Challenges Encountered in Conducting Rare Disease Clinical Trials
*Min Min, FDA 


Rare diseases literally mean diseases that are occurring in small patient populations. In extreme cases, the number of patients who can participate in a clinical trial can potentially be fewer than ten. At times it is not feasible to conduct a traditional, so-called, “adequate and well controlled” clinical trial with type I error control. In studying these ultra-rare diseases, two major challenges are encountered: 1) lack of sufficient study power to detect the treatment effect; and 2) lack of comprehensive disease knowledge for identifying suitable endpoints for assessing drug effect.

To tackle these challenges, potential solutions can be: 1) exploring treatment effect in terms of patient profiles; 2) adopting innovative trial designs; and 3) comparing trial data with natural history data. In particular, regarding patient profiles, we can use them to examine patients’ disease progress and improvement in regards to the timing of treatment interventions. In designing a trial for a rare disease it might be unethical to include a concurrent control. In this circumstance, a multi-arm trial that includes a well-defined natural history study may be considered. In including an historical control, it may useful to consider whether we should potentially use a Bayesian statistical approach. .

In my presentation, I will share my FDA experience in evaluating rare disease clinical trials. Several cases with different types of challenges will be discussed. I will also offer the audience my recommendations.