Super-enhancer gene C1QTNF6 as a prognostic biomarker and therapeutic target in bladder cancer: A machine learning-based multi-omics analysis

Introduction: Super-enhancers play crucial roles in tumor development as key transcriptional regulatory elements, yet their prognostic value in bladder cancer (BLCA) remains to be systematically elucidated. Objective: This study aimed to comprehensively analyze the regulatory mechanisms of super-enhancer-related genes (SERGs) in predicting BLCA prognosis and immune therapy response. Methods: This study integrated BLCA RNA sequencing data from the cancer genome atlas (training set) and gene expression omnibus (validation set) databases and obtained SERG sets from the SEdb database. Using 101 machine learning ensemble frameworks, we screened and validated SERG sets with significant prognostic value and constructed a risk score model was constructed based on based on CoxBoost and plsRcox. Model performance was evaluated through nomograms. We conducted an in-depth analysis of the association between the risk model and tumor immune microenvironment, identified key hub genes through differential expression analysis, survival analysis, and receiver operating characteristic curve analysis, and performed multi-dimensional validation using immunohistochemistry and single-cell sequencing data. Results: Through machine learning algorithm optimization, we identified eight core genes (AHNAK, NT5DC3, NFIC, MTHFD1L, C1QTNF6, SLC45A3, QRICH2, KRT8). High-risk group patients exhibited poor prognosis and elevated immune and tumor immune dysfunction and exclusion scores, suggesting potential immune therapy resistance. The single-sample gene set enrichment analysis analysis revealed significant positive correlations between risk scores and multiple key signaling pathways, including extracellular matrix-receptor interaction, regulation of actin cytoskeleton, and pathogenic Escherichia coli infection, focal adhesion, melanoma, and gap junction pathways. Further analysis indicated that C1QTNF6 and MTHFD1L displayed significant potential as biomarkers based on expression profiles across cell types. Conclusion: This study pioneered the construction of a prognostic prediction model for BLCA based on SERGs, revealing the crucial role of super-enhancers in regulating the tumor immune microenvironment and identifying potential therapeutic targets and prognostic markers. This research provides a new molecular typing strategy for the precision treatment of BLCA while establishing a theoretical foundation for personalized immunotherapy.