In-vitro permeability screening of melt extrudate formulations containing poorly water-soluble drug compounds using the phospholipid vesicle-based barrier Johanna Kanzer a,b , Ingunn Tho a , Gøril Eide Flaten a , Markus Mägerlein c , Peter Hölig c , Gert Fricker b and Martin Brandl a,d a Drug Transport and Delivery Research Group, Department of Pharmacy, University of Tromsø, Tromsø, Norway, b Department of Pharmaceutical Technology and Biopharmacy, Institute of Pharmacy and Molecular Biotechnology, University of Heidelberg, Heidelberg, Germany, c SOLIQS, Abbott GmbH and Co. KG, Ludwigshafen, Germany, and d Department of Physics and Chemistry, University of Southern Denmark, Odense M, Denmark Abstract Objectives The phospholipid vesicle-based barrier has recently been introduced as an in-vitro permeation model mimicking gastro-epithelial barriers in terms of passive diffusion of drugs. The aim of this study was to investigate whether the phospholipid vesicle-based barrier was suitable for permeability screening of complex formulations such as solid dispersions. Methods Solid dispersions containing the poorly water-soluble drugs HIV-PI 1 (log P = 6.2, molar mass = 628.80 g/mol) and HIV-PI 2 (log P = 5.3, molar mass = 720.95 g/ mol), a hydrophilic polymer and different surfactants were tested with respect to their influence on integrity of the barrier in terms of electrical resistance and permeability for calcein. Furthermore, utilisation of a more biologically relevant medium, Hank’s balanced salt solution supplemented with Mg 2+ - and Ca 2+ -ions (HBSS (Mg 2+ , Ca 2+ )), has been tested. Key findings Except for the polyoxyl 40 hydrogenated castor oil-containing solid disper- sion, no influence on the phospholipid vesicle-based barrier could be observed from the tested samples. Presence of active pharmaceutical ingredients (APIs) in the solid dispersions led to the same results as the corresponding placebo results. First experiments analysing the passive diffusion of both APIs in HBSS (Mg 2+ , Ca 2+ ), evaluated as suitable transport medium, have shown promising results regarding the suitability of the phospholipid vesicle- based barrier for investigation of solid dispersions. Conclusions The study indicated that the phospholipid vesicle-based barrier was compat- ible with selected melt extrudate formulations. The model seemed capable to reveal different transport routes in comparison with Caco-2 cell permeability tests. Keywords artificial membrane; melt extrudate; permeability; poorly water-soluble drug; surfactant Introduction The oral route is the preferred route of drug administration due to its cost efficiency, convenience and patient compliance. In recent years, in-vitro permeability screening has become an integral part of drug discovery regimes to pre-estimate oral absorption potential of new drug compounds. Cell culture based models such as the Caco-2 cells, originating from a human colon adenocarcinoma, and the Mardin-Darby canine kidney (MDCK) cell line are commonly used for investigation of passive and active transport properties of drug compounds. [1–3] Furthermore, noncellular models based on lipids, such as the parallel arti- ficial membrane permeability assay (PAMPA) and immobilised artificial membranes (IAM), have successfully been proven to assess passive absorption of active pharmaceutical ingre- dients (APIs). [4,5] Recently, a novel model, the phospholipid vesicle-based barrier, has been introduced. It has been shown to deliver a good correlation with in-vivo data for passive diffusion for a diverse test set of model drugs. [6] The model is based on a tight barrier of phospholipid vesicles, deposited on a filter support by centrifugation and partly fused by freeze–thaw Research Paper JPP 2010, 62: 1591–1598 © 2010 The Authors JPP © 2010 Royal Pharmaceutical Society of Great Britain Received February 11, 2010 Accepted July 2, 2010 DOI 10.1111/j.2042-7158.2010.01172.x ISSN 0022-3573 Correspondence: Martin Brandl, Department of Physics and Chemistry, University of Southern Denmark, Campusvej 55, DK-5230 Odense M, Denmark. E-mail: mmb@ifk.sdu.dk 1591