Search

Background: Vosaroxin, is a first-in-class anti-cancer quinolone derived (AQD) DNA topoisomerase II inhibitor, which is not a substrate for p53 or P-glycoprotein, has limited toxicity, and is currently under evaluation for the treatment of pts with AML and high-risk MDS. Methods: Pts were eligible if they had AML or high-risk MDS (defined as having ³ 10% blasts in the bone marrow), were 60 years of age or older, and had adequate performance status (ECOG ² 2) and organ function. Pts younger than 60 who were unsuitable for standard chemotherapy were also eligible. In the phase I part of the study the first six pts received vosaroxin 90 mg/m2 daily on days 1 and 4 with decitabine 20 mg/m2 daily for 5 days. This dose was well tolerated with no dose limiting toxicity identified and pts were then enrolled on phase II at this dose and schedule. However, due to mucositis in a few subsequent pts, the induction dose of vosaroxin was reduced to 70 mg /m2, starting with pt #25. The vosaroxin dose could be maintained at 70 mg/m2 or reduced to 50 mg/m2 in consolidation cycles, which were repeated in approximately 4 to 5-week intervals for a total of up to 7 cycles. Dose adjustments and dose delays of one or both agents, were allowed based on toxicity. The primary endpoint was to determine the overall response rate including complete response (CR) + CR without platelet recovery (CRp) + CR with insufficient hematological recovery (CRi). Secondary endpoints were: CR duration, disease-free survival, overall survival, safety, and early mortality. Results: To date, 35 pts (32 AML, 3 high-risk MDS) with a median age of 71 years (range, 41-78) have been enrolled; 34 (97%) pts were older than 60 years. They included 15 (43%) pts with diploid cytogenetics, 12 (34%) with complex cytogenetic abnormalities including chromosome 5 and/or 7 abnormalities, and 8 (23%) with other miscellaneous abnormalities. 13 (37%) pts with AML had antecedent hematological disorders (AHD) including 7 (20%) with MDS, 4 (11%) with myeloproliferative neoplasm and 2 (6%) with MDS/MPN. Three pts with AHD had received prior therapy including 5-azacytidine (n=1), ruxolitinib + 5-azacytidine (n=1), and lenalidomide (n=1). Additionally, 5 (15%) pts had therapy-related disease with prior exposure to chemotherapy or radiation therapy. Median bone marrow blast %, median white blood cell, hemoglobin, & platelet counts were 40% (9-97), 4.1 x 109/L (0.4-57.0), 9.4 g/dL (6.8-11.5), and 40 x 109/L (7-333), respectively. 34 pts were evaluable for response; 17 (50%) achieved CR, 6 (18%) CRp, and 3 (9%) CRi for an overall response rate of 77%. One pt is too early for response assessment. Responses by age, cytogenetic and molecular characteristics are shown in table 1. The median follow-up is 5.1 months (range, 0.9-11.0). Pts have received a median of 2 (1-6) treatment cycles with the median number of cycles to response being 1 (1-4). Three pts have relapsed and the median duration of CR/CRp/CRi has not been reached (0.5-9.9+ months). Four (12%) pts have proceeded to allogeneic stem cell transplant (ASCT). 4-week and 8-week mortality were 0% and 14%, respectively. The regimen is well tolerated with the main grade ³ 3 toxicity being mucositis in 9 (26%) pts and liver enzyme elevation in 3 (9%). Conclusion: Combination of vosaroxin and decitabine is effective and feasible in older pts with AML and high-risk MDS. Enrollment is ongoing. Table 1: Response by baseline characteristics Parameter Category N Overall response Too early to evaluate Age* 60-70 16 14/16 (88%) 0 >70 18 12/17 (71%) 1 Cytogenetics Diploid 15 11/14 (79%) 1 -5/-7/other adverse 12 8/12 (67%) 0 Miscellaneous 8 5/8 (63%) 0 Mutation Status IDH2 7 7/7 (100%) 0 IDH1 5 1/4 (25%) 1 TP53 9 6/9 (67%) 0 RAS 9 4/9 (44%) 0 *1 pt below the age of 60 years was unsuitable for standard chemotherapy and was enrolled on study. Disclosures Naval: Sunesis: Advisory board membership Other, Research Funding. Off Label Use: Vosaroxin (Qinprezo) has not been approved for AML. In this clinical trial we are evaluating the efficacy and safety of vosaroxin (Qinprezo) in combination with decitabine in elderly patients with AML. . Kantarjian:ARIAD, Pfizer, Amgen: Research Funding. Kadia:GSK: Research Funding; ARIAD: Honoraria. Borthakur:Tetralogic Pharmaceuticals: Research Funding. Jabbour:Ariad, Novartis, BMS, Pfizer, and Teva: Consultancy. Cortes:Ariad: Consultancy, Research Funding; BMS: Consultancy, Research Funding; Novartis: Consultancy, Research Funding; Pfizer: Consultancy, Research Funding; Teva: Consultancy, Research Funding. Craig:Sunesis: Employment. Ravandi:Sunesis: Advisory board membership Other, Research Funding.

The prognosis of patients with untreated MDS or AML with poor performance, organ dysfunction, or other comorbidities is very poor. Most patients do not receive therapy, and a large majority of them are excluded from investigational clinical trials. We hypothesized that this group of patients would significantly benefit from participation in clinical trials. To test this concept, we performed an initial proof-of-principle study of 30 patients with these characteristics using a combination of azacitidine and vorinostat, a histone deacetylase inhibitor, with the stopping rules being survival at 60 days, a response rate (30%), and toxicity. This study indicated that treatment was associated with longer-than-expected survival and response rate and acceptable toxicity as well as demonstrating that this group of patients can be safely and efficiently enrolled in clinical trials (Garcia-Manero ASH 2010 abstract #604). To further verify these results, we subsequently expanded this trial as a randomized study comparing azacitidine versus azacitidine + vorinostat, the results of which are shown here. This study would allow further experience in the development of clinical trials for this at-risk population as well as explore the role of vorinostat in combination with azacitidine. The primary objective was survival at 60 days. The study was independently monitored by the Dept of Biostatistics at our institution. Patients were randomized using an adaptive procedure. The first 40 patients were randomized in a 1:1 ratio, and after that will be unbalanced in favor of the superior arm as efficacy (survival at 60 days) data accrue. Inclusion criteria included patients with newly diagnosed AML or higher-risk MDS who were not eligible for high-dose chemotherapy or clinical trials due to comorbidities, organ dysfunction, or poor performance status. Treatment consisted of azacitidine 75 mg/m2 IV daily x 5 (A arm) or azacitidine at the same dose schedule with vorinostat at 200 mg 3 times a day on days 1 to 5 (A+V arm). From September 2011 to March 2014, 79 patients were enrolled: 27 (34%) in the A arm and 52 (66%) in the A+V arm. Median age was 70 (30-90); median bone marrow blasts were 8% (1-89), median WBC 2.8 (0.2-102.0), median peripheral blasts 0% (0-78), poor cytogenetic risk in 41 (52%), diploidy in 20 (25%), and intermediate risk in 18 (22%). Reasons for inclusion were concurrence or previous history of other malignancies (36; 46%); ≥2 ECOG performance status (9; 11%); comorbidities including lung fibrosis, renal or liver dysfunction, or HIV positive status (17; 21.5%); or non-eligibility for others trials due to higher priority (17; 21.5%). The number of median cycles was 1 (1-12) for the A arm and 3 (1-12) for A+V arm. Study drugs were tolerated with no drug-related early stoppage (8 week mortality). Incidence of any grade 3-4 toxicity was less than 9%. Sixty-day survival was seen in 18 patients (67%) for the A arm and 44 patients (85%) for the A+V. The A+V arm was associated with a better outcome (OR 0.311, 95% CI [0.101-0.962]; p=0.043). The ORR (CR+CRp) was 12 (44%) for the A arm and 18 (35%) for the A+V arm (p=0.395). The median number of cycles for response was 1 in the A arm and 1.5 in the A+V arm. Survival was confounded by death unrelated to leukemia. The median overall (OS) and relapse-free survival (RFS) for the whole group were 7.1 months (5.5-8.7) and 5.9 months (3.1-8.7), respectively. Median OS for the A and A+V arms was 6.1 and 5.9 months, respectively (p=0.9), and median RFS for the A and A+V arms was 5.9 and 8.7 months (p=0.5), respectively. Patients who achieved CR/CRp had better OS: median 12 vs 6 months (HR: 0.41, [0.22-0.78] p=0.007). After a median follow-up of 9.5 months, 23 patients (29%) are alive at last follow-up. In conclusion, azacitidine+vorinostat was associated with a better 60-day survival than single-agent azacitidine in a patient population that is generally excluded from clinical trials. This study confirms the possibility of treating patients that are generally not considered eligible for clinical trials. Disclosures No relevant conflicts of interest to declare.

