Leveraging historical data for use in rare disease trials

Rare disease clinical trials have several challenges that make them more difficult to conduct than clinical trials for more common diseases. Small patient populations, pediatric populations and the desire by patients and their caregivers to receive active therapy while on a clinical trial make it much more likely that these trials will need to incorporate historical data into the design.

Both the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) have shown a willingness to accept studies with historical controls. There are two primary ways to analyse historical data - the Bayesian approach and the Frequentist approach. This paper aims to:

• Increase awareness among drug developers of these two approaches
• Provide an overview on how to use Frequentist and Bayesian methods to incorporate historical data into the study design
• Outline the differences between the two approaches
• Review recent developments in the regulatory landscape
• Demonstrate how this approach has been used by the US FDA and by ICON clients