Construction of a Prognostic Model and Drug Prediction Based on Endoplasmic Reticulum Stress Response Genes in Esophageal Cancer

Endoplasmic reticulum stress (ERS) has profound implications for various cancer cells, including esophageal cancer (ESCA) cells. Currently, there is no study about the function of ERS-related genes (ERSRGs) prognosis in ESCA. We developed an ERS risk scoring model by TCGA-ESCA dataset to validate predictive ability of ERSRGs for ESCA prognosis and to screen reliable prognostic biomarkers. Enrichment analysis demonstrated that differentially expressed ERSRGs were mainly associated with the ERS response, regulation of apoptosis signaling pathways, and the p53 signaling pathway. Major differences among ESCA patients with different ERS risk scores were found by survival analysis. According to mutation analysis, high-risk group had a higher mutation frequency than the other group. Drug prediction indicated that fulvestrant, elesclomol, and methylprednisolone were promising small molecule candidates that could potentially target model genes. In summary, we developed ERS risk score model serving as an independent prognostic factor and simultaneously predicted several potential small molecule drugs. These results offer new directions for future investigation on prognosis and treatment.