Background: Therapy with HMA is now the standard of care for pts with MDS and chronic myelomonocytic leukemia (CMML) with complete response (CR) rates of 7% to 35%, median response durations of 9 to 10 months, and median survival of 20 to 24 months. While allogeneic stem cell transplantation (ASCT) is curative in pts with MDS, the long-term outcome of pts treated with HMA remains unknown. Aims: The aims of the study are to assess the long-term outcome of pts with MDS treated with HMA and to identify prognostic factors of long-term outcome. This may help selecting patients for this long-term treatment in whom ASCT may not be indicated. Methods: We reviewed the records of 511 pts with diagnosed MDS (n=409) and CMML (n=102) treated from 4/2000 to 4/2014 and who were treated with HMA. Pts who received ASCT (n=65) were excluded. Thus, a total of 446 pts were evaluable for the study. The probabilities of leukemia-free survival (LFS) and overall survival (OS) were estimated using the method of Kaplan and Meier. Univariate and multivariate analyses were performed to identify potential factors associated with the achievement of response with logistic regression models and survival with Cox proportional hazard regression models. Results: The median follow-up for the entire cohort was 13.6 months. Pt characteristics are outcomes described in Table 1. Best responses to HMA were CR in 124 (28%) pts, CRp in 27 (6%), PR in 9 (2%), HI in 31 (7%). Median duration of response was 7 months (1-68). 130 (29%) transformed into AML after a median of 11 months (11-60). At the last follow-up 133 (30%) remained alive. The median LFS and OS were 16.1 and 16.2 months respectively. The 2- and 5-year LFS and OS rates were 29% and 11% and 34% and 12%, respectively. By multivariate analysis, baseline characteristics associated with OS included WBC (>4 vs. ≤4), ferritin (>500 vs. ≤500), hemoglobin (>10 vs. ≤10), platelets (>500 vs. ≤500) and cytogenetics (high vs. intermediate vs. low risk) (p4 vs. ≤4), ferritin (>500 vs. ≤500), platelets (>500 vs. ≤500) and cytogenetics (0=low risk; 1=intermediate risk; and 2=high risk) as independently associated with survival improvement Patients with 0-2 (n=244) or 3-5 (n=162) adverse factors had a median survival of 19 and 13 months, respectively (p Conclusion: Our current analyses identified a small subset of pts with MDS in whom outcome of therapy with HMA is excellent and can be differentially predicted. Table 1. Patient Characteristics and Outcomes Parameter (N=446) Number (%); Median [range] Age (years) 70 (13-92) White Blood Cell Count (x 109/L) 3.5 (0.5-212) Ferritin 465.0 (0-10971) Hemoglobin (g/dL) 9.7 (6-16) Platelets (x 109/L) 68.0 (4-987) Bone marrow blasts (%) 6.0 (0-19) Prior malignancy 198 (44) Prior chemotherapy 133 (30) Prior radiotherapy 85 (19) Prior Transfusion 134 (30) Cytogenetics (by IPSS) Low 213 (48) Intermediate 78 (17.5) High 144 (8) Missing 11 (2.5) WHO RA 47 (10.5) RARS 16 (4) RCMD 75 (17) RAEB 204 (46) MDS-U 8 (2) CMML 95 (21) Missing 1 (0.2) IPSS Low 46 (10) Intermediate-1 193 (43) Intermediate-2 156 (35) High 36 (8) Missing 15 (3) MDA Score Low 59 (13) Intermediate-1 113 (25) Intermediate-2 124 (28) High 113 (25) Missing 37 (8) Type of HMA Azacitidine/ AZA+ 189 (42) Decitabine/DAC 257 (58) Response to HMA CR 124 ( 28) CRp 27 (6) PR 9 (2) HI 31 (7) NR 121 (27) Died on therapy 23 (5) NE 6 (1.4) Missing 105 (23.5) Median duration of response (mos) 7.2 (1-68) Transformed into AML 130 (29) Dead 313 (70) Disclosures No relevant conflicts of interest to declare.

Patients with acute myeloid leukemia (AML) experience sustained and profound neutropenia during induction chemotherapy; infections remain the leading cause of morbidity and mortality in myelosuppressed leukemia patients. The current standard of care is to provide patients with broad-spectrum antibiotic, antifungal and antiviral therapy; however resistance to available anti-infective agents is increasing. Transfusion of functional non-irradiated allogeneic granulocytes to neutropenic leukemia patients may treat, delay or prevent infections in patients with leukemia, and importantly may also have anti-leukemia benefits. Patients treated at MD Anderson Cancer Center with a diagnosis of AML undergoing front-line AML therapy were eligible. Patients were required to be free of signs and symptoms of infection at the time of study entry, and have sufficient volunteer donors to administer prophylactic white cell transfusions approximately twice a week for six weeks. Allogeneic white cell transfusions (≥ 4 x 10¹⁰ cells per transfusion) were administered every 3-4 days during the induction chemotherapy cycle, so long as patients remained neutropenic with an absolute neutrophil count < 500. Prophylactic transfusions continued until the time of sustained ANC recovery, initiation of a new treatment regimen, or after the completion of 6 weeks on protocol. Herein we provide the results of the first 21 patients. Median age was 67 (range 23-78); 12 patients were male and 9 were female. All patients were enrolled at the time of initiation of AML induction therapy. Seven patients had a diagnosis of therapy-related AML (6 with an antecedent hematologic disorder and 1 with prior solid malignancy). Various treatment regimens were administered and are displayed in Table 1, along with cytogenetic and MD Anderson prognostic model classification. Patients received a median of 5 transfusions (range 0 – 11). Of 21 patients, 1 withdrew consent prior to the first transfusion, and 3 patients discontinued transfusions prior to the end of study (due to a diagnosis of pneumonia in one with potential concern for increased pulmonary toxicity with continued transfusions, and two at patient requests due to fevers and myalgias post transfusions). Median time to ANC recovery was 38 days (range 14 – 137) and platelet recovery 29 days (range 12 – 160), consistent with the elderly population and the allotted treatment regimens. All patients experienced at least one neutropenic fever, often within 48 hours of white cell administration. Documented infections included 2 instances of bacteremia (1 streptococcal and 1 E. coli), 2 urinary tract infections (both coagulase negative staphylococci), 8 cases of pneumonia (1 documented E. coli, 7 culture-negative and presumed fungal pneumonia per imaging). There was one patient with neutropenic colitis per imaging. There were no induction-related deaths. 8-week mortality was 5% (n=1) and 12-week mortality was 20% (n=4). With a median follow-up time on study of 20.3 weeks, overall survival is 67% with a median survival which has not yet been reached (Figure 1). Overall response to initial induction chemotherapy was 67% (9 CR, 1 CRi, 4 PR). A total of 15 patients (71%) ultimately attained a CR (after the first cycle or with additional cycles of therapy), and 4 patients received a stem cell transplant at the time of CR/CRi. In conclusion, prophylactic transfusion of allogeneic non-irradiated white blood cells for newly diagnosed AML patients during induction chemotherapy was associated with a decreased incidence of life-threatening infections, decreased induction mortality (6-week OS of 100%), and may also lead to improved remission rates and overall survival. This feasibility study has been expanded to enroll a total of 50 patients, with updated results to be provided at the annual meeting. Figure 1 Figure 1. Figure 2 Figure 2. Disclosures No relevant conflicts of interest to declare.

The RUNX family of transcription factors forms the DNA binding α-chain partners of the heterodimeric core binding factor (CBF) complex. Each of the RUNX proteins, RUNX1, RUNX2, and RUNX3, can form heterodimers with CBFβ. In the M4Eo subtype of human acute leukemia, the chromosomal translocation resulting in inversion 16 encodes a chimeric protein in which CBFβ is fused to smooth muscle myosin heavy chain (SMMHC). Although the exact mechanism of leukemogenesis by this chimera is unknown, it is thought that CBFβ-SMMHC sequesters RUNX1 in the cytoplasm and antagonizes its normal function. Although the role of RUNX1 in hematopoiesis has been previously well-established, recent data have indicated that the RUNX3 gene may also play a key role in the development of human acute leukemias. To clarify the role of RUNX3 in acute myeloid leukemia (AML), we investigated its expression and promoter DNA methylation in leukemia cell lines and patient samples. Eleven human leukemia cell lines of myeloid origin and twelve of lymphoid origin were used in this study. Cell suspensions from bone marrow aspirate specimens from patients with AML (69 cases), MDS (19 cases) and ALL (6 cases) were obtained prior to therapy from established tissue blocks. Peripheral blood samples were obtained from four healthy volunteers, and CD34+ cells were obtained from another four individuals. Methylation status of the gene promoters of RUNX1, RUNX2 and RUNX3 were evaluated using the Pyrosequencing Methylation Assay (PMA) method, and expression of RUNX3 was analyzed by quantitative real-time PCR and immunohistochemical staining. Hypermethylation of RUNX1 and RUNX2 was rare in cell lines; RUNX1 was not hypermethylated in any of the studied samples, and RUNX2 was hypermethylated in only two cell lines. In contrast, we found that the RUNX3 promoter was hypermethylated in 17 of the cell lines (74%). Interestingly, we observed a trend toward higher frequency of hypermethylation of RUNX3 in cell lines of myeloid (90%) compared to lymphoid (57%) origin. In patient samples, RUNX3 promoter methylation was below 15% in normal samples, and hypermethylation was found in 32/69 AML samples (46%), 4/19 MDS samples (21%), and 6/6 ALL samples (100%). Of the 69 AML samples, 19 were classified as AML M4Eo, and 50 were other types of AML. 84% of the human AML M4Eo samples were hypermethylated at the RUNX3 promoter region, whereas only 34% of the other AML subtypes were hypermethylated. We also evaluated DNA methylation of RUNX1 and RUNX2 in a subgroup of these samples (66 samples for RUNX1 and 72 for RUNX2) and found that, as in cell lines, these genes are almost universally unmethylated; with the exception of a single AML case, all studied samples presented no promoter methylation. As support of functional outcome, hypermethylation of RUNX3 was correlated with both lower levels of mRNA and protein, as confirmed by qRT-PCR and immunohistochemistry analysis in cell lines and patient samples, and treatment with the DNA demethylating agent Decitabine resulted in mRNA re-expression of RUNX3 concomitantly with decreased promoter methylation. Finally, we compared clinicopathological features of patients with and without RUNX3 methylation. In this analysis, only non-M4Eo AML cases were compared because of the small number of non-methylated patients in the M4Eo group. Differences were found neither for blood counts nor for overall survival probability. However, relapse-free survival was significantly better for the unmethylated group (p=0.016). In summary, we showed that promoter methylation of the RUNX3 gene and down regulation of RUNX3 expression occurs almost universally in M4Eo/inversion 16 AMLs, and that in cell lines, RUNX3 repression can be reversed by treatment with the hypomethylating agent decitabine. These results suggest that silencing of RUNX3 is likely an important target in CBF leukemia and that future studies should be dedicated to further characterize the role of RUNX3 in inversion 16 AML and its predictive value of relapse-free survival in AML. Disclosures No relevant conflicts of interest to declare.

