A Delayed Response Model to Assess the Proof of Concept Clinical Design in Treatment to Rheumatoid Arthritis

• Activity Number: 616
• Type: Contributed
• Date/Time: Thursday, August 2, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Feng Liu*+ and Daren Austin and Steven Julious
• Companies: GlaxoSmithKline/University of Sheffield and GlaxoSmithKline and University of Sheffield
• Keywords: proof of concept ; Clinical Trial design ; Bayesian ; Adaptive design ; Delayed response ; Rheumatoid Arthritis

Abstract

Rheumatoid Arthritis (RA) is an auto-immume disease and is major cause of disability affecting about 1% of the world's population. The sponsor is currently developing a compound that is in phase 2a clinical trial design (proof of concept study). This abstract describes a case study of flexible design and study implementation in Rheumatoid arthritis proof of concept dose finding study. The proof of concept study design has two parts, part A: learning and part B confirm. Part A is a Bayesian dose finding design and part B is group sequential design with the dose level selected from part A. Bayesian dose finding design uses Bayesian model to inform next dose level for future patients and has more power compared with fixed level design. The Merits and potential setback of the design is to be discussed. A new PoC design using earlier endpoint to predict late time point is proposed to expedite the design making in future RA trials. Simulations have been used to explore the better design options. The proposed delayed response model would significantly reduce the time for decision making in adaptive design.