A Computational Physics-based Approach to Predict Unbound Brain-to-Plasma Partition Coefficient, K_p,uu

Area of Science:

• Pharmacology
• Computational Chemistry
• Neuroscience

Background:

• The blood-brain barrier (BBB) is crucial for protecting the brain from harmful substances.
• Effective central nervous system (CNS) drug delivery requires optimal penetration of the BBB.
• Current methods for predicting BBB penetration, like medicinal chemistry strategies and in silico models, have limitations in directly forecasting unbound brain/plasma ratio (K_p,uu).

Purpose of the Study:

• To introduce a novel physics-based computational approach for predicting K_p,uu.
• To evaluate the accuracy and utility of this new method in CNS drug discovery.

Main Methods:

• Developed a quantum mechanics (QM)-based energy of solvation (E-sol) computational model.
• Applied the E-sol method to predict K_p,uu for CNS drug candidates.
• Validated the model's performance using internal drug discovery program data.

Main Results:

• The E-sol method demonstrated strong predictive capability for K_p,uu.
• Achieved a categorical accuracy of 79% in predictions.
• Linear regression analysis yielded an R² value of 0.61, indicating good model fit.

Conclusions:

• The QM-based E-sol approach offers a powerful and accurate tool for predicting BBB penetration.
• This method can significantly enhance the efficiency of CNS drug discovery programs.
• The physics-based model provides a valuable alternative to existing predictive strategies for K_p,uu.