Physiologically based pharmacokinetic modeling of the unusual kinetics of ketoconazole due to nonlinear liver binding and metabolism in rats.

Ketoconazole (KTZ) is a well known inhibitor of intestinal and liver CYP3A4/5, and its kinetics is being re-evaluated in rats. In vitro binding and microsomal incubation studies confirmed saturable liver binding and metabolism and linear plasma binding of KTZ. These binding constants led to a decreasing trend of the liver tissue to blood partitioning ratio with increasing KTZ blood/plasma concentration. KTZ, assayed by liquid chromatography-tandem mass spectrometry, from single intravenous (2.5, 5, and 10 mg/kg) and oral (10, 20, and 40 mg/kg in 1% methylcellulose via gavage) doses displayed primarily nonlinear liver but negligible intestinal metabolism and low renal and biliary clearances. Improved fits to the KTZ blood and liver concentrations in physiologically based pharmacokinetic modeling were obtained upon upgrading the in vitro Km,L and Vmax,L for liver metabolism and the plasma binding constant, ѱ (ratio of plasma bound/unbound concentrations). However, minor improvement resulted upon fitting parameters for the intestinal metabolic intrinsic clearance and liver binding capacity and dissociation constant. The absorption rate constant, ka, obtained from the combined intravenous and oral fits and sequential fitting of the intravenous then with oral data, was similar. The optimized model, which described KTZ nonlinear liver binding and metabolism, predicted the intravenous and oral data (area under the curve) well. Simulations were further conducted under nonlinear and linear conditions based on the segregated intestinal flow model (with 10% flow to enterocyte and 90% to serosal regions) and the traditional model to illustrate the roles of intestine versus liver controlling the extent of first-pass metabolism of drugs. SIGNIFICANCE STATEMENT: The complexity of nonlinear liver tissue binding, metabolism, and solubility issues toward ketoconazole was demonstrated in stepwise fitting of blood and liver data in physiologically based pharmacokinetic modeling to remove bias and errors. The optimized model, predicated on predominanat liver and minor intestinal metabolism, predicted the blood and liver ketoconazole area under the curve well. Simulations suggest a greater presystemic fractional intestinal metabolism for drugs given orally according to the segregated intestinal flow model (10% to enterocyte region) over the traditional model.