Finke, Christy M.

Background: Myelodysplastic syndrome (MDS)/myeloproliferative neoplasm (MPN) with ring sideroblasts and thrombocytosis (MDS/MPN-RS-T) was formally included as a unique WHO-defined entity in the 2016, iteration of the classification of hematological neoplasms (under MDS/MPN overlap syndromes). Patients with MDS/MPN-RS-T have features of MDS-RS-single lineage dysplasia (SLD), persistent thrombocytosis (platelet count ≥450x109/L), with proliferation of large atypical megakaryocytes. Anemia, including red blood cell transfusion dependency (RBC-TD) is a major problem. While extensive studies have documented erythroid response rates (ERR) to erythropoiesis stimulating agents (ESA) in MDS and MDS-RS-SLD, the data in MDS/MPN-RS-T is less clear. We carried out this study to asses ERR in ESA treated patients with MDS/MPN-RS-T. Methods: After approval by the IRB, patients with 2016, WHO-defined MDS/MPN-RS-T were included in the study. The bone marrow (BM) morphology, cytogenetics and BM RS% were retrospectively reviewed. All study patients underwent next generation sequencing for 29 myeloid-relevant genes, obtained on BM mononuclear cells, at diagnosis, or at first referral, by previously described methods (Patnaik et al., BCJ 2016). Treatment details, including the type of ESA used, dose administered, side effects and reason for discontinuation were retrospectively abstracted. Erythroid responses were assessed using the 2006, international working group (IWG) MDS and the 2015, IWG MDS/MPN response criteria. Results: Forty seven patients with MDS/MPN-RS-T were identified, of which 37 (79%) patients received ESA treatment at any time point during their disease course; median age 73 years (range, 52-93), 46% male.The hemoglobin at diagnosis (HB) and baseline erythropoietin (EPO) levels were 9.1 gm/dL (range, 6.6-12.1) and 39 IU/L (range, 8-500), respectively. Ten (29%) patients were RBC-TD at baseline. None of the 37, ESA-treated patients were on concomitant cyto-reductive therapy for thrombocytosis. Twenty (54%) patients were treated with erythropoietin-a, 13 (35%) with darbepoetin, and 4 (11%) with both sequentially. Median doses were 40,000 IU weekly (range, 20,000-80,000) for erythropoietin-a and 200 mcg every 2 weeks (range, 100-500), for darbepoetin. Median treatment duration was 14 months (range, 1-173) and causes for discontinuation included; ESA failure/loss of response in 20 (54%), sustained HB rise in 1 (3%) and other causes, including adverse events in 4 (11%) patients. With regards to safety, the following events were considered potentially ESA related; treatment emergent hypertension in 2 (5%), venous thromboembolism, ischemic stroke and splenic infarction in 1 (3%) patient each, respectively. All patients with thrombotic/ischemic events had been on low dose aspirin prophylaxis. Erythroid responses by the IWG MDS and MDS/MPN response criteria were assessable in 35 (95%) patients: with 16 (46%) meeting criteria for an erythroid response, by both assessment systems. The median duration of ER was 20 months (range, 2-172). In comparison to ESA non-responders, those that responded were more likely to have baseline EPO levels ≤44 IU/L (ROC analysis; median, 28 vs 112 IU/L; p=0.0080), while there was a trend for inferior responses associated with RBC-TD (p=0.0857). Age, gender, baseline white blood cell and platelet counts, type of ESA used, BM RS %, JAK2V617F, SF3B1, DNMT3A, TET2 and ASXL1 mutational status did not impact ESA response (Table 1). On a univariate survival analysis (OS), survival was not significantly different between ESA responders vs non-responder (median OS, 76 vs 45 months; p=0.2929). After ESA failure, 7 (19%) patients received hypomethylating agents (HMA), while 5 (14%) received lenalidomide. There were no erythroid responses to HMA, while 1 of 3 (33%) assessable lenalidomide treated patients achieved a morphological complete response. Conclusion: Erythropoiesis stimulating agents are effective first line therapies for the management of anemia in patients with MDS/MPN-RS-T, with 46% achieving an erythroid response (median duration of response of 20 months). Low baseline endogenous EPO level (<44 IU/L) was the best predictor of response, with a trend towards an inferior response in the presence of RBC-TD. In the context of clonal thrombocytosis, potentially related thrombotic/ischemic events were documented in 8% of patients. Disclosures Al-Kali: Astex Pharmaceuticals, Inc.: Research Funding. Patnaik:Stem Line Pharmaceuticals.: Membership on an entity's Board of Directors or advisory committees.

