This thesis investigated the usage and limitations of a cassette approach for determining the permeability coefficients for a large number of active pharmaceutical ingredients with Caco-2 and Rat intestine permeability models. We conducted our research using 66 active substances from all 4 Biopharmaceutical Classification System classes at pre-determined donor concentration, and their 1/2 and 1/5 dilution. Active substances were previously chosen based on their compatibility in complex donor solutions and usefulness in acceptor sample analysis.
We investigated the results of a cell model and the rat intestine model. We also compared cassette and individually determined permeability coefficients on a similar cell model. The measured permeability coefficients and efflux ratios were presented in correlation with the fraction absorbed. Half concentration was shown to be significantly less toxic to the cell line during the experiment, while at the same time the obtained results were almost the same as with the full concentration. Lowering the concentration to 1/5 did not prove to be more beneficial regarding the toxicity, but several more problems with the detection of certain compounds in the acceptor samples were encountered. As expected, the Receiver Operating Characteristic curve has shown that the individual permeability measurement approach on the Caco-2 cell model has a better ability to separate and classify between categories. Besides that, the cassette approach of the Caco-2 model is superior for classification compared to the diffusion cell model using rat intestine. We successfully used the cassette approach to determine a large number of permeability coefficients in a single experiment and showed that further use will require a selection based on our results to a narrower set of compounds that are appropriately classified by different experimental models for determining permeability.
|