Biotransformationbased Pharmacokinetic Interactions

Asthma Free Forever

Asthma Free Forever By Jerry Ericson

Get Instant Access

A number of prominent drug products have been withdrawn in recent years because of severe drug-drug interactions and despite preclinical safety assessment. Mibefradil, a novel calcium antagonist, for example, was approved in Switzerland in 1996 and was also launched in the U.S. in 1997 as well as in several other European countries. Shortly following its launch as an antihypertensive and antianginal agent, reports about serious pharmacokinetic and pharmacodynamic interactions with other drugs frequently administered to patients with cardiovascular diseases were noted. These interacting drugs are to a great extent metabolized by Cytochrome P45o (CYP45o)-dependent microsomal enzymes, including widely prescribed drugs like quinidine, digoxin, cyclosporin A, terfenadine, and metoprolol. In addition, reports on severe rhabdomyolysis in patients on mibefradil who were simultaneously receiving lovastatin or simvastatin were issued. Mibefradil was reported to mainly inhibit CYP2D6 and 3A4 isoenzymes. In 1998 the drug was withdrawn from the market due to the information gathered about the severity of drug-drug interactions in patients receiving mibefradil and other medications [3]. Another example of clinically important interactions between CYP3A4 inhibitors and drugs largely eliminated by oxidative biotransformation is between ketoconazole, itraconazole, clarithromycin, erythromycin, nefazodone, and ritonavir as inhibitors, when these are coadministered with terfenadine, astemizole, cisapride, or pimozide.

In that case, Torsades de pointes, a life-threatening ventricular arrhythmia associated with QT prolongation has been shown to occur as a consequence of decreased clearance of the arrhythmia-causing parent compound or metabolite [4]. Finally, a drug-drug interaction between sorivudine, an antiviral drug, and 5-fluorouracil, an anticancer drug, caused one of the most serious cases of toxicity ever seen in Japan. The interaction is based on the irreversible inhibition (mechanism-based inhibition) of dihydropyrimidine dehydrogenase, a rate limiting enzyme in the metabolism of 5-fluorouracil by a metabolite of sorivudine, which is formed by gut flora [5]. On the basis of these case reports on drug-drug interactions due to decreased metabolic clearance of the active compound and the clinical experience, several recommendations have been made for the regulatory assessment of new active substances with respect to drug-drug interactions. These include the requirement for a detailed understanding about the mechanism of biotransformation of the parent compound and its metabolites primarily by in vitro studies with human liver enzymes in which the potential for metabolic interactions with other drugs is outlined. This first screen then may serve as a start for identification of drugs that are commonly used in the target population and that may represent a particular risk by pharmacoepidemiological studies. Here, particular attention is to be put on drugs with "a high first-pass metabolism" and "a narrow therapeutic index." These may then be studied in interaction studies in the patient population or in healthy volunteers before their introduction into clinical practice. Particular attention needs to be put on the interpretation with respect to the severity of a drug-drug interaction. Here, not only the mean of the interaction effect, but also the observed and the theoretically conceivable extreme effects in individual subjects need to be addressed. In particular, the mibefradil case has shown that for drugs that are expected to be co-administered in the target population and that may represent a particular risk, a labelling in the product information indicating the possibility of an interaction should not be acceptable as a substitute for performing the appropriate interaction studies before introduction of the new drug into clinical practice.

Biotransformation-based drug-drug interactions may occur presystemically, i.e., at the level of the intestine and in the liver (gastrointestinal and hepatic first-pass effect) and thus may affect the bioavailability and the clearance of a drug. The intrinsic organ clearance is defined as:

