Prediction of time-dependent CYP3A4 drug–drug interactions by physiologically based pharmacokinetic modelling: Impact of inactivation parameters and enzyme turnover

https://doi.org/10.1016/j.ejps.2011.04.008Get rights and content

Abstract

Predicting the magnitude of time-dependent metabolic drug–drug (mDDIs) interactions involving cytochrome P-450 3A4 (CYP3A4) from in vitro data requires accurate knowledge of the inactivation parameters of the inhibitor (KI, kinact) and of the turnover of the enzyme (kdeg) in both the gut and the liver. We have predicted the magnitude of mDDIs observed in 29 in vivo studies involving six CYP3A4 probe substrates and five mechanism based inhibitors of CYP3A4 of variable potency (azithromycin, clarithromycin, diltiazem, erythromycin and verapamil). Inactivation parameters determined anew in a single laboratory under standardised conditions together with data from substrate and inhibitor files within the Simcyp Simulator (Version 9.3) were used to determine a value of the hepatic kdeg (0.0193 or 0.0077 h−1) most appropriate for the prediction of mDDIs involving time-dependent inhibition of CYP3A4. The higher value resulted in decreased bias (geometric mean fold error – 1.05 versus 1.30) and increased precision (root mean squared error – 1.29 versus 2.30) of predictions of mean ratios of AUC in the absence and presence of inhibitor. Depending on the kdeg value used (0.0193 versus 0.0077 h−1), predicted mean ratios of AUC were within 2-fold of the observed values for all (100%) and 27 (93%) of the 29 studies, respectively and within 1.5-fold for 24 (83%) and 17 (59%) of the 29 studies, respectively. Comprehensive PBPK models were applied for accurate assessment of the potential for mDDIs involving time-dependent inhibition of CYP3A4 using a hepatic kdeg value of 0.0193 h−1 in conjunction with inactivation parameters determined by the conventional experimental approach.

Introduction

Many clinically relevant drug–drug interactions (DDIs) involving cytochrome P450 3A4 (CYP3A4), the most abundant P450 in the human liver and intestine, are due to time-dependent inhibition consequent to metabolism of the inhibitor to a reactive intermediate that binds irreversibly or quasi-irreversibly to the enzyme (Silverman, 1995, Venkatakrishnan and Obach, 2007). These DDIs are persistent since they require synthesis of new enzyme for recovery. Examples of the so-called ‘mechanism-based’ inhibitors of CYP3A4 include anti-retroviral agents such as ritonavir and saquinavir, the macrolides erythromycin, clarithromycin and azithromycin and the calcium channel blockers verapamil, diltiazem and mibefradil (Posicor®) (FDA, 2006). The withdrawal of the potent mechanism-based inhibitor mibefradil from the market a year after its introduction in 1997, as a consequence of reports of serious DDIs with CYP3A4 substrates, illustrates the importance of assessing the potential impact of mechanism based inhibitors in drug development (Huang and Lesko, 2004). Prediction of the magnitude of DDIs mediated by mechanism-based inhibition is now of significant interest, both academically and in an industrial context (Grimm et al., 2009), and there has been an evaluation of early decision making based on different in vitro methods and prediction methods (Grime et al., 2009).

Prediction of the magnitude of CYP3A4-based DDIs mediated by mechanism-based inhibition requires values of substrate-dependent, inhibitor-dependent and intrinsic parameters (Venkatakrishnan and Obach, 2007, Yang et al., 2007a) Substrate parameters include the fractions of the drug metabolised by the inhibited pathway (fm) in the gut wall and liver. Time-dependent inhibition is characterized by the maximal inactivation rate constant (kinact) and the concentration of inhibitor that produces half-maximal inactivation (KI). The significant impact the of experimental design on the values of these parameters has been discussed in detail (Ghanbari et al., 2006, Yang et al., 2007a). A key intrinsic factor is the turnover of the inhibited enzyme, characterized by a degradation rate constant (kdeg). This value cannot be determined directly in vivo in humans. Therefore, estimates of the degradation half-life of CYP3A4 have been derived from a variety of indirect sources including rats (in vivo) and Caco-2 cells (range 14–35 h; Mayhew et al., 2000), human hepatocytes and liver slices (range 26–44 h; Maurel, 1996, Pichard et al., 1992) and the time-course of recovery of baseline CYP3A4 activity as marked by plasma drug concentrations after mechanism-based inhibition or enzyme induction (range 43–140 h; collated by Yang et al. (2008). It should be noted that the higher values from the latter source do not correct for the residence time of the compound used to mark enzyme activity. Estimates of the half-life of intestinal CYP3A4 range from 12 to 33 h (Greenblatt et al., 2003, Lilja et al., 2000, Lundahl et al., 1995, Takanaga et al., 2000), largely reflecting the shorter turnover of enterocytes relative to that of the enzyme.

