ALBERT

Description: Myeloid sarcoma (MS) is defined as a tumor mass consisting of myeloid blasts with or without maturation occurring at an anatomical site other than bone marrow (BM). MS may occur before, concurrently or after a characterized acute myeloid leukemia (AML). Cytogenetic abnormalities are found in 50% of the cases but molecular alterations are less well described and involved FLT3 and/or NPM1 mutations. Mutations in IDH1 and IDH2 genes are found in 15% to 20% of patients with AML but have never been described in MS. Mutated IDH enzymes produce in vast excess D-2-hydroxyglutarate (2-HG) in leukemic cells, which can act as a biomarker predictive of the presence of IDH1 and IDH2 mutations. As availability of DNA sequencing techniques on paraffin samples are limited, molecular characterization of MS remained difficult. We asked whether in MS, serum 2-HG would predict the presence of IDH1/2 mutations at diagnosis, and could provide a biomarker for follow up. Tissue samples and serum samples from 8 patients with a MS diagnosis were analyzed. High quality genomic DNA was extracted from frozen MS samples using conventional phenol/chloroforme extraction procedures. Exon 4 of IDH1 and IDH2 genes (IDH1/R132 and IDH2/R140 and /R172 codons) was amplified by PCR using HotStar Taq polymeraze (Qiagen) and primers. Direct sequencing was performed using the Sanger method as previously described. In case of MS relapse or AML evolution, IDH1 and IDH2 genes were analyzed in the same way from frozen tissue sample or bone marrow sample. Serum samples at MS diagnosis were analyzed for total 2-HG, D-2-HG and L-2-HG by reverse-phase liquid chromatography coupled to mass spectrometry. In case of myeloid sarcoma with IDH1/2 mutation, 2-HG values were compared to 18F-FDG-PET results when available during remission phase and at relapse. Three patients (3/8; 37.5%) had an IDH2 R140Q mutation at diagnosis of MS localized to lymph node, soft tissue, skin or pharynx. At MS diagnosis, serum total 2-HG, D-2-HG and ratio D/L-2-HG were significantly higher in case of myeloid sarcoma with IDH2 R140Q mutation compared to patients with no IDH mutation (Table 1). Serum total 2-HG level ≥2µM or D-2-HG level ≥1.8µM or ratio D/L 2-HG 〉2.5 were significantly associated with the presence of IDH2 mutation (Fisher's exact test P≤0.02). Table 1. Myeloid sarcoma with IDH2 R140Q mutation (N=3) Myeloid sarcoma without IDH2 R140Q mutation (N=5) Median total 2-HG (µM) 4.1 (range: 3.1-30.1) 1.4 (range: 1-1.6) Median D-2-HG (µM) 3.7 (range: 2.3-28) 0.6 (range: 0.5-0.8) Median L-2-HG (µM) 0.8 (range: 0.4-2.1) 0.8 (range: 0.4-0.8) Median ratio D/L 2-HG 8.3 (range: 2.9-18.8) 1 (range: 0.7-1.7) All 3 patients with IDH2 R140Q mutated MS received intensive chemotherapy treatment and achieved complete remission (CR). Two patients relapsed: one experienced isolated extramedullary relapse (thigh muscle); one had a bone marrow relapse. IDH2 R140Q mutation was found at the site of relapse in both cases. When available, serum 2-HG values and 18F-fluorodeoxyglucose-positron-emission tomography (FDG-PET) were compared at different time points (at diagnosis, remission and relapse; Table 2). Table 2. Patient #1 Patient #2 Patient #3 FDG-PET at diagnosis (SUVmax) - 17 5.25 Serum 2-HG at diagnosis (µM) 4.1 3.1 30.1 FDG-PET in remission (SUVmax) 0 - 9.6 (N=4) 0 (N=4) - Serum 2-HG in remission (µM) 0.5 - 1.1 (N=4) 0.6 - 3.1 (N=4) 3.4 - 17.9 (N=3) FDG-PET at MS relapse/evolution to AML (SUVmax) 6.1 - 0 Serum 2-HG at MS relapse/evolution to AML (µM) 2.3 - 16.7 Time