Exploring the utility of the pig model for predicting bioavailability in humans

Purpose It has been increasingly recognised that there is a need to reduce costly drug development delays and facilitate earlier access to market for new medicines. Within drug development, there is considerable debate on which preclinical animal model is most appropriate for assessing oral bioavailability in humans. In order to streamline preclinical development, and reduce repeated testing in various animal models, it is necessary to get a better understanding which particular animal model is suitable for a specific drug candidate. In pharmaceutical drug development, preclinical tests in animal models are essential to demonstrate whether the new drug is orally bioavailable and to gain a first insight into in vivo pharmacokinetics (PK) parameters that can subsequently be used to predict human values. Preclinical testing, therefore, provides essential scientific evidence in advance of pivotal clinical testing in humans on the safety and efficacy of new drug candidate. Despite significant advances in development of bio-predictive in vitro models and increasing ethical expectations for reducing the number of animals used for research purposes, there is still a need for appropriately selected preclinical in vivo testing to provide guidance on the decision to progress to testing in humans. The selection of appropriate animal models is essential both to maximise the learning that can be obtained from such experiments and to avoid unnecessary testing in a range of species. Selection of the right species for the various questions and compounds is therefore a critical step. While there has been extensive discussion in the literature on the use of dog and non-human primate models to predict oral bioavailability in humans (Musther et al., 2014), in relation to the pig model there are significant gaps in our understanding. Methods The suitability of the pig as a preclinical model within pharmaceutical research was evaluated using bioenabling formulations of the model drugs fenofibrate and venetoclax. In addition, the ability of predicting food dependent bioavailability was assessed for both compounds. The formulations were evaluated in vivo in landrace (LR) pigs. Followed by a post-mortem assessment of the gastrointestinal (GI) fluids, to assess fasted and fed study protocols. Subsequently the physiological composition of the porcine GI fluids has been characterised in terms of pH, buffer capacity, osmolality, surface tension, as well as the bile salt, phospholipid and free fatty acid content. Based on the findings a porcine biorelevant media (FaSSIFp) was proposed. Porcine GI transit conditions were studied with a telemetric motility capsule (SmartPill®) under fasted and postprandial conditions, by determining pH values, pressure and temperature. The physiological GI knowledge of pigs has been combined in a pig specific absorption model, to predict drug levels prospectively. Results This thesis has firstly demonstrated the suitability of the pig as a preclinical model for predicting oral bioavailability in humans. Moreover, it was the first report in the literature, where the pig model was used successfully capturing clinical reported food dependent bioavailability of the two model drugs fenofibrate and venetoclax. It was further shown that established study protocols, achieved fasted state conditions in pigs. Secondly a porcine biorelevant medium was proposed and has shown the advantage to predict luminal solubility in pigs more accurately in comparison to human biorelevant media. Thirdly an improved porcine physiologically based absorption model was used to prospectively simulate the impact of food on the bioavailability of venetoclax. By using porcine biorelevant media in combination with an in silico pig specific absorption model, it was possible to correctly predict the observed pig plasma concentration profile and food effect. Conclusion The thesis demonstrated that pigs are a suitable model in preclinical research to predict drug product performance and oral bioavailability in humans, including challenging compounds and different formulation strategies. The established porcine physiologically based pharmacokinetic (PBPK) model in combination with the proposed porcine biorelevant medium, can lead to more efficient preclinical processes and provide for more informed decision making on the formulation properties, as summarized in a species specific development guide. Establishing reliable and appropriate in vitro models to reflect preclinical conditions, can lead to reduction and replacement of animal experiments in preclinical development processes. Overall, the impact of this thesis is that the range of application of the pig model can be extended, to address parallel goals in in vivo animal studies, by evaluating optimal formulation strategies to improve drug absorption, as well as predicting impact of food on oral drug product performance early in the product lifecycle