Prediction Characteristics of Oral Absorption Simulation Software Evaluated Using Structurally Diverse Low-Solubility...

Area of Science:

• Pharmacokinetics and Biopharmaceutics
• Computational Modeling and Simulation
• Drug Absorption and Metabolism

Background:

• Accurate prediction of the fraction of a dose absorbed (Fa) is crucial for drug development.
• Biopharmaceutics modeling and simulation software are increasingly used to predict in vivo drug performance.
• Understanding the performance of these tools is essential for reliable in silico predictions.

Purpose of the Study:

• To characterize and compare the Fa prediction capabilities of commercial software (GastroPlus™, Simcyp®) and a theoretical framework.
• To evaluate the performance of these models using a comprehensive dataset of clinical Fa data.
• To identify factors influencing the accuracy of oral drug absorption predictions.

Main Methods:

• Systematic evaluation of GastroPlus™, Simcyp®, and the gastrointestinal unified theoretical framework.
• Utilized 96 clinical Fa data points from 27 model drugs with default software settings.
• Input parameters included molecular weight, pKa, logP, solubility, dose, and particle size.

Main Results:

• Significant differences in Fa prediction characteristics were observed among the evaluated models.
• Model performance varied depending on the identified rate-limiting steps in oral drug absorption.
• Despite using identical input parameters, each model exhibited unique prediction behaviors.

Conclusions:

• The study highlights the distinct predictive capabilities of different biopharmaceutics modeling approaches.
• Understanding model-specific behaviors is key to selecting appropriate tools for Fa prediction.
• These findings contribute to the advancement of physiologically based absorption models in pharmaceutical research.