Background: Allogeneic stem cell transplant (ASCT) is a curative option for patients with higher risk myelodysplastic syndromes (MDS). Hypomethylating agents (HMA) have been shown to improve survival of patients with MDS and have an excellent toxicity profile. In eligible patients, HMA and ASCT are used as complementary strategies. The aim of this study is to compare outcome with HMA alone vs HMA+ASCT where ASCT is used as consolidation approach or as salvage therapy after HMA failure. Methods: We performed a retrospective analysis of 216 patients with high-risk MDS who received HMA treatment at our institution from between April 2004 and October 2012; after HMA therapy, 61 (28%) patients underwent ASCT: 25 (41%) of them received as a consolidation treatment and 36 (59%) as a salvage therapy. The remaining 155 patients continued on HMA therapy until relapse or progression and did not receive a further ASCT. We used SPSS v.20 for all statistical analysis. Categorical and continuous variables were compared by chi-square and Student’s t test, respectively, and survival analysis was conducted using Kaplan-Meier analysis with the log-rank test and Cox regression for multivariate models. We also implemented a landmark survival analysis that considered median time to transplant. Results: Median age was 65 (20-89), and patients were older in the group that did not receive ASCT (69 vs 58 years; p=0.000). WHO diagnoses were RA/RARS in 31 patients (15%), RCMD in 43 patients (19.9%), RAEB in 133 patients (62%), CMML in 23 patients (11%), and unclassifiable MDS in 18 patients (8.3%). IPSS risk was int-2 in 107 patients (50%) and high in 61 (28%), and the percentage of inttermediate-2/high-risk MDS was higher in the ASCT group (89% vs 73%; p=0.001). High-risk cytogenetics were found in 71% of patients (82% in ASCT vs 67% in HMA alone; p=0.02). Patients received a median of 6 (1-58) courses of HMA. Overall response rate (ORR) to HMA was 45% (n=97), with 38% (n=82) having complete response (CR), 2% (n=5) partial response, and 5% (n=10) hematologic improvement. There were no significant differences between response to HMA in the ASCT group when compared to the group that did not receive ASCT (ORR: 46% vs 43%, p=0.3; CR: 38% vs 37%, p= 0.5). When we further analyzed the 61 ASCT patients, 25 (41%) received it as a consolidation after achieving response and 36 (59%) as a salvage therapy after treatment failure. Response to ASCT was CR in 65% of patients, and 20% were not evaluable due to early mortality. Median overall survival (OS) for the whole series was 14 months (12-16), with 1- and 2-year OS rates of 57% and 24%, respectively. To adjust for early mortality after ASCT and to eliminate any bias, we performed a landmark analysis after a median time of 7 months after ASCT. Patients who received ASCT had better survival. This advantage was more evident among patients who received ASCT as salvage therapy, although there were no differences between both strategies, with respective median survivals of 14 months for consolidation ASCT and 23 months for salvage ASCT (p=0.29)Furthermore, no significant differences in survival were observed between patients who received HMA alone and those who received ASCT as a consolidation therapy (median survival of 14 and 16 months; p=0.498), although there was a tendency for a better OS after 2 years of follow up: the OS for HMA treatment and HMA+ASCT consolidation were 68% and 56% at 1 year, 25% and 31% at 2 years, and 10% and 31% at 3 years. The 1-year survival rates for patients who received HMA alone, HMA followed by ASCT as consolidation, and HMA followed by ASCT as salvage were 68%, 56%, and 78%, respectively, and the 2-year survival rates were 25%, 31%, and 42%, respectively. In a Cox regression model to analyze effects on OS, receiving an ASCT (median OS of 13 (7-19) vs 10 months (8-12); HR 0.62 [0.42 – 0.92], p=0.018) and hemoglobin levels at diagnosis (HR 0.829 [0.73 – 0.93], p=0.002) had a significant impact. Conclusions: ASCT is a feasible and curative strategy for patients with MDS, both as a consolidation or a salvage therapy, and thus it can be a good option after HMA failure. However, its benefits as a consolidation therapy after HMA treatment compared with continuation of HMA treatment are not clear owing to early mortality related to procedure, so ASCT should be carefully considered in patients responding to HMA. Disclosures Borthakur: Tetralogic Pharmaceuticals: Research Funding. Cortes:Ariad: Consultancy, Research Funding; BMS: Consultancy, Research Funding; Novartis: Consultancy, Research Funding; Pfizer: Consultancy, Research Funding; Teva: Consultancy, Research Funding. Ravandi:Incyte Corporation: Research Funding. Kadia:GSK: Research Funding; ARIAD: Honoraria. Champlin:Otsuka: Research Funding. Kantarjian:ARIAD: Research Funding; Pfizer: Research Funding; Amgen: Research Funding.

Background: The outcome of patients with myelodysplastic syndromes (MDS) after failure of hypomethylating agents (HMAs) failure is poor with a median overall survival (OS) of only 4-6 months. Omacetaxine mepesuccinate (OM) is safe and effective in myeloid malignancies but has not been studied in MDS with HMA failure. Methods: We conducted a phase II study of OM in patients with MDS or chronic myelomonocytic leukemia (CMML) who had previously failed or been intolerant to HMAs. Patients received OM at a dose of 1.25 mg/m2 subcutaneously every 12 hours for 3 consecutive days on a 4-7 week schedule. The primary endpoints were the overall response rate (ORR) and OS. Findings: Forty-two patients were enrolled with a median age of 76 years. The ORR was 33%. Patients with diploid cytogenetics were more likely to respond to OM than were those with cytogenetic abnormalities (58% versus 23%, respectively; P=0.03). Overall, the median OS was 7.5 months and 1-year OS rate was 25%. Patients with diploid cytogenetics had superior OS to those with cytogenetic abnormalities (median OS 14.8 versus 6.8 months, respectively; P=0.01). Two patients had ongoing response to OM of 2 years or longer (both MDS with diploid cytogenetics and RUNX1 mutation). The most common grade ≥3 adverse events were infections in 11 patients (26%), febrile neutropenia in 4 (10%), and hemorrhage in 3 (7%). Interpretation: Overall, OM was safe and active in patients with MDS or CMML who experienced HMA failure. These results support the further development of OM in this setting. Clinical Trial Registry: The trial was registered at ClinicalTrials.gov with the identifier NCT02159872. Funding: This trial was funded by the MD Anderson Cancer Center Support Grant CA016672. Disclosure of Conflicts of Interest: The authors report no relevant conflicts of interest. Ethics Approval Statement: This study was approved by the Institutional Review Board of The University of Texas MD Anderson Cancer Center.

Background: Vosaroxin (formerly voreloxin), is a quinolone derived DNA topoisomerase II inhibitor, which is not a substrate for p53 or P-glycoprotein and has limited toxicity; it is currently under evaluation for the treatment of pts with AML and high-risk MDS. Methods: Pts were eligible if they had AML or high-risk MDS (defined as having ≥ 10% blasts in the bone marrow), were 60 years of age or older, and had adequate performance status (ECOG ≤ 2) and organ function. Pts younger than 60 who were unsuitable for standard chemotherapy were also eligible. The treatment regimen included vosaroxin 90 mg/m2 daily on days 1 and 4 with decitabine 20 mg/m2 daily for 5 days. Vosaroxin dose was reduced to 70 mg/m2 in consolidation cycles, which were repeated in approximately 4 to 5-week intervals for a total of up to 7 cycles. Dose adjustments and dose delays of one or both agents, were allowed based on toxicity. The primary endpoint was to determine the complete remission (CR) rate. Secondary endpoints were: CR duration, disease-free survival, overall survival, safety, and early mortality. Results: To date, 12 pts (9 AML, 3 high-risk MDS) with a median age of 71 years (range, 41-76) have been enrolled; 11 were older than 60 years. They included 6 (50%) pts with diploid cytogenetics, 4 (34%) with complex cytogenetic abnormalities including chromosome 5 and/or 7 abnormalities, and 2 (16%) with other miscellaneous abnormalities. 4 (34%) pts with AML had antecedent hematological disorders (AHD) including 1 (8%) with MDS, 1 (8%) with myeloproliferative neoplasm and 2 (17%) with MDS/MPN. Three pts with AHD had received prior therapy including 5-azacytidine (n=1), ruxolitinib + 5-azacytidine (n=1) and lenalidomide (n=1). RAS mutations were identified in 4 (34%) and IDH2 mutations in 2 (16%) pts. Median bone marrow blast %, median white blood cell, hemoglobin, & platelet counts were 29% (11-73), 3.7 x 109/L (1.0-127.1), 9.5 g/dL (8.2-10.8), and 92 x 109/L (11-361), respectively. 9 pts were evaluable for response; 5 (42%) achieved CR, 1 (8%) CRp, and 2 (17%) CRi, for an overall response rate of 89%. 1 pt had no response after cycle 1 and is currently undergoing re-induction. Three are too early for response assessment. Pts have received a median of 2 (1-3) treatment cycles with median number of cycles to response being 1 (1-2) and median time to achieving response, 26 days (21-74 days). The median duration of CR/CRp/CRi is 53+ days (6-89+ days). One pt has proceeded to allogeneic stem cell transplant. No pts died during induction. One pt died on study on day 47 from sepsis and acute myocardial infarction. The regimen is well tolerated with the main grade ≥ 3 toxicity being mucositis in 3 (25%) pts. Conclusion: Combination of vosaroxin and decitabine is effective and well tolerated in older patients with AML and high-risk MDS. Enrollment is ongoing. Citation Format: Naval Daver, Hagop M. Kantarjian, Guillermo Garcia - Manero, Naveen Pemmaraju, Tapan Kadia, Courtney DiNardo, Nitin Jain, Gautum Borthakur, Jorge Cortes, Craig Adam, Farhad Ravandi. Phase I/II study of vosaroxin and decitabine in older patients (pts) with acute myeloid leukemia (AML) and high risk myelodysplastic syndrome (MDS). [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr CT307. doi:10.1158/1538-7445.AM2014-CT307

7098 Background: Vosaroxin is a first-in-class anticancer quinolone derivative (AQD) that intercalates DNA and inhibits topoisomerase II and has previously demonstrated activity in AML. We conducte...

7097 Background: A majority of patients (pts) with AML age ≥ 60 are unable to tolerate intensive chemotherapy (IC) induction. Front-line therapy with clofarabine plus low dose cytarabine (CLDA) has...

6506 Background: Application of cognitive computing technology in cancer has potential to democratize expert knowledge and enable personalized evidenced based oncology care. Methods: We developed a web based prototype of MD Anderson’s Oncology Expert Advisor (OEA), a cognitive clinical decision support tool powered by IBM Watson. The Watson technology is IBM’s third generation cognitive computing system based on its unique capabilities in natural language processing and deep QA (question-answer). OEA is built with four core capabilities: 1) Patient Evaluation through interpretation of structured and unstructured clinical data to create a dynamic case summary with longitudinal view of the pertinent events 2) Treatment and Management suggestions based on patient profile weighed against consensus guidelines, relevant literature, and MD Anderson expertise, which will include approved therapies, genomic based therapies as well as automatic matching to eligibility criteria of clinical trials at MD Anderson, 3) ...

