Hydroxychloroquine is metabolized by CYP2D6, CYP3A4 and CYP2C8, and inhibits CYP2D6, while its metabolites also inhibit CYP3A4 in vitro

Introduction: Hydroxychloroquine is used for the treatment of malaria,
rheumatoid arthritis and lupus erythematosus. In 2020, hydroxychloroquine
was also repurposed for the treatment of COVID-19. Although
current evidence does not encourage the use of hydroxychloroquine to
treat COVID-19, its therapeutic and prophylactic use against COVID-19
is still investigated in clinical trials. Despite being in clinical use for more
than 60 years, its clinical pharmacology is not well understood.
Hydroxychloroquine is metabolized into three active metabolites, but
the key metabolizing enzymes have not been unambiguously identified.
Moreover, little is known about the inhibitory effects of hydroxychloroquine
on cytochrome P450 (CYP) enzymes.
Objectives: This study aimed to investigate the CYP metabolic and inhibitory
profile of hydroxychloroquine and its three metabolites in vitro.
Methods: Hydroxychloroquine metabolism was studied in human liver
microsomes (HLM) and recombinant CYP enzymes using substrate
depletion and CYP-selective inhibitors. The inhibitory effects
of hydroxychloroquine and its metabolites on nine CYP
enzymes were also determined in HLM, using automated probe
substrate cocktail assays.
Results: Based on screening experiments, CYP3A4, CYP2D6 and
CYP2C8 were the key enzymes involved in hydroxychloroquine metabolism
in vitro. Although the intrinsic clearance (CLint) value of hydroxychloroquine
depletion by recombinant CYP2D6 (0.87 μl/min/pmol) was
more than 10-fold higher than that by CYP3A4 (0.075 μl/min/pmol),
scaling of the recombinant data to HLM level resulted in similar CLint
values for CYP2D6 and CYP3A4 (11 and 14 μl/min/mg) because of the
much greater abundancy of CYP3A4 than that of CYP2D6. The scaled
HLM CLint of CYP2C8 was 5.7 μl/min/mg. Data in HLM with CYPselective
inhibitors also suggested relatively equal roles for CYP2D6 and
CYP3A4 in hydroxychloroquine metabolism, and a smaller contribution
for CYP2C8. In CYP inhibition experiments, hydroxychloroquine and its
three metabolites were direct CYP2D6 inhibitors (50% inhibitory concentration
IC50 18-135 μM), while all metabolites were CYP3A timedependent
inhibitors (IC50 12-117 μM, IC50 shift 2.2-3.4-fold).
CYP2D6 inhibition explains the reported clinical drug-drug interaction
between hydroxychloroquine and the CYP2D6 substrate metoprolol. The
present data, together with the inhibitors’ estimated intracellular
hepatocyte concentrations, were successfully used in a static model
to predict the fold increase in metoprolol AUC (predicted: 2.3-
2.8-fold, observed: 1.65-fold).
Conclusion: The present study unambiguously demonstrates that hydroxychloroquine
is metabolized mainly by CYP2D6, CYP3A4 and
CYP2C8 in vitro. Moreover, hydroxychloroquine and its three metabolites
are CYP2D6 reversible inhibitors, and hydroxychloroquine metabolites
are CYP3A time-dependent inhibitors. The current data can be used
in static and physiologically-based pharmacokinetic models to predict
hydroxychloroquine drug-drug interaction potential, as shown with the successful prediction of hydroxychloroquine - metoprolol drug-drug interaction.