Introduction: Myelodysplastic syndrome/myeloproliferative neoplasm (MDS/MPN) overlap syndromes consist of 5 distinct WHO-defined entities; namely chronic myelomonocytic leukemia (CMML), atypical chronic myeloid leukemia, BCR/ABL1- (aCML), juvenile myelomonocytic leukemia (JMML), MDS/MPN with ring sideroblasts and thrombocytosis (MDS/MPN-RS-T), and MDS/MPN, unclassifiable (MDS/MPN-U) (Arber et al., Blood 2016). With the notable exception of JMML, a bona fide RASopathy, the other entities are characterized by clinical heterogeneity and molecular diversity. Loss of function TET2 mutations (TET2MT) are common in myeloid neoplasms, especially CMML (60%), and are known leukemogenic drivers. We carried out this study to assess the TET2 mutational landscape and phenotypic correlates in patients with MDS/MPN overlap syndromes. Methods: After approval by the institutional review board, adult patients with WHO defined MDS/MPN overlap syndromes were included; with the exception of JMML. The BM morphology, cytogenetics and 2016, WHO-diagnoses were retrospectively reviewed and all patients underwent targeted next generation sequencing for 29 myeloid-relevant genes, obtained on BM mononuclear cells, at diagnosis, or at first referral, by previously described methods (Patnaik et al., BCJ 2016). Results: Five hundred and four patients were included in the study; including 387 (77%) with CMML, 48 (10%) with MDS/MPN-RS-T, 17 (3%) with aCML and 52 (10%) with MDS/MPN-U. The median age at diagnosis was 71 (range, 18-99) years, and 333 (66%) were male. TET2MT were seen in 212 (42%) patients, with the frequency of other mutations being: ASXL1 45%, SRSF2 40%, NRAS 15%, SF3B1 13%, CBL, RUNX1 and SETBP1 12% each, and JAK2 V617F 11% (Figure B). Among the MDS/MPN overlap syndromes, TET2 was more frequently mutated in CMML (49%) and aCML (47%) compared to MDS/MPN-RS-T (10%) and MDS/MPN-U (15%). The prevalence of patients with TET2MT increased with age, a finding consistent across all MDS/MPN subtypes (Figure C). Overall, 341 TET2MT were identified in 212 patients (mean 1.6 variants/patient, range 0-5): 120 (24%) had >1 TET2MT, while 113 (22%), 5 (1%) and 2 (0.4%) had 2, 3 and 5 mutations, respectively. CMML and aCML patients were more likely to have an age-independent increase in multiple TET2MT (28% and 24%), in comparison to MDS/MPN-RS-T (4%) and MDS/MPN-U (8%). TET2 MT spanned the entire coding sequence and were mostly truncating (78%, Figure A): 59 (17%) were missense, 14 (4%) involved the splice-donor/acceptor sites, 2 (0.5%) were in-frame deletions, 129 (38%) were nonsense, and 137 (40%) were frameshift mutations. Overall, the distribution of TET2MT was superimposable across CMML, aCML, and MDS/MPN-U; the only exception being the absence of splice site mutations in the latter two. One hundred and eighty-seven (55%) TET2MT were secondary to pathogenic single nucleotide variants (SNV), while the remainders were secondary to deletions (25%) and insertions (15%). Transitions comprised the most frequent type of SNV (65%), with the C:G>T:A being the most common (56%). Patients with MDS/MPN overlap syndrome and TET2MT were more likely to have additional gene mutations compared to wild type patients (mean mutation number 3.1 vs 2.1, p<0.0001), with common co-mutations being SRSF2 (51%), ASXL1 (42%), and CBL (17%). The median overall survival (OS) of the entire cohort was 29 (range, 0-170) months; 29 months for CMML, 63 months for MDS/MPN-RS-T, 14 months for aCML, and 25 months for MDS/MPN-U. On univariate analysis, OS was superior in CMML patients with TET2MT (35 months) compared to wild type cases (21 months, p<0.0001, Figure D), and in CMML patients with >1 TET2MT (41 months) in comparison to wild type (21 months, p<0.0001) and single TET2MT (29 months, p=0.0476) cases (Figure E). These observations were not seen in patients with aCML, MDS/MPN-RS-T, and MDS/MPN-U. Conclusion: Our study demonstrates that TET2MT are among the most frequent mutations in patients with MDS/MPN overlap syndromes (42%), especially CMML (49%), with an age-dependent increase in the frequency and number of TET2MT. Mutations in TET2 were found to span the entire coding sequence, with truncating mutations being more common (78%). Importantly, in CMML, TET2MT, including number of TET2MT, were associated with favorable survival outcomes. Figure Disclosures Al-Kali: Astex Pharmaceuticals, Inc.: Research Funding. Patnaik:Stem Line Pharmaceuticals.: Membership on an entity's Board of Directors or advisory committees.

Introduction: Isocitrate dehydrogenase 1 and 2 (IDH1/2) are metabolic enzymes in the citric acid cycle, producing alpha-ketoglutarate (αKG). Mutations in specific regions of these genes have been characterized in gliomas, AML and chronic myeloid neoplasms such as MDS and MPN. These mutations produce an oncometabolite 2-hydroxyglutarate (2HG) that disrupts the epigenetic phenotype of myeloid cells promoting oncogenesis. While this process has been characterized for arginine hot spot (R132/R140) mutations, there exist variants of unclear significance (IDHVUS), with potential pathogenic predictions using in silico assessment models. With the availability of mutant IDH inhibitors in practice, it has become important to characterize these VUS. We carried out this study to functionally interrogate potentially pathogenic IDHVUS in MDS and MPN. Methods: After institutional approval, bone marrow (BM) DNA of patients with MDS and MPN, including overlap syndromes were subjected to next generation sequencing (29-gene, including the entire coding region of IDH1/2) by previously described methods (Patnaik et al., BCJ 2016). A sub-set of patients with available plasma [IDH wild type (IDHWT), bona fide IDH mutations (IDHMT) and IDHVUS with predicted pathogenicity (≤0.01% minor allele frequency, and predicted damaging in silico scores)] were submitted for mass spectrometry (LC-MS) based detection of IDH-related metabolites; D and L isomers of 2HG, αKG, and total 2HG. The impact of variations on protein