Maciejewski, Jaroslaw P.

Introduction: The NHLBI National MDS Study (NCT02775383) is a prospective cohort study conducted at 92 community hospitals and 29 academic centers. It enrolls patients undergoing work up for suspected MDS to understand the genetic, epigenetic, and biological factors associated with the initiation and progression of the disease. Previously untreated, cytopenic participants undergo both local and centralized pathology review and are assigned a diagnosis, including MDS, MDS/MPN, AML with blasts < 30%, and "Other". Emerging data suggests that Next Generation Sequencing (NGS), along with cytogenetics and clinical variables, may improve MDS diagnostic precision. Given that our study relies on central review (with additional tertiary pathology review used to adjudicate disagreements), we examined whether targeted gene sequencing data could be used to increase the agreement between local and central pathologic diagnosis of MDS vs. Other. Methods: Peripheral blood and bone marrow (BM) biopsy specimens from cytopenic patients, along with clinical history, CBC, and other results including karyotyping, FISH and pathology reports from local pathologists were reviewed by central pathologists. The updated 2016 WHO classifications were used to diagnose MDS. Targeted exon sequencing of 96 genes was performed using BM specimens. A subset of 648 individuals that were classified as MDS (n=212) or Other (n=436, including 90 CCUS and 89 individuals with other cancers) by pathology assessments were selected. A mean coverage of 1,317X was achieved and variants had a minimum variant allele frequency (VAF) of 2% (except FLT3). Variants for 596 subjects were manually reviewed to retain likely disease-causing variants to build a binary classifier (MDS vs. Other) using the maximum VAF per gene as input (Figure 1). Subjects diagnosed with MDS or Other by both central and local pathology were used for training, validation, and testing, and were considered "gold standard" (GS) cases (n=546). These subjects were split into 4 random groups with equal proportions of MDS cases. 75% of the GS cases were used to train and validate lasso-regularized logistic regression models using 3-fold cross validation. ROC curve analysis was carried out using the remaining 25% of GS cases (Test Set 1) on the best model to identify an optimal probability cut off point for classifying subjects as MDS. Model performance was then tested on 50 subjects for which the central and local pathology diagnosis disagreed (Test Set 2), as well as on 52 additional subjects irrespective of agreement (Test Set 3). Results : The best performing logistic regression model retained 7 genes as most informative in a discriminating diagnosis of MDS from Other based on their VAFs, in order of impact: TP53, SF3B1, U2AF1, ASXL1, TET2,STAG2, and SRSF2. We used this model to assign probabilities for each of the subjects in Test Set 1 and to estimate the performance using ROC analysis (Figure 1), resulting in a high area under the curve (AUC) of 0.89. We chose a probability cut-off of ≥0.17, being associated with a high percentage of correct classification of MDS with a sensitivity and specificity of 0.90 and 0.81, respectively. Among the cohort of 50 subjects with a discordant local and central pathology diagnosis (Test Set 2), the classifier accurately reassigned 37 subjects (accuracy = 74%) from the local to the central pathology. The blinded tertiary pathology reviewer agreed with central in all Test Set 2 cases. This included 24/34 MDS cases that had been labeled as Other by local pathology (positive predictive value [PPV]=0.89). 3/16 final pathology-classified Other cases were mis-classified as MDS by the local pathologist (negative predictive value [NPV] = 0.57). Next, we assessed the ability of the model to predict MDS vs. Other for 52 additional independent subjects using the third pathologist's diagnosis to break any ties (Test Set 3). The classifier correctly predicted 15/21 MDS cases (PPV=0.83) and misclassified 6/31 Others as MDS (NPV=0.82). The overall accuracy was 83%. Conclusions: We identified that VAFs for 7 genes can correctly re-classify subjects as either MDS or Other in 74% of cases that were misclassified between local and central pathology review. Further assessment on an independent cohort showed an accuracy of 83% of the model. Taken together, these data suggest that complementing pathology reviews with targeted sequencing of 7 genes could improve MDS