where Vmax,■ and Kmj are the maximum reaction velocity and substrateenzyme affinity constant for the ith enzyme. Drug-drug interactions may affect intrinsic clearance. In the case of competitive enzyme inhibition, Km is increased, whereas for noncompetitive inhibition, a decrease in Vmax is noted. Enzyme induction, on the other hand, results in an increase of Vmax. In particular, for low hepatic extraction drugs (E<0.2), clearance is primarily dependent upon intrinsic clearance (enzyme activity) and not liver blood flow. Consequently for these drugs, small changes in intrinsic clearance, e.g., due to enzyme induction or inhibition, may result in severe changes of drug clearance. On the other hand, high hepatic-extraction drugs (E>0.6) have an intrinsic hepatic clearance which exceeds the hepatic blood flow. Clearance of these drugs is therefore primarily dependent on liver blood flow and not on intrinsic hepatic clearance. High ratios of the area under the curves in the presence and absence of an inhibitor are to be expected when the value of (1+I/K) is large, i.e., at high concentrations of a high affinity inhibitor, and/or when the fraction of the dose eliminated by a pathway which can be inhibited by the metabolic inhibitor is large. A particular issue is the relevance of I and K values for the likelihood of an in vivo drug-drug interaction. In the case of reversible inhibition, a drug-drug interaction (potential for in vivo inhibition) is considered "highly likely," if K<1 pM and I/K,>1 [6]. When K is between 1 and 50 pM and I/Kj equals 0.1-1, an in vivo interaction is deemed possible, and when K>50 pM and I/K<0.1 the potential of an in vivo

interaction is rather remote. Consequently, if the I/Kj value is larger than 0.3-1, it has been suggested to consider designing the appropriate in vivo drug interaction studies [7]. The principle has been depicted again schematically in Fig. 1. It needs to be pointed out though, that the zone of medium risk is a gray zone and the definition of universal cut-off values is not uniquely agreed upon by several researchers. Nevertheless, high I/K, values for a particular metabolic pathway suggest that the possibility of occurrence of a drug-drug interaction in vivo because it is likely that the inhibitor also inhibits other metabolic pathways which have not been identified yet. For mechanismbased inhibition, K, values<20 pM for the inhibitor have "likely" potential for in vivo inhibition, whereas K values in the range of 20 to 100 pM and >100 pM have "possible" and "remote" potential for causing an in vivo interaction, respectively. The principle has successfully been applied e.g., for the prediction of the absence of an interaction between warfarin and tenoxicam, both of which are eliminated by CYP2C9 [8]. Similarly, an in vivo interaction has been predicted between warfarin and lornoxicam [8], tolbutamide and sulfaphenazole, and triazolam and ketoconazole [7]. For the CYP2D6-mediated dehydration of sparteine and the interaction with the CYP2D6 inhibitor quinidine, the interaction between the CYP1A2 inhibitor ciprofloxacin and the CYP1A2 substrate caffeine, and the CYP3A4 substrate cyclosporin and the CYP3A4 inhibitor erythromycin as well as for the interaction between the CYP3A4 substrate terfenadine and the CYP3A4

FIGURE 1 Impact of [l]/Ki on the ratio of the AUC of substrate ([S]<KJin the presence and absence of a competitive inhibitor. The equation governing the relationship is: AUCi/AUC=1+[l]/Ki, where AUCi and AUC are the areas under the substrate concentration-time curve in the presence and absence of the inhibitor, respectively. The figure was redrawn according to Tucker et al. [14].

FIGURE 1 Impact of [l]/Ki on the ratio of the AUC of substrate ([S]<KJin the presence and absence of a competitive inhibitor. The equation governing the relationship is: AUCi/AUC=1+[l]/Ki, where AUCi and AUC are the areas under the substrate concentration-time curve in the presence and absence of the inhibitor, respectively. The figure was redrawn according to Tucker et al. [14].

inhibitor ketoconazole, the magnitude of the interactions was underpredicted by factors of approximately 2, 1.5, 1.3, and 7, respectively [7]. The reasons for this underprediction may include estimation errors for K,, the possibility that other elimination pathways may also be reduced by the inhibitor and the possible accumulation of the inhibitor in the liver. The latter leads to an underprediction of the inhibitor concentration at the site of metabolism, which may be the case when carrier-mediated transport processes promote the uptake of the inhibitor into hepatocytes, e.g., in the case of ciprofloxacin.

How to predict inhibitory effects of co-administered drugs on hepatic metabolism of other drugs?

The procedure for predicting the metabolic inhibition by one drug that is expected to be administered together with the study drug involves several steps. First, the metabolic pathway of the drug under consideration and possibly the P450 isozyme(s) most relevant for its degradation should be identified. This can be done either from metabolic pharmacokinetic drug interaction databases [9] or it can be determined experimentally e.g., by human P450 expression systems or by inhibition studies with human liver microsomes using P450 antibodies or inhibitors specific for each isozyme. A list of P450 isozymes and their inhibitors is given in Table 1. Secondly, pharmacokinetic data for the co-administered drug that possibly inhibits the isozyme responsible for the metabolism of the study drug are assembled and the maximum concentration of the co-administered inhibitor is estimated. Thirdly, the K of the inhibitor for the metabolism of the study drug is determined using e.g., human liver microsomes or human P450 expression systems and the I/K ratio is calculated. For more detailed information on in vitro metabolic methodology, see Chapter 5.

