This study aimed to develop a prognostic model for clear cell renal cell carcinoma (ccRCC) based on transcriptome analysis. We screened Gene Expression Omnibus (GEO) database and the Cancer Genome Atlas (TCGA) database for gene expression data and clinical characteristics of ccRCC. After differentially expression analysis, we identified 533 key genes of the development of ccRCC. Next, a weighted gene co-expression network analysis (WGCNA) was executed to investigate the association between differentially expressed genes and clinical characteristics. Then, based on protein-protein interaction (PPI) network, least absolute shrinkage and selection operator (LASSO) regression and Cox regression, a four-gene (COL4A5, ABCB1, NR3C2 and PLG) prognostic model has been constructed in TCGA training cohort. Finally, we examined the predictive power of this model with survival analysis and receiver operating characteristic (ROC) curve both in training cohort and in validation cohorts. And we found this model was significantly correlated with infiltrating immune cells. The four-gene prognosis model could be a potential prediction tool with high accuracy and reliability for ccRCC patients.
Copyright © 2021. Published by Elsevier Inc.

Author