Komrokji, Rami S.

Abstract Introduction: CMML is a heterogeneous malignancy characterized by peripheral monocytosis and a propensity for AML transformation. Given the low incidence of CMML and its broad range complexity, a large disease-specific data set is necessary to discern independent co-variates predictive of disease behavior. Several prognostic models derived from smaller data sets have been developed to stratify CMML patients into distinct groups that are predictive for OS. However, the validity of these models in a large international data set has never been investigated. Here, we present the largest International CMML data base established to date and report key baseline characteristics predictive for disease outcome. We externally validate and perform a detailed statistical comparison between the International Prognostic Scoring System (IPSS), Revised (R)-IPSS, Global MD Anderson Scoring System (MDASC), MD Anderson Prognostic Score (MDAPS), Dusseldorf Score (DS), Mayo, and Spanish Scoring Systems (CPSS). Methods: Data were abstracted by each contributing institution and deposited for central data review at the Moffitt Cancer Center. Manual review of cases was performed to ensure data quality prior to analysis. Only WHO defined CMML was included. The primary objective of this study was to establish an international CMML data set and validate the above models calculated at the time of presentation to each center. All prognostic models were calculated as previously described. The Kaplan–Meier (KM) method was used to estimate median OS and the log rank test was used to compare KM survival estimates with SPSS version 21.0. Random Forest Survival (RSF) and ROC analysis was done with R. Results: Between July 1981 and June 2014, 1832 CMML patients were captured in the International CMML database. Median age at diagnosis was 70 (16-93) years with a male (67%) predominance. By the WHO classification, the majority of patients had CMML-1 (79.9% vs. 20.1%) and most patients were evenly subcategorized as MPN-CMML (49.8%) versus MDS-CMML (50.2%) by FAB criteria. Splenomegaly was demonstrable in 25%. Most patients had favorable cytogenetics by both IPSS (70%) and CPSS (67.6%) classification schemas. Mean BM blast % was 5.6 and mean monocyte count was 4.85x103/dL. Median OS of the entire data set was 31.6 months. All tested prognostic models were valid and able to predict OS (p<0.0001). To compare relative model performance, 1013 complete cases with sufficient data to calculate all risk models were compared using time dependent receiver operator characteristic (ROC) curves and their area under the curves (AUC). ROC curves were calculated for OS at 36 months. The R-IPSS model had the highest AUC (0.694) while the DS model had the lowest (0.635). The difference in AUC between the R-IPSS and DS models was statistically significant (p = 0.003) whereas there was no significant difference from other models tested. To determine which models were most vulnerable to reclassification from low-risk to higher risk, we calculated a vulnerability score defined by the number of models able to up-stage low-risk disease in more than 15% of cases. Using this metric, the Mayo and MDASC scores were least vulnerable to up-staging by other models. Lastly, we used an orthogonal method for determination of variable importance (RSF for OS at 36 months). RSF is an ensemble method of classification and regression tree methodology. Ranking of variables by their weighted importance determined by this procedure identified the top five variables were hemoglobin, circulating blasts, platelet count, number of cytopenias, and karyotype. A new risk model was developed using the results of the RSF with an AUC of 0.720, which was comparable but not significantly different than other modern scoring systems. Conclusions: This represents the largest international CMML-specific data set derived from eight unique centers of excellence. All modern prognostic models were valid and performed comparably. Despite novel strategies for variable importance discovery, an improved model could not be constructed. This data suggest that the power of clinical variables has reached an asymptote and highlights the need for incorporation of molecular and other novel biological features. This data set is currently being populated with molecular data and is being leveraged to test published CMML molecular prognostic models which will be disclosed at the time of presentation. Disclosures Garcia-Manero: Epizyme, Inc: Research Funding. Fenaux:Celgene, Janssenm, Novartis: Research Funding. Kantarjian:ARIAD, Pfizer, Amgen: Research Funding. Sekeres:Celgene: Membership on an entity's Board of Directors or advisory committees; Amgen Corp: Membership on an entity's Board of Directors or advisory committees; Boehringer-Ingelheim Corp: Membership on an entity's Board of Directors or advisory committees.

