Establishing Efficacy of a New Experimental Treatment in the ‘Gold Standard’ Design

Provided that there are no ethical concerns, the comparison of an active drug with placebo in a randomized two‐arm clinical trial provides the most convincing way to demonstrate the efficacy of a new experimental treatment. However, in a placebo‐controlled clinical trial it is not sufficient to demonstrate merely a statistically significant treatment difference. Regulatory authorities strongly recommend to assess additionally whether the observed treatment difference is also of clinical relevance. The inherent issue is the necessity of the a priori definition of what constitutes a clinically relevant difference in efficacy. This problem can be solved in a three‐arm study by including an active control group. We address the necessary conditions in the gold standard design which allow the claim of efficacy for the new treatment with particular focus on assay sensitivity. (© 2005 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)