between diagnosis and MS relapse/evolution to AML (months) 30 - 9 Serum 2-HG values were in accordance with FDG-PET interpretations except in patient #1 who presented a transient hypermetabolic splenic nodule (SUVmax 9.6) without serum 2-HG increase. Patient #2 remained in CR but had recently increased 2-HG values without overt relapse. Patient #3 presented relapse as a refractory anemia with excess of blast without extra-medullary localization. FDG-PET didn't find any abnormality contrary to the persistent increased value of serum 2-HG (total 2-HG: 16.7µM). These data show that myeloid sarcoma can be associated with IDH2 R140Q mutation and suggest that 2-HG measurement in the serum predicts the presence of IDH1/2 mutations at diagnosis. During follow-up, serum 2-HG values could be representative of the disease status. Because of IDH inhibitors promising results in AML, 2-HG screening at MS diagnosis could be useful. Disclosures Ribrag: Celgene: Research Funding; Esai: Membership on an entity's Board of Directors or advisory committees, Research Funding; Pharmamar: Honoraria, Membership on an entity's Board of Directors or advisory committees; Gilead: Membership on an entity's Board of Directors or advisory committees; Servier: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding. De Botton:Agios pharmaceuticals: Research Funding.

Description: Introduction: Mutations in IDH1 and IDH2 genes are found in 15% to 20% of patients with acute myeloid leukemia (AML). Mutated IDH enzymes produce in excess the D stereoisomer 2-hydroxyglutarate (2-HG) in leukemic cells, which can act as a biomarker predictive of the presence of IDH1 and IDH2 mutations. Total serum 2-HG (D and L stereoisomers) 〉2µM and ratio D/L 2-HG 〉2.5 are strong predictors of the presence of IDH1/2 mutations at diagnosis (Janin et al; JCO 2014). We measured 2-HG levels in serial samples of 47 patients with IDH1/2 mutations during induction or first salvage therapy to determine whether 2HG can serve as a surrogate marker of treatment efficacy. Methods: Between 2007 and 2015, 47 AML patients with IDH1 or IDH2 mutation received intensive chemotherapy as induction therapy (n=42) or as first salvage regimen (n=5) in our center (Table 1). Genomic DNA was extracted from BM samples using conventional procedures. IDH1-(codon R132) and IDH2- targeted (codons R140 and R172) regions were amplified by PCR with primers containing Ion Torrent adapters and unique barcodes to generate libraries. Pooled amplicons libraries were clonally amplified on Ion Spheres using the Ion Xpress Template 200 Kit. All available serum samples (n=394) prior and during induction/salvage chemotherapy were analyzed for total 2-HG, D-2-HG and L-2-HG by reverse-phase liquid chromatography coupled to mass spectrometry. We studied the serum 2-HG values (D-2-HG and ratio D/L 2-HG) evolution during IC according to complete remission (CR) evaluated at the end of IC. When a patient had a missing value for 2-HG, the missing value was replaced by the last available observation (up to 7 days before the missing value). At each day we tested the difference between the dosage values of the responders and the non-responders with non-parametric Wilcoxon 2-sided tests. The analyses were done with SAS software (version 9.3; SAS Institute, Cary, NC). Table 1. Patients and treatment characteristics: AML at diagnosis / RR AML (N) 42 / 5 Median age (years) 58.2 (23.1 - 75.7) Sex ratio (M/F) 23/24 Conventional cytogenetic: - Normal karyotype - Trisomy 8 - Monosomy 7 39 (83%) 3 (6%) 2 (4%) Associated gene