Maziarz, Richard Thomas

Abstract Abstract 680 Cytogenetics play an essential role in determining the prognosis of patients with AML. However, there is still no validated cytogenetics grouping scheme that specifically applies to patients undergoing allogeneic stem cell transplantation, which hampers accurate prognostication and risk stratification. We studied 821 adult patients (median age 41, range 18–74) who underwent SCT between 1999 and 2004 for AML (excluding APL) in CR1 or CR2 and who were reported to the CIBMTR from centers with >20 patients meeting study criteria. 75% of patients received a myeloablative conditioning. 496 patients had a normal karyotype. The primary cytogenetics reports were manually reviewed for 92% of the patients with an abnormal karyotype. We compared the ability of the 6 existing grouping schemes (MRC, CALGB, EORTC/GIMEMA, SWOG/ECOG, DFCI, and Monosomal Karyotype (MK) classifications) to stratify patients, using both the Akaike Information Criterion in multivariable models and the C-statistic. Among all existing schemes, the DFCI system provided a marginally superior stratification for overall and leukemia-free survival. We also built a new classification using individual cytogenetic abnormalities in a Cox model that included other significant covariates (performance status, therapy-related disease, conditioning intensity, graft source, donor match, duration of CR1, and gender match). This CIBMTR scheme (see Table), which is similar to the DFCI scheme except for the inclusion of patients with t(8;21) in the intermediate group, could stratify patients into 3 groups with similar treatment-related mortality but significantly different overall survival, leukemia-free survival, and incidence of relapse. This scheme appeared to apply to both patients in CR1 and in CR2 (see FiguresF2). This transplant-specific scheme could be adopted for prognostication purposes and to stratify patients with karyotypic abnormalities entering transplantation clinical trials. Future studies may clarify the relative outcome of patients with t(8;21) and refine this scheme with the inclusion of molecular abnormalities. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 1987 Background. MultiStem®, a multipotent adult progenitor cells (MAPC) collection, is an immune-modulatory, bone marrow-derived adult adherent allogeneic stromal cell product manufactured to large scale, assuring consistency of universal donor cell product during clinical evaluation. Pre-clinical studies showed safety of IV infusion and survival benefit in a haploidentical transplantation rat model of acute GVHD (Kovacsovics 2008, 2009). Methods. An open label Phase I clinical dose escalation study is being conducted with the primary goal to assess safety of MAPC as an adjunct treatment for human subjects undergoing myeloablative allogeneic HSCT for hematologic malignancies. Patients have been enrolled for MultiStem® administration as a single dose or in multiple weekly doses. Infusional toxicity and RRTs are assessed for 30 days following the last study drug dose administration. Secondary endpoints include incidence of acute GVHD, infection and survival through day 100. Dose escalation is guided by the Continual Reassessment Method (CRM). Results. In the single dose arm, 18 patients with AML (9), MDS (4), CML (3) and ALL (2) from 5 clinical centers were administered MultiStem® IV at 1, 5, or 10 million cells per kg at day 2 after allogeneic HSCT. There was no observed infusional toxicity. Two patients experienced Bearman RRTs (Gr 3 mucositis; Gr 3 renal & pulmonary failure) deemed unrelated to study product. Engraftment occurred in all 18 patients. The median time to neutrophil engraftment for related (n=7) and unrelated (n=11) HSCT was 15 days (range, 12–16 days) and 14 days (range, 11–25 days), respectively. The 100-day cumulative incidence of Grade II-IV and III-IV GVHD was 28% and 6% (n=18). In the highest MultiStem® dose group (n=9) there was no grade III-IV GVHD, and only one case of grade II GVHD, which subsequently resolved with treatment. 17/18 patients have been enrolled in the multidose arm. Results of safety endpoints and interim results of all secondary endpoints of repeated MultiStem® administrations following HSCT will be presented. Conclusion. Single dose administration of MultiStem® is well-tolerated, without observation of infusional toxicity or graft failure. Together with the observed low incidence of severe aGVHD, these findings are supportive of the concept that stromal stem cell therapy is safe and may be harnessed as an efficacious therapeutic option for acute GVHD prophylaxis following HSCT. Disclosures: Maziarz: Athersys, Inc: Patents & Royalties, Research Funding. Perry:Athersys, Inc.: Employment. Deans:Athersys, Inc: Employment. Van't Hof:Athersys, Inc.: Employment.

