Go to Home Page

Psychopharmacology Bulletin 30(2):251-259, 1994.

 

Inhibition of Hepatic P-450 Isoenzymes by Serotonin Selective

Reuptake Inhibitors: In Vitro and In Vivo Findings and Their

Implications for Patient Care

 

Sheldon H. Preskorn, M.D., and Ryan D. Magnus, M.D.

 

Department of Psychiatry, University of Kansas School of Medicine, and Psychiatric Research Institute, Wichita, KS.

Reprint requests: Dr. Sheldon H. Preskorn, Department of Psychiatry, University of Kansas School of Medicine in Wichita, 1010 North Kansas, Wichita, KS 67214.

Abstract

The effect on hepatic isoenzymes is emerging as the major clinically important distinguishing characteristic among the selective serotonin reuptake inhibitors (SSRIs). Although this fact has only recently gained widespread attention, the knowledge that some SSRIs inhibit hepatic metabolism dates back almost 20 years. This paper will first provide an overview of hepatic isoenzymes and then present the history and our current understanding of the effects of different SSRIs on different hepatic isoenzymes. Moat of the attention in this area has focused on drug-drug Interactions. This paper will also review recent work indicating that genetically determined differences in hepatic isoenzymes function can be risk factors in the development of a variety of diseases. The possible implication of this work relative to the long-term use of SSRIs will be discussed.

Over the last several years, there has been growing interest in the effects of drugs on hepatic isoenzyme function. This interest has been stimulated by reports of serious toxic reactions in patients whose clearance of TCAs and terfenadine was delayed by selective serotonin reuptake inhibitors (SSRIs) such as fluoxetine and by antifungal agents such as ketoconazole respectively. This paper will review the history and our current understanding of this area.

 

Hepatic Isoenzymes

There are two major classes of P450 isoenzymes: those that catalyze the formation of a variety of endogenous substances (e g., steroids, prostaglandins, fatty acids) and those that metabolize a wide range of foreign chemicals (e g., drugs, environmental pollutants, natural plant and animal products, and alcohols). The metabolism of foreign substances may produce metabolites implicated in initiation and progression of various disease (e g., tumors) (Nebert et al. 1991; Nelson et al. 1993). The former are predominantly located in mitochondria and the latter predominantly in the smooth endoplasmic reticulum. There is modest overlap between these two groups. Out of 26 gene families so far described in animal and plant species, 11 families comprised of 14 subfamilies have been identified so far in man (Table 1). Of these 11, 3 families (1, 2, and 3) have so far been implicated in most drug metabolism. Each family and subfamily is defined by the degree of similarity in the amino acid sequence of the isoenzyme. Thus, there is structural similarity between these isoenzymes such that a drug that affects one may affect others. For example, fluoxetine appears to inhibit at least three hepatic isoenzymes from different subfamilies: 2D6, 3A3/4, and one or more of the 2C series (Table 2).

Our knowledge is rudimentary but expanding rapidly with regard to what are the substrates for these enzymes in terms of both drugs and naturally occurring substances. Most of our current knowledge is restricted to drugs rather than to other xenobiotics, even though the latter may also have considerable health consequences.

 

TABLE 1. Human Hepatic Isoenzymes as Classified

By Family, Subfamily and Gene*

1A1

1A2

 2A6

2A7

2B6

2C8

2C9

2C9

2C18

2C19

2D6

2F1

2E1

 3A3/4

3A5

3A7

 4A9

4B1

4F2

4F3

 5 7 11A1

11B1

11B2

 17 19 21A2 27

*Key to Classification

  1. The first Arabic numeral represents the family.
  2. The following alphabetic letter represents the subfamily.
  3. The second Arabic numeral represents the individual gene within the subfamily.

