Construction of a Two-Gene Prognostic Model Related to Ferroptosis in Renal Cell Carcinoma.
Lei Li,Yawei Xu,Wuping Yang,Kenan Zhang,Zedan Zhang,Jingcheng Zhou,Yanqing Gong,Kan Gong
DOI: https://doi.org/10.21037/tau-23-346
2023-01-01
Translational Andrology and Urology
Abstract:Background:Renal cell carcinoma (RCC) is a common and aggressive tumor. A newly discovered form of programmed cell death, ferroptosis, plays an important role in tumor development and progression. However, a clear prognostic correlation between Ferroptosis-related genes (FRGs) and RCC has not yet been established. In this study, prognostic markers associated with FRGs were investigated to improve the therapeutic, diagnostic, and preventive strategies available to patients with renal cancer.Methods:The present study analyzed the predictive value of 23 FRGs in RCC through bioinformatics techniques, including Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) tools, Kaplan-Meier survival analysis, Cox regression modeling, tumor mutational burden (TMB), CIBERSORT, and half maximal inhibitory concentration (IC50) difference analysis.Results:We screened FRGs by differentially expressed genes (DEGs) and overall survival (OS). Four candidate genes were obtained by hybridization. Then, we constructed a two-gene prognostic signature (NCOA4 and CDKN1A) via univariate Cox regression and multivariate stepwise Cox regression, which classified RCC patients into high- and low-risk groups, and patients in the high-risk group were found to have worse OS and progression-free survival (PFS). We also found that patients with higher TNM stage, T stage, and M stage had higher risk scores than those with lower TNM stage, T stage, and M stage (P<0.05). Males had higher risk scores than females. This signature was identified as an independent prognostic indicator for RCC. These results were validated in both the test cohort and the entire cohort. In addition, we also constructed a nomogram that predicted the OS in RCC patients, the consistency index (C-index) of the nomogram was 0.731 [95% confidence interval (CI): 0.672-0.790], the areas under the receiver operating characteristic (ROC) curves (AUCs) were 0.728, 0.704, and 0.898 at 1-, 3-, and 5-year, respectively, which shows that nomogram has good prediction ability. and we also analyzed the immune status and drug sensitivity between the high- and low-risk groups.Conclusions:We constructed a prognostic model associated with ferroptosis, which may provide clinicians with a reliable predictive assessment tool and offer new perspectives for the future clinical management of RCC.