Karl Box began his presentation by expressing gratitude to the organisers and introducing the topic of in vitro dissolution absorption tools, focusing on using flux as a predictor of oral absorption. He provided an overview of the in vitro tools available, followed by case studies and a discussion on predicting in vivo human fraction absorption based on in vitro flux profiles. He also examined the evolution of these models, their effectiveness, and potential improvements.
Box highlighted that Pion was the first company to commercialise the PAMPA technique, a high-throughput method for ranking the permeability of compounds. This technique involved measuring the permeation of dissolved drugs across a phospholipid-plated membrane. He explained that drug absorption is a composite process involving both solubilisation and permeability, and Pion developed various tools to assist in drug development, from low-volume assays to larger-scale assays for final formulation development.
He detailed the apparatus used, including the rainbow spectrometer, which allowed for direct in situ concentration analysis. Box described the optimisation of artificial membranes using a mixture of phospholipid components and the importance of maintaining sink conditions in the receiver side of the apparatus. He defined flux as a measure of transport and explained how it could be calculated from the concentration in the receiver compartment.
Box presented case studies, including the use of the microflux apparatus with carbamazepine, demonstrating the spring and parachute effect of polymers in preventing precipitation. He discussed the importance of measuring flux to understand the composite effect of solubility and permeation on drug performance. He introduced the Pion Predictor software, developed in collaboration with Kiyohiko Sugano, to predict the fraction of drug absorbed using flux data.
Box concluded by discussing the evolution of the models used to predict human fraction absorbed, highlighting improvements made to address issues such as over-prediction and the particle drifting effect. He emphasised the need for more data to test the robustness of the models and suggested combining flux data with in vivo parameters to simulate plasma concentration-time profiles.
