Glycogen Metabolism Predicts the Efficacy of Immunotherapy for Urothelial Carcinoma

Abstract

Urothelial cancer (UC) is one of the common refractory tumors and chemotherapy is the primary treatment for it. The advent of immune checkpoint inhibitors (ICI) has facilitated the development of treatment strategies for UC patients. To screen out UC patients sensitive to ICI, researchers have proposed that PD-L1, tumor mutation burden and TCGA molecular subtypes can be used as predictors of ICI efficacy. However, the performance of these predictors needs further validation. We need to identify novel biomarkers to screen out UC patients sensitive to ICI. In our study, we collected the data of two clinical cohorts: the ICI cohort and the TCGA cohort. The result of the multivariate Cox regression analysis showed that glycogen metabolism score (GMS) (HR = 1.26, p = 0.017) was the negative predictor of prognosis for UC patients receiving ICI treatment. Low-GMS patients had a higher proportion of patients achieving complete response or partial response to ICI. After the comparison of gene mutation status between high-GMS and low-GMS patients, we identified six genes with significant differences in mutation frequencies, which may provide new directions for potential drug targets. Moreover, we analyzed the immune infiltration status and immune-related genes expression between high-GMS and low-GMS patients. A reduced proportion of tumor-associated fibroblasts and elevated proportion of CD8+ T cells can be observed in low-GMS patients while several immunosuppressive molecules were elevated in the high-GMS patients. Using the sequencing data of the GSE164042 dataset, we also found that myeloid-derived suppressor cell and neutrophil related signature scores were lower in α-glucosidase knockout bladder carcinoma cells when compared to the control group. In addition, angiogenesis, classic carcinogenic pathways, immunosuppressive cells related pathways and immunosuppressive cytokine secretion were mainly enriched in high-GMS patients and cell samples from the control group. Finally, we suspected that the combination treatment of ICI and histone deacetylase inhibitors may achieve better clinical responses in UC patients based on the analysis of drug sensitivity data. In conclusion, our study revealed the predictive value of GMS for ICI efficacy of UC patients, providing a novel perspective for the exploration of new drug targets and potential treatment strategies.