Genomic Subtypes and Computational Biomarkers in Non-Muscle-Invasive Bladder Cancer Guiding Optimal Timing of Radical Cystectomy and BCG Response Prediction

Non-muscle-invasive bladder cancer (NMIBC) accounts for approximately 70% of newly diagnosed bladder cancer cases but exhibits significant clinical heterogeneity in treatment response and progression risk. While intravesical bacillus Calmette–GuérinCa (BCG) therapy remains the gold standard for high-risk disease, approximately 30–50% of patients experience BCG failure, creating a critical decision point between additional bladder-sparing therapy (BST) and early radical cystectomy (RC). Recent clinical data from the CISTO study suggest that, in appropriately selected patients, RC may be associated with higher 12-month recurrence-free survival while maintaining comparable cancer-specific survival and physical functioning. In this narrative review, we synthesize contemporary evidence on NMIBC genomic and transcriptomic subtypes, immune contexture, and clinicopathologic features associated with BCG response and progression risk, with emphasis on clinically oriented classification systems such as BCG Response Subtypes (BRS1–3) and UROMOL21. We highlight how tumor-intrinsic biology (e.g., EMT-associated programs), immune phenotypes (inflamed vs. immune-cold microenvironments), and genomic alterations may help refine risk stratification beyond traditional clinicopathologic models. To facilitate clinical integration, we propose a conceptual decisional framework that combines molecular subtype assignment, immune profiling, key pathologic risk factors, and patient considerations to generate probabilistic risk tiers that support selection among early RC, BST, and clinical trial strategies. Standardized multicenter cohorts and prospective evaluation are needed to validate integrated models and define their clinical utility for the precision timing of cystectomy in BCG-unresponsive NMIBC.