Abstract
OBJECTIVE: To identify peripheral blood biomarker candidates for coronary artery disease (CAD) and to test whether bioinformatically prioritized genes replicate in an angiography-confirmed cohort, with emphasis on HIST1H2AE as an epigenetically relevant marker. METHODS: Differentially expressed genes (DEGs) were identified from three GEO datasets (GSE42148, GSE98583, GSE12288) after within-dataset normalization; DEGs were then intersected across datasets to prioritize robust candidates while avoiding direct cross-study merging and associated batch effects. Enrichment analysis and protein-protein interaction network prioritization were performed, followed by independent in silico evaluation in GSE20681 using receiver operating characteristic (ROC) analysis. HIST1H2AE and CXCL14 were then tested by qPCR in peripheral blood from 20 participants (10 angiography-confirmed CAD; 10 non-CAD controls). Immune cell proportions in GSE20681 were estimated with CIBERSORTx, and correlations with hub-gene expression were assessed. RESULTS: Eight DEGs were consistently shared across the three discovery datasets, and HIST1H2AE and CXCL14 were prioritized as hub candidates. In GSE20681, ROC analysis suggested discriminatory ability for both genes (AUC=0.711 for HIST1H2AE; AUC=0.878 for CXCL14). In the angiography-confirmed qPCR cohort, CXCL14 was not consistently different between groups, whereas HIST1H2AE was significantly downregulated in CAD. HIST1H2AE expression showed no significant correlation with estimated immune cell proportions. CONCLUSION: This multi-dataset discovery and comparative validation framework prioritizes HIST1H2AE as a peripheral blood biomarker candidate for CAD with an immune-independent expression profile in the analyses performed. Given the small qPCR cohort, these findings are preliminary and require confirmation in larger, independent cohorts and mechanistic studies of chromatin-level regulation.