secondary structure was assessed by the GorIV prediction algorithm. Potential alterations within the tertiary structure of the proteins was evaluated by in silico mutagenesis. Results: One hundred and seventeen patients were included in the study; median age 68 years (range, 39-94), 68% male. Of these, 40 (34%) had MDS, 42 (36%) MPN, and 35 (30%) MDS/MPN overlap syndromes. Eighty six (74%) had IDHMT, including n=28 (24%) with IDH1MT (R132C/G/H/S) (11 MDS, 8 MPN, 9 MDS/MPN) and n=59 (50%) with IDH2MT (R140G/Q/W) (22 MDS, 20 MPN, 17 MDS/MPN). IDHVUS were detected in 32 (27%) patients, including 1 IDH1MT patient, and involved IDH1 in 14 (12%) cases and IDH2 in 18 (15%) cases. All IDHVUS were missense, except for a 2 amino acid insertion in IDH2, with the most frequent VUS being IDH1Y183C (7%) and IDH2T352P (7%). While there were no phenotypic differences between IDH1/2MUT and IDH1/2VUS, in comparison to patients with IDH1/2MT, those with IDH1VUS were more likely to have higher WBC counts (median 8.1 vs 3.6 x109/L, p=0.0322), higher neutrophil counts (median 3.3 vs 1.3 x109/L, p=0.0203), and lower BM blast % (median 1 vs 3%, p=0.02061). Compared to IDH1/2MT patients, those with sole IDH1/2VUS more frequently had mutations in SF3B1 (p=0.0167), while they were associated with a lower frequency of SRSF2 mutations (p=0.0002). Interestingly, TET2 mutations were detected in only 5% of patients with IDHMT and 16% of IDHVUS patients, although without reaching statistical significance. LC-MS analysis on 75 plasma samples (n=43 IDHMT, n=9 IDHVUS, n=20 IDHWT, 3 normal) revealed a significant increase in total 2HG levels in IDHMT vs IDHVUS (p=0.0089), and IDHWT (p=0.0020) cases (Figure 2A-D). While there were no significant differences in αKG levels between the groups, the D-2HG enantiomer levels were higher in IDHMT samples vs IDHVUS (p=0.0199) and IDHWT (p=0.0055). On examination of the crystal structures of the IDH genes using PDB, while we observed a direct interaction between R132/R140/R172 and the catalytic core, the IDHVUS were located away from this active center and were hence thought to be allosteric in nature (Figure 2F). These VUS however, did demonstrate predictive steric clashes in the tertiary structure context of IDH1 using the PyMol artificial mutagenesis tool. The absence of wild-type crystal structure for IDH2precluded a similar analysis for IDH2. Median overall survival (OS) of the entire cohort (Figure 1) was 36 (range, 0-190) months, with OS being inferior in IDH1/2MT patients (34 months) vs those with IDH1/2VUS (50 months, p=0.0255). Conclusion: Our study demonstrates that in spite of predictive pathogenicity of IDH1/2VUS using in silico predictive tools and crystal studies, the lack of involvement of the arginine hot spot precludes IDH neomorphic activity, as shown by normal 2HG levels. In addition, in comparison to IDHMT patients, those with IDHVUS had fewer BM blasts and a better survival. Disclosures Patnaik: Stem Line Pharmaceuticals.: Membership on an entity's Board of Directors or advisory committees.

Introduction: Germline RUNX1 mutations/deletions result in a Familial Platelet Disorder with propensity to Myeloid Malignancy (FPDMM); an autosomal dominant condition characterized by thrombocytopenia, qualitative platelet defects, and myeloid clonal evolution (MDS and AML). The advent of next generation sequencing (NGS) has allowed detection of clonal cytopenias of unclear significance (CCUS) prior to bone marrow (BM) morphological changes. Somatic RUNX1 mutations (RUNX1MT) can be seen in myeloid malignances, including MDS/MPN overlap syndromes, with an indeterminate prognostic impact. We carried out this study to define the landscape of clonal evolution in FPDMM patients and to compare germline RUNX1MT with somatic mutations. Methods After IRB approval, families with RUNX1MT FPDMM (detected by germline sequencing and deletion/duplication assays) and patients with 2016, WHO-defined MDS/MPN overlap syndromes (except JMML) were included in the study. Index patients with FPDMM underwent BM biopsies and in select cases NGS at diagnosis and then as indicated for follow up, or at disease progression. All MDS/MPN overlap syndrome patients underwent a 29-gene panel NGS assay on BM samples obtained at diagnosis by previously described methods (Patnaik et al., BCJ 2016). Results Eight index patients with FPDMM were included; median age at diagnosis 58 (range, 0-71) years, 3 (38%) males (Figure A). Four (50%) had intragenic deletions involving multiple exons in 3 cases and 2 nucleotides in 1 case, 1 (13%) each had missense, frameshift and splice site mutations, while 1 (13%) patient had a whole gene deletion (Figure C). Notably, in 4 (100%) of 4 patients with large deletions, NGS testing was negative for RUNX1 abnormalities. Four (50%) index patients presented with a myeloid neoplasm; JAK2, SRSF2, ASXL1 mutant CMML, MDS/MPN-U with trisomy 8, AML with del7q and normal karyotype AML (negative NGS). Three (38%) index patients had CCUS at diagnosis with TET2 (67%) being the most common acquired mutation, and others including SRSF2, EZH2, and SF3B1. Cumulatively, myeloid clonal evolution was seen in 7 (88%) patients. Three (38%) index patients (AML-2, CMML-1) underwent allogeneic HCT, and at last follow up were in remission. Median age at myeloid clonal evolution was 61 (range, 4-71) years; median overall survival (OS) from date of myeloid clonal evolution was 12 (range, 7-53) months. For the comparative analysis, 499 patients with MDS/MPN overlap syndromes were included; 382 (77%) with CMML, 48 (10%) with MDS/MPN-RS-T (MDS/MPN with ring sideroblasts and thrombocytosis), 17 (3%) with aCML (atypical CML), and 52 (10%) with MDS-MPN-U (unclassifiable). Median age at diagnosis was 71 (range, 18-99) years, and 329 (66%) were male. RUNX1MT were detected in 57 (11%) patients; 43 (11%) with CMML, 5 (30%) with aCML, and 9 (17%) with MDS-MPN-U. No RUNX1MT were found in MDS-MPN-RS-T. Overall, 62 RUNX1MT were identified in 57 patients, with 52 (39 