Sage Journals: Discover world-class research

Abstract

Hepatic drug metabolizing enzyme (DME) induction complicates the development of new drugs owing to altered efficacy of concomitant treatments, reduction in exposure resulting from autoinduction, and potential generation of toxic metabolites. Risk assessment of DME induction during clinical evaluation is confounded by several uncertainties pertaining to hazard identification and dose response analysis. Hepatic DME induction rarely leads to clinical evidence of altered metabolism and toxicity in the patient, which typically occur only if the DME induction is relatively severe. High drug doses are associated with a greater likelihood of hepatic DME induction and downstream effects; therefore, drugs of low potency requiring higher dosing tend to lead to a greater risk of drug–drug interactions. Vigilance in clinical trials for increased or diminished drug effect and, specifically, pharmacokinetic studies in the presence of other drugs and concomitant diseases are necessary for a drug risk assessment profile.

Efforts to remove hepatic DME-inducing drugs from development can be facilitated with current in vitro and in vivo assessments and will improve with the development of newer technologies. A carefully tailored case-by-case approach will lead to the development of efficacious drugs with an acceptable risk/benefit profile available to patients.

Keywords

metabolism induction hepatic drug-metabolizing enzymes enzyme induction risk assessment

Introduction

Drug development involves characterization of a safety profile for the new compound and conducting risk assessment based on considerations of drug benefits versus the safety risks for a patient population. Much of the safety profile is determined through in vitro assays and in vivo nonclinical animal studies (Graham and Lake 2008; Hewitt et al. 2007). From these nonclinical studies, information is extrapolated to clinical studies in patient populations in order to assess the risk of undesired effects, which may occur during drug development or after regulatory approval when the drug is marketed. The development of a risk assessment strategy in early compound development may be challenging because of difficulties inherent in the extrapolation of nonclinical data to humans. These difficulties are explored in greater detail later in this manuscript and include notable species differences in DME induction mediated by nuclear hormone receptors. In addition, the lack of knowledge concerning the efficacious circulating concentration of a molecule in development prevents in vitro assay results from being readily put into a clinical perspective.

Hepatic DME induction by compounds in development complicates their risk assessment, since DME induction can alter the metabolism of the compound under development and/or other co-administered xenobiotics and drugs. Phase I metabolic processes introduce or reveal polar functional groups onto a drug molecule and are often catalyzed by the members of the cytochrome P450 (CYP) enzyme system. The CYPs are particularly important in the safety risk assessment process of a drug in development, since they are commonly involved in the metabolism of xenobiotics. Two-thirds of the top 200 drugs prescribed in the United States in 2002 are cleared through metabolism that involves CYPs (Williams et al. 2004), and many of the CYPs are also inducible (Hewitt et al. 2007). Phase II enzymes, which add an endogenous polar compound such as a sugar to a polar functional group of the metabolized molecule, are fewer in proportion and not inducible to the same extent as are some CYPs (Soars et al. 2004). However, induction of these enzymes, such as UDP-glucuronosyltransferases (UGTs), can further add to the complexity of drug safety risk assessment.

The most common effect of hepatic DME induction by a compound is enhanced metabolism of other co-administered xenobiotics with accelerated rates of clearance and possibly increased production of reactive intermediates or toxic metabolites (Lin 2006; Smith 2000). Even though induction of hepatic DMEs leads to a more complex risk assessment and may slow development of a compound, in itself, DME induction does not preclude development of compounds and is observed with a number of marketed drugs and various other xenobiotics, including carbamazepine, phenytoin, efavirenz, rifampicin, phenobarbital, and St. John’s wort.

Both induction and the more commonly reported inhibition of the activity of the major hepatic DMEs of the CYP superfamily need to be considered in risk assessment (Tucker et al. 2001). This manuscript focuses on current preclinical testing and risk assessment strategies of compounds in development with potential hepatic DME induction properties and provides relevant clinical examples.

Screening Tests for Human CYP Induction with Implications for Risk Assessment

To study the induction potential of a compound and allow for a risk assessment before substantial resources are spent on a compound’s development, screens have been developed over recent years. Induction of DMEs can occur through various mechanisms, most often by nuclear receptor–mediated transactivation, such as pregnane X receptor (PXR), aryl hydrocarbon receptor (AhR), and constitutive androstane receptor (CAR), and in rodents, the peroxisome proliferator-activated receptor α (PPAR α). The study of induction of DMEs by PXR is more important than induction by CAR or Ahr. The PXR nuclear receptor is responsible for strong induction of CYP3A4, the CYP enzyme most commonly involved in the metabolism of drugs. Although CAR can be involved in the induction of CYP3A4, it is usually involved to a lesser extent (Goodwin et al. 2002). Screens assess the possibility of an increase in enzyme levels through activation of these nuclear receptors. A more thorough discussion of the mechanisms of DME induction was provided in the introductory manuscript for this series (Botts et al. 2010).

The ability of a prospective drug to cause enzyme induction is best assessed early in drug discovery/development to enable the synthesis of new molecules with low potential for human hepatic enzyme induction or to mitigate and understand the risk associated with DME induction. A significant limitation in screening compounds early in the discovery process of drug development is limited knowledge of what the therapeutically efficacious circulating plasma concentration of the molecule will be in humans. This limitation makes the extrapolation of data from the in vitro screens/preclinical models to the clinical situation more complicated and imprecise. The models differ in sophistication and utility, but all give some information on the risk of induction of DMEs.

In Vitro Models

The simplest in vitro model to study CYP3A hepatic DME induction is a competition assay between a compound and a radiolabeled substrate for the ligand-binding domain of PXR, attached to a scintillation-containing bead. Binding of the compound to PXR displaces radiolabeled ligand, resulting in less signal observed (Moore and Kliewer 2000). Although the method is reproducible and allows for very high throughput, it is prone to a high number of false positive results because a compound’s success in binding and displacing the ligand does not necessarily mean that the compound will be an activator/inducer in vivo and is thus rarely used. Another simple assay uses a cell line transfected with a reporter gene, such as the firefly luciferase gene, which is under the control of a PXR promoter (Raucy et al. 2002). Both of these simple in vitro assays involve testing of a compound’s ability to bind to the PXR receptor with the assumption that binding to the receptor leads to induction in vivo. Since this is not always the case, positive results of receptor binding indicate only the potential for induction.

Assays to study the activation of the AhR receptor have also been developed (Quattrochi and Tukey 1993), but they are used less often in drug development (Chu et al. 2009). It has not been possible to develop a reporter assay for CAR, since the receptor has high constitutive activity, especially in continuously cultured cells.

Whereas the previously mentioned in vitro assays measured receptor binding and activation, other models measure changes in CYP mRNA transcription following exposure of hepatocytes to chemical inducers. Although increased mRNA levels do not necessarily indicate increased protein translation, good correlation between the increased CYP3A4 mRNA and CYP3A4 enzymatic activity has been shown with several prototypical inducers through a PXR-mediated mechanism (Luo et al. 2002). Other mechanisms for increased enzyme activity are possible, including stabilization of the enzyme by the inducer, leading to reduced turnover, the postulated mechanism for CYP2E1 induction by ethanol and other ligands (Chien et al. 1997).

The gold standard for the in vitro determination of DME induction potential is the measurement of enzyme activity in plated primary human hepatocytes following exposure to a potential inducer. As compared to the aforementioned in vitro assays, the primary human hepatocyte assay will determine with greater certainty whether a xenobiotic/compound is a likely inducer in vivo (U.S. Department of Health and Human Services 2006). This assay can determine induction potential arising from activation of nuclear receptors along with mRNA stabilization, increased translation efficiency, and protein stabilization. Many of these latter mechanisms cannot be determined using the more basic in vitro assays.

Very good in vitro/in vivo extrapolation has been shown with the primary human hepatocyte in vitro assay. For example, lack of enzyme induction, as determined by activity measurement in primary human hepatocytes, may be sufficient information not to require a clinical study by the Center for Drug Evaluation and Research (CDER) of the United States Food and Drug Administration (FDA) (U.S. Department of Health and Human Services 2006). The in vitro/in vivo extrapolation is very much dependent on the accurate estimation of the pharmacologically efficacious circulating drug concentration in human serum. This estimation often improves as the drug moves through the stages of its development, and thus risk assessment becomes more informed as more data become available.