mutations: - NPM1- FLT3-ITD- FLT3-TKD 24 (51%) 10 (21%) 5 (11%) IDH mutation: - IDH1 R132 (N) - IDH2 R140 (N) - IDH2 R172 (N) 8 (17%) 32 (65%) 7 (15%) Intensive treatment: - "3+7" regimen - Intensified induction chemotherapy - High-dose cytarabine based chemotherapy 24 (51%) 19 (40%) 4 (8%) Results: Complete remission (CR) 1 was obtained in 35/42 patients (83%) and CR2 in 3/5 (60%) after IC. All (8/8) IDH1 -R132 mutated patients achieved CR1, 25/28 (89%) IDH2 -R140 mutated patients and 3/4 achieved CR1 and CR2 respectively; 2/6 (33%) IDH2 -R172 mutated patients and 0/1 achieved CR1 and CR2, respectively. Median serum 2-HG values before IC were as follow: total 2-HG 10.9 µM (range: 2.1-135.5); D-2-HG 10.7 µM (range: 1.7-132.6) and ratio D/L 36.9 (range: 3.9-131.2). Median number of serum 2-HG values per patient was 7 (range: 5-10). Serum D-2-HG and ratio D/L 2-HG evolution according to CR status after IC began to differ significantly at day 5 after IC (Wilcoxon test P=0.02, Figure 1). The difference between responders and non-responders increased overtime (Wilcoxon test P

Description: In 2015, a germline copy number variation of chromosome 14 (CNVdup14) including ATG2B and GSKIP genes was described as a predisposition genetic factor responsible of familial myeloproliferative neoplasms from French West Indies (Saliba et al, Nat Gen 2015), frequently progressing to AML. In this study, we looked at the presence of this CNVdup14 in a cohort of Caribbean islands patients (pts) with non-secondary aggressive hematological malignancies (HM). We also studied the expression of ATG2B and GSKIP genes in a cohort of acquired AML pts. This is a retrospective multicenter study of adults Caribbean islands pts treated at Gustave Roussy Cancer Center (Villejuif, France) and at the French West Indian hospitals (Martinique and Guadeloupe) between May 2000 and May 2018. We included pts with AML, acute undifferentiated leukemia (AUL), acute lymphoblastic leukemia (ALL) and lymphoblastic lymphoma (LL). Pts with personal history of myeloproliferative or myelodysplastic syndromes before the onset of aggressive HM were not included in this study. The presence of the CNVdup14 was carried out by PCR analyses in all the pts. For the second part of the study, expression of ATG2B and GSKIP genes were assessed in newly diagnosed de novo AML pts with normal karyotype or trisomy 8 (samples from the GOELAMSTHEQUE) by quantitative RT-PCR and expressed as relative expression PPIA/HPRT/H2A.Z. One hundred pts were analyzed. Median age was 52 years (IQ 40-62) with male predominance (61%). Fifty eight pts came from Martinique, 42 pts from the rest of the Caribbean islands (including 28, 4, 3, 2 pts from Guadeloupe, Haiti, Saint Martin, Dominican Republic, respectively). Seventy eight pts had AML. Among them, according to revised MRC cytogenetic classification, 11 (14%) were favorable, 47 (60%) intermediate and 20 (26%) adverse. Seventeen pts had ALL, 3 LL, and 2 AUL. On the entire cohort, all except nine pts were treated with intensive chemotherapy, 80 reached complete remission, 29 relapsed, 46 pts died. Thirty two pts received hematopoietic stem cell transplantation (HSCT). Six pts were positive for the CNVdup14 by PCR (confirmed by SNP array in the 5 pts with leftover DNA available). All had an AML (no pts with favorable AML) and were originated from Martinique. These pts represented 14% of the 43 AML from Martinique in our cohort (17% if we excluded favorable AML). One was known to be part of an ATG2B/GSKIP family, 2 pts had no familial history