CEBPA and DNMT3A Methylation and Expression Profiles Predict Survival in Acute Myeloid Leukemia.

Abstract

Introduction: Acute myeloid leukemia (AML) is a genetically heterogeneous malignancy with variable clinical outcomes. Epigenetic alterations, particularly DNA methylation, have emerged as key modulators of gene expression and disease progression. This study investigates the prognostic relevance of methylation and expression profiles of CEBPA and DNMT3A in AML.

Methods: A cohort of adult AML patients from the TCGA-LAML dataset was analyzed for gene expression and methylation profiles across seven AML-associated genes. Survival associations were evaluated using Kaplan-Meier curves and Cox proportional hazards models. An initial panel of seven AML-associated genes (CEBPA, DNMT3A, HDAC1, IDH1, IDH2, NPM1, and RUNX1) was screened for prognostic relevance using matched expression and methylation data; genes with significant survival associations were retained for model construction. Patients were stratified into low, intermediate, and high-risk groups based on this score.

Results: Significant survival associations were observed for DNMT3A expression (p = 0.0062) and CEBPA methylation (p = 0.0074). Of the seven candidates, only DNMT3A expression and CEBPA methylation demonstrated independent prognostic value and were integrated into the final composite score. The composite score yielded strong survival separation across risk groups (p = 0.0033). In the final multivariate Cox model (n = 158), CEBPA methylation and DNMT3A expression remained independently predictive of overall survival.

Conclusion: Methylation and expression profiles of CEBPA and DNMT3A could be integrated into a composite molecular score to complement conventional genetic and cytogenetic classifications to stratify survival risk in AML; however, further studies with established clinical risk systems are required.