Cell lines that allow the study of induction in a manner similar to primary human hepatocytes through measurement of increased CYP3A activity after exposure to an inducer are also being developed. Although assays performed with the cell lines are useful tools for the study of induction via certain mechanisms (Mills et al. 2004; Ripp et al. 2006), recent studies have shown their limitations compared to the gold standard, the primary human hepatocyte culture assay. These limitations are in part owing to the fact that cell lines express less OATP1B1 (SLCO1B1) and OATP1B3 (SLCO1B3) uptake transporter proteins following immortalization, thus limiting access of substrates into the cell (Hariparsad et al. 2008), with possible false negative results. Similarly, other cells express low levels of a functional CAR receptor, limiting identification of compounds that induce through this mechanism (Gupta et al. 2008). The HepaRG cells, however, do appear to maintain much of the metabolic, transport, and induction ability that is lacking in other cell lines and seem to be a more valuable tool than other available cell lines for the study of induction (McGinnity et al. 2009).

Cell lines were primarily developed to overcome some of the limitations of the primary hepatocyte culture assay, including the high cost, limited availability, and sample variability between donors. In addition, the use of human hepatocytes in culture is generally limited to short-duration studies, as the cells rapidly undergo phenotypic alterations, which render them less useful for accurate evaluations of long-term toxicity as a result of compound-associated enzyme induction. However, recent studies suggest that this limitation may be overcome by fresh human hepatocytes co-cultured in optimized micropatterns with fibroblasts, which allows the hepatocytes to maintain their metabolic and transport activities much longer in culture than unorganized hepatocytes, thus permitting the study of both induction and hepatotoxicity (Khetani and Bhatia 2008).

An even more informed risk assessment of the potential of drug–drug interactions via induction may be achieved through in vitro–to–in vivo extrapolation of the in vitro data generated in the primary hepatocyte assay. This process includes modeling of data on the effective concentration and maximal induction response of the inducing compound, binding of the compound to plasma proteins and the hepatocytes, potential clinical concentrations of the compound, and fractions of drugs cleared by CYP3A4. This modeling results in a more quantitative prediction of drug–drug interactions aiming for a better informed termination of a compound earlier in development, or the planning of clinical drug–drug interaction studies (Shou et al. 2008).

Although a compound may induce hepatic enzymes in the in vitro primary human hepatocyte assay, clinical studies may not reveal increases in hepatic DME activity at therapeutic doses. In addition, the degree of DME induction may be clinically variable because of genetic variations in CYP expression among humans and different plasma/blood levels of the inducer. For example, omeprazole, a known in vitro inducer of CYP1A, has been shown in some clinical studies not to be an inducer of hepatic enzymes in vivo (Andersson et al. 1991; Rizzo et al. 1996; Xiaodong et al. 1994). However, CYP1A induction was observed at therapeutic doses in patients who had higher circulating serum concentrations of omeprazole, because these patients expressed the “poor metabolizer” phenotype of CYP2C19, omeprazole’s main clearance enzyme (Rost et al. 1992; Rost et al. 1994). This is an example of how genetic variability may affect blood levels of the compound in patients and complicate the extrapolation of in vitro findings to risk in the patient population.

Animal Models

Although preclinical animal species are used to evaluate risk of hepatotoxicity, translation of these findings to humans may be unpredictable. The animal model for hepatic DME induction is most often the rat, but other species such as the mouse, dog, and monkey are also used. The different animal models have been shown to have many differences in the ligand-binding domain of the various nuclear hormone receptors compared with humans, resulting in different induction responses to chemicals between different species.

Induction is an important observation in preclinical animal testing, but DME in animals may or may not correlate with enzyme induction in humans, as is discussed below. Therefore, before abandoning a molecule because of its induction potential in animals, it may be worth testing it using human in vitro systems.

Drug-metabolizing enzyme induction through PXR, the most common nuclear receptor involved in human induction, illustrates issues in predicting induction in humans using preclinical species: only 76% sequence identity in the PXR ligand binding domain occurs between human and mouse (Lehmann et al. 1998). This difference may lead to substantial variations in induction by the same chemical between species. There are many examples of chemicals that cause induction in certain species but have no effect in others. Rifampicin is one example; it induces CYP3A in humans and rabbits, but has no effect on this enzyme in rats or mice (Kocarek et al. 1995). Another example is pregnenolone-16-α-carbonitrile (PCN), which gives rise to strong induction of CYP3A in mice and rats, but causes no induction response in rabbits and humans. On the other hand, CYP3A induction by 5α-pregnane-3,20-dione is seen only in humans and mice, but not in rats and rabbits (Wrighton et al. 1985).

Other studies have explored and used the differences in the amino acid sequence of the putative binding site of the CAR receptor (Jyrkkarinne et al. 2005) and the AhR receptor (Ramadoss and Perdew 2004) of humans and mice to explain the differences seen in a compound’s induction potential between these two species.

The azole antifungal voriconazole, for example, shows autoinduction observed as increased levels of total CYP, inducing the enzyme responsible for its own clearance, which results in decreased exposure upon multiple dosing in the mouse, rat, and dog. In humans, however, this autoinduction is not observed (Roffey et al. 2003). The steady-state free concentration of voriconazole in humans overlaps with concentrations that cause induction in animals, which suggests that the lack of induction in humans is a result of receptor differences. This example illustrates the limitations of the use of animal models to predict induction in humans. There are, however, other examples where induction of DMEs in rats is also observed in humans. For example, TSU-68, an experimental receptor tyrosine kinase inhibitor, showed autoinduction of CYP1A in rats (Kitamura et al. 2007; Kitamura, Matsuoka et al. 2008) and humans (Kitamura, Asanoma et al. 2008; Sessa et al. 2006).

New methods for dealing with these species differences are being developed. For example, new mouse models are commercially available in which the PXR and CAR genes have been knocked out, and mice in which the human PXR and CAR genes have been knocked in are also available (Scheer et al. 2008). Recently, the generation of a PXR-humanized mouse model by bacterial artificial chromosome transgenesis and its utility in predicting a rifampicin-midazolam interaction were reported (Ma et al. 2007).

Clinical Implications and Toxicity Resulting From Hepatic DME Induction

The following section emphasizes the importance of understanding the consequences and complexities of hepatic DME induction. Questions that need to be answered in a risk assessment approach include: what are the enzymes with increased levels following the induction process; what are the mechanisms of induction; what are the human genetic differences in the metabolic steps involved; and what other xenobiotic exposures and exogenous factors affect microsomal enzyme activity?

In addition to the induction potential of a molecule, the preclinical assessment takes into account issues such as enzyme specificity and potential for generation of reactive metabolites. Preclinical toxicity is evaluated for evidence of a role of reactive metabolites. There are few examples, some of which are mentioned below, that have shown impact of enzyme induction, both on progressing drugs to market and/or on the clinical use of drugs. The fact that there are few documented clinical issues supports the approach to assess the risk:benefit ratio before determining a molecule’s suitability for development.

A drug may interfere with another compound during concomitant drug therapy (e.g., retroviral therapy; Fichtenbaum and Gerber 2002), concomitant use of herbal remedies (Pal and Mitra 2006), or exposure to environmental chemicals (Conney 1967; Kluwe and Hook 1981; Kluwe et al. 1982). A toxicological outcome of hepatic DME induction will be a result of any change in the balance of detoxification versus production of toxic metabolites, which is most apparent for inducers of phase I or phase II metabolism. More explicitly, drug toxicity could result from a variety of factors including: (a) enhanced activation, (b) decreased detoxification, (c) decreased inactivation, and (d) an altered balance between activation and inactivation (Park et al. 1996). In addition, there has been increasing awareness that chemically reactive metabolites may produce a range of toxic drug reactions by reacting covalently with essential cellular components. The proportion of a drug that is converted into its reactive metabolite will depend largely on the activity and amount of the enzyme(s) responsible for generating the reactive metabolite; alternatively, the activity and amount of the enzyme(s) involved in deactivating chemically reactive metabolites is also important (Breckenridge 1987). The toxicity of a drug has often been shown to correlate with dose, in that the higher the dose, the higher the amount of toxic metabolite formed and the greater the likelihood of clinically relevant toxicity (Smith and Obach 2005).