of myeloid malignancies and 2 new families were discovered. Median white blood cell, hemoglobin and platelets counts were 17.7 G/l (IQ 6.7-48), 8.15 g/dl (6.9-9.7) and 58 G/l (20-132), respectively. Median age at AML diagnosis was 49 years (34-55), 3/6 (50%) had extramedullary localization compared to 11/78 (14%) for others AML pts. Karyotype was normal for 4, or showed a monosomy 7 for 2 pts. NGS panel showed distinct abnormalities compared to the entire cohort (Fig A). None had JAK2, MPL, CALR, P53, RUNX1, DNMT3A, FLT3-ITD mutations. All harbored an epigenetic and/or spliceosome mutation (IDH n=3, TET2 n=3, ASLX1 n=3, SRSF2 n=3). Five out of the six pts received intensive treatment and 4 achieved complete remission. Two received HSCT, 2 relapsed and 4 died. Median overall survival (OS) of the entire cohort was 35.7 months (22.5-89.5) and progression free survival (PFS) 27.6 months (15.6 -56.1). As CNVdup14 pts had AML only, we next evaluated survival according to the predisposition status in the AML cohort. Pts with CNVdup14 had a median OS and PFS of 19 (6.5-29) and 11.4 (6.5-29) months, respectively, compared with 52.6 (22.9-100.2) and 30 months (15.6-60) in CNV wild-type counterparts (PFS Fig B). We next evaluated ATG2B and GSKIP expression in a cohort of 46 random de novo AML pts (GOELAMS-LAM-IR-2006 multicenter trial). Median expression of ATG2B and GSKIP were 4.8 (2.6-12.9) and 5.3 (0.3-5.1) respectively. No CNVdup14 was detected. Interestingly we found a correlation between the two genes expression (Pearson Correlation Coefficients 0.55 and linear regression p〈 0.001, Fig C). Expression of ATG2B and GSKIP was also correlated with leukocytosis (p=0.003 and p=0.07) (Fig D). We found no impact on OS and PFS. For the first time, we described a high percentage of the germline CNVdup14 in de novo AML pts from Martinique (14%). Moreover evaluation of ATG2B and GSKIP expression suggested that the role of theses 2 genes in leukemogenesis is not limited to pts with the CNVdup14. Figure. Figure. Disclosures de Botton: Agios: Research Funding; Celgene: Honoraria, Research Funding. Benabelali:CERBA laboratory: Employment.

Description: Key Points The use of low-molecular-weight heparin did not improve live-birth rates in nonthrombophilic women with consecutive recurrent miscarriage. Prophylactic doses of low-molecular-weight heparin should no longer be prescribed in this clinical setting.

Description: There is a Blood Commentary on this article in this issue.

Description: Introduction: Therapy-related Myeloid Neoplasms (TRMN) arise after cytotoxic chemotherapy and/or radiotherapy administered for a prior neoplasm and have dismal outcome. Inherited predisposition or direct induction of fusion transcripts can be responsible of TRMN. Recent evidence suggests that patient with Clonal Hematopoiesis of Indeterminate Potential (CHIP) may have an increased risk of TRMN. In gynecological and breast cancers, CHIP mutations have been described in around 25 % of patients (Coombs et al., Cell Stem Cell 2017). In this setting, we aimed to identify the impact of CHIP-associated mutations in overall survival of TRMN. Methods: In this retrospective study, we included patients with TRMN diagnosed and/or treated at Gustave Roussy Cancer Center between January 2004 and December 2018 if they had a previous breast or gynecological cancer, DNA samples available at TRMN diagnosis and a signed informed consent. We performed a targeted 77 genes mutational analysis using Next Generation Sequencing (NGS), using Haloplex technique (Agilent), sequencing on MiSeq (Illumina). If any somatic mutation associated with hematological malignancies could define CHIP, the most frequent genes mutated in original CHIP papers are: ASXL1, ASXL2, ATM, BCOR, CBL, CHEK2, DNMT3A, IDH1, IDH2, JAK2, PPM1D, SF3B1, SRSF2, TET2 and TP53. These will define "CHIP-associated mutations" at TRMN diagnosis. According to results, patients were classified into "CHIP-associated mutations" or "no-CHIP" categories (no mutations detected or all the other mutations detected at TRMN diagnosis). Moreover, patients were also classified into 3 subgroups according to a modified genetic ontogeny-based classifier (Lindsley et al., Blood 2015): "P53/PPM1D" subgroup, "MDS and AML with MDS mutations" subgroup, and "de novo/pan-AML" subgroup. Survival analyses were performed using GraphPad software. Results: 77 patients were identified: 49 therapy-related AML (t-AML) (64%) and 28 therapy-related MDS (t-MDS) (36%). Median age at TRMN diagnosis was 62 years [36-86] and median time interval between primary cancer and TRMN was 5.1 years. Primary cancers were breast (70%), ovarian (23%), endometrial (4%) and cervical (3%) cancers. Patients were treated with radiotherapy alone (13%), cytotoxic agent alone (19%), or chemotherapy/radiotherapy (68%). The most frequently mutated genes at TRMN diagnosis were: TP53 (31%), DNMT3A (19%), NRAS (13%), TET2 (12%), NPM1 (10%), PPM1D (9%), PTPN11 (9%) (Fig1A). 10% of patients had no gene mutation detected. According to 2017 ELN risk stratification, genetic risk for t-AML was favorable, intermediate and adverse in 19 (39%), 14 (28%) and 16 patients (33%), respectively. According to IPSS score, 86% of the t-MDS patients were classified as High risk/Intermediate 2 and 14% as Intermediate 1/Low risk. Treatment options included best supportive care for 16 patients (21%), low dose chemotherapy for 26 patients (34%), or intensive chemotherapy/allogenic transplant for 34 patients (45%). Based on Lindsley's modified classifier median overall survival for "P53/PPM1D", "MDS" and "de novo/pan-AML" subgroups were 12, 17 and 25 months, respectively (p=0.009) (Fig1B). "CHIP-associated mutations" were detected in 53 patients (69%) with no significant impact on overall survival (Fig1C). Interestingly, age at TRMN diagnosis in patients with "CHIP-associated mutations" vs patients with "no-CHIP" was higher (65 vs 56 years old, p=0.002) and the time interval between cancer diagnosis and TRMN was longer (6.6 [0.9-38.1] vs 2.9 [1.2-8.5] years, p〈 0.001) (Fig1D). CHIP emergence was not correlated with type of cancer's treatment or with number of treatment lines. "P53/PPM1D" subgroup was more frequent in patients treated with 2 lines or more for their primary cancer than in patients who received only 1 line of treatment (50% and 25% respectively, p=0.03). Conclusion: TRMN occurring after gynecological or breast cancers are of bad prognosis, especially for P53 and PPM1D mutated patients. Our results show that CHIP related mutations are found in a large percentage of patients and could be responsible for emergence of TRMN, especially in older patients. Figure 1 Disclosures de Botton: Forma: Consultancy, Research Funding; Bayer: Consultancy; Daiichi: Consultancy; Novartis: Consultancy; Astellas: Consultancy; Abbvie: Consultancy; Pierre Fabre: Consultancy; Syros: Consultancy; Agios: Consultancy, Research Funding; Janssen: Consultancy; Pfizer: Consultancy; Servier: Consultancy; Celgene: Consultancy, Speakers Bureau. Micol:AbbVie: Consultancy; Jazz Pharmaceuticals: Consultancy.

