Brunner, Andrew M.

Abstract Myelodysplastic syndromes (MDS) are clinically and biologically diverse disorders, emphasizing the need for personalized treatment approaches. The International Working Group for Prognostication of MDS (IWG_PM) recently introduced a molecular classification, referred to as the MDS taxonomy, that categorizes patients into 16 subgroups based on 21 gene mutations, 6 cytogenetic abnormalities, and loss of heterozygosity (LOH) at TP53 and TET2 loci. This study sought to validate and enhance the clinical relevance of the MDS taxonomy by analyzing a large retrospective cohort ( n = 5136) and transcriptomic data from a prospective cohort ( n = 477). The taxonomy successfully identified subgroups with distinct clinical characteristics and disease progression patterns. However, incorporating gene interactions from taxonomy subgroups did not improve the prognostic performance of the Molecular International Prognostic Scoring System (IPSS‐M). We further assessed whether the taxonomy could guide management in patients receiving disease‐modifying therapies. Except for the “ TP53 ‐complex” subgroup, taxonomy classifications were not predictive of hypomethylating agent response or transplant outcomes. Nonetheless, they correlated with overall survival, suggesting that while both IPSS‐M and the taxonomy capture disease biology, other non‐genetic factors may influence treatment response. RNA sequencing confirmed the biological distinctiveness of the taxonomy groups. Transcriptomic profiling of CD34+ bone marrow cells revealed unique, homogeneous gene expression patterns, particularly within the AML‐like, bi TET2 , SF3B1 , and TP53 ‐complex subgroups. Further integration of multi‐omics data may refine MDS classification, improving clinical decision‐making and guiding the development of targeted therapies.

Abstract Introduction: HMA monotherapy (HMAm) continues to be the standard of care for HR-MDS. While the addition of VEN to azacitidine (AZA) improved overall survival (OS) and complete remission (CR) rates among older pts with acute myeloid leukemia (AML) over AZA monotherapy, the recently completed phase III VERONA trial of AZA+/-VEN in HR-MDS failed to show an OS benefit for AZA/VEN over AZA monotherapy. Using the icMDS VALIDATE database, we compared outcomes of pts with HR-MDS treated with HMA or HMA/VEN as frontline therapy. Methods: The icMDS VALIDATE database includes over 4000 adult MDS pts who received HMA-based therapy from 32 international centers in 9 countries. For this analysis, we included pts with HR-MDS defined by International Prognostic Scoring System [IPSS] ≥1.5 or IPSS-revised [IPSS-R] ≥4 (at diagnosis or HMA initiation) and treated with HMAm or HMA/VEN as frontline therapy for HR-MDS. We assessed responses based on IWG 2023 criteria (Zeidan A et al, Blood 2023). Composite complete remission (cCR) was defined as CR+CRbi+CRuni+CRh+CRequivalent. We calculated OS from time of HMA initiation to date of death from any cause. Pts lost to follow-up were censored at the date they were last known to be alive. We used Kaplan-Meier methods to calculate OS and compared OS between different groups using the log-rank test. We used Cox and logistic regression models to adjust for baseline prognostic variables as appropriate. The study was funded by an independent grant from Abbvie. Results: A total of 1907 pts met eligibility and were included (1552 AZA, 221 decitabine, 134 HMA+VEN). 