Several classic examples that either implicate hepatic DME induction, at least in part in terms of mechanism, or focus on hepatotoxicity as an untoward outcome of enzyme induction are outlined below. Other effects of induction, such as alterations in reproductive and thyroid hormones, are briefly discussed in the companion Clinical Pathology publication (Ennulat et al. 2010).

Human DMEs can be induced to a clinically relevant extent by various xenobiotics, including certain drugs, cigarette smoke, alcohol, and dietary factors (Ronis and Ingelman-Sundberg 1999). In the clinic, the nutritional state of patients will vary greatly from those on high-protein Western diets to those who are cachectic owing to serious illness or malnourished for other reasons.

A historical example of enhanced toxicity from activated metabolites owing to hepatic DME induction of CYP2E1 is carbon tetrachloride. Although clinical toxicity owing to the formation of greater levels of a toxic metabolite via hepatic DME induction is not readily observed in humans, several examples of toxicity have been reported in rats. For example, different nutritional states of rats have been shown to modulate carbon tetrachloride–induced hepatic DME activity and the resulting toxicity (McLean and McLean 1966). Other animal studies showed that heavy consumption of alcohol (Maling et al. 1975) or high-protein diets may also result in an increase in a compound’s enzymatic metabolism, possibly leading to increased toxicity of chemicals that are activated by the liver (Dickerson et al. 1976; Seawright and McLean 1967).

Alcohol-induced liver disease is a classic example of a toxicity related in part to hepatic DME induction. The fact that many of the proposed mechanisms are still speculative illustrates the complexity of parameters informing risk assessment considerations. Ethanol intake is associated with a two-fold increase in CYP2E1 protein content (Perrot et al. 1989) and a three-fold elevation of CYP2E1 mRNA (Takahashi et al. 1993) in human liver biopsy samples. The main enzymes involved in the metabolism of alcohol are the alcohol dehydrogenases and CYP2E1. A significant role for CYP2E1 in the pathogenesis of alcohol-induced liver disease has been postulated because of the occurrence of antibodies directed against CYP2E1 proteins that have been modified by hydroxyethyl free radicals (Ronis et al. 1996). These radicals are putative metabolites of CYP-mediated ethanol oxidation (Albano et al. 1991). An increase in transcription rate, along with the post-transcriptional stabilization resulting from exposure to alcohol, may be sufficient to overcome detoxification mechanisms for the inactivation of the ethoxyl radicals and thus cause toxicity owing to antigen formation (Albano 2006; Park et al. 1996).

Even greater than the consumption of alcohol alone is the enhanced risk of liver damage secondary to consumption of acetaminophen (referred to as paracetamol in Europe) in heavy alcohol consumers resulting from CYP2E1 induction (Seeff et al. 1986). CYP2E1, secondary to alcohol induction, can catalyze the activation of acetaminophen to N-acetyl-p-benzoquinoneimine (NAPQI), without concomitant induction of the major detoxification pathways of acetaminophen (Nelson 1990).

Similar to acetaminophen-related hepatotoxicity, an increased risk of hepatotoxicity and other toxic side effects is observed in heavy alcohol consumers exposed to the anesthetics halothane, enflurane, and isoflurane, since these are also metabolized by CYP2E1. Halothane hepatitis, which is now regarded as a model for immune-mediated hepatotoxicity, appears to be caused by the formation of neoantigens derived from oxidative defluorination of the drug, a pathway dominated by CYP2E1 (Gut et al. 1993).

Tacrine, a centrally acting cholinesterase antagonist for the treatment of Alzheimer’s disease, has been shown to cause substantial elevations of ALT levels in as many as 20–50% of patients on the drug (Gracon et al. 1998; Watkins et al. 1994). The hepatic metabolism of tacrine, including bioactivation, is catalyzed exclusively by CYP1A2 (Spaldin et al. 1995), which might explain individual susceptibility, since there is wide individual variation in expression of this highly inducible enzyme. In addition, cigarette smoking, which induces CYP1A2, has been shown to alter the pharmacokinetics of tacrine, resulting in a two-thirds lower exposure in smokers (Sciele 2009). In vitro experiments with human liver microsomes have shown that tacrine undergoes metabolic activation to chemically reactive cytotoxic quinone-methide metabolites (Madden et al. 1995).

Induction of CYP3A4 enzymes has been demonstrated in humans following treatment with the anticonvulsants phenytoin and carbamazepine (Friedmann et al. 1994). Either drug may produce a hypersensitivity syndrome in some patients treated over a prolonged period, resulting in skin and liver damage (Bu et al. 2005; Pearce et al. 2008; Pearce et al. 2005; Wolkenstein et al. 1998). The principal pathways of metabolism of carbamazepine include glucuronidation, a postulated toxic metabolic pathway 10,11-epoxidation (Bu et al. 2005), followed by hydrolysis producing a dihydrodiol (indirect detoxification), and aromatic hydroxylation. Other minor pathways may have some responsibility for the observed hepatotoxicity (Pearce et al. 2008; Pearce et al. 2005).

Drug-metabolizing enzyme induction may have other outcomes besides toxicologic effects. Drug-metabolizing enzyme induction may result in an accelerated metabolism with generation of increased inactive or nontoxic products. In this case, a pharmacokinetic effect may be noted as increased drug clearance and reduced (potentially subtherapeutic) blood levels. The dosage may consequently need to be increased to restore the desired effect. In this context, numerous examples of CYP3A induction resulting in changes in pharmacokinetics and/or pharmacodynamics have been reported in the literature (Niemi et al. 2003). For example, rifampicin, a very strong inducer of DMEs through PXR, was shown to increase the metabolism and clearance of warfarin, leading to decreased activity of the anticoagulant (Venkatesan 1992). In addition to pharmaceuticals, herbal products may be CYP inducers. A report of the induction effects of St. John’s wort on the pharmacokinetics and the pharmacodynamics of the CYP3A4 substrate cyclosporine was described in a heart transplant patient; the effects led to an acute rejection episode (Ruschitzka et al. 2000).

Autoinduction, mentioned previously, may also result in a need for medication dosage adjustments over time. An example is the anticonvulsive carbamazepine, which induces (primarily via PXR) and is substantially metabolized by CYP3A4 (Kerr et al. 1994; Oscarson et al. 2006). Because of this autoinduction, chronic administration of carbamazepine often requires increases from the initial dose of 400 mg per day up to 1800 mg per day (Novartis 2009).

Not all autoinducing drugs require increasing doses because of the short duration for which they are given. For example, the antimalarial drug artemisinin, which increases the expression of CYP2B6 through activation of both the CAR and PXR pathways (Burk et al. 2005), also increases its own clearance to reduce drug exposure (Simonsson et al. 2003). Despite the induction of drug metabolism, the dosage of artemisinin is usually not changed and efficacy is maintained because the duration of drug administration is generally short (Le et al. 1999).

Another consequence of DME induction applies to drugs (Schoedel and Tyndale 2003) and chemicals (the classic example is carbon tetrachloride) that require metabolism for activation (McLean and McLean 1966). Prodrugs rely on the conversion of the administered molecule to the active form before the drug can exert its pharmacologic effect. Drugs such as oseltamivir and valacyclovir, for example, are almost completely (nearly 100%) converted to the active drug by first-pass intestinal and/or hepatic metabolism before entering the systemic circulation, and thus higher levels of the activating enzymes would not result in a more active compound being formed. In another situation, a large amount of prodrug may circulate and only slowly be converted to the active drug. Drug-metabolizing enzyme induction may thus result in a more rapid onset and greater intensity of the desired effect (driven by a shorter time to maximal concentration, or t_max, and a higher maximal concentration, or C_max), however, with a shorter duration of action. Consequently, the dose administered may have to be modified (i.e., lower dose given more frequently) to restore the desired intensity and duration of the therapeutic effect. An example of a modest increase in formation of the active drug from the prodrug has been shown with the antiplatelet aggregation drug clopidogrel following administration of rifampicin. A modest though statistically significant decrease in platelet aggregation was observed following pretreatment with rifampicin (Lau et al. 2004), initially thought to be a result of induction of CYP3A4; increased levels of CYP2C19 may also be responsible (Shuldiner et al. 2009).

