Abstract
The goal of clinical trial research is to deliver safe and efficacious new treatments to patients in need in a timely and cost-effective manner. There is precedent in using historical control data to reduce the number of concurrent control subjects required in developing medicines for rare diseases and other areas of unmet need. The purpose of this paper is to provide a review for a regulatory and industry audience of the current state of relevant statistical methods, and of the uptake of these approaches and the opportunities for broader use of historical data in confirmatory clinical trials. General principles to consider when incorporating historical control data in a new trial are presented. Bayesian and frequentist approaches are outlined including how the operating characteristics for such a trial can be obtained. Finally, examples of approved new treatments that incorporated historical controls in their confirmatory trials are presented.
Highlights
In 2004, the United States (US) Food and Drug Administration (FDA) launched the Critical Path Initiative, which sought to determine the root causes for the latency between laboratory discoveries and their translation into clinical therapies delivered to patients.[1]
If we assume that one-third of those subjects were in control arms, and that by using historical control data we could conservatively reduce that number by 25%, we might have saved over 12,000 subjects from painful procedures that reproduced information that already existed while accelerating clinical trial decision timelines
The method may discard some concurrent control subjects, which can be difficult to justify in the context of a confirmatory clinical trial. To adapt this method for clinical trials, we propose finding matches from the treatment arm for all the concurrent control subjects
Summary
In 2004, the United States (US) Food and Drug Administration (FDA) launched the Critical Path Initiative, which sought to determine the root causes for the latency between laboratory discoveries and their translation into clinical therapies delivered to patients.[1]. The Bayesian paradigm of formally quantifying current knowledge and updating that knowledge in the light of new data fits perfectly with the idea that there is useful information contained within historical data available prior to a clinical trial. Lee and Chu identified 121 publications reporting a Bayesian analysis of a clinical trial; 54 of these publications described use of an informative prior,[18] leading us to speculate that many of these used historical data. Spiegelhalter et al and Sung et al both provide helpful guidelines with a great deal of overlap.[19,20] When using priors that incorporate historical data, details of the historical data sources, the method used to identify these sources, and the weight assigned to each historical dataset need to be reported to ensure the reproducibility of the Bayesian analysis
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