diagnosis. Disclosures Lindsley: MedImmune: Research Funding; Jazz Pharmaceuticals: Consultancy, Research Funding; Bluebird Bio: Consultancy; Takeda Pharmaceuticals: Consultancy. Bejar:Aptose Biosciences: Current Employment; AbbVie/Genentech: Honoraria; Astex/Otsuka: Honoraria; Takeda: Honoraria, Research Funding; Celgene/BMS: Honoraria, Research Funding; Daiichi-Sankyo: Honoraria; Forty-Seven/Gilead: Honoraria; Genoptix/NeoGenomics: Honoraria. DeZern:MEI: Consultancy; Astex: Research Funding; Abbvie: Consultancy; Celgene: Consultancy, Honoraria. Foran:H3Biosciences: Research Funding; Aptose: Research Funding; Kura Oncology: Research Funding; Trillium: Research Funding; Takeda: Research Funding; Revolution Medicine: Consultancy; Xencor: Research Funding; Agios: Honoraria, Research Funding; Aprea: Research Funding; Actinium: Research Funding; Servier: Membership on an entity's Board of Directors or advisory committees; Novartis: Membership on an entity's Board of Directors or advisory committees; Abbvie: Research Funding; BMS: Membership on an entity's Board of Directors or advisory committees; Pfizer: Membership on an entity's Board of Directors or advisory committees; Boehringer Ingelheim: Research Funding. Gore:Abbvie: Consultancy, Honoraria, Research Funding. Komrokji:Acceleron: Honoraria; Incyte: Honoraria; Abbvie: Honoraria; Agios: Speakers Bureau; BMS: Honoraria, Speakers Bureau; Jazz: Honoraria, Speakers Bureau; Geron: Honoraria; Novartis: Honoraria. Maciejewski:Alexion, BMS: Speakers Bureau; Novartis, Roche: Consultancy, Honoraria. Padron:Novartis: Honoraria; BMS: Research Funding; Incyte: Research Funding; Kura: Research Funding. Starczynowski:Captor Therapeutics: Consultancy; Tolero Therapeutics: Research Funding; Kurome Therapeutics: Consultancy, Current equity holder in private company, Research Funding. Sekeres:BMS: Consultancy; Takeda/Millenium: Consultancy; Pfizer: Consultancy.

PURPOSE Impaired response to erythropoietin underlies ineffective erythropoiesis and anemia in myelodysplastic syndromes (MDS). We investigated whether treatment with lenalidomide (LEN), which augments erythropoietin receptor signaling in vitro, can restore and improve hemoglobin response to epoetin (EPO) alfa in patients with lower-risk, non-del(5q) MDS who have anemia that is refractory to or have low probability of benefit from treatment with recombinant erythropoietin. METHODS In a phase III, US intergroup trial, we randomly assigned patients to receive either LEN and EPO alfa or LEN alone following stratification by serum erythropoietin concentration and prior erythropoietin treatment. RESULTS A total of 195 evaluable patients were randomly assigned: 99 patients to the LEN-EPO alfa cohort and 96 to LEN alone. After four cycles of treatment, the primary end point of major erythroid response (MER) was significantly higher (28.3%) with the combination compared with LEN alone (11.5%) ( P = .004). Among 136 patients who completed 16 weeks of study treatment, 38.9% and 15.6% achieved MER, respectively ( P = .004). Additionally, minor erythroid response was achieved in 18.2% and 20.8% of patients, for an overall erythroid response rate of 46.5% versus 32.3%. Among LEN nonresponders, 38 crossed over to the addition of EPO alfa with 10 patients (26.3%) achieving a MER. Responses to the combined treatment were highly durable with a median MER duration of 23.8 months compared with 13 months with LEN alone. CONCLUSION LEN restores sensitivity to recombinant erythropoietin in growth factor–insensitive, lower-risk, non-del(5q) MDS, to yield a significantly higher rate and duration of MER compared with LEN alone (funded by the National Cancer Institute; E2905 ClinicalTrials.gov identifier: NCT02048813 ).

Abstract Introduction: Recent reviews indicate that transfusional hemosiderosis may be associated with an increased risk of mortality in lower-risk pts with MDS. This trial is designed to evaluate the efficacy and long-term safety of deferasirox (Exjade®, ICL670) in MDS. Deferasirox is an oral iron chelator approved for use in pts with transfusional iron overload. Methods: This is a Phase II, open-label, 3-yr clinical trial in 55 US centers, enrolling 150 pts (aged ≥18 years) with Low- or Int-1-risk MDS (by IPSS criteria) and transfusional iron overload (serum ferritin ≥1000 ng/mL and >20 units RBC transfusions). Deferasirox dosing is 20–30 mg/kg/day. Serum ferritin, iron, transferrin and transferrin saturation are being assessed at screening and monthly in