Hepatic DME Induction Leading to Increased Metabolites Unique to Humans versus Preclinical Animal Species

Preclinical safety assessment studies are critical to identify potential toxicity in humans and to ensure the test drug’s safety at proposed clinical doses. Safety studies in animal models generally test compounds at doses that are multiples of the projected human treatment doses. The effects of test compounds on most phase I and phase II reactions are thus often relatively greater in the animal models than in humans. Therefore, in the vast majority of cases, both the qualitative and quantitative predictions from preclinical safety studies concerning drug safety are adequate to assess safety. Occasionally, however, metabolites that are generated uniquely by humans are identified and present a challenge in the risk assessment of a xenobiotic.

Because these metabolites are not formed in the surrogate preclinical animal species, risk associated with them cannot be predicted by preclinical studies. As mentioned previously, there are often important quantitative differences in the rate and the extent of metabolism across surrogate animal species and humans. In a small number of cases, when the major versus minor pathways of metabolism are dissimilar across species, there may be human-specific “unique” or “dominant metabolites” that may not have been adequately tested in nonclinical safety studies. When a human-only metabolite is not formed in the surrogate species, an incomplete and/or inadequate preclinical drug safety assessment may result in an underestimation of hazard. This has recently been the subject of regulatory concern (ICH Topic M 3 [R2] 2009). The note for guidance states that a unique human metabolite must be tested for toxicity “when that metabolite(s) is observed at exposures greater than 10% of total drug-related exposure and at significantly greater levels in humans than the maximum exposure seen in the toxicity studies.” This has recently been the subject of FDA concerns (FDA/CDER guidance on safety testing of drug metabolites) (U.S. Department of Health and Human Services 2008). The FDA/CDER draft guidance states that a unique human metabolite must be tested for toxicity when the metabolite level reaches > 10% parent systemic exposure at steady state. Some argue that the amount of a metabolite presented as milligrams or molar equivalents of metabolite, rather than a percentage of AUC, is more relevant to toxicity (Smith and Obach 2005), along with the time of exposure to the chemical (Smith and Obach 2009). This argument would support, as previously mentioned, that lower doses and shorter exposure periods of a compound are less likely to result in toxicity.

Based on the information above, hepatic DME induction by xenobiotics can complicate human risk assessment through various mechanisms at any point of drug development, both preclinically and clinically. Clinically, hepatic DME induction through drug–drug or drug–herbal interactions is a well-known, important variable among patients, leading to alterations of pharmacokinetic parameters and/or unpredictable nonoptimal dosage regimens. Clinically relevant pharmacodynamic changes owing to hepatic DME induction are less common, although they may occasionally account for idiosyncratic reactions, possibly attributable to active metabolites unique to individual human beings. If a metabolite is responsible for toxicity, the quantitative differences in exposure to a major toxic metabolite between humans and laboratory animals can lead to misrepresentation of safety margins if these are based on the parent drug exposure.

The most common effect of hepatic DME induction, however, is enhanced metabolism and an accelerated rate of clearance. Alteration of pharmacokinetic parameters in this way can lead to variations that may require appropriate dosage adjustments and thus complicate hazard estimations preclinically and clinically.

Risk Assessment and Risk Mitigation

As outlined above, hepatic DME induction in the clinical setting typically results in more rapid metabolism of xenobiotics, with possible unwanted effects including decreased activity of the parent compound or possibly increased formation of reactive metabolites (Lin 2006; Smith 2000). From the perspective of the pharmaceutical industry, potent enzyme inducers are less likely to be selected for drug development, predominantly because of concerns regarding possible drug–drug interactions or an increased propensity for hepatotoxicity.

Hazard Identification and Weight of Evidence

The FDA and other regulatory agencies have provided guidance on how to study and address hepatic DME induction for regulatory submissions based on which drug–drug interactions can be relatively confidently predicted (U.S. Department of Health and Human Services 2006). However, there are only a few well-understood examples of drug–drug interactions demonstrating clear, known toxicity resulting from hepatic DME induction. In certain instances, hepatic DME induction can be demonstrated with hepatotoxic drugs, but linking the two phenomena remains problematic (Ayrton et al. 1991). The previously mentioned example of acetaminophen and ethanol interactions exemplifies the complexity involved in formulating a risk assessment for hepatic DME induction resulting in liver toxicity; the precise pathological mechanism, however, continues to be debated. Although CYP2E1 induction by ethanol is known to result in excess enzymatic capacity, leading to increased formation of a toxic intermediate, this effect is short-lived following cessation of alcohol intake (Thummel et al. 2000). A recent study in patients with a history of chronic alcohol consumption showed that the maximal recommended dose of acetaminophen for three days did not result in any increases in parameters measuring liver injury, including serum transaminases (Kuffner et al. 2007). In addition, possible changes in glutathione (GSH) levels in some alcohol abusers need to be considered. CYP2E1 induction importance is apparent only in overdose, because once saturation of sulfation and GSH conjugation has occurred, even small changes in NAPQI formation could have a marked effect on the dose–toxicity relationship.

A recent clinical review showed that acetaminophen toxicity is mostly a matter of deliberate overdosing (10× therapeutic dose), and is not due to marginal differences in sensitivity, such as those that might be observed with 1.5- to 2-fold changes in CYP2E1 enzyme levels (Hersh et al. 2007).

Limitations in risk assessment of drug safety pertain also to hazard identification, because it is not easy to precisely qualify clinically the adverse nature of hepatic DME induction. Uncertainty in correlating enzyme induction observed in in vitro models, in preclinical species, and/or in small clinical study groups to the entire population of potential patients makes early risk assessment challenging. Part of these early difficulties is inherent in the previously mentioned extrapolation of results from high-dose experimental rodent bioassays to the typically lower therapeutic doses used in humans. Furthermore, a dose-response analysis within the risk assessment of hepatic DME induction is often limited, because the relationship between dose and the probability of the manifestation of an adverse effect secondary to hepatic DME induction is often not reliable. During later phases of drug development, the possibility of drug–drug interactions via DME induction will be better characterized, and the results of interaction studies are included in the drug prescribing information. Although at that point the induction potential of a drug may be predictable for a large part of the population, the induction potential in specific human subpopulations is often still unknown until the drug is used more widely and for longer durations.

Drugs that cause hepatic DME induction may increase microsomal enzymes, leading to hypertrophy of hepatocytes in a dose-dependent fashion. The larger the administered dose, the greater the likelihood that there will be hepatic DME induction, hepatic hypertrophy, and, in preclinical species, liver weight increase (Maronpot et al. 2010).

The traditional thinking has been that hepatotoxic idiosyncratic drug reactions are not dose related. Although there is no population-wide dose response for idiosyncratic drug reactions, a certain population of individuals who are susceptible to the toxicity because of inflammation or other stresses does show a dose response (Roth and Ganey 2010). However, more recent evidence suggests that drug hepatotoxicity occurs mainly with high-dose (i.e., low-potency) drugs (Lammert et al. 2008; Smith 2000; Uetrecht 2007). Hepatotoxicity is now thought to be associated with drugs that must be used at higher concentrations and that are, de facto, less potent agents. Low potency may therefore be a negative selection parameter during drug development (Smith 2000). As an example, troglitazone (used at up to 600 mg daily) was associated with a very low incidence of “idiosyncratic” hepatotoxicity and also showed induction of CYP3A4 (Frantz and Nguyen 1998; Koup et al. 1998; Loi et al. 1999; Park et al. 1998). Similar drugs with higher potency that were not or have not been associated with idiosyncratic hepatotoxicity (rosiglitazone used at up to 8 mg daily and pioglitazone used at up to 40 mg daily) have been marketed. In addition to not causing idiosyncratic drug interactions, the two newer agents have been shown not to result in clinically significant hepatic DME induction at these low doses (Harris et al. 1999; Nowak et al. 2002) even though they do induce enzymes in vitro at fairly high concentrations (Sahi et al. 2003). Thus, lower exposure to a drug that decreases the incidence of idiosyncratic drug reactions should decrease the incidence of the reaction and lower incidence of induction.

