Abstract
The role of sphingolipid metabolism (SM) in promoting the progression of bladder cancer (BLCA) and its impact on patient prognosis has been established. To improve therapeutic outcomes, it is essential to identify specific molecular pathways in BLCA and develop a predictive signature underlying SM-related genes. In this study, 430 BLCA samples were analyzed using univariate Cox regression to identify critical SM-relevant genes (SMGs) involved in BLCA development. LASSO regression analysis was then employed to reduce the possibility of overfitting. A multivariable Cox regression analysis was employed to develop a prognostic signature underlying SMGs, which was subsequently validated in a separate cohort. Our research revealed that dysregulated SM leads to worse prognosis in BLCA patients, and important prognostic genes (PCSK2, NFASC, NTF3, NR2F1, ATP13A2, SREBF1, GSDMB, and LGALS4) were identified. Using these SMGs, we developed a prognostic BLCA-risk model that effectively predicted the prognosis of BLCA patients (AUC was 0.772 for the training cohort and 0.725 for the validation cohort). Interestingly, patients identified as high-risk by this model had a significantly more active immunological milieu, characterized by higher immune scores and increased 26 types of immune function and cell like NK cells, CD8+T cells, and cytolytic activity. These findings suggest that dysregulated SM may contribute to immune microenvironment dysregulation in BLCA. Our research provides a better awareness of the role of SM in the emergence of BLCA and has the potential to offer customized care to high-risk patients based on their SM-related gene expression signature.