Adapted from Nelson et al. 1993

 

The clearance of many drugs and of xenobiotics is dependent on P450 enzyme-mediated biotransformation into polar metabolites, which are then filtered through the kidneys. Such biotransformation is often the rate-limiting step and may be dependent on the functional status of a single enzyme. If the enzyme is not functional because of genetic deficiency, concomitant ingestion of an inhibitor, or concurrent disease, then the target's (i.e., drug or xenobiotic) clearance can be appreciably delayed, resulting in accumulation substantially greater than expected and a prolongation of its half-life due to the fact that its elimination becomes dependent on direct filtration of the parent drug or on biotransformation by enzymes that have low affinity for the substrate. This scenario can have profound consequences in terms of the patient's response to the target (e.g., a drug), such as serious toxicity if that drug has a narrow therapeutic index or reduced efficacy, if efficacy is in part or wholly mediated by the metabolites rather than by the parent drug.

 

Serotonin Selective Reuptake Inhibitors (SSRI's) and Hepatic Isoenzyme Function

There has been a growing awareness that specific SSRIs can inhibit specific P450 isoenzymes. The inhibition produced by SSRIs is competitive and reversible in contrast to suicidal (i.e., noncompetitive and irreversible) inhibition such as the inhibition of monoamine oxidase produced by drugs (such as phenelzine and tranylcypromine). Hence, the degree of inhibition is a function of several variables: The affinity of the SSRI for the enzyme x its concentration at the enzyme in relation to the affinity of the substrate (i.e., drug, other xenobiotic, foreign chemical, or intrinsic substance) whose metabolism and clearance is being inhibited and the concentration of this substrate at the enzyme. This concept is central to understanding the rest of the paper.

The experience with fluoxetine demonstrates how long the delay can be between identification of a potential problem and its clarification (Table 2). In 1976, fluoxetine was reported to inhibit the metabolism of hexobarbital and ethinamate in rats (Fuller et al. 1976). In 1988, the first cases of a serious interaction between fluoxetine and TCAs were reported from our department. The addition of fluoxetine to patients on stable doses of TCAs resulted in an appreciable increase in TCA plasma levels (Vaughan 1988). The first report of this phenomenon came 12 years after the initial discovery of an effect of fluoxetine on the hepatic metabolism of other drugs and less than I year after marketing. During that year of marketing, over S00,000 people were exposed to this risk. Fortunately, therapeutic drug monitoring in the case of TCAs was available to simultaneously detect the phenomenon and reduce the likelihood of serious adverse outcome by permitting the clinician to adjust the dose to compensate for the reduction in drug clearance induced by fluoxetine Despite this report and the ones that followed, several years elapsed between identification of the problem and the cause: fluoxetine -induced inhibition of the hepatic isoenzyme 2D6 (Crewe et al. 1992).

 

TABLE 2. History of P450 Isoenzymes and SSRI’s.

Year Finding Reference
1976 Fluoxetine inhibits the metabolism of barbiturates in rats Fuller et al. 1976
1988 Fluoxetine is marketed  
1988 Fluoxetine reported to increase plasma levels of TCAs Vaughan 1988
1988 Fluoxetine reported to decrease clearance of diazepam suggesting an effect on P450 2C series. Lemberger et al. 1988
1991 SSRI's reported to inhibit P450 2D6 in vitro Crewe et al. 1992
1991 Fluoxetine reported to inhibit metabolism of alprazolam indicating effects on P450 3A3/4 Lasher et al. 1991 Greenblatt et al. 1992
1993 Fluoxetine demonstrated in vitro von moltke et al.
1994 Fluoxetine demonstrated in vitro to inhibit alprazolam probably via an effect on 3A3/4 von Moltke et al.
1994a
SSRI=serotonin selective reuptake inhibitors
TCA=tricyclic antidepressants

Copyright S. Preskorn.

 

Since that time, significant progress has been made in understanding the phenomenon. Whereas there was previously debate as to how common the interaction was, it is now recognized to occur in the vast majority of the population if they take tricyclic antidepressants or similar drugs dependent on 2D6 for their elimination in the presence of fluoxetine . The hepatic isoenzyme 2D6 is the rate-limiting enzyme for the clearance of drugs like TCAs in over 95 percent of Caucasians and 99 percent of Asians (Alvan et al. 1990; Meyer 1990). At the minimally effective, recommended dose of fluoxetine (20 mg/day), individuals who combine these drugs become phenocopies of individuals genetically deficient of 2D6 (Preskorn et al. 1994). On conventional doses of TCAs, such patients will develop excessively high concentrations (i.e., 450 ng/mL). Such concentrations can cause delirium, seizures, heart blocks, and sudden death (Preskorn 1993c; Preskorn & Fast 1991,1992; Preskorn & Jerkovich 1990).

