Inflammatory response-based prognostication and personalized therapy decisions in clear cell renal cell cancer to aid precision oncology

Abstract

Abstract Objective The impact of inflammatory response on tumor development and therapeutic response is of significant importance in clear cell renal cell carcinoma (ccRCC). The customization of specialized prognostication approaches and the exploration of supplementary treatment options hold critical clinical implications in relation to the inflammatory response. Methods In the present study, unsupervised clustering was implemented on TCGA-KIRC tumors using transcriptome profiles of inflammatory response genes, which was then validated in two ccRCC datasets (E-MATB-1980 and ICGC) and two immunotherapy datasets (IMvigor210 and Liu et al.) via SubMap and NTP algorithms. Combining co-expression and LASSO analyses, inflammatory response-based scoring system was defined, which was evaluated in pan-cancer. Results Three reproducible inflammatory response subtypes (named IR1, IR2 and IR3) were determined and independently verified, each exhibiting distinct molecular, clinical, and immunological characteristics. Among these subtypes, IR2 had the best OS outcomes, followed by IR3 and IR1. In terms of anti-angiogenic agents, sunitinib may be appropriate for IR1 patients, while axitinib and pazopanib may be suitable for IR2 patients, and sorafenib for IR3 patients. Additionally, IR1 patients might benefit from anti-CTLA4 therapy. A scoring system called IRscore was defined for individual ccRCC patients. Patients with high IRscore presented a lower response rate to anti-PD-L1 therapy and worse prognostic outcomes. Pan-cancer analysis demonstrated the immunological features and prognostic relevance of the IRscore. Conclusion Altogether, characterization of inflammatory response subtypes and IRscore provides a roadmap for patient risk stratification and personalized treatment decisions, not only in ccRCC, but also in pan-cancer.