Abstract Introduction: Patients with RR-AML, particularly older adults, have dismal outcomes and limited therapy options. Given low response rates and high toxicity with salvage intensive chemotherapy, and frequent ineligibility for allogeneic stem cell transplantation (alloSCT), many patients are treated with HMAs. Robust data regarding use of HMAs in AML predominates in the frontline setting, while their use in RR-AML has limited supportive data. Here wesought to analyze theoutcomes and their predictors in patients with RR-AML treated with HMAs. Methods:We collected data, spanning a period from 2006 to 2016, from 7 centers in the United States and 4 centers in Europe regarding patients treated with HMAs for RR-AML. Responses were defined by International Working Group criteria. Kaplan-Meier methods estimated overall survival (OS) from initiation of HMAs to death or end of follow-up. Multivariable logistic regression models estimated odds for response, and multivariable Cox Proportional Hazard (CPH) models estimated hazards ratios (HR) for OS. Covariates considered included HMA received, age at diagnosis (in years), AML classification at diagnosis (AML with myelodysplasia-related changes [AML-MRC], therapy-related [t]-AML), disease status (relapsed vs. refractory), number of therapy lines prior to HMA (1 vs. 2 vs. >=3), duration of first complete remission (CR1), white blood cell count, peripheral blood blast percentage, bone marrow (BM) cellularity (<=20% vs. > 20%), BM blast percentage (<=20% vs. >20%), cytogenetic risk group, and the presence of complex or chromosome 7 abnormalities. Results: Of 514 patients, 217 patients (42.2%) had refractory and 297 (58%) had relapsed AML. By end of study, 415 patients (88.5%) had died. Median follow-up for living patients was 11.6 months.Median age at diagnosis was 64 years (range [R], 16-92). AML-MRC was diagnosed in 29.0% while 8.2% had t-AML. Median number of prior therapies was 2 (R, 1-7), with 48.3% receiving 1 prior line, 30.2% receiving 2 prior lines, and 21.5% receiving >=3 prior lines. Prior alloSCT was performed in 21.2%. Only 1.9% had good risk (core binding factor) karyotype, while 56.2% had intermediate risk karyotype, and 41.9% had poor risk karyotype. Azacitidine was used in 45.8% and decitabine in 54.2%; median number of azacitidine cycles was 4 (Interquartile range [IQR], 2-6) compared to 2 for decitabine (IQR, 1-4, p <0.001). Best response to HMAs was CR in 11.7% (95%CI, 9%-14%), CRi in 6.4% (95%CI, 4.3%-8.8%), hematologic improvement (HI) in 8% (95%CI, 5.7%-10.5%), stable disease (SD) in 9.8% (95%CI, 7.2%-12.5%), while 64.1% (95%CI, 57.7%-66.2%) had progressive disease (PD). Median OS from HMA initiation for all patients was 6.9 months (IQR, 3.0-13.3). There was a significant difference in OS based on best response achieved [Figure 1]. Unadjusted OS showed an insignificant trend for worsening with increasing number of prior lines of therapy [Figure 2A]. In unadjusted analyses, there was no difference in OS based on HMA received in all patients [Figure 2B] or the subset who received only 1 prior line of therapy (median OS: Azacitidine vs. decitabine 8.4 vs 7.3 months, p=0.88). Following HMA therapy, the median number of subsequent therapies was 0 (R, 0-6), and only 12.8% underwent alloSCT. In multivariate CPH models, HMA used was not significantly associated with OS (HR=0.80, 95%CI, 0.42-1.51, p=0.49), while increasing age, and presence of complex cytogenetics and chromosome 7 abnormalities were significantly associated with risk of death [Table 1]. In multivariable logistic regression models, HMA used was not associated with achieving CR+CRi (Odds ratio=0.56, p=0.32). Conclusions: In this largest reported cohort of patients with RR-AML treated with HMAs, we found that HMAs are often used as alast line of therapy, with a minority of patients receiving subsequent treatment. Nonetheless, the minority of patients who achieve CR (11.7%) with HMA therapy had a median OS of 25.6 months. Therefore, use of HMAs for management of RR-AML is a reasonable intervention in the absence of clinical trial options. There appears to be no difference in OS or probability of achieving CR+CRi based on HMA used. Ongoing analyses in this dataset include further evaluations of predictors, including genetic mutations, and the development of prediction tools for clinical outcomes with HMA therapy. Figure 1. Figure 1. Disclosures Podoltsev: Ariad: Consultancy, Honoraria; Incyte: Consultancy, Honoraria. Ritchie:Novartis: Honoraria; Incyte: Speakers Bureau; Arian: Speakers Bureau; Pfizer: Honoraria; Celgene: Speakers Bureau. Sekeres:Millenium/Takeda: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees. Komrokji:Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Consultancy, Speakers Bureau. Al-Kali:Onconova Therapeutics, Inc.: Research Funding; Celgene: Research Funding. Santini:Celgene: Consultancy, Honoraria, Research Funding; Janssen: Consultancy, Honoraria; Novartis: Consultancy, Honoraria; Onconova: Consultancy; Amgen: Consultancy; Astex: Consultancy. Roboz:Cellectis: Research Funding; Agios, Amgen, Amphivena, Astex, AstraZeneca, Boehringer Ingelheim, Celator, Celgene, Genoptix, Janssen, Juno, MEI Pharma, MedImmune, Novartis, Onconova, Pfizer, Roche/Genentech, Sunesis, Teva: Consultancy. Fenaux:Celgene, Janssen,Novartis, Astex, Teva: Honoraria, Research Funding. Prebet:celgene: Consultancy, Honoraria; Novartis: Consultancy, Honoraria. Gore:celgene: Consultancy, Honoraria. Zeidan:Ariad: Consultancy, Honoraria; Pfizer: Consultancy, Honoraria; Incyte: Consultancy, Honoraria; Celgene: Consultancy, Honoraria.