Abstract BMT CTN 0201 was a randomized study of unrelated donor bone marrow (BM) vs. peripheral blood (PB) (N=551) in hematopoietic cell transplantation (HCT) for hematologic malignancies. The primary analysis after 2 years of follow up showed similar survival, disease-free survival and treatment-related mortality between the graft types. There was a higher rate of graft failure with bone marrow (9% vs. 3%, p=0.002) and a higher rate of chronic GVHD with peripheral blood (53% vs. 41%, p=0.01). (Anasetti et al, NEJM 2012) Patient reported outcomes (PROs) were collected from patients > 16 years old at enrollment and 0.5, 1, 2 and 5 years after transplantation with the Functional Assessment of Cancer Therapies - Bone Marrow Transplant (FACT-BMT), Mental Health Inventory (MHI), and Lee chronic graft-versus-host disease (cGVHD) symptom scale. At 5 years, 102 BM and 93 PB participants were alive and eligible for the quality of life (QOL) study. Clinically meaningful differences (changes in scores that are noticeable to patients) were considered 0.5 x standard deviation of the total population based on the distribution method. Results: Data as of 5/25/15 were analyzed. 70% completed the pre-randomization assessment, and 74-78% of adult survivors completed the 5 year assessment. Age < 30 at transplant and high risk disease were associated with missing 5 year data, but not graft source. Response rates at 0.5, 1 and 2 years were 28-43% of survivors. There are no differences in any of the primary QOL scores in the first two years after HCT using univariate comparisons, although missing data severely limit conclusions during this period. At 5 years, the FACT-Trial Outcome Index (TOI), the MHI Psychological Well-Being, and the cGVHD symptom scale scores are all significantly better for BM patients, although only the latter two are still significant after adjustment for multiple testing (p<0.0125 because of 4 primary QOL variables). Results were similar when tested in multivariate models adjusting for baseline patient-reported scores and also imputing missing values based on patient characteristics (Table). Of the 7 chronic GVHD subscales, symptoms in the eye, lung, and energy were significantly better with BM (p<0.01). A diagnosis of cGVHD was highly associated with patient-reported cGVHD symptoms but not with QOL or psychological status. Inclusion of extensive cGVHD in the multivariate models did not change the significance of PRO differences suggesting that differences in cGVHD incidence do not explain the PRO findings. With a median follow up of 73 months for survivors, no difference in survival between PB and BM is observed (p=0.84, Figure). Conclusion: At 5 years after transplant, recipients of unrelated donor BM have better psychological well-being and less burdensome chronic GVHD symptoms than recipients of PB. Survival rates are similar. Table. Comparisons of primary QOL variables at 5 years, adjusted for QOL values at baseline and missing data using inverse probability weighting using significant clinical characteristics. QOL scale Bone marrow (n=102) Peripheral blood (n=93) P value Clinically significant difference* Difference between BM and PB (95% CI) FACT-BMT TOI, mean +/- SE (higher scores better) 76.7 +/- 1.6 (n=79) 70.5 +/- 1.9 (n=69) 0.014 8.5 6.2 (1.3-11.1) MHI - Psychological well-being, mean +/- SE (higher scores better) 78.9 +/- 1.7 (n=80) 72.2 +/- 1.9 (n=72) 0.011 8.4 6.7 (1.6-11.8) MHI-Psychological Distress, mean +/- SE (lower scores better) 16.0 +/- 1.3 (n=80) 19.0 +/- 1.5 (n=71) 0.128 6.5 -3.0 (-6.8,0.9) Chronic GVHD symptoms, mean +/- SE (lower scores better) 13.1 +/- 1.5 (n=80) 19.3 +/- 1.6 (n=72) 0.004 7.1 -6.3 (-10.5, -2.0) SE, standard error *0.5 x STD Figure 1. Figure 1. Disclosures Lee: Bristol-Myers Squibb: Consultancy; Kadmon: Consultancy. Maziarz:Athersys: Consultancy, Patents & Royalties, Research Funding; Novartis: Consultancy.

