414059 The Use of Gastroplus Modeling in Enabling Drug Discovery Research

Tuesday, November 10, 2015: 3:15 PM
Ballroom D (Salt Palace Convention Center)
Ching Kim Tye1, Zhanbin Wang1, Ching Chiang Su1, John R. Crison2 and Michael J. Hageman1, (1)Bristol-Myers Squibb, Princeton, NJ, (2)Bristol-Myers Squibb, New Brunswick, NJ

The use of in vivo modeling software, for example, GastroPlus, for oral drug absorption is well established in drug development research; however, their related usage in Discovery is much less explored.  The main reasons for its low usage in discovery are the inadequate definition of solid forms, the general lack of data, and the inadequate understanding of how to translate in vivo modeling output from animal data to predict human responses.  While it is difficult to fully predict human responses from animal data, the aim of drug delivery scientists in discovery is to assess the developability of potential drug molecules to provide directional guidance and not absolute guidance.  In vivo modeling software can serve as a powerful risk assessment tool in discovery.

A case example of using GastroPlus in combination to other more conventional discovery tools to enable project progression in discovery research will be presented.  The combined information package provided clarification about the risks and its potential mitigation strategy.  To evaluate the risk of pH-dependent absorption effect for a potential drug molecule (weak base) in human, a 3-pronged approach - GastroPlus simulation, in vitro, and in vivo experiments - was applied.  For GastroPlus simulation, as the first step, in vitro experiments were carried out to define related physicochemical properties of the compounds including solubility, permeability, melting points, pKa, Log D, followed by the gathering of pharmacokinetics and oral bioavailability of the compound in rodents with intravenous injection, and oral solutions and suspensions at several doses. In vitromicro dissolution experiments were also carried out to understand the speed of precipitation from supersaturation in Fasted Simulated Intestinal Fluid using a pH swift method. This pH swift dissolution test itself is also a method to evaluate pH-dependent effect. 

GastroPlus simulations were then used to build a model for describing the compound absorption process based on all available data, including particle size. Multiple rodent-base physiology models in the GastroPlus program were tested, modified, and optimized to obtain a best fit model.  A parameter sensitivity analysis was then used to analyze the effect of stomach pH to AUC and Cmax in rats based on microsuspension.  The model provided predictions that were useful from multiple perspectives:

  1. The potential of stomach pH variability in affecting total bioavailability
  2. The impact of drug dose in pH-dependent absorption
  3. The potential impact of particle size in bioavailability and its relationship to pH variability

An in vivorodent study with the use of famotidine (proton pump inhibitor) and pentagastrin (acidic promoter) was also conducted to evaluate the pH-dependent absorption, and the results aligned with the GastroPlus prediction.

The information package was then used to provide guidance for evaluating the risk of pH-dependent absorption effect in human.  This exercise formed the basis of risk assessment output for the pH-dependent liability for the program and served as guidance information for the creation of clinical development plan. The study demonstrates the use of GastroPlus in discovery setting in enable decision making and project progression.

Extended Abstract: File Uploaded