Currently, no effective prognostic model of clear cell renal cell carcinoma (ccRCC) based on immune cell infiltration has been developed. Recent studies have identified 6 immune groups (IS) in 33 solid tumors. We aimed to characterize the expression pattern of IS in ccRCC and evaluate the potential in predicting patient prognosis. The clinical information, immune subgroup, somatic mutation, copy number variation, and methylation score of patients with TCGA ccRCC cohort were downloaded from UCSC Xena for further analysis. The most dominant IS in ccRCC was the inflammatory subgroup (immune C3) (86.5%), regardless of different pathological stages, pathological grades, and genders. In the C3 subgroup, stage IV (69.1%) and grade 4 (69.9%) were the least presented. Survival analysis showed that the IS could effectively predict the overall survival (OS) (P < .0001) and disease-specific survival (DSS) (P < .0001) of ccRCC alone, of which group C3 (OS, HR = 2.3, P < .001; DSS, HR = 2.84, P < .001) exhibited the best prognosis. Among the most frequently mutated ccRCC genes, only VHL and PBRM1 were found to be common in the C3 group. The homologous recombination deficiency score was also lower. High heterogeneity was observed in immune cells and immunoregulatory genes of IS. Notably, CD4+ memory resting T cells were highly infiltrating, regulatory T cells (Treg) showed low infiltration, and most immunoregulatory genes (such as CX3CL1, IFNA2, TLR4, SELP, HMGB1, and TNFRSF14) were highly expressed in the C3 subgroup than in other subgroups. Enrichment analysis showed that adipogenesis, apical junction, hypoxia, IL2 STAT5 signaling, TGF-beta signaling, and UV response DN were activated, whereas E2F targets, G2M checkpoint, and MYC targets V2 were downregulated in the C3 group. Immune classification can more accurately classify ccRCC patients and predict OS and DSS. Thus, IS-based classification may be a valuable tool that enables individualized treatment of patients with ccRCC.

Compelling evidence indicates that immune function is correlated with the prognosis of bladder cancer (BC). Here, we aimed to develop a clinically translatable immune-related gene pairs (IRGPs) prognostic signature to estimate the overall survival (OS) of bladder cancer. From the 251 prognostic-related IRGPs, 37 prognostic-related IRGPs were identified using LASSO regression. We identified IRGPs with the potential to be prognostic markers. The established risk scores divided BC patients into high and low risk score groups, and the survival analysis showed that risk score was related to OS in the TCGA-training set ( p < 0.001 ; HR = 7.5 [5.3, 10]). ROC curve analysis showed that the AUC for the 1-year, 3-year, and 5-year follow-up was 0.820, 0.883, and 0.879, respectively. The model was verified in the TCGA-testing set and external dataset GSE13507. Multivariate analysis showed that risk score was an independent prognostic predictor in patients with BC. In addition, significant differences were found in gene mutations, copy number variations, and gene expression levels in patients with BC between the high and low risk score groups. Gene set enrichment analysis showed that, in the high-risk score group, multiple immune-related pathways were inhibited, and multiple mesenchymal phenotype-related pathways were activated. Immune infiltration analysis revealed that immune cells associated with poor prognosis of BC were upregulated in the high-risk score group, whereas immune cells associated with a better prognosis of BC were downregulated in the high-risk score group. Other immunoregulatory genes were also differentially expressed between high and low risk score groups. A 37 IRGPs-based risk score signature is presented in this study. This signature can efficiently classify BC patients into high and low risk score groups. This signature can be exploited to select high-risk BC patients for more targeted treatment.

No abstract