Abstract
A mathematical equation model was developed by building the relationship between the f(u,b)/f(u,p) ratio and the computed physicochemical properties of candidate compounds, thereby predicting K(p,uu,brain) based on a single experimentally measured K(p,brain) value. A total of 256 compounds and 36 marketed published drugs including acidic, basic, neutral, zwitterionic, CNS-penetrant, and non-CNS penetrant compounds with diverse structures and physicochemical properties were involved in this study. A strong correlation was demonstrated between the f(u,b)/f(u,p) ratio and physicochemical parameters (CLogP and ionized fraction). The model showed good performance in both internal and external validations. The percentages of compounds with K(p,uu,brain) predictions within 2-fold variability were 80.0 %-83.3 %, and more than 90 % were within a 3-fold variability. Meanwhile, "black box" QSAR models constructed by machine learning approaches for predicting f(u,b)/f(u,p) ratio based on the chemical descriptors are also presented, and the ANN model displayed the highest accuracy with an RMSE value of 0.27 and 86.7 % of the test set drugs fell within a 2-fold window of linear regression. These models demonstrated strong predictive power and could be helpful tools for evaluating the K(p,uu,brain) by a single measurement parameter of K(p,brain) during lead optimization for CNS penetration evaluation and ranking CNS drug candidate molecules in the early stages of CNS drug discovery.