Citations

Abstract Introduction: Hypocellular myelodysplastic syndrome (h-MDS) represents only a small portion of MDS, of which, the clinical significance and outcomes are not well understood. Both laboratory and clinical evidence suggests that h-MDS shares immune-mediated pathogenic mechanisms similar to acquired aplastic anemia. As a result, immunosuppressive therapy (IST) has been the mainstay treatment. Since h-MDS is poorly understood and reported response rates to IST vary, we retrospectively reviewed the outcome of our patients with hypocellular MDS and identified the effectiveness of alternative therapy using hypomethylating agents (HMA). Methods: AllMDS patients over a 13-year period (June 1997 to May 2010) were captured after an IRB approval was obtained. Patients' demographics, labs, bone marrow aspirates and biopsies as well as cytogenetic data were collected. We used the currently accepted age-adjusted criteria to define hypocellularity as ≤ 30% in patients <70 years old, and ≤20% in >70 years old. Only patients with available response data were included in the final analysis. Survival estimates were calculated using Kaplan-Meier curves. Results: We identified 65 (8%) h-MDS patients from a cohort of 827 adult MDS. Median age was 68 years (range, 20-90), with 69% males. The largest group of patients were RAEB-1 at 9 (14%) and RAEB-2 at 14 (22%) along with RCMD at 18 (28%) and 15 (23%) MDS-U, 2 (3%) MDS 5q, 4 (6%) RARS and 2 (3%) were considered atypical. Cytogenetic analysis was normal in 17 (26%) with 7 (11%) having trisomy 8, and 11 (16%) and 14 (22%) having abnormalities of chromosome 7 and 5, respectively. IPSS-R were very low, low, intermediate, high, and very high in 4%, 30%, 28%, 21% and 17%, respectively. Leukemic transformation (LT) occurred in 6 (9%). The 4 year overall survival and progression free survival was 35% and 34%, respectively. Median overall survival and progression free survival was 30.2 and 29.7 months, respectively. Nine out 65 h-MDS patients were treated with a HMA, azacitidine (4) or decitabine (5). The median age of patients was 71 years (range, 66-85) with 6 (66%) being males. The median cellularity of the diagnostic bone marrow was 15% with the majority of patients presenting with RAEB-1 (11%) and RAEB-2 (55%) classification and 2 (22%) cases of RCMD as well as 1 (11%) case of MDS with isolated 5q. Cytogenetic analysis was normal in 3 (33%). IPSS-R were low in 1 (11%), intermediate in 4 (44%), high in 1 (11%) and very high risk in 3 (33%). The median time to initiation of a HMA from diagnosis was 5 months with response assessed at a median of 4.2 months. Two patients had received prior therapy including lenalidomide (for MDS 5q) and one had 1 cycle of trial therapy. Two patients (22%) achieved a marrow complete remission at a median time of 2.95 months; with one case of relapse at 9.8 months using HMA. Three patients (33%) had LT at a median time of 7.9 months post initiation of HMA. Median overall survival of patients with LT was 20 days. A total of 4 (44%) patients remain alive with a leukemia free survival of 67% at 2 years. In comparing to h-MDS patients without HMA treatment versus HMA treated, median OS was 28.2 vs 50.9 months (p=0.41) and PFS was 28.1 vs 40.6 months (p=0.69), respectively. Conclusions: The results from our clinical experience confirm that hypocellular MDS is a rare variant and that it is distinct from normocellular/hypercellular MDS. Two thirds of h-MDS received HMA for the treatment of higher risk disease (RAEB1-2). HMA has some efficacy in h-MDS. Further studies are warranted to better understand the clinical significance and outcomes of hypocellular MDS. Disclosures Binder: American Society of Hematology: Research Funding. Al-Kali:Celgene: Research Funding.

