[Biometrical methods for the proof of efficacy in regulatory submissions. Non-inferiority in clinical studies].
The increasing number of approved efficacious therapies for various indications raises the question of whether the inclusion of a placebo group is still justified. In addition, pharmaceutical companies and regulatory agencies acknowledge that it may be sufficient to prove that a new therapy is comparable to an approved reference therapy regarding efficacy and safety in some situations. This becomes especially striking for the approval to market a generic drug. Another, perhaps even more important example is the increasing resistance of bacteria which calls for new antibiotics based on new therapeutic principles without having the claim for better efficacy. In these situations, a comparable efficacy would constitute progress. In the present paper, we discuss the numerous methodological challenges and approaches to overcome these problems that occur when it is not possible or even not wanted to use the classic approach of a randomized placebo-controlled superiority trial. Here, the field of medical biometry, which has proved in the last 25 years to be an integral part of the development of new drugs, demonstrates its suitability as a flexible and scientifically based means to fulfil the requirements resulting from clinical practice. Starting from the fact that statistical methods are not able to prove "equality" of two treatments, "shifted" hypotheses are considered and their importance for the different study designs is discussed. We show how the classic hypotheses known from placebo-controlled clinical trials can be embedded in this concept. The implications of this approach for the analysis and interpretation of study results is further discussed. The relevant guidelines of the European and US regulatory agencies are taken into account.