pneumonia and renal failure) and 2 developed gd III gastro-intestinal and renal tox. 44 pts were then treated at 2 mg/m2: 3 pts died of regimen-related tox (2 ED due to pulmonary bleeding and sepsis, and 1 death within the first 100 days due to pneumonia, renal failure/TTP/HUS). 100-day treatment-related mortality (including 2 deaths due to aGVHD and 1 due to infection) was 15% (n=8). Tox included reversible gd III-IV bilirubin (n=2), transaminase elevation (n=5), moderate hepatic VOD (n=1) and gd I-II GI tract (n=39). CR rate was 90%; 1 pt failed to respond. Gd II-IV and III-IV aGVHD rates were 44% and 23%, respectively, and 58% developed cGVHD. Median follow-up is 9 months (2.5–36) for surviving pts (n=29); 17 pts have relapsed. Median event-free-survival (EFS) is 3.8 mos. Median EFS of a historic control group treated with FM140 (n=36) was 2.2 mos (Figure). Tox of GOFM is similar to that documented by our group with FM140 mg/m2 in a somewhat better prognosis population.(de Lima et al. Blood.2004; 104:865–72) Conclusion: GO 2mg/m2 can be safely added to FM, and may improve the anti-leukemic efficacy of the regimen. EFS - patients with active disease at HSCT EFS - patients with active disease at HSCT

Description: After autologous or allogeneic transplants of peripheral blood stem cells (PBSC), an adequate dose of CD34+ cells is necessary to ensure early and sustained hematopoietic engraftment and favorable clinical outcome. There are no comparable data on the relationship between CD34+ cell dose and recovery after allogeneic bone marrow transplants (BMT). Twenty-eight patients with hematologic malignancies received a BMT from an HLA-identical sibling, using T-cell depletion and cyclosporin for graft-versus-host disease prophylaxis and delayed donor lymphocyte transfusions in an attempt to prevent leukemia relapse. The treatment-related mortality (TRM), primarily due to infections and cytopenias, was significantly higher for 13 patients receiving less than 1 x 10(6) CD34+ cells/kg (64.9% +/- 12.8% v 6.9% +/- 6.4%, P = .003). Survival at a median follow-up of 1 year was also lower in the group receiving less than 1 x 10(6) CD34+ cells/kg (30.8% +/- 12.8 v 74.3% +/- 13.7%, P = .005). The CD34+ cell dose was the only variable significantly associated with TRM. The dose of CD34+ cells also correlated with speed of hematopoistic recovery. Patients receiving more than 2 x 10(6) CD34+ cells/kg showed significantly earlier recovery of monocytes and a trend for earlier recovery of lymphocytes. They achieved platelet and red blood cell transfusion independence earlier, required less granulocyte colony-stimulating factor support during ganciclovir treatment, and spent fewer days in the hospital after transplantation. These results suggest that, for allogeneic T-cell-depleted BMT, the higher CD34+ cell doses may improve outcome in engrafting patients.

Description: Donor-recipient histocompatibility is a major determinant of outcome after UD HSCT, and matching pairs by high-resolution (Hi-res) HLA typing is likely to improve results. Methods: We studied 66 pts with myeloid leukemias in complete remission (CR) transplanted with UD bone marrow (n=58) or peripheral blood (n=8) from 01/02 to 12/04. Preparative regimens were ablative [Busulfan (Bu)/Cyclophosphamide (Cy)(n=13), Cy/TBI (n=2), Fludarabine (Flu)/Bu 130 mg/m2x4 doses (n=33)] and reduced intensity [Flu/Bu 130 mg/m2x2 doses+Gleevec (n=7), Flu/Melphalan 140mg/m2 (n=11)]. ATG was given in 61 cases. GVHD prophylaxis was tacrolimus and mini-methotrexate, with additional pentostatin in 24 pts. Hi-res typing of HLA-A, B, DRB1 and DQB1 loci was done prospectively; HLA-DPB1 and C loci were typed prospectively in 50% of cases and retrospectively in 50%. Hi-res typing was sequencing-based for HLA-A, B, DRB1, SSP-based for DRB3/4/5, DQB1 and DPB1; and SBT/SSOP-based for HLA-C. A Cox’s proportional hazards regression model was used to study overall (S), event-free (EFS) and aGVHD-free survival. Variables with a p-value