The concern is more than just with fluoxetine and with fluoxetine it is more than just its effects on 2D6. Different SSRIs inhibit different hepatic isoenzymes to varying degrees. fluoxetine inhibits at least three hepatic isoenzymes: 2D6 (Lemberger et al.1988; von Moltke et al. 1994a) (Table 4), 3A3/4 (Greenblatt et al. 1992; Lasher et al. 1991; von Moltke et al. l994b) and probably one or more of the 2C series (Lemberger et al.1988). The effect on 3A3/4 is suggested by the fact that fluoxetine inhibits the clearance of alprozolam. In a similar way, the effect on a 2C enzyme is suggested by the original finding in rodents since the barbiturates are metabolized by the 2C series (Fuller et al. 1976). fluoxetine has also been found to delay in humans the clearance of diazepam which appears to be dependent on one of the members of the P4S0 2C subfamily. The effect on 2D6 is substantially greater than the effect on the other two enzymes at fluoxetine's usually effective minimum dose of 20 mg/day (Bertilsson et al. 1989).

The magnitude of the effect of fluoxetine at the usually effective, minimum dose (20 mg/day) on 3A3/4 and the enzymes responsible for the metabolism of alprozolam and diazopam are not known. The problem is the long half-life of fluoxetine and its active metabolite, norfluoxetine . In the drug interaction studies published, volunteers have generally received the fluoxetine for less than 10 days. Since it takes I to 2 months to reach steady state of norfluoxetine on the usually effective minimum dose (i.c, 20 mg/day) and longer at higher doses, these reports underestimate the full effect of fluoxetine on the clearance of drugs dependent on the functional integrity of these enzymes. The degree of hepatic enzyme inhibition is dependent on the concentration of fluoxetine and norfluoxetine (Table 3) (Preskorn et al.1994; von Moltke et al., 1994b). Since fluoxetine inhibits its own clearance and thus has nonlinear pharmacokinetics (Preskorn 1993a), higher doses of fluoxetine will produce a disproportionately greater degree of enzyme inhibition until the enzyme is completely inhibited. The long half-life of norfluoxetine means the effect on the hepatic isoenzymes will persist for a considerable period of time after fluoxetine has been discontinued. In the case of 2D6, the enzyme was still substantially inhibited 3 weeks after fluoxetine discontinuation in volunteers who had received 20 mg/day of fluoxetine for only 3 weeks (Preskorn et al. 1994). The autoinhibition of fluoxetine clearance also means that its half-life will be longer at higher doses. Thus, the effect of fluoxetine on hepatic isoenzymes would be both greater and more prolonged at higher doses (Preskorn et al. 1994). The clinician must keep these facts in mind when treating patients with drugs dependent on these hepatic isoenzymes for their clearance, even many weeks after the discontinuation of fluoxetine .

Paroxetine is the most potent inhibitor of 2D6 in vitro (Table 4). However, it produces approximately the same degree of in vivo inhibition in humans as does fluoxetine at each drug's respective usually effective, minimum antidepressant dose (i.e., 20 mg/day) due to the difference in drug concentrations produced by that dose of those two drugs: approximately 40 ng/mL of Paroxetine vs. approximately 22S ng/mL of fluoxetine plus norfluoxetine (Preskorn 1993b; Preskorn et al. 1994). The degree of enzyme inhibition is a function of the potency of the drug for inhibiting the enzyme x the drug concentration achieved by a given dose. Like fluoxetine , Paroxetine inhibits its own clearance so that the half-life is prolonged at higher doses and there is a disproportionate increase in Paroxetine concentrations with dose increases (Preskorn 1993a). There has not been sufficient work done to determine whether Paroxetin affects 3A3/4 or the 2C series.