Abstract Abstract 3224 Poster Board III-161 Background While most centers use 2 × 106 CD34+ cells/kg as the minimal cell dose for autologous hematopoietic stem cell (HSC) transplantation (auto-HSCT), infusion of higher CD34+ cell dose is associated with better outcomes in patients with multiple myeloma (MM) or non-Hodgkin's lymphoma (NHL). Recent evidence suggests a correlation between CD34+ cell yield on Day 1 of collection and total CD34+ cell yield as well as post-transplant outcomes. This analysis was designed to: 1) compare Day 1 collection between patients with NHL or MM mobilized with plerixafor plus G-CSF or placebo plus G-CSF; and 2) determine whether Day 1 CD34+ cell yields correlated with the total mobilization yield and number of apheresis days. Methods Data were obtained from two prospective, randomized, double-blind, placebo-controlled, phase 3 clinical trials that compared the safety and efficacy of plerixafor (0.24 mg/kg/day SQ) plus G-CSF (10 μg/kg/day) with placebo plus G-CSF for mobilization of HSC for auto-HSCT in patients with NHL (3101 Study) or MM (3102 Study). Pearson correlation coefficient was used to evaluate the association of day 1 CD34+ cell collection with total CD34+ cell yield and the number of days of apheresis. Results In the NHL trial, 150 patients were mobilized with plerixafor plus G-CSF and 148 patients underwent mobilization with placebo plus G-CSF. More than half the patients (55.3%) in the plerixafor group collected ≥2 × 106 CD34+ cells/kg on Day 1 of apheresis (Figure 1A). In contrast, 19.6% patients in the placebo group collected ≥ 2 × 106 CD34+ cells/kg on Day 1 of apheresis (p< 0.001). In the MM study, 148 patients were mobilized with plerixafor plus G-CSF and 154 patients were mobilized with placebo plus G-CSF. More than half the patients (52.7%) in the plerixafor group collected ≥6 × 106 CD34+ cells/kg on the first day of collection compared to only 16.9% patients in the placebo group (p<0.001; Figure 1B). There was a strong positive correlation between day 1 collection and the total CD34+ cell yield in patients with NHL (r= 0.86, p-value= <0.0001) or MM (r= 0.87, p-value= <0.0001) in both the plerixafor and placebo groups. For NHL patients, the median Day 1 collection was higher in the plerixafor group compared to the placebo group: 2.66 × 106 vs. 0.77 × 106 CD34+ cells/kg (p<0.001) and this translated into higher total CD34+ cell yields in the two groups respectively: 5.69 × 106 vs. 1.98 × 106 CD34+ cells/kg (p<0.001). Similarly, for MM patients, the median CD34+ cells/kg collected on Day 1 was higher in the plerixafor group compared to the placebo group: 7.01 × 106 vs. 2.29 × 106 CD34+ cells/kg (p<0.001) and this translated into better overall collection in the plerixafor vs. placebo groups: 10.96 × 106 vs. 6.18 × 106 CD34+ cells/kg (p<0.001). A negative correlation was observed between CD34+ cells collected on Day 1 and the number of days of apheresis performed in patients with NHL (r= -0.67, p-value=<0.0001) or MM (r= -0.50, p-value= <0.0001) in both the plerixafor and placebo groups. Consequently, better Day 1 collection in plerixafor-treated NHL or MM patients translated into significantly fewer apheresis days to achieve the target collection compared to placebo treated patients. Conclusions These data support previous reports demonstrating a strong correlation between day 1 CD34+ cell collection and total CD34+ cell yield and apheresis days. These data also demonstrate that addition of plerixafor to G-CSF allows significantly more patients to achieve the target cell collection within 1 day of apheresis compared to G-CSF alone. These findings support the observation that mobilization with plerixafor plus G-CSF reduces the number of apheresis days required to achieve the minimal or optimal cell dose to proceed to transplantation. Disclosures Bolwell: Genzyme Corporation: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Nademanee:Genzyme Corporation: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Stiff:Genzyme Corp.: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Stadtmauer:Genzyme Corporation: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Maziarz:Genzyme Corp.: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Micallef:Genzyme Corporation: Membership on an entity's Board of Directors or advisory committees, Research Funding. Marulkar:Genzyme Corporation: Employment, Equity Ownership. Gandhi:Genzyme Corporation: Employment, Equity Ownership. DiPersio:Genzyme: Honoraria.

