Abstract:
|
In phase II clinical trials, new treatments/therapies are studied relative to both treatment efficacy and safety. We propose a procedure for selecting a random size subset that contains all experimental treatments that are better than the standard. The comparison is with regard to two binary endpoints, in order to account for both treatment efficacy and safety. An experimental treatment is considered to be better than the standard if its two endpoints have successful rates higher than those associated with the standard. We derive the probability of a correct selection for our procedure and show that, whenever an experimental treatment is significantly better than the standard, our procedure achieves the probability requirements predetermined by the experimenter.
|