Abstract Background: Adult T-cell acute lymphoblastic leukemia / lymphoma (T-ALL) is a rare and aggressive hematological malignancy. Unlike B-cell ALL, apart from age and presenting WBC, additional prognostic factors remain largely unknown. Additionally, the prognostic impact of a monosomal karyotype, which has a negative impact on survival in myeloid neoplasms, remains to be determined in T-ALL (Kenderian et al, BCJ 2013) Aim: To study predictors of survival including the role of a monosomal karyotype in T-ALL. Methods: After due IRB approval, adults diagnosed with T-ALL from 1990-2014 at Mayo Clinic Rochester were identified. All clinical and pathologic data including FISH studies and cytogenetics were retrospectively reviewed. Chi-square test was used to compare variables. Survival was estimated and compared using the Kaplan-Meier Method and log-rank test. Univariate and multivariate analysis was performed using the Cox regression model. Results: Between 1990 and 2014, a total of 92 consecutive patients (pts) with T-ALL were identified, 41 had acute lymphoblastic lymphoma (LBL), and 51 (55%) acute lymphoblastic leukemia (ALL). Median age at diagnosis was 33 years (range; 18-88 years) with 72%males. Median follow up of our cohort was 25 months (range; 0.9-260 months) during which time 42 deaths (45%) were documented. A. Baseline characteristics: Fifty six (67%) had constitutional symptoms, and 63 (77%) had lymphadenopathy at diagnosis. Mediastinal mass was identified in 49 (55%) pts with palpable splenomegaly in 16 (20%) pts. Fifty one (58%) pts had extranodal disease, and 17 (19%) pts had CNS involvement. Pleural and pericardial effusion were evident in 16 (19%), and 6 (7%) pts, respectively. Median laboratory values included; LDH 288 U/L (115-6590), hemoglobin (Hb) 12.8 g/dl (4.5-18.2), WBC 8.65 x109 (1.4-406), platelets 169 x109 (8-485), bone marrow blasts 29% (0-99). B. Treatments: Induction therapies were variable and included; 11 (12%) CHOP like regimens, 7 (8%) modified NHL regimens (Asparginase+CHOP), 54 (58%) ALL like regimens (CALGB 9111, E2993, or HyperCVAD), 13 (14%) Pediatric regimens (ALL 10403, CCG 1961, BFM, Augmented BFM, GRAALL 2003, or Hoelzer regimen), with 7 (8%) patients' receiving palliative care only. Seventy-one (77%) pts achieved complete remission (CR1) during induction therapy while eight pts had refractory disease (8%). Thirty-three (35%) pts underwent stem cell transplant; 7 (21%) autologous and 26 (79%) allogeneic. Forty one (44%) pts had disease relapse and median time to relapse was 10.8 months (range; 8.2-16 months) C. Predictors of survival: In univariate analysis, that included age > 60 years, leukocyte count Hb, Plt, Blast percentage, mediastinal mass, LDH and CNS involvement, only age >60 and CNS involvement were predictors of inferior survival and both retained significance in the multivariable analysis with HR 3.9 (CI 1.5 - 9; p=0.0055) and 2.8 (1.3 - 5.7; p =0.0085), respectively (Figure 1 and 2). In this cohort age >60 was a better stratification that the conventionally used age above 40 (p = 0.1). Among 56 pts with evaluable cytogenetics data, 7 (13%) pts had MK which did not have an impact on survival with HR 1.9 (CI 0.4-5.6; p=0.33). Conclusion: In this large cohort of adult T-ALL patients, we observed that age >60 years and CNS involvement were independent predictors of inferior survival. MK was infrequent, but unlike in myeloid neoplasms, does not seem to impact overall survival. Figure 1. Figure 1. Figure 2. Figure 2. Disclosures Al-Kali: Celgene: Research Funding. Thompson:Kite Pharma: Research Funding. Witzig:Novartis: Research Funding; Celgene: Research Funding; Amgen: Honoraria; Spectrum: Research Funding; Valeant Pharma: Equity Ownership.

Abstract Background The impact of JAK inhibitor (JAKi) therapy on the natural history of myelofibrosis remains unclear. While a Phase 3 study showed that overall survival of ruxolitinib-treated intermediate-2 or high-risk myelofibrosis patients was superior as compared to placebo-treated patients, it remains unclear whether such an advantage would persist in patients receiving standard therapy, and whether this result is generalizable to other JAK inhibitors. We conducted a sponsor-independent, single-center retrospective analysis of overall and leukemia-free survival after long-term follow up in patients receiving JAKi treatment with comparison to two control populations: (i) a contemporaneously diagnosed (2005 onwards) primary myelofibrosis (PMF) patient cohort receiving standard therapy (JAKi and pomalidomide naïve); and (ii) a pomalidomide treated patient cohort (JAKi-naïve). Methods Baseline clinical, cytogenetic and molecular data were collected at the onset of study treatment with the first JAKi or pomalidomide, or at the time of first referral for patients receiving standard therapy. Information regarding bone marrow karyotype and DIPSS-plus risk status was available for all patients. Information on survival and leukemic transformation was updated in July 2013. Results The 