Abstract Abstract 33 Background: Circulating levels of peripheral blood (PB) CD34+ cells/μl are a strong predictor of hematopoietic stem cell (HSC) yields in patients with non-Hodgkin's lymphoma (NHL) undergoing autologous HSC transplantation (auto-HSCT), and are routinely monitored to optimize the timing and success of HSC collection after cytokine ± chemotherapy mobilization. The threshold PB CD34+ cell count to initiate apheresis varies from 5-20 cells/μl, depending on the institution. This analysis compared the efficacy of plerixafor + G-CSF to placebo + G-CSF for HSC mobilization in NHL patients with pre-apheresis PB CD34+ cells/μl <10. Methods: Data were obtained from a randomized, double-blind, phase 3 clinical trial comparing the safety and efficacy of plerixafor (0.24 mg/kg/day SC) + G-CSF (10 μg/kg/day) to placebo + G-CSF for mobilization and auto-HSCT in NHL patients. PB CD34+ cell count was measured on Day 4, before the first plerixafor/placebo dose, and on Day 5, 10-11 hours post study drug treatment. The proportion of patients collecting ≥2 × 106 (minimal) or ≥5 × 106 (optimal) CD34+ cells/kg, apheresis yields, and time to engraftment were compared between the plerixafor and placebo groups for patients with PB CD34+ cells/μl <10. Results: 77/150 (51%) patients and 73/148 (49%) patients in the plerixafor and placebo groups, respectively, had PB CD34+ cells/μl <10 on Day 4, prior to the first dose of study drug. Patient characteristics were similar between both groups. As shown in Table 1, addition of plerixafor to G-CSF resulted in a statistically significant increase in the absolute PB CD34+ cells/μl on Day 5 compared to G-CSF alone (p<0.001). The median fold increase in PB CD34+ cells from Day 4 to Day 5 was significantly higher in plerixafor-treated patients vs. placebo-treated patients: 6.0-fold vs. 1.6-fold (p<0.001). In these hard to mobilize patients, the median CD34+ cell yield after 2 days was significantly higher with plerixafor + G-CSF compared to placebo + G-CSF: 2.92 vs.0.94 × 106 cells/kg (p<0.001), respectively. The median CD34+ cell yield after 4 days was also significantly higher in the plerixafor vs. placebo groups: 3.96 vs.1.25 × 106 cells/kg (p<0.001). Plerixafor + G-CSF allowed a significantly greater proportion of patients with PB CD34+ cells/μl <10 to collect the minimal cell dose in 4 apheresis days: 77.9% vs. 34.2 % patients in the plerixafor and placebo group, respectively collected ≥2 × 106 CD34+ cells/kg in 4 days (p<0.001). The success rate of collecting the ≥5 × 106 CD34+ cells/kg in 4 days was 40.3% with plerixafor + G-CSF compared to only 9.6% with placebo + G-CSF (p<0.001). The proportion of patients collecting the optimal or minimal cell dose in 2 days was also significantly higher in the plerixafor vs. placebo group (p<0.001). The median time to platelet (19-21 days) and neutrophil (10-11 days) engraftment was similar in both groups. Similar, statistically significant increases in PB CD34+ cell collections were obtained when the efficacy of plerixafor + G-CSF was compared to placebo + G-CSF, in patients with a Day 4 PB CD34 cells/μl<20 (data not shown). Conclusions: Plerixafor + G-CSF allowed collection of the minimal transplantable cell dose in 78% patients with PB CD34+ cells/μl <10. These data show that adding plerixafor to G-CSF substantially reduces the risk of mobilization failure in patients with NHL predicted to be poor mobilizers based on low PB CD34+ cell counts. Thus, mobilization with plerixafor + G-CSF in patients with PB CD34+ cells/μl <10 increases stem cell collection efficiency allowing patients to pursue auto-HSCT with more optimal cell doses. Disclosures: Maziarz: Genzyme Corp.: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Micallef:Genzyme Corporation: Membership on an entity's Board of Directors or advisory committees, Research Funding. Stiff:Genzyme Corp.: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Bolwell:Genzyme Corporation: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. Marulkar:Genzyme Corporation: Employment, Equity Ownership. Calandra:Genzyme Corporation: Consultancy, Equity Ownership. DiPersio:Genzyme Corporation: Honoraria.

