Citations

Background. Study of myeloid neoplasms (MN) has been rapidly transformed by genome characterization. Gene mutations have been reported to influence disease phenotype and risk of disease progression, and the evaluation of mutational status can provide valuable insights to improve the decision-making process. In this international study, we evaluated the clinical impact of mutational screening in patients with CMML, which is characterized by significant clinical and genomic heterogeneity, including a substantial proportion of patients with poor outcomes and unmet needs. Currently, only limited molecular information has been incorporated into CMML classification and prognostication. Methods. We retrospectively collected 3182 CMML patients. We used MOSAIC, an AI-based framework for multimodal analysis in rare cancers (PMID: 38875514) to develop innovative, molecular-based tools for classification and prognostication. An independent validation of the results on 516 prospectively collected patients was performed. Results. Bayesian Networs and Hierarchical Dirichlet Processes were used to identify genomic associations and define a CMML molecular classification. We identified 11 clusters with distinct clinical and genomic features, including splicing machinery, signal transduction and tyrosine kinase pathways aberrations, and high-risk molecular signatures (i.e., AML-like and TP53 mutations). Up to 15% of patients showed clear genomic overlap with other MN. Importantly, extensive multi-color flow cytometry on T lymphocytes, NK and myeloid cells (BD FACSymphony A5 Cell Analyzer) revealed specific immunologic and inflammation signatures associated with molecular subgroups. We then integrated molecular and clinical information to build an international CMML Prognostic Scoring System (iCPSS). We assessed the performance of different statistical and AI-based methods; L1-penalized Cox Model resulted as the best performing method. Selected features included hematological parameters (WBC, Hb, PLT and marrow blasts), cytogenetic abnormalities, and mutations in 10 genes (ASXL1, DNMT3A, EZH2, NRAS, RUNX1, SETBP1, STAG2, TET2, TP53, U2AF1). By Bayesian thresholds optimization, we identified 5 risk classes (i.e., very-low, low, intermediate, high and very-high) with median overall survival (OS) ranging from 99 to 9 months (P<.001), and median leukemia-free survival (LFS) ranging from 210 to 18 months (P<.001). iCPSS provided better patient discrimination across all clinical endpoints compared to currently available prognostic tools, with a Concordance Index (CI) for OS of 0.75 versus 0.62-0.64, respectively. Notably, compared to available scoring systems, up to 40% of patients were reassigned to higher or lower risk classes by the iCPSS. External validation on an independent prospective cohort confirmed that the performance of iCPSS is superior to existing prognostic models (CI for OS 0.71 vs 0.54-0.61, respectively) To demonstrate the clinical utility of iCPSS, we focused on 753 patients who underwent allogeneic stem cell transplantation (HSCT). In multivariable analysis, iCPSS stratified the probability of OS post-HSCT (P<.001) and identified groups of patients with different probabilities of disease relapse, ranging from 9% to 62% (P<.001). We employed a clinical- and genomic-based decision support system (PMID: 38723212) to determine the optimal timing of HSCT. Preliminary analyses indicated that higher-risk patients according to iCPSS (i.e., intermediate, high, and very high) benefit from an immediate HSCT strategy. Lower-risk patients (i.e., very low and low) had life expectancy maximized with delayed HSCT. Modeling decision analysis using iCPSS compared to conventional scores, resulted in a change in transplantation policy for a significant proportion of patients (up to 25%). Conclusion. Molecular information significantly improves the classification of CMML patients, providing a basis for refining diagnostic boundaries with other MN and for a more rational inclusion of patients in clinical trials. The iCPSS demonstrated superior performance over currently available scoring systems and improved HSCT decision making process.