CMML, 4 aCML, 9 MDS-MPN-U) and 5 (4 CMML, 1 aCML) patients harboring 1 and 2 RUNX1MT, respectively. RUNX1MT spanned the entire coding sequence without specific hotspots (Figure B): 27 (44%) were missense, 2 (3%) involved the splicing sites, 5 (8%) were nonsense and 28 (45%) were frameshift. There was only one RUNX1MT (R273K), seen in both germline and somatic patients (Figures B and C). Compared to wild type patients, RUNX1MT patients with MDS/MPN overlap syndromes had significantly lower platelet counts (median 72 vs 117 x109/L, p<0.0001), without any other phenotypic differences. In the somatic RUNX1MT cohort, co-occurring mutations included ASXL1 (54%), SRSF2 (54%), TET2 (49%), and NRAS (28%), with NRAS (28% vs 13%, p=0.0028), and SRSF2 (54% vs 39%, p=0.0218) mutations clustering with RUNX1MT. Median OS of the entire cohort was 29 (range, 0-170) months. On univariate analysis, while there was a trend for an inferior OS in RUNX1MTvs wild type patients (20 vs 29 months, p=0.0523), the leukemia-free survival was clearly inferior (p=0.0175, Figure D). Conclusion RUNX1 MT FPDMM is a germline predisposition syndrome with a high rate of myeloid clonal evolution (88%), including clonal cytopenias (38%). Germline RUNX1MT have minimal overlap with somatic mutations seen in myeloid malignancies, with NGS testing often missing RUNX1 deletions. Somatic RUNX1MT in myeloid malignancies are significantly associated with thrombocytopenia and an inferior leukemia-free survival. Figure Disclosures Patnaik: Stem Line Pharmaceuticals.: Membership on an entity's Board of Directors or advisory committees.

Abstract Background: Current prognostication in primary myelofibrosis (PMF) utilizes MIPSS70 (mutation-enhanced international prognostic scoring system for transplant-age patients), MIPSS70+ version 2.0 (karyotype-enhanced MIPSS70) and GIPSS (genetically-inspired prognostic scoring system, which is based on mutations and karyotype) (JCO 2018;36:310; JCO doi: 10.1200/JCO.2018.78.9867; Leukemia. 2018;doi:10.1038/s41375-018-0107). These contemporary models are inter-complimentary and highly efficient in predicting overall survival. However, specific risk factors for very long survival (e.g. 20+ years) and short-term mortality (i.e. death within 5 years of diagnosis) have not been systematically looked into. Methods: Study patients were recruited from the Mayo Clinic, Rochester, MN, USA. Diagnoses were according to the 2016 World Health Organization criteria (Blood. 2016;127:2391). Logistic regression statistics was used to identify predictors of 20-year survival and 5-year mortality. Variables evaluated in the logistic model included those that are currently listed in MIPSS70, MIPSS70+ version 2.0 and GIPSS, as well as age (≤70 vs >70 years) and sex. Receiver operating characteristic (ROC) curves and area under the curve (AUC) were used to estimate predictive accuracy. The JMP® Pro 13.0.0 software from SAS Institute, Cary, NC, USA, was used for all calculations. Results: A total of 1,282 consecutive patients with PMF (median age 65 years, range 19-92; 63% males) were used to identify 26 (2%) patients (median age 51 years, range 28-71; 38% males) who survived their disease for at least 20 years (very long-lived patients) and 626 (49%) patients (median age 68 years, range 27-92; 66% males) who died within 5 years of their diagnosis. Clinically-derived information, including the clinical variables used in MIPSS70 and MIPSS70+ version 2.0, with the exception of fibrosis grade (information available in 60%), was available in 99% of the study population; cytogenetic information was available in 94%, driver mutational status in 71% and the full profile of high molecular risk (HMR) mutations, including ASXL1, SRSF2, U2AF1 Q157, EZH2, IDH1 and IDH2, in 42%. MIPSS70+ version 2.0 risk distribution for the entire study population (n=1,282) was 4% very low risk, 16% low risk, 19% intermediate risk, 40% high risk and 21% very high risk; the corresponding percentages for the 26 very long-lived patients were 14%, 29%, 14%, 43% and 0% and for the 626 short-term survivors were 0%, 4%, 10%, 47% and 39% (p<0.001). Cytogenetic, driver mutation and HMR mutation information was available in 100%, 58% and 27% of the 26 long-term survivors, respectively, and 98%, 63% and 39% of the 626 short-term survivors. Multivariable logistic regression analysis of 20-year survival identified the following seven variables as being favorable: age ≤70 years (p=0.002); female sex (p=0.03); hemoglobin level ≥10 g/dl for women and ≥11 g/dl for men (p=0.03), leukocyte count ≤25 x 109/l (p=0.009), platelet count ≥100 x 109/l (p=0.002), circulating blasts ≤1% (p=0.03) and absence of constitutional symptoms (p=0.04). The particular phenotypic profile exhibited a high degree of predictive accuracy with an estimated AUC of 0.90 (Figure 1a); karyotype and mutations did not retain significance in this particular analysis; in fact, 5 (71%) of the 7 long-term survivors who were informative harbored at least one HMR mutation. A similar analysis for 5-year mortality identified the following risk factors: high molecular risk mutations (p<0.001); unfavorable or very high risk karyotype (p<0.001); absence of type 1/like CALR mutations (p<0.001); age >70 years (p<0.001); constitutional symptoms (p<0.001); hemoglobin level <10 g/dl for women and <11 g/dl for men (p<0.001); leukocyte count >25 x 109/l (p=0.004); and circulating blasts >1% (p=0.001); predictive accuracy was estimated at AUC 0.87 (Figure 1b). Conclusions: The phenotypic profile of long-lived patients in PMF includes young age (≤70 years old), female sex, asymptomatic disease (i.e. no constitutional symptoms) and absence of moderate to severe anemia, thrombocytopenia or leukocytosis and ≤1% circulating blasts; each one of these parameters, with the exception of female sex, were present in ≥80% of long-term survivors. Karyotype and mutations were more relevant in predicting short-term mortality. Disclosures No relevant conflicts of interest to declare.