yr 1, then quarterly in yrs 2 and 3, while labile plasma iron (LPI) is assessed quarterly in yr 1. In addition, creatinine, calculated creatinine clearance, echocardiograms and endocrine and hematological status are being assessed. This report describes baseline data in these pts. Results: As of June 2006, 84 pts have enrolled. Demographic data are available from 79 pts: median age 71 years (range 47–87); sex (52 male, 27 female); ethnicity (74 Caucasian, 2 Black, 2 Hispanic, 1 Oriental); and IPSS Risk Group (Low: 22 pts; Int-1: 56 pts). Iron status is summarized in the table: Parameter n Mean ± SD Median Range Normal range n/a, not applicable Serum ferritin,μg/L 84 3779 ± 4070 2951 1160–36280 12–370 Serum iron, μg/dL 84 205 ± 64 201 48–409 37–180 Transferrin, mg/dL 82 153 ± 31 152 83–244 190–375 Transferrin saturation, % 83 85 ± 15 91 20–94 15–50 LPI, μmol/L 38 0.52 ± 0.63 0.25 0–2.9 0 Total transfusions, n 78 63.3 ± 66.3 41.5 14–435 n/a Years of transfusion 75 3.4 ± 1.9 3 1–12 n/a Baseline concurrent therapies: 5-azacytidine (Vidaza): 5 pts; lenalidomide (Revlimid): 1 pt. Calculated creatinine clearance: normal (>80 mL/min): 37 pts; mildly abnormal (51–80 mL/min): 30 pts; moderately abnormal (30–50 mL/min): 9 pts. Hematological parameters: Anemia was present in all pts; other cytopenias included: neutropenia (<1800/μL): 13 pts, thrombocytopenia (<100,000/μL): 15 pts; neutropenia and thrombocytopenia: 12 pts. A total of 53 pts had received chelation prior to enrolling: 51 deferoxamine (Desferal®); and 2 deferasirox. Conclusions: Despite the prior availability of deferoxamine, these baseline data demonstrate significant levels of iron overload among transfused pts with myelodysplasia. Serum iron, ferritin and LPI are all well above the clinically significant thresholds associated with increased complications. Since recent data has suggested that iron overload may be a poor prognostic indicator in MDS, increased attention to maintaining appropriate iron balance is warranted. The recent availability of an oral iron chelator may be more acceptable to MDS pts and their physicians. This ongoing trial is designed to assess the long-term efficacy, safety, and clinical benefits of deferasirox in pts with MDS.

Background: Only a small subset of Lower risk (LR) MDS patients benefit from treatment with rhu-Erythropoietin (Epo). We previously reported that lenalidomide (LEN) restores sensitivity to Epo in MDS progenitors by inducing the formation of lipid rafts that are enriched for signaling competent, JAK2/Epo-receptor complexes (McGraw K, et. al. PLoS One 2014; Basiorka A, et. al. Cancer Res 2016). In the MDS-002 and MDS-005 trials, treatment with LEN monotherapy gave rise to RBC transfusion-independence (TI) in 26% of azanucleoside-naïve, transfusion-dependent (TD) LR, non-del(5q) MDS patients for a median of 10.2 and 7.75 months, respectively. In a pilot study of Epo-refractory LR-MDS patients, the addition of epoetin alfa (EA) to LEN treatment yielded erythroid responses in 28% of patients who were unresponsive to LEN alone, suggesting that LEN may overcome clinical resistance to augment response to rhEpo (Komrokji R, et. al. Blood 2012). To test this hypothesis, we performed a randomized phase III trial comparing treatment with LEN to LEN+EA in LR non-del(5q) MDS patients who were refractory to, or not candidates for treatment with rhEpo. Methods: Patients with Low or Intermediate-1 IPSS risk MDS with hemoglobin <9.5 g/dL who were unresponsive to rhEpo treatment or were TD (>2 units/month) with serum Epo >500mU/mL were eligible. Patients were stratified by serum Epo level and prior rhEpo (EA vs. darbepoetin vs. none) then randomized to treatment with LEN 10 mg/d x21d q4wk (Arm A) or LEN + EA 60,000U SC/wk (Arm B). The primary endpoint was major erythroid response (MER) at week 16 which was defined according to transfusion status at baseline: (1) achievement of RBC-TI for ≥ 8 consecutive weeks AND a sustained ≥1 g/dL hemoglobin rise compared to mean pre-transfusion baseline value in TD patients; and (2) a >2 g/dL rise in hemoglobin without transfusion for ≥ 8 consecutive weeks in non-TD patients (<4U RBC/8 wks). Arm A non-responders were offered cross-over to combined therapy. Results: Among 205 patients randomized, 14 were excluded from the primary analysis due to a 4 month interruption in drug supply. Among the 195 evaluable patients, 96 were assigned to Arm A and 99 to Arm B. Baseline characteristics were well balanced between arms