Premature termination of genitourinary cancer clinical trials.

Abstract

288 Background: The GU oncology literature is inundated with clinical trials that terminated prematurely, particularly in bladder cancer. Such trials require substantial resource expenditure and entail the time, trust, and commitment of patients, yet contribute minimally to the scientific knowledgebase and divert resources from answering critical questions. We sought to determine the scope of this problem within the clinical trials enterprise. Methods: ClinicalTrials.gov was queried to identify all phase II-III interventional adult cancer clinical trials registered between 9/11/05 and 11/11/11. Prematurely terminated trials were “stopped early” as defined by the registry. Kaplan-Meier methods and Cox regression were used to determine risk of premature trial termination. Results: We identified 7,776 trials, including 491 prostate (PCa), 142 kidney, 75 bladder, and 34 testis cancer trials. The risk of premature termination due to any cause for all cancers was 25% (95% CI 19-31%) and the risk due to poor accrual was 10% (95% CI 9-12%). Poor accrual was the most common reason for premature termination (Table). Risk was not significantly different for kidney, bladder or testis cancers compared to other cancer types with the exception of PCa (HR 1.35 [1.03-1.78]). Industry-funded trials were more likely to terminate prematurely (HR 2.26 [1.83-2.80]). Trials with sites outside of the USA (HR 0.63 [0.54-0.74]) or both within and outside of the USA (HR 0.68 [0.54-0.74]) were less likely to terminate prematurely as were trials with multiple sites (HR 0.56 [0.48-0.64]). Conclusions: In this large cohort of clinical trials, ~1 in 4 trials terminated prematurely (1 in 10 due to poor accrual). GU cancer trials were at similar risk of termination compared to other cancer clinical trials with the exception of PCa. Novel approaches are needed to improve the efficiency of the clinical cancer research enterprise. [Table: see text]