3 patient groups were: (i) JAKi-treated: n=157, median age 65 years, range 34-89. DIPSS-plus risk: Low/Int-1 20 (13%), Int-2 66 (42%), and High 71 (45%). Median follow up from first JAKi treatment was 30 months (range 1-67 months); during this period, 72 deaths (46%) and 14 (9%) leukemic transformations were documented. The first JAKi was ruxolitinib in 51 patients (33%), momelotinib (CYT387) in 79 (50%), and fedratinib (SAR302503) in 27 (17%). The proportion of patient deaths and leukemic transformation (median follow up, range) per specific JAKi was: ruxolitinib 57% and 12% (40 months, 2-67), momelotinib 43% and 9% (29 months, 1-42), and fedratinib 33% and 4% (21 months, 2-53), respectively. (ii) Control#1 (Pomalidomide-treated): n=69, median age 67 years, range 36-87. DIPSS-plus risk: Low/Int-1 0 (0%), Int-2 28 (41%), and High 41 (59%). Median follow up was 25 months (range 2-70 months); during this period, 47 deaths (68%) and 7 (10%) leukemic transformations were documented. (iii) Control# 2 (Standard therapy): n=319, median age 66 years, range 22-90. DIPSS-plus risk: Low/Int-1 98 (31%), Int-2 122 (38%), and High 99 (31%). Median follow up from time of referral was 18 months (range 1-92 months); during this period, 101 deaths (32%) and 18 (6%) leukemic transformations were documented. For patients with Int-2 or high-risk disease (n=221), 92 deaths (42%) and 13 (6%) leukemic transformations were recorded (median follow up 16 months, range 1-92). The survival analysis focused on DIPSS-plus Int-2 and high-risk patients for all groups, reflecting the population predominantly selected for treatment on JAKi clinical trials. There was no significant difference in the median overall survival of myelofibrosis patients treated with JAKi (39 months) versus standard therapy (35 months) versus pomalidomide (26 months) (p=0.12) (Figure). The 5-year survival rates were 31%, 33% and 26%, respectively. Similarly, the median overall survival of patients treated with ruxolitinib (35 months), momelotinib (37 months) or fedratinib (42 months) was not significantly different (p=0.4). Baseline DIPSS-plus status (ie, Int-2 vs. high-risk) effectively stratified JAKi treated patients in terms of overall survival (median survival 31 vs. 60 months; p=0.0002). There was no significant difference in leukemia-free survival when comparing JAKi-treated patients versus standard therapy verus pomalidomide-treated patients (p=0.6) (Figure). The estimated 5-year risk of leukemic transformation was 17%, 12% and 13%, respectively. Conclusions With long-term follow up, myelofibrosis patients receiving JAKi therapy exhibited similar overall and leukemia-free survival as compared to two control populations (DIPSS-plus matched standard therapy and pomalidomide-treated patients). Baseline DIPSS-plus status was a powerful predictor of overall survival in JAKi treated patients. While underscoring the limitation of a non-head-to-head retrospective comparison, there was no significant difference in overall or leukemia-free survival when comparing individual JAKi drugs. Disclosures: Pardanani: Bristol Myers Squibb: Clinical trial support Other; Sanofi: Clinical trial support, Publication support services, Clinical trial support, Publication support services Other; PharmaMar: Clinical trial support, Clinical trial support Other; JW Pharmaceutical Corp.: Clinical trial support, Clinical trial support Other. Off Label Use: Use of Ruxolitinib, Momelotinib, Fedratinib and Pomalidomide for treatment of Myelofibrosis in Clinical Trial setting.

Abstract Background: Acute T-cell lymphoblastic leukemia (T-ALL)/ lymphoma (T-LBL) are immature T-cell malignancies with very poor outcomes. In patients with relapsed disease, survival is often < 12 months (Marks et al, Blood, 2009). Prognostic correlates and treatment outcomes after relapse remain largely unknown. Aims: To evaluate prognostic correlates and treatment outcomes in patients with relapsed T-ALL/LBL. Methods: After IRB approval, 92 consecutive patients with T-cell ALL/LBL were identified. 41 patients with relapsed disease after achieving a first complete remission (CR1) were included in the study group. Features at the time of relapse were retrospectively abstracted and analyzed. Survival time after relapse was calculated from the time of relapse until death or last follow-up. Conventional methods were used for statistical analysis. Results: A. Baseline characteristics of all patients 92 patients were identified [41 LBL, 51 ALL]. Median age was 33 years (range; 18-88) and 66 (72%) were males. Median follow-up was 25 months (range 0.9-260). 13(14%) received palliative care or were lost to follow-up before treatment initiation. At last follow-up, there were 41 (52%, n = 79) relapses and 42 (46%, n = 92) deaths. Of 79 patients, 71 (90%) achieved a CR1 and 8 (10%) had primary refractory disease. Comparisons between ALL and LBL revealed expected differences in presenting features such as frequency of primary mediastinal mass at presentation in LBL and high leukocyte count in ALL. Significant differences in treatment were seen. Asparaginase-based and ALL-directed therapy were more common in the ALL group (71% vs. 45%, p = 0.009; 88% vs. 71%, p = 0.04). 