Prognostic Model for Clear-cell Renal Cell Carcinoma Based on Natural Killer Cell-related Genes.

Natural killer cells play a fundamental role in immune surveillance of ccRCC. A NKRPS was established by univariate Cox and LASSO-Cox regression based on the TCGA cohort (n = 515). The high-risk group exhibited superior immune infiltration and higher expression of ICI-related genes. This suggests that the NKRPS is valuable in predicting prognosis and providing a therapeutic reference for ccRCC patients. Background: Natural killer (NK) cells are a key factor affecting progression and immune surveillance of clear-cell renal cell carcinoma (ccRCC). This study sought to construct a natural killer cell-related prognostic signature (NKRPS) to predict the outcome of ccRCC patients and to furnish guidance for finding appropriate treatment strategies. Methods: From the TCGA and ArrayExpress databases, transcriptomic profiles and relevant clinical information of ccRCC patients were downloaded for the TCGA cohort (n = 515) and the E-MTAB-1980 cohort (n = 101). With the univariate Cox analysis and LASSO-Cox regression algorithm, a NKRPS was built to evaluate patients' prognosis. Receiver operating characteristic (ROC) cur ves and calibration cur ves were drawn to estimate the predictive power of the prognostic model. Then, tumor microenvironment (TME), tumor mutational burden (TMB), sensitization to immune checkpoint inhibitors (ICIs) therapy and targeted drug treatment were analyzed in ccRCC patients. Results: Nine genes (BID, CCL7, CSF2, IL23A, KNSTRN, RHBDD3, PIK3R3, RNF19B and VAV3) were identified to construct a NKRPS. High-risk group displayed undesirable survival compared to low-risk group ( P < .05). Moreover, the area under the curve (AUC) of ROC at 1-, 3and 5-year were 0.766, 0.755, and 0.757, respectively. High-risk group was characterized by superior immune infiltration, higher TMB, and higher expression of 5 ICI-related genes. Additionally, this model enabled to predict the sensitivity of patients to chemotherapy drugs. Conclusion: NKRPS had a strong predictive power on prognosis of ccRCC patients, which may facilitate individualized treatment and medical decision making.