Introduction: Response assessment is used to evaluate effectiveness of hypomethylating agent (HMA)-based therapy in myelodysplastic syndromes/neoplasms (MDS) in clinical practice and trials. Recently, the clinical response criteria (IWG 2023) and the clinical-molecular risk assessment (IPSS-M) in MDS were published. However, it remains unknown whether the IWG 2023 response criteria have an independent impact on overall survival (OS) when adjusting for baseline IPSS-M risk. Additionally, the association of IWG 2023 response criteria with OS in patients (pts) undergoing allogeneic stem cell transplantation (allo-HCT) versus those who do not is unknown. Lastly, it is unclear if optimal time-to-transplant impacts OS in MDS pts who achieve a IWG 2023 defined remission with HMA therapy. We examined these questions in a large population of MDS pts within the multi-institutional international consortium for MDS (icMDS) VALIDATE database. Methods: High-risk MDS pts with IPSS ≥1.5 or IPSS-R >3.5 treated with HMA-based therapy were included in the study. Baseline IPSS-M risk was stratified into very low/low/moderate low (VL+L+ML), moderate-high/high (MH+H), and very high (VH). Response to treatment was evaluated using the IWG 2023 criteria and grouped into: composite complete response (cCR) and no cCR. cCR included: complete response (CR), CR equivalent (CRequ), CR bilineage (CRbi), CR unilineage (CRuni), and CR with partial hematologic recovery (CRh). We analyzed the differential impact of achieving cCR on HMA-based therapy prior to transplant on OS in pts who either did or did not undergo allo-HCT, while adjusting for their baseline IPSS-M risk in Cox regression analysis. Optimal time-to-transplant was determined based on age and IPSS-M category as described by Tentori et al. (JCO 2024). Pts proceeding to allo-HCT were stratified on whether they underwent transplant within the optimal time-to-transplant timeframe after MDS diagnosis or not. VALIDATE protocol was exempted, acknowledged, or approved by all centers' IRBs. VALIDATE is supported by an independent grant from AbbVie. Results: We analyzed 762 MDS pts. The median age at MDS diagnosis was 68 years (range: 19-95 years) and 64% of pts were male. Distribution of the baseline IPSS-M risk was: VL+L+ML in 110 (14%) pts, MH+H in 347 (46%) pts, VH in 305 (40%) pts. Pts received azacitidine monotherapy, decitabine monotherapy, and HMA-based combination therapy (e.g. HMA + venetoclax) in 52%, 21%, and 28% of cases, respectively. A total of 282/762 (37%) pts achieved cCR as their best response to HMA-based therapy. The median OS for all pts was 18.7 months (95% CI: 16.6-21.3 months). Pts who did not undergo allo-HCT (n=436, 57%) had an improved median OS if they achieved cCR when compared to pts in the no cCR group in all IPSS-M risk groups (VL+L+ML: 23.5 months vs. 10.7 months; MH+H: 20.8 months vs. 11.6 months; VH: 13.8 months vs. 7.5 months). Cox analysis demonstrated an independent effect of cCR on OS in these pts when adjusted for baseline IPSS-M risk (cCR vs. no cCR: HR 0.60, 95% CI 0.48-0.76, p<0.001). In contrast, in pts who subsequently received an allo-HCT (n=326, 43%), achieving cCR with HMA prior to allo-HCT did not provide an additional OS benefit, when adjusted for baseline IPSS-M risk (cCR vs. no cCR: HR 1.00, 95% CI 0.73-1.38, p=0.986). The median time-to-transplant was 6.7 months and 34% of transplanted MDS pts received their allo-HCT during the optimal timeframe. Among pts who achieved cCR with HMA-based therapy prior to allo-HCT, optimal time-to-transplant was associated with significantly improved OS (optimal vs. non-optimal timing: HR 0.55; p=0.033). Conclusion: Baseline IPSS-M risk and allo-HCT had an independent impact on OS, while IWG 2023 defined cCR had an independent effect on OS only in non-transplanted MDS pts. While this retrospective analysis occurred at large academic centers with specific expertise in the management of MDS, our data may have an impact on clinical practice: (i) failure to achieve cCR on HMA therapy prior to transplant should not impact the decision to proceed to allo-HCT, (ii) upfront allo-HCT without prior HMA therapy might also be an option if the clinical situation and availability of donors allow, and (iii) allo-HCT should ideally be performed during the timeframe that is considered optimal as defined by pt age and baseline IPSS-M.