Attempts to predict the extent of DDIs involving CYP3A4-mediated time-dependent inhibition have been based on both static (Galetin et al., 2006, Mayhew et al., 2000, Wang et al., 2004) and dynamic models (Ito et al., 2003, Rowland Yeo et al., 2010, Zhang et al., 2009, Obach et al., 2007). The former evaluated simply the expected change in overall exposure of the victim drug, as indicated by a change in plasma AUC, and assume a constant concentration of inhibitor. The latter models account for temporal changes in inhibitor concentration, active CYP3A4 and the exposure to victim substrate. Although the algorithms used to model these changes were similar across the studies, values of kdeg ranged from 0.0077 h−1 (t1/2 – 90 h) to 0.03 h−1 (t1/2 – 23 h), and there has been considerable debate in the literature regarding the appropriate value to use for accurate predictions. Recently, Wang (2010) used the physiologically based pharmacokinetic (PBPK) model within the Simcyp Simulator (Version 9.3) to assess the impact of the value of kdeg on the accuracy of prediction of 54 DDIs. They concluded that there was a significant improvement in prediction accuracy when the value of kdeg was set at 0.03 h−1 (t1/2 – 23 h) rather than the value of 0.0077 h−1 (t1/2 – 90 h). Although the results appeared to be conclusive, the inactivation parameters were taken from a variety of literature sources and, in some cases, the predicted times-courses of plasma concentrations of substrates and inhibitors varied appreciably from the observed data.

The aim of this study was to use inactivation parameters determined anew in a single laboratory under standardised conditions together with data from substrate and inhibitor files within the Simcyp Simulator (Version 9.3) to determine a value of the hepatic kdeg appropriate for the prediction of DDIs involving time-dependent inhibition of CYP3A4. The conversion of midazolam to its 1′hydroxy metabolite (1’OH midazolam) by human liver microsomes was used as a marker of CYP3A4 activity. The inhibitors studied were azithromycin, clarithromycin, diltiazem and its N-demethylated metabolite (MA) and verapamil.

Section snippets

Chemicals and reagents

Midazolam and N-desmethyldiltiazem were obtained from Cerilliant Corporation (Round Rock, TX). Azithromycin, clarithromycin and erythromycin were purchased from Sequoia Research Products (Oxford, UK). Diltiazem, NADPH and verapamil were obtained from Sigma Chemical Company (St. Louis, MO). The metabolites 1′-hydroxymidazolam and [D4]1′-hydroxymidazolam were synthesised at Pfizer, Inc. (Groton, CT). Human liver microsomes were prepared from a pool of 50 donors provided by BD Biosciences (Woburn,

Time-dependent inhibition of midazolam metabolism

The effects of increasing pre-incubation time and concentration of azithromycin, clarithromycin, erythromycin, verapamil, diltiazem and its metabolite MA on the 1-hydroxylation of midazolam by human liver microsomes are shown in Fig. 1A–F. Inactivation parameters are shown in Table 5, in comparison with those previously reported in the literature. As indicated by the kinact/KI ratio, verapamil was the most potent inhibitor followed by MA, clarithromycin, diltiazem, erythromycin, and

Discussion

We have used inactivation parameters for azithromycin, clarithromycin, diltiazem, MA, erythromycin, and verapamil determined anew in a single laboratory under standardised conditions, together with data from substrate and inhibitor files within the Simcyp Population based ADME Simulator (Version 9.3), to predict DDIs involving time-dependent inhibition of CYP3A4, particularly with respect to the choice of kdeg value. While our findings are broadly in line with those of Wang (2010), who also

Acknowledgement

We thank James Kay for his assistance with the preparation of this manuscript.

