Physiologically based pharmacokinetic modeling of fluconazole: interpretation of gender-dependent pharmacokinetics and cytochrome P450 inhibition.

The accurate prediction of clinical drug-drug interactions (DDIs) in modeling analysis relies on the fraction of metabolism and transport of the substrate drugs and on the inhibitory potency and plasma exposure of index inhibitors. Fluconazole is frequently used as a clinical index inhibitor of cytochrome P450 (CYP) 2C9 and CYP3A4 to assess DDI liabilities involving these CYP enzymes. It also inhibits uridine diphosphate glucuronosyltransferase (UGT) 2B7 and exhibits gender-dependent pharmacokinetics (PK). The currently available model profile of fluconazole does not account for these population-specific PK variations. Therefore, extrapolating from one fluconazole DDI dataset to another remains questionable when it is used as an index inhibitor. In the present study, a physiologically based pharmacokinetic (PBPK) model of fluconazole was developed using PK data from male and female subjects. A minimal PBPK model with advanced dissolution, absorption, and metabolism (ADAM) for fluconazole was built in the Simcyp Simulator (V22). The base model from the Simcyp compound library captured fluconazole plasma profiles following oral administration only in male subjects, but not in female subjects. The volume of distribution at steady state (Vss) of fluconazole is close to the total body water volume, which can contribute to differences in plasma profiles between male and female subjects. Therefore, Vss in female subjects was optimized by curve fitting. The total clearance (CL) value in female subjects was also adjusted to reflect the similar fluconazole elimination half-life (t1/2) between male and female subjects. The plasma profiles of fluconazole in the general population were captured well when the predicted mean Vss and total CL were calibrated by the proportion of female subjects. The developed model was further validated by simulating reported clinical DDI studies with fluconazole. Using the refined PBPK model, the inhibitory constants (Ki) for fluconazole against CYP2C9 and UGT2B7 were optimized. The final fluconazole model captured the precipitant DDIs with midazolam (CYP3A probe substrate), tolbutamide (CYP2C9 probe substrate), omeprazole (CYP2C19 probe substrate), and zidovudine (UGT2B7 probe substrate), indicating that the refined fluconazole PBPK model in the present study can accurately predict clinical DDIs across studies in different populations with diverse gender mixes.