Abstract Background The Janus-kinase (JAK) pathway plays a critical role in normal hematopoiesis. In two phase 3 studies, the JAK1/JAK2 inhibitor ruxolitinib (RUX) demonstrated clinical benefit over placebo (COMFORT-I) or best available therapy (COMFORT-II) in patients with intermediate-2 or high-risk myelofibrosis (MF). Anemia and thrombocytopenia, the most common adverse events associated with RUX treatment, were dose-dependent and expected based on RUX's mechanism of action. These events were successfully managed and few RUX-treated patients discontinued for anemia or thrombocytopenia (<1% for each event). The ability to identify patients who would most likely benefit from a dose adjustment may contribute to improved patient management. Therefore, an analysis was conducted in patients who received RUX in COMFORT-I to identify patient characteristics that could help select patients who would potentially benefit from closer monitoring and dose titration in the early period after initiation of therapy. Methods In COMFORT-I, starting doses in the patients randomized to RUX were 15 mg twice daily (BID) for patients with baseline platelet counts of 100–200 ×109/L (n=55) and 20 mg BID for those with platelet counts >200 ×109/L (n=100). Dose reductions were required for protocol-defined decreases in platelet counts; there was no protocol-mandated dose change for anemia. The time course of dose adjustments and the most common titrated doses by starting dose were evaluated. Logistic regression analyses with baseline hemoglobin (Hgb) level, platelet count, IPSS risk category, sex, and MF type as model effects were conducted to determine predictors of: 1) titration to a RUX dose of ≤10 mg BID within 8 weeks of starting therapy (early dose titration event); or 2) development of grade ≥3 anemia or new need for RBC transfusions (≥2 units of packed RBCs) within 12 weeks of therapy initiation (anemia event). Only baseline platelet count and Hgb emerged as significant predictors (alpha=0.05). Receiver operating characteristics (ROC) analyses were also conducted for these two predictors to determine a clinically relevant cutoff of the underlying predictor variable with a positive predictive value (PPV), true positive rate (TPR), and false positive rate (FPR) obtained at all possible cutoffs and a general goal of obtaining a TPR ≥ 50%, an FPR ≤ 10%, and a high PPV. Results Approximately 70% of patients had a dose adjustment within the first 12 weeks of initiating RUX treatment. At week 24, 77% of patients with baseline platelet counts of 100–200 ×109/L and 39% of those with baseline platelet counts >200 ×109/L had titrated to a reduced RUX dose relative to their starting dose. With longer-term follow-up (median 149 weeks), the mean dose over time in patients who continued on study was ∼10 mg BID for patients starting at 15 mg BID and ∼15 mg BID for those starting at 20 mg BID (Figure). ROC analyses identified that a baseline platelet count range of less than 150–175 x109/L may predict which patients would experience an early titration event, while a baseline Hgb level of less than 10 mg/dL may predict which patients would experience an anemia event (Table). Conclusions In COMFORT-I, most patients experienced a RUX dose adjustment in the first 12 weeks of treatment and stable doses were achieved with long-term therapy. Dose titration allowed most patients to maintain adequate platelet counts. After starting therapy at 15 or 20 mg BID, the majority of patients titrated to RUX doses of 10 mg BID or higher, which have been previously been shown to produce meaningful spleen volume reductions and symptom improvements (Mesa et al. 8th Annual Hematologic Malignancies Conference. Abstract 305; Verstovsek et al. ASH 2012. Abstract 800; Talpaz et al. ASH 2012. Abstract 176). A baseline platelet count of less than 150 ×109/L and baseline Hgb level of less than 10 g/dL may be clinically useful cutoffs for predicting which patients may benefit from closer monitoring shortly after treatment initiation of RUX and appropriate dose adjustments as needed. Disclosures: Mesa: Lilly: Research Funding; Genentech: Research Funding; NS Pharma: Research Funding; Gilead: Research Funding; Incyte Corporation: Research Funding. Komrokji:Incyte Corporation: Membership on an entity’s Board of Directors or advisory committees. Sun:Incyte Corporation: Equity Ownership; Incyte Corporation: Employment. Sandor:Incyte Corporation: Equity Ownership; Incyte Corporation: Employment. Verstovsek:Incyte Corporation: Research Funding.