References (124)

  • P.M. Shaw et al.

    Reconstitution premixes for assays using purified recombinant human cytochrome P450, NADPH-cytochrome P450 reductase, and cytochrome b5

    Arch. Biochem. Biophys.

    (1997)
  • R.B. Silverman

    Mechanism-based enzyme inactivators

    Methods Enzymol.

    (1995)
  • C.H. Ahsan et al.

    The influence of dose and ethnic origins on the pharmacokinetics of nifedipine

    Clin. Pharmacol. Ther.

    (1993)
  • H. Allonen et al.

    Midazolam kinetics

    Clin. Pharmacol. Ther.

    (1981)
  • J. Amchin et al.

    Effect of venlafaxine on the pharmacokinetics of alprazolam

    Psychopharmacol. Bull.

    (1998)
  • K.L. Austin et al.

    Intersubject and dose-related variability after intravenous administration of erythromycin

    Br. J. Clin. Pharmacol.

    (1980)
  • R.P. Austin et al.

    The influence of nonspecific microsomal binding on apparent intrinsic clearance, and its prediction from physicochemical properties

    Drug Metab. Dispos.

    (2002)
  • J.T. Backman et al.

    Dose of midazolam should be reduced during diltiazem and verapamil treatments

    Br. J. Clin. Pharmacol.

    (1994)
  • J.T. Backman et al.

    Azithromycin does not increase plasma concentrations of oral midazolam

    Int. J. Clin. Pharmacol. Ther.

    (1995)
  • J. Barre et al.

    Pharmacokinetics of erythromycin in patients with severe cirrhosis. Respective influence of decreased serum binding and impaired liver metabolic capacity

    Br. J. Clin. Pharmacol.

    (1987)
  • D.J. Birkett et al.

    Single oral dose pharmacokinetics of erythromycin and roxithromycin and the effects of chronic dosing

    Ther. Drug. Monit.

    (1990)
  • B. Carr et al.

    Characterization of the rhesus monkey CYP3A64 enzyme: species comparisons of CYP3A substrate specificity and kinetics using baculovirus-expressed recombinant enzymes

    Drug Metab. Dispos.

    (2006)
  • E. Chung et al.

    Comparison of midazolam and simvastatin as cytochrome P450 3A probes

    Clin. Pharmacol. Ther.

    (2006)
  • P. Damkier et al.

    Effect of diclofenac, disulfiram, itraconazole, grapefruit juice and erythromycin on the pharmacokinetics of quinidine

    Br. J. Clin. Pharmacol.

    (1999)
  • D. Darbar et al.

    Dietary salt increases first-pass elimination of oral quinidine

    Clin. Pharmacol. Ther.

    (1997)
  • D.J. Edwards et al.

    The effect of coadministration of verapamil on the pharmacokinetics and metabolism of quinidine

    Clin. Pharmacol. Ther.

    (1987)
  • M. Eichelbaum et al.

    Pharmacokinetics of (+)-, (−)- and (+/−)-verapamil after intravenous administration

    Br. J. Clin. Pharmacol.

    (1984)
  • H. El-Eraky et al.

    Effects of sex on the pharmacokinetic and pharmacodynamic properties of quinidine

    Br. J. Clin. Pharmacol.

    (2003)
  • C. Emoto et al.

    Enzymatic characteristics of CYP3A5 and CYP3A4: a comparison of in vitro kinetic and drug–drug interaction patterns

    Xenobiotica

    (2006)
  • C. Emoto et al.

    Approach to predict the contribution of cytochrome P450 enzymes to drug metabolism in the early drug-discovery stage: the effect of the expression of cytochrome b(5) with recombinant P450 enzymes

    Xenobiotica

    (2007)
  • C. Emoto et al.

    Relative roles of CYP2C19 and CYP3A4/5 in midazolam 1′-hydroxylation

    Xenobiotica

    (2007)
  • Food & Drug Administration, 2006. FDA guidance for industry on drug interaction studies – study design, data analysis...
  • G. Foulds et al.

    The pharmacokinetics of azithromycin in human serum and tissues

    J. Antimicrob. Chemother.

    (1990)
  • G. Foulds et al.