In addition to the selection of a particular in vitro model, particular probe substrates and inhibitors have to be chosen for the drug-drug interaction study. Table 1 is a compilation of suitable compounds for each of the human CYPs. These compounds currently present the most useful tools to provide in vitro enzyme-kinetic parameters with respect to the various CYP isoforms [10]. For a variety of reasons, e.g., not approved as a drug product and/or toxicity in humans, several of the compounds listed in Table 1 are not suitable for in vivo drug-drug interaction studies in humans. Therefore, Table 2 contains a list of probe substrates and inhibitors of CYP isoenzymes which may be used for in vivo studies in humans. The conduct of in vivo studies is most relevant to confirm positive outcomes of drug-drug interactions from in vitro findings and cases are known, in which compounds prove to be potent inhibitors of CYP isoenzymes in vitro in liver microsomes, yet have no inhibitory effect on the AUC of various probe substrates in vivo. This may, for example, be explained by the fact that microsomes are poor

Substrates

Inhibitors

2C19

2E1 3A4

Ethoxyresorufin, Phenacetin, Caffeine, Theophylline, Acetanilide, Methoxyresorufin Coumarin

S-Mephenytoin, Bupropion Paclitaxel

S-Warfarin, Diclofenac, (Tolbutamide) S-Mephenytoin, Omeprazole Bufuralol, Dextromethorphan, Metoprolol,

Debrisoquine, Codeine Chlorzoxazone, 4-Notrophenol, lauric acid Midazolam, Testosterone, Nifedipine, Felodipine, Cyclosporine, Terfenadine, Erythromycin, Simvastatin, Tacrolimus

Furafylline, a-Naphthoflavone1

Coumarin

Sertraline2

Troglitazone3

Sulphaphenazole

Ticlopidine,2 Nootkatone2

Quinidine

4-Methyl pyrazole Ketoconazole, Troleandomycin, Tacrolimus, Cyclosporin

1Can also activate and inhibit CYP3A4. 2Also inhibits CYP2D6. 3Also inhibits CYP2C9.

TABLE 2 In vivo Probe Substrates and Inhibitors for CYPs

CYP

Substrates

Inhibitors

1A2

Caffeine, Theophylline

Enoxacin, Ofloxacin

2B6

Bupropion

Sertraline2

2C8

Paclitaxel1

Troglitazone3

2C9

Tolbutamide, Flurbiprofen,

Sulphaphenazole, Fluconazole'

Diclofenac, Phenytoin

2C19

Mephenytoin, Omeprazole

Ticlopidine

2D6

Debrisoquine, Dextromethorphan,

Quinidine

Metoprolol, Desipramine, Codeine

2E1

Chlorzoxazone

Disulfiram

3A4

Midazolam, Erythromycin,

Grapefruit juice, Ketoconazole,1

Simvastatin, Atorvastatin Itraconazole

Simvastatin, Atorvastatin Itraconazole

1Cannot be administered to healthy volunteers.

2Also inhibits CYP2D6 at high doses exceeding 150 mg/day

3Also inhibits CYP2C9.

4Also moderately inhibits CYP3A4.

5Also an inhibitor of 2C19.

