|
Abstract:
|
Clinical development has traditionally relied on large sample central limit theory in a frequentist framework. In fact, almost all the operational characteristics that health regulatory agencies evaluate in a clinical trial design are the direct results of the framework. Recently, more and more genetically specific rare diseases have been discovered and consequently more specific treatments have been developed. On the other hand, with an increasing number of regulatory incentive programs many sponsors start to invest on such treatments’ research. The landscape of drug development has been forced to adapt accordingly with smaller and more specific patient population. Bayesian framework, which has been modernized with technology advancement over the last few decades, has its natural mechanism to leverage external data and built-in simulation infrastructure to handle more complicated models. Hence, it has been brought to the center stage in small-population clinical development. We will use case studies to illustrate how Bayesian statistics can be introduced to evaluate efficacy in a clinical trial. Advantage of the framework and caveats in trial designing will be discussed.
|