479 pts subsequently underwent allogeneic stem cell transplant (allo-SCT; 430 after HMAm [32.6%]; 49 after HMA/VEN [37.1%]). Our cohort was enriched for pts with adverse disease characteristics including complex karyotype (33.6%) and TP53 mutations (28.7%). Compared to pts treated with HMA/VEN, pts treated with HMAm were more likely to be male (p=0.012), have lower baseline platelet counts (p=0.037), peripheral blood (PB) blasts (p=0.004), or bone marrow blasts (p<0.001), and were less likely to have very high risk IPSS-M (p<0.001), TP53 mutations (p<0.001), or complex karyotype (p=0.020). In unadjusted analyses, the rate of cCR was significantly lower in HMAm treated pts compared to HMA/VEN treated pts (27.7% vs. 48.8%, p<0.001). However, CR rate was not significantly different (11.7% vs. 17.1% for HMAm and HMA/VEN, respectively, p=0.22). After adjusting for age, sex, and IPSS-M, pts treated with HMA/VEN were more likely to achieve cCR compared to HMAm with an odds ratio (OR) of 2.49 (95% CI: 1.56-3.96, p<0.001). When separated by TP53 mutation status, the OR for cCR for HMA/VEN vs HMAm was 4.17 (95%CI: 2.14-8.31; p<0.001) and 1.27 (95%CI: 0.54-2.86; p=0.57) for TP53 wild-type and TP53-mutant pts, respectively. In unadjusted analyses, median OS was 17.4 months for HMAm and 19.6 months for HMA+VEN, and OS did not differ significantly by treatment type (p=0.33). For HMAm treated pts, the OS rates at year 1, 2, 3 were 0.65 (95%CI: 0.63-0.67), 0.38 (95%CI: 0.36-0.40), and 0.27 (95%CI: 0.25-0.29), respectively. For HMA/VEN treated pts, the OS rates at year 1, 2, 3 were 0.64 (95%CI: 0.56-0.73), 0.45 (95%CI: 0.36-0.56), and 0.37 (95%CI: 0.27-0.51), respectively. In a multivariate analysis adjusting for age, sex, PB blast, IPSS-M and allo-SCT as a time-varying covariate, treatment with HMA/VEN vs HMAm did not significantly influence OS (hazard ratio (HR) of OS for HMA/VEN vs HMAm: 0.89, 95%CI: 0.69-1.13, p=0.33). When stratified by TP53 mutation status, HR of HMA/VEN vs. HMAm was favorable for TP53 wild-type (HR: 0.47; 95%CI: 0.29-0.74) but not TP53-mutant pts (HR=0.96; 95%CI: 0.66-1.41). Adjusting for other variables in the model, HR for pts who underwent vs. those who did not undergo allo-SCT were 0.65 (95% CI: 0.49-0.85) and 0.36 (95%CI: 0.24-0.53) in TP53 wild-type and mutant patients, respectively. Conclusion: In this large, international retrospective analysis from the icMDS VALIDATE database, HMA/VEN was associated with higher cCR rates—but not CR rates—compared to HMAm among pts with HR-MDS, though we observed no significant differences in OS. In our analysis, clinical benefit from the addition of VEN was most clear in TP53 wild-type pts, and allo-SCT was associated with improved OS regardless of baseline TP53 status. Our analysis was limited by the relatively small number of HMA/VEN-treated pts and heterogeneous real-world treatment patterns.