13 (81%) LBL and 14 (27%) ALL underwent allogeneic SCT. The remaining three in each group underwent autologous SCT in CR1. B. Characteristics of relapsed patients 41 relapsed after achieving CR1 [23 LBL, 18 ALL]. Median time to relapse was 8.0 months (range 2.3-130), with no difference between ALL and LBL groups. Median age was 34 years (range 19-71) and 29 (71%) were males. Median follow-up after relapse was 7.2 months (range 0.3-207). At time of last follow-up, there were 30 (73%) deaths, 25 (61%) achieved a second complete remission (CR2) and 14 (34%) had persistent disease. 12 (48%) had relapse following CR2. 34 (83%) relapsed following CR1 after induction chemotherapy alone. 6 (15%) and 1 (2%) relapsed following allogeneic and autologous SCT respectively. 27 (66%) had BM involvement, 5 (12%) had CNS involvement, 13 (32%) had mediastinal involvement, and 3 (7%) had isolated extramedullary relapse. Ten (43%) patients with LBL relapsed with ALL. Of 27 patients with available data, 9 (33%) had high-risk cytogenetics at relapse, defined as ≥ 5 chromosomal abnormalities and/or markers known to confer an unfavorable prognosis. 16 of these 27 patients had follow-up cytogenetic studies performed, of which 8 (50%) had evidence of clonal evolution. Patients were treated with up to five regimens. 12 (31%) received asparaginase-based therapy and 13 (33%) received nelarabine with CR2 rates of 75% and 25% respectively. Median number of regimens needed to achieve CR2 was one. 18 (46%) went onto SCT following CR2: 14 (allogeneic), 3 (autologous), and 1 (donor lymphocyte infusion). The majority received myeloablative conditioning (92%). C. Outcomes and prognostic correlates for survival after relapse Median survival following relapse was 8.2 months (IQR 4.1-21.3) and did not differ between LBL and ALL. Nine (36%) who achieved CR2 are alive at time of last follow-up. 25 (61%) achieved CR2, but 12 (48%) relapsed following CR2, with median time to second relapse of 7.6 months (IQR 4.7-14). Nine of 18 who underwent SCT following CR2 had disease relapse (8 allogeneic, 1 autologous). Only BM involvement at relapse significantly correlated with post-relapse survival (HR 2.4 [1.1, 5.1], p = 0.03). Clonal evolution, high-risk cytogenetics, nelarabine therapy, achieving CR2, and SCT were not independent predictors of survival in our small subset of patients. Conclusion: Patients with relapsed T-ALL/LBL have dismal outcomes, in spite of advances in therapy with newer agents such as nelarabine and the use of allogeneic SCT. Regardless of T-ALL/LBL subtype, BM involvement at relapse appears to be a significant factor negatively impacting post-relapse survival. Further studies are needed to validate these findings. Disclosures Al-Kali: Celgene: Research Funding. Thompson:Kite Pharma: Research Funding.

Abstract Background JAK inhibitors (JAKi) currently represent a frontline treatment modality in patients with symptomatic intermediate- or high-risk myelofibrosis (MF). While the efficacy of JAKi in decreasing spleen size and improving disease-related symptoms is well documented, the effect of this drug class on overall survival and risk of leukemic transformation in MF patients is less well understood, primarily due to short follow up in individual patient cohorts. Furthermore, there is scant information on baseline predictors of overall and leukemia-free survival, and also spleen and/or anemia responses, at the onset of JAKi therapy. We conducted a sponsor-independent single-center retrospective review of patients treated with JAKi with long follow up to address the aforementioned issues. Here, we report our findings regarding baseline predictors of clinical outcome and treatment response in MF patients treated with JAKi. Methods Clinical, cytogenetic and molecular data were retrospectively collected on patients treated with one or more JAKi on a clinical trial at the Mayo Clinic, Rochester, MN. Baseline clinical data pertained to the onset of treatment with the first JAKi. Karyotype information was available for all patients. Screening for JAK2V617F, MPLW515, ASXL1, and SRSF2 mutations was performed as previously described (Leukemia. 