Abstract
Background: Mesenchymal stem cells (MSCs) have been identified to have a unique migratory pattern toward tumor sites across diverse cancer types, playing a crucial role in cancer progression, treatment resistance, and immunosuppression. This study aims to formulate a prognostic model focused on MSC-associated markers to efficiently predict the clinical outcomes and responses to therapy in individuals with bladder cancer (BC). Methods: Clinical and transcriptome profiling data were extracted from The Cancer Genome Atlas Urothelial Bladder Carcinoma (TCGA-BLCA) and GSE31684 databases. Systematic quantification of MSC prevalences and stromal indices was undertaken, culminating in the discernment of genes correlated with stromal MSCs following a thorough application of weighted gene coexpression network analysis techniques. Subsequently, an exhaustive risk signature pertinent to MSC was formulated by amalgamating methods from univariate and Least Absolute Shrinkage and Selection Operator (LASSO) Cox regression models. Drugs targeting genes associated with MSCs were screened using molecular docking. Results: The prognostic model for MSC incorporated five critical genes: ZNF165, matrix remodeling-associated 7 (MXRA7), CEMIP, ADP-ribosylation factor-like 4C (ARL4C), and cerebral endothelial cell adhesion molecule (CERCAM). In the case of BC patients, stratification was performed into discrete risk categories, utilizing the median MSC risk score as a criterion. It was striking that those classified within the high-MSC-risk bracket demonstrated correlations with unfavorable prognostic implications. Enhanced responsiveness to immunotherapy in low-MSC-risk patients was delineated compared to their high-MSC-risk counterparts. A heightened receptivity was noted toward particular chemotherapy drugs, encompassing gemcitabine, vincristine, paclitaxel, gefitinib, and sorafenib, within this high-risk group. Conversely, a superior reaction to cisplatin was distinctly evident among those marked by low MSC scores. The results of molecular docking demonstrated that kaempferol exhibited favorable docking with ZNF165, quercetin exhibited favorable docking with MXRA7, mairin exhibited favorable docking with CEMIP, and limonin diosphenol exhibited favorable docking with ARL4C. Conclusions: The five-gene MSC prognostic model demonstrates substantial efficacy in prognosticating clinical outcomes and gauging responsiveness to chemotherapy and immunotherapy regimens. The genes ZNF165, MXRA7, CEMIP, ARL4C, and CERCAM are underscored as promising candidates warranting further exploration for anti-MSC therapeutic strategies, thereby offering novel insights for personalized treatment approaches in BC.