Testing Effect of a Drug Using Multiple Modelsfor the Dose-Response

• Activity Number: 688
• Type: Contributed
• Date/Time: Thursday, August 8, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Corine Baayen*+ and Philip Hougaard
• Companies: H. Lundbeck A/S and Lundbeck
• Keywords: dose finding ; multiple testing ; model uncertainty

Abstract

During development of a drug, typically the choice of dose is based on a Phase 2 dose-finding trial, where selected doses are included with placebo. Two common statistical dose finding methods are separate comparisons of each dose to placebo (using a multiple comparison procedure) or a model-based strategy (where a model is fitted to all data). The first approach works best when patients are concentrated on few doses, but cannot conclude on doses not tested. Model-based methods allow for interpolation between doses, but the validity depends on the correctness of the assumed dose-response model. Recently, Bretz et al. (2005; Biometrics 61, 738-748) suggested a combined approach, which selects one or more suitable models from a set of candidate models using a multiple comparison procedure. Candidate models each have 2 parameters; the outcome at dose 0 and a factor to a fully specified function of dose. The method requires a choice of these functions, which makes it somewhat ad hoc. We propose an alternative multiple model, error-controlled, approach, which has good power for simple (e.g. linear) models, but which also tests for more complex nonlinear models with estimated parameters.