A prognosis model for predicting immunotherapy response of esophageal cancer based on oxidative stress-related signatures.

Oxidative stress (OS) is intimately associated with tumorigenesis and has been considered a potential therapeutic strategy. However, the OS-associated therapeutic target for esophageal squamous cell carcinoma (ESCC) remains unconfirmed. In our study, gene expression data of ESCC and clinical information from public databases were downloaded. Through LASSO-Cox regression analysis, a risk score (RS) signature map of prognosis was constructed and performed external verification with the GSE53625 cohort. The ESTIMATE, xCell, CIBERSORT, TIMER, and ImmuCellAI algorithms were employed to analyze infiltrating immune cells and generate an immune microenvironment (IM). Afterward, functional enrichment analysis clarified the underlying mechanism of the model. Nomogram was utilized for forecasting the survival rate of individual ESCC cases. As a result, we successfully constructed an OS-related genes (OSRGs) model and found that the survival rate of high-risk groups was lower than that of low-risk groups. The AUC of the ROC verified the strong prediction performance of the signal in these two cohorts further. According to independent prognostic analysis, the RS was identified as an independent risk factor for ESCC. The nomogram and follow-up data revealed that the RS possesses favorable predictive value for the prognosis of ESCC patients. qRT-PCR detection demonstrated increased expression of MPC1, COX6C, CYB5R3, CASP7, and CYCS in esophageal cancer patients. In conclusion, we have constructed an OSRGs model for ESCC to predict patients' prognosis, offering a novel insight into the potential application of the OSRGs model in ESCC.