 

TABLE 3. Relationship Between Drug Response and Pharmacodynamics and Pharmacokinetics.

RESPONSE = Potency for mechanism of action x Concentration at the effector site.

OUTCOME = Pharmacodynamics x Pharmacokinetics

Copyright S. Preskorn

 

Sertraline does inhibit 2D6 in vitro but is the weakest of these three SSRIs in this regard (Table 4). In addition, plasma levels of sertraline are the lowest of these three SSRI's at its usually effective, minimum dose (i.e., 50 mg/day) (Preskorn 1993b; Preskorn et al. 1994). These two facts appear to explain why sertraline at this dose produces substantially less in vivo inhibition of the enzyme in volunteers at its usually effective, minimum dose than do fluoxetine and Paroxetine (Table 5). At higher concentrations, sertraline would be expected to produce a greater degree of enzyme inhibition (Preskorn et al. 1994). Since sertraline does not inhibit its clearance, the half-life of this drug does not change over its clinically relevant dose range, and dose increases produce proportional changes in drug concentration (Preskorn 1993a). Sertraline is a less potent in vitro inhibitor of 3A3/4 than is norfluoxetine (von Moltke et al. 1994b). For the same reasons as with 2D6, sertraline would be expected to produce less inhibition of this enzyme in vivo. Sertraline produces a negligible change in diazepam clearance in normal volunteers suggesting that sertraline is a less potent inhibitor of the 2C isoenzyme responsible for the biotransformation of diazepam than is fluoxetine (Gardner et al., in press).

Fluvoxamine is an even weaker in vitro inhibitor of 2D6 than is sertraline. The available data also suggests that it does not produce clinically meaningful inhibition of 2D6 in vivo (Crewe et al. 1992). Its effects on 3A3/4 and the 2C series have not been adequately studied to make any statement. In contrast to the other three SSRIs, fluvoxamine is a potent inhibitor of the hepatic isoenzyme 1A2 in vitro (Brosen et al.1993b). The usually effective, minimum dose of fluvoxamine produces clinically meaningful in vivo inhibition of this enzyme in humans (Bertschy et al. 1991; Spina et al. 1992).

The fact that several SSRIs share both the ability to inhibit 2D6 and the neuronal uptake pump for serotonin suggests that there may be a structural similarity between these two mechanisms. However, there is a difference in the rank order of the potency of the drugs for this effect. Based on in vitro studies using hepatic microsomes, the rank order for the inhibition of 2D6 for SSRIs marketed in the United States is: Paroxetine > fluoxetine > > sertraline (Table 4). Based on in vitro studies using rat brain synaptosomes, the rank order for the inhibition of the neuronal uptake for serotonin is: Paroxetine = sertraline > > fluoxetine (Shank et al. 1988). Citalopram, an SSRI marketed in Europe, is 30 times weaker than paroxetine as an inhibitor of 2D6 but is only 10 times weaker than paroxetine as an inhibitor of the neuronal uptake pump for serotonin (Hyttel & Larsen 1985).

 

TABLE 4. The Relative Potency (Ki, um) of Three Different Selective Serotonin Reuptake Inhibitors (SSRIs) and Their Metabolites for Inhibiting the Functional Integrity of the Hepatic Isoenzyme 2D6.

SSRI and metabolite Crewe et al

1992.

 Skjelbo et al.

1992

 von Moltke et al.

1993

 Otton et al.

1994

 Otton et al.

1993
Paroxetine .15 .36 - .065 -
M2 .5 - - - -
Fluoxetine .6 .92 3 .15 .17
Norfluoxetine .45 .33 3.5 - .19
Sertraline .7 - 22.7 1.2 1.5
Desmethylsertraline - - 16.9 - -

NOTE: All scores based on in vitro studies using human hapatic microsomes.

Copyright S. Preskorn

 

TABLE 5. Variables Determining the Magnitude of Effect on P450 2D6.