Description: RIC with FM has extended the use of HSCT to patients otherwise not eligible for this treatment. Longer follow-up and larger number of patients now allow for more robust evaluation of risk factors and outcomes. Herein are the results of such an evaluation. Patients and Methods: We evaluated outcomes of 112 patients with high-risk AML/MDS treated from August 1996 to December 2003 with FM (fludarabine 100–180 mg/m2 and melphalan 100–180 mg/m2) and unmanipulated HSCT. Eligibility included age 〉54 yrs. or comorbidity precluding an ablative preparative regimen. Disease status at HSCT was relapsed/refractory (n=43, 38.4%), primary induction failure (n=32, 28.6%), untreated (n=7, 6.3%) or complete remission (CR, n=30, 26.8). Cytogenetic risk was intermediate (n=59, 53%), high (n=47, 42%), low (n=3, 2.5%) or unknown (n=3, 2.5%). Donors were HLA matched related (MRD; n=59) or unrelated (UD; n=53). GVHD prophylaxis was tacrolimus based in all but one patient. Anti-thymocyte-globulin was added in 31 UD HSCT. Stem cell sources were bone marrow (n=56) or peripheral blood (n=57). Median age was 55 (range 22–74). Evaluated were the following variables and their influence on disease progression and overall survival: - age, donor type, duration of first CR, disease status at transplant (categorized as CR, No CR with (NoCR/CB) and without circulating blasts (NoCR/NoCB)), cytogenetics, acute and chronic GVHD (time dependent variables), and blood counts on day 30 (lymphocytes, monocytes and platelets). We used a Cox’s regression analysis. Results: Median time of follow up among survivors (n=43) was 28.4 mo (3.3–88.9). CR rate at day 30 post transplant was 87% (n=97), 8 patients died early and 7 did not respond. 25 (26%) of 97 patients progressed after day 30. All but 3 patients relapsed within the first year post HSCT, and only one relapsed more than 2 years after HSCT. In a landmark analysis, disease status at transplant was the only significant risk factor for progression among these 97 patients (HR of 3.7 for the NoCR/CB group compared to the CR group). 69 of 112 patients died with a median survival of 4.6 mo. Seven deaths (10% of all deaths) were observed more than 2 yrs. after HSCT, due to GVHD (n=3), infection (n=2), relapse (n=1) and unknown causes (n=1). Two-year OS and PFS was 44% and 69% respectively. Disease status at HSCT and grade II-IV aGVHD were the only significant predictors of OS on univariate and multivariate analysis. Blood counts on day 30 were associated with disease status at transplant, donor type and aGVHD. Their independent effect on outcome could not be evaluated given sample size. Conclusion: A significant portion of older patients with high-risk AML/MDS may achieve long-term PFS, but early relapses are the major cause of treatment failure in this context. Prognostic factors for event-free and overall survival Variables Multivariate analysis for disease progression CB=circulating blasts Disease status n Events (n) HR 95% CI p 2-yr PFS CR 30 6 1.0 57% (39–72) NoCR/NoCB 41 7 1.1 0.4–3.2 0.9 46% (30–60) NoCR/CB 26 12 3.7 1.4–9.8 0.001 22% (9–38) Multivariate analysis for overall survival (OS) Disease status HR 95% CI p 2-yr OS CR 30 12 1.0 66% (48–80) NoCR/NoCB 49 29 1.8 0.9–3.5 0.06 40% (26–53) NoCR/CB 34 28 2.8 1.4–3.5 0.002 23% (11–37) gd II-IV aGVHD 2.8 1.8–4.6

Description: We previously demonstrated early safety and efficacy of fixed-dose IV BuFlu. Here we compare IV BuFlu and IV BuCy2, with long follow-up in a large number of pts. Since pts were treated on consecutive, not randomized programs, we used pts receiving Flu-Melphalan (M) in a program spanning the time of the IV Bu-based studies to estimate/correct for this non-randomized “period” effect. Methods: 293 pts received Bu 130 mg/m2, Flu 40 mg/m2 (after April 2001; n=148); Bu 0.8 mg/kg × 16, Cy 60 mg/kg × 2 (before April 2001; n=67); Flu 30mg/m2 × 4, M 140–180 mg/m2 (FM; before and after April 2001; n=78). Patients received FM mainly based on higher age and coorbidities. Graft-vs-host dx (GVHD) prophylaxis was tacrolimus-based. Bayesian log-normal regression models were used to assess covariate- and treatment effects on survival, non-relapse mortality (NRM) and time to progressive disease (TPD). Covariates and survival model are shown in the table. The model accounts for the “period” effect, the effects of Bu-vs-FM, and Flu-vs-Cy within IV Bu pts. Patient and treatment characteristics were (BuFlu, BuCy2 and FM): median age 46 (19–66), 39 (13–64) and 54 (22–74) years. At HSCT 47%, 48% and 21% of the patients were in CR1-3. Donors were HLA-identical siblings in 53%, 79% and 41%. 30% of the Bu groups and 47% of FM had poor prognosis cytogenetics. Results: Median follow-up in months (BuFlu, BuCy2 and FM; alive pts): 30 (9–56; n=80; all but 2 pts followed for at least 1 yr), 69 (24–102; n=23) and 48 (17–89, n=25). 100-day NRM for pts in CR at HSCT was 2% and 9% after Bu-Flu and BuCy2, while 1 yr-NRM was 7% and 18%. BuFlu produced significantly better EFS (P=0.04) and overall NRM (P=0.02) than BuCy2, but TPD was similar (P=0.34). 2-yr actuarial survival (CR pts): 75% (BuFlu) vs. 50% (BuCy), P=0.01. Conclusion: After accounting for covariate and treatment period effects, there was a significant benefit to being treated with BuFlu compared to BuCy2 in spite of a higher median age and proportion of MUDs. The benefit was most evident for CR-pts due to a significant reduction in NRM, without loss of antileukemic activity due to exchanging Cy with Flu. Fitted Bayesian log-normal overall survival model Variable Mean SD Posterior 95% credible interval 2.5% 97.5%* Probability of beneficial effect *1=beneficial effect; 0=low likelihood of beneficial effect Cytogenetics −0.85 0.4 −1.652 −0.045 0.017 Disease status (CR vs other) 1.12 0.55 0.019 2.15 0.975 Donor (sibling vs other) 0.606 0.237 0.149 1.056 0.992 Age −0.022 0.010 −0.041 −0.003 0.017 Circulating blasts (0 vs 〉0) 0.951 0.268 0.454 1.463 1 Platelet count 0.002 0.001 0.000 0.004 0.967 Period effect −0.914 0.458 −1.795 0.041 0.029 IV Bu (vs FM) −0.513 0.481 −1.455 0.420 0.148 BuFlu (vs Bucy2) 1.476 0.544 0.406 2.555 0.995 Survival – patients in CR at transplant Survival – patients in CR at transplant