Multi-omics features-based machine learning method improve immunotherapy response in clear cell renal cell carcinoma

Abstract

Programmed cell death 1 (PD-1) or PD-ligand 1 (PD-L1) blocker-based strategies have improved the survival outcomes of clear cell renal cell carcinomas (ccRCCs) in recent years, but only a small number of patients have benefited from them. In this study, we identified three inflammatory features through over 1900 autoimmune nephropathy patients-related bulk RNA sequencing, single-cell RNA sequencing analysis, and three immunogenic signatures using genomics (TIs), both of which are associated with response to immune checkpoint blocks (ICBs) and the survival of ccRCC patients. Here, we developed a framework with a TIs-based machine learning approach to accurately predict ICB efficacy. We enrolled more than 1000 ccRCC patients with ICB treatment from five cohorts to apply the model and demonstrated its excellent specificity and robustness. Moreover, our model outperforms well-known ICB predictive biomarkers such as tumor mutational burden (TMB), PD-L1 expression, and tumor immune microenvironment (TME). Overall, the TIs-ML model provides a novel method for guiding precise immunotherapy in ccRCC.