Evaluation of Using Dogs to Predict Fraction of Oral Dose Absorbed in Humans for Poorly Water-Soluble Drugs

Global Analysis of Models for Predicting Human Absorption: QSAR, In Vitro, and Preclinical Models

Models intended to predict intestinal absorption are an essential part of the drug development process. Although many models exist for capturing intestinal absorption, many questions still exist around the applicability of these models to drug types like "beyond rule of 5" (bRo5) and low absorption compounds. This presents a challenge as current models have not been rigorously tested to understand intestinal absorption. Here, we assembled a large, structurally diverse dataset of ∼1000 compounds with known in vitro, preclinical, and human permeability and/or absorption data. In silico (quantitative structure-activity relationship), in vitro (Caco-2), and in vivo (rat) models were statistically evaluated for predictive performance against this human intestinal absorption dataset. We expect this evaluation to serve as a resource for DMPK scientists and medicinal/computational chemists to increase their understanding of permeability and absorption model utility and applications for academia and industry.

Predicting the effect of prandial stage and particle size on absorption of ODM-204

The prediction of absorption properties plays a key role in formulation development when the compound under development shows poor solubility and its absorption is therefore presumed to be solubility limited. In our work, we combined and compared data obtained from in vitro dissolution tests, transit intestinal model studies (TIM-1) and physiologically based pharmacokinetic modelling. Our aim was to determine the ability of these methods to predict performance of poorly soluble lipophilic weak base in vivo. The validity of the predictive methods was evaluated against the in vivo clinical pharmacokinetic (PK) data obtained after administration of the first test formulation, T1. The aim of our study was to utilize the models in evaluating absorption properties of the second test formulation, T2, which has not yet been clinically administered. The compound in the studies was ODM-204, which is a novel, orally administered, investigational, nonsteroidal dual inhibitor of CYP17A1 and androgen receptor. Owing to its physicochemical properties ODM-204 is prone to low or variable bioavailability. The models examined provided congruent data on dose dependent absorption, food effect at a dose of 200 mg and on the effect of API (active pharmaceutical ingredient) particle size on absorption. Our study shows that the predictive tools of in vitro dissolution, TIM-1 system and the PBPK (physiologically based pharmacokinetic) simulation, showed predictive power of different mechanisms of bioavailability and together provided valuable information for decision making.

Prediction Characteristics of Oral Absorption Simulation Software Evaluated Using Structurally Diverse Low-Solubility Drugs

The purpose of the present study was to characterize current biopharmaceutics modeling and simulation software regarding the prediction of the fraction of a dose absorbed (Fa) in humans. As commercial software products, GastroPlus™ and Simcyp® were used. In addition, the gastrointestinal unified theoretical framework, a simple and publicly accessible model, was used as a benchmark. The Fa prediction characteristics for a total of 96 clinical Fa data of 27 model drugs were systematically evaluated using the default settings of each software product. The molecular weight, dissociation constant, octanol-water partition coefficient, solubility in biorelevant media, dose, and particle size of model drugs were used as input data. Although the same input parameters were used, GastroPlus™, Simcyp®, and the gastrointestinal unified theoretical framework showed different Fa prediction characteristics depending on the rate-limiting steps of oral drug absorption. The results of the present study would be of great help for the overall progression of physiologically based absorption models.

Reconciling Human-Canine Differences in Oral Bioavailability: Looking beyond the Biopharmaceutics Classification System

The extrapolation of oral bioavailability (F) information between dogs and humans has had an important role in the drug development process, whether it be to support an assessment of potential human pharmaceutical formulations or to identify the bioavailability challenges that may be encountered in dogs. Accordingly, these interspecies extrapolations could benefit from a tool that helps identify those drug characteristics consistent with species similarities in F. Our initial effort to find such a tool led to an exploration of species differences as it pertained to the biopharmaceutics classification system (BCS). However, using a range of compounds, we concluded that solubility and permeability alone could not explain interspecies inconsistencies in estimates of F. Therefore, we have now extended our evaluation to include canine versus human comparisons of F based upon the biopharmaceutics drug disposition classification system (BDDCS) and the extended clearance classification system (ECCS). Using the same data as that in our initial BCS assessments, we conclude that although neither the BDDCS nor the ECCS can reliably improve our ability to determine when F will be similar in humans and dogs, the ECCS provides a mechanism to help define possible causes for observed human-canine inconsistencies.

Prediction Accuracy of Mechanism-Based Oral Absorption Model for Dogs

The purpose of the present study was to evaluate the prediction accuracy of a mechanism-based oral absorption model for the fraction of a dose absorbed (Fa) in dogs, focusing on poorly soluble drugs. As an open mechanism-based model, the gastrointestinal unified theoretical framework was used in this study. The prediction accuracy of the gastrointestinal unified theoretical framework was evaluated using Fa data in dogs (63 data sets for marketed drugs and proprietary compounds). For neutral compounds, Fa was accurately predicted, suggesting that the physiological parameters of dogs were appropriate except for gastrointestinal pH. An extensive literature survey on the small intestinal pH of dogs was then conducted. The result suggested that the pH value ranged between 6.5 and 7.5, with the midst value of 7.0, but there was a great variation among the literature. To confirm the appropriateness of this pH value, the Fa of free acid compounds was predicted by setting the small intestinal pH to 6.5, 7.0, and 7.5. The proportions of compounds with <2-fold error were 57%, 90%, and 76%, respectively. The results of the present study would enable an appropriate use of a mechanism-based model for drug discovery and development.