In vitro potenciesa

Paroxetine > fluoxetine > Sertraline

Plasma drug concentrations on minimum, effective doseb

Paroxetine on 20 mg/day . 40 ng/mL

fluoxetine on 20 mg/day, 2W ng/mL

Sertraline on 50 mg/day = 25 ng/mL

Type of pharmacokineticsc

Paroxetine - nonlinear

fluoxetine - nonlinear

Sertraline - linear

In vivo effect on desipramine (DMI)d

Paroxetine , 20 mg/day -

400% prorogation of DMI clearance

fluoxetine , 20 mg/day -

400% prorogation of DMI clearance

Sertraline, 50 mg/day

30% prorogation of DMI clearance

aSee table 4

bPlasma levels in young healthy volunteers. Plasma level9 of Paroxetine are considerably higher in healthy elderly (265 years). Minimal effect is seen with sertraline (Preskorn 1993b). fluoxetine has not been adequately studied.

cPreskorn 1993a

dBrosen et al. 1993a; Preskorn et al. t994.

Copyright S. Preskorn

 

Studies demonstrating the in vivo differences between fluoxetine , sertraline, and paroxetine have used the tricyclic antidepressant, desipramine (DMI), as the model substrate for the enzyme (Brosen et al. 1992; Preskorn et al. (1994). Beyond demonstrating that there is a differential effect of these three SSRIs on the clearance of DMI at each SSRI's respective usually effective, minimum dose, the results with DMI are a reflection of the functional inhibition of 2D6 activity for any substrate (e.g., another drug or a xenobiotic) which depends on 2D6 mediated biotransformation for its clearance from the body. The question then is: What is the absolute magnitude of the 2D6 inhibition?

The results from the normal volunteer studies of DMI suggest that paroxetine and fluoxetine at 20 mg/day produce substantial inhibition of 2D6 (Brosen et al. 1993b; Preskorn et al. 1994). This conclusion is based on the observation that almost the same DMI plasma level is achieved per mg/day DMI dose in the average patient concomitantly being treated with these two SSRIs at this dose as are achieved in patients who are genetically deficient in the enzyme (Brosen et al.1993a; Preskorn 1993a, Preskorn et al. 1994). In vitro modeling supports the same conclusion suggesting that the concentrations of fluoxetine and norfluoxetine that would be expected to be achieved on 20 mg/day should produce 80-o inhibition of 2D6. In contrast, combined levels of sertraline and desmethylsertraline that would occur as a result of treatment with 50 mg/day should produce approximately a 15% inhibition of the enzyme (von Moltke et al. 1994a).

 

Examples of Types of Increased Drug Toxicity Associated With Specific

Hepatic Isoenzyme Deficiency

The concern about the effect of SSRIs and other drugs (e.g., antifungal agents such as ketoconazole) on hepatic isoenzymes has focused on the immediate problem, decreased clearance of another drug leading to the possibility of acute toxic accumulations of that drug on conventional doses. To put such interactions in perspective, it is useful to review the work that has been done to understand genetically determined deficiency in hepatic isoenzymes in increasing the risk of both acute and more chronic drug toxicity.

Due to a mutation, some individuals either: (1) produce an abnormal form of a specific hepatic isoenzyme in terms of its affinity for the substrate, its maximal velocity, or its stereoselectivity for the reaction, or (2) have a decreased rate of synthesis, and/or (3) an increased rate of degradation of the enzyme. Such mutations produce individuals who are functionally deficient in the enzyme. As would be expected, such individuals have been found to be at increased risk for acute drug toxicity when given drugs that are dependent on the affected enzyme for their clearance and that have narrow therapeutic indexes.

The ability to detect such serious interactions is directly related to the rapidity of toxicity onset and its severity. The ability to make the casual connection between reduced clearance and drug toxicity is also helped when the investigator can measure the drug level and demonstrate that the toxicity is associated with unexpectedly high levels of the drug or its metabolites. There are multiple examples of such toxicity. Perhaps the best known in psychiatry has already been discussed, the toxicity due to the accumulation of excessive concentrations of TCAs in individuals functionally deficient in P4SO 2D6 activity (Preskorn & Fast 1991, 1992; Preskorn & Jerkovich 1990).

