Real-World Survival Outcomes of Hypomethylating Agents and Allogeneic Hematopoietic Stem Cell Transplantation in Myelodysplastic Syndromes: A Retrospective Single-Center Experience

Background: Treatment strategies for myelodysplastic syndromes (MDS) range from supportive care to disease-modifying therapies, depending on risk stratification and patient characteristics. Hypomethylating agents (HMAs) remain the standard treatment for higher-risk disease, whereas allogeneic hematopoietic stem cell transplantation (allo-HSCT) is the only potentially curative option. This study evaluated real-world survival outcomes and prognostic factors in patients with MDS treated with HMAs or allo-HSCT. Methods: A total of 79 patients with MDS who received azacitidine, decitabine, or allo-HSCT were retrospectively analyzed. Overall survival (OS) was evaluated using Kaplan–Meier and Cox proportional hazards analyses. To address potential immortal time bias, a prespecified 6-month landmark analysis was additionally performed. Results: Median OS was 14.9 months (95% CI, 9.9–20.0) with azacitidine, 10.0 months (95% CI, 6.8–13.2) with decitabine, and 48.0 months (95% CI, 19.9–76.1) after allo-HSCT. Patients undergoing allo-HSCT were significantly younger and had better ECOG performance status than those receiving HMAs. After adjustment for age, ECOG performance status, and IPSS-R risk category, treatment modality remained significantly associated with OS in the multivariable Cox regression model. In the prespecified 6-month landmark analysis, the survival advantage associated with allo-HSCT remained significant, although the adjusted association was attenuated. Conclusions: In this real-world cohort of actively treated patients with MDS, allo-HSCT was associated with longer overall survival than HMAs. Although this association persisted after adjustment for major clinical confounders and in a landmark analysis addressing immortal time bias, residual confounding related to treatment selection cannot be excluded. These findings should therefore be interpreted as an association rather than evidence of a causal treatment effect.