Hypoxia score for predicting prognosis and tumor response to immunotherapy in bladder urothelial carcinoma

Abstract Numerous types of cancers are characterized by hypoxia. However, the effects of hypoxia on cancer immunology, tumor growth, and therapeutic responses in bladder urothelial carcinoma (BLCA) remain unknown. In this study, we examined the biological characteristics of hypoxia and hypoxia-related gene clusters in BLCA. Clustering analysis was performed to identify the subtypes of BLCA based on hypoxia biomarker genes. Principal component analysis and hypoxia-related genes were used to calculate a hypoxia score for BLCA. The level of infiltration of immune and stromal cells in the BLCA immune microenvironment was quantified using CIBERSORT and ESTIMATE. A hazard ratio was determined for the hypoxia score through meta-analysis of three separate BLCA cohorts. The association between the hypoxia score and the tumor microenvironment was determined using pan-cancer analysis. A tumor immunotherapy cohort and the Genomics of Drug Sensitivity in Cancer database were used to examine the predictive value of the hypoxia score in relation to various treatment options. We evaluated 1,449 BLCA samples, 5 cell lines, and 9,858 samples from 32 different tumor types. In the BLCA cohort, three hypoxia and two hypoxia-related gene clusters with different hypoxia levels in the tumor microenvironment were found. The final hypoxia scoring system considered 77 hypoxia-related genes. A low hypoxia score was associated with high KDM6A, FGFR3, and OBSCN mutation rates; better prognosis; and a strong response to immunotherapy. The understanding of tumor invasion and metastasis can be improved by analyzing the hypoxia patterns in different tumor microenvironments. The hypoxia scoring system may be useful for determining patient prognosis and personalized immune treatment plans.