Abstract
INTRODUCTION: Clear cell renal cell carcinoma (ccRCC) is a malignant tumor characterized by poor prognosis and difficult treatment. Ferroptosis is a relatively new form of programmed cell death that involved in cancer development and therapy resistance. Studies have shown that targeted ferroptosis may be a novel option for the treatment of ccRCC, but key genes and their roles between ferroptosis and ccRCC are limited so far. This study aims to develop a ccRCC stratified model based on ferroptosis-related genes to provide a reference for the prognosis prediction and the individualized treatment of ccRCC. MATERIALS AND METHODS: The mRNAs expression data of ccRCC and FRGs were obtained from TCGA and FerrDb database, respectively. Through multiple analysis, a 4-FRG based prognostic stratified model was constructed and its predictive performance was validated through various methods. Then, a nomogram based on the model was constructed and ccRCC patients stratified by the model were analyzed for tumor microenvironment, immune infiltration, sensitivity for immune checkpoint inhibitors (ICIs)/traditional anti-tumor therapy and tumor mutation burden (TMB). Functional enrichment analysis was performed to explore potential biological pathways. Finally, we verified our model by RT-qPCR, siRNA transfection, scratch assay and CCK-8 assay. RESULTS: In this study, the stratified model and a model-based nomogram can accurately predict the prognosis of ccRCC patients in TCGA database. The patients stratified by the model showed different tumor microenvironments, immune infiltration, TMB, resistance to traditional and ICIs therapy, and sensitivity to ferroptosis. Functional enrichment analysis suggested several biological pathways related to the process and prognosis of ccRCC. RT-qPCR confirmed the differential expression of ferroptosis-related genes. Scratch assay and CCK-8 assay indicated the promotion effects of CD44 on the proliferation and migration of ccRCC. CONCLUSION: In this study, we established a novel ccRCC stratified model based on FRGs, which can accurately predict the prognosis of ccRCC patients and provide a reference for clinical individualized treatment.