In Silico Quantitative Structure Pharmacokinetic Relationship Modeling for Quinolone Drugs: Biological Half-Life

Abstract

The use of in silico approaches for successful prediction of pharmacokinetic properties of compounds during new drug discovery has been increasing exponentially. These in silico models, for the prognosis of absorption, distribution, metabolism and excretion (ADME) are invariably based upon the implementation of quantitative structure pharmacokinetic relationship (QSPkR) techniques. This study was conducted to investigate QSPkR for biological half-life (t_1/2) in humans for 28 quinolone drugs employing extra-thermodynamic multi-linear regression analysis (MLRA) approach. The overall predictability was found to be high (R² = 0.8752, F = 20.24, S² = 9.3212, Q² = 0.7384, p < 0.001). Topological, steric and electrostatic parameters were found to primarily ascribe the variation in t_1/2. Logarithmic transformations of t_1/2 tend to improve the degree of correlations during one-parameter and two-parameter studies. However, the inverse transformations of t_1/2 remarkably enhance the degree of correlations (both R² and Q²). Maximum predictability for quinolones was found to be 94.16 %.