with 85% of patients heavily TD, receiving a median of 4 units/8 weeks. Overall, 93% of patients received prior treatment with Epo and 18% azanucleosides. In an intent to treat analysis, 28/99 patients (28.3%) in Arm B achieved MER compared to 11/96 (11.5%) in Arm A (P=0.004). Among 136 patients who completed 16 weeks of study treatment, 28/72 (38.9%) and 10/64 (15.6%) achieved MER, respectively (P=0.004). Forty-four Arm A non-responders crossed over to combination-therapy with 11 patients (25%) experiencing a MER. Multivariable logistic regression analysis identified only treatment with LEN + EA as an independent covariate for erythroid response (P=0.01). Responses were durable with MER median duration of 23.8 months in Arm B compared to 13 months in Arm A. There was no significant difference in the frequency or distribution of >Grade 3, non-hematologic adverse events between treatment arms. Two patients progressed to AML while on study (Arm A), and no thromboembolic events were reported. Conclusions: LEN restores sensitivity to rhEpo in otherwise refractory, LR-non-del(5q) MDS patients to yield a significantly higher frequency of durable major erythroid responses compared to LEN alone. The addition of LEN to EA treatment is an effective strategy for the management of Epo-refractory patients with a potential duration of benefit extending to years. Disclosures List: Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding. Verma:Janssen: Research Funding; BMS: Research Funding; Stelexis: Equity Ownership, Honoraria; Acceleron: Honoraria; Celgene: Honoraria. Maciejewski:Novartis: Consultancy; Alexion: Consultancy. Komrokji:JAZZ: Speakers Bureau; Novartis: Speakers Bureau; JAZZ: Consultancy; Agios: Consultancy; Incyte: Consultancy; DSI: Consultancy; celgene: Consultancy; pfizer: Consultancy. Luger:Onconova: Research Funding; Pfizer: Honoraria; Seattle Genetics: Research Funding; Cyslacel: Research Funding; Biosight: Research Funding; Ariad: Research Funding; Agios: Honoraria; Genetech: Research Funding; Jazz: Honoraria; Daichi Sankyo: Honoraria; Kura: Research Funding; Celgene: Research Funding. Mattison:Pfizer: Membership on an entity's Board of Directors or advisory committees. Altman:Abbvie: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Theradex: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Glycomimetics: Consultancy, Honoraria, Other: Data Safety and Monitoring Committee; Agios: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; France Foundation: Speakers Bureau; Daiichi Sankyo: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; prIME Oncology: Speakers Bureau; PeerView: Speakers Bureau; Cancer Expert Now: Consultancy; Novartis: Consultancy; Biosight: Other: US Lead. Claxton:Astellas Pharma: Other: Pharma support of clinical studies; Merck Sharp & Dohme Corp.: Other: Pharma support of clinical studies; Cyclacel Pharmaceuticals, Inc.: Other: Pharma support of clinical studies; Medimmune Inc.: Other: Pharma support of clinical studies; Novartis Pharmaceuticals: Other: Pharma support of clinical studies; Celgene Corporation: Other: Pharma support of clinical studies; Incyte Corporation: Other: Cyclacel Pharmaceuticals, Inc; Daiichi Sankyo Co. and Ambit Biosciences Corp: Other: Pharma support of clinical studies. Artz:Miltenyi: Research Funding. Tallman:Daiichi-Sankyo: Consultancy, Membership on an entity's Board of Directors or advisory committees; ADC Therapeutics: Research Funding; Oncolyze: Consultancy, Membership on an entity's Board of Directors or advisory committees; Tetraphase: Consultancy, Membership on an entity's Board of Directors or advisory committees; UpToDate: Patents & Royalties; Jazz Pharmaceuticals: Consultancy, Membership on an entity's Board of Directors or advisory committees; Cellerant: Research Funding; Rigel: Consultancy, Membership on an entity's Board of Directors or advisory committees; KAHR: Consultancy, Membership on an entity's Board of Directors or advisory committees; Biosight: Research Funding; Abbvie: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Nohla: Consultancy, Membership on an entity's Board of Directors or advisory committees; Orsenix: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Delta Fly Pharma: Consultancy, Membership on an entity's Board of Directors or advisory committees; BioLineRx: Consultancy, Membership on an entity's Board of Directors or advisory committees. OffLabel Disclosure: Lenalidomide used for treatment non-del 5q myelodysplastic syndromes.

