Machine Learning-Based Prognostic Modeling of Seven Signatures Associated with Lysosomes for Predicting Prognosis and Immune Status in Clear Cell Renal Cell Carcinoma

INTRODUCTION: Clear cell renal cell carcinoma (ccRCC) is the most common subtype of kidney cancer and is associated with poor prognosis in advanced stages. This study aims to develop a prognostic model for patients with ccRCC based on a lysosome-related gene signature. METHODS: The clinical and transcriptomic data of kidney renal clear cell carcinoma (KIRC) patients were downloaded from TCGA, cBioportal, and GEO databases, and lysosome-related gene sets were acquired in the previous study. TCGA data were used as a training set to investigate the prognostic role of lysosomal-related genes in ccRCC, and cBioportal and GEO databases were used for validation. After the lysosome-related differentially expressed genes were found, machine learning method was used to construct a risk model, and Kaplan-Meier (K-M) and receiver operating characteristic curves were used to evaluate the performance of the model. RESULTS: Machine learning methods were utilized to identify seven gene signatures related to lysosome, which accurately predict the prognosis of ccRCC. Patients with higher risk scores demonstrate poorer overall survival (OS; HR: 2.467, 95% CI: 1.642-3.706, p < 0.001), and significant disparities in immune infiltration, immune score, and response to anticancer drugs are observed between the high-risk group and the low-risk group (p < 0.001). CONCLUSION: The prognostic model developed in this study demonstrates a high efficacy in accurately predicting the OS of ccRCC patients, thereby offering a novel perspective for the advancement of ccRCC treatment.