Commentary
Factors confounding the successful extrapolation of in vitro CYP3A inhibition information to the in vivo condition

https://doi.org/10.1016/S0928-0987(02)00002-7Get rights and content

Abstract

For the most part, the majority of adverse drug–drug interactions, which are pharmacokinetic in origin, can be understood in terms of alterations of cytochrome P450-catalyzed reactions. Much is known about the human P450 enzymes, and in many cases it is possible to apply this information to clinically related issues. Of the relatively small subset of the total number of human P450s, CYP3A4 appears to be responsible for the largest fraction of the drug oxidation reactions. As a consequence many important drug–drug interactions observed in the clinic are associated with drugs which are principally metabolized by CYP3A4. The two major reasons for drug–drug interactions involving CYP3A4 are induction and inhibition, with inhibition appearing to be the more important in terms of known clinical problems. Fortunately, with the available knowledge of human P450s and in vitro reagents, it is possible to do experiments with drugs to predict the in vivo condition. The goal of these studies is not only to improve predictions about which drugs might show serious P450 interaction problems, but also to decrease the number of in vivo interaction studies that must be performed in drug development. The focus of the current report is to describe some of the confounding factors associated with in vitro drug inhibition studies and the impact of these issues on in vitro/in vivo extrapolations.

Introduction

The issue of metabolism-based drug interactions continues to draw attention from the medical community, regulatory agencies, and the pharmaceutical industry. Over the past 5 years several prominent drugs have been withdrawn from the market due to serious adverse events associated with drug–drug interactions (e.g. Posicor (mibefradil); Hismanal (astemizole)). In this light, drug–drug interactions are not only a medical problem for clinicians and patients, but also represent a profound economic loss for the sponsoring pharmaceutical companies. As a means of avoiding disastrous in vivo drug interactions, the FDA requires identification of the specific metabolic pathways from which potential inhibition or induction interactions may be inferred and, most recently, the effect of the new drug on hepatic P450 metabolism (Guidance for Industry, 1997). As a consequence, many pharmaceutical companies employ in vitro drug–drug assays early in drug discovery to predict potential interactions of new drug candidates in an attempt to minimize untoward characteristics associated with novel compounds and therefore underwrite the health and safety of patients.

The cytochrome P450s are a super-family of drug metabolizing enzymes found principally in liver and intestine and are responsible for the oxidative metabolism of a wide range of xenobiotic and endobiotic compounds (Guengerich, 2001). In humans P450 3A4 (CYP3A4) is the most abundant P450 enzyme and is responsible for the biotransformation of the majority of drugs currently on the market (Bertz and Granneman, 1997). As a consequence, there are several well-documented clinical drug interactions involving compounds that are principally metabolized by CYP3A4 (Thummel and Wilkinson, 1998).

Comment is provided here to discuss some of the limitations of extrapolating drug–drug interaction data derived from in vitro experiments to the in vivo situation. Emphasis is placed on the confounding features associated with in vitro CYP inhibition studies in particular with CYP3A4, and the possible constraints these findings place on the ability to extrapolate in vitro data in order to predict in vivo inhibitory drug interaction potential.

Section snippets

In vitro approaches to predict drug interactions in vivo

Historically, most drug–drug interaction studies were conducted relatively late in development (typically Phase II or III clinical trials focusing primarily on the therapeutic indices of candidate drug and the likelihood of concurrent medications). Unfortunately, the assessment of interaction potential at such a late stage was not very practical for if clinical studies revealed that a drug candidate caused serious drug interactions, it was probably too late to terminate the development of the

Factors which impact accurate Ki determinations

There are numerous experimental factors that may affect the accuracy of Ki estimation. For many drug inhibition studies, inaccuracy in the Ki estimation is a reflection of inappropriate incubation conditions. There exist a variety of conditions that confound one or more key assumptions that form the basis of this type of analysis. For instance, complications arise when the concentration of substrate and/or inhibitor in solution is significantly reduced by nonspecific binding to components of

CYP3A4-mediated drug metabolism

The active site of CYP3A4 is generally considered to be spacious as evidenced by its ability to oxidize a wide range of structurally diverse molecules. Moreover, CYP3A4 may also exhibit atypical kinetic profiles including positive co-operativity (Ueng et al., 1997) and substrate inhibition (Wang et al., 2000). In deference to the large CYP3A4 active site and the atypical kinetics associated with certain CYP3A4 mediated oxidations, it has been hypothesized that two substrates may physically

Conclusion

As the end result, the various anomalies associated with CYP3A4 inhibition kinetics suggest that in vitro conditions and resulting substrate-velocity curves may not be consistent with the Michaelis–Menten kinetic model and thus, may not be appropriate in predicting drug interactions in vivo. Moreover, the observed atypical kinetics is not restricted to CYP3A4, since nonMichaelis–Menten kinetics has been reported for other P450s enzymes (Hutzler et al., 2001, Lin et al., 2001). From a pragmatic

References (39)

  • J.M. Grace et al.

    Metabolism of artelinic acid to dihydroqinqhaosu by human liver cytochrome P4503A

    Xenobiotica

    (1999)
  • J.M. Grace et al.

    Metabolism of beta-arteether to dihydroqinghaosu by human liver microsomes and recombinant cytochrome P450

    Drug Metab. Dispos.

    (1998)
  • F.P. Guengerich

    Common and uncommon cytochrome P450 reactions related to metabolism and chemical toxicity

    Chem. Res. Toxicol.

    (2001)
  • Guidance for Industry, 1997. Drug Metabolism/Drug Interaction Studies in the Drug Development Process: Studies In...
  • G.R. Harlow et al.

    Analysis of human cytochrome P450 3A4 cooperativity: Construction and characterization of a site-directed mutant that displays hyperbolic steroid hydroxylation kinetics

    Proc. Natl. Acad. Sci. USA

    (1998)
  • L.D. Heimark et al.

    The mechanism of the interaction between amiodarone and warfarin in humans

    Clin. Pharmacol. Ther.

    (1992)
  • N.A. Hosea et al.

    Elucidation of distinct ligand binding sites for cytochrome P450 3A4

    Biochemistry

    (2000)
  • J.B. Houston et al.

    In vitro–in vivo scaling of CYP kinetic data not consistent with the classical Michaelis–Menten model

    Drug Metab. Dispos.

    (2000)
  • J.M. Hutzler et al.

    Dapsone activation of CYP2C9-mediated metabolism: evidence for activation of multiple substrates and a two-site model

    Drug Metab. Dispos.

    (2001)
  • Cited by (16)

    • Catalytic turnover of pyrene by CYP3A4: Evidence that cytochrome b <inf>5</inf> directly induces positive cooperativity

      2005, Archives of Biochemistry and Biophysics
      Citation Excerpt :

      The practice of extrapolating in vitro P450 data to predict human clearance and metabolic inhibitory potency is a fundamental activity of the drug discovery process. However, environmental factors such as the ionic strength of the buffer, pH, lipids, and divalent cations can affect P450 function [8,17,37,38] and may confound drug metabolism predictions. Biological factors such as reductase [39,40], P450 [25], and cytochrome b5[24,41] may also alter P450 structure/function.

    View all citing articles on Scopus
    View full text