On the efficiency of targeted clinical trials

摘要: The development of genomics-based technologies is demonstrating that many common diseases are heterogeneous collections molecularly distinct entities. Molecularly targeted therapeutics often effective only for some subsets patients with a conventionally defined disease. We consider the problem design phase III randomized clinical trials evaluation treatment when there an assay predictive which will be more responsive to experimental than control regimen. compare conventional trial based on randomizing predicted preferentially benefit from new treatment. Trial designs compared required number and expected screened randomization eligibility. Relative efficiency depends upon distribution effect across patient subsets, prevalence subset who respond treatment, performance.