Fraction of a dose absorbed estimation for structurally diverse low solubility compounds.

The purpose of the present study was to investigate the prediction accuracy of the fully mechanistic gastrointestinal unified theoretical (GUT) framework for in vivo oral absorption of low solubility drugs. Solubility in biorelevant media, molecular weight, logP(oct), pK(a), Caco-2 permeability, dose and particle size were used as the input parameters. To neglect the effect of the low stomach pH on dissolution of a drug, the fraction of a dose absorbed (Fa%) of undissociable and free acids were used. In addition, Fa% of free base drugs with the high pH stomach was also included to increase the number of model drugs. In total twenty nine structurally diverse compounds were used as the model drugs. Fa% data at several doses and particle sizes in humans and dogs were collated from the literature (total 110 Fa% data). In approximately 80% cases, the prediction error was within 2 fold, suggesting that the GUT framework has practical predictability for drug discovery, but not for drug development. The GUT framework appropriately captured the dose and particle size dependency of Fa% as the particle drifting effect was taken into account. It should be noted that the present validation results cannot be applied for salt form cases and other special formulations such as solid dispersions and emulsion formulations.