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.

Abstract Despite advances in understanding the genetic abnormalities in myeloproliferative neoplasms (MPN) and the development of JAK2 inhibitors, there is an urgent need to devise new treatment strategies, particularly for patients with triple-negative (TN) myelofibrosis (MF) who lack mutations in the JAK2 kinase pathway and have very poor clinical outcomes. Here we report that MYC copy number gain and increased MYC expression frequently occur in TN-MF and that MYC-directed activation of S100A9, an alarmin protein that plays pivotal roles in inflammation and innate immunity, is necessary and sufficient to drive development and progression of MF. Notably, the MYC-S100A9 circuit provokes a complex network of inflammatory signaling that involves numerous hematopoietic cell types in the bone marrow microenvironment. Accordingly, genetic ablation of S100A9 or treatment with small molecules targeting the MYC-S100A9 pathway effectively ameliorates MF phenotypes, highlighting the MYC–alarmin axis as a novel therapeutic vulnerability for this subgroup of MPNs. Significance: This study establishes that MYC expression is increased in TN-MPNs via trisomy 8, that a MYC-S100A9 circuit manifest in these cases is sufficient to provoke myelofibrosis and inflammation in diverse hematopoietic cell types in the BM niche, and that the MYC-S100A9 circuit is targetable in TN-MPNs.

SummaryAmong 210 patients with myelodysplastic syndromes (MDSs) with del(5q), molecular information was available at diagnosis or at least 3 months before leukaemic transformation in 146 cases. Multivariate analysis identified therapy‐related setting (p = 0.02; HR 2.3) and TP53 variant allele frequency (VAF) ≥22% (p < 0.01; HR 2.8), but not SF3B1 mutation (p = 0.65), as independent risk factors for survival. Median survival was 11.7 versus 4 years (5/10‐year survival 73%/52% vs. 42%/14%) in the absence (N = 112) versus presence (N = 34) of ≥1 risk factors; leukaemia‐free survival was affected by TP53 VAF ≥22% (p < 0.01). Such information might inform treatment decision‐making in MDS‐del(5q) regarding allogeneic stem cell transplant.

Primary Myelofibrosis (PMF) is an aggressive myeloproliferative neoplasm (MPN) characterized by constitutional symptoms, cytopenias, splenomegaly, extramedullary hematopoiesis, bone marrow (BM) fibrosis, and a propensity to transform to acute myeloid leukemia (AML). Despite advances in understanding the underlying genetic abnormalities in MPN and the development of JAK2 inhibitors to treat MF, there is an urgent need to devise new treatment strategies, particularly for triple negative myelofibrosis (TN-MF) cases that lack mutations in the JAK2 kinase pathway and who have inferior outcomes. To identify potential oncogenic drivers in TN-MF, we performed cytogenetic analyses and targeted exome sequencing of 98 genes commonly mutated in myeloid malignancies in 584 MF patients identified from the Total Cancer Care database at Moffitt Cancer Center. Notably, there were no significant differences in the somatic mutation profiles (other than JAK2 activating mutations) between TN vs. JAK2/CALR/MPL mutant MF patients. However, trisomy of chromosome 8 occurs more frequently in TN-MF vs. other subtypes (26.7% vs. 6.2%, p=0.0001). To identify oncogenic drivers in trisomy 8+ TN-MF, we performed scRNA-seq analysis from 18,651 cells ( n=15,850 from 3 normal donors and n=2,801 from a trisomy 8+ TN-MF patient) and identified a total of 1,260 genes that are