Background Epigenetic therapy with HMA is now the standard of care for pts with MDS with complete response (CR) rates of 7% to 35% and median survival of 20 to 24 months (mos). Large studies have demonstrated the relevance of karyotype abnormalities as an independent prognostic factor in predicting outcome in MDS pts. In pts with chronic myeloproliferative neoplasms (e.g. chronic myeloid leukemia), the achievement of CCyR has proven to correlate with improved outcomes: less transformation into acute myeloid leukemia and better overall survival. However this phenomenon has not been examined thus far in pts with MDS, where the response criteria are still predominantly based on morphologic basis. Aims To assess the impact of CCyR on outcome in pts with MDS treated with HMA Methods 216 consecutive pts with MDS and abnormal karyotype treated with HMA between 4/04 and 10/12 were reviewed. Response criteria were based on the International Working Group (IWG) 2006 criteria. CCyR was defined as achievement of a diploid karyotype among at least 20 metaphases analyzed. Results Median age was 66 years (21-90). IPSS was high-risk in 22%, intermediate-2 in 55%, intermediate-1 in 21%, and low-risk disease in 3%. 54% were secondary MDS; 43% were therapy-related MDS. Cytogenetic analysis revealed a complex karyotype in 24%, predominantly demonstrating abnormalities in chromosomes 5 or 7 (37%) and other abnormalities (39%). 69% were treated with decitabine based therapy (single agent in 47%) while 31% were treated with 5-azacitidine (single agent in 20%). Best responses were CR in 64 pts (33%), CR without platelet recovery (CRp) in 15 pts (8%), partial response (PR) in 4 pts (2%), hematologic improvement (HI) in 10 pts (5%), and stable disease (SD) in 5 pts (3%). Furthermore, CCyR was achieved in 68 pts (31%): 45/79 pts with CR/CRp (57%), 3/19 pts with PR/HI/SD (16%), and 20/118 pts with no morphologic response (17%) (p Conclusion Our results indicate that CCyR can occur at a rate of 31% in pts with MDS treated with HMA. Pts presenting with primary MDS and attaining a morphologic response to HMA are more likely to achieve a CCyR. Furthermore we identified a significant association/correlation between the achievement of CCyR and better OS and TFS, independent of attaining a morphologic response. Therefore, the achievement of CCyR should be considered a new milestone in the management of pts with MDS. Disclosures: No relevant conflicts of interest to declare.

The pathogenesis of MDS is multifactorial including both cell intrinsic alterations, such as mutations, and cell extrinsic stimuli such as immune deregulation. The prognosis of patients that lose response to hypomethylating agents (HMAs) is very poor and the mechanisms are not understood. Leukemia cells develop multiple resistance mechanisms to escape host immune response. Programmed death-1 (PD-1) is a negative costimulatory receptor on activated T lymphocytes. Expression of PD-1 ligands on tumor cells can induce immunosuppressive T-cells. Recent studies have indicated that demethylation of PD-1 leads to exhausted CD8+ T cells after chronic virus infection. We hypothesized that dysregulated PD-1 immune checkpoint signaling may be involved in pathogenesis of MDS and resistance to HMAs. We first evaluated the mRNA expression levels of PD-L1, PD-L2, PD-1 and CTLA4 by Q-PCR in bone marrow CD34+ cells from 124 patients which include 69 with MDS, 46 with CMML and 9 with AML. Seventy two (58%) patients were previously untreated. We observed aberrant up-regulated mRNA expression of PD-L1 in 39 patients (34% with fold 0-843.3), PD-L2 in 17 patients (14% with fold 0-22.5), PD-1 in 18 patients (15% with fold 0-76.7) and CTLA4 in 10 patients (8% with fold 0-25.1). No significant differences in gene expression were observed when comparing patients that had and had not received prior therapy. Statistically significant relationships were identified between elevated PD-1 expression and increased age (p=0.008), while elevated PD-L2 expression correlated significantly with female gender (p=0.005). Both elevated PD-L1 and CTLA4 expression correlated with MDS subtype, (p=0.034) and (p=0.012), respectively. Additionally, elevated CTLA4 expression correlated with higher white blood cell count (p=0.021), lack of prior therapy (p=0.02) and lower MDS IPSS score (p=0.027). We then performed an analysis of the impact on survival of the 4 gene expression in patients that had not received prior therapy. Patients with lower expression of PD-L1 had a non-significant trend towards better survival (31.5 months versus 16.2, p=0.24). Forty six of 124 patients analyzed received HMAs, and within these 46 patients lower PD-L1 expression correlated with a significantly improved overall response rate (67% vs 25%, p=0.038). Aberrant up-regulation of these 4 genes was also observed in peripheral blood mononuclear cell (PBMNC) from 61 MDS, CMML and AML patients. The relative expression of PD-L1 was significantly higher in MDS (p=0.018) and CMML (p=0.0128) compared to AML. Of interest, mRNA expression of these 4 genes was significantly higher in PBMNC than in CD34+ cells except PD-L1. By immunohistochemical (IHC) analysis, a strong correlation was observed between protein and mRNA expression. By IHC, we observed that leukemia blasts were positive for PD-L1 whereas stroma/non-blast cellular compartment was positive for PD-1 in bone marrow biopsies from MDS, CMML and AML patients. We then analyzed effect of HMAs on these four gene expression in cohort of 61 patients treated with different trials of epigenetic therapy. Treatment resulted in up regulated expression of PD-L1 in 57% of the patients (with maximum induction fold of 4.8), PD-L2 in 57% (maximum fold of 15.77), PD-1 in 58% (maximum fold of 50.26) and CTLA4 in 66% (maximum fold of 29.9). Of importance, patients resistant to therapy had increased gene expression compared to patients that achieved response (fold 5.3 vs 0.4 for PD-L1, 6.2 vs 0.4 for PD-L2, 3.0 vs 0.4 for PD-1 and 5.4 vs 0.7 for CTLA4). Comparing patients without and with expression induction, the median survival was 11.7 and 6.6 months (p=0.122) for PD-L1, 12.5 and 4.7 months (p=0.029) for PD-L2, indicating a better prognosis in patients without PD-L1 and PD-L2 induction treated with HMAs. To model this observation, we treated KG-1 and THP1 cells with different concentrations of decitabine (0 to 10 uM) and observed a dose dependent up-regulation of PD-L1 and PD-L2 in THP1 cells, and CTLA4, PD-1, PD-L1 in KG1 cells. Exposure to decitabine resulted in demethylation of PD-1 in these cell lines, and the demethylation effect was also observed in HMAs treated MDS and AML patients. This study suggests immune checkpoint PD-1/PD-1 ligands signaling may be involved in MDS pathogenesis and resistance mechanisms to HMAs. Blockade of this pathway can be a potential therapy in MDS and AML. Fig.1 PD-L1 expression in MDS CD34+ cells by IHC. Fig.1. PD-L1 expression in MDS CD34+ cells by IHC. Disclosures: No relevant conflicts of interest to declare.

Background In 2001, the World Health Organization modified the French-American-British (FAB) classification for myelodysplastic syndrome (MDS) by folding the refractory anemia with excess blasts in transformation (RAEB-t) category into acute myeloid leukemia (AML). Whether this group of patients (pts) should be treated with AML versus MDS therapy remains controversial. A subset analysis of the AZA-001 trial showed that azacitidine prolongs overall survival (OS) in elderly pts with low blasts AML (bone marrow blasts [BM] 20-30%). Aim To compare the clinical outcome and OS of patients with MDS or AML with BM blasts between 10-30%, treated with hypomethylating agents (HMA) vs intensive chemotherapy (IC). Patients and Methods We conducted a retrospective analysis of newly diagnosed pts with MDS (or AML by WHO) and BM blasts between 10-30% treated with either HMA (alone or in combination with investigational therapies) or IC on clinical trials. Eligibility was based on pt characteristics and specific protocol inclusion criteria. A univariate Cox proportional hazards regression model was used to evaluate the overall effects of treatments and outcome (remission duration (RD), and OS). Then a regression model with the interactions between treatments and baseline covariates were used for subgroup analysis. The final model was obtained by a stepwise selection using 0.05 as a cut off of significant values. Results 330 patients were included in the final analysis, with 93 (28%) HMA-treated pts and 237 (72%) pts treated with IC. Clinical characteristics at diagnosis are summarized in Table1. The overall response rate (ORR= CR+CR p) was 42% for the pts who treated with HMA and 60% for pts treated with IC (P = 0.01). The median RD was similar between the two groups (14.7 mos (m) vs. 14.7 mos, respectively, P = 0.74). Early induction mortality was also similar among the two groups (4-week mortality was 5% vs. 7%, respectively, and the 8-week mortality was 10% vs. 13 %, respectively). With median follow up of 37 mos (range, 1-94 mos), the median OS was 18.8 mos for pts who treated with HMA vs. 14.6 mos for pts treated with IC (P = 0.32). Moreover, the BM blasts percentage did not impact the overall outcome. In multivariate analysis, treatment with IC was associated with worse OS compared to HMA (HR 2.09, 95% CI 2.07-3.17, P = 0.003) but not for RD (HR 0.43, 95% CI 0.89-2.68, P = 0.13). Conclusion Although patients with MDS or AML with BM blasts between 10-30% initially achieve a higher ORR when treated with IC compared to HMA-based therapy, the OS was better for pts treated with HMA after accounting for all other covariates. Interestingly, BM blast percentages within this cohort did not impact overall outcome suggesting that pts with BM blasts 20-30% may achieve better outcome with MDS therapy. Disclosures: No relevant conflicts of interest to declare.

