Identification of cancer-associated fibroblast characteristics for predicting outcome and response to immunotherapy in renal cell carcinoma

To elucidate the prognostic effect of cancer-associated fibroblasts (CAFs) on renal cell carcinoma. CAFs and stromal scores were calculated using various algorithms including Estimating the Proportions of Immune and Cancer cells (EPIC), Microenvironment Cell Populations-counter (MCP counter), Tumor immune dysfunction and exclusion (TIDE) and xCell. Weighted gene co-expression network analysis (WGCNA) was conducted to determine the CAF-associated modules and key genes. The functional pathways of key genes in important CAF modules were analyzed using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses. CAF-associated signatures were established through univariate Cox and Least Absolute Shrinkage and Selection Operator (LASSO) analyses. Kaplan–Meier and receiver operating characteristic (ROC) analyses were carried out to assess the predictive value of CAF signatures. WGCNA analysis distinguished several CAF-associated modules for renal clear cell carcinoma (KIRC), renal chromophobe cell carcinoma (KICH) and renal papillary cell carcinoma (KIRP) respectively. CAF signatures were established containing two and four genes for KIRC and KIRP, respectively. In KIRC and KIRP, patients with high-risk scores had unfavorable outcome than those with low-risk scores. Additionally, in both KIRC and KIRP, the ratio of patients responding to immunotherapy was obviously higher in low-risk group than in high-risk group. Finally, the mutation frequency of some genes differed significantly between two groups. Our study provided valuable CAF signatures for predicting the outcome of KIRC and KIRP patients. These CAF signatures were also used to predict immunotherapy response, providing strategies for individualized therapy of patients.