In addition to acute toxic drug reactions, differences in hepatic isoenzyme function are associated with adverse consequences that have longer latencies and therefore are more difficult to detect. For example, individuals functionally deficient in P450 2D6 activity are at substantially greater risk for peripheral neuropathy or hepatotoxicity when they use perhexiline for antianginal therapy (Cooper et al.1984). This fact severely restricts the use of this otherwise effective agent in these patients. There are a number of other examples of delayed toxicity caused by reduced drug clearance. There is an increased incidence of isoniazid-induced peripheral neuropathy in individuals who are slow N-acetylators (SAs) (Devadatta et al. 1960; Hughes et al. 1954). N-acetylation controls the rate of elimination of isoniazid. Hydralazine-induced lupus erythematous is overwhelmingly a disease of SAs (Uetrecht & Woosley 1991).

Toxicity may be due to either the delayed clearance of a toxic drug or its metabolite or the increased production of such a metabolite. In the latter case, an aberrant, or usually minor enzymatic pathway, is used to clear a drug because a deficiency exists in the usual pathway. For example, SA status determines the incidence and severity of toxic reactions in patients being treated with salicylazopyridine. Sulfapyridine is produced in the gut by bacterial enzymes that cleave the parent drug, salicylazopyridine, at the azo linkage. The sulfapyridine is then absorbed and acetylated as a step in its elimination. Higher concentrations of this toxic metabolite therefore accumulate in SA individuals, resulting in the increased incidence and severity of adverse effects (Schroeder & Evans 1972).

 

Examples of Increased Toxicity Caused by Environmental Exposure Associated With Specific Hepatic Isoenzyme Deficiency

The health implications of these findings extend beyond these findings. The cytochrome P450 enzymes play an important role in the metabolism of many toxins, mutagens, and carcinogens (Caporaso et al. 1991; Guengerich 1988). Depending on the specific example, this metabolism can convert a protoxin, promutagen, or procarcinogen to their ultimate hazardous forms or may be responsible for clearing the hazardous form. In the former case, a deficiency in the enzyme could decrease the risk of exposure by decreasing the rate or extent of conversion to the hazardous form; in the latter case, the enzyme deficiency could increase the risk by decreasing the rate of clearance of the hazardous forms leading to prolonged exposure to higher concentrations. Not surprisingly, there is evidence that both of these scenarios occur with regard to specific environmental exposures and specific diseases. Drugs are analogous to other environmental agents (cg., xenobiotics, foreign chemicals) but are prescribed in pharmacological doses. A drug or its metabolites can be toxins depending on their pharmacological actions. By analogy, the finding that genetic deficiency in hepatic enzymes can increase the risk of drug toxicity on conventional doses raises the issue of whether such deficiency may also modify the risk of adverse consequences arising out of exposure to environmental toxins which are normally cleared by biotransformation via these enzymes.

As with drug toxicity, toxicity due to exposure to an environmental toxin can be either acute or delayed. Genetic selection should mitigate against propagation of a deficiency that would result in high risk of acute toxicity early in life. Hence, delayed toxicity would seem to be the more likely finding. As discussed above, it is naturally more difficult to establish a cause and effect relationship when toxicity is delayed. Nonetheless, a growing body of evidence suggests that late-onset toxicity does result from chronically high accumulation of environmental toxins resulting from delayed clearance of these toxins due to a functional deficiency in a hepatic isoenzyme needed for its biotransformation and elimination.

N-acetylation is an example of genetic polymorphism that has been identified as a risk factor in the development of a number of diseases. The risk of developing Gilbert's disease, diabetes mellitus, leprosy, and early-onset thyrotoxicosis differs substantially between rapid and slow acetylators (Evans 1989; Weber 1987).

Genetically determined deficiency in the functional activity of P450 2D6 has been suggested to be a risk factor for the development of early-onset Parkinson's disease (Barbeau et al. 1985). This hypothesis is quite controversial with negative as well as positive studies. A full review is beyond the scope of this paper but suffice to say that the results of such studies have to be carefully examined. For example, even a reportedly negative study (Steiger et al. 1992) found a statistically significant higher incidence of 2D6 deficiency in patients with Parkinson's disease versus controls (p < 0.01) and a significant inverse correlation between 2D6 activity and age of onset of the disease even though the rate of 2D6 deficiency in the young-onset group just failed to reach statistical significance (p= 0.05).