Abstract BackgroundPacritinib, a JAK2, JAK2 V617F and FLT3 kinase inhibitor is in Phase III development in myelofibrosis. In preclinical models and in Phase 1 studies in both non-Hodgkin lymphoma and myelofibrosis, it appeared unique among this class of agents in its lack of myelosuppression at doses that inhibited the JAK2/STAT3 pathway and resulted in significant clinical efficacy. This potentially important clinical observation was verified in two Phase II studies in myelofibrosis that had no exclusion restriction for low platelets. Pacritinib was effective at reducing spleen size and symptoms regardless of starting platelet counts and there was no apparent decline in either platelet count or hemoglobin levels, or an increase in transfusion requirements. Aims and Methods: To help elucidate the mechanisms for pacritinib’s lack of myelosuppression, a kinome screening analysis against 429 recombinant kinases at 100 nM pacritinib concentration was performed, followed by titration from 1-100nM against those that were >50% inhibited at 100nM (Anastassiadis et al.; Nature Biotechnology; 2011). In humans, pacritinib achieves steady state free drug levels of approximately 200nM at the standard 400 mg daily dose. Results:All kinases with IC50 values of <50nM are included in the table. Table Kinase IC50 (nM) JAK2; JAK2(V617F) 6.0; 9.4 FLT3; FLT3-ITD; FLT3 (D835Y) 6.4; 12.0; 8.3 TYK2 27.0 JAK3 18.3 IRAK1 (IL-1 receptor kinase) 13.6 HIPK4 (PT kinase for P53 ser-9) 14.5 FMS 39.5 c-KIT and most common mutations except (V560G), (V559D/V654A) 43.8 c-Src; (c-Src (T341M)) 47.0 (55.5) Of note, pacritinib did not affect JAK1 (82% control at 100nM). As the concentrations of pacritinib achieved clinically are likely inhibit signaling the kinases listed in the table, these data suggest that the lack of myelosuppression with pacritinib may be either due to inhibition of JAK2 without affecting JAK1 or to suppression of IRAK1, an IL-1 receptor kinase, downstream from Toll-Like Receptors, and associated with the inflammatory response and suppression of normal hematopoiesis in myelodysplasia models (Rhyasen et al.; Cancer Cell; 24:90, 2013). Conclusions: The kinase profile for pacritinib is unique among agents in development for myelofibrosis combining lack of suppression of JAK1 with suppression of inflammatory signaling through IRAK1. Moreover, the kinase inhibition profile of pacritinib suggests its potential therapeutic utility in AML, MDS, and particularly in CMML due to CMML’s upregulation of JAK2, c-FMS, and IRAK1. Disclosures Singer: CTI BioPharma, Corp: Employment, Equity Ownership. Mesa:Incyte Corporation, CTI, NS Pharma, Inc., Gilead Science Inc., Celgene: Research Funding. Verstovsek:CTI BioPharma, Corp: Research Funding.

Key Points In the largest study of HMAs in RR-AML to date, 16% of patients achieved CR/CRi and experienced a median OS of 21 months. Low proliferative disease (peripheral blood blasts <5%) was associated with improved response and OS.

BACKGROUNDTransfusion of blood products is a key component of the supportive management in patients with acute leukemia (AL). However high‐quality trial evidence and clinical outcome data to support specific transfusion goals for blood products for patients with AL remain limited leading to diverse transfusion practices. The primary objective of this study was to determine the spectrum of transfusion patterns in a variety of care settings among providers who treat AL patients.STUDY DESIGN AND METHODSA 31‐question survey queried providers caring for AL patients about the existence of institutional guidelines for transfusion of blood products, transfusion triggers for hemoglobin (Hb), platelets (PLTs), and fibrinogen in various settings including inpatient and outpatient and before procedures.RESULTSWe analyzed 130 responses and identified divergent transfusion Hb goals in hospitalized and ambulatory patients, fibrinogen goals for cryoprecipitate transfusions, and variation in practice for use of certain PLTs and red blood cell products. The least variable transfusion patterns were reported for PLT goals in thrombocytopenia and in the setting of invasive procedures such as bone marrow biopsy and lumbar punctures.CONCLUSIONSThis survey confirmed wide variations in blood product transfusion practices across several clinical scenarios in patients with AL. The findings emphasized the need for large prospective randomized trials to develop standardized evidence‐based guidelines for blood product transfusions in patients with AL with the goal of limiting unnecessary transfusions without compromising outcomes.