Varying the intensity of cystoscopic surveillance for high-risk non-muscle-invasive bladder cancer.

Abstract

To compare the clinical, economic, and health utility outcomes associated with alternative cystoscopic surveillance regimens for high-risk non-muscle-invasive bladder cancer (HRNMIBC). We performed real-world clinical data-driven microsimulations of a hypothetical cohort of 100 000 patients diagnosed with HRNMIBC at age 70 years. The cohort was simulated to undergo alternative surveillance regimens recommended by five guidelines, and two hypothetical regimens-surveillance intensity escalation and de-escalation-which had a surveillance intensity moderately higher and lower, respectively, than the guideline-recommended regimens. We evaluated the 10-year cumulative incidence of muscle-invasive bladder cancer (MIBC), cancer-specific survival (CSS), overall survival (OS), and cost-effectiveness from a United States healthcare payer perspective. The guideline-recommended surveillance regimens led to an estimated 10-year cumulative incidence of MIBC ranging from 11.0% to 11.6%, CSS 95.0% to 95.2%, and OS 69.7% to 69.8%. Surveillance intensity escalation resulted in a 10-year cumulative incidence of MIBC of 10.5% (95% confidence interval [CI] 10.3-10.7%), CSS of 95.4% (95% CI 95.2-95.5%), and OS of 69.9% (95% CI 69.6-70.1%), vs 11.9% (95% CI 11.7-12.1%), 94.9% (95% CI 94.8-95.1%), and 69.6% (95% CI 69.3-69.9%), respectively, from surveillance intensity de-escalation. By increasing surveillance intensity, the number-needed-to-treat to prevent one additional MIBC progression over 10 years was ≥80, and ≥257 to avoid one additional cancer-related mortality. Compared to surveillance intensity de-escalation, higher-intensity regimens incurred an incremental cost of ≥$336 000 per incremental quality-adjusted life year gained, which well exceeded conventional willingness-to-pay thresholds, ≥$686 000 per additional MIBC progression prevented, and ≥$2.2 million per additional cancer-related mortality avoided. In microsimulations testing a wide range of cystoscopic surveillance intensity for patients newly diagnosed with HRNMIBC, moderate surveillance de-escalation appears associated with an insignificant change in 10-year OS and furthermore is cost-effective vs higher-intensity surveillance regimens. These results suggest that moderate surveillance de-escalation can reduce costs of care without compromising life expectancy for many patients.