Introduction: Myelodysplastic syndromes (MDS) are a heterogenous group of blood disorders defined by peripheral cytopenia(s), bone marrow failure, morphological dysplasia, and risk of progression. To understand the genetic, epigenetic and biological factors associated with the initiation and progression of MDS, NHLBI created the National MDS Study (NCT02775383). This is a prospective cohort study conducted at 92 community hospitals and 29 academic centers enrolling patients undergoing diagnostic work up for suspected MDS or MDS/myeloproliferative neoplasm (MPN) overlap syndrome. Eligible patients have yet to receive any therapy directed at their cytopenias. Previously untreated cytopenic participants underwent centralized histopathology and data review at the time of enrollment for assignment into distinct subcategories: MDS, MDS/MPN overlap, AML, and Other. Targeted exon sequencing of 96 genes was performed using marrow specimens from the first 300 consecutive individuals in the study. Here we report the genetic mutations for this cohort. Methods: NovaSeq 6000 was used for deep sequencing at a mean coverage of 1,286X and mean breadth (bases covered at ≥100X) of 99.8%. Reads were aligned against build GRCh38 using BWA-MEM, and VarScan2 was used to detect SNVs and INDELS. Variants were filtered for those with an allele base quality of >25 in combination with rule-based and manual review criteria. Subjects in the Other category without an identified malignancy were considered clonal cytopenias of undetermined significance (CCUS) when a mutation or a clonal cytogenetic change was present. Fisher's exact and Wilcoxon rank sum tests in combination with Bonferroni correction were applied to compare groups. Results : A total of 350 putative nonsynonymous pathogenic variants in 36 genes with an allele frequency of >.05 were identified across 150 patients (50%). At least one variant was noted in the following proportion of individuals: 61/72 (85%) with MDS, 13/13 (100%) with MDS/MPN, 15/17 (88%) with AML, and 61/198 (31%) in the Other category, of which 48 were CCUS and 13 were other cancers. Two CCUS patients only had a cytogenetic abnormality. Table 1 shows the distribution of variants in each subcategory of patients for the most commonly mutated genes in our cohort of 300 subjects. Mutations in these genes were enriched in specific groups: SF3B1, STAG2,TP53, and ASXL1 in MDS; TET2 in MDS/MPN; and IDH2 and TP53 in AML (one-sided p<0.0012). None were enriched in the Other group. Within the CCUS subset, 21 genes were mutated, with 37 of 50 (74%) patients having a mutation in TET2, ASXL1, SRSF2, SF3B1, or DNMT3A. The heatmap presented in Figure 1 summarizes variants by subject and allele frequency. Pair-wise comparisons of baseline characteristics of subjects between MDS, MDS/MPN, AML, or CCUS groups revealed no significant differences for age or sex. The CCUS group had significantly higher hemoglobins than the MDS group with median hemoglobin levels of 11.35 and 9.40 g/dL, respectively (p<0.0084). AML had significantly lower platelets compared to all other groups (median of 53 vs. > 110x109/L , respectively, p<0.0084). All groups significantly differed in their median ANC with AML having the lowest (0.8x109/L), followed by MDS (1.5x109/L), CCUS (2.45x109/L), and MDS/MPN overlap (5.95x109/L) (p<0.0084). There was no difference in the median number of variants per patient between groups or correlation with age (rs=0.11, p=0.18). The maximum variant allele frequency (maxVAF) per patient was highest in the MDS/MPN group (median = 0.42, range = 0.38-0.91) and lowest in the CCUS group (median = 0.37, range =0.06-0.98) with the MDS/MPN group having a significantly higher maxVAF compared to the MDS and the CCUS groups (p<0.0084). The proportions of subjects with mutations was similar for those who had abnormal (92% [34/37]) and normal (91% [80/88]) cytogenetics. Conclusions: Incorporation of gene-panel sequencing in the comprehensive evaluation of 300 adult cytopenic patients identified half of the cohort with potentially pathogenic variants. Ultimately, a diagnosis of CCUS was possible in 48 of 183 subjects (26%) not diagnosed with MDS, MDS/MPN overlap syndrome, AML, other cancers or clonal cytogenetics. This study continues to serially bank samples from patients with CCUS, in addition to MDS, MDS/MPN, and ICUS, with the goal to better understand the natural history of these diseases and their progression. Disclosures Lindsley: Jazz Pharmaceuticals: Research Funding; Takeda Pharmaceuticals: Consultancy; Medlmmune: Research Funding. Bejar:Celgene: Consultancy; Takeda Pharmaceuticals: Research Funding; AbbVie/Genentech: Consultancy, Honoraria; Astex/Otsuka: Consultancy; Modus Outcomes: Consultancy; Daiichi-Sankyo: Consultancy. Al Baghdadi:Celgene: Consultancy, Honoraria; Heron: Consultancy, Honoraria; Tracon: Equity Ownership; Epizyme: Equity Ownership; Bristol Myer Squibb: Consultancy, Honoraria; Sunesis: Equity Ownership; Portola: Equity Ownership; Heron therapeutics: Equity Ownership; Cardinal health: Consultancy, Honoraria; Bristol Myer Squibb: Equity Ownership; Celgene: Equity Ownership; Spectrum pharmaceutical: Equity Ownership; Astrazeneca: Equity Ownership; Seattle genetics: Equity Ownership; Roche: Consultancy, Honoraria. DeZern:Astex Pharmaceuticals, Inc.: Consultancy; Celgene: Consultancy. Foran:Agios: Honoraria, Research Funding. Gore:Celgene Corporation: Consultancy, Research Funding. Komrokji:DSI: Consultancy; Agios: Consultancy; Novartis: Speakers Bureau; JAZZ: Speakers Bureau; pfizer: Consultancy; celgene: Consultancy; Incyte: Consultancy; JAZZ: Consultancy. Maciejewski:Alexion: Consultancy; Novartis: Consultancy. Starczynowski:Kurome Therapeutics: Consultancy. Sekeres:Millenium: Membership on an entity's Board of Directors or advisory committees; Syros: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees.