Background Ofatumumab is a human IgG1-kappa monoclonal antibody that binds to CD20 on normal B cells and on B chronic lymphocytic leukemia cells, resulting in B cell lysis. Ofatumumab has single-agent activity in patients (pts) with refractory CLL (Wierda, JCO 2010). Pts with CLL who have early stage disease (Rai 0-II) but have high-risk prognostic markers such as deletion 17p or deletion 11q have an increased risk of disease progression. Currently, these pts are offered watch-and-wait approach. The objective of this Phase II study is to evaluate the efficacy of ofatumumab in treating these pts with the goal to delay time to first chemoimmunotherapy treatment. Methods Pts with CLL/SLL were eligible provided they had high-risk for progression based on the presence of at least one of the following features: Rai stage II, serum beta-2 microglobulin (β2M) ≥3 mg/L, absolute lymphocyte count ≥25,000/µL, unmutated (≤2%) IGHV gene or mutated IGHV3-21, ZAP70 positive, CD38 positive (≥ 30%), or 11q or 17p deletion by FISH. Pts having a 2008 IWCLL/NCI-WG indication for CLL treatment were not eligible. Pts with Rai stage III-IV CLL were not eligible. Ofatumumab 300 mg dose 1, then 1000 mg weekly for 7 additional weeks (8 doses) was administered. Response assessment, including bone marrow evaluation, was done at least 2 months after last dose of ofatumumab and pts were followed for progression-free survival and time to first chemoimmunotherapy. Results Twenty-five pts (9 women, 16 men) were enrolled so far. The median age was 59 yrs (range, 40-81). The baseline characteristics are listed in the Table. Fifty percent of pts had unmutated IGHV gene. Thirty-four percent of pts had high-risk cytogenetic by FISH analysis. The median follow-up on the study is 4.7 months (range, 0.5-26). Grade 3-4 adverse events included infusion reaction in 6 pts. Autoimmune hepatitis with grade 4 ALT, grade 4 AST, and grade 4 alkaline phosphatase elevations was seen in 1 pt. Other grade 3-4 toxicities included rash (1 pt), shingles (1 pt), and tumor lysis (1 pt). Tumor lysis was seen in the pt with the WBC count of 207 K/µL. Eighteen pts (7 too early) are evaluable for response. Responses are as follows: 6 CR, 3 nPR, 3 PR, and 6 with stable disease. Three pts have progressed at 18.8, 14.1 and 3.2 months after starting the study treatment; 2 pts had unmutated IGHV gene (FISH negative) and one pt had trisomy 12 (IGHV mutation status unknown). None of the pts with deletion 17p or deletion 11q have progressed. All pts are alive at this time. The median overall follow up time is 7.6 months (range, 1-28). Conclusions Ofatumumab is well tolerated in pts with early stage CLL and may delay time to first chemoimmunotherapy. Disclosures: Ferrajoli: GlaxoSmithKline: Research Funding.

Background Comorbidites play a crucial role in the prognosis, treatment and outcomes of patients with hematological malignancies. We have previously reported the impact of comorbidities independent of age and IPSS score (1). The aim of this study was to determine whether comorbidities continued to have a similar impact when patients were reclassified according to the Revised-International Prognostic Scoring System (RIPSS). Methods We reviewed the medical records of 600 consecutive MDS patients who presented to MD Anderson Cancer Center from January 2002 to June 2004. The Adult Comorbidity Evaluation-27 (ACE-27), a validated 27-item comorbidity index for cancer patients (2), was used to assess the severity of comorbid conditions. For each patient, we obtained demographic data and specific staging information based on the R-IPSS. We also collected information on stem cell transplantation (SCT), mortality and survival. Kaplan-Meier methods and log-rank tests were used to assess survival. Results Of the 600 patients included in this study, 402 (67%) were male, and 517 (86%) were white; median age at presentation was 66 years (17 – 94); mean duration of follow-up was 54 months (1 - 100). A total of 34, 117, 131, 129, 165 patients were R-IPSS very low, low, intermediate, high and very high respectively. The ACE-27 comorbidity scores were as follows: none, 137 patients (23%); mild, 254 (42%); moderate, 127 (21%); and severe, 82 (14%). 476 (79%) patients died, and 54 (9%) patients underwent SCT. Overall median survival using the Kaplan-Meier method was 16 months (1 – 100). Median survival according to ACE-27 scores was: 28 months for no comorbidity, 16 months for mild comorbidity, 14 months for moderate comorbidity, and 9 months for severe comorbidity with a P-value of P65 years (P=0.18). Conclusion Comorbidities had a significant impact on the survival of patients with myelodysplastic syndrome, especially among those with more advanced disease and younger age ( References 1. Naqvi K, Garcia-Manero G, Sardesai S. Association of comorbidities with overall survival in myelodysplastic syndrome: Development of a prognostic mode. J Clin Oncol. 2011;29(16):2240-6. 2. Piccirillo JF, Tierney RM, Costas I, et al. Prognostic importance of comorbidity in a hospital-based cancer registry. JAMA. 2004;291:2441-47. Disclosures: Ravandi: Sunesis: Consultancy, Honoraria, Research Funding; Celgene: Honoraria, Research Funding; Novartis: Honoraria, Research Funding; Bristol Myers Squibb: Honoraria, Research Funding; Teva: Consultancy, Honoraria; Pfizer: Honoraria; Merck: Research Funding; Bayer/Onyx: Consultancy, Honoraria; EMD Serono: Research Funding; Medimmune: Research Funding; Amgen: Consultancy, Honoraria; Seattle Genetics: Consultancy, Honoraria. Garcia-Manero:Novartis Pharmaceutical: Research Funding. Kantarjian:Sanofi-Aventis: Research Funding.

Introduction Monoclonal or oligoclonal T cell receptor positive populations (TCR+) can be observed in elderly individuals with viral infections, patients (pts) with pancytopenia, bone marrow failure syndromes, and in large granular lymphocytic leukemia. Expansion of these T cell clones may indicate that an antigen driven myelosuppressive mechanism is active in pts with MDS. However the clinical significance of clonal T cell population/TCR+ in MDS is not clear. In the present study we have reported on the clinical characteristics and prognostic relevance of monoclonal or oligoclonal TCR+ cells in pts with MDS. Methods A retrospective chart review was performed for all pts with a diagnosis of MDS in whom TCR clonal assessment was done at our institution between 2005 and 2013. Using labeled V-primers, TCR (beta and gamma) were assessed in the bone marrow (BM) samples by semi quantitative polymerase chain reaction (PCR) assay with a sensitivity of 1:100 to 1:10,000 lymphocytes. Results The clinical characteristics and outcomes of 50 pts, with a median age of 67 yrs (range 35-85) were analyzed. 17 (34 %) pts had TCRβ+ clones, 22 pts (44 %) had TCRγ+ clones. Pts with any TCR+ clones (n=27; 54 %) were compared with those pts who were negative for TCR clonality (n=23; 46 %). The median follow-up for the pts with or without TCR+ clones was 17 months vs. 19 months, respectively. Clinical characteristics of the pts with and without TCR+ clones are outlined in Table 1. Overall, the groups were similar with regards to blood counts, blast percentages, bone marrow cellularity, and karyotype. There were trends towards higher percentages of males, lower neutrophil count, higher IPSS category, and more pts with >10% BM blasts in the TCR+ group, but these did not reach statistical significance. Three pts (6%) transformed to acute myeloid leukemia and two of these pts had TCR+ clones in the BM. The majority of these pts were previously untreated 34/50 (68%). We then looked at response to therapy with hypomethylating agents in those pts who were previously untreated and/or who had previously received one type of therapy (N=42). In 20 pts with TCR+ clones, 8 (40%) received hypomethylating agents and 2 went into complete remission (25%), 2 had stable disease and 4 (20%) were non responders. In 22 pts without TCR+ clones 4 (18%) received hypomethylating agents and there were no (0%) responders. The estimated 3 year overall survival was 21% vs 47% between pts with andwithout TCR+ clones, respectively (P=0.67). In a subset analysis restricted to pts with available TCRγ assessment (n=67), the estimated 3 year OS for pts with TCRγ+ vs. those without TCRγ clones was 17% vs 57%, respectively (P=0.08) (Figure 1). No difference in OS was observed when pts with or without TCRβ clones were compared. Conclusions We report our experience of the clinical impact of clonal TCR+ populations in pts with MDS. TCR+ clonal T-cells were present in about half of pts with MDS. Although the study is limited with number of pts, those with MDS and clonal TCR+ population tend to have higher proportions of poor risk features and exhibit a trend of inferior outcome as compared to those pts without TCR+ clones. Larger prospective studies are needed to better define the clinical and mechanistic relevance of clonal TCR+ populations in pts with MDS. Disclosures: No relevant conflicts of interest to declare.

Background The aminopeptidase inhibitor tosedostat is an orally bioavailable novel chemotherapeutic, functioning through enzymatic blockade of intracellular protein degradation and re-synthesis. Encouraging results have been observed with tosedostat monotherapy in Phase I and II trials of relapsed and refractory elderly AML patients. Aims This Phase I/II trial explores the safety, tolerability and activity of tosedostat in combination with cytarabine or azacitidine in older patients with AML or high-risk MDS. Methods Eligibility includes age >60 years, performance status of 0-2, with relapsed/refractory AML or high-risk MDS having failed prior hypomethylating agent (HMA) and/or lenalidomide therapy. All subjects received tosedostat 120mg orally once daily for 28 day cycles, with either subcutaneous (SQ) cytarabine at starting dose of 7.5mg twice daily for 10 days, or azacitidine (AZA) IV/SQ at starting dose of 50mg/m2 daily for 7 days, per investigator’s choice. A modified 3+3 dose escalation design was used in the Phase I portion to identify the maximum tolerated dose independently for both arms. Escalation to the predefined target dose levels of 10mg SQ cytarabine twice daily for 10 days, or azacitidine 75mg/m² IV/SQ daily for 7 days per 28 day cycle was achieved. Dose escalation to 180mg daily tosedostat was allowable for patients not achieving a CR after 4 weeks on therapy. Results From November 2012 to April 2013, the Phase I portion completed enrollment with a total of 18 patients (10 with AZA, 8 cytarabine). Six patients discontinued prior to completion of cycle 1 for reasons other than study-drug related toxicity or progressive disease and were replaced; they are considered as non-responders in the primary efficacy analysis. Median age was 73 (range 60-81), and 56% were male. 13 patients (72%) had secondary or therapy-related AML, of which 11 (61%) had prior MDS. The median number of prior treatments for MDS and/or AML was 2 (range 1-6), and 11 (61%) had received prior HMA therapy. Median white blood cell (WBC) count at study start was 5x10⁹/L (range 0.3 – 44.2), with median peripheral blood blasts of 27% (range 0-95%), and bone marrow blasts 57% (range 6-90%). The majority (61%) of patients had complex cytogenetics or abnormal cytogenetics involving chromosome 5 and/or 7; six patients (33%) had diploid cytogenetics. Molecular analysis identified 1 FLT3-ITD, 3 DNMT3A, 2 JAK2 V617F, 2 NRAS, 2 KRAS, and 1 IDH1 mutation. Median duration on study was 49 days (range 13 – 254), with median overall survival (OS) of 3.1 months (range 0.4-8.5) and an overall response rate (ORR) of 33% (CR/CRp/MLFS 17%, PR 17%). ORR was 50% (CR/CRp/MLFS 25%, PR 25%) in fully evaluable patients remaining on study for >28 days. Median OS for patients with CR/CRp/MLFS and PR was 7.3 and 3.5 months, respectively, while OS for unevaluable patients or those with progressive disease was 29 days. On univariate analysis, the presence of a diploid karyotype (p=0.001) or WBC count The most common non-hematologic toxicities regardless of attribution (all %; grade >3) included pleural and pericardial effusions (72%; 0%), dyspnea (67%; 0%), peripheral edema (61%; 6%), pneumonia (56%; 50%), fatigue (50%; 0%), diarrhea (50%; 0%), and cough (39%; 0%). Additional cardiac toxicities included systolic dysfunction (17%; 6%), QTc prolongation (50%; 6%) and myocardial infarction (MI) (11%; 11%). One fatal acute coronary event on cycle 2, day 2 of tosedostat 180mg and AZA 75mg/m² was considered possibly related to study drug. Conclusions The combination of tosedostat with low-dose cytarabine or azacitidine appears effective in a population with an overall very poor prognosis. Additional safety and efficacy evaluations are ongoing. Updated results and follow-up will be presented at the annual meeting. Disclosures: No relevant conflicts of interest to declare.