The potential link between genetically determined deficiency in 2D6 and Parkinson's disease has been spurred in part because MPTP, a dopamine neurotoxin capable of producing Parkinsonism, is metabolized by 2D6 (Fonne-Pfister et al. 1987). This substance can be an environmental contaminant particularly in areas with a high density of chemical manufacturing. Chronic long grade exposure can destroy central dopamine neurons. Impaired ability to detoxify this agent once consumed presumably could decrease the amount of exposure necessary to develop a clinically meaningful impairment of dopamine. This scenario is clearly speculative. Nonetheless, this hypothesis provides an interesting model for the potential role for hepatic enzyme deficiency as a risk factor in the development of other diseases.

There have been several observations indicating that hepatic enzyme genotype is a risk factor in the development of some forms of cancer. Slow acetylators have an increased incidence of transitional cell carcinoma of the urinary bladder, and carcinoma of the lung and breast. In contrast, individuals deficient in 2D6 are underrepresented in patients with certain types of carcinoma of the lung (Agundez et al. 1994; Hirvonen et al. 1992) and transitional cell carcinoma of the bladder (Kaisary et al. 1987). In the latter case, deficiency of 2D6 may protect against the risk of recurrence (Fleming et al. 1992). On the other hand, deficiency in 3A3/4 has been associated with a six-fold increase in the relative risk for an aggressive form of bladder cancer (personal communication, J. Porter).

This area of research is in its infancy. The bulk of studies are epidemiological which have several limitations. They often are underpowered (i.e., the size of the population studied is too small leading to the possibility of false negative results). The disease population and the control population may not be appropriately matched. An example is the study by Agundez and colleagues (1994), which reported a decreased incidence of lung cancer in individuals deficient in 2D6. The control population was 20 years younger than the disease population. Hence, they may not have lived long enough to develop a disease requiring a longer incubation period.

At best, epidemiological studies indicate whether an association is likely but not whether there is a cause and effect relationship. Even a strong association may simply reflect a linkage phenomenon (e g., the close association of the gene for the hepatic isoenzyme with an oncogene). In the case where there is a cause and effect relationship, the enzyme activity probably modifies the relative risk but is not the primary determinant. The individual must be exposed to the causative agent. If no exposure occurs, then no disease results. Conversely, sufficiently high and prolonged exposure to the causative agent will most likely produce the disease regardless of the enzyme activity.

The point is that there is sufficient evidence from such epidemiological studies to suspect an association between the functional activity of hepatic isoenzymes and the development of specific diseases. This work has implications for clinical psychopharmacology. If the altered incidence of these diseases is due to the alteration in formation or clearance of an environmental toxin, then a chronically administered drug which produces a phenocopy of the enzyme deficiency might be expected to produce a similarly altered incidence rate. This issue is pertinent to the use of psychiatric medications such as antidepressants given the recent emphasis on the need for long-term treatment for a substantial percentage of patients with recurrent major depression (Table 6).

 

Conclusion

There is a growing understanding of the effect of drugs on hepatic isoenzyme function. The selective serotonin reuptake inhibitors (SSRI's) have made this issue important to psychiatry. A major difference among this class is their differential effects on hepatic isoenzymes. These effects account for differences among these drugs in terms of their pharmacokinetics: their half-lives and whether they demonstrate linear vs. nonlinear pharmacokinetics over their clinically relevant dosing range. These differences are also important with regard to the potential for causing pharmacokinetic interactions with specific concomitantly prescribed drugs. Given the frequent recommendation for long-term antidepressant therapy, an area that requires further study is the potential long-term health consequences of substantially altering hepatic isoenzyme function.

