Senescence-related Genes as Prognostic Markers for STEMI Patients: LASSO Regression-Based Bioinformatics and External Validation.

Abstract

The prognostic value of differentially expressed senescence-related genes(DESRGs) in ST-segment elevation myocardial infarction(STEMI) patients is unclear. We used GEO2R to identify DESRGs from GSE60993 and performed functional enrichment analysis. We built an optimal prognostic model with LASSO penalized Cox regression via GSE49925. We evaluated the model with survival analysis, ROC curve, decision curve analysis, nomogram, and external validation with plasma samples. We created a prognostic signature with three dysregulated DESRGs (CDC25B, FKBP5, and ECHDC3) and two clinical variables (serum creatinine, Gensini score). The signature stratified patients into low- and high-risk groups and showed strong predictive performance within two years. The external validation confirmed the survival difference between the groups. We identified three DESRGs that were differentially expressed and prognostic in STEMI patients. The model incorporating three DESRGs showed promising prediction and utility for stratifying patients and estimating survival risk in STEMI patients.