Abstract Introduction TP53MT are associated with poor outcomes in MDS, especially in the context of biallelic MT. We have previously reported on outcomes of pts with HR-MDS with TP53MT who received HMA in the VALIDATE dataset (Kewan T et al, ASH 2023). Here we provide expanded analyses focusing on outcomes post allo-HCT, and we explore the prognostic significance of different classification methods for biallelic TP53MT. Methods The VALIDATE database includes 4,732 pts with MDS who received HMA-based frontline therapy from 32 international centers in 9 countries. For this study, only pts with HR-MDS with molecular data at time of diagnosis were included. Biallelic TP53MT status was assessed using three classification schemes: WHO 2022, ICC 2022, and a VAF-based Cleveland Clinic Foundation (CCF) method.1TP53 copy neutral loss of heterozygosity (CN-LOH) was evaluable in 386 pts (118 had TP53MT). Treatment responses were evaluated using IWG 2023 criteria (Zeidan A et al, Blood 2023). Composite complete remission (cCR) was defined as CR+CRbi+CRuni+CRh+CRequivalent. We calculated OS from time of HMA initiation to date of death or last follow up. Unsupervised clustering (MOSAIC framework) was applied to identify molecular subgroups within the TP53MT cohort. The study was funded by an independent grant from Abbvie. Results A total of 1,485 pts were included. Median age was 69 years (IQR: 62–76), with 66% male pts. Median bone marrow blast was 7%, and 28% had a complex karyotype. Treatment regimens included azacitidine (68%), decitabine (15%), HMA/Venetoclax (8%), and other HMA combinations (9%); 34% underwent allo-HCT. Overall, 392 (26%) had TP53MT with 430 TP53 mutations (missense: 316; truncating: 89). Biallelic TP53MT % varied by classification: 72% (WHO), 84% (ICC), and 85% (CCF). Compared to TP53 wild-type (TP53WT) pts, pts with TP53MT had worse OS (median: 11.5 vs. 23.9 months [mo], p< 0.001), were younger (68 vs. 70 years, p=0.039), and had lower counts at diagnosis. Complex karyotype (81% vs. 14%) and chromosomal abnormalities were more frequent in TP53MT pts. Pts with TP53MT also had fewer co-mutations in other genes. Response to HMA was evaluable in 1,094 pts: cCR and CR were 32% and 13%, respectively. TP53MT pts had higher cCR (39% vs. 29%, p=0.001) than TP53WT, but similar CR rates (16% vs. 12%, p=0.093). According to WHO 2022, biallelic TP53MT had higher CR rates than monoallelic TP53MT (18% vs. 8%, p=0.03), but similar cCR. ICC and CCF classifications did not differentiate response rates. Median OS did not differ by TP53MT allelic status using WHO or ICC definitions. However, using the CCF method, pts with biallelic TP53MT had shorter OS than monoallelic TP53MT (10.4 vs. 17.2 mo, p=0.038). In multivariable models adjusted for age, gender, and allo-HCT, only the CCF-defined biallelic TP53MT group had significantly worse OS (HR: 1.6; 95% CI: 1.1–2.4). TP53MT pts undergoing allo-HCT had improved OS regardless of allelic status. Median OS for TP53MT pts with allo-HCT was 18.9 vs. 9.2 mo without all-HCT. This benefit was observed across WHO, ICC, and CCF classifications. However, a TP53MT VAF ≥19% predicted worse OS even post allo-HCT (14.6 vs. 47.1 mo, p<0.001). At 1 year, the survival rate was 78% (95%CI: 69-89%) in TP53MT pts who underwent allo-HCT and had cCR vs. 68% (95%CI: 55-84%) in pts who did not have cCR. However, 1 year survival rate was significantly higher for pts with TP53MT who did not have allo-HCT and achieved cCR (60%, 95%CI: 48-76%) vs. no cCR (26%, 95%CI: 20-35%). Unsupervised clustering of 335 TP53MT pts identified 9 molecular clusters (MCs). MC2, characterized by low TP53MT VAF (median 17%), fewer co-mutations, and frequent del17p, was associated with significantly improved OS (17.9 mo) and predicted favorable outcome (HR for death: 0.6; 95% CI: 0.4–0.8). Other clusters were enriched for adverse features (e.g., high VAF, multi-hit TP53MT).Conclusions The CCF method appears to best classify biallelic TP53MT pts with worse survival in the absence of accurate assessment of LOH status. Treatment response per IWG 2023 for TP53MT pts was associated with OS only in non-transplanted pts. Allo-HCT improved survival across TP53MT subsets, but TP53 VAF ≥19% at diagnosis remained a strong negative prognostic factor. Integrative clustering revealed a less aggresive molecular subset among TP53MT pts, suggesting refined risk stratification may guide therapy in this high-risk population. Ref: 1.Bahaj et al. J Hematol Oncol 2023.

