Clinical Trials in Urology: Predictors of Successes and Failures.

Abstract

Clinical trials serve as a critical source of information to guide evidence-based practices in urology. Conversely, trials that are abandoned consume significant resources and results are underreported in the literature. ClinicalTrials.gov was queried for urology trials from 2006 to 2016. Trials were screened by 2 screeners for applicability to urology and disputes were resolved by a third independent reviewer. Overall 1,340 trials met final inclusion criteria (722 successful trials, 618 failed trials). Univariable analysis used Fisher's exact, chi-squared and Wilcoxon rank sum tests. Trial characteristics, including AUA (American Urological Association) section, phase, subspecialty, intervention type, source of funding and randomization were examined for association with failure using multivariable logistic regression. Trial failure is associated with oncology subspecialty (adjusted odds ratio 2.25, 95% CI 1.60-3.18), infertility/andrology subspecialty (AOR 4.99, CI 1.60-17.61), device trials (AOR 1.64, CI 1.00-2.70) and combination funding by industry/government/grants (AOR 3.13, CI 2.21-4.48). Clinical trials in AUA sections were less likely to fail than international and multisectional trials. Among trials that failed, poor accrual was the primary reason for trial failure, comprising 41% of all failures. Other reasons for failure include inadequate budget (9%), sponsor cancellation (7%), poor interim results (7%) and toxicity (3%). Despite their significance, many urological trials fail prematurely due to poor accrual. Complex features inherent to oncology, andrology/infertility, devices and multisectional trials pose significant barriers to success.