Importance of reversed-phase chromatographic parameters in predicting biopharmaceutical and pharmacokinetic descriptors on the group of androgen derivatives

Abstract

© 2017 Elsevier B.V. Nowadays the standard measure of lipophilicity, the logarithm of n-octanol-water partition coefficient, logP, is proposed to be replaced with chromatographic techniques. Chromatography techniques (reversed phase thin layer chromatography RPTLC and reversed phase thin layer chromatography RPHPLC) are the most widely used alternatives to the shake flask method. However, it is shown that, by changing the temperature or concentration of organic modifier in the chromatography experiment, it is possible to derive data matrix of retention parameters from which, by principle component analysis, structural characteristics of the examined molecules can be gained. The question may be asked which of the chromatography experimentally obtained and calculated parameters: capacity factor k, ΔGx (the change in Gibbs energy of binding of molecule for stationary phase), ΔHx (the change in enthalpy of binding of molecule for stationary phase) or ΔSx (the change in the entropy of binding of molecule for stationary phase) is the most suitable in describing hydrophobicity. The canonical correlation analysis (CCA) method is used to evaluate the importance of the n functions in explaining the variance of molecular descriptors connected to pharmaceutical processes and wherein molecule's hydrophobicity is expressed and possible differences between molecular descriptors with realistic conformations of the analyzed molecules steroid skeleton are discussed. Conformational analysis showed that structure of steroid skeleton in hydrophobicity is most completely described with k or ΔGx, and connection between conformation of the steroid skeleton and hydrophobicity to a lesser extent is projected on temperature dependence on ΔHx and similarly on ΔSx, so in describing molecules hydrophobicity it is necessary to observe entropic as well as enthalpic contribution together, expressed with ΔGx function. Canonical conformation analysis (CCA) showed that hydrophobicity contained in ΔGx and k explains 61% of variance represented in in silico descriptors. Analyzed molecular descriptors, derived from different molecules fragments don't map conformational specifics of those molecules in small groups so recommendation is to use them complementary with chromatographic data in describing hydrophobicity.