Kelley Kidwell, of the Department of Biostatistics at the University of Michigan, will present: “Design and Methodological Improvements for Rare Disease Research: Use of snSMART Design”

Abstract: A small n, sequential, multiple assignment, randomized trial (snSMART) is a small sample, multi-stage design where subjects may be re-randomized to treatment based on intermediate endpoints. Our methods are motivated by ARAMIS (NCT09239573), an ongoing snSMART focusing on the evaluation of three drugs for isolated skin vasculitis. We will compare and contrast the snSMART design and analyses to standard SMART design and analyses. We will also present an overview of Bayesian and Frequentist methods to identify the best first stage treatment and adequately size an snSMART with the goal of comparing the best and second-best treatments.