Biomarkers in NMIBC: Promise without clinical utility yet

As of March 2026, no biomarker—whether immunohistochemistry, PD-L1, tumor mutational burden, or AI-based signatures—reliably predicts BCG response in non-muscle-invasive bladder cancer (NMIBC) with sufficient discriminatory power to drive clinical decision-making, although artificial intelligence-based markers have been applied selectively during BCG shortages to guide rationing.

In this video, recorded at the 41st Annual Congress of the European Association of Urology in London, UK, Ashish M. Kamat, MD, MBBS, FACS, an endowed professor of Urologic Oncology (Surgery) and Cancer Research, and the director of the Bladder Cancer Support Program at University of Texas MD Anderson Cancer Center in Houston, emphasized that effective biomarker strategy in NMIBC will require integrating tumor-based, immune microenvironment, and systemic markers rather than relying on any single assay, given the complexity of oncologic pathogenesis and treatment response along this disease continuum.

Regarding combination BCG plus checkpoint inhibitor therapy, Kamat noted that despite positive results from the CREST and POTOMAC trials, clinicians remain reluctant to adopt IO-BCG combinations broadly due to added toxicity and the perception that benefit accrues primarily in recurrence-free survival rather than in progression or overall survival. He argued that a validated predictive biomarker demonstrating that combination therapy reduces cystectomy rates, disease progression, or mortality for a defined patient subgroup would substantially lower the threshold for accepting that toxicity burden and would enable more informed shared decision-making—although no such marker currently exists.