Abstract Background: p38a mitogen-activated protein kinase (MAPK) mediates production of proinflammatory cytokines and other factors including PGE-2 induced RANKL and IL-1 induced IL-6. SCIO-469 inhibits p38a MAPK blocking synthesis of Il-1ß, PGE2, VEGF, MIP-1α and TNFα thus reducing in vitro MM tumor growth and survival, and drug resistance. Objectives: Assess efficacy, safety, tolerability of SCIO-469 as M or MB in patients (pts) with relapsed and refractory multiple myeloma. Methods: Pts were treated with SCIO-469 alone (60 mg po tid)and assessed on Days 15, 30, 52 and 73. Bortezomib (1.0–1.3mg/m2 d 1,4,8,11 q21d) was added to M for PD or SD. Pts who had clinical benefit on either M or MB at Day 73 were allowed to continue on extension study for up to a total of 168 d, to further evaluate efficacy and saftey of M or MB. Results: 62 pts were enrolled: 39 M, 23 F, median age was 63 years (range 48–84), D-S stage at Dx: I (6%), II (32), III (60%%); median time from diagnosis to treatment 3.1 years; median # prior Rx = 5; including 29% with > 6; 71% prior transplant. 27 had received prior bortezomib and 17 were considered refractory. All 62 pts received study drug; 34 pts received MB at some time during the treatment period. 5 of the 28 (18%) pts treated with M and 25 of 34 (74%) pts treated with MB continued on the extension study. Median time on study was 77.5 d (range 7–304) for M and 180.5 d (range 34–275) for MB. Best response (EBMT; Blade 1998) to M was stable disease in 24% of pts; the best response to MB was 9/34 PR (26%), 2/34 MR (6%),3/34 SD (9%). Median time to progression (TTP) was 50 days for M compared to 140 days for MB. Safety/Tolerability: Adverse events on M were tabulated separately from those on MB. Neutropenia was minimal, < 3% with either regimen; anemia (47 vs 29%) and thrombocytopenia (26 vs 16%) were more prevalent with MB vs M, respectively. Epistaxis, fatigue and diarrhea were seen in > 15% of pts receiving M. Nausea, vomiting, diarrhea, constipation, abdominal pain, fatigue, pyrexia, anorexia, URI, arthralgia, muscle cramp, dizziness, peripheral neuropathy and headache occurred in > 15% pts receiving MB. 10 pts died on study, 9 during primary phase and 1 during extension phase. 3 deaths were the result of sepsis, the remaining 7 were due to progressive disease. Conclusions: SCIO-469, 60 mg tid, as monotherapy or in combination with Bortezomib was well tolerated. Although there were no objective responses in M (compared to 32% (11/34) in MB-including 4 patients who had failed prior bortezomib), 24% had stable disease at the end of monotherapy. Further studies to refine the dose of SCIO-469 as monotherapy and of SCIO-469 in combination with other agents are warranted.

Abstract AMD3100, Plerixafor (A)+G-CSF (G) have safely and effectively allowed aHSC mobilization in Phase I and II studies. This Phase III, multicenter, randomized, double-blind, placebo controlled study compared the safety and efficacy of A+G versus Placebo (P)+G to mobilize and transplant patients with NHL. Methods: Adult NHL patients requiring an aHSC transplant, in first or second CR or PR were eligible to participate. Patients received G (10μg/kg/day) subcutaneously (SQ) for 4 days. On the evening of Day 4, patients received either A (240μg/kg SQ) or P. Patients underwent apheresis on Day 5 after an AM dose of G and 10–11 hours after administration of study treatment. Patients continued to receive the evening dose of study treatment, followed by AM dose of G and apheresis for up to a total of 4 aphereses or until ≥5 x 106 CD34+cells/kg were collected. Only study cells were used for transplant. Patients who failed to mobilize ≥2 x 106 CD34+cells/kg could enter into a rescue arm of A+G, without unblinding of randomized treatment. The primary endpoint was the percentage of patients who achieved ≥5 x 106 CD34+cells/kg in 4 or less aphereses days. All patients will be followed for ≥12 months post-transplant. Results: 298 patients were enrolled and randomized. All patients have completed 100 days follow-up and are included in this intent-to-treat analysis. Baseline characteristics were similar between groups. The primary endpoint was met in 89/150 (59%) patients in the A+G group and 29/148 (20%) patients in the P+G group, p<0.0001. 130/150 (87%) patients in the A+G group and 70/148 (47%) patients in the P+G group collected ≥2 x 106 CD34+cells/kg in 4 or less aphereses days, p<0.0001. Kaplan-Meir Estimates of Time to Trget Collection of ≥ × 106 CD34 + cells/kg Kaplan-Meir Estimates of Time to Trget Collection of ≥ × 106 CD34 + cells/kg The figure shows that more patients in the A+G group reached target after 1 day of apheresis than patients in the P+G group after 4 days of apheresis. A+G rescue was successful in 33/52 rescue patients in the P+G group and 4/10 rescue patients in the A+G group. 135 patients (90%) in A+G group and 82 patients (55%) in P+G group underwent transplantation. Median time to engraftment was Day 10 for PMN and Day 20 for platelets in both groups. Through 100 days, grafts were durable in all but 2 A+G (133/135) patients and in all P+G (82) patients. A+G patients experienced a higher incidence of GI effects (mild to moderate) and injection site erythema than P+G patients. There were 2 drug related serious adverse events in the A+G group and 1 in the P+G group. Conclusions: In this study, the addition of AMD3100 to G-CSF is generally safe and well tolerated and is superior to G-CSF alone for aHSC mobilization in NHL patients. A+G patients were statistically significantly more likely to achieve target cell collection quicker and to achieve sufficient cells for successful transplant than P+G patients.