Abstract Background Myelodysplastic syndromes (MDS) and chronic myelomonocytic leukemia (CMML) are mainly diagnosed based on morphological changes in the bone marrow. The diagnosis can be challenging in patients (pts) with pancytopenia with minimal dysplasia, and is subject to inter-observer variability. Somatic mutations can be identified in either disease but no genes, in isolation or in combination, are specific for disease phenotype. We developed a geno-clinical model that uses mutational data, peripheral blood values, and clinical variables to predict an MDS vs. CMML diagnosis in pts who presented with cytopenias, in the absence of bone marrow biopsy results. Method We combined genomic and clinical data from 1897 pts treated at our institution (593) and the Munich Leukemia Laboratory (1304). Pts were diagnosed with MDS or CMML according to 2008 WHO criteria. Diagnosis of MDS and CMML was confirmed by independent hematopathologists that were not associated with the study. A genomic panel of 40 genes commonly mutated in myeloid malignancies was included. The initial cohort was randomly (computer generated) divided into learner (80%) and validation (20%) cohorts. Multiple machine learning algorithms were applied to predict the phenotype. Feature extraction algorithms were used to extract genomic/clinical variables that impacted the algorithm decision and to visualize the impact of each variable on phenotype. Prediction performance was evaluated according to the area under the curve of the receiver operator characteristic (ROC-AUC) and confusion/accuracy matrices. Results Of 1897 pts included, 1368 pts had MDS and 529 had CMML. The median age for the entire cohort was 72 years (range, 11-102); 37% were female. The median white blood cell count (WBC) was 5.1x109/L (range, 0.60-176), absolute monocyte count (AMC) 0.19 x109/L (range, 0-96), absolute lymphocyte count (ALC) 0.77x109/L (range, 0-62), absolute neutrophil count (ANC) 2.44x109/L (range, 0-170), hemoglobin (Hgb) 10.2 (range, 3.9-19.6), and platelet (Plt) count 111x103/mL (range 2-1491). The most commonly mutated genes in all pts were: TET2 (33%), ASXL1 (26%), SF3B1 (21%), SRSF2 (16%), RUNX1 (12%), DNMT3A (10%), CBL (7%), U2AF1 (7%), STAG2 (6%), EZH2 (6%), ZRSR2 (6%), NRAS (6%). In CMML, they were: TET2 (51%), ASXL1 (43%), SRSF2 (25%), RUNX1 (18%), CBL (16%), KRAS (12%), NRAS (11%), EZH2 (9%), JAK2 (6%), U2AF1 (5%), SF3B1 (4%), and DNMT3A (3%). In MDS, they were: TET2 (27%), SF3B1 (24%), ASXL1 (21%), SRSF2 (13%), DNMT3A (12%), RUNX1 (10%), STAG2 (8%), U2AF1 (8%), ZRSR2 (7%), TP53 (7%), BCOR (5%), and EZH2 (5%).The median total number of mutations/sample was 2 (range 0-27) for all pts, 2 (range 0-8) for CMML, and 2 (range 0-27) for MDS. A set of 83 genomic/clinical variables were evaluated and several feature extraction algorithms were used to identify the least number of variables that have the most significant impact on the algorithm's decision. These variables included: AMC, ALC, TET2, ANC, ASXL1, SF3B1, Hgb, number of mutations/sample, AEC, age, Plt, splenomegaly, RUNX1, NRAS, CBL, U2AF1, STAG2, DNMT3A, TP53, EZH2, SRSF2, and ZRSR2, Figure 1. When applying the model to the validation cohort, the ROC-AUC was .98 with an accuracy of 94%, with other statistical values as follows: specificities CMML 93%, MDS 96%; sensitivities CMML 96%, MDS 93%; positive predictive values CMML 84%, MDS 98%; negative predictive values CMML 98%, MDS 84%. Individual pt data can also be entered into the model, with a probability of whether the diagnosis is MDS vs. CMML provided along with the impact of each variable on the decision, as shown in Figure 1. When the analysis was restricted to mutations only, the accuracy of the model dropped dramatically (77%, ROC-AUC .85). Conclusions We propose a novel approach using interpretable, individualized modeling to predict MDS vs. CMML phenotypes based on genomic and clinical data without the need for bone marrow biopsy data. This approach can aid clinicians and hematopathologists when encountering pts with cytopenias and a diagnosis suspicious for MDS vs. CMML. The model also provides feature attributions that allow for quantitative understanding of the complex interplay among genotype, clinical variables, and phenotype. Disclosures Sekeres: Celgene: Membership on an entity's Board of Directors or advisory committees; Opsona: Membership on an entity's Board of Directors or advisory committees; Opsona: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees. Meggendorfer:MLL Munich Leukemia Laboratory: Employment. Walter:MLL Munich Leukemia Laboratory: Employment. Hutter:MLL Munich Leukemia Laboratory: Employment. Savona:Boehringer Ingelheim: Consultancy; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Incyte: Membership on an entity's Board of Directors or advisory committees, Research Funding. Gerds:CTI Biopharma: Consultancy; Celgene: Consultancy; Apexx Oncology: Consultancy; Incyte: Consultancy. Sallman:Celgene: Research Funding, Speakers Bureau. Komrokji:Celgene: Honoraria, Research Funding; Novartis: Honoraria, Speakers Bureau; Celgene: Honoraria, Research Funding; Novartis: Honoraria, Speakers Bureau; Novartis: Honoraria, Speakers Bureau; Novartis: Honoraria, Speakers Bureau. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Maciejewski:Apellis Pharmaceuticals: Consultancy; Ra Pharmaceuticals, Inc: Consultancy; Apellis Pharmaceuticals: Consultancy; Ra Pharmaceuticals, Inc: Consultancy; Alexion Pharmaceuticals, Inc.: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Alexion Pharmaceuticals, Inc.: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Nazha:MEI: Consultancy.

Abstract Background/Objective: The association between paroxysmal nocturnal hemoglobinuria (PNH) granulocyte clone size at disease onset and outcomes in patients with PNH remains unclear. Most but not all reports examining the relationship have shown a positive correlation between clone size and thrombotic events [TEs] in patients with PNH, but without a clear temporal association showing the prognostic value of clone size on the risk of TEs. The ongoing International PNH Registry (NCT01374360) is the largest prospective, observational study of patients with PNH conducted to date. The objective of this analysis was to examine the relationship between PNH granulocyte clone size at disease onset and risk of thrombosis after disease onset while untreated with a complement inhibitor in patients enrolled in the Registry. Methods: The current analysis included patients enrolled in the Registry as of April 2018 who had known demographics, were untreated with complement inhibitor therapy at enrollment, and had ≥12 months of untreated follow-up after disease onset. Baseline was defined as disease onset (earliest of a reported PNH clone, date of PNH diagnosis, or reported PNH symptom), and patients were stratified into 5 cohorts based on clone size at baseline (using the earliest reported clone prior to enrollment): cohort 1, clone size: 0.01-1%; cohort 2, clone size: >1-5%; cohort 3, clone size: >5-10%; cohort 4, clone size: >10-50%; cohort 5, clone size: >50%. Event rates for TEs and all major adverse vascular events (MAVEs; including TEs) were calculated for the time period from baseline to last follow-up. Other outcomes of interest included LDH ratio (LDH/LDH upper limit of normal [ULN]), hemoglobin levels, platelet counts, absolute neutrophil counts, and absolute reticulocyte counts at last follow-up. Results: A total of 2489 patients were eligible for the analysis. The majority of patients in the overall study population were female (54.0% [1343/2489]) and white (79.2% [1967/2484]). Mean (standard deviation [SD]) age at PNH start ranged from 36.4 (16.53) in cohort 5 to 44.2 (20.70) in cohort 1. Median time from baseline to last follow-up was 3.7 years in cohort 1, 4.3 years in cohort 2, 4.7 years in cohort 3, 5.6 years in cohort 4, and 6.8 years in cohort 5. Results for the outcomes of interest are summarized in the Table. All cohorts showed a risk of MAVE and TE during follow-up. Although estimated rates of MAVE and TE were highest in the >50% clone size cohort, there was no difference in the rate of MAVE or TE during follow-up across the 4 cohorts with clone size <50% at disease onset. Mean LDH ratio (LDH/LDH ULN) at last follow-up showed a statistically significant difference by clone size at baseline among the cohorts, ranging from a mean (SD) of 1.1 (1.34) in the patients with clone size <1% and increasing to 5.1 (3.81) in the patients with clone size >50% (P<0.0001). No difference in hemoglobin level at last follow-up was observed in the smaller clone size cohorts, although mean hemoglobin level was lower in patients with clone size >50% (P<0.0001). Similar trends were seen in mean platelet and absolute neutrophil counts across the smaller clone size cohorts, while patients with clone size >50% showed higher values (P<0.0001). Mean absolute reticulocyte count at last follow-up was lowest in patients with clone size 0.01-1%, and were incrementally higher in each successive clone-size