Background The combination of vorinostat, idarubicin and cytarabine (IA+vorinostat) is associated with high response rate in patients with newly diagnosed acute myelogenous leukemia (AML) or higher-risk myelodysplastic syndromes (MDS) (JCO 2012;30:2204). In that study, presence of FLT3-ITD was associated with 100% overall response rate (ORR) in 11 patients. To confirm the efficacy of this combination, we extended the phase II study to treat 2 additional cohorts: one for patients with newly diagnosed (untreated cohort) and the other with relapsed and refractory (R/R cohort) AML or higher-risk MDS with FLT3 alteration (both ITD and D835 mutation). Methods Patients with the above diagnosis, ages 15 to 65 years, with appropriate organ function (measured cardiac ejection fraction ≥ 50%, serum creatinine ≤ 2 mg/dl, total bilirubin ≤ 2 mg/dl, and GPT/GOT ≤ 2.5 x upper limit of normal) whose eastern cooperative group (ECOG) defined performance status ≤ 2 were eligible for the study. Study treatment comprised of vorinostat 500 mg orally three times a day (days 1 to 3), idarubicin 12 mg/m2 intravenously (IV) daily x 3 days (days 4 to 6), and cytarabine 1.5 g/m2 IV as a continuous infusion daily x 3 - 4 days (days 4 to 7). Patients in remission could be treated with five cycles of consolidation therapy with lower dose combination and up to 12 months of maintenance therapy with single-agent vorinostat. Result Untreated cohort included 26 patients, whereas 13 patients were treated in R/R cohort (total 39 patients). Thirty six patients had de novo AML, 1 had de novo MDS and 2 had therapy-related AML. For the R/R patients, the median number of prior therapies was 3 (range: 1-6). The median age of the entire study group was 49 (range: 19-64) and 17 (44%) were male. Among the patients whose cytogenetic result were available, 20 (51%) patients had normal and 15 (39%) had abnormal karyotype. By Medical Research Council (MRC) criteria, 30 (77%) patients had intermediate risk and 9 (23%) had poor risk karyotype. Thirty three (85%) patients had FLT3-ITD only, 4 (10%) had both FLT3-ITD and D835 mutation, and 2 (5%) had D835 mutation only. Seventeen (44%) patients had NPM1 mutation. In untreated cohort (N = 26), CR and CRp were documented in 21 (80%) and 2 (8%) patients, respectively (ORR = 88%). In R/R cohort (N = 13), overall response (OR) was observed in 4 (30%) patients (CR in 2 [15%] and CRp in 2[15%]). Of those 4 patients who had OR in R/R cohort, 2 patients were refractory to other high-dose cytarabine-based regimen. The median duration of CR or CRp was 9.2 months (range: 0.1-48.4) in untreated cohort and was 2.9 months (range: 1.6-4.7) in R/R cohort. Twelve (46%) patients in the untreated cohort were bridged to stem cell transplant (SCT) while they were in 1st CR. None of the patients in R/R cohort were bridged to SCT. No difference in response was observed in 1) younger (Age < 60) vs. older patients, 2) normal vs. abnormal karyotype, 3) intermediate vs. poor risk cytogenetics by MRC criteria, 4) presence of RAS mutation, 5) presence of NPM1 mutation, or 6) de novo vs. therapy-related disease. The median overall survival (OS) was 21.7 months (95% CI: 8.1-35.3) in the untreated cohort and was 4.9 months (95% CI: 0.1-10.4) in the R/R cohort. Early treatment related mortality (defined by the death within 4 weeks of the induction) was documented in 1 (4%) patient in the untreated cohort and 2 (15%) patients in the R/R cohort. Toxicity profiles were similar to that reported in the original phase II study (JCO 2012;30:2204). Discussion Vorinostat in combination with IA provides high response rate and durable remission in previously untreateed AML or higher risk MDS patients with FLT3 alteration but is less effective in patients with R/R disease. Phase III randomized study of IA+/- vorinostat in previously untreated AML patients is ongoing (SWOG S1203). Disclosures: Off Label Use: vorinostat in MDS and AML. Cortes:Ambit: Research Funding; Astellas: Research Funding; Argo: Research Funding; Novartis: Research Funding.

Background Treatment with hypomethylating agents (HMA) such as azacitidine and decitabine has changed the overall outcome of patients with MDS. Failure to responed to or relapse from treatment with HMA carries a poor outcome and no approved therapy for these patients exists. Clofarabine is a second generation nucleoside analog with single agent activity in MDS. Aim This is a phase II trial to evaluate the safety and activity of the combination of clofarabine and low dose cytarabine in the treatment of patients with high risk MDS who failed prior HMA therapy. Material and Method Eligible were adults older than 18 years with MDS who have had no response, progressed, or relapsed following at least 4 cycles of therapy with either azacitidine and/or decitabine. Patients were required to have an ECOG performance status of Results Between January 2012 and March 2013, 29 patients were enrolled. The clinical characteristics are summarized in Table 1. Twenty four patients were evaluable for response (5 patients were too early for response). The overall response rate (ORR) was 50% (8 [33%] achieved complete remission (CR)/ complete remission with incomplete platelets recovery (CRp)/marrow CR, and 4 [17%] had stable disease with hematological improvement). With a median follow up of 4.9 months (range, 1.9-16.7), the median overall survival (OS) was 4.8 months (range, 0.5-13.5). Most toxicities were grade /= 3 toxicities included: elevated liver enzymes (3%) and elevated bilirubin (3%). Four-week mortality was 12%. Conclusion The combination of clofarabine and low-dose cytarabine has an ORR of 50% in patients with MDS who failed prior therapy with HMA. The study continues to accrue and updated results with longer follow up will be presented at the meeting. Disclosures: Off Label Use: Clofarabine use in MDS. Faderl:Sanofi-Aventis: Consultancy, Membership on an entity’s Board of Directors or advisory committees, Research Funding.

Background AML has been extensively studied in a gene-to-gene and pathway-to-pathway fashion over the years, unraveling insightful local patterns that capture heterogeneity in patients and identify potential drug targets. However, our understanding of AML from a global and systems perspective is still lacking. A global proteomic pathway map is yet to be drawn to integrate local activity patterns and to translate patient classifications across pathways. This will not only improve our scientific understanding of how different functional pathways are inter-related, but will also enable us to develop more robust and effective therapeutic regimens based on pathway cross-talks. Methods A proteomic profile, containing the expression pattern of 231 proteins in each of the 415 newly diagnosed AML patients at UT MD Anderson Cancer Center, was generated using the Reverse Phase Protein Array (RPPA) technology. We grouped these proteins into 23 functional pathways based on protein association known in literature and correlation shown in the proteomic profile. Principal component analysis and scree plot were used to refine the pathway construction. The AML patients were clustered by their protein expression patterns within each individual pathway, and were then compared across pathways. The association of patient clusters between different pathways was measured by Goodman-Kruskal's (GK) tau method, indicating the predictability of patient clustering in one pathway given that in the other. This association between pathways and interchangeability of patient groupings were visualized in a circos plot (Figure 1), depicting a global proteomic pathway map. Results The global proteomic pathway map illustrates how strongly protein expression patterns of different pathways are associated, and how patient classifications under different pathways could be translated from one to another. Here, we highlight some of the key insights surfaced from this analysis. First, we identified ‘social' pathways that have intensive cross-talks with multiple other pathways, including some of the cell signal transduction pathways (MEK, PI3K, mTOR), genetic information processing pathways (transcription, histone methylation), and cell survival/death pathways (apoptosis, autophagy). We also identified ‘orphan' pathways that are more independent and are poorly associated with others. These include a subset of signal transduction pathways (pkc, tp53, S6rp, Src, Creb, Wnt), cytoskeleton and differentiation. As the association is directional, each pathway could be further characterized as either a ‘sender' or a ‘receiver' pathway based on whether it is acting more as the origin or the target of the link. The patient clusters from the ‘sender' pathways (e.g. Apoptosis, mTOR, Fli), could be easily translated to other pathways, while the patient clusters in ‘receiver' pathways (e.g. Hippo and Transcription), are highly predictable by patient clusters from multiple other pathways. We further constructed and compared the global pathway maps for patients in different cytogenetic groups. Comparison of pathway maps from patients with favorable, intermediate and unfavorable cytogenetics shows the power of this methodology to discern differences in the degree of correlation between protein functional groups. Favorable cytogenetics (T8;21) and inversion 16, because they are more similar have less patient to patient variation and thus have a more consistent and highly correlated pathway map with a higher number of connections. Conclusions Based on the RPPA data in AML patients, we built a global proteomic pathway map that captures the association between protein expression patterns in defined protein functional groups. We identified intensive interacting pathways as well as independent pathways, which indicate potential hubs and modulators of leukemic cell behavior. We further compared maps of different cytogenetic groups and revealed different correlation mappings. We are further refining the algorithms in order to study more focused changes within lower population subsets. Ultimately we believe that this will enable the matching of targeted agents to specific settings where the target is expressed and highly interactive based on proteomic data. Disclosures: No relevant conflicts of interest to declare.