2013 Apr 26. doi: 10.1038/leu.2013.119). Spleen size was measured by palpation by clinical investigators. Anemia and spleen responses were adjudicated as per IWG criteria (Tefferi, Blood, 2006). Information on survival and leukemic transformation was updated in July 2013. Results A total of 157 MF patients were studied (60% male; median age 65 years, range 34-89 years). Ninety seven patients (62%) had primary MF, 42 (27%) post-polycythemia vera MF, and 18 (11%) post-essential thrombocythemia MF. The DIPSS-plus distribution was: Low/Int-1 20 (13%), Int-2 66 (42%), and High 71 (45%). Other characteristics were: JAK2V617F positive 78%, unfavorable karyotype 17%, hemoglobin <10 g/dL 58%, leukocyte count >25 x 109/L 31%, platelet count <100 x 109/L 15%, peripheral blasts ≥1% 69%, red blood cell transfusion-dependent 37%, and constitutional symptoms 56%. Forty one patients (38%) were ASXL1 mutated (n=108) and 20 (18%) SRSF2 mutated (n=110). One hundred and thirty nine patients (89%) were evaluable for spleen response (median spleen size by palpation 19 cm, range 6-32 cm) and 91 (58%) for anemia response by IWG criteria. The first JAKi was momelotinib (CYT387) in 79 patients (50%), ruxolitinib in 51 (33%) and fedratinib (SAR302503) in 27 (17%). Median follow up from first JAKi treatment was 30 months (range 1-67 months); during this period, 72 deaths (46%) and 14 (9%) leukemic transformations have been documented. The median overall survival of the entire cohort was 41 months. Multivariable analysis identified the following DIPSS-plus independent predictors for survival: palpable spleen size >20 cm (HR 2.5), ASXL1 mutated status (HR 3.1) and SRSF2 mutated status (HR 3.8) (all p<0.01) (Figure). In contrast, there was no difference in survival with regards to the first JAKi used for MF treatment (p=0.3). Multivariable analysis identified the following predictors for leukemia-free survival: SRSF2 mutated (HR 9.7), unfavorable karyotype (HR 8.7), and palpable spleen size >20 cm (HR 9.1) (all p<0.01). Predictors of spleen response included baseline spleen size (≤20 cm 56%, >20 cm 34%; p=0.01), with borderline significance for ASXL1 mutation status (unmutated 51%, mutated 31%; p=0.06). Predictors of anemia response included RBC transfusion status at baseline (transfusion dependent 38%, transfusion independent 12%; p<0.01) and JAK2V617F status (unmutated 46%, mutated 22%; p=0.02). Conclusions Larger baseline spleen size and ASXL1 and SRSF2 mutated status predicted for inferior overall survival, independent of DIPSS-plus, at the onset of first JAKi treatment in MF patients. Similarly, multivariable analysis identified larger baseline spleen size and presence of unfavorable karyotype and SRSF2 mutations as being predictive for inferior leukemia-free survival. Spleen response rates were significantly superior in patients with smaller baseline spleen size and possibly ASXL1 unmutated status. Anemia response rates were significantly higher in patients who were transfusion-dependent at baseline and had JAK2V617F unmutated status. Disclosures: Pardanani: Bristol Myers Squibb: Clinical trial support Other; Sanofi: Clinical trial support, Publication support services, Clinical trial support, Publication support services Other; PharmaMar: Clinical trial support, Clinical trial support Other; JW Pharmaceutical Corp.: Clinical trial support, Clinical trial support Other. Off Label Use: Use of Ruxolitinib, Momelotinib, Fedratinib and Pomalidomide for treatment of Myelofibrosis in Clinical Trial setting.

Abstract Introduction Steroid refractory chronic graft-versus-host-disease (SR-cGVHD) is the leading cause of late morbidity and mortality in allogeneic transplant patients. Safety of extracorporeal photophoresis (ECP) has led to multiple studies evaluating its role in SR-cGVHD. Overall and organ specific response rates (RR) of SR-cGVHD have varied significantly between studies. We conducted a systemic review and meta-analysis to assess the efficacy of ECP in SR-cGVHD. Methods A comprehensive search of several databases (Ovid Medline In-Process & Other Non-Indexed Citations, MEDLINE, EMBASE, Cochrane Central register of Controlled Trials,, Scopus) for articles published in any language between 1984-2012 was conducted. Unpublished studies using electronic databases from the annual meetings abstracts of hematology/transplant societies were searched. Experts in the field were also contacted for selection of studies and missing data from selected trials. End points included complete response rate (CR), overall RR (ORR), and organ-specific RR's. Results Initial search generated 312 studies, out of which 18 studies (4 prospective) met selection criteria (N=595). Random effects model was used for pooled rates. Pooled CR rate (11 studies) was 29% (CI 19%-42%). Pooled ORR (16 studies) was 64% (CI 65-82%). 1 year overall survival was 49% (CI 29-70%). There was significant heterogeneity between studies (I²=72) likely due to differences in ECP schedules utilized, and a variety of scales used to measure outcomes. Pooled RR's for SR-cGVHD of skin was 74% (CI 60-85%), liver was 68% (CI 57-77%), ocular was 60% (CI 40-78%), oral was 72% (CI 51-86%), lung was 48% (CI 33-63%), gastrointestinal (GI) was 53% (CI 21-83%), and musculoskeletal (MSK) was 64% (CI 18-94%). No significant differences in responses to ECP for pediatric and adult populations were found. Sensitivity analysis could not be done due to limited number of prospective studies. Conclusions ECP is an effective therapy for oral, skin & liver SR-cGVHD, with modest activity in lung and GI SR-cGVHD. Further studies must report standardized outcomes with a uniform schedule of ECP. Disclosures: No relevant conflicts of interest to declare.