Drug safety assessment has to be conducted on a case-by-case basis, with hepatic DME induction being just one of many factors in determining the risk/benefit profile of a drug. A number of factors need to be taken into consideration, including seriousness of the disease state/condition for which the product is indicated, the seriousness of the potential adverse effects, the estimated incidence of the reaction, and the available alternative therapies. In the case of hepatic DME induction, it is relatively straightforward to identify the effect on metabolism of common concomitantly administered drugs, but it is difficult to identify all possible interactions with all drugs and xenobiotics and to identify the toxicities that may result from induced metabolism of endogenous and exogenous substances. Examples of therapeutic failure relating to hepatic DME induction include unintended pregnancies while on birth control medication (Back et al. 1988). Other drug interactions through induction are often more difficult to identify than interactions through inhibition because of the nature of the diseases being treated. A pharmacodynamic effect of enzyme inhibition often manifests itself as a side effect because of the supratherapeutic drug levels observed after enzyme inhibition. The effect of hepatic enzyme induction by a drug may not be observed clinically because the specific disease treated (e.g., increased cholesterol levels) may not cause symptoms that are experienced adversely by the patient. In addition, available diagnostic tests for drug efficacy (e.g., blood pressure reading) could be taken long after the DME induction took place. If treatment with the inducer is stopped, the enzyme levels may have returned to normal, and thus the change in status would not be noticed. On the other hand, if the test reading was considered out of range because it was taken during the period of enzyme induction, the clinician may only adjust the dosage or substitute a different therapeutic agent without necessarily connecting the loss of efficacy to hepatic DME induction.

In conclusion, these factors will influence the product’s benefit/risk profile, which will in turn influence the choice of risk-mitigation strategy. This complex interaction of potential effects makes development of a risk mitigation strategy difficult. It would be simple to mitigate risk by eliminating from development any drug that had the ability to induce hepatic enzymes in humans. However, this approach would exclude many effective drugs with a favorable risk/benefit profile for patients. Efforts to screen against DME induction must factor in the potential therapeutic benefit of a compound within the context of the disease to be treated.

As mentioned in previous sections, understanding species differences as applied to preclinical studies and their extrapolation to humans is important for risk assessment and consequently for developing a risk mitigation approach. The CYP profile of laboratory animal species differs from that of humans, which complicates extrapolation from preclinical results to the clinic (Martignoni et al. 2006). New approaches such as metaboomics and transcriptomics may also help in 1 risk assessment by allowing for an earlier and more extensive assessment of the effects of an enzyme inducer (Waterman et al. 2010). Risk mitigation consists of efforts taken to reduce both the probability of the occurrence of an undesired outcome and any detrimental consequences. Risk mitigation strategies take many forms, including eliminating certain kinds of drugs from development based on structure or metabolism profiles, developing of more potent, lower-dose drugs, understanding the concomitant diseases and treatments and possibility of interactions, clinical monitoring, labeling and issuance of information, applying restrictions to patient populations, balancing risk with efficacy, enforcing compliance, post-market monitoring of adverse events, and developing safer drug molecule alternatives (Senior 2007).

Footnotes

The views expressed in this manuscript are those of the authors and do not necessarily represent the policies, positions or opinions of their respective agencies and organizations.

Abbreviations

References

Albano

-2006-. Alcohol, oxidative stress and free radical damage. Proc Nutr Soc 65, 278–90.

Albano

Tomasi

Persson

J. O.

Terelius

Goria-Gatti

Ingelman-Sundberg

Dianzani

M. U.

-1991-. Role of ethanol-inducible cytochrome P450 -P450IIE1- in catalysing the free radical activation of aliphatic alcohols. Biochem Pharmacol 41, 1895–902.

Andersson

Bergstrand

Cederberg

Eriksson

Lagerstrom

P. O.

Skanberg

-1991-. Omeprazole treatment does not affect the metabolism of caffeine. Gastroenterology 101, 943–47.

Ayrton

A. D.

Ioannides

Parke

D. V.

-1991-. Induction of the cytochrome P450 I and IV families and peroxisomal proliferation in the liver of rats treated with benoxaprofen. Possible implications in its hepatotoxicity. Biochem Pharmacol 42, 109–15.

Back

D. J.

Grimmer

S. F.

Orme

M. L.

Proudlove

Mann

R. D.

Breckenridge

A. M.

-1988-. Evaluation of Committee on Safety of Medicines yellow card reports on oral contraceptive-drug interactions with anticonvulsants and antibiotics. Br J Clin Pharmacol 25, 527–32.

Botts

Ennulat

Francke-Carroll

Graham

Maronpot

Mohutsky

-2010-. Introduction to hepatic drug metabolizing enzyme induction studies in drug safety evaluation studies. Toxicol Pathol 38, 794–96.

Breckenridge

-1987-. Enzyme induction in humans. Clinical aspects—an overview. Pharmacol Ther 33, 95–9.

H. Z.

Kang

Deese

A. J.

Zhao

Pool

W. F.

-2005-. Human in vitro glutathionyl and protein adducts of carbamazepine-10,11-epoxide, a stable and pharmacologically active metabolite of carbamazepine. Drug Metab Dispos 33, 1920–24.

Burk

Arnold

K. A.

Nussler

A. K.

Schaeffeler

Efimova

Avery

B. A.

Avery

M. A.

Fromm

M. F.

Eichelbaum

-2005-. Antimalarial artemisinin drugs induce cytochrome P450 and MDR1 expression by activation of xenosensors pregnane X receptor and constitutive androstane receptor. Mol Pharmacol 67, 1954–65.

10.

Chien

J. Y.

Thummel

K. E.

Slattery

J. T.

-1997-. Pharmacokinetic consequences of induction of CYP2E1 by ligand stabilization. Drug Metab Dispos 25, 1165–75.

11.

Chu

Einolf

H. J.

Evers

Kumar

Moore

Ripp

Silva

Sinha

Sinz

Skerjanec

-2009-. In vitro and in vivo induction of cytochrome p450: A survey of the current practices and recommendations: A Pharmaceutical Research and Manufacturers of America perspective. Drug Metab Dispos 37, 1339–54.

12.

Conney

A. H.

-1967-. Pharmacological implications of microsomal enzyme induction. Pharmacol Rev 19, 317–66.

13.

Dickerson

J. W.

Basu

T. K.

Parke

D. V.

-1976-. Effect of protein-energy nutrition on the activity of hepatic microsomal drug-metabolizing enzymes in growing rats. J Nutr 106, 258–64.

14.

Ennulat

Boone

Clemo

Graham

Ledieu

Magid-Slav

Walker

Botts

-2010-. Effects of hepatic drug metabolizing enzyme induction on clinical pathology parameters in animals and man. Toxicol Pathol 38, 808–26.

15.

Fichtenbaum

C. J.

Gerber

J. G.

-2002-. Interactions between antiretroviral drugs and drugs used for the therapy of the metabolic complications encountered during HIV infection. Clin Pharmacokinet 41, 1195–211.

16.

Frantz

R. P.

Nguyen

T. T.

-1998-. Rezulin -troglitazone- greatly increases cyclosporine metabolism. J Heart Lung Transplant 17, 1037–38.

17.

Friedmann

P. S.

Strickland

Pirmohamed

Park

B. K.

-1994-. Investigation of mechanisms in toxic epidermal necrolysis induced by carbamazepine. Arch Dermatol 130, 598–604.

18.

Goodwin

Hodgson

D’Costa

D. J.

Robertson

G. R.

Liddle

-2002-. Transcriptional regulation of the human CYP3A4 gene by the constitutive androstane receptor. Mol Pharmacol 62, 359–65.