Description: Telomeres - the terminal regions of human chromosomes, and enzyme telomerase - a ribonucloprotein that synthesizes telomeric DNA onto chromosomal ends, have been thoroughly investigated as potential markers for the prognosis of various cancers including leukemia. However, it is important to consider both parameters and only few studies have investigated the prognostic value of these two combined biomarkers in patients with acute myeloid leukemia (AML). Our work was designed to determine the impact of telomere length together with telomerase activity (TA) on survival in patients with AML. In this current retrospective study, TA (reflected by the quantitative expression of the catalytic subunit of telomerase i.e the hTERT mRNA/18s RNA ratio measured by Q-RT-PCR) and telomere length (determined by southern blot analysis of terminal restriction fragments) were assayed in the bone marrow of 40 patients diagnosed with AML between 1999 and 2003 at Institute Gustave Roussy’s division of Hematology. The patients’ characteristics are shown on table 1. All patients were treated according to standard AML-type chemotherapy protocols. The median of TA (hTERT mRNA/18s RNA ratio) was 0.0458. TA was not detectable in 4 patients. The median of telomere length was 7.6 Kb (range: 3.5–11.2 Kb). No correlation was found between TA and telomere length. A negative correlation existed between telomere length and age (r= −0.42; p=0.0097). The Kaplan-Meier statistical method and logrank test were used for univariate survival analysis and the Cox proportional hazard regression models for multivariate survival analysis. In multivariate analysis, when adjusted for age (〉= 50 years versus younger), cytogenetics findings (poor prognosis versus others) and the nature of leukemia (secondary versus de novo), improved survival was found in patients with a combination of short telomere length (=0.09) (hazard ratio=9.91; 95% CI: 1.75–56.03; p=0.01). Our results suggest that the combination of telomere length and telomerase activity can be considered as an independent prognostic factor for survival in patients with AML. Table 1: Patients characteristics Sex (no. of patients) * AML post solid tumor (n=7), post myelodysplastic syndrome (n=1), post chronic myeloid leukemia (n=1) M 18 F 22 Age (years) Median 50 Range 22–74 Leukocyte count (Giga/L) Median 24.2 Range 1.3–360 Bone marrow blast percentage Median 76.5 Range 20–99 FAB Classification (no. of patients) M0 6 M1 9 M2 5 M4–M5 12 M4Eo 3 M6 1 Biphenotypic AL 3 NK AML 1 Type of leukemia (no. of patients) De novo 31 Secondary 9* Prognosis (based on karyotype) Good 5 Intermediate 21 Poor 10 Missing 4

Description: Point mutations in isocitrate dehydrogenase (IDH) define distinct subsets of acute myelogenous leukemia (AML). IDH is a metabolic enzyme that interconverts isocitrate and α-ketoglutarate (α-KG), but cancer-associated point mutations in IDH1 and IDH2 confer a neomorphic activity that allows reduction of α-KG to the oncometabolite R-2-hydroxyglutarate (2-HG). High levels of 2-HG have been shown to inhibit α-KG-dependent dioxygenases including histone and DNA demethylases, which play a key role in regulating the epigenetic state of cells, but the relationship between 2-HG and oncogenesis is not completely understood. Consistent with 2-HG promoting cancer via an effect on chromatin structure, patients harboring IDH mutations display a CpG island methylator phenotype (CIMP) and several studies have shown that overexpression of IDH mutant enzymes can induce histone and DNA hypermethylation as well as block cellular differentiation. In addition, mice engineered to express IDH1-R132H in hematopoietic tissue have increased early hematopoietic progenitors, splenomegaly, anemia, hypermethylated histones and altered DNA methylation patterns similar to those found in AML patients harboring IDH1/2 mutations.