Gastric cancer is one of the major cancers with high cancer mortality and shows significant heterogeneity. The development of precise prognostic models is crucial for advancing treatment strategies. Recognizing the pivotal role of DNA damage in tumor progression, we conducted a consensus clustering analysis of DNA damage-related genes to categorize gastric cancer patients from the TCGA clinical cohort into distinct subtypes. Prognostic models were then constructed utilizing machine learning algorithms following Cox regression with differentially expressed genes. Validation was performed using the GSE gastric cancer cohort. Additionally, we investigated other characteristic responses of patients through gene mapping and drug sensitivity analysis. This study 12 differentially prognostic signature genes between the 2 DNA damage subtypes identified were used to calculate risk scores for the patients. This score predicts the prognosis of patients with gastric cancer and their overall survival time. Higher risk scores mean less drug sensitivity, lower survival, and possibly a poorer response to immunotherapy. Our findings provide the basis for future studies targeting DNA damage and its immune microenvironment to improve prognosis and response to immunotherapy.
Keywords: DNA damage; Gastric cancer; Machine learning; Prognostic models; Risk score.
© King Abdulaziz City for Science and Technology 2025. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.