performers with respect to phase II metabolic reactions and the scavenging of potentially inhibitory phase I metabolites is not an issue in whole functional hepatocytes. An alternative to the use of very specific enzyme inhibitors in clinical studies is the application of inhibitors with broad inhibition specificity. Examples include Cimetidine (3A4, 2D6, 1A2, 2C9) and Ritonavir (3A4, 2D6, 2C9, 2C19). Furthermore, genetic polymorphisms need to be taken into account. The polymorphic variability of drug metabolism was empirically recognized before the P450 system was well understood. Slow and rapid acetylators of isoniazid were recognized in the 1950s. Glucose-6-phosphate dehydrogenase deficiency leading to hemolytic anemia was appreciated as a genetically based variation in drug metabolism. In the 1970s, Ziegler and Biggs [15] noted that African-American patients had significantly higher nortriptyline levels than did other patients, and these investigators assumed there were genetic differences [11]. The differences in nortriptyline metabolism are now believed to result from genetic polymorphisms related to 2D6, 2C9, and/or 2C19. CYP P450 polymorphisms known today are tabulated in Table 3. Due to very active research in this field, in particular in the area of genotyping or phenotyping of individuals with respect to P450 enzymes, it is expected that this list will continue to grow. Taking the information on the different metabolism capacities of individuals it may thus be possible to predict that only those individuals in whom a major metabolic pathway is inhibited may show profound drug-drug interactions. On the other hand, the same drug combination may be estimated as having no interactions, when administered to a subject who is genetically deficient with respect to the isoenzyme responsible for drug clearance.

In addition to enzyme inhibition, induction processes by some xenobiotics, both drugs and environmental substances such as cigarette smoke, may increase the synthesis of P450 proteins. This induction process may lead to decrease in circulating plasma levels of the parent drug administered and increase in the concentrations of metabolites produced and is one of the major underlying mechanisms for time-dependent pharmacokinetics. For example, co-administration of the potent inducers rifampin or nevirapine [12, 13], and methadone has led to opiate withdrawal symptoms. Cytochromes P450 3A4, 1A2, 2C9, 2C19, and 2E1 may all be induced. Important inducers are e.g., carbamazepine, oxcarbazepine, phenytoin, phenobarbital, rifampin, rifabutin, nevirapine, troglitazone, dexamethasone, prednisone, St. John's wort, and primidone (3A4), tobacco smoke, brussel sprouts, broccoli, cabbage and other cruciferous vegetables, charbroiled foods, e.g., burned meats (1A2), rifampin, phenytoin, secobarbital (2C9), rifampin (2C19), alcohol, and isoniazid [14-17].

TABLE 3 CYP P450 Polymorphisms [Cozza, Armstrong, 2001]

CYP

Polymorphisms

Type(s)

Phenotype substrate probe

Examples of drugs that warrant caution for PMs

2D6

Yes

PM, EM, UEM

Debrisoquin, dextromethorphan, sparteine

Antiarrhythmics, codeine, tramadol

3A4

Possible

Unknown

Cyclosporine, erythromycin, lidocain,

Unknown

midazolam

1A2

Probable

Unknown

Caffeine, theophylline

Unknown

2C9

Yes

PM, EM

Naproxen, phenytoin, tolbutamide

S-Warfarin

2C19

Yes

PM, EM

Proguanil

Cyclophosphamide, ifosfamide

2E1

Yes

Unclear

Chlorzoxazone

Anesthetic agents

Numerous examples of documented and clinically relevant drug-drug interactions exist with respect to enzyme induction. For example, induction of 3A4 by oxcarbazepine can induce the metabolism of oral contraceptives rendering them less effective [18]. Plasma concentrations of mirtazapine, a nonadrenergic and specific serotonergic antidepressant which is mainly metabolized by CYP 2D6 and CYP 3A4 are decreased by 60% following enzyme induction by carbamazepine [19]. Rifampicin and rifapentine induction can decrease plasma concentrations of protease inhibitors or non-nucleoside reverse transcriptase inhibitors which may lead to viral resistance (decreased sensitivity to the protease inhibitor or NNRTIs). St. John's wort was recently found to decrease mean trough plasma concentrations of indinavir by 81% (20), cyclosporine A by 43% (21), digoxin AUC by 25%, and trough concentrations by 33% (22). Interestingly, the effect of St. John's wort was only seen following chronic dosing of the hypericum extract and not after single dose, indicating that the mechanism of action is by induction of protein expression and not by direct competition with the concomitantly administered drug.

Was this article helpful?

0 0
Dealing With Asthma Naturally

Dealing With Asthma Naturally

Do You Suffer From ASTHMA Chronic asthma is a paralyzing, suffocating and socially isolating condition that can cause anxiety that can trigger even more attacks. Before you know it you are caught in a vicious cycle Put an end to the dependence on inhalers, buying expensive prescription drugs and avoidance of allergenic situations and animals. Get control of your life again and Deal With Asthma Naturally

Get My Free Ebook


Post a comment