Abstract Background: Mutations in ethanolamine kinase-1 (ETNK1) have been identified in a wide range of myeloid neoplasms (MN), but their clinical significance remains poorly defined. Key gaps remain, including the true prevalence, variant- and disease-specific co-mutation patterns, cytogenetic patterns, and prognostic impact, highlighting the need for broader multi-institutional studies to refine our understanding and risk stratification. We present findings from an ongoing multi-institutional study of ETNK1-mutated (ETNK1mut) MN. Methods We analyzed 106 patients with ETNK1mut MN across the Mayo Clinic Comprehensive Cancer Center (Minnesota, Arizona, and Florida), H. Lee Moffitt Cancer Center (Tampa, FL), Massachusetts General Hospital (Boston, MA), and University of Milano-Bicocca (Monza, Italy), integrating clinical, hematologic, cytogenetic, and next-generation sequencing (NGS) data. Appropriate IRB approval was obtained. Diagnoses were assigned by the 5th edition of the World Health Organization and 2022 European LeukemiaNet criteria. A clone was considered dominant if it had the highest variant allele frequency (VAF) or VAF within 5% of the highest co-mutation. Overall survival (OS) was assessed by Kaplan-Meier. Comparison cohorts for OS included 83 ETNK1 wild-type (ETNK1wt) myelodysplastic syndrome (MDS) and 415 ETNK1wt acute myeloid leukemia (AML) patients without TP53 mutations. GraphPad Prism v10 was used for statistical analysis. Results Median age at diagnosis was 70 years (range 30.4-90), with a male predominance of 72.6% (n=77). ETNK1mut were identified in MDS (n=47, 44.3%), myelodysplastic/myeloproliferative neoplasms (MDS/MPN; n=23, 21.7%), myeloproliferative neoplasms (MPN; n=13, 12.3%), AML (n=12, 11.3%), clonal cytopenia of undetermined significance (CCUS; n=5, 4.7%), and other entities (n=6, 5.7%) including atypical chronic myeloid leukemia (n=1) and systemic mastocytosis (n=1). Over a median follow-up of 33.6 months (IQR 15-91.8), 49.1% (n=52) patients died and 19.4% (n=18 of 93 at risk) progressed to AML. Cytogenetics were available in 103 patients; 57.3% (n=59) had a normal karyotype, while 30.1% (n=31) had abnormal and 12.6% (n=13) had complex karyotypes. None harbored classical MDS-associated deletions such as del(5q), del(7q), or del(20q). The most frequent cytogenetic abnormalities included translocations (n=17), monosomy 7 (n=9), and trisomy 8 (n=8). Of 17 translocations, 8 (47%) involved chromosome 1, two of which (11.8%) involved der(1;7)(q10;p10). The most common ETNK1 mutation was N244S (n=70, 66.0%), followed by K378N (n=12, 11.3%) and G245 variants (n=9, 8.5%), including G245V and G245D. VAF data were available for 103 patients; the median ETNK1 VAF was 31.4% (range 2.0-54.8%). High VAF (≥40%) was observed in 27.2% (n=28) of patients. ETNK1 was the dominant mutation in 33% (n=34) of cases, shared equal dominance with a co-mutation in 13% (n=14), and was a subclone in 56% (n=58). Co-mutations were detected in 77% (n=82) of patients, with a median of 4 (range 0-13) co-mutated genes. A third of patients (n=37, 35%) had ≥5 co-mutations. The most frequently co-mutated genes were epigenetic regulators including ASXL1 (n=29, 27%), TET2 (n=17, 16%), EZH2 (n=12, 11%), and DNMT3A (n=12, 11%), followed by splicing factor genes including U2AF1 (n=13, 12%), SRSF2 (n=10, 9%), and ZRSR2 (n=10, 9%). Mutations in signaling genes (JAK2, NRAS, SETBP1) and transcription factors (RUNX1, BCOR) were also observed. High VAFs were seen in co-mutations of EZH2 (median 49%), TET2 (46%), and NRAS (46%). ASXL1 was most enriched in MDS/MPN (n=8, 35%) and MPN (n=3, 23%). Survival analysis showed better survival in ETNK1mut compared to ETNK1wt within AML (65 vs 19 months; p=0.2002) and MDS (107 vs 61 months; p=0.1993), although statistical significance was not reached. Conclusion ETNK1 mut frequently co-occur with mutations in epigenetic and splicing regulators, and demonstrate distinct molecular and cytogenetic features, including the absence of classic MDS-associated deletions and a high prevalence of translocations involving chromosome 1. ETNK1 was the dominant clone in a third of cases with a high co-mutation burden. Overall survival was better in ETNK1mut AML and MDS, but the difference was not statistically significant. These findings highlight the need for larger collaborative studies.

