Identification of ferroptosis-related genes for the prediction of prognosis and chemotherapy benefit of gastric cancer

OBJECTIVE: As a newly reported programmed cell death, ferroptosis plays a crucial role in tumor progression. We aimed to comprehensively analyze the prognostic value of ferroptosis-related genes and identify the genes for the prediction of prognosis and chemotherapy benefit of gastric cancer (GC) patients. MATERIALS AND METHODS: Public microarray data and corresponding clinical information of GC were downloaded from GEO databases. Patients in GSE66229 were randomly separated into discovery and internal validation at a ratio of 2:1. GSE15459 was set as the external validation set. LASSO Cox regression model was per-formed to identify the most significant prognostic ferroptosis-related genes. RESULTS: Twenty ferroptosis-related genes were finally identified to establish the predictive signature. In the discovery data set, the signture could divide patients into low-and high-risk groups with significantly different overall survival (HR: 2.11, 95% CI: 1.40-3.17, p < 0.001). These results were successfully validated in the internal validation cohort (HR 2.04, 95% CI 1.18-3.52; p=0.008) and external validation cohort (HR 4.87, 95% CI 2.99-7.93; p=0.008). Survival ROC at 5-year revealed a remarkably higher predictive ability of the ferroptosis classifier (AUC=0.835) compared with other clinical features in predicting prognosis. Further GSEA analysis showed that samples of high risk were related to several established tumor invasion and metastatic signaling pathways. Further experiments revealed that VLDLR and GCH1 were two newly identified genes associated with chemotherapy sensitivity in GC. CONCLUSIONS: The developed ferroptosis gene-set based prognostic signature indicated superior prognostic and predictive value, suggesting new possibilities for individualized treatment of GC patients.