Construction of T cell exhaustion model for predicting survival and immunotherapy effect of bladder cancer based on WGCNA

Introduction: The prognosis of bladder cancer (BLCA) and response to immune checkpoint inhibitors (ICIs) are determined by multiple factors. Existed biomarkers for predicting the effect of immunotherapy cannot accurately predict the response of BLCA patients to ICIs.

Methods: To further accurately stratify patients' response to ICIs and identify potential novel predictive biomarkers, we used the known T cell exhaustion (TEX)-related specific pathways, including tumor necrosis factor (TNF), interleukin (IL)-2, interferon (IFN)-g, and T- cell cytotoxicpathways, combined with weighted correlation network analysis (WGCNA) to analyze the characteristics of TEX in BLCA in detail, constructed a TEX model.

Results: This model including 28 genes can robustly predict the survival of BLCA and immunotherapeutic efficacy. This model could divide BLCA into two groups, TEXhigh and TEXlow, with significantly different prognoses, clinical features, and reactivity to ICIs. The critical characteristic genes, such as potential biomarkers Charged Multivesicular Body Protein 4C (CHMP4C), SH2 Domain Containing 2A (SH2D2A), Prickle Planar Cell Polarity Protein 3 (PRICKLE3) and Zinc Finger Protein 165 (ZNF165) were verified in BLCA clinical samples by real-time quantitative chain reaction (qPCR) and immunohistochemistry (IHC).

Discussion: Our findings show that the TEX model can serve as biological markers for predicting the response to ICIs, and the involving molecules in the TEX model might provide new potential targets for immunotherapy in BLCA.