differentially regulated in hematopoietic stem cells (HSCs) of trisomy 8+ TN-MF compared to normal donors. Among these, 131 genes were located on chromosome 8 and MYC was one of the top 5 genes that are significantly upregulated in trisomy 8+ HSCs. Additional immunohistochemistry staining demonstrated that MYC protein levels in BM cells were significantly higher in trisomy 8+ TN-MF patients vs. trisomy 8- patients. To assess whether MYC might be a driver of MPN, we established an Mx1-Cre +/-;Rosa26 LSL-MYC/LSL-MYC transgenic mouse model that inducibly overexpresses MYC in HSCs following treatment of polyinosinic-polycytidylic acid (pIpC). Forced MYC expression induced profound anemia, monocytosis, megakaryocytic atypia, splenomegaly, increased BM collagen deposition/fibrosis, extramedullary hematopoiesis in spleen and liver, and significantly reduced OS vs. Mx1-Cre +/- wild type controls (median OS 258 days vs. not reached (NR), p&lt;0.0001). MYC also promoted expansion of HSCs, myeloid progenitors, and Gr-1 +/CD11b + mature myeloid cells in BM and spleen; thus, MYC was sufficient to provoke MF-like disease independent of JAK2 pathway mutations. Subsequent scRNA-seq analysis of 25,232 cells ( n=13,552 from control and n=11,680 from a MYC transgenic mouse) revealed that S100a9 mRNA levels were increased by MYC in most major hematopoietic cell types. Supporting oncogenic roles of S100A9 in MYC-driven MF, deletion of S100a9 in Mx1-Cre +/-;Rosa26 LSL-MYC/LSL-MYC;S100a9 -/- mice abolished MYC-induced MF phenotypes and significantly improved OS (median OS NR vs. 225 days, p=0.0345). To test whether inhibition of MYC can suppress MF disease progression in vivo, lethally irradiated CD45.1 +/CD45.2 + WT mice were transplanted with BM cells harvested from CD45.2 + Mx1-Cre +/-;Rosa26 LSL-MYC/LSL-MYC mice that were treated with pIpC 20 weeks prior to transplant. MYCi975, a small molecule that inhibits MYC, reduced MYC and S100a9 protein levels in BM and spleen cells. Further, MYCi975 effectively suppressed MYC-driven MF phenotypes and disease progression, and significantly improved OS (median OS NR vs. 252 days, p=0.0109). In summary, our studies are the first to describe an oncogenic role of MYC in MF pathogenesis, where MYC provokes an alarmin-driven inflammatory circuit, and where the MYC-S100A9 axis represents a therapeutic vulnerability for TN-MF patients. Accordingly, our results provide a strong rationale for testing agents targeting MYC or S100A9 in early phase clinical trials in MPNs having increased MYC levels or activity.

Introduction: MDS are primarily diseases of the elderly, with median age at diagnosis of 72 years. Focused data on clinical and molecular characteristics, as well as outcomes, in younger pts under age 60 years are limited. We conducted this multicenter retrospective study to describe pt characteristics, cytogenetic and molecular profiles, treatment modalities, and outcomes in younger pts with MDS, including adolescent and young adults (AYA). Methods: The study included pts diagnosed with MDS between 2002-2023 at 7 academic centers. We compared baseline characteristics, treatment, and outcomes of AYA (18-39 years) and younger adult (YA) pts (40-59 years). Responses were assessed using the modified IWG 2006 response criteria which included complete remission (CR), marrow CR (mCR), and hematologic improvement (HI). Wilcoxon rank sum test and Fisher's exact (and Chi-square) test were used to compare pt and disease characteristics for continuous and categorical variables, respectively. Overall