Predicting Caco-2/MDCK intrinsic membrane permeability from HDM-PAMPA-derived hexadecane/water partition...

Area of Science:

• Drug discovery and development
• Computational chemistry
• Biophysical chemistry

Background:

• Accurate membrane permeability data are vital for early drug development.
• Predictive tools for membrane permeability are needed for high-throughput screening.
• Previous work showed the solubility-diffusion model predicts passive permeability using Khex/w.

Purpose of the Study:

• Investigate the HDM-PAMPA method for determining Khex/w.
• Validate Khex/w predictions using experimental and in silico methods.
• Assess the utility of Khex/w for high-throughput permeability estimation.

Main Methods:

• Measured Khex/w for 64 compounds using HDM-PAMPA.
• Compared HDM-PAMPA Khex/w with black lipid membrane (BLM) and two-phase systems.
• Evaluated in silico Khex/w prediction using COSMOtherm and LSER.

Main Results:

• HDM-PAMPA showed good agreement with established Khex/w determination methods.
• Experimentally determined Khex/w values accurately predicted Caco-2 and MDCK cell membrane permeability (RMSE = 0.8).
• COSMOtherm in silico Khex/w prediction performed well (RMSE = 1.20), outperforming LSER (RMSE = 1.63).

Conclusions:

• The HDM-PAMPA method is a practical tool for determining Khex/w.
• Experimentally derived Khex/w values enable accurate passive permeability predictions.
• In silico Khex/w prediction, especially with COSMOtherm, offers a viable alternative for high-throughput screening.