Switching between superiority and non-inferiority: an introductory note

The purpose of this note is to introduce a guideline recently released by the Committee for Proprietary Medicinal Products (CPMP) and reproduced on the following pages of this journal. This guideline, described as a ‘Points to Consider’ (PtC) by the CPMP, covers a methodological topic of specific relevance to controlled clinical trials that aim to compare a new medicinal product with an established product. Such a trial may have the objective of showing that the new product is superior to the control treatment. Alternatively the objective may be to show that the new product is not inferior to the control. After the completion of the trial there may be interest in the possibility of changing the original objective. This PtC covers the methodological issues relevant to this change and outlines circumstances when it may be appropriate and possible.

Guidelines relating to the evaluation of clinical efficacy and safety, such as this one, are developed by the Efficacy Working Party (EWP) of the CPMP before being approved by the CPMP itself. They are released for a period of public consultation and the comments received are used to assist with the revision of the document before it is finalized. The topic covered by this PtC was addressed by the EWP because it had arisen in the context of a number of new products for which licences had been sought in recent years. The debate surrounding these products at CPMP had led to the realization that some clarification of the issues would be helpful in the regulatory arena. However, interest in this topic is by no means confined to those involved in the regulation of medicines. It was felt that the publication of this PtC might be helpful to other researchers with the same or similar problems and might also encourage a broader consensus in this area of methodology.

Studies directed at establishing noninferiority have received a great deal of recent regulatory attention. The subject is covered briefly in the Note for Guidance ICH E9 entitled ‘Statistical Principles for Clinical Trials’ [1, 2]. [ICH – International Conference on Harmonization – is a collaborative enterprise involving the pharmaceutical industry and regulatory authorities in Europe, Japan and the USA]. The more recently released Note for Guidance ICH E10 ‘Choice of Control Group’ [3] deals extensively with the topic of noninferiority and provides much fuller guidance on this more general topic. These guidelines and others are available in Europe on the web-site of the EMEA (European Agency for the Evaluation of Medicinal Products) which is http://www.eudra.org/emea.html. The new PtC addresses a very specific topic relating to noninferiority and those wishing to amplify the background will find this additional material useful.

Further methodological guidelines are under development by the CPMP Efficacy Working Party. One of these is expected to cover the choice of the acceptable degree of inferiority in a noninferiority trial (the noninferiority margin – ‘delta’), illustrating once again the high level of current interest in this topic and the difficulties that the increasing use of these designs is highlighting. It is to be hoped that the PtC that follows will help to illuminate some of these difficulties.