A group sequential type design for three-arm non-inferiority trials with binary endpoints

• Published in: Biometrical journal Vol. 52; no. 4; pp. 504 - 518
• Main Authors: Li, Gang, Gao, Shan
• Format: Journal Article
• Language: English
• Published: Weinheim WILEY-VCH Verlag 01.08.2010; WILEY‐VCH Verlag; Wiley-VCH
• Subjects: Applications; Biology, psychology, social sciences; Biometrics; Clinical trials; Clinical Trials as Topic - methods; Effect preservation test; Exact sciences and technology; Experimental design; General topics; Group sequential design; Humans; Mathematics; Monte Carlo Method; Monte Carlo simulation; Non-inferiority; Preservation; Probability and statistics; Risk; Sciences and techniques of general use; Sequential methods; Statistics; Three-arm trial; Biometrics; Paired comparison; Statistical distribution; Stochastic method; Variance; Hypothesis test; Statistical test; Optimal allocation; Sequential method; Non-inferiority; Approximation theory; Three-arm trial; Effect preservation test; Monte Carlo method; Sequential test; Sequential design; Test statistic; Covariance analysis; Life science; Group sequential design; Multiple comparison; Variance analysis; Placebo effect; Group design; Statistical method; Statistical regression; Numerical analysis; Simulation; Experimental design; Maximum likelihood; Sequential estimation
• ISSN: 0323-3847; 1521-4036; 1521-4036
• DOI: 10.1002/bimj.200900188

The three‐arm design with a test treatment, an active control and a placebo group is the gold standard design for non‐inferiority trials if it is ethically justifiable to expose patients to placebo. In this paper, we first use the closed testing principle to establish the hierarchical testing procedure for the multiple comparisons involved in the three‐arm design. For the effect preservation test we derive the explicit formula for the optimal allocation ratios. We propose a group sequential type design, which naturally accommodates the hierarchical testing procedure. Under this proposed design, Monte Carlo simulations are conducted to evaluate the performance of the sequential effect preservation test when the variance of the test statistic is estimated based on the restricted maximum likelihood estimators of the response rates under the null hypothesis. When there are uncertainties for the placebo response rate, the proposed design demonstrates better operating characteristics than the fixed sample design.