Development of a coupled in vitro lipid digestion – in vivo absorption model to evaluate oral drug absorption from lipid-based formulations

2017-02-27T00:03:03Z (GMT) by Crum, Matthew Frederick
Good absorption after oral administration is a requirement for most commercially successful drugs. For many drug candidates, however, low solubility in the gastrointestinal (GI) fluids limits drug absorption and is therefore a potential limitation to successful development. Formulation of poorly water-soluble drugs (PWSD) with lipids to generate lipid-based formulations (LBF) provides one means of potentially overcoming inadequate absorption due to low water solubility, but the mechanisms by which LBFs enhance drug absorption remain incompletely understood. This thesis explored the mechanisms of drug absorption from LBFs using the PWSD fenofibrate as a model compound. Particular focus was directed to the development of improved screening tests that may be used to scrutinise the mechanisms of drug absorption from LBFs. Specifically the work described in this thesis sought to modify traditional in vitro lipid digestion tests via the addition of in-line absorption/permeability assessment, a model analogous to the dissolution-permeability tests previously described for traditional formulations. This was achieved through coupling the in vitro model of lipid digestion directly to a single pass in situ intestinal perfusion in an anaesthetised rat. This allowed for real time observations of formulation dispersion, formulation digestion and drug absorption to occur. The advantage of this approach is that it provides a direct read-out of the impact of rapid changes to solubilisation (and supersaturation) on drug flux, using an absorption sink that is resistant to bile salt and has appropriately high surface area. Using this new model, the generation of supersaturation during LBF dispersion and digestion was found to be an effective driver for drug absorption, and for the highly permeable drug fenofibrate, where only fleeting supersaturation was required to facilitate robust absorption, at least over short time scales. Over longer time periods and when assessed using longer perfused intestinal segments, however, maintenance of drug solubilisation and supersaturation throughout the absorption period became more important. Consistent with the poor dissolution of PWSDs, the presence of precipitated or suspended fenofibrate drug from a LBF had little impact on absorptive drug flux. Subsequent studies sought to compare the in vitro digestion – in situ absorption data with data obtained after oral administration in rats. Interestingly, no differences in fenofibrate exposure were evident after oral administration of the three different LBFs. This was consistent with previous published data for fenofibrate absorption, and also with the data obtained with the in vitro digestion – in vivo absorption model at early time points, but inconsistent with the data obtained from the in vitro digestion – in vivo absorption model employing longer perfused intestinal segments over longer time periods. These findings stimulated a more detailed evaluation of the enzymatic and hydrodynamic conditions employed during the in vitro LBF digestion experiment (to clarify whether the conditions employed in the in vitro digestion – in vivo absorption model reflected those in vivo). Two major differences were investigated. First in vitro studies were conducted using lower lipase enzyme levels that were more consistent with lipase levels in the rat (rather than the higher levels found in the dog and human and previously used in in vitro lipid digestion tests). Replication of the in vitro digestion – in vivo absorption model under rat relevant digestion, however, reproduced the same in vitro solubilisation and in situ absorption trends. In contrast, studies conducted using lower stirring rates, in an attempt to better replicate in vivo hydrodynamics, revealed reduced precipitation rates and enhanced solubilisation and in situ absorption data that were more closely aligned with the in vivo bioavailability data. The data suggests that rapid stirring in in vitro lipid digestion tests, may overestimate drug precipitation during in vivo dispersion and digestion of LBFs. In summary, the work presented in this thesis demonstrates the feasibility of using a coupled in vitro lipid digestion – single pass in situ rat jejunum intestinal perfusion model to provide a mechanistic understanding of drug absorption from digesting LBFs in vivo. The studies confirm that LBFs provide a robust means to enhance absorption for PWSDs, that the generation of supersaturation during intestinal processing of LBFs is a significant driver of absorption, and that the stirring rates employed within in vitro digestion models significantly affect estimates of formulation performance. The latter suggests the potential benefit of future studies to better understand the relationship between in vitro and in vivo hydrodynamics on drug precipitation and absorption from LBFs.