19.

Gracon

S. I.

Knapp

M. J.

Berghoff

W. G.

Pierce

DeJong

Lobbestael

S. J.

Symons

Dombey

S. L.

Luscombe

F. A.

Kraemer

-1998-. Safety of tacrine: Clinical trials, treatment IND, and postmarketing experience. Alzheimer Dis Assoc Disord 12, 93–101.

20.

Graham

M. J.

Lake

B. G.

-2008-. Induction of drug metabolism: Species differences and toxicological relevance. Toxicology. 254-3-, 184–91.

21.

Gupta

Mugundu

G. M.

Desai

P. B.

Thummel

K. E.

Unadkat

J. D.

-2008-. Intestinal human colon adenocarcinoma cell line LS180 is an excellent model to study pregnane X receptor, but not constitutive androstane receptor, mediated CYP3A4 and multidrug resistance transporter 1 induction: Studies with anti-human immunodeficiency virus protease inhibitors. Drug Metab Dispos 36, 1172–80.

22.

Gut

Christen

Huwyler

-1993-. Mechanisms of halothane toxicity: Novel insights. Pharmacol Ther 58, 133–55.

23.

Hariparsad

Carr

B. A.

Evers

Chu

-2008-. Comparison of immortalized Fa2N-4 cells and human hepatocytes as in vitro models for cytochrome P450 induction. Drug Metab Dispos 36, 1046–55.

24.

Harris

R. Z.

Inglis

A. M.

Miller

A. K.

Thompson

K. A.

Finnerty

Patterson

Jorkasky

D. K.

Freed

M. I.

-1999-. Rosiglitazone has no clinically significant effect on nifedipine pharmacokinetics. J Clin Pharmacol 39, 1189–94.

25.

Hersh

E. V.

Pinto

Moore

P. A.

-2007-. Adverse drug interactions involving common prescription and over-the-counter analgesic agents. Clin Ther 29 Suppl, 2477–97.

26.

Hewitt

N. J.

Lecluyse

E. L.

Ferguson

S. S.

-2007-. Induction of hepatic cytochrome P450 enzymes: Methods, mechanisms, recommendations, and in vitro-in vivo correlations. Xenobiotica 37, 1196–224.

27.

Jyrkkarinne

Windshugel

Makinen

Ylisirnio

Perakyla

Poso

Sippl

Honkakoski

-2005-. Amino acids important for ligand specificity of the human constitutive androstane receptor. J Biol Chem 280, 5960–71.

28.

Kerr

B. M.

Thummel

K. E.

Wurden

C. J.

Klein

S. M.

Kroetz

D. L.

Gonzalez

F. J.

Levy

R. H.

-1994-. Human liver carbamazepine metabolism. Role of CYP3A4 and CYP2C8 in 10,11-epoxide formation. Biochem Pharmacol 47, 1969–79.

29.

Khetani

S. R.

Bhatia

S. N.

-2008-. Microscale culture of human liver cells for drug development. Nat Biotechnol 26, 120–26.

30.

Kitamura

Asanoma

Nagayama

Otagiri

-2008-. Identification of human liver cytochrome P450 isoforms involved in autoinduced metabolism of the antiangiogenic agent -Z--5-[-1,2-dihydro-2-oxo-3H-indol-3-ylidene-methyl]-2,4-dimethyl-1H-pyrro le-3-propanoic acid -TSU-68-. Drug Metab Dispos 36, 1003–9.

31.

Kitamura

Matsuoka

Nagayama

Otagiri

-2008-. Time-dependent induction of rat hepatic CYP1A1 and CYP1A2 expression after single-dose administration of the anti-angiogenic agent TSU-68. Drug Metab Pharmacokinet 23, 421–27.

32.

Kitamura

Yamamoto

Nagayama

Otagiri

-2007-. Decrease in plasma concentrations of antiangiogenic agent TSU-68 --Z--5-[-1,2-dihydro-2-oxo-3H-indol-3-ylidene-methyl]-2,4-dimethyl-1H-pyrr ole-3-propanoic acid- during oral administration twice a day to rats. Drug Metab Dispos 35, 1611–16.

33.

Kluwe

W. M.

Hook

J. B.

-1981-. Comparative induction of xenobiotic metabolism in rodent kidney, testis and liver by commercial mixtures of polybrominated biphenyls and polychlorinated biphenyls, phenobarbital and 3-methylcholanthrene: Absolute and temporal effects. Toxicology 20, 259–73.

34.

Kluwe

W. M.

Hook

J. B.

Bernstein

-1982-. Synergistic toxicity of carbon tetrachloride and several aromatic organohalide compounds. Toxicology 23, 321–36.

35.

Kocarek

T. A.

Schuetz

E. G.

Strom

S. C.

Fisher

R. A.

Guzelian

P. S.

-1995-. Comparative analysis of cytochrome P4503A induction in primary cultures of rat, rabbit, and human hepatocytes. Drug Metab Dispos 23, 415–21.

36.

Koup

J. R.

Anderson

G. D.

Loi

C. M.

-1998-. Effect of troglitazone on urinary excretion of 6beta-hydroxycortisol. J Clin Pharmacol 38, 815–18.

37.

Kuffner

E. K.

Green

J. L.

Bogdan

G. M.

Knox

P. C.

Palmer

R. B.

Heard

Slattery

J. T.

Dart

R. C.

-2007-. The effect of acetaminophen -four grams a day for three consecutive days- on hepatic tests in alcoholic patients—a multicenter randomized study. BMC Med 5, 13

38.

Lammert

Einarsson

Saha

Niklasson

Bjornsson

Chalasani

-2008-. Relationship between daily dose of oral medications and idiosyncratic drug-induced liver injury: Search for signals. Hepatology 47, 2003–9.

39.

Lau

W. C.

Gurbel

P. A.

Watkins

P. B.

Neer

C. J.

Hopp

A. S.

Carville

D. G.

Guyer

K. E.

Tait

A. R.

Bates

E. R.

-2004-. Contribution of hepatic cytochrome P450 3A4 metabolic activity to the phenomenon of clopidogrel resistance. Circulation 109, 166–71.

40.

N. B.

Pham

T. Y.

Nguyen

B. N.

Dang

C. T.

Pham

T. L.

D. C.

-1999-. Efficacy and effectiveness of five day treatment of uncomplicated falciparum with artemisinin or artesunate in Vietnam. Southeast Asian J Trop Med Public Health 30, 3–6.

41.

Lehmann

J. M.

McKee

D. D.

Watson

M. A.

Willson

T. M.

Moore

J. T.

Kliewer

S. A.

-1998-. The human orphan nuclear receptor PXR is activated by compounds that regulate CYP3A4 gene expression and cause drug interactions. J Clin Invest 102, 1016–23.

42.

Lin

J. H.

-2006-. CYP induction-mediated drug interactions: In vitro assessment and clinical implications. Pharm Res 23, 1089–116.

43.

Loi

C. M.

Stern

Koup

J. R.

Vassos

A. B.

Knowlton

Sedman

A. J.

-1999-. Effect of troglitazone on the pharmacokinetics of an oral contraceptive agent. J Clin Pharmacol 39, 410–7.

44.

Luo

Cunningham

Kim

Burn

Lin

Sinz

Hamilton

Rizzo

Jolley

Gilbert

Downey

Mudra

Graham

Carroll

Xie

Madan

Parkinson

Christ

Selling

LeCluyse

Gan

L. S.

-2002-. CYP3A4 induction by drugs: Correlation between a pregnane X receptor reporter gene assay and CYP3A4 expression in human hepatocytes. Drug Metab Dispos 30, 795–804.

45.

Shah

Cheung

Guo

G. L.

Feigenbaum

Krausz

K. W.

Idle

J. R.

Gonzalez

F. J.

-2007-. The PREgnane X receptor gene-humanized mouse: A model for investigating drug–drug interactions mediated by cytochromes P450 3A. Drug Metab Dispos 35, 194–200.

46.

