Abstract / Summary
Background. Myelodysplastic syndromes (MDS) are characterized by aberrant DNA methylation, and mutations in epigenetic modifiers are frequently found in these patients. Although DNA methyltransferase inhibitors (DNMTi) are used to treat MDS, response variability remains a challenge in the clinic, with limited predictive markers. Methods. We integrated genomic, epigenomic, and transcriptomic analyses of 98 MDS patients. Patients were classified into epigenetic subtypes via hierarchical clustering. Random forest classifiers were developed and validated using internal stratified testing and an independent external cohort to predict AZA response. Results. MDS is characterized by widespread DNA hypomethylation affecting distal regulatory elements. We identified seven epigenetic clusters correlated with distinct molecular drivers. Notably, Cluster VI exhibited low mutational burden but a high AZA response rate of 71% (P ≤ 0.01). While transcriptional profiles alone failed to distinguish responders, a DNAme-based classifier achieved an area under the curve (AUC) of 0.82. An integrative model combining DNAme, gene expression, mutations, and clinical parameters achieved an AUC of 0.93 in internal validation and 0.88 in the external cohort. Conclusion. Epigenetic signatures at distal genomic elements provide superior predictive power for AZA response compared to promoter-centric or transcriptional analyses. These findings establish a robust framework for personalized treatment strategies in MDS.