    The absence of an effect of food on the bioavailability of azithromycin administered as tablets, sachet or suspension

    J. Antimicrob. Chemother.

    (1996)
  • A.D. Fraser et al.

    Urinary screening for alprazolam and its major metabolites by the Abbott ADx and TDx analyzers with confirmation by GC/MS

    J. Anal. Toxicol.

    (1991)
  • A. Galetin et al.

    Utility of recombinant enzyme kinetics in prediction of human clearance: impact of variability, CYP3A5, and CYP2C19 on CYP3A4 probe substrates

    Drug Metab. Dispos.

    (2004)
  • A. Galetin et al.

    Prediction of time-dependent CYP3A4 drug–drug interactions: impact of enzyme degradation, parallel elimination pathways, and intestinal inhibition

    Drug Metab. Dispos.

    (2006)
  • A. Galetin et al.

    Intestinal and hepatic metabolic activity of five cytochrome P450 enzymes: impact on prediction of first-pass metabolism

    J. Pharmacol. Exp. Ther.

    (2006)
  • F. Ghanbari et al.

    A critical evaluation of the experimental design of studies of mechanism based enzyme inhibition, with implications for in vitro–in vivo extrapolation

    Curr. Drug Metab.

    (2006)
  • J.C. Gorski et al.

    The contribution of intestinal and hepatic CYP3A to the interaction between midazolam and clarithromycin

    Clin. Pharmacol. Ther.

    (1998)
  • D.J. Greenblatt et al.

    Alprazolam kinetics in the elderly. Relation to antipyrine disposition

    Arch. Gen. Psychiatry

    (1983)
  • D.J. Greenblatt et al.

    Effect of age, gender, and obesity on midazolam kinetics

    Anesthesiology

    (1984)
  • D.J. Greenblatt et al.

    Inhibition of triazolam clearance by macrolide antimicrobial agents: in vitro correlates and dynamic consequences

    Clin. Pharmacol. Ther.

    (1998)
  • D.J. Greenblatt et al.

    Alprazolam–ritonavir interaction: implications for product labeling

    Clin. Pharmacol. Ther.

    (2000)
  • D.J. Greenblatt et al.

    Time course of recovery of cytochrome p450 3A function after single doses of grapefruit juice

    Clin. Pharmacol. Ther.

    (2003)
  • S.E. Grimm et al.

    The conduct of in vitro studies to address time-dependent inhibition of drug metabolizing enzymes: a perspective of the Pharmaceutical Research and Manufacturers of America

    Drug Metab. Dispos.

    (2009)
  • E.J. Guest et al.

    Critique of the two-fold measure of prediction success for ratios: application for the assessment of drug–drug interactions

    Drug Metab. Dispos.

    (2011)
  • P. Heizmann et al.

    Pharmacokinetics and bioavailability of midazolam in man

    Br. J. Clin. Pharmacol.

    (1983)
  • N. Hirota et al.

    In vitro/in vivo scaling of alprazolam metabolism by CYP3A4 and CYP3A5 in humans

    Biopharm. Drug Dispos.

    (2001)
  • P.C. Ho et al.

    Effect of grapefruit juice on pharmacokinetics and pharmacodynamics of verapamil enantiomers in healthy volunteers

    Eur. J. Clin. Pharmacol.

    (2000)
  • Cited by (98)

    • Treating asthma in the time of COVID

      2023, Journal of Allergy and Clinical Immunology
    • Inhibition of drug-metabolizing enzymes by Qingfei Paidu decoction: Implication of herb-drug interactions in COVID-19 pharmacotherapy

      2021, Food and Chemical Toxicology
      Citation Excerpt :

      The mixtures were centrifuged at 20,000×g, 4 °C for 20 min, then the supernatant (100 μL) was mixed with ultrapure water (100 μL) in a 1:1 ratio for LC-MS/MS analysis as described in Table S2 (please refer to the details in supplementary material). CYP3A inactivation kinetic experiments were carried out as previous reports (Rowland et al., 2011; Kent et al., 2002; Ji et al., 2015). The incubation mixtures consisted of inactivation groups and activity evaluation groups.

    View all citing articles on Scopus
    View full text