A Robust Ferroptosis-Related Prognostic Model Associated With Immune Infiltration, Tumor Mutation Burden in Bladder Cancer

Abstract Background: Bladder cancer (BC) is a common clinical disease with a poor prognosis caused by both genetic and environmental factors. Despite many treatments available, the risk of recurrence and metastasis remains high. Ferroptosis is a newly discovered iron-dependent programmed cell death. More and more scientific studies have shown that inducing ferroptosis of tumor cells can inhibit tumor cell growth and disease progression, especially for some tumors that are not sensitive to traditional treatments. However, whether the ferroptosis-related genes(FRGs) can accurately predict the prognosis of BC patients is still not very clear and significant biomarkers are still insufficient. Results: Six genes (EGFR, FADS1, ISCU, PGRMC1, PTPN6, and TRIM26) were identified to construct a prognostic risk model. The Cancer Genome Atlas (TCGA) training cohort was divided into high- and low-risk groups according to the median risk score. Kaplan–Meier survival analysis indicated that the overall survival (OS) of the high-risk group was worse than that of the low-risk group. The receiver operating characteristic(ROC) curves showed excellent predictive accuracy. TCGA validation cohort and three independent Gene Expression Omnibus (GEO) datasets were used to conduct further external validation. A series of functional analyses demonstrated the relationship between tumor microenvironment and FRGs, and between tumor mutation burden and immunotherapy in the high- and low-risk groups. Conclusion: A robust prognostic risk model was established, which has independent predictive value for the prognosis of BC patients. The correlations between ferroptosis and tumor immune infiltration, immunotherapy, and tumor mutation burden were studied, providing insights into the treatment of bladder cancer patients in the future. Methods: We downloaded the gene expression data and corresponding clinical information of bladder cancer samples from TCGA database in the UCSC-Xena and GEO public database, and obtained FRGs from the FerrDb platform. Univariate Cox regression analysis, multivariate Cox regression analysis, least absolute shrinkage and selection operator (LASSO) regression were used to screen out FRGs with clinical predictive value. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis were performed to explore the classical signaling pathways related to ferroptosis. CIBERSORT was used to quantify the infiltration of 22 kinds of immune cells.