survival (OS) was estimated using the Kaplan-Meier method and compared using log-rank test. Results: A total of 173 pts were identified (AYA, N=50; YA, N=123). Median age at diagnosis was 50 years (range, 18-59), with 51% females. Baseline characteristics are shown in Table 1. Therapy-related MDS (t-MDS) was observed in 40 pts (23%) with the remaining classified as de novo MDS (N=133; 77%). Inherited bone marrow failure syndromes were documented in 15 pts (9%), with Shwachman-Diamond Syndrome (N=4) and Fanconi anemia (N=4) being the most frequent. Using the WHO 2016 classification, the most common subtypes of MDS in both the AYA and YA cohorts were MDS with multilineage dysplasia (38% vs 19%), MDS with excess blasts-2 (MDS-EB) (24% vs 29%), and MDS-EB-1 (16% vs 20%). Similarly, using the WHO 2022 classification, the most common subtypes were MDS with low blasts (50% vs 21%), MDS with increased blasts-2 (MDS-IB-2) (24% vs 24%), and MDS-IB-1 (16% vs 18%). Stratifying pts using the IPSS-R cytogenetic (CG) risk group, the most common CG risk category identified was good CG (34% vs 52%, p=0.012). Poor/very poor CG were present in 34% and 37% of the AYA and YA groups, respectively. STAG2 and NRAS mutations occurred more frequently in the AYA pts compared to the YA group, while the YA cohort was enriched for the following mutations: TP53, DNMT3A, SF3B1, SRSF2, and TET2 (Figure 1). In the AYA population, 18 pts (38%) had very low/low, 10 (21%) intermediate, and 20 pts (42%) had high/very high risk MDS using IPSS-R. In the YA population, 32 pts (26%) had very low/low, 32 (26%) intermediate, and 57 pts (47%) had high/very high risk MDS using IPSS-R. Twenty-seven percent of pts with very low to intermediate-risk MDS using IPSS-R were upstaged to moderate high to very high-risk MDS using IPSS-M. The majority of pts (AYA: 75%; YA: 63%, p=0.5) were transfusion independent at baseline. Single-agent hypomethylating agents were the most frequently used frontline therapy, used in 17 AYA pts (34%) and 58 YA pts (48%), with more AYA pts receiving upfront allogeneic stem cell transplantation (Allo-SCT) compared to YA pts (28% vs 5%). After frontline therapy (Table 1), an additional 14 AYA (28%) and 52 YA (42%) pts received Allo-SCT. In pts with IPSS-R intermediate to very high-risk, response rates (HI+CR+mCR+mCR with HI) to frontline therapies were 46% and 54.5% in the AYA and YA groups, respectively. After median follow-up of 24 months [range, 1-218], median OS was numerically longer in the AYA versus YA group (155 vs 53 months, p=0.2). AML transformation occurred in 24% and 20% of the pts in the AYA and YA groups, respectively. Median OS was significantly longer in de novo compared to t-MDS pts (56 vs 27 months, p=0.03). Median OS among Allo-SCT compared to non-SCT pts were not reached vs 155 months ( p=0.9) in the AYA group and 53 vs 54 months ( p=0.4) in the YA group. IPSS-R (very low+low vs intermediate vs high+very high) was able to distinguish differences in OS in the YA (133 vs 53 vs 17 months, p=0.0002) and AYA (155 vs not reached vs 18 months, p=0.006) pts, with the exception of the intermediate-risk group in the AYA cohort. Conclusions: In this study, d e novoMDS was seen in the majority of younger pts with MDS (both AYA and YA). Mutations in STAG2 and NRAS were more common in the AYA pts, while YA pts had MDS more enriched for TP53, DNMT3A, TET2 and splicing mutations. This study is still ongoing with a plan to compare the abovementioned groups to older pts (≥60 years) with MDS.