Madden

Spaldin

Hayes

R. N.

Woolf

T. F.

Pool

W. F.

Park

B. K.

-1995-. Species variation in the bioactivation of tacrine by hepatic microsomes. Xenobiotica 25, 103–16.

47.

Maling

H. M.

Stripp

Sipes

I. G.

Highman

Saul

Williams

M. A.

-1975-. Enhanced hepatotoxicity of carbon tetrachloride, thioacetamide, and dimethylnitrosamine by pretreatment of rats with ethanol and some comparisons with potentiation by isopropanol. Toxicol Appl Pharmacol 33, 291–308.

48.

Maronpot

Dambach

Flake

Takanori

Mueller

Nyska

Singh

Ward

Yoshizawa

-2010-. Hepatic enzyme induction—Microscopic and ultrastructural appearance. Toxicol Pathol

49.

Martignoni

Groothuis

G. M.

de Kanter

-2006-. Species differences between mouse, rat, dog, monkey and human CYP-mediated drug metabolism, inhibition and induction. Expert Opin Drug Metab Toxicol 2, 875–94.

50.

McGinnity

D. F.

Zhang

Kenny

J. R.

Hamilton

G. A.

Otmani

Stams

K. R.

Haney

Brassil

Stresser

D. M.

Riley

R. J.

-2009-. Evaluation of multiple in vitro systems for assessment of CYP3A4 induction in drug discovery: Human hepatocytes, pregnane X receptor reporter gene, and Fa2N-4 and HepaRG cells. Drug Metab Dispos 37, 1259–68.

51.

McLean

A. E.

McLean

E. K.

-1966-. The effect of diet and 1,1,1-trichloro-2,2-bis--p-chlorophenyl-ethane -DDT- on microsomal hydroxylating enzymes and on sensitivity of rats to carbon tetrachloride poisoning. Biochem J 100, 564–71.

52.

Mills

J. B.

Rose

K. A.

Sadagopan

Sahi

de Morais

S. M.

-2004-. Induction of drug metabolism enzymes and MDR1 using a novel human hepatocyte cell line. J Pharmacol Exp Ther 309, 303–9.

53.

Moore

J. T.

Kliewer

S. A.

-2000-. Use of the nuclear receptor PXR to predict drug interactions. Toxicology 153, 1–10.

54.

Nelson

S. D.

-1990-. Molecular mechanisms of the hepatotoxicity caused by acetaminophen. Semin Liver Dis 10, 267–78.

55.

Niemi

Backman

J. T.

Fromm

M. F.

Neuvonen

P. J.

Kivisto

K. T.

-2003-. Pharmacokinetic interactions with rifampicin: Clinical relevance. Clin Pharmacokinet 42, 819–50.

56.

Novartis -2009-. Tegretol—Prescribing Information. Available at: http://www.pharma.us.novartis.com/product/pi/pdf/tegretol.pdf. Accessed November 2009.

57.

Nowak

S. N.

Edwards

D. J.

Clarke

Anderson

G. D.

Jaber

L. A.

-2002-. Pioglitazone: Effect on CYP3A4 activity. J Clin Pharmacol 42, 1299–302.

58.

Oscarson

Zanger

U. M.

Rifki

O. F.

Klein

Eichelbaum

Meyer

U. A.

-2006-. Transcriptional profiling of genes induced in the livers of patients treated with carbamazepine. Clin Pharmacol Ther 80, 440–56.

59.

Pal

Mitra

A. K.

-2006-. MDR- and CYP3A4-mediated drug-herbal interactions. Life Sci 78, 2131–45.

60.

Park

B. K.

Kitteringham

N. R.

Pirmohamed

Tucker

G. T.

-1996-. Relevance of induction of human drug-metabolizing enzymes: Pharmacological and toxicological implications. Br J Clin Pharmacol 41, 477–91.

61.

Park

M. H.

Pelegrin

Haug

M. T.

3rd Young

J. B.

-1998-. Troglitazone, a new antidiabetic agent, decreases cyclosporine level. J Heart Lung Transplant 17, 1139–40.

62.

Pearce

R. E.

Wang

Uetrecht

J. P.

Correia

M. A.

Leeder

J. S.

-2008-. Pathways of carbamazepine bioactivation in vitro. III. The role of human cytochrome P450 enzymes in the formation of 2,3-dihydroxycarbamazepine. Drug Metab Dispos 36, 1637–49.

63.

Pearce

R. E.

Uetrecht

J. P.

Leeder

J. S.

-2005-. Pathways of carbamazepine bioactivation in vitro: II. The role of human cytochrome P450 enzymes in the formation of 2-hydroxyiminostilbene. Drug Metab Dispos 33, 1819–26.

64.

Perrot

Nalpas

Yang

C. S.

Beaune

P. H.

-1989-. Modulation of cytochrome P450 isozymes in human liver, by ethanol and drug intake. Eur J Clin Invest 19, 549–55.

65.

Quattrochi

L. C.

Tukey

R. H.

-1993-. Nuclear uptake of the Ah -dioxin- receptor in response to omeprazole: Transcriptional activation of the human CYP1A1 gene. Mol Pharmacol 43, 504–8.

66.

Ramadoss

Perdew

G. H.

-2004-. Use of 2-azido-3-[125I]iodo-7,8-dibromodibenzo-p-dioxin as a probe to determine the relative ligand affinity of human versus mouse aryl hydrocarbon receptor in cultured cells. Mol Pharmacol 66, 129–36.

67.

Raucy

Warfe

Yueh

M. F.

Allen

S. W.

-2002-. A cell-based reporter gene assay for determining induction of CYP3A4 in a high-volume system. J Pharmacol Exp Ther 303, 412–23.

68.

Ripp

S. L.

Mills

J. B.

Fahmi

O. A.

Trevena

K. A.

Liras

J. L.

Maurer

T. S.

de Morais

S. M.

-2006-. Use of immortalized human hepatocytes to predict the magnitude of clinical drug–drug interactions caused by CYP3A4 induction. Drug Metab Dispos 34, 1742–48.

69.

Rizzo

Padoin

Palombo

Scherrmann

J. M.

Girre

-1996-. Omeprazole and lansoprazole are not inducers of cytochrome P4501A2 under conventional therapeutic conditions. Eur J Clin Pharmacol 49, 491–95.

70.

Roffey

S. J.

Cole

Comby

Gibson

Jezequel

S. G.

Nedderman

A. N.

Smith

D. A.

Walker

D. K.

Wood

-2003-. The disposition of voriconazole in mouse, rat, rabbit, guinea pig, dog, and human. Drug Metab Dispos 31, 731–41.

71.

Ronis

M. J. J.

Ingelman-Sundberg

-1999-. Induction of human drug-metabolizing enzymes: Mechanisms and implications. In Handbook of Drug Metabolism. - Woolf

T. F.

, ed.-, pp. 239–62, Marcel Dekker, Inc., New York, NY.

72.

Ronis

M. J. J.

Lindros

K. O.

Ingelman-Sundberg

-1996-. The CYP2E family.76. Cytochrome P450: Metabolic and Toxicological Aspects - Ioannides

, ed.-, pp. 211–39, CRC Press, Boca Raton, FL.

73.

Rost

K. L.

Brosicke

Brockmoller

Scheffler

Helge

Roots

-1992-. Increase of cytochrome P450IA2 activity by omeprazole: Evidence by the 13C-[N-3-methyl]-caffeine breath test in poor and extensive metabolizers of S-mephenytoin. Clin Pharmacol Ther 52, 170–80.

74.

Rost

K. L.

Brosicke

Heinemeyer

Roots

-1994-. Specific and dose-dependent enzyme induction by omeprazole in human beings. Hepatology 20, 1204–12.

75.

Roth

R. A.

Ganey

P. E.

-2010-. Intrinsic versus idiosyncratic drug-induced hepatotoxicity—two villains or one? J Pharmacol Exp Ther 332-3-, 692–97.

76.

Ruschitzka

Meier

P. J.

Turina

Luscher

T. F.

Noll

-2000-. Acute heart transplant rejection due to Saint John’s wort. Lancet 355, 548–49.

77.

