Prediction of clear cell renal cell carcinoma prognosis based on an immunogenomic landscape analysis

Abstract

Immune cell infiltration and tumor-related immune molecules play key roles in tumorigenesis and tumor progression. The influence of immune interactions on the molecular characteristics and prognosis of clear cell renal cell carcinoma (ccRCC) remains unclear. A machine learning algorithm was applied to the transcriptome data from The Cancer Genome Atlas database to determine the immunophenotypic and immunological characteristics of ccRCC patients. These algorithms included single-sample gene set enrichment analyses and cell type identification. Using bioinformatics techniques, we examined the prognostic potential and regulatory networks of immune-related genes (IRGs) involved in ccRCC immune interactions. Fifteen IRGs (CCL7, CHGA, CMA1, CRABP2, IFNE, ISG15, NPR3, PDIA2, PGLYRP2, PLA2G2A, SAA1, TEK, TGFA, TNFSF14, and UCN2) were identified as prognostic IRGs associated with overall survival and were used to construct a prognostic model. The area under the receiver operating characteristic curve at 1 year was 0.927; 3 years, 0.822; and 5 years, 0.717, indicating good predictive accuracy. Molecular regulatory networks were found to govern immune interactions in ccRCC. Additionally, we developed a nomogram containing the model and clinical characteristics with high prognostic potential. By systematically examining the sophisticated regulatory mechanisms, molecular characteristics, and prognostic potential of ccRCC immune interactions, we provided an important framework for understanding the molecular mechanisms of ccRCC and identifying new prognostic markers and therapeutic targets for future research.

Keywords

• Clear cell renal cell carcinoma
• Immune-related gene
• Prognostic model
• ssGSEA
• CIBERSORT