Background Myelodysplastic/myeloproliferative neoplasm, Unclassifiable (MDS/MPN-U) is a distinct entity among the myeloid neoplasms, characterized by the simultaneous presence of both dysplastic and proliferative features. The natural history, clinical characteristics, response to treatment and patient outcome of patients with MDS/MPN-U has not been well described, apart from the MDS/MPN-U provisional entity of Refractory Anemia with Ringed Sideroblasts and Thrombocytosis (RARS-T). Aims Clinicopathologic features of 85 patients with MDS/MPN-U and without RARS-T were evaluated, along with clinical outcomes based on treatment strategy. Methods All patients with MDS/MPNs treated from July 1986 to February 2013 at MDACC were reviewed. Patients seen prior to 2006 with a recorded diagnosis of MDS/MPD or chronic MPD-U were reviewed by two hematopathologists and diagnoses were revised as appropriate according to the current WHO criteria. All RARS-T patients per WHO 2008 definition were excluded, and patients were excluded if the de novo presence of both myelodysplastic and proliferative features at diagnostic presentation per WHO criteria was unable to be confirmed. Results MDS/MPN-U patients were predominantly over the age of sixty (92%), with a median age of 70 years (range 22 – 90), and 61 patients (72%) were male. 29 patients (35%) had splenomegaly at presentation. Median presenting white blood cell (WBC) count was 17 x 10⁹/L (1 – 141 x 10⁹/L), median hemoglobin was 10 g/dL (5 – 15 g/dL), and median platelet (PLT) count was 85 x 10⁹/L (6 – 1168 x 10⁹/L). Of 56 patients with known JAK2 status, 17 patients (30%) had the presence of the JAK2 V617 mutation. The majority of patients had either diploid cytogenetics (49%) or trisomy 8 as the sole abnormality (15%); 10 patients (12%) had a complex karyotype. Median OS was 12.4 months (range 0.3 – 138.7). A uniquely favorable outcome was seen in the 11 patients presenting with a PLT count ≥ 450 x 10⁹/L, with median OS 52.5 vs 11.9 months (p Overall, 36 patients (42%) received hypomethylating agent (HMA) therapy. 13 (15%) were treated with an immunomodulatory approach (e.g. lenalidomide, interferon-α), five (6%) underwent splenectomy and five (6%) received stem cell transplant (SCT). HMA-treated patients had an OS of 16.4 vs 11.5 months with other approaches (p=0.57); patients on immunotherapy had a non-significant trend towards worse OS of 11.4 vs 15.4 months (p=0.07). The five patients receiving splenectomy had OS 51.1 m vs 10.0 mo (p=0.12), and with SCT 12.4 vs 12.5 months (p=0.16). Conclusion Overall, our analysis validates the diagnosis of MDS/MPN-U as a unique pathologic entity, with features and prognostic variables distinctive from either PMF or MDS. While no statistically significant differences in OS were identified based on therapy; in terms of the various diagnostic, prognostic and treatment strategies, MDS/MPN-U patients without RARS-T appear to have characteristics and outcome most analogous to MDS. The favorable outcome of surgical splenectomy among those MDS/MPN-U patients with splenomegaly suggests this may also be part of a a reasonable therapeutic approach. Disclosures: No relevant conflicts of interest to declare.

MYD88 is a key mediator of Toll-like receptor innate immunity signaling. Oncogenically active MYD88 mutations have recently been reported in lymphoid malignancies, but has not been described in MDS. To characterize MYD88 in MDS, we sequenced the coding region of the MYD88 gene in 40 MDS patients. No MYD88 mutation was detected. We next characterized MYD88 expression in bone marrow CD34+ cells (N = 64). Increased MYD88 RNA was detected in 40% of patients. Patients with higher MYD88 expression in CD34+ cells had a tendency for shorter survival compared to the ones with lower MYD88, which was significant when controlled for IPSS and age. We then evaluated effect of MYD88 blockade in the CD34+ cells of patients with lower-risk MDS. Colony formation assays indicated that MYD88 blockade using a MYD88 inhibitor resulted in increased erythroid colony formation. MYD88 blockade also negatively regulated the secretion of interleukin-8. Treatment of MDS CD34+ cells with an IL-8 antibody also increased formation of erythroid colonies. These results indicate that MYD88 plays a role in the pathobiology of MDS and may have prognostic and therapeutic value in the management of patients with this disease.

Abstract 2481 Purpose: Cancer-associated IDH mutations produce the metabolite 2-hydroxyglutarate (2HG), but the clinical utility of serum 2HG measurements has not been previously established. We studied whether 2HG measurements in AML patients correlate with the presence of IDH mutations and whether diagnostic or remission 2HG measurements predict survival. Patients and Methods: Serum samples from 223 previously untreated adults (≤ 60 years of age) with de novo AML from the Eastern Cooperative Oncology Group E1900 clinical trial (62 IDH mutated, 161 IDH wild-type) were analyzed for 2HG concentration by reverse-phase liquid chromatography coupled to mass spectrometry (GC-MS). Results: Pretreatment 2HG levels ranged from 10 to 30000 ng/ml and were significantly elevated in IDH-mutant samples (median 3004.1 ng/ml), as compared to the wild-type cohort (median 61.2 ng/ml) (p < 0.0005). 2HG levels did not differ among the specific IDH1 or IDH2 allelic variants. In ROC analysis, a discriminatory level of 700 ng/ml segregated patients with and without IDH mutations with 86.9% sensitivity and 90.7% specificity. On repeat mutational analysis of 13 IDH wild-type samples with 2HG levels >700 ng/ml, IDH mutations were identified in nine samples, most often at low allele burden. IDH mutant patients with 2HG levels ≤ 200 ng/ml at complete remission experienced improved overall survival compared to those with higher 2HG levels (HR 3.5, p = 0.02) (Figure 1). Conclusion: We establish a firm association between IDH mutations and elevated serum 2HG concentration in AML. These data confirm that peripheral blood measurement of an oncometabolite provides useful diagnostic and prognostic information for cancer therapy, and furthermore can inform patient selection of IDH mutant targeted therapies. Disclosures: Levine: Agios Pharmaceuticals: Research Funding. Straley:Agios Pharmaceuticals: Employment. Yen:Agios Pharmaceuticals: Employment. Agresta:Agios Pharmaceuticals: Employment. Carroll:Agios Pharmaceuticals: Research Funding.

Abstract 3576 Eltrombopag, a non-peptide, small molecule, thrombopoietin receptor agonist, has been shown to increase platelet counts in patients with idiopathic thrombocytopenic purpura (ITP), hepatitis C associated thrombocytopenia and severe aplastic anemia. Pre-clinical studies show that eltrombopag stimulates megakaryopoiesis in bone marrow samples from patients with acute myelogenous leukemia (AML) and inhibits leukemic cell growth. The clinical use of eltrombopag in the AML patient population has not been previously explored. We report the results of an investigator-initiated, phase I dose escalation clinical trial to evaluate the safety and efficacy of eltrombopag monotherapy in older patients with AML. Eligible patients were ≥ 60 years old with non-M3 AML and a baseline platelet count ≤75,000/ul. A standard 3+3 design was used employing dose escalation in 4 sequential cohorts of 50mg, 100mg, 200mg and 300mg of eltrombopag daily. Subjects who did not complete at least one cycle of therapy (4 weeks) or experience a dose-limiting toxicity (DLT) were replaced. The primary endpoint was safety. Secondary endpoints included platelet and disease response. Bone marrow biopsies to evaluate for disease response and fibrosis were planned at one month, 3 months and 6 months. Twenty-three subjects were enrolled from June 2010 to February 2012 at the Hospital of the University of Pennsylvania and received at least one dose of study drug. The median patient age was 73 (range 60–86). Twenty-one of 23 subjects were relapsed or refractory after prior therapy. Two of 23 subjects (ages 85 and 86) were previously untreated. Nineteen subjects were platelet transfusion-dependent. Thirteen subjects had a history of antecedent myelodysplastic syndrome (MDS) and 6 were requiring hydroxyurea at time of study entry. Fourteen of 23 subjects completed the first cycle of therapy. Three remained on study for the entire 6-month study period. At the time of this report one subject remains on eltrombopag in an extension phase (8+ months of therapy). Indications for drug cessation included: possible DLT=1; progressive disease=5; death due to sepsis=5; lack of response=6, loss of response=1; and pursuit of hospice or alternative therapy=4. Overall eltrombopag was well tolerated at all doses studied and a maximally tolerated dose (MTD) was not reached. One subject at the 300mg dose level developed Grade 4 hepatic laboratory abnormalities possibly related to study drug, which met criteria for a DLT. The subject had concurrent polymicrobial sepsis and subsequently died. No other grade 3 or greater liver function abnormalities were observed. Asymptomatic skin discoloration with associated abnormal serum pigmentation was noted in subjects at the 200mg and 300mg dose levels. There were no platelet responses in the 50 mg or 100 mg cohorts (7 subjects total). Two of 7 subjects treated at the 200mg dose level and 2 of 9 subjects treated at the 300mg dose level achieved a platelet response (25% platelet response rate at these higher dose levels). All 4 of these subjects became platelet transfusion independent. The duration of platelet responses were 5 weeks, 12 weeks, 6 months and 8+ months (ongoing at time of this report). One subject with primary refractory AML, associated with monosomy 7, obtained a complete morphologic and cytogenetic response by 3 months of therapy at the 300mg dose level. This response was sustained at most recent bone marrow biopsy at 6 months of therapy and the patient remains on study drug. Increased reticulin fibrosis was also observed in this subject over time. No other subjects experienced a PR or CR. 6 patients had stable disease (SD) at one month, which was maintained at 3 months for 4 patients. Our study shows that eltrombopag is well tolerated in patients with AML at doses well above those typically used to treat ITP. At higher doses, we observed clinically meaningful platelet and antileukemic responses. Further exploration of eltrombopag as both a platelet agonist and disease-modifying agent in the treatment of AML is justified. Disclosures: Frey: GSK: Consultancy, Research Funding. Porter:Novatis: Patents & Royalties; Celgene: Honoraria; Genentech: Employment; Pfizer: Research Funding. Perl:Astellas: Consultancy. Carroll:GlaxoSmithKlein: Research Funding.

