Statistical Monitoring of Semi-Competing Risk Outcomes in Clinical Trials

• Activity Number: 674 - Adaptive Design - 3
• Type: Contributed
• Date/Time: Thursday, August 2, 2018 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Toshi Hamasaki* and Scott Evans and Tomoyuki Sugimoto and Koko Asakura and Susan Halabi
• Companies: and Harvard University and Kagoshima University and National Cerebral and Cardiovascular Center and Duke University
• Keywords: alpha level allocation; Bivariate time-to-event outcomes; co-primary endpoints; Multiple primary endpoints; Type I error adjustment

Abstract

We discuss logrank test-based methods for efficacy or futility evaluation in group-sequential clinical trials that compare two interventions with respect to two time-to-event outcomes. Evaluation is conducted under two situations: (a) both events are non-composite but one event is fatal, and (b) one event is composite but the other is fatal and non-composite. Based on group-sequential boundaries, we consider several decision-making frameworks for evaluating efficacy or futility. We consider two inferential goals, evaluating if a test intervention is superior to a control intervention on: (i) both outcomes (co-primary endpoints: CPE), and (ii) at least one outcome (multiple primary endpoints: MPE). For the CPE goal, we incorporate the correlations among the outcomes into the calculations for non-binding futility boundaries and sample sizes (or event numbers) as a function of other design parameters, including the mean differences, the number of analyses, and efficacy boundaries. We investigate the operating characteristics of the decision-making frameworks in terms of efficacy/futility boundaries, power, the Type I error rate, sample sizes, event numbers, while varying the number of