Abstract No laboratory test can predict non-relapse mortality (NRM) after hematopoietic cellular transplantation (HCT) prior to the onset graft-versus-host disease (GVHD). Recently, we have shown that a signature of three GVHD plasma biomarkers (TNFR1, ST2, and REG3α) can predict response to GVHD therapy and NRM at the onset of clinical GVHD (Levine, Lancet Haem, 2015). Our goal in the current study was to identify a blood biomarker signature that could predict lethal GVHD and six-month NRM well in advance of the onset of GVHD symptoms. Patient samples on day +7 after HCT were obtained from 1,287 patients from 11 HCT centers in the Mount Sinai Acute GVHD International Consortium (MAGIC). Samples from two large centers (n = 929) were combined and randomly assigned to a training set (n = 620) and test set (n = 309). 358 patients from nine others centers constituted an independent validation set. The overall cumulative incidences of 6-month NRM were 11%, 12%, and 13% for the training, test, and validation sets respectively. The incidence of lethal GVHD, defined as death without preceding relapse while under steroid treatment for acute GVHD, were 18%, 24%, and 14% in the same groups, respectively. The median day of GVHD onset was 28 days in the training set and 29 days in the test and validation sets. We measured four GVHD related biomarkers [ST2, REG3α, TNFR1, and IL2Rα] in all samples and used the training set alone to develop competing risks regression models that used all 13 possible combinations of one to four biomarkers to predict 6-month NRM. The best algorithm, which we rigorously confirmed through Monte Carlo cross-validation of 75 different combinations of training sets, included ST2 and REG3α. No combination of one, three, or four biomarkers was superior to the combination of these two biomarkers. The day 7 algorithm identified high risk (HR) and low risk (LR) groups with 6-month NRMs of 28% and 7%, respectively (p<0.001) (Fig 1A). The relapse rates did not differ between risk groups so that overall survival (OS) was 60% for HR and 84% for LR (p<0.001) (Fig 1B). When applied to the test set (Fig 1C/D), the algorithm identified 54/309 (17%) of the patients as HR with an NRM of 33% vs 7% for LR patients (p<0.001) and 6-month OS of 57% and 81% for HR and LR patients, respectively (p<0.001). In the independent validation set (Fig 1 E/F), the algorithm identified 72/358 (20%) of the patients as HR with an NRM of 26% vs 10% for LR patients (p<0.001) and OS of 68% and 85% for HR and LR patients, respectively (p<0.001). High risk patients were three times more likely to die from GVHD than LR patients in each cohort (p<0.001) (Fig 2). The GI tract is the GVHD target organ that is most resistant to treatment and represents a major cause of NRM, and we observed twice as much severe GI GVHD (stage 3 or 4) in HR patients as in LR patients (p<0.001, data not shown). The algorithm successfully separated HR and LR strata for 6 month NRM in several groups with differing risks for GVHD and NRM, including donor type, degree. of HLA-match, age group, and conditioning regimen intensity (Fig 3). In conclusion, we have developed a blood biomarker algorithm that predicts the development of lethal GVHD seven days after HCT, which performed successfully in large multicenter validation sets. The GVH reaction is already in progress by day +7, even though clinical symptoms may not occur until days or weeks later. We speculate that the blood biomarker concentrations at this early time point reflect subclinical GI pathology, a notion that is reinforced by the fact that ST2 and REG3α, the two biomarkers in the algorithm, are closely associated with GI GVHD. The algorithm identified HR and LR strata in several patient groups with different overall risk for lethal GVHD (donor, HLA match, conditioning regimen intensity, age). This day +7 algorithm should prove useful in clinical BMT research by identifying patients at high risk for lethal GVHD who might benefit from aggressive preemptive treatment strategies. Disclosures Chen: Novartis: Research Funding; Incyte Corporation: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding. Jagasia:Therakos: Consultancy. Kitko:Therakos: Honoraria, Speakers Bureau. Kroeger:Novartis: Honoraria, Research Funding. Levine:Viracor: Patents & Royalties: GVHD biomarkers patent. Ferrara:Viracor: Patents & Royalties: GVHD biomarkers patent.

Abstract Background: Molecular profiling has revolutionized the field of hematology, and in particular myeloid neoplasms. NPM1+/FLT3- mutational status has been shown to have a favorable prognostic impact in patients with acute myeloid leukemia (AML). However, little is known about its interaction with age. Methods: We retrospectively analyzed the overall survival (OS) in newly diagnosed AML patients, whose FLT3 and NPM1 data were available, and were treated with cytotoxic chemotherapy at our institution between 2003 and 2013. Patients 18 to 60 years of age were categorized as adult, and those above 60 years were classified as elderly. Each of the elderly or adult groups were further divided according to the presence or absence of combined NPM1+/FLT3- mutational status. Estimated probabilities of overall survival were calculated using the Kaplan-Meier method, with the log-rank test to compare groups. Demographics and continuous variables were compared using the Wilcoxon rank sum test. Appropriate IRB approval was obtained. Results: Out of 835 patients with AML, 86 patients with both known FLT3/NPM1 mutation status and receiving induction chemotherapy were found. NPM1+/FLT3- mutational status was present in 11 elderly (median age 68 yr, hemoglobin (Hb) 8.7 g/dL, white blood cells (WBC) 9.6 x109/L, platelets (PLT) 100 x109/L, peripheral blood blasts percentage (PB%) 31), and in 11 adult patients (median age 49 years, Hb 9.1 g/dL, WBC 14 x109/L, PLT 67 x109/L, and PB% 21). Twenty-eight elderly patients (median age 66 yr, Hb 9.4 g/dl, WBC 15.7 x109/L, PLT 66 x109/L, and PB% 33), and 36 adult patients (median age 49 years, Hb 9.1 g/dL, WBC 