[i] Taken together, these data suggest that cancer-associated IDH mutations may induce a block in cellular differentiation to promote tumorigenesis. To investigate whether selective pharmacological inhibition of the mutant IDH1 enzyme could provide an effective way to lower intracellular 2-HG levels and restore normal differentiation, we treated TF-1 cells or primary human AML patient samples expressing mutant IDH1 with AG-120, an oral, selective, first-in-class, potent IDH1 mutant inhibitor currently in phase I clinical trials. Treatment with AG-120 decreased intracellular 2-HG levels, inhibited growth factor independent proliferation and restored erythropoietin (EPO)-induced differentiation in TF-1 IDH1-R132H cells. Similarly, pharmacological inhibition of mutant IDH1 enzyme with AG-120 in primary human blast cells cultured ex vivo provided an effective way to lower intracellular 2-HG levels and induced myeloid differentiation. Taken together, these data demonstrate that AG-120 is effective at lowering 2-HG levels and restoring cellular differentiation, and support further clinical development of this compound. Figure 1: Diagnosis and karyotypes of primary AML patient samples used in ex vivo studies Figure 1:. Diagnosis and karyotypes of primary AML patient samples used in ex vivo studies PB = peripheral blood, BM = bone marrow Figure 2: Percent 2-HG remaining relative to DMSO control after 6-day treatment with AG-120 in IDH1 R132H or IDH1 R132C patient samples Figure 2:. Percent 2-HG remaining relative to DMSO control after 6-day treatment with AG-120 in IDH1 R132H or IDH1 R132C patient samples or following 6 days of treatment with control (DMSO) or AG-120 (0.5, 1.0, and 5.0 μM) Figure 3: Relative proportion of cell types in human AML bone marrow samples untreated Figure 3:. Relative proportion of cell types in human AML bone marrow samples untreated [i] M. Sasaki et al., IDH1(R132H) mutation increases murine haematopoietic progenitors and alters epigenetics. Nature 488(7413):656-9, 2012. Disclosures Hansen: Agios Pharmaceuticals: Employment, Stockholder Other. Quivoron:Institut National de la Santé Et de la Recherche Médicale (INSERM): Grant Other; Association Laurette Fugain: Grant, Grant Other; Institut National du Cancer (INCa): Grant, Grant Other; Association pour la recherche contre le Cancer (ARC): Grant, Grant Other; AGIOS: Grant Other. Straley:Agios Pharmaceuticals: Employment, Stockholder Other. Lemieux:Agios Pharmaceuticals: Employment, Stockholder Other, US20130190249 (pending) Patents & Royalties. Popovici-Muller:Agios Pharmaceuticals: Employment, Stockholder Other. Fathi:Agios Pharmaceuticals: Advisory board participation Other. Gliser:Agios Pharmaceuticals: Employment, Stockholder Other. David:Institut National de la Santé Et de la Recherche Médicale (INSERM): Grant Other; Institut National du Cancer (INCa): Grant, Grant Other; Association pour la Recherche contre le Cancer (ARC): Grant, Grant Other; Association Laurette Fugain: Grant, Grant Other; AGIOS: Grant Other. Bernard:Institut National de la Santé Et de la Recherche Médicale (INSERM): Grant Other; Association Laurette Fugain: Grant, Grant Other; Institut National du Cancer (INCa): Grant, Grant Other; Ligue Nationale contre le cancer (LNCC): Grant, Grant Other; AGIOS: Grant Other. Dorsch:Agios Pharmaceuticals: Employment, Stockholder Other. Yang:Agios Pharmaceuticals: Employment, Stockholder Other. Su:Agios Pharmaceuticals: Employment, Stockholder Other. Agresta:Agios Pharmaceuticals: Employment, Stockholder Other. de Botton:AGIOS: Grant Other. Penard-Lacronique:Institut National de la Santé Et de la Recherche Médicale (INSERM): Grant Other; Association Laurette Fugain: Grant, Grant Other; Institut National du Cancer (INCa): Grant, Grant Other; Association pour la recherche contre le Cancer (ARC): Grant, Grant Other; AGIOS: Grant Other. Yen:Agios Pharmaceuticals: Employment, Stockholder Other.

