Abstract
Clear cell renal cell carcinoma (ccRCC) is the most frequent type of renal cell carcinoma (RCC). The present study aimed to examine prognostic markers and construct a prognostic prediction system for ccRCC. The mRNA sequencing data of ccRCC was downloaded from The Cancer Genome Atlas (TCGA) database, and the GSE40435 dataset was obtained from the Gene Expression Omnibus database. Using the Limma package, the differentially expressed genes (DEGs) in the TCGA dataset and GSE40435 dataset were obtained, respectively, and the overlapped DEGs were selected. Subsequently, Cox regression analysis was applied for screening prognosis‑associated genes. Following visualization of the co‑expression network using Cytoscape software, the network modules were examined using the GraphWeb tool. Functional annotation for genes in the network was performed using the clusterProfiler package. Finally, a prognostic prediction system was constructed through Bayes discriminant analysis and confirmed with the GSE29609 validation dataset. The results revealed a total of 263 overlapped DEGs and 161 prognosis‑associated genes. Following construction of the co‑expression network, 16 functional terms and three pathways were obtained for genes in the network. In addition, red, yellow (involving chemokine ligand 10 (CXCL10), CD27 molecule (CD27) and runt‑related transcription factor 3 (RUNX3)], green (involving angiopoietin‑like 4 (ANGPTL4), stanniocalcin 2 (STC2), and sperm associated antigen 4 (SPAG4)], and cyan modules were extracted from the co‑expression network. Additionally, the prognostic prediction system involving 44 signature genes, including ANGPTL4, STC2, CXCL10, SPAG4, CD27, matrix metalloproteinase (MMP9) and RUNX3, was identified and confirmed. In conclusion, the 44‑gene prognostic prediction system involving ANGPTL4, STC2, CXCL10, SPAG4, CD27, MMP9 and RUNX3 may be utilized for predicting the prognosis of patients with ccRCC.