14 x109/L, PLT 67 x109/L, and PB% 21) were carrying genotypes other than NPM1+/FlT3-. There was no statistical significant difference in the median age, Hb, WBC, PLT, and PB% at the time of diagnosis. Out of 86 patients, 79 had intermediate, 6 had adverse, and 1 had favorable cytogenetic risk. All elderly and adult patients with NPM1+/FLT3- genotype had intermediate risk cytogenetics. When treated with cytotoxic chemotherapy, elderly patients with NPM1+/FLT3- genotype demonstrated a significantly better overall survival compared to elderly patients without this genotypes (p=0.015), and interestingly also to adult AML carrying other genotypes (p=0.028). Surprisingly, there was not an improved survival in adults with NPM1+/FLT3- genotype compared to adults carrying other genotypes (p=0.14), but this may have been related to the small numbers in each group. No significant difference in the overall survival was observed between elderly and adult patients who were carrying NPM1+/FLT3- mutational status (p=0.4). The estimated 5-year survival rates for elderly with and without NPM1+/FLT3- status were 71% vs. 14%, and for adults with and without this genotype were 49% vs. 19%, respectively. Conclusion: Age does not have an impact on the OS in AML patients with NPM1+/FLT3- mutational status, arguing strongly for intense chemotherapy in this group. Elderly AML patients with NPM1+/FLT3- genotype have a superior OS compared to both adult and elderly patients carrying other genotypes, when treated with cytotoxic chemotherapy. Further validation in large prospective studies is warranted. Figure 1 Figure 1. Figure 2 Figure 2. Disclosures No relevant conflicts of interest to declare.

Abstract Background: Myelodysplastic syndrome (MDS) is a heterogeneous disease characterized by cytopenias and a propensity to progress into acute myeloid leukemia in a subset of patients. Clinical trials which led to FDA approval of hypomethylating agents (HMAs) stratified patient's risk status per the IPSS. In 2008, WHO further categorized patients into different groups and the IPSS was revised into the IPSS-R. Therefore, there is a paucity of data regarding the response rates and clinical outcomes of HMA therapy when stratified per WHO subtype and IPSS-R. Goal: We carried this review to assess whether WHO subtypes or IPSS-R had a differential response with corresponding impact on survival. Methods: After IRB approval, all MDS pts who received hypomethylating agents (HMA) (azacitidine or decitabine) with available response data (2000-2014) were included. Clinical characteristics were obtained retrospectively. IPSS-R (Greenberg 2012) was used for risk stratification. Response assessment was based on IWGR 2006 criteria. Statistical comparisons were performed with the χ2 test for categorical variables and the Wilcoxon test for continuous variables. Survival analysis was based on the Kaplan-Meier method and log rank test. For all analyses, P ≤ 0.05 was considered statistically significant. Results: We identified 823 pt in our MDS database; among those that received HMA therapy 70 patients (pts) had evaluable response data. The median age at diagnosis was 67 years (31-85) and 49 (70%) were males. MDS subtypes per WHO 2008 classification were 31(44%) RAEB1/2, 26(37%) RCMD, 10(14%) MDS-U, 2(3%) RARS, and 1(1%) MDS 5q. Fifteen (33%) patients had secondary MDS. 23(32%) developed AML, and median time to leukemic transformation was 14 months (7.8-21). Median time to start HMA was 2.7 months (0-77). Median follow up from initial diagnosis was 20 months (3-118) at which time 22(31%) deaths were documented. A. Patient characteristics: Prior to HMA therapy median laboratory values included; Hemoglobin 9.6 g/dL (6.6-14.8), ANC 1.1 x1010 (0.01-16), Platelet 66 x1010 (4-937), peripheral blast % 0 (0-18), and bone marrow blast % 5 (0-17). IPSS-R for our cohort was as the following, 15(27%) Very high, 14(25%) high, 10(18%) intermediate 12(21%) low, and 5(9%) very low risk. B. HMA Response rates based on WHO subtype: We observed no significant difference in response (p = 0.6) when comparing response rate among different groups. Response rates per WHO subtype were as follows: i. RAEB: 7 (22%) pts achieved CR/Marrow CR, 1(3%) had HI/PR, 17(55%), and 6(20%) had stable disease (SD) and progressive disease (PD), respectively. ii. RCMD: 2 (8%) pts achieved CR/Marrow CR, 4(16%) had HI/PR, 12(46%), and 8(30%) had SD and PD, respectively. iii. MDS-U: 1 (10%) pt achieved CR/Marrow CR, 1(10%) had HI/PR, 6(60%), and 2(20%) had SD and PD, respectively. iv. MDS 5q/RARS: 1(33%) had SD, and the remaining three pts had PD. C. HMA Response rates based on IPSS-R: i. Very low: 1(20%) had HI/PR, 3(60%), and 1(20%) had SD. ii. Low: 2(17%) pts achieved CR/Marrow CR, 3(25%) had HI/PR, 4(33%), and 3(25%) had SD and PD, respectively. iii. Intermediate: 2 (20%) pts achieved CR/Marrow CR, 5(50%), and 3(30%) had SD and PD, respectively. iv. High: 2(14%) pts achieved CR/Marrow CR, 2(14%) had HI/PR, 7(50%), and 3(21%) had SD and PD, respectively. v. Very high: 1(7%) pts achieved CR/Marrow CR, 9(60%), and 5(33%) had SD and PD, respectively. Comparing responses among IPSS-R risk stratification groups, was not statistically significant (p = 0.49). We did not observe a difference in response rate when we further classified pts into low risk (very low, low & intermediate), and high risk (high, very high) (p = 0.27). D. Response effect on Survival: Overall survival was not statistically different among responders (CR/Marrow CR/HI &PR) and non-responder (SD & PD); 22.3 vs 14.7 months respectively (p = 0.21) (Figure1). Conclusion: Using contemporary MDS risk stratification and WHO classification was not predictive of response to hypomethylating agents. Responders did not seem to benefit from prolongation of survival. Given the small sample size in this cohort, this should be further explored in larger groups to identify if there are any potential subgroups that would benefit more from such therapy. Figure 1. Figure 1. Disclosures Al-Kali: Celgene: Research Funding.

