top of page
Search

Statistical Principles for Confirmatory Clinical Trials

The ICH E9 is a published guidance which provides the statistical principles, specific methods and specific procedural steps to ensure that its principles are implemented properly across the clinical trials.


The principles outlined in the ICH E9 guidance are largely relevant to the clinical trials which are conducted in the later phases of clinical development, where many of these are confirmatory trials of evaluation of clinical efficacy. Such confirmatory trials may have a safety variable which is referred as a primary variable such as an adverse event (AE).


In May 2021, the US FDA issued an Addendum guidance titled, E9 (R1) Statistical Principles for Clinical Trials: Addendum: Estimands and Sensitivity Analysis in Clinical Trials. The purpose of this guidance is to present a structured framework to strengthen the dialogue between disciplines involved in the planning of human clinical trial objectives, the study design, conduct, analysis and interpretation, and the communication between the sponsor and regulator regarding the treatment effect (or effects) of interest that a clinical trial should address.


This addendum clarifies and extends ICH E9 in respect to the following topics: (1) the Intention-To-Treat (ITT) principle; (2) issues considered generally under data handling and missing data; and (3) issues related to the concept of analysis sets which are considered in the clinical trial framework. The principles which are outlined in this addendum guidance are relevant in using wherever a treatment effect is estimated or a hypothesis related to a treatment effect is tested, whether related to efficacy or safety. While this addendum applies primary focus on the randomized clinical trials, these principles are also applicable for single-arm trials and observational studies. Overall, from regulatory perspective, the principles outlined in this document would be a great interest when applied for the confirmatory clinical trials and to estimate and generate confirmatory conclusions, for data integrated across the clinical trials.

Comments


bottom of page