Predicting the exposure and antibacterial activity of fluoroquinolones based on physicochemical properties.

The purposes of this study are to evaluate if the PAMPA (Parallel Artificial Membrane Permeability Assay) permeability and the true partition coefficient could be useful for predicting AUC and MIC data of a group of antibacterial fluoroquinolones (FQs). The protonation macro- and microconstants, the n-octanol/water partition coefficients at isoelectric pHs, and the PAMPA permeability of 11 selected FQs were determined, and used to calculate the true partition coefficient, the interactivity parameter between the acidic and basic group, and the apparent intrinsic permeability. It has been shown that the apparent intrinsic permeability correlates well with the AUC in human, whereas the true partition coefficient and the interactivity parameter correlate with 1/MIC values on two Gram-positive bacteria, namely Streptococcus pneumonia and Staphylococcus aureus (methicillin-susceptible). The AUC/MIC ratios predicted from these correlations have shown to be in good agreement with the literature values. It is envisaged that the models described in this study could be useful in the development of new FQs by enabling an early prediction of AUC/MIC ratios based on physicochemical properties.