|
Activity Number:
|
539
- SPEED: Bayesian Methods and Applications in the Life and Social Sciences
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2018 : 11:35 AM to 12:20 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
|
Abstract #332686
|
|
|
Title:
|
A Novel Bayesian PK/PD Model for Synergy: Challenges and Opportunities for Sequential Knowledge Integration
|
|
Author(s):
|
Fabiola La Gamba* and Tom Jacobs and Helena Geys and Christel Faes
|
|
Companies:
|
and Janssen R&D and Janssen R&D and Hasselt University
|
|
Keywords:
|
Bayesian inference;
Nonlinear Mixed Models;
Pharmacodynamics;
Pharmacokinetics;
Sequential integration;
Meta-analysis
|
|
Abstract:
|
Studies on pharmacodynamic interactions are usually performed in an in-vitro setting. In this work the co-administration of a novel molecule with a marketed treatment is studied through in-vivo studies performed sequentially. The body temperature change over time is expressed through a turnover model where a virtual pharmacokinetic profile of the marketed treatment drives the effect. A pharmacodynamic interaction is assumed at IC50. The aim of this work is to discuss the implications of performing the model in a Bayesian sequential manner, so that the posteriors resulting from a study are used to determine the priors of the next study. The assessed modelling aspects are: 1.Impact of prior elicitation; 2.Speci fication of random effect; 3.Impact of different Bayesian sequential integrations. The model worked well with informative priors, random baseline and when a wide dose range was investigated in each study. The integration of studies where one or few dose combinations were assessed produced biased results. This highlights the importance of a careful design of experiment for a successful sequential integration.
|
Authors who are presenting talks have a * after their name.
