Novel Simulation Model of Non-Muscle Invasive Bladder Cancer: A Platform for a Virtual Randomized Trial of Conservative Therapy vs. Cystectomy in BCG Refractory Patients

Introduction: There have been no randomized controlled trials (RCTs) evaluating the clinical or economic benefit of mitomycin C intravesical therapy vs. radical cystectomy in patients with high-risk non-muscle invasive bladder cancer (NMIBC). We used the Archimedes computational model to simulate RCT comparing radical cystectomy versus intravesical mitomycin C (MMC) therapy to evaluate the clinical and economic outcomes for BCG-refractory NMIBC as well demonstrate the utility of computer based models to simulate a clinical trial. Methods: The Archimedes model was developed to generate a virtual population using the Surveillance Epidemiology and End Results database, other clinical trials, and expert opinions. Patients selected were diagnosed with NMIBC (<cT2 disease) who recurred or progressed despite BCG therapy and were randomized to 1) immediate radical cystectomy vs. 2) MMC induction intravesical therapy. Clinical (progression, overall survival, and disease specific survival) and economic outcomes were reported. Results: A total of 1300 virtual patients were evaluation. Progression to MIBC in the MMC treatment arm was 30% over the lifetime. Disease specific death at 5 years was 1.6% and 8.7% for the immediate cystectomy and MMC treatment arms respectively; while, overall death was 17.8% and 23.8% at 5 years. Over a 5-year period the average cost of immediate cystectomy was $64,675 vs $68,517 in the MMC arm. Conclusion: Immediate radical cystectomy after BCG failure for NMIBC has improved survival and is more cost-effective when compared to those undergoing MMC. Simulation of clinical trials using computational models similar to the Archimedes model can overcome shortcomings of real-world clinical trials and may prove useful in the face of current medical cost-conscious era.