Fluoxetine- and norfluoxetine-mediated complex drug-drug interactions: in vitro to in vivo correlation of effects on CYP2D6, CYP2C19, and CYP3A4

Abstract

Fluoxetine and its circulating metabolite norfluoxetine comprise a complex multiple-inhibitor system that causes reversible or time-dependent inhibition of the cytochrome P450 (CYP) family members CYP2D6, CYP3A4, and CYP2C19 in vitro. Although significant inhibition of all three enzymes in vivo was predicted, the areas under the concentration-time curve (AUCs) for midazolam and lovastatin were unaffected by 2-week dosing of fluoxetine, whereas the AUCs of dextromethorphan and omeprazole were increased by 27- and 7.1-fold, respectively. This observed discrepancy between in vitro risk assessment and in vivo drug-drug interaction (DDI) profile was rationalized by time-varying dynamic pharmacokinetic models that incorporated circulating concentrations of fluoxetine and norfluoxetine enantiomers, mutual inhibitor-inhibitor interactions, and CYP3A4 induction. The dynamic models predicted all DDIs with less than twofold error. This study demonstrates that complex DDIs that involve multiple mechanisms, pathways, and inhibitors with their metabolites can be predicted and rationalized via characterization of all the inhibitory species in vitro.

Figure 1

The effect of fluoxetine administration on dextromethorphan and omeprazole pharmacokinetics. The mean (with standard deviation) plasma concentration versus time curves for dextromethorphan (A), dextrorphan (B) omeprazole (D), and 5-hydroxyomeprazole (E) in the presence (circles) and absence (triangles) of fluoxetine (n=10) are shown with the effect of fluoxetine on the AUC_0-∞ of dextromethorphan and omeprazole in each individual subject shown in panels C and F.

Figure 2

Disposition of caffeine (A and D), midazolam (B and E) and lovastatin (C and F) in the presence and absence of fluoxetine administration. Mean and standard deviation (n=10) plasma concentration versus time curves are displayed in the presence (circles) and absence (triangles) of fluoxetine. AUC_0-∞ changes are shown for individual subjects.

Figure 3

Simulated and observed concentration profiles of dextromethorphan (A and D), omeprazole (B and E) and midazolam (C and F) on day 12 of the study following fluoxetine administration (right panel) and in the control day (left panel). Observed mean and standard deviation plasma concentration versus time curves are shown as circles and simulated curves are shown as lines. The grey lines represent the 95 and 5% confidence intervals of the simulated data in 100 subjects. In panels E and F the red lines represent the simulated concentration versus time curve of omeprazole and midazolam in the absence of CYP3A4 inactivation, respectively. The control day observed C_max values were 8.7 ± 11.4 nM, 660 ±380 nM, and 17±16nM for dextromethorphan, omeprazole and midazolam, respectively. The corresponding predicted control day C_max values were 6.2±4nM, 340±230nM, and 15±15nM. The observed day 12 C_max values were 74±31nM, 2500 ± 1400 nM and 14.5 ± 10 nM for dextromethorphan, omeprazole and midazolam, respectively. The corresponding predicted day 12 C_max values were 40±25nM, 1700±940 nM and 58 ± 62nM.

Figure 4

Induction of CYP3A4 by fluoxetine and norfluoxetine enantiomers. Concentration dependent effects of fluoxetine and norfluoxetine on CYP3A4 mRNA (A) and activity (B) are shown for three donors. Rifampicin was used as the positive control for CYP3A4 induction. The mRNA induction parameters obtained were I_max of 2.8 fold and EC₅₀ of 3.5µM for (S)-fluoxetine and I_max of 2.6 fold and EC₅₀ of 3.9 µM for (S)-norfluoxetine. For (R)-fluoxetine and (R)-norfluoxetine toxicity to the hepatocytes prevented treatments at concentrations that would be high enough to show saturation of induction and hence the induction slope was determined. The slopes were 0.3 µM⁻¹ for (R)-fluoxetine and 0.8 µM⁻¹ for (R)-norfluoxetine respectively.

Figure 5

Simulated and observed concentration profiles following fluoxetine administration when CYP3A4 induction is incorporated. Mean and standard deviation plasma concentration versus time curves along with the simulated curves are displayed for omeprazole (A) and midazolam (B) Black circles represent observed concentrations, the solid black line represents the simulated concentration versus time curve resulting from the incorporation of CYP3A4 induction. Grey lines represent the 95 and 5% confidence intervals of the simulated curve. The observed C_max values for omeprazole and midazolam were 2500 ± 1400 nM and 15 ± 15 nM, respectively. The corresponding predicted C_max values were 980 ± 550 nM and 10 ± 11 nM.