Classification and Prognostic Stratification Based on Genomic Features in Myelodysplastic and Myeloproliferative...

Background/Objectives: Myeloid neoplasms encompass a diverse group of disorders. In this study, we aimed to analyze the clinical and genomic data of patients with myeloproliferative neoplasm (MPN), myelodysplastic neoplasm (MDS), and their overlapping conditions, such as MDS/MPN and aplastic anemia (AA), to help redefine the disease classification. Methods: Clinico-genomic data of 1585 patients diagnosed with MPN (n = 715), MDS (n = 698), MDS/MPN (n = 78), and AA (n = 94) were collected. Patterns of 53 recurrent genomic abnormalities were compartmentalized into 10 groups using a Dirichlet process (DP). Results: These genomic groups were correlated with specific genomic features, survival outcomes, and disease subtypes. Groups DP1 and DP5, characterized by JAK2 and CALR mutations, respectively, showed very favorable prognoses among the patients with MPN. Groups DP2, DP7, and DP9 demonstrated very adverse prognoses across the disease subtypes. DP2 included patients with MDS harboring TP53 mutations and complex karyotypes; DP9 comprised patients with acute myeloid leukemia-related mutations, including NPM1; and DP7 included patients with SETBP1 mutations. Groups DP10 and DP8, linked to SF3B1 and DDX41 mutations or chromosome 1q derivatives, presented a favorable risk profile. Improved survival was observed with transplantation in groups DP2, DP7, and DP9. Conclusions: These findings highlight the role of genomic classifications in guiding personalized treatment strategies, ultimately enhancing the understanding and management of myeloid neoplasms.