Abstract Introduction: Marrow complete remission (mCR) in 2006 IWG criteria for MDS (Cheson B et al, Blood 2006) reflects achieving bone marrow (BM) blast percentage below 5% but is frequently not associated with meaningful count recovery. While mCR + hematologic improvement (HI) does reflect wide ranges of count improvement or transfusion reductions, HI is difficult to measure accurately in routine clinical practice. The 2023 International Working Group (IWG) response criteria (Zeidan A et al, Blood 2023) eliminated the mCR category without meaningful count recovery and reclassified mCR associated with meaningful count recovery into specific “less-than-CR” categories. Still, many clinical trials continue to use mCR as a secondary efficacy endpoint or as a part of overall response rate (ORR). Within the icMDS VALIDATE database, we assessed the association of mCR with OS in hypomethylating agent (HMA)-treated higher risk (HR)-MDS patients among both (1) transplanted and (2) non-transplanted patients. Methods We included patients with HR-MDS (defined as IPSS ≥1.5 and/or IPSS-R >3.5) who received HMA-based therapies from the international icMDS VALIDATE dataset. We evaluated the best response by IWG 2006 criteria into CR, mCR, or no CR/mCR. We then reclassified patients with mCR per IWG 2023 criteria into either (A) having achieved a composite CR (cCR) defined as CR + CR equivalent (CRequ) + CR with bilineage count recovery (CRbi) + CR with unilineage count recovery (CRuni) + CR with partial hematologic recovery (CRh) or (B) not having achieved cCR (no cCR). OS was calculated from time of HMA initiation to date of death or last follow-up by using 6-month landmark analyses. We used Kaplan-Meier methods to estimate OS, and the Log-rank test to compare OS. VALIDATE is supported by an independent grant from AbbVie. Results A total of1049 patients met eligibility.Median age of patients was 69 (range 21-95), 64% were men. While 61% received azacitidine, 17% received decitabine, and 22% received HMA-based combination therapy. In total, 37% underwent subsequent allogenic stem cell transplantation (allo-HCT). Based on IWG 2006 criteria, 17%, 33%, and 50% achieved CR, mCR, and no CR/mCR as their best response, respectively. Patients achieving CR had superior median OS (25 months, 95% CI: 21-34 months) compared to those with mCR (16 months, 95% CI: 13-18 months; p=0.038) and no CR/mCR (15 months, 95% CI: 12-17 months; p=0.001). In patients who received a subsequent allo-HCT, the median OS for patients with mCR (30 months, 95% CI: 21-54 months; 1 year OS: 71%, 2 year OS: 55%) did not significantly differ compared to patients with CR (59 months, 95% CI: 25 months-not reached, p=0.80; 1 year OS: 73%, 2 year OS 62%) and patients with no CR/mCR (32 months, 95% CI: 21-45 months, p=0.46; 1 year OS: 69%, 2 year OS 54%). In contrast, in patients treated with HMA therapy alone without subsequent allo-HCT, median OS for patients with mCR (11 months, 95% CI: 9-15 months; 1 year OS: 46%, 2 year OS: 23%) did not significantly differ compared to patients with no CR/mCR (11 months, 95% CI: 9-14 months, p=0.85; 1 year OS: 47%, 2 year OS: 22%) but was significantly shorter compared to patients with CR (20 months, 95% CI: 17-28 months, p<0.001; 1 year OS: 69%, 2 year OS: 46%). Patients with mCR by IWG 2006 (N=347) were reclassified per IWG 2023 into having either achieved a cCR (N=219, 63%) or no cCR (N=128, 37%). Specifically, patients