A Statistical Design for Investigating Skin Diffusion-Based Dose Response in Humans

• Activity Number: 514
• Type: Contributed
• Date/Time: Wednesday, August 4, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Jihao Zhou*+ and Thomas Lin and Susan Guo and Barbara Scholz
• Companies: Allergan Pharmaceuticals, Inc. and Allergan Pharmaceuticals, Inc. and Allergan Pharmaceuticals, Inc. and Allergan Pharmaceuticals, Inc.
• Keywords: experimental design ; dose response ; clinical trial ; skin diffusion ; optimal design ; biostatistics

Abstract

A statistical design for investigating skin-diffusion dose-response is proposed to assess a compound with 4 doses and 1 placebo when injected in the 6 skin areas (2 columns by 3 rows) of the middle back at 3 levels at the right and left of the spine. A total of 15 sampling schemes with unique dose distribution patterns are proposed. Design properties are: 1) Each subject receives all the doses; a dose-response curve can be built for a subject. 2) Each dose is tested at both areas on the same row the same number of times; it's assumed that the diffusion difference between left and right on the same row is trivial. 3) The number of each dose at each row is equal to six (balanced). 4) The number of each dose at each column is equal to nine (balanced). The design allows for population-based dose-response modeling and seems appealing in assessing intra- and inter-subject variability.