Abstract Introduction Midostaurin is an orally available multikinase inhibitor that blocks FLT3 kinase activity. FLT3 mutations are associated with more frequent and earlier relapses and worse survival. The phase 3 RATIFY trial showed that midostaurin compared with placebo improved overall and event-free survival in newly diagnosed pts with FLT3-mutated AML when administered in combination with standard chemotherapy and as single-agent maintenance (Stone et al, ASH 2015). The ongoing RADIUS trial (NCT01883362) is investigating whether adding midostaurin to SOC after alloHSCT reduces the risk of relapse in pts with FLT3-ITD-mutated AML. Here we report safety data from the first 56 enrolled pts. Methods RADIUS is a randomized, open-label, phase 2 study comparing SOC vs midostaurin + SOC after alloHSCT in adult pts with AML with FLT3-ITD mutations (planned enrollment, N = 60). Study treatment started 28-60 days after alloHSCT. SOC was dictated by the treating physician. Pts were randomized to either SOC or midostaurin 50 mg twice daily + SOC (hereafter called the midostaurin arm) continuously for ≤ 12 months and will be followed up for ≥ 24 months. The study was designed to look for any safety or efficacy signals and not powered to find differences between study treatments. The primary endpoint is relapse-free survival at 18 months after alloHSCT. Adverse events (AEs) were followed up for 30 days after treatment. Key inclusion criteria are documented FLT3-ITD mutation, age 18-70 years, and first complete remission status. Pts could enroll after the date of engraftment and hematologic recovery to an absolute neutrophil count > 1000/μL and platelet count ≥ 20,000/μL without requiring transfusion. Results Pts were randomized from Feb 5, 2014, to Jun 13, 2016. The 2 arms (n = 28 each) were balanced regarding age, sex, and race. Most pts (93%) had de novo AML. At data cutoff (Jun 3, 2016), data were not mature enough to evaluate efficacy. Median (range) follow-up in the SOC and midostaurin arms was 240 (3-786) and 234 (3-656) days, respectively. Overall, 18 pts (64%) in the SOC arm and 19 pts (68%) in the midostaurin arm stopped treatment. Of these, 10 (36%) and 8 (29%) in the SOC and midostaurin arms, respectively, completed 12 months of treatment. Other reasons for stopping treatment were relapse (2 [7%] and 2 [7%]), death (2 [7%] and 0 [0%]), administrative problems (2 [7%] and 1 [4%]), withdrawn consent (1 [4%] and 3 [11%]), abnormal test results (1 [4%] and 0 [0%]), and AEs (0 [0%] and 4 [14%]). In the 24 pts who received ≥ 1 dose in the midostaurin arm, the median midostaurin dose was 76.2 (range, 25-100) mg daily. In the 15 pts (54%) who required a dose change, the reasons for dose changes were AEs (13 [46%]), dosing error (3 [11%]), re-escalation (3 [11%]), abnormal test results (2 [7%]), per protocol (2 [7%]), use of concomitant strong CYP3A4 inhibitors (1 [4%]), and other reasons (4 [14%]). The most common any-grade (Gr) AEs were fatigue (29%) and AST, headache, nausea and vomiting (all 25%) in the SOC arm and nausea and vomiting (both 64%) and diarrhea (43%) in the midostaurin arm (Figure 1). The most common Gr 3/4 AEs were nausea and hypertension in the SOC arm (all 3 [11%] each) and diarrhea, increased ALT, neutrophil count decreased (all 3 [11%] each) and platelet count decreased (5 [18%]) in the midostaurin arm. No on-treatment deaths occurred in the midostaurin arm. AEs led to discontinuation of midostaurin in 4 pts and included Gr 1 nausea, Gr 2 nausea and vomiting, Gr 3 lung infection, and Gr 2 elevated liver enzymes (n = 1 each). All of these except the lung infection were considered related to midostaurin. Graft-vs-host disease (GVHD) occurred in 16 pts (57%) in the SOC arm and 18 pts (64%) in the midostaurin arm (Figure 2). No stage ≥ 3 organ involvement occurred. Most cases of GVHD (11 [39%] in the SOC arm and 17 [61%] in the midostaurin arm) were acute. The most commonly affected organ was the skin (11 [39%] and 13 [46%], respectively, of which 5 [18%] and 4 [14%], respectively, were stage 2). Conclusions The preliminary safety data in the post-alloHSCT setting was consistent with data from other studies. Rates of Gr 1/2 nausea, vomiting, and diarrhea were higher in the midostaurin arm. Midostaurin did not change rates of GVHD. Because of the limited duration of follow-up, effects on chronic GVHD are unknown. Follow-up is ongoing, with efficacy data anticipated in 2017. Disclosures Maziarz: Athersys: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Incyte: Membership on an entity's Board of Directors or advisory committees; Novartis Pharmaceuticals Corporation: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding. Scott:Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Novartis: Consultancy, Research Funding, Speakers Bureau; Alexion: Speakers Bureau; Agios: Membership on an entity's Board of Directors or advisory committees. Deol:Jazz Pharmaceuticals: Consultancy. Haines:Novartis: Employment. Bonifacio:Novartis: Employment. Rine:Novartis: Employment. Fernandez:Fate Pharmaceuticals: Honoraria; Chimerix: Honoraria; Sanofi: Speakers Bureau.

