Predicting human pharmacokinetics from preclinical data: absorption.

Area of Science:

• Pharmacokinetics
• Drug Discovery
• Computational Biology

Background:

• Predicting oral drug absorption in humans presents significant challenges in drug development.
• Established in vitro and physiologically-based pharmacokinetic (PBPK) methods have been developed over decades to address this.
• Accurate prediction of drug bioavailability is crucial for efficient drug development.

Purpose of the Study:

• To review in vitro and PBPK methods for predicting human oral drug absorption.
• To discuss the key parameters and challenges involved in these prediction models.
• To emphasize the importance of user expertise in applying these predictive tools.

Main Methods:

• Utilizing physicochemical properties and permeability data (e.g., Caco-2 system) to determine fraction absorbed (F_a).
• Employing intrinsic clearance from human microsomes or hepatocytes to predict gut (F_g) and hepatic (F_h) bioavailability.
• Calculating overall bioavailability (F) using the formula F = F_a × F_g × F_h and predicting the absorption rate constant (k_a).

Main Results:

• Physicochemical and permeability data are essential for predicting F_a.
• Intrinsic clearance is required for F_g and F_h predictions, though challenges like inter-laboratory variability and CYP3A4 dependency exist.
• The absorption rate constant (k_a) can be predicted from in vitro and in vivo data, but requires careful interpretation.

Conclusions:

• PBPK models offer a framework for predicting oral absorption, but require careful application due to inherent assumptions and variances.
• Issues such as inter-laboratory variability and specific metabolic pathway dependencies need to be addressed for reliable predictions.
• User knowledge and experience in PBPK and in vitro techniques are critical for accurate human pharmacokinetic predictions.