Background: According to the international consensus classification (ICC), diagnosis of myelodysplastic syndromes (MDS) with del(5q) requires presence of i) del5q, alone or associated with no more than one other cytogenetic abnormality, exclusive of -7/del(7q), ii) &lt;5% bone marrow (BM) and &lt;2% circulating blasts, and iii) the absence of “multi-hit” TP53 ( Arber et al. Blood 2022;140: 1200). We have recently reported on TP53 variant allele frequency (VAF) ≥22% being the sole genetic predictor of inferior survival in MDS-del(5q) (Fleti. et al. AJH 2023;98:E76). In the current study, we examined the additional prognostic impact of therapy-related setting, in the context of TP53 VAF and other risk factors. Methods : The current study included patients with MDS-del(5q) who were seen at the Mayo Clinic or Moffitt Cancer Center, USA. Diagnosis of MDS-del(5q) and therapy-related qualification was according to ICC criteria ( Arber et al. Blood 2022;140: 1200), with the exception that patients with biallelic or multi-hit TP53 were included to allow unbiased analysis of impact on prognosis. Patient selection was based on availability of mutation information from next-generation sequencing (NGS). The latter was assessed either at diagnosis or at different time points during the disease course, but at least 3 months before leukemic transformation. Otherwise, all analyses, including estimation of overall and leukemia-free survival, were based on variables collected at time of initial diagnosis of MDS-del(5q). TP53 mutation VAF threshold of 22% was considered based on our previously published report cited above. Age cutoff at 67 years was determined by ROC analysis. Conventional statistical methods were applied using JMP Pro 16.0.0 software (SAS Institute, Cary, NC, USA). Results: A total of 210 patients (median age 72 years; 65% females), meeting the above-stipulated criteria for MDS-del(5q), were considered: 172 primary and 38 therapy-related. Table 1 outlines clinical and laboratory characteristics obtained at the time of diagnosis: 42% were transfusion-dependent, 69% had hemoglobin level &lt;10 g/dL, 30% platelet count &lt;150 x 10(9)/L, and 15% carried additional cytogenetic abnormality. NGS data performed either at diagnosis or at least 3 months prior to leukemic transformation was available in 146 patients; the most frequent mutations were TP53 (21%), SF3B1 (16%), DNMT3A (12%), TET2 (12%), ASXL1 (10%), and JAK2 (7%) (Table 1). At a median follow-up of 3.9 years (range 0.002-18.5) for living patients, 95 (45%) deaths and 34 (16%) leukemic transformations were recorded; treatment undertaken included lenalidomide in 162 (77%) cases. NGS data-exclusive multivariable analysis that included all 210 patients identified therapy-related setting (p&lt;0.01; HR 2.9; Figure 1a), age &gt;67 years (p=0.01; HR 1.8), transfusion-dependency (p=0.04; HR 1.5), and platelet count &lt;150 x 10(9)/L (p=0.03; HR 1.6), as independent predictors of inferior survival. The significant effect of age and platelet count was sustained when analysis was limited to patients with primary MDS-del(5q) but lost when applied to therapy-related disease. NGS data-inclusive multivariable analysis, which was applicable to the subset of 146 patients, identified, as risk factors, therapy-related setting (p&lt;0.01; HR 2.3) and TP53 VAF ≥22% (p&lt;0.01; HR 2.4; Figure 1b) while age &gt;67 years sustained borderline significance (p=0.07; HR 1.8). Figure 1c depicts survival data based on the presence or absence of TP53 VAF ≥22% and therapy-related qualification; low-risk with no risk factors (N=112; median survival 11.7 years; 5/10-year survival 73%/52%) and high-risk with one or both risk factors (N=34; median survival 4 years; 5/10-year survival 42%/14%); leukemia-free survival was affected by TP53 VAF ≥22% (p&lt;0.01) but not therapy-related setting (p=0.13). Conclusions : The current study highlights TP53 VAF and therapy-related setting as the most prominent risk factors for survival in MDS-del(5q) and proposes a simple risk model that distinguishes patients with a projected 10-year survival that exceeds 50%, in the absence of both risk factors, and plummets to 14%, in the presence of one of the two risk factors. Accordingly, allogeneic stem cell transplant might be deferred in the former instance while it should be pursued sooner than later, in the latter instance.

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&gt;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 &lt;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 &lt;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.