Statistical considerations in evaluating pharmacogenomics-based clinical effect for confirmatory trials.

摘要: Background The current practice for seeking genomically favorable patients in randomized controlled clinical trials using genomic convenience samples.Purpose To discuss the extent of imbalance, confounding, bias, design efficiency loss, type I error, and II error that can occur evaluation samples, particularly when they are small samples. articulate statistical considerations a reasonable sample size to minimize chance and, highlight importance replicating subgroup finding independent studies.Methods Four case examples reflecting recent regulatory experiences used underscore problems with Probability imbalance pre-specified is provided elucidate needed imbalance. We use an example drug development level scientific rigor needed, evidence replicated claim.Results samples evalua...