Abstract Background: Chronic GVHD contributes to morbidity and mortality after allogeneic transplantation. Several phase 2 and 3 open label studies suggested that ATLG (formerly referred to as ATG-F) reduces cGVHD without impacting relapse or survival. This report describes the first prospective randomized double blind phase 3 trial to assess the effect of ATLG (Neovii) when added to tacrolimus / methotrexate prophylaxis on cGVHD free survival. Patients and Methods: This study was conducted at 27 sites in the United States and Australia. 260 patients were randomized (132 placebo, 128 ATLG). 6 patients who were randomized never received ATLG/placebo (4 placebo and 2 ATLG). Data presented include the 128 placebo and 126 ATLG patients who underwent treatment and transplant. Patients were ages 18-65 years with a diagnosis of acute myeloid (AML, 64%) or lymphoblastic (ALL, 21%) leukemia in first or subsequent complete remission or myelodysplastic syndrome (MDS, 15%) with less than 10% marrow blasts. Median age was 48 years (18-65) with 55% males. All patients received 8/8 HLA (A, B, C, DR) allele matched unrelated products (80% mobilized peripheral blood, 20% bone marrow). Myeloablative conditioning was with cyclophosphamide and total body irradiation (Cy/TBI, 27%), busulfan and cyclophosphamide (Bu/Cy, 33%), or busulfan and fludarabine (Bu/Flu, 40%). All patients received tacrolimus starting on Day -1 and standard mtx on days 1, 3, 6, and 11 for GVHD prophylaxis. Patients received ATLG or placebo on Days -3, -2, -1 at an ATLG dose of 20 mg/kg per day (60 mg/kg total). The primary endpoint of the study was moderate/severe chronic GVHD free survival. cGVHD was scored by the investigator and then confirmed or overturned by an independent adjudication committee. Secondary endpoints included engraftment, acute GVHD , moderate/severe cGVHD, non-relapse mortality, GVHD and relapse free survival (GRFS), progression free (PFS) and overall survival (OS). The impact of ATLG on immune reconstitution was assessed in 91 patients. Results: Treatment arms were balanced with respect to age, diagnosis, remission status, cytogenetics, graft source (PB or BM), CMV serostatus and conditioning regimen. Median follow up of survivors was 745 days (61, 1425). Regarding the primary endpoint, there was no difference in 2 year moderate/severe cGVHD free survival between ATLG and placebo (48 vs 44%, p = 0.57), nor was there a difference in GRFS (Table 1). Neutrophil engraftment at Day 30 was less in the ATLG arm (85% vs 95%, p=0.00001). Grade 2 or higher infusion reactions were higher in the ATLG arm (16% vs 2.3%, p=0.0001). Reactivation of CMV was similar (27% vs 31%, p=0.58). The Day 180 cumulative incidence of grades 2-4 acute GVHD was lower in the ATLG arm (23 vs 40%, p = 0.003) as was the 2-year cumulative incidence of moderate/severe cGVHD (12 vs 33%, p = 0.000006). However, both 2-year OS and PFS were lower in ATLG treated patients (58 v 76% and 47 vs 67%, respectively) due in part to a higher incidence of relapse (32 vs 19%, p=0.068). Multivariable models confirmed that ATLG was associated with inferior OS, HR 1.85 (1.1-2.9, p=.0075) and PFS, HR 1.63 (1.1-2.41, p=0.015). We then conducted an unplanned post-hoc analysis and discovered a striking influence of conditioning regimen on outcomes. There were no differences in OS or PFS between ATLG and placebo arms in patients receiving Bu/Cy or Bu/Flu but a dramatic difference in patients receiving Cy/TBI conditioning (OS, 88% vs 48% p =0.006 and PFS, 75 vs 29%, p=0.007 in placebo and ATLG arms, respectively). 