Identification of four Ferroptosis Gene signatures and Establishment of a predictive model for the overall survival rate of gastric cancer

Abstract Background: Gastric cancer (GC) is highly aggressive and recurrent. Ferroptosis is believed to be closely related to the occurrence and development of tumors, but the specific mechanism is still unclear. This study aims to construct a new prognostic model based on ferroptosis-related gene scores to assess the prognosis of GC patients and guide clinical treatment.Materials and Methods: The gene expression information of GC patients with clinical data were obtained from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO). Differentially expressed genes (DEGs) between tumor and normal tissue were screened from TCGA dataset, and ferroptosis-related genes were downloaded from the ferroptosis database. The ferroptosis-related DEGs were identified by intersecting the ferroptosis-related genes with DEGs. Univariate Cox and LASSO regression analysis was applied to identify survival-related intersection genes (SRIGs) and 4 hub genes. The TCGA dataset was randomly divided into 2 cohorts in a ratio of 0.7:0.3, the training cohort for construction of signature and the testing cohort for internal validation. The GSE26901 was used for external validation. Receiver operating characteristic (ROC), Kaplan-Meier curve and risk curve of 3 cohorts were plotted to evaluate performance of the signature. Multivariate Cox regression analysis was used to determine independent prognostic factors for gastric cancer, and then a prognostic nomogram was established based on risk score and clinicopathological parameters.Results: A four-gene signature comprising MYB, CHAC1, NOX4 and AFT3 was constructed to predict overall survival of GC. Using the best cut-off value to divide patients into high-risk and low-risk groups, we found that the survival rate of the low-risk group in the training group, test group and external validation group was significantly higher than that of the high-risk group (P<0.05). The area under the ROC curve for 1 year, 3 year, and 5 year were (0.635, 0.666, 0.713), (0.63, 0.681, 0.602) and (0.758, 0.679, 0.677). Both univariate and multivariate Cox regression analysis showed that 4 gene markers are independent prognostic factors for GC. The 1-year, 3-year and 5-year overall survival consistency tests show that the nomogram has good predictability. Immune correlation analysis of high-risk population indicated that tumor progression may be related to the high expression of Macrophages.Conclusion: Our study identified a signature with 4 ferroptosis-related genes and established a reliable prognostic nomogram for predicting the overall survival rate of GC patients. The results may help medical decision-making, and provide a novel reference for predicting the prognosis of GC and even potential molecular targets.