cohort (P<0.0001). Conclusions: In this study, all patients with a PNH clone at disease onset were at risk for TEs and other MAVEs. Patients in the highest baseline PNH clone size strata (clone size >50%) had an approximately 2-times higher risk of TEs than patients with smaller clone sizes; Patients with small clone sizes (0.01-1%) were older and showed a higher prevalence of BMD. There was no difference in the prognostic value of clone size at disease onset on the risk of TEs and other MAVEs in patients with small (0.01-1% and 1-5%) versus medium-sized (5-10% or 10-50%) clones. Table Table. Disclosures Peffault De Latour: Alexion Pharmaceuticals, Inc.: Consultancy, Honoraria, Research Funding; Novartis: Consultancy, Honoraria, Research Funding; Pfizer Inc.: Consultancy, Honoraria, Research Funding; Amgen Inc.: Research Funding. Maciejewski:Alexion Pharmaceuticals, Inc.: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Alexion Pharmaceuticals, Inc.: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Apellis Pharmaceuticals: Consultancy; Apellis Pharmaceuticals: Consultancy; Ra Pharmaceuticals, Inc: Consultancy; Ra Pharmaceuticals, Inc: Consultancy. Kulasekararaj:Alexion Pharmaceuticals, Inc.: Consultancy, Honoraria, Other: Travel Support . Larratt:Alexion Pharmaceuticals, Inc.: Honoraria, Research Funding. Dingli:Millennium Takeda: Research Funding; Alexion Pharmaceuticals, Inc.: Other: Participates in the International PNH Registry (for Mayo Clinic, Rochester) for Alexion Pharmaceuticals, Inc.; Millennium Takeda: Research Funding; Alexion Pharmaceuticals, Inc.: Other: Participates in the International PNH Registry (for Mayo Clinic, Rochester) for Alexion Pharmaceuticals, Inc.. Wilson:Alexion Pharmaceuticals, Inc.: Employment, Equity Ownership. Gustovic:Alexion Pharma GmbH: Employment, Equity Ownership. Kulagin:Alexion Pharmaceuticals, Inc.: Consultancy, Honoraria.

Abstract MDS with chromosome 5q deletion [del(5q)] is uniquely sensitive to clonal suppression by lenalidomide (LEN), yielding high rates of cytogenetic and erythroid response. Although haploinsufficiency for a candidate tumor suppressor gene within the common deleted segment (CDS) has been implicated in the pathogenesis of del(5q) MDS, no specific gene products account for the karyotype-dependent LEN sensitivity. Using primary AML cells, we showed that LEN is selectively cytotoxic to del(5q) cells, inducing cell cycle arrest at the G2 > G1 junctures and apoptosis in a concentration-dependent fashion. Now we show that the CDS-encoded Cdc25C and PP2A phosphatases (PPTs) serve as molecular targets for LEN. Cdc25C and PP2A are dual specificity PPTs that regulate cell cycle. The genes encoding the catalytic subunit alpha (PP2ACa), the dominant catalytic isoform, and the regulatory subunit B (beta isoform PR 52) of the PP2A holoenzyme reside within the del(5q) CDS. Real-time RT-PCR showed that Cdc25C and PP2ACa gene expression is significantly reduced in primary del(5q) AML and MDS BM-MNC compared to non-del5q and normal controls. A Cdc25C FISH probe confirmed presence of only a single allele in del(5q) specimens [n=5] vs. normal karyotype [n=5] with no overlap with the EGR1 locus. We found that LEN inhibited rHu-Cdc25C activity with IC90 inhibitory concentrations ≤10nM. Similarly, PP2A PPT activity was inhibited by LEN in enzyme immunoprecipitates from Ramos cells at sub-micromolar concentrations. In U937 cells, we confirmed intracellular Cdc25C and PP2A inhibition, showing concentration-dependent retention of the inhibitory p-Tyr15 -Cdc2 substrates for Cdc25C with corresponding phosphorylation of the PP2A substrate Ser216 Cdc25C, findings reproduced by treatment with the PP2A selective antagonist, fostreicin. Cdc25C immunoprecipitation showed that phosphorylation of Ser216 Cdc25C increased binding to the 14-3-3 protein with consequent cytoplasmic sequestration confirmed by fluorescence microscopy. Phosphorylation of histone H2A.X Ser139 was unchanged after LEN treatment, indicating that phosphorylation of the inhibitory Cdc25C serine does not arise from upstream DNA damage activated kinases. Importantly, siRNA suppression of Cdc25C gene expression in U937 cells promoted LEN-specific apoptosis in 45% of cells, which increased to 68% after fostreicin inhibition of PP2A. These data indicate that haplodeficiency for the Cdc25C and PP2ACa genes promotes selective sensitivity of del(5q) cells to LEN-induced apoptosis. Haplodeficiency of critical drug-able targets may represent a novel strategy for tumor selective therapy.