Modified nonsink equation for permeability estimation in cell monolayers: comparison with standard methods.

Cell culture permeability experiments are valuable tools in drug development and candidate selection, but the monolayer preparation protocols and the calculations procedures can affect the permeability estimation. Hence, standardization and method suitability demonstration are necessary steps for using permeability data for regulatory and in vivo prediction purposes. Much attention is usually paid to experimental procedure validation and less to the mathematical analysis of the results although the standard equations used imply several assumptions that many times do not hold. The aim of this study was to use a simulation strategy to explore the performance of a new proposed modified nonsink equation (MNS) for unidirectional apparent permeability estimation in different types of profiles (of cumulative drug amounts versus time) including those in which the initial permeation rate is altered, considering several levels of experimental variability. The second objective was to compare the MNS method with the classical sink and nonsink approaches and finally to explore its usefulness for BCS classification. Real data from permeability experiments representing atypical profiles have been used for fitting with the three approaches, MNS, sink, and nonsink equations, in order to validate the performance of the new proposed model. The results demonstrated that the MNS method is a precise and accurate equation for calculating the apparent unidirectional permeability in any type of profile and different scenarios of variability, in any sink and nonsink conditions, while the standard nonsink equation fails in obtaining good permeability estimations in those situations in which the initial permeation rate is altered. Linear regression models (S and SC) are not valid under nonsink conditions, as expected, as the underlying assumptions (sink conditions) do not hold, but also in situations in which sink conditions are fulfilled but the system variability is high.