A survival-informed CD8+ T cell-associated gene signature and its prognostic relevance in clear cell renal cell carcinoma
Year: 2026
Type: article
Abstract: Background: Clear cell renal cell carcinoma (ccRCC) is characterized by marked clinical and immunological heterogeneity, and conventional clinicopathological variables do not fully capture patient survival risk. CD8+ T cells play a central role in antitumor immunity, yet their functional states can encompass both effective cytotoxic responses and dysfunctional immune programs. This study aimed to derive a survival-informed CD8+ T cell-associated gene signature and evaluate its prognostic associations, cellular expression, and relationship to immune dysfunction.Methods: Tumor-infiltrating CD8+ T cells from GSE178481 were analyzed using Semi-supervised Iterative Deep Learning for Identifying Single-cell High-Risk Populations (SIDISH) with survival information from The Cancer Genome Atlas (TCGA) training cohort. Candidate genes underwent cell-level differential expression analysis, sample-level pseudobulk screening, in silico perturbation, univariable Cox screening and benchmarking of 246 machine-learning model combinations. The final risk score was assessed for survival stratification, independent prognostic value, nomogram performance, pathway enrichment, tumor microenvironment features, and survival associations in a programmed cell death protein 1 (PD-1) blockade-treated cohort.Results: SIDISH identified a subset of high-risk CD8+ T cells and corresponding candidate marker genes for subsequent model construction. The selected elastic net (α =0.8) plus least absolute shrinkage and selection operator (LASSO) Cox regression model generated a multigene risk score that consistently stratified overall survival (OS) in the TCGA training set, TCGA validation set, E-MTAB-1980 cohort, and CPTAC cohort. Higher risk scores were associated with advanced stage and grade and remained independently prognostic after adjustment for clinicopathological covariates in TCGA and E-MTAB-1980 cohorts. Nomograms incorporating the risk score improved individualized prediction of OS and progression-free survival (PFS). The risk score showed potential associations with immune-related pathways, metabolic pathways, and the tumor immune microenvironment, and was also associated with prognosis in the CheckMate 025 cohort.Conclusions: This survival-informed CD8+ T cell-associated signature was associated with prognosis in ccRCC and showed potential associations with immune features. These findings suggest that integrating single-cell immune states with bulk transcriptomic survival modeling may help identify prognosis-related signatures, although further validation is needed.
Source: Translational Cancer Research
Authors: Hui Wang, Fan Li, Yi Lu, Lifeng Ding
Institutions: Sir Run Run Shaw Hospital, Zhejiang University
Language: English
Subfield: Oncology
Field: Medicine
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W7224131753