All Times EDT
Keywords: type i error control, alpha spending, interim analysis, multiplicity issue, rare disease
Although it has been more than three decades since Orphan Drug Act was signed into law, most rare diseases don’t have approved treatment yet. Among them, many diseases are progressive, life-threatening, and genetic inherited which even affect children at early age. Therefore, there is a growing need to speed up the development of treatment for rare disease. As interim analyses provide evidence at the middle of the study, it allows to make early decision, such as sample size re-estimation or study termination, and it may help reduce the time to market. However, there are many challenges in study design for interim analysis in rare disease which worth bringing to the discussion. For example, due to the small sample size and potential normality assumption violation, the alpha spending could be very different from the traditional clinical trials between interim and final analyses. The same issue could also lead to the complexity of multiplicity issue. For example, when the same endpoint are assesses at two or multiple timepoints, how to order endpoints and define critical values between these endpoints with their association appropriately estimated. In this round table discussion, we will share and discuss well-developed and agency accepted approaches allowing type I error control when small sample size issues are addressed.