Abstract Background : Cytogenetic abnormalities occur in approximately a third of patients with primary myelofibrosis (PMF) and 5-20% of those with essential thrombocythemia (ET) or polycythemia vera (PV). Abnormal karyotype in PV and specific cytogenetic abnormalities in PMF have been shown to adversely affect survival (Leukemia. 2018;32:1189; Br J Haematol. 2017 Jun 9. doi: 10.1111/bjh.14798). In the current retrospective study, we considered 650 Mayo Clinic patients with myeloproliferative neoplasms (MPN), including ET, PV and PMF, with at least two documented bone marrow (BM) biopsies, in order to examine the incidence and pattern of changes in karyotype and their clinical correlates. Methods : Study patients were recruited from the Mayo Clinic, Rochester, MN, USA, based on documentation of at least two serial BM biopsies. Diagnoses were according to the 2016 World Health Organization criteria (Blood. 2016;127:2391). BM biopsy #1 referred to the baseline karyotype at time of initial diagnosis/referral and BM biopsy #2 to the first repeat BM biopsy after diagnosis. In addition to documenting the presence or absence of changes in karyotype, a notation was made regarding disease phase, at the time of the repeat biopsy, in order to allow accurate interpretation of the data and evaluation of survival impact. Results: 650 patients with MPN, including 153 ET, 105 PV and 392 PMF, were included in the current study and had undergone at least two BM biopsies; 227 patients had three, 108 four, 48 five, 19 six, and 4 seven repeat BM biopsies during their clinical course. Cytogenetic clonal evolution in patients with normal karyotype at baseline : Baseline karyotype was normal in 466 (72%) patients, including 139 (91%) ET, 91 (87%) PV and 236 (60%) PMF. Clonal evolution, which constituted a change from "normal" to "abnormal" karyotype, was documented in 16 (12%) patients with ET, 22 (24%) with PV and 75 (33%) with PMF; the latter included favorable karyotype in 11% of the cases, unfavorable in 12% and very high risk (VHR) in 10%. 11 (69%) of the 16 documented cases of clonal evolution in ET, 10 (45%) of 22 in PV and 11 (15%) of 75 in PMF were accompanied by clinically evident progression to blast or fibrotic phase disease, at the time of the repeat BM biopsy. Cytogenetic clonal evolution in patients with abnormal karyotype at baseline : Baseline karyotype was abnormal in 187 (28%) patients, including 17 (9%) ET, 14 (13%) PV and 156 (40%) PMF. Additional changes in karyotype, during serial BM biopsies, were documented in 6 (35%) ET, 5 (36%) PV and 71 (46%) PMF patients. In PMF, the pattern of changes in karyotype included "favorable" to "favorable" in 7%; "favorable" to "unfavorable" in 13%; "favorable" to "VHR" in 5%; "unfavorable" to "unfavorable" in 13%; "unfavorable" to "VHR" in 3%; and "VHR" to "VHR" in 6%. All 11 (100%) patients with ET or PV who underwent clonal evolution displayed clinically overt disease transformation into leukemic or fibrotic phase disease, at the time of the repeat BM biopsy. Leukemic transformation at the time of the repeat BM biopsy was evident in 13 (18%) of the 71 PMF cases with clonal evolution. Survival impact of cytogenetic clonal evolution without change in disease phase : The survival impact of cytogenetic clonal evolution was examined under the following provisions: i) analysis was limited to events occurring at BM biopsy #2 with survival calculated from the date of BM biopsy #2; ii) patients with leukemic or fibrotic transformation at time of BM biopsy #2 were excluded from analysis; and iii) clonal evolution for the purposes of survival analysis constituted changes from "normal" to "abnormal", for ET and PV, while the type of changes were taken under consideration for PMF. Under these stipulations, clonal evolution appeared to affect survival in general, although statistical significance was apparent for only ET and PMF (Figures 1a, 1b and 1c). Conclusions: Cytogenetic clonal evolution in ET and PV is infrequent, in the absence of clinically overt disease transformation, and is more likely to occur in PMF. Such events, when documented in chronic phase disease, appeared to predict shortened survival in all three MPNs, which was most apparent in ET and with emergence of VHR or unfavorable abnormalities in PMF. These observations warrant prospective studies that account for indication bias. Disclosures No relevant conflicts of interest to declare.