with mCR were reclassified per IWG 2023 into CR (4.9%), CRequ (0.6%), CRbi (26.8%), CRuni (29.4%), CRh (1.4%), PR (0%), HI (10%), no response (21.9%) and PD (4.9%). Patients with mCR who were reclassified into cCR had an improved median OS (16 months, 95% CI: 15-26 months; 1 year OS: 62%; 2 year OS: 43%) compared to patients reclassified into no cCR per IWG 2023 (13 months, 95% CI: 11-17 months, p=0.023; 1 year OS: 52%; 2 year OS: 31%). Conclusions Using the large icMDS VALIDATE dataset, we found that achievement of mCR by 2006 IWG criteria was not associated with improved OS compared to patients who did not achieve CR/mCR. Reclassification of mCR using IWG 2023 highlights the heterogeneity within the mCR group and suggests that IWG 2023 criteria stratifies outcomes in patients previously labeled as mCR by IWG 2006. IWG 2023 should be the standard used for ORR assessment in all future clinical trials involving patients with HR-MDS.

The NCCN Guidelines for Myelodysplastic Syndromes (MDS) provide recommendations for the evaluation, diagnosis, and comprehensive care of patients with MDS based on a review of recent clinical evidence that has led to important advances in treatment or has yielded new information on biologic factors that may have prognostic significance in MDS. The multidisciplinary panel of MDS experts is convened at least on an annual basis. During the annual meeting, the panel evaluates new and emerging data to inform their recommendations. These NCCN Guidelines Insights review the recent updates, including treatment recommendations both for lower-risk and higher-risk MDS, preference stratification of therapeutic agents, and emerging data on novel therapeutics.

Abstract Clonal cytopenia of undetermined significance (CCUS) represents a distinct disease entity characterized by myeloid-related somatic mutations with a variant allele fraction of ≥2% in individuals with unexplained cytopenia(s) but without a myeloid neoplasm (MN). Notably, CCUS carries a risk of progressing to MN, particularly in cases featuring high-risk mutations. Understanding CCUS requires dedicated studies to elucidate its risk factors and natural history. Our analysis of 357 patients with CCUS investigated the interplay between clonality, cytopenia, and prognosis. Multivariate analysis identified 3 key adverse prognostic factors: the presence of splicing mutation(s) (score = 2 points), platelet count of <100 × 109/L (score = 2.5), and ≥2 mutations (score = 3). Variable scores were based on the coefficients from the Cox proportional hazards model. This led to the development of the clonal cytopenia risk score (CCRS), which stratified patients into low- (score of <2.5 points), intermediate- (score of 2.5 to <5), and high-risk (score of ≥5) groups. The CCRS effectively predicted 2-year cumulative incidence of MN for low- (6.4%), intermediate- (14.1%), and high-risk (37.2%) groups, respectively, by the Gray test (P < .0001). We further validated the CCRS by applying it to an independent CCUS cohort of 104 patients, demonstrating a c-index of 0.64 (P = .005) in stratifying the cumulative incidence of MN. Our study underscores the importance of integrating clinical and molecular data to assess the risk of CCUS progression, making the CCRS a valuable tool that is practical and easily calculable. These findings are clinically relevant, shaping the management strategies for CCUS and informing future clinical trial designs.

