CHOICE OF CONTROL IN CLINICAL TRIALS - ISSUES AND IMPLICATIONS OF ICH-E10 On behalf of the European Federation of Statisticians in the Pharmaceutical Industry

The ICH-E10 guideline on the choice of control group in clinical trials was a long awaited document. The concept paper gave reason to hope that it would give clear and harmonised guidance to an area, which has been very much disputed, including the burning issue of the role of placebo control. In the main, the final guideline was disappointing in spite of the long preparation time. In many respects it did not succeed to provide harmonised guidance across regions and it is not specific enough on a number of issues leaving the drug development stakeholders uncertain about what needs to be done. Examples of such issues are: (i) the focus on individual studies rather than a whole clinical development program, (ii) the choice of active control and the non-inferiority margin(δ) in a noninferiority trial, (iii) the interpretation of “assay sensitivity”, (iv) the bias in favour of placebo-control, (v) the failure of proposed alternative designs to resolve the unethical use of placebo. The guideline does not acknowledge the gradual change of the clinical trial environment where placebo-controlled trials will be more and more difficult to conduct. This is driven by the existence of effective treatments in most therapeutic areas in combination with the new version of the Helsinki declaration. Sooner or later efficacy for a new drug will need to be demonstrated using active-controlled non-inferiority studies for most indication areas. In order to meet this inevitable evolution efforts must be spent to further develop the methodology for non-inferiority trials, and to ensure that published meta-analyses provide the necessary information to allow the design of high quality non-inferiority studies in the future.