Abstract Background : Survival prediction in essential thrombocythemia (ET) and polycythemia vera (PV) is currently based on clinically-derived risk variables, including age, thrombosis history and leukocyte count (Blood. 2012;120:1197; Leukemia. 2013;27:1874). We have previously reported on the prognostic contribution of mutations, in both ET and PV (Blood Adv. 2016;1:21). The current study examines the possibility of integrating genetic and clinical information for predicting overall (OS), leukemia-free (LFS) and myelofibrosis-free (MFFS) survival in ET and PV. Methods : Study patients were recruited from the Mayo Clinic, Rochester, MN, USA and the University of Florence, Florence, Italy, based on availability of next-generation sequencing (NGS)-derived mutation information. Diagnoses were according to the 2016 World Health Organization criteria (Blood. 2016;127:2391). NGS-detected coding region variants were filtered through the Exome Aggregation Consortium (ExAC) database and annotated by the Catalogue of Somatic Mutations in Cancer (COSMIC) database as mutants or variants of uncertain significance (VUS). Conventional statistics was employed for outcome analysis and assessment of model performance. Results: 906 molecularly-annotated patients (416 from Mayo and 490 from Florence), including 502 ET and 404 PV cases, were included in the current study. The Mayo/Florence cohorts included 270/232 ET (median age 57/54 years, 60%/59% females) and 146/258 PV (median age 63/58 years, 52%/44% females) patients; median follow-up was 9.9/12.9 years for ET and 10.7/12.4 years for PV; during this time 39%/38% deaths and 4.4%/6.5% leukemic and 16%/33% fibrotic progressions were documented for ET, with the corresponding figures for PV being 50%/26%, 4.8%/3% and 13%/36%. Cytogenetic information was available in 84%/63% ET and 80%/49% PV cases in the Mayo/Florence cohorts. In both Mayo and Florence cohorts, multivariable analysis of mutations identified primarily spliceosome mutations to adversely affect OS (SF3B1 and SRSF2 in ET and SRSF2 in PV) and MFFS (U2AF1 and SF3B1 in ET); in addition, TP53 mutations was associated with leukemic transformation in ET, in both patient cohorts. Other mutations with independently significant contribution, validated in one but not both cohorts, included EZH2 (OS and LFS in ET), IDH2 (OS and LFS in PV) and RUNX1 (LFS in PV and ET). For the purposes of the current study, mutations considered "adverse" for subsequent analysis and development of prognostic models required validation in both Mayo and Florence cohorts. Age-adjusted, all-inclusive multivariable analysis of genetic and clinical variables identified the following as independent risk factors for OS, in both Mayo and Florence cohorts: ET ─ SRSF2/SF3B1 mutations, age >60 years and male sex; PV ─ SRSF2 mutations, age >67 years and leukocyte count ≥11 x 109/l. In addition, multivariable analysis flagged leukocytosis in ET (Florence cohort only) and abnormal karyotype in PV (Mayo cohort only) as additional factors. Development of mutation-enhanced international prognostic systems for ET and PV In order to optimize the number of informative cases, the two Mayo and Florence databases were subsequently combined and subjected to multivariable analysis that confirmed the independent survival effect, in ET, of SRSF2/SF3B1 mutations (HR 2.8, 95% CI 1.8-4.3), age >60 years (HR 6.7, 95% CI 4.8-9.4) and male sex (HR 1.8, 95% CI 1.4-2.4) and, in PV, of SRSF2 mutations (HR 7.0, 95% CI 2.3-17.4), age >60 years (HR 5.7, 95% CI 3.3-10.1), and leukocyte count ≥11 x 109/l (HR 2.4, 95% CI 1.5-3.9); the combined analysis also flagged independent prognostic contribution from abnormal karyotype in PV (HR 2.1, 95% CI 1.1-3.6). HR-weighted risk point allocation resulted in new, mutation-enhanced 4-tiered risk models for ET (MIPSS-ET based on mutations, age and sex; figure 1a) and PV (MIPSS-PV based on mutations, karyotype, leukocyte count and age; figure 1b). Both models displayed high predictive accuracy and were internally validated by bootstrapping. The combined analysis also confirmed the impact of TP53 mutations on LFS (figure 1c) and U2AF1/SF3B1 mutations on MFFS (figure 1d), in ET. Conclusions: Spliceosome mutation information enhances survival prediction in ET and PV and identifies those at risk for fibrotic progression. TP53 mutations predict leukemic transformation in ET. Disclosures No relevant conflicts of interest to declare.

Abstract Background : Myeloproliferative leukemia virus oncogene (MPL)-mutated essential thrombocythemia (ET) is uncommon, with a cited incidence of less than 5%, while that of MPL-mutated primary myelofibrosis (PMF) is at least twice as frequent (Blood. 2006;108:3472;Blood.2008;112:141). MPL-mutated ET cohorts also have higher reported rates of fibrotic progression than their MPL wild type counterparts (Blood. 2014;124:2507). These observations suggest the possibility that some instances of MPL-mutated ET might actually represent prefibrotic PMF. Methods : Patients were recruited from institutional databases of the Mayo Clinic, Rochester, MN, USA. Diagnoses were according to the 2016 World Health Organization (WHO) criteria (Blood. 2016;127:2391). Study inclusion criteria required the availability of bone marrow biopsy slides for central review by one of the authors (C.A.H.). Central pathology review included assessment of bone marrow cellularity and extent of tri-lineage proliferation, megakaryocyte morphology and grading of reticulin fibrosis. Data was additionally collected from MPL-mutated patients with PMF, for comparison. Standard statistical methods were used for statistical analysis, including calculation of survival data (JMP® Pro 13.0.0, SAS Institute, Cary, NC, USA). Results: A total of 665 patients with ET were annotated for their driver mutational status; 18 (2.7%) were reported out as being MPL-mutated; by comparison, among 867 patients with PMF, 47 (5.4%) were signed out as MPL-mutated. Among the 18 cases with MPL-mutated ET, bone marrow slides were available for central pathology review in 14 patients (Table 1). The latter were subsequently reassigned the diagnosis of either prefibrotic PMF (n=8; 57%) or were felt to be morphologically consistent with true WHO-defined ET (n=6; 43%). Comparison of these two distinct histopathological patterns, i.e. true ET vs reassigned prefibrotic PMF, was respectively characterized by lower (median 35%, range 30-50) vs higher (median 65%, range 40-80) bone marrow cellularity (P<0.001), ET (n=6) vs PMF (n=8) consistent megakaryocyte morphology (P<0.001) and presence of tri-lineage proliferation (0% vs 100%; P<0.001); in contrast the