Background. The inclusion of gene mutations and chromosomal abnormalities in the 2022 WHO and ICC Classifications of MDS has enhanced diagnostic precision and is expected to improve clinical decision-making process. Although these two systems share similarities, clinically relevant discrepancies still exist and potentially cause inconsistency in their adoption in a clinical setting. In this study on behalf of the International Consortium for MDS (icMDS), we adopted a data-driven approach to provide a harmonization roadmap between the 2022 WHO and ICC classification for MDS. A modified Delphi Process consensus approach is currently ongoing among icMDS experts to finalize a harmonized MDS classification scheme. Methods. We analyzed retrospective international cohorts of patients with a diagnosis of MDS (n=7017) and AML (n=1002) according to WHO 2016 criteria. Hierarchical Dirichlet Processes were applied to define clusters capturing broad dependencies among all gene mutations and cytogenetic abnormalities. To investigate the features of importance and their impact on the clustering process, we employed the SHapley Additive exPlanations approach (SHAP). This allowed to define harmonized labels for each clinical entity. The clinical relevance of the unsupervised clustering was assessed through the analysis of phenotypic features and outcomes among each group. ( Blood 2022;140: 9828-9830) Results. Patients' characteristics are summarized in Table 1. We identified 9 clusters, defined by specific genomic features. The cluster of highest hierarchical importance was characterized by biallelic inactivation of TP53 (biTP53). According to SHAP analysis, bi TP53 was defined as 2 or more TP53 mutations, or 1 mutation with copy number loss or cnLOH. Most patients assigned to bi TP53 cluster had TP53 VAF>10% (77.9%) and complex karyotype (70.1%). Assignment to bi TP53 cluster was irrespective of blast count. Patients with monoallelic TP53 mutation segregated into other clusters. Hierarchically, the second cluster included patients with del(5q). SHAP analysis highlighted 5q deletion alone, or with one other chromosomal abnormality other than -7, and absence of bi TP53, as the most relevant features. Most of these patients had blast counts <5% (88.1%). The third distinct cluster included patients with SF3B1 mutations (in the absence of concurrent del(7q), abn3q26.2, complex karyotype or RUNX1 mutation). Most patients with MDS and SF3B1 mutation had <5% blasts (94.2%). Common co-mutated variants in the SF3B1 cluster included mutant DNMT3A (25.2%) and TET2 (38.3%). Morphologically defined MDS cases (i.e., not meeting criteria for bi TP53, del(5q) or SF3B1) were preferentially assigned to the following additional clusters: SF3B1 and concurrent higher-risk mutations (e.g., RUNX1 and ASXL1); SRSF2 and concomitant TET2 mutations; U2AF1 mutations with del(20q), del(7q) or -7; SRSF2 with TET2 mutations and co-mutational patterns including RUNX1 and ASXL1; and AML-like genomic signatures. Our analyses suggest that morphologically defined MDS is characterized by a large heterogeneity in terms of mutation profiles, not entirely captured by the presence of unilineage versus multilineage dysplasia, percentage of bone marrow blasts, and presence of hypocellularity and fibrosis. To better investigate the continuum between high risk MDS (i.e., patients with ≥10% blasts) and AML, an exploratory comparison was made using a cohort of AML (defined according to WHO 2016) patients analyzed using the same statistical methods. Only a partial overlap in genetic signatures was observed between MDS with ≥10% blasts and AML. However, similarities were observed between the AML-like MDS clusters (characterized by mutant NPM1, bZIP CEBPA, and Core Binding Factor abnormalities) and AML clusters defined by the same genetic signature, thus supporting the classification of these entities as AML, irrespective of blast count. Conclusion. Our study demonstrated the utility of a data-driven approach based on advanced statistical methods to generate a harmonized classification for MDS. Table 2 shows a provisional, hierarchical classification algorithm. Further refinement of entity labels and classification criteria is the subject of the ongoing modified Delphi Process consensus approach among icMDS experts.