Design and analysis of biomarker-integrated clinical trials with adaptive threshold detection and flexible patient enrichment

摘要: We propose a new adaptive threshold detection and enrichment design in which the biomarker is adaptively estimated updated by optimizing trade-off between size of positive population magnitude treatment effect that population. Enrichment based on an enrollment criterion accounts for uncertainty estimation threshold. Early termination futility allowed predictive success probability. Valid testing techniques overall inpatient subgroups are studied. Simulations example demonstrate advantages proposed over existing designs.