APPLICATION OF NEURAL NETWORKS FOR THE PREDICTION OF HUMAN PHARMACOKINETIC PARAMETERS

Artificial neural network (ANN) is a method used in the prediction of response variables from a set of input and target parameters. The most commonly used network in the area of pattern recognition is the feed forward/back propagation (BPN) network. A method to predict human phar macokinetic parameters has been proposed using BPN with a combination of physicochemical properties and animal pharmacokinetic parameters. T he results were compared with in vitro estimation of the same pharmaco kinetic parameters. Fourteen network models, using a variety of input variables, were developed. Protein binding, partition coefficients, di ssociation constants, and the total clearance (Cl-tot) and volume of d istribution distribution (V-z) in rat and dog species of 41 drugs were evaluated for prediction of human total clearance and volume of distr ibution using the EDBD algorithm. The observation showed highest predi ction for Cl-tot and V-z when rat and dog pharmacokinetics, combined w ith protein binding and partition coefficients of the drugs, were used as input parameters. Drugs with a partition coefficient (log P) < 1.1 7 showed predictability of 63.41% for Cl-tot and 48.78% for V-z. Drugs with low protein binding (similar to 20%) showed predictability of 19 .51% for Cl-tot and 41.46% for V-z. Comparison with in vitro estimatio n showed no bias in the prediction of either either clearance (p < 0.2 ) or volume of distribution (p < 0.5) by the two methods.