Whole Body Physiologically Based Pharmacokinetic Model to Explain A Patient With Drug–Drug Interaction Between Voriconazole and Flucloxacillin

Background and Objectives

Voriconazole administered concomitantly with flucloxacillin may result in subtherapeutic plasma concentrations as shown in a patient with Staphylococcus aureus sepsis and a probable pulmonary aspergillosis. After switching our patient to posaconazole, therapeutic concentrations were reached. The aim of this study was to first test our hypothesis that flucloxacillin competes with voriconazole not posaconazole for binding to albumin ex vivo, leading to lower total concentrations in plasma.

Methods

A physiologically based pharmacokinetic (PBPK) model was then applied to predict the mechanism of action of the drug–drug interaction (DDI). The model included non-linear hepatic metabolism and the effect of a severe infectious disease on cytochrome P450 (CYP) enzymes activity.

Results

The unbound voriconazole concentration remained unchanged in plasma after adding flucloxacillin, thereby rejecting our hypothesis of albumin-binding site competition. The PBPK model was able to adequately predict the plasma concentration of both voriconazole and posaconazole over time in healthy volunteers. Upregulation of CYP3A4, CYP2C9, and CYP2C19 through the pregnane X receptor (PXR) gene by flucloxacillin resulted in decreased voriconazole plasma concentrations, reflecting the DDI observations in our patient. Posaconazole metabolism was not affected, or was only limitedly affected, by the changes through the PXR gene, which agrees with the observed plasma concentrations within the target range in our patient.

Conclusions

Ex vivo experiments reported that the unbound voriconazole plasma concentration remained unchanged after adding flucloxacillin. The PBPK model describes the potential mechanism driving the drug–drug and drug–disease interaction of voriconazole and flucloxacillin, highlighting the large substantial influence of flucloxacillin on the PXR gene and the influence of infection on voriconazole plasma concentrations, and suggests a more limited effect on other triazoles.