Artificial intelligence for compound pharmacokinetics prediction.

Area of Science:

• Pharmacology
• Drug Discovery
• Computational Chemistry

Background:

• Compound pharmacokinetics (PK) optimization is crucial in drug discovery and development.
• In vivo animal data and in vitro systems are routinely used to assess human PK.
• Machine learning (ML) and artificial intelligence (AI) are increasingly vital tools in this field.

Purpose of the Study:

• To review recent advancements in ML and AI models for predicting in vivo animal and human PK.
• To highlight the application of these models in guiding drug design and prioritizing molecules.
• To showcase examples for both small-molecule compounds and antibody therapeutics.

Main Methods:

• Review of recent literature on ML and AI applications in PK modeling.
• Analysis of models predicting PK from chemical structures.
• Examination of studies utilizing in vivo animal and in vitro data for human PK prediction.

Main Results:

• ML and AI enable early prediction of PK properties from chemical structures.
• These technologies facilitate the design and prioritization of molecules with desired in vivo characteristics.
• AI models offer improved prediction of human PK at the molecular design stage.

Conclusions:

• ML and AI are powerful tools for optimizing compound pharmacokinetics.
• These approaches accelerate drug discovery by enabling early and accurate PK prediction.
• The integration of AI is transforming the prediction of human PK for small molecules and biologics.