Early Termination of Phase II-III Genitourinary Oncology Trials: A Two-Decade Analysis of Causes and Trial-Level Risk Factors.

PURPOSE: Early termination of clinical trials delays scientific progress and wastes resources, yet factors associated with trial completion in genitourinary (GU) oncology remain incompletely characterized. We investigated trial-level factors associated with premature discontinuation of phase II-III GU oncology clinical trials.

MATERIALS AND

METHODS

We analyzed US phase II-III interventional GU oncology trials registered on ClinicalTrials.gov that were completed or terminated between January 1, 2005, and January 1, 2025. Prespecified operational and trial-level characteristics at initiation included disease site, phase, randomization, masking, sponsor type, number of sites, multinational status, data monitoring committee (DMC) presence, and intervention category. Reasons for early termination were categorized, and associations with premature discontinuation were assessed using multivariable logistic regression.

RESULTS

Among 1,597 trials, 445 (27.9%) were terminated early and 1,152 (72.1%) were completed. Poor accrual accounted for 41.8% of terminations, followed by sponsor or business decisions (14.6%) and interim futility or efficacy findings (10.3%). In multivariable analysis, prostate cancer trials were more likely to complete than bladder cancer trials (odds ratio [OR], 1.77, 95% CI, 1.25 to 2.50), as were multinational (OR, 1.96, 95% CI, 1.27 to 3.02) and multisite studies (OR, 1.72, 95% CI, 1.28 to 2.30). Trials enrolling metastatic-only (OR, 0.45, 95% CI, 0.29 to 0.72) or mixed-stage populations (OR, 0.55, 95% CI, 0.36 to 0.86), as well as those with DMC oversight (OR, 0.62, 95% CI, 0.47 to 0.81), were less likely to complete.

CONCLUSION

More than one quarter of GU oncology trials terminate prematurely, most often due to accrual failure. Early termination is associated with identifiable disease-specific, structural, and operational trial characteristics. Recognition of these patterns may help inform feasibility assessment, trial planning, and resource allocation, with the goal of reducing preventable trial failure in GU oncology research.