Description: Mutations in genes encoding RNA splicing factors constitute the most common class of alterations in patients with myelodysplasticsyndromes (MDS). These occur predominantly as heterozygous point mutations at restricted residues in SF3B1, SRSF2, and U2AF1 in a mutually exclusive manner, suggesting that spliceosomal gene mutations confer gain-of-function with converging biological effects. However, recent studies suggest that mutations in each splicing factor result in distinct alterations in pre-mRNA splicing. It is therefore unclear if such mutual exclusivity is due to overlapping biological effects and/or synthetic lethal interactions. Furthermore, although cells bearing mutant splicing factors have been shown to require the wildtype allele for survival, whether these mutations can exist in a homozygous state is unknown. Here we addressed these questions by analyzing the effects of expressing SF3B1 and SRSF2 mutations simultaneously or in a homozygous state in vivo. Re-analysis of published sequencing data revealed that only 2% of MDS patients (86/4,032) have mutations in 〉1 splicing factor simultaneously (whether these mutations are present in the same cell or not is unclear). Of these 86 patients, co-mutations in SRSF2 and SF3B1 represent the most prevalent combination (n=23/86) with all SRSF2 mutations affecting the P95 residue while all co-existing SF3B1 mutations occurred outside of the most commonly mutated K700 residue. To understand the basis for exclusivity of SRSF2P95 and SF3B1K700 mutations, we generated mice for inducible heterozygous expression of Sf3b1K700E/+ and Srsf2P95H/+ mutations simultaneously (Mx1-cre Sf3b1K700E/+/Srsf2P95H/+). We next performed competitive bone marrow transplantation (BMT) assays where each mutation was induced, alone or together, following stable engraftment (Figure 1A). Simultaneous expression of Sf3b1K700E and Srsf2P95H mutations resulted in severe defects on the self-renewal and differentiation of hematopoietic stem and progenitor cells (HSPC), which were outcompeted by wildtype and single-mutant HSPCs (Figure 1B). In noncompetitive BMT assays, HSPCs co-expressing Sf3b1K700E and Srsf2P95H mutations had severe defects in multi-lineage reconstitution (Figure 1C). Analyses of hematopoietic organs 6 months post-BMT revealed a near complete absence of Sf3b1K700E/+/Srsf2P95H/+ double-mutant cells, which was distinct from expression of Sf3b1K700E or Srsf2P95H mutationalone. In addition, mice with conditional homozygous expression of the SRSF2P95H mutation (Mx1-cre Srsf2P95H/P95H) had severe defects in HSPC self-renewal as well as multi-lineage reconstitution, analogous to those seen with hemizygous Srsf2P95H expression (Mx1-cre Srsf2P95H/KO) (Figure 1D). As noted earlier, SF3B1 and SRSF2 mutations cause different effects on mRNA splicing. However, there has never been a direct comparison of the effects of each of these mutations in an isogenic context. To address this and to understand the mechanistic basis for exclusivity of these mutations, we performed RNA-seq on lineage- c-Kit+ cells from Mx1-cre Sf3b1K700E/+/Srsf2P95H/+ andcontrols 2 weeks after conditionally expressing each mutation alone or together. As evidence of the intolerability of combined SF3B1/SRSF2 mutations, mean allelic ratio of Sf3b1K700E and Srsf2P95H expressed in double-mutant mice was 20.7% and 33.5%, respectively, markedly lower than the near 50% expression seen in single-mutant controls. Despite this, principle component analysis of differentially spliced genes revealed distinct changes mediated by expression of Sf3b1K700E and Srsf2P95H mutations (Figure 1E). Moreover, previously described changes to alternative 3" splice site selection as well as cassette exon splicing were seen in cells bearing Sf3b1K700E and Srsf2P95H mutation, respectively, as well as in Sf3b1K700E/+/Srsf2P95H/+ double-mutantcells. These findings indicate thatspliceosomalgene mutations, despite imparting distinct alterations on gene expression and splicing, are not tolerated when co-expressed in the same cell, thus providing a basis for their strong mutual exclusivity in MDS. These data, combined with the fact that neitherhemizygousnor homozygous expression of splicing factor mutations is tolerated, further establishes the unique requirement ofspliceosomalmutant cells on the remaining function ofwildtypespliceosomecomponents. Figure 1. Figure 1. Disclosures Palacino: H3 Biomedicine Inc.: Employment. Seiler:H3 Biomedicine: Employment. Buonamici:H3 Biomedicine: Employment. Smith:H3 Biomedicine Inc.: Employment.