degree of reticulin fibrosis was similar between the two (P=0.1) (Table 2). The reassigned prefibrotic PMF (n=8), vs confirmed ET (n=6), cases also displayed a higher frequency of increased serum levels of lactate dehydrogenase (60% vs 0%; P=0.02), higher likelihood of displaying hemoglobin levels below the sex-adjusted reference range values (29% vs 0%; P=0.1), leukoerythroblastosis (14% vs 0%; P=0.2), constitutional symptoms (13% vs 0%; P=0.2) and a higher incidence of thrombosis history at presentation (38% vs 0%; P=0.04) (Table 2). Interestingly, reassigned prefibrotic PMF also displayed a narrower MPL mutational spectrum compared to those confirmed as ET (MPLW515L/K incidence 100% vs 60%; P=0.04). The incidences of abnormal karyotype and high risk molecular mutations (ASXL1, SRSF2 and U2AF1) were similar between the two. We documented a higher incidence of post-diagnosis thrombosis in prefibrotic PMF (25% vs 0%; P=0.1) but similar rates of leukemic transformation (0% for both) and fibrotic progression (38% vs 33%); when all 665 ET patients were assessed for myelofibrosis-free survival, MPL-mutated cases (prior to central review) displayed significantly worse outcome, compared to patients with other driver mutations (Figure 1a). Median overall survival in confirmed MPL-mutated ET, ET re-classified as PMF, and MPL-mutated PMF was not reached, 11.6 and 5.3 years, respectively (confirmed ET vs PMF, P=0.01; ET re-classified vs PMF, P=0.04; confirmed ET vs ET re-classified as PMF, P=0.54) (Figure 1b). Conclusions: The current study suggests that the majority of routinely-assigned MPL-mutated ET probably represents prefibrotic PMF, when morphologically scrutinized. Prior to central pathology review, we show a higher rate of fibrotic progression in MPL-mutated ET, compared to patients with other driver mutations; accordingly, after central pathology review, the similar rate of fibrotic progression between morphologically confirmed MPL-mutated ET and those reassigned as prefibrotic PMF further suggests the latter to be biologically more akin to PMF, despite its ET-consistent morphologic features. Disclosures No relevant conflicts of interest to declare.

Abstract Background: Vitamin B1 (Thiamine) deficiency might result in Wernicke's encephalopathy (WE); the latter was reported in clinical studies with the JAK2 inhibitor fedratinib resulting in disruption of clinical development, despite efficacy in patients with myelofibrosis (MF) (JAMA Oncol, 2015, 5). Our objectives were i) to determine the incidence of thiamine deficiency in patients with myeloproliferative neoplasms (MPN), ii) assess clinical correlations of thiamine level and iii) determine the impact of JAK2 inhibitor therapy on thiamine level in MPN. Methods: After Institutional review board approval, patients referred with a suspected diagnosis of MPN were enrolled. All clinical and laboratory variables including treatment details were collected at the time of referral. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis of thiamine diphosphate in whole blood was performed at the time of referral. Reference range for thiamine level in whole blood was 70-180 nmol/L. The JMP® Pro 13.0.0 software from SAS Institute, Cary, NC, USA, was used for all statistical analysis. Results: A total of 115 patients were prospectively enrolled and included 28 patients without MPN and 87 with MPN. 87 MPN patients (median age, 65 years, 52% males) included 32 with primary myelofibrosis (PMF), 17 with polycythemia vera (PV), 17 with essential thrombocythemia (ET), 11 with post PV MF, 8 with post ET MF, and 2 with MPN-U. The MPN and non-MPN patients were similar in their gender distribution, pattern of alcohol use, malnutrition and multivitamin use (p=0.24, 0.12, 0.54 and 0.78 respectively), although the former were older (p=0.09). Median thiamine level of the study patients was 167 nmol/L (range; 60-442 nmol/L) with only two (1.7%) female patients displaying level below the normal reference range. One belonged to the non-MPN group (thiamine level 68 nmol/l); a 26 year old female with erythroid leukemia in the absence of alcohol use or malnutrition; the second patient (thiamine level 60 nmol/l) was a 38 year old female with ET, CALR type 2 mutated on hydroxyurea, without a history of alcohol use or malnutrition. Overall, both MPN and non-MPN patients displayed similar thiamine level (p=0.89). Clinical correlations of thiamine level in MPN Median thiamine level was similar amongst post ET MF, ET, PMF, PV, and post PV PMF patients (127, 162, 166, 190 and 209 nmol/L respectively, p=0.13). No significant effect on thiamine level was apparent for age, gender, hemoglobin, platelet count, dynamic international prognostic scoring system (DIPSS) score, unfavorable cytogenetics, palpable splenomegaly, or constitutional symptoms. Interestingly, thiamine level was significantly higher in JAK2 mutated cases (median 177 nmol/L vs 145 nmol/L in unmutated cases, p=0.03). Furthermore, when analysis was restricted to PMF patients, thiamine level positively correlated with JAK2V617F mutant allele burden (p=0.01) and leukocyte count > 10 x 109/L (p=0.03). JAK2 mutated PMF patients with leukocyte count > 10 x 109/L depicted significantly higher thiamine levels (median 313 nmol/L) compared to JAK2 mutated patients without leukocytosis (median 164 nmol/L) and JAK2 unmutated patients with or without leukocytosis (median 192 nmol/L and 143 nmol/L respectively)(p=0.004, Figure 1a). Impact of treatment on thiamine level Treatment with hydroxyurea was documented in 35 patients with no significant impact on thiamine level (p=0.28). 27 patients were on JAK inhibitors (median duration at referral; 41 months); 12 patients on ruxolitinib and 15 MF patients on momelotinib as part of a clinical trial. Thiamine level was similar amongst JAK inhibitor treated vs untreated patients (median 166 vs 169.5 nmol/L respectively, p=0.72). Annual thiamine measurements were performed on 15 patients treated with momelotinib (median duration 52 months, range 18-60 months). Despite fluctuations in thiamine levels, only two patients were noted to have thiamine level below normal (65 and 69 nmol/L at year 3), which subsequently improved to 101 and 121 nmol/L by year 4, while still on treatment (Figure 1b). Conclusions: The current study confirms the rarity of subnormal thiamine levels in MPN and the limited impact of treatment on thiamine levels, including with JAK2 inhibitors. Our novel observation regarding the correlation of thiamine level with JAK2V617F mutant allele burden and leukocyte count requires further exploration. Disclosures No relevant conflicts of interest to declare.