91 patients (44 ATLG, 47 placebo) participated in the immune reconstitution study. ATLG treatment was associated with lower CD3, CD4, and CD8 counts at Days 30, 100, and 360, respectively. In turn lower CD3, CD4, and CD8 counts assessed as time dependent variables were all associated with inferior OS and PFS as well as higher NRM (data not shown). Conclusion: In this first ever prospective, randomized, double blind trial of ATLG added to tacrolimus /methotrexate in recipients of matched unrelated donor grafts after myeloablative conditioning, there was no significant difference in 2 year moderate/severe cGVHD free survival or GRFS. ATLG did significantly reduce grades 2-4 acute and moderate/severe chronic GVHD. However, overall survival was higher in the placebo arm, driven mainly by patients who received Cy/TBI conditioning. Further analyses are needed to understand the proper role for ATLG in HCT. Disclosures Soiffer: Juno: Membership on an entity's Board of Directors or advisory committees; Kiadis: Membership on an entity's Board of Directors or advisory committees; Jazz Pharmaceuticals: Membership on an entity's Board of Directors or advisory committees. Porter:Genentech: Employment; Novartis: Patents & Royalties, Research Funding. Jagasia:Therakos: Consultancy. Szer:Alnylam: Honoraria, Membership on an entity's Board of Directors or advisory committees; Ra Pharma: Honoraria, Membership on an entity's Board of Directors or advisory committees; Alexion Pharmaceuticals, Inc.: Honoraria, Membership on an entity's Board of Directors or advisory committees. Ritz:Kiadis: Membership on an entity's Board of Directors or advisory committees. Glavin:Neovii Biotech: Employment. Chen:Incyte Corporation: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Research Funding.

Abstract Background: AMD3100 (plerixafor)+G-CSF (A+G) was generally safe and more effective in mobilizing CD34+ hematopoietic stem cells compared to placebo+G-CSF (P+G) in a Phase III, multicenter, randomized, double-blind, placebo controlled study conducted in patients with non-Hodgkin’s lymphoma (NHL). In the intent-to-treat analysis, 78/148 (53%) patients in the P+G group and 20/150 (13%) patients in the A+G group failed to mobilize ≥ 2 x 106 CD34+ cells/kg, p<0.0001. Of these failure patients, 52 in the P+G group and 10 in the A+G group consented to enter into the rescue arm of the study. This abstract reports the outcomes of these rescue patients. Methods: Patients enrolled in the Phase III study who failed to mobilize ≥2 x 106 CD34+ cells/kg could enter into the rescue arm of the study and receive A+G. The patient’s treatment assignment in the study remained blinded. Following a rest period of ≥7 days, patients were given G (10 μg/kg/day) subcutaneously (SQ) for 4 days. In the evening around 10PM on Day 4, patients were given a dose of A (240 μg/kg SQ). On Day 5, all patients returned to clinic to receive a morning dose of G before apheresis. Patients continued to receive the evening dose of A followed by morning dose of G and apheresis the next day for up to a total of 4 aphereses or until ≥5 x 106 CD34+ cells/kg were collected. Pooled cells were allowed for transplantation. Results: After rescue therapy, 33/52 (63%) patients who previously failed P+G and 4/10 (40%) patients who previously failed A+G collected ≥2 x 106 CD34+ cells/kg. Seven of the 33 patients who previously failed with P+G collected ≥ 5 x 106 CD34+ cells/kg. Six of the ten A+G patients and 46/52 P+G patients were transplanted. Median time to PMN engraftment was Day 10 and Day 11 for the A+G and P+G groups, respectively. Median time to platelet engraftment was Day 22 and Day 20 for the A+G and P+G groups, respectively. One patient in the P+G group never had platelet counts > 50,000 and was counted as a failure to engraft, even though the patient was clinically stable through 6 months post-transplant. In the re-mobilized group, 9 patients died (2 patients were never transplanted and 7 patients had disease progression). Grafts were durable for ≥100 days in all 6 rescued A+G patients and all 41 rescued P+G patients who engrafted. The predominant adverse events were mild gastrointestinal effects (diarrhea, nausea, vomiting, and abdominal pain) and injection site reactions. There were no drug-related serious adverse events reported in these re-mobilized patients. Conclusions: The addition of AMD3100 to G-CSF successfully mobilized sufficient stem cells for transplant in 63% of NHL patients who previously failed mobilization with G-CSF alone. The high success rate is consistent with the Compassionate Use Protocol experience. AMD3100 was generally well tolerated. The 40% success rate in the patients who previously failed AMD3100+G-CSF suggests that re-mobilization with the same combination is a potentially useful strategy for similar patients.