Insights into the biopharmaceutical behavior of oral lipid-based formulations using advanced analytical techniques and mathematical modeling

Abstract

Lipid-based formulations (LBFs) are effective means for the oral delivery of poorly water-soluble compounds. The drug is already solubilized in the formulation and, thus, the critical dissolution step is circumvented. However, the oral bioavailability is also determined by the fate of the LBF in the gastrointestinal (GI) tract. Formulation dispersion and lipid digestion are particularly critical steps in this regard. The ability to maintain the drug in a solubilized state may be reduced leading to an increased risk of drug precipitation and erratic drug absorption. The present thesis consists of five studies, which aim at improving the biopharmaceutical understanding of LBF performance in the GI tract. To this end, in vitro dispersion and digestion assays are employed along with advanced analytical techniques and mathematical modeling. The findings may improve the predictability of LBF performance upon oral administration. In the first study, we analyzed surfactant/co-solvent systems during aqueous dilution. A theoretical model was proposed to analyze the role of excipient interaction for drug solubilization during dilution. This model indicated that, in undiluted formulations, co-solvent/surfactant domains were responsible for drug solubilization. In contrast, in diluted formulations the co-solvent partitioned out of the surfactant microstructure. This loss of excipient interaction caused formulation-specific supersaturation, which was indicative for the risk of drug precipitation. The analysis of excipient interactions and drug supersaturation facilitated the identification of critical drug-loadings in LBFs that are prone to drug precipitation. The second study focused on the in vitro lipolysis test. We evaluated Raman spectroscopy as an analytical technique for real-time monitoring of lipolysis-triggered drug precipitation. Despite the complex and varying medium composition, in-line analytics provided robust and highly time-resolved drug precipitation profiles. This allowed further analysis of the precipitation kinetics using a theoretical nucleation and growth model. The combination of real-time Raman spectroscopy and mathematical modeling provided valuable insights into the time evolution of lipolysis-triggered drug crystallization. The simulation of formulation digestion in an absorption environment was the purpose of the third study. Current in vitro lipolysis tests are performed in a single compartment and, therefore, they include no absorption sink. In this study, we developed a physiologically based model of formulation digestion in the GI tract based on in vitro lipolysis and in vivo pharmacokinetic data. The resulting system of differential equations allowed the calculation of drug supersaturation during the intestinal transit of LBFs. This approach provided clear evidence that an absorption sink significantly lowers the risk for lipolysis-triggered drug precipitation. Hence, in vitro lipolysis provides the worst-case prediction of LBF performance. Moreover, our results suggested that the intestinal digestion of LBFs is less critical than expected with respect to drug precipitation, especially with highly permeable drugs. The fourth study focused on the solid-state of precipitated weakly basic drugs. Drug-loaded LBFs were dispersed in a simulated intestinal medium with and without digestive enzymes and the resulting precipitate was analyzed by X-ray diffraction and re-dissolution. The study revealed that in vitro conditions can influence the solid-state properties of precipitating weak bases. While a crystalline precipitate was observed upon dispersion, the presence of digestive enzymes led to an amorphous precipitate. These findings are of high practical importance for the prediction of LBF performance in vivo. In contrast to the crystalline form, an amorphous precipitate may re-dissolve rapidly and, hence, become again available for absorption. Finally, in the fifth study, Raman spectroscopy and ultrasound resonator technology were evaluated as process analytical tools for drug quantification in LBFs. This study evidenced the excellence of Raman spectroscopy for drug quantification in complex lipidic matrices and was the basis for using Raman spectroscopy with biopharmaceutical tests. This thesis provided novel insights into the biopharmaceutical behavior of LBFs in the GI tract. The establishment of real-time techniques allowed the examination of highly dynamic formulation changes during dispersion and digestion. Moreover, mathematical modeling provided key insights into biopharmaceutical processes that are hardly accessible using in vitro methods. These advancements may improve the ability to predict LBF performance in vivo.