Mechanistic Understanding of Enhanced Human Oral Bioavailability of Fenofibric Acid from Novel Lipid Carriers Using Semi- Physiologically Based Pharmacokinetic Model and Various Analytical Approaches Including Biorelevant Dissolution Testing (English shop)

Preview

Extract, PDF (230 KB)
Table of Contents, PDF (21 KB)

The subject of this dissertation was to determine the improvement in the peroral bioavailability of fenofibrate (FFB) achieved by lipid-based formulation (LBF). A further task was to apply various analytical methods for assessing the improvement in the oral bioavailability of fenofibrate. These included in vitro biorelevant solubility, dispersion, dissolution and precipitation tests. On the basis of the analytical results, PBPK models were then used to simulate human plasma profiles following administration of the FFB formulations. The resulting in silico predictions were in agreement with the in vivo observations. By applying parameter sensitivity analysis, it was furthermore possible to achieve a mechanistic understanding of the rate-limiting steps involved.

Lipid-based formulations can be categorised according to the Pouton classification system. Type I formulations consist exclusively of oils, whereas at the other end of the scale the Type IV formulation consists largely of surfactants. In this work, Type IIIA and Type IIIB lipid formulations were primarily investigated.

It was shown that dispersion tests on FFB lipid formulations are best carried out using the USP 3 apparatus, since in vivo GI motility is best reflected in this apparatus. In order to compare the hydrodynamics in different dissolution apparatuses, the dissolution experiment on LBF no. 1 – no. 4 of FFB was also carried out using USP 2. Irrespective of compendial or biorelevant media, most of these lipid formulations led to the dissolution of the majority of the loaded drug, in contrast to the unformulated fenofibrate, which barely dissolves in the fasted state. Furthermore, the transfer model experiments on the lipid formulations of FFB showed that intestinal precipitation following gastric dissolution is unlikely.

Through mathematical transformation of the Noyes-Whitney equation, an Excel toolkit can be used to calculate the z-value from in vitro dissolution profiles. The z-values are then implemented in the physiologically based pharmacokinetic in silico models STELLA® and Simcyp®. On the basis of the required post-absorptive parameters, these models can be used to predict the plasma drug concentration following oral administration of various formulations. In addition, the Simcyp® simulator enables a series of virtual trials to determine the PK variability of the drug substance in different populations. In order to test these possibilities for LBF of fenofibrate, LBF no. 4 was modelled. The simulation result from Simcyp® corresponded to that from the STELLA® software. Furthermore, the plasma fenofibric acid concentration profiles were accurately predicted by the models. The point estimates for Cmax and AUC, calculated from the in silico and in vivo plasma profiles, even lay within the range of 0.8–1.25 for the SMEDDS solution and capsule formulations. This agreement of in vitro–in silico with in vivo was further supported by calculation of the respective f2 factors.

Based on these results, it appears that the in vitro–in silico–in vivo approach represents a useful tool for identifying and comparing limitations to oral absorption for lipid-based formulations and for optimising lipid formulation development of poorly soluble drugs.