Abstract
BACKGROUND: While immune cells are pivotal in clear cell renal cell carcinoma (ccRCC), functional alterations of specific subsets and their prognostic implications remain unclear. We aimed to identify key immune cells, characterize their functional states, and develop a prognostic model by integrating single-cell RNA sequencing (scRNA-seq) and bulk RNA sequencing (RNA-seq) data. METHODS: scRNA-seq dataset GSE210038 was analyzed to annotate tumor-infiltrating immune cells. Key immune cells were identified from deconvolved The Cancer Genome Atlas-Kidney Renal Clear Cell Carcinoma (TCGA-KIRC) and Gene Expression Omnibus Series (GSE)105261 datasets using support vector machine-recursive feature elimination (SVM-RFE), least absolute shrinkage and selection operator (LASSO), and random forest (RF) algorithms. Immune Response Enrichment Analysis (IREA) delineated polarization states of key immune cells in response to cytokines. Cellular communication was profiled via CellChat. Prognostic genes associated with key immune cells were screened by univariate Cox and LASSO regression. A risk score (RS) model was constructed and validated in TCGA-KIRC (training, n=511) and E-MTAB-1980 (validation, n=101) cohorts. Receiver operating characteristic (ROC) curves evaluated overall survival (OS) prediction efficacy. Nomograms and drug sensitivity analyses were performed. RESULTS: Seven immune cell types infiltrated ccRCC. CD8(+) T and natural killer (NK) cells were identified as the key immune cells. IREA revealed T8-c polarization in CD8(+) T cells [enriched for interferon (IFN)-α1/interleukin (IL)-36α] and NK-f polarization in NK cells (enriched for IL-18/IL-2). CellChat analysis showed amplified protease activated receptors (PARs) signaling from key immune cells to cancer-associated fibroblasts, but attenuated Annexin A1 (ANXA1)-formyl peptide receptor 1 (FPR1) signaling to macrophages. Ten immune-cell-associated prognostic genes (PTTG1, CLEC2D, PLIN2, LRBA, ABCB1, MIR155HG, ADAM8, P2RY8, SORL1, CD82) were used to build the RS model. The model stratified patients into high/low-risk groups with distinct OS (log-rank P<0.001). The area under the curves (AUCs) for 1-/3-/5-year OS were 0.751/0.734/0.747 (training) and 0.742/0.740/0.778 (validation). The nomogram [incorporating RS and metastatic stage (M stage)] showed robust calibration. Drug sensitivity analysis revealed low-risk patients responded better to tyrosine kinase inhibitors, while high-risk patients were sensitive to a B-cell lymphoma-2 inhibitor. CONCLUSIONS: CD8+ T and NK cells exhibit functional polarization and altered cellular communication indicative of augmented anti-tumor immunity in ccRCC. The immune-cell-derived 10-gene signature provides a reliable prognostic tool and guides personalized therapy.