6579 Background: Increased risks of solid and lymphoid malignancies are described in patients with autoimmune (AI) disease. The association between AI disease, AI treatment (particularly anti-TNF therapy), and AML is less established. Moreover, the pathologic characteristics of AML in this population have not been evaluated. Here we describe our AML experience in patients with prior AI disease. Methods: Patients with AML from 1992 to 2011 were identified from our database, and AI disorder was verified through chart review. Patients were included if they had an AI disorder that required systemic treatment, and diagnosed > 1 year prior to AML. Patients were excluded if details including the year of AI diagnosis or treatment(s) received were not documented. A total of 22 patients were included for analysis. Results: Median age at AML diagnosis was 59 years (range 32 – 78), 4 (18%) were men. 10 (48%) had received an anti-TNF agent, for a median duration of 30 months (range 2 – 120). Median latency from AI diagnosis to AML was 9.4 yrs. All FAB subtypes were represented except M0 and M6, 9 (41%) had M4 or M5 disease. 10/21 patients (45%) had normal cytogenetics, and 5 had favorable cytogenetics (2 with acute promyelocytic leukemia (APML)). Standard induction was given to 19 patients, and APML-directed therapy for those with APML. One patient received an autologous transplant in CR1, and 5 an allogeneic transplant (1 in CR1, 4 in CR2). 9 patients are alive in CR1. 13 relapsed, and 4 (31%) are alive in CR2 or CR3 (3 following allogeneic transplant in later CR). Median OS was 68.1 months. In univariate analysis, factors associated with OS were cytogenetics, FAB subtype, and gender. Anti-TNF therapy may be associated with poorer prognosis; this was not statistically significant. Median OS was 14.1 months for poor, 68.1 months for intermediate, and not reached for favorable risk cytogenetics. Conclusions: Median OS was higher (> 5 years) than expected. Younger age and an increased incidence of favorable risk AML may account for this finding: whether this is a meaningful biologic association or stochastic event can only be verified with larger studies. Our observations do not support a label of “secondary leukemia” or therapy-related neoplasm in this population.

Abstract 2366 Background: Allogeneic hematopoietic stem cell transplant (HSCT) has been shown to be of some benefit for patients with acute myelogenous leukemia (AML) with untreated early relapse. However, outcomes for patients who have morphologic evidence of relapsed leukemia are not well defined. We describe our experience with HSCT for patients with active AML at the time of transplant, to determine prognostic factors for outcome after transplant. Patients and Methods: We analyzed recipients of myeloablative HSCT from 1997 to 2009 at our institution with morphologically active AML at the time of transplant. Patients were identified through our transplant database, and disease status was verified through medical record review. Forty patients were included based on the presence of circulating blasts at the time of admission for transplant, and/or a positive bone marrow biopsy (>5% bone marrow blasts) immediately prior to transplant. There were 6 patients coded as “relapsed/refractory” in our database whose disease status at the time of transplant was unable to be confirmed and thus were excluded from analysis. Results: Baseline patient and transplant characteristics are shown in Table 1. 30% of patients were transplanted for primary induction failure, 22% for first untreated relapse (no treatment between first documented relapse and HSCT), 35% for first refractory relapse (did not attain CR with treatment administered at time of first relapse), and 13% had second or greater refractory relapse. The overall survival for all patients was 82% at 30 days; 55% at 100 days; 28% at one year; and 18% at two years post-HSCT. Of 31 deaths, 71% were attributable to disease, 19% to regimen-related toxicity and infection, and 10% to graft-versus-host-disease (GVHD). Seven patients remain alive at the time of analysis, with 6 patients (15%) alive and free of disease more than 3 years post-HSCT. The median follow-up of the disease-free survivors is 9 years; Table 1 shows characteristics for this subgroup. We identified several patterns of interest. Four disease-free survivors were transplanted in first untreated relapse, yielding 44% (4/9 patients) with first untreated relapse that are long-term disease-free survivors. In addition, while most patients were transplanted with circulating blasts, 5 of 6 disease-free survivors did not have circulating disease. Although 40% of transplants were from unrelated donors, 5 of 6 disease-free survivors received sibling transplants. Conclusions: Our review of patients with morphologically apparent relapsed or refractory AML confirmed that a subset of patients can achieve a durable remission from HSCT. The majority of these long-term survivors shared important characteristics, including (1) lack of circulating blasts at transplant, (2) sibling donors, and (3) first untreated relapse disease status. This information may serve a prognostic purpose, and may assist in identifying appropriate candidates for transplant or for alternative therapies. However, it was not possible based on our analysis to predict reliably those who would not experience long-term survival. Except for second or greater refractory relapse, there was no factor that identified patients who had no benefit from HSCT. As such, HSCT remains a viable option with the potential for long-term disease-free survival in this population. Disclosures: No relevant conflicts of interest to declare.

7002 Background: Smoking increases the relative risk of AML by 40% and 25% in active and former smokers, respectively, compared with non-smokers (Fircanis et al., 2014). While the relationship of smoking with AML development is recognized, whether smoking impacts underling AML biology and clinical outcome remains ill-defined. Methods: Newly diagnosed, treatment naïve AML pts seen at MDACC between 2012 and 2017 with available smoking history were evaluated, along with baseline parameters, co-occurring mutations, cytogenetics and clinical outcome. Results: We identified 858 pts [486 (57%) male; median age 67 yrs (14-97)], representing 535 (62%) treatment naïve and 323 (38%) salvage pts. Smoking status was recorded as smokers (active = 39 pt, former = 380 pt), versus never smoker (439 pt). In tx naïve group, smoking is associated with lower remission rates (OR 0.63, 95% CI 0.43-0.94, p = 0.02) and inferior OS (HR = 1.6, 95% CI 1.27-2.02, p < 0.001). Smoking status was not significant in multivariate analysis including AML biologic characteristics and ELN 2017 risk stratification. Therefore we postulated that worse OS may be driven by smoking associated AML biology rather than smoking associated comorbidities. Indeed, in univariate analysis smoking was associated with poor ELN risk (p = 0.015), complex karyotype (p = 0.0002), and TP53 mutation (p = 0.0235) while negatively associated with NPM1 (p = 0.018), FLT3-ITD (p = 0.032) and GATA2 (p = 0.0497). Age was a significant cofounder between smokers vs non-smoker ( < 0.0001). After controlling for age, significance was retained for ELN risk, complex karyotype and GATA2 at p = 0.0454, p = 0.0006, p = 0.048 respectively, while significance was lost for NPM1 (p = 0.079), FLT3-ITD (p = 0.1) and TP53 (p = 0.084). In analysis of young pts ( < 60 yr), smoking is positively associated with complex karyotype (p = 0.0042) and TP53 (p = 0.0289), and negatively associated with RUNX1 (p = 0.0143) and IDH2 (p = 0.0357). Conclusions: We report the largest analysis of smoking status and impact on molecular, cytogenetics, and AML clinical outcomes. Smoking history is associated with poorer risk molecular and cytogenetics, lower response rate and shorter survival in treatment naïve patients.

7010 Background: Isocitrate dehydrogenase 2 ( IDH2) mutations occur in 5% of patients (pts) with MDS. Enasidenib (ENA) is a selective oral inhibitor of the mutant IDH2 enzyme with single-agent activity in relapsed/refractory acute myeloid leukemia (AML). We report the results of the open label phase II study designed to evaluate the efficacy and tolerability of ENA, as monotherapy or in combination with azacitidine (AZA) in pts with higher-risk IDH2-mutated MDS (NCT03383575). Methods: Pts with higher-risk [Revised International Prognostic Scoring System risk > 3 or high molecular risk (HMR)] MDS/CMML or LB AML naïve to hypomethylating agents (HMA) received ENA100 mg orally daily for 28 d of each 28-d cycle + AZA 75 mg/m2 IV or SC on d 1-7 of each cycle (ENA+AZA), and pts with refractory or progressive MDS to prior HMA therapy received ENA alone (ENA), in 28-d cycles until unacceptable toxicity, relapse, transformation to AML, or progression. The primary endpoint was overall response rate (ORR) [complete remission (CR), marrow CR (mCR), partial remission (PR) and hematologic improvement (HI)]. Other endpoints include safety, and survival outcomes. Results: 48 pts received ENA+AZA (n = 26) or ENA (n = 22). The median age was 73 yrs (range, 46-83). Most pts (72%) had HMR: ASXL1 (39%), and RUNX1 (17%). Median number Tx cycles was 4 (2–32) in the ENA+AZA, and 7 (1–23) in the ENA arm. Common Tx-related grade 3–4 AEs in the ENA+AZA arm were neutropenia (64%), thrombocytopenia (28%), and anemia (8%); these occurred in 10%, 0%, and 5%, in the ENA arm. Grade 3–4 infections occurred in 32% (ENA+AZA) and 14% (ENA). IDH differentiation syndrome occurred in 3 pts (12%) in the ENA+AZA and 5 pts (24%) in the ENA arm. Two deaths occurred during the initial 60 d, both unrelated to study and due to COVID. In response-evaluable pts (n=46), ORR was 84% (n = 21/25; 24% CR + 8% PR+44% mCR+ 8% HI] in the treatment naïve ENA+AZA and 43% (n = 9/21; 24% CR+5%PR+5% mCR+10% HI) in the HMA failure ENA arm (Table). Most common reason for Tx discontinuation was disease progression (ENA+AZA 20%, ENA 33%).5 pts (20%) received HCT in the ENA+AZA and 1 (5%) in the ENA arm. 7 pts in the ENA+AZA and 5 in the ENA arm were ongoing at data cutoff (Dec 31, 2020). After a median follow up of 12.6 mo, median OS was 32.2 mo in the ENA+AZA and 21.3 mo in the ENA arm. Conclusions: ENA is well tolerated and shows promising efficacy in IDH2-mutated higher risk MDS. Follow up and accrual is ongoing to better define duration and biomarkers of response. Clinical trial information: NCT03383575. [Table: see text]