Abstract:
|
Use of multiple endpoints creates challenges in the evaluation of power and the calculation of sample size during trial design. Clinical trials with multiple time-to-event outcomes are common in many disease areas including infectious disease, oncology, and cardiovascular disease. Sample size determination for time-to-event outcomes is more complex compared with continuous or binary outcomes. We discuss methods for calculating the power and sample size for randomized superiority clinical trials with two correlated time-to-event outcomes in the three independent or dependent censoring schemes: (i) where two events are non-fatal, (ii) when one event is fatal (semi-competing risk), and (iii) when both are fatal. We derive an asymptotic form of the bivariate logrank statistic in all three censoring schemes including when considering a composite endpoint consisting of fatal and non-fatal components. We investigate the behavior of power and the required sample sizes in clinical trials with the two inferential goals, where the trial is designed to evaluated if the intervention is superior (or noninferior) on:(i) all of the endpoints, and (ii) at least one endpoint (multiple primary).
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.