Abstract Background: Young patients with myeloproliferative neoplasms (MPN), including essential thrombocythemia (ET), polycythemia vera (PV) and primary myelofibrosis (PMF), constitute a distinct, steadily growing subpopulation (Eur J Haematol. 2017;98:85). Most available data, including current prognostic models, are excerpted from older populations and may not accurately mirror the unique phenotypic and prognostic patterns seen in younger patients. The objective of the current study was to comprehensively assess the phenotypic features and long-term outcomes in MPN patients age ≤ 40 years. Methods: The study population was recruited from a master database of MPN patients seen at our institution between 1967 and 2017. Diagnoses were in accordance with 2016 World Health Organization criteria (Blood. 2016;127:2391). Data was abstracted from the time of referral, which coincided in the majority with the time/within one year of diagnosis. Additional molecular data was obtained based on availability of next-generation sequencing (NGS) information. Conventional statistics was utilized for group comparisons and calculation of survival data. Results: The study population was recruited from a total of 3,023 MPN patients: 665 PV, 1076 ET, and 1282 PMF. Of these, 361 (12%) were ≤ 40 years of age at the time of initial diagnosis; the corresponding incidences were 12% in PV (n=79), 20% in ET (n=219) and 5% in PMF (n= 63). Young PV patients (n=79): Median age (range) was 32 years (18-40) and males constituted 54% (Table 1). Over a median follow-up of 11.3 years, fibrotic transformation rates were 22% vs 25% and 10% in those 41-60 and >60 years of age, respectively (P<0.001). Leukemic conversion rates were balanced across age categories (4% vs 5% vs 3% for successively older cohorts) (P=0.4). Interestingly, young patients had lower rates of arterial (P<0.001) and higher of venous thrombosis (P=0.01) prior to/at diagnosis vs older cohorts while events post-diagnosis were comparable across age groups (P=0.6). Young ET patients (n=219): Median age (range) was 32 years (18-40) with a minority (31%) constituted of males. During a median follow-up of 13 years, fibrotic progressions rates were 16% vs 16% and 9% in those aged 41-60 and >60 years, respectively (P<0.001). The leukemic conversion rate (2%) was analogous to that of older cohorts (5% and 3%, respectively) (P=0.1). Young patients also experienced significantly fewer arterial thrombotic events prior to/at diagnosis (6% vs 14% and 18%) and after diagnosis (8% vs 16% and 22%) compared to older cohorts (P<0.0001). Young PMF patients (n=63): Median age (range) was 37 years (19-40) with a preponderance of males (59%). High risk karyotype was not seen in young patients vs in 5% and 8% of those 41-60 and > 60 years of age, respectively (P=0.004). During median follow-up of 7.1 years, leukemic conversions were recorded in 10% of young patients vs 10% and 9% of respectively older cohorts (P=0.9). Young subjects also experienced fewer arterial thrombotic events prior to/at diagnosis (2% vs 6% and 13%) and lower rates of venous thrombosis after diagnosis (2% vs 10% and 4%) compared to cohorts 41-60 and > 60 years of age, respectively (P<0.0001). NGS data disclosed significant clustering of high risk molecular mutations with successively older subgroups (≤ 40 vs 41-60 vs > 60 years); ASXL1: 19 vs 40 vs 45% (P=0.005), SRSF2: 6 vs 8 vs 18% (P=0.0004), and U2AF1: 3 vs 11 vs 19% (P=0.004) (Table 1). Survival data Kaplan-Meier survival curves stratified by age bracket (≤ 40 vs 41-60 vs > 60 years) disclosed median OS in successively older age groups to be 37, 22, and 10 years for PV (Figure 1A), 35, 22, and 11 years for ET (1B), and 20, 8, and 3 years for PMF (1C); exposing significantly longer median survival estimates in younger patients, systematically across all disease subtypes (P<0.001). When appraised as a function of MPN subtype, survival data confirmed the significantly reduced OS in PMF relative to PV and ET (P<0.001), though PMF patients aged ≤ 40 years were still projected to have a reasonably favorable median life expectancy of 20 years (Figure 2). Conclusions: Young MPN patients comprise a unique disease subset defined by an attenuated-risk cytogenetic and mutational backdrop compared to their older counterparts. Caution must be exercised in counseling and managing this population as historical 'one-size-fits all' data fails to reflect the true natural history and disease biology in the young. Disclosures No relevant conflicts of interest to declare.