Sahi

Black

C. B.

Hamilton

G. A.

Zheng

Jolley

Rose

K. A.

Gilbert

LeCluyse

E. L.

Sinz

M. W.

-2003-. Comparative effects of thiazolidinediones on in vitro P450 enzyme induction and inhibition. Drug Metab Dispos 31, 439–46.

78.

Scheer

Ross

Rode

Zevnik

Niehaves

Faust

Wolf

C. R.

-2008-. A novel panel of mouse models to evaluate the role of human pregnane X receptor and constitutive androstane receptor in drug response. J Clin Invest 118, 3228–39.

79.

Schoedel

K. A.

Tyndale

R. F.

-2003-. Induction of nicotine-metabolizing CYP2B1 by ethanol and ethanol-metabolizing CYP2E1 by nicotine: Summary and implications. Biochim Biophys Acta 1619, 283–90.

80.

Sciele Pharma. Cognex - Prescribing Information. Available at: http://www.sciele.com/PI/Cognex%20PI.pdf. Accessed November 2009.

81.

Seawright

A. A.

McLean

A. E.

-1967-. The effect of diet on carbon tetrachloride metabolism. Biochem J 105, 1055–60.

82.

Seeff

L. B.

Cuccherini

B. A.

Zimmerman

H. J.

Adler

Benjamin

S. B.

-1986-. Acetaminophen hepatotoxicity in alcoholics. A therapeutic misadventure. Ann Intern Med 104, 399–404.

83.

Senior

J. R.

-2007-. Regulatory perspectives. In Drug-Induced Liver Disease, 2nd edition - Kaplowitz

DeLeve

L. D.

, eds.-, pp. 771–89, Informa Healthcare, New York, NY.

84.

Sessa

Vigano

Grasselli

Trigo

Marimon

Llado

Locatelli

Ielmini

Marsoni

Gianni

-2006-. Phase I clinical and pharmacological evaluation of the multi-tyrosine kinase inhibitor SU006668 by chronic oral dosing. Eur J Cancer 42, 171–78.

85.

Shou

Hayashi

Pan

Morrissey

Skiles

G. L.

-2008-. Modeling, prediction, and in vitro in vivo correlation of CYP3A4 induction. Drug Metab Dispos 36, 2355–70.

86.

Shuldiner

A. R.

O’Connell

J. R.

Bliden

K. P.

Gandhi

Ryan

Horenstein

R. B.

Damcott

C. M.

Pakyz

Tantry

U. S.

Gibson

Pollin

T. I.

Post

Parsa

Mitchell

B. D.

Faraday

Herzog

Gurbel

P. A.

-2009-. Association of cytochrome P450 2C19 genotype with the antiplatelet effect and clinical efficacy of clopidogrel therapy. JAMA 302, 849–57.

87.

Simonsson

U. S.

Jansson

Hai

T. N.

Huong

D. X.

Tybring

Ashton

-2003-. Artemisinin autoinduction is caused by involvement of cytochrome P450 2B6 but not 2C9. Clin Pharmacol Ther 74, 32–43.

88.

Smith

D. A.

-2000-. Induction and drug development. Eur J Pharm Sci 11, 185–89.

89.

Smith

D. A.

Obach

R. S.

-2005-. Seeing through the mist: Abundance versus percentage. Commentary on metabolites in safety testing. Drug Metab Dispos 33, 1409–17.

90.

Smith

D. A.

Obach

R. S.

-2009-. Metabolites in safety testing -MIST-: Considerations of mechanisms of toxicity with dose, abundance, and duration of treatment. Chem Res Toxicol 22, 267–79.

91.

Soars

M. G.

Petullo

D. M.

Eckstein

J. A.

Kasper

S. C.

Wrighton

S. A.

-2004-. An assessment of UDP-glucuronosyltransferase induction using primary human hepatocytes. Drug Metab Dispos 32, 140–48.

92.

Spaldin

Madden

Adams

D. A.

Edwards

R. J.

Davies

D. S.

Park

B. K.

-1995-. Determination of human hepatic cytochrome P4501A2 activity in vitro use of tacrine as an isoenzyme-specific probe. Drug Metab Dispos 23, 929–34.

93.

Takahashi

Lasker

J. M.

Rosman

A. S.

Lieber

C. S.

-1993-. Induction of cytochrome P-4502E1 in the human liver by ethanol is caused by a corresponding increase in encoding messenger RNA. Hepatology 17, 236–45.

94.

Thummel

K. E.

Slattery

J. T.

Chien

J. Y.

Nelson

S. D.

Lown

K. E.

Watkins

P. B.

-2000-. Ethanol and production of the hepatotoxic metabolite of acetaminophen in healthy adults. Clin Pharmacol Ther 67, 591–99.

95.

Tucker

G. T.

Houston

J. B.

Huang

S. M.

-2001-. Optimizing drug development: Strategies to assess drug metabolism/transporter interaction potential—toward a consensus. Pharm Res 18, 1071–80.

96.

U.S. Department of Health Human Services F. D. A. Center for Drug Evaluation Research -CDER- -2006-. Guidance for Industry Drug Interaction Studies - Study Design, Data Analysis, and Implications for Dosing and Labeling, Draft Guidance. http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM072101.pdf. Accessed November 2009.

97.

U.S. Department of Health Human Services F. D. A. Center for Drug Evaluation Research -CDER- - 2008-. Safety Testing of Drug Metabolites. http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm079266.pdf. Accessed November 2009.

98.

Uetrecht

-2007-. Idiosyncratic drug reactions: current understanding. Annu Rev Pharmacol Toxicol 47, 513–39.

99.

Venkatesan

-1992-. Pharmacokinetic drug interactions with rifampicin. Clin Pharmacokinet 22, 47–65.

100.

Waterman

C. L.

Currie

R. A.

Cottrell

L. A.

Dow

Wright

Waterfield

C. J.

Griffin

J. L.

-2010- An integrated functional genomic study of acute phenobarbital exposure in the rat. BMC Genomics 11, 9.

101.

Watkins

P. B.

Zimmerman

H. J.

Knapp

M. J.

Gracon

S. I.

Lewis

K. W.

-1994-. Hepatotoxic effects of tacrine administration in patients with Alzheimer’s disease. JAMA 271, 992–98.

102.

Williams

J. A.

Hyland

Jones

B. C.

Smith

D. A.

Hurst

Goosen

T. C.

Peterkin

Koup

J. R.

Ball

S. E.

-2004-. Drug–drug interactions for UDP-glucuronosyltransferase substrates: A pharmacokinetic explanation for typically observed low exposure -AUCi/AUC- ratios. Drug Metab Dispos 32, 1201–8.

103.

Wolkenstein

Tan

Lecoeur

Wechsler

Garcia-Martin

Charue

Bagot

Beaune

-1998-. Covalent binding of carbamazepine reactive metabolites to P450 isoforms present in the skin. Chem Biol Interact 113, 39–50.

104.

Wrighton

S. A.

Schuetz

E. G.

Watkins

P. B.

Maurel

Barwick

Bailey

B. S.

Hartle

H. T.

Young

Guzelian

-1985-. Demonstration in multiple species of inducible hepatic cytochromes P-450 and their mRNAs related to the glucocorticoid-inducible cytochrome P-450 of the rat. Mol Pharmacol 28, 312–21.

105.

Xiaodong

Gatti

Bartoli

Cipolla

Crema

Perucca

-1994-. Omeprazole does not enhance the metabolism of phenacetin, a marker of CYP1A2 activity, in healthy volunteers. Ther Drug Monit 16, 248–50.

Hepatic Drug-Metabolizing Enzyme Induction and Implications for Preclinical and Clinical Risk Assessment

Abstract

Keywords

Introduction

Screening Tests for Human CYP Induction with Implications for Risk Assessment

In Vitro Models

Animal Models

Clinical Implications and Toxicity Resulting From Hepatic DME Induction

Hepatic DME Induction Leading to Increased Metabolites Unique to Humans versus Preclinical Animal Species

Risk Assessment and Risk Mitigation

Hazard Identification and Weight of Evidence

Footnotes

Abbreviations

References