A novel combined lipolysis-permeation screening approach in 96-well format: Method development using type I lipid-based formulations

For release testing of lipid-based formulations (LBFs), lipid digestion using the pH-stat approach is widely used despite the fact that it is a laborious exercise and predictivity towards in vivo performance could not be demonstrated. A probable reason is the lack of differentiation between readily absorbable (molecularly dissolved) and less absorbable (colloid associated) drug fractions. This work describes the development and testing of an alternative approach designed to address both issues of the pH-stat method, the labor-intensive nature and the limited ability to estimate in vivo behavior. The proposed solution involves combined lipolysis-permeation testing on 96-well microtiter sandwich plates.

In this new time-efficient and material-sparing approach, a highly buffered lipolysis medium was used in the donor chamber to avoid the need for pH stabilization by titration. Moreover, the method was optimized to minimize non-specific adsorption of cinnarizine to the plates and polytetrafluoroethylene (PTFE)-coated stirring bars.

The predictive power of the new high throughput screening (HTS) to estimate in vivo oral bioavailability of cinnarizine-loaded type I LBFs was evaluated against recent oral bioavailability data of the very same formulations in rats and compared to in vitro data generated by the pH-stat lipolysis approach. The following variables were studied: supersaturation, lipase inhibition, lipid chain length, and presence of an amphiphilic polymer (precipitation inhibitor).

While the HTS method correctly captured the in vivo impact of both supersaturation and lipase inhibition for all formulations, the pH-stat method revealed opposite trends for one out of four combinations in each formulation sets. In vivo there had been no effect observed for lipid chain length nor presence of the amphiphilic polymer. In contrast, both in vitro approaches wrongly predicted such effects in some cases. A better prediction for the long-chain systems was found with the HTS method as with the laborious pH-stat approach. The HTS lipolysis-permeation method can test multiple formulations in the 96-well plate format within hours and gave IVIVCs of up to 0.91 for grouped type I LBFs. In particular the in vivo performance of supersaturated formulations was correctly captured. This study demonstrates that this new method represents a promising alternative to existing tools for prediction of the in vivo performance of type I LBFs.