Using a Pattern of Response Outcome for a Pilot Off-on-Off Crossover Design

• Activity Number: 350
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Jonathan D. Mahnken*+ and Russell H. Swerdlow
• Companies: University of Kansas Medical Center and University of Kansas Medical Center
• Keywords: Randomization ; Single-arm trial ; Small sample size ; Study design

Abstract

An initial design for a pilot-phase trial of a new treatment is presented. The design is a single-blind, single-arm trial where each subject is measured following a placebo phase, after treatment, and after drug is discontinued. As no preliminary data are provided the sample size is justified based on the anticipated pattern of response. The research hypothesis is for an increase in response to treatment, so observations are defined as responders (i.e., a "successes") if their individual change from the measure following placebo to that following treatment has a greater slope than that of the change from the measure following the active treatment period to following the wash-out period. Using this approach of identifying successes based on response patterns is compared to an alternative, randomized design where the small sample size supported by this trial divided between treatments. The proposed design provides some protection against temporal trends in case subjects have consistent increasing or decreasing trends in the outcome measure.