Background: The regulatory agencies in Taiwan defines NCE2 (New Chemical Entity 2) as those compound drugs that have been approved and marketed for five years in the top ten pharmaceutically advanced countries but are new in Taiwan. To apply registration of NCE2 in Taiwan, a clinical trial may be conducted in Taiwan to evaluate the efficacy and safety. However, if the trial design is based conventional significance testing, the requirement of large sample is inevitable. Objectives: We will design a clinical trial to evaluate the efficacy and safety of NCE2 based on the concept of consistency. Methods: We propose a Bayeian approach to design and evaluation of NCE2 trial. More specifically, an empirical Bayes method with a mixture prior information is suggested to synthesize the data from both Taiwan and other countries to assess the onsistency of the treatment effect of NCE2 in Taiwan with other countries. Results: Since the NCE2 has been approved in at least one of in the top ten pharmaceutically advanced countries, we can construct the empirical prior information for the treatment effect for the NCE2 trial on the basis of all of the observed data from other countries. We will conclude similarity between the NCE2 trial and the results from other countries if the posterior probability of deriving a positive treatment effect in the NCE2 trial is large, say 80%. Numerical examples illustrate applications of the proposed approach in different scenarios. Methods for sample size determination for the NCE2 trial are also proposed. Conclusions: We have developed a Bayeian approach to design and evaluation of NCE2 trial. For assessment of the treatment effect of NCE2, the similarity criterion is established by a Bayesian approach. The concept of similarity is based on statistical prediction instead of conventional significance testing. With this approach, the total sample size required for the NCE2 trial might be reduced. That is, shortening the total duration of conducting the NCE2 trial may be possible.
Date:
2014-10
Relation:
Pharmacoepidemiology and Drug Safety. 2014 Oct;23(S1):432.