Abstract Background: Acute erythroleukemia (AML-M6) is an aggressive myeloid leukemia, with poor outcomes (median survival < 12months). In 2008, the WHO further sub-classified this disease into two subgroups- erythroid leukemia and pure erythroid leukemia, based on the number of erythroblasts and myeloblasts. We carried out this study to evaluate differences in presenting features and outcomes between erythroid leukemia and pure erythroid leukemia. Methods: After due IRB approval, out of 923 AML, 43(4.6%) pts who met the 2008 WHO criteria for acute erythroleukemia (2000-2014) were identified, and clinical characteristics retrospectively reviewed. A diagnosis of erythroid leukemia (M6a) was considered if > 50% of the total nucleate BM cells were of erythroid lineage and > 20% of non-erythroid cells were myeloblasts. Pure erythroid leukemia (M6b) was diagnosed in pts who presented with > 80% of BM nucleate cells consisting of immature erythroid precursors. Cytogenetic risk stratification was based on NCCN AML guidelines. Statistical comparisons were performed with the χ2 test for categorical variables and the Wilcoxon test for continuous variables. Univariate and multivariate analyses were performed with the Cox logistical regression. Survival analysis was based on the Kaplan-Meier method and log rank test. For all analyses, P ≤ 0.05 was considered statistically significant. Results: The median age at diagnosis was 65 years (27-90) and 34(79%) were male. 30 (69%) had M6a and 13 (30%) had M6b. 27 (62%) had de novo AML M6, whereas 16 (37%) had secondary AML (9 underlying MDS and 7 therapy related).The median time to leukemic transformation for pts with underlying MDS was 8 months (2-83). Median follow up in our cohort was 7.2 months (2-157) at which time 13 (30%) deaths were documented. A. Patient characteristics: At time of diagnosis median laboratory values included; Hemoglobin 8.5 g/dL (4.6-11), WBC 2x109 (0.7-11.1), ANC 0.47x109 (0.01-4.07), Platelet 33x 109 (7-177), peripheral blast % 2 (0-35), and bone marrow blast % 11 (0-84). Cytogenetic information was available in 38 pts; of which 6(16%) were diploid, 21(55%) were complex/monosmal and 1(3%) was complex/non monosomal. Intermediate risk cytogenetic was observed in 15 pts (39%), and poor risk cytogenetic was identified in 23(61%). Comparing M6a and M6b, we observed that pts with M6a tend to have more leukopenia and neutropenia and higher marrow blast percentage compared to M6b (Table 1). None of the patients experienced extramedullary disease or CNS involvement. B. Outcome and prognostics: We did observe a trend towards worse survival in pts with M6b, 7 months vs 12.9 months with M6a (p= 0.08) (Figure 1). There was no difference in survival in acute erythroleukemia compared to other acute myeloid leukemias, 10.1 vs 14 months (p= 0.28), respectively (Figure 2). In univariate analysis, WBC, ANC, and cytogenetic risk impacted survival (Table 2), however only cytogenetic risks retained significance in the multivariate model; Intermediate risk HR 0.18 (0.05-0.51; p = 0.0009) and poor risk HR 5.4 (1.9-17.6; p = 0.0009). Conclusion: Acute Erythroid Leukemia is a rare leukemia with ~5% incidence in our cohort. We observed that regardless of the percentage of bone marrow erythroblasts and myeloblasts, survival in patients with Acute Erythroleukemia are predominantly impacted by cytogenetic risk stratification. This raises a question on the clinical relevance of the sub-categorization of acute erythroleukemia into erythroid leukemia and pure erythroid leukemia. Larger studies are needed to validate our results. Figure 1. Figure 1. Disclosures Al-Kali: Celgene: Research Funding.