 

TABLE 6. - Factors Which Increase the Potential Risk of Long-term Adverse Consequences of Hepatic Isoenzyme Inhibition

 

References

Nebert, DW; Nelson, DR; Coon, MJ; Estabrook, RW; Feyereisen, R; Fujii-Kuriyama, Y; Gonzalez, FJ; Guengerich, FP; Gunsalus, IC; Johnson, EF; Loper, JC; Sato, R; Waterman, MR; and Waxman, DJ. The P-450 superfamily: Update on new sequences, gene mapping, and recommended nomenclature. DNA Cell Biol. 10:1-14, 1991.

Nelson, DR; Kamataki, T; Waxman, DJ; Guengerich, P; Estabrook , RW; Feyereisen , R; Gonzalez, FJ ; Coon , M J ; Gunsalus, IC; Gotoh, O; Okuda, K; and Nebert, DW. The P450 superfamily: Update on new sequences, gene mapping, accession numbers, early trivial names of enzymes and nomenclature. DNA Cell BioL 12:1-51, 1993.

Otton, SV; Ball, SE; Cheung, SW; Inaba, T; and Sellers, EM. Comparative inhibition of the polymorphic enzyme CYP2D6 by venlafaxine and other 5HT uptake inhibitors Clin. Pharmacol. Ther. 55(2):141, 1994.

Otton, SV; Wu, D; Joffe, RT; Cheung, SW; and Sellers, EM. inhibition by fluoxetine of cytochrome P-450 2D6 activity. Clin. Pharmacol. Ther. 53:401-409, 1993.

Preskorn, SH. Pharmacokinetics of antidepressants: Why and how they are relative to treatment. J. Clin. Psychiatry 54:9 (Suppl.), 1993a.

Preskorn, SH. Recent pharmacologic advances in antidepressant therapy for the elderly. Am. J. Med 94(Suppl.5A).42-12, 1993b.

Preskorn, SH. Sudden death and tricyclic antidepressants (TCA): A rare adverse event linked to high TCA plasma levels. Nord J. Psychiatry47(Suppl. 30):49-56, 1993c.

Preskorn, SH; Alderman, J; Chung, M; Harrison, W; Messig, M; and Harris, S. Pharmacokinetics of desipramine coadministered with sertraline or fluoxetine . J. Clin. Psychopharmacol 14(2):90-98, 1994.

Preskorn, SH, and Fast, GA. Therapeutic drug monitoring for antidepressants: Efficacy, safety, and cost effectiveness. J. Clin. Psychiatry 52(6):23-33, 1991.

Preskorn, SH, and Fast, GA. Tricyclic antidepressant-induced seizures and plasma drug concentration. J. Clin. Psychiatry 53:160-162, 1992.

Preskorn, SH, and Jerkovich, GS. Central nervous system toxicity of tricyclic antidepressants: Phenomenology, course, risk factors, and role of therapeutic drug monitoring. J. Clin. Psychopharmacol. 10(2):88-95, 1990.

Schroeder, H, and Evans, DAP. Acetylator phenotype and adverse effects of sulphasalazine in healthy subjects. GUT 13:278-284, 1972.

Shank, RP; Vaught, JL; Pelley, KA; Sether, PE; McComsey, DF; and Maryanott, BE. McN-5652: A highly potent inhibitor of serotonin uptake. J. Pharmacol Exp. Ther. 247:1032-1038, 1988.

Skjelbo, E, and Brosen, K. Inhibitors of imipramine metabolism by human liver microsomes. Br. J. Clin. Pharmacol. 34:256-261, 1992.

Spina, E; Campo, GM; Avenoso, A; Pollicino, MA; and Caputi, AP. Interaction between fluvoxamine and imipramine/despramine in four patients. Ther. Drug Monit. 14:194-196, 1992.

Steiger, MJ; Lledo, P; Quinn, NP; Marsden, CD; Turner, P; and Jenner, PG. Debrisoquine hydroxylation in Parkinson's disease. Acta Neurol. Scand. 86:159-164, 1992.

Uetrecht, JP, and Woosley, RL. Acetylator phenotype and lupus erythematosus. Clin. Pharmacokinet. 6:118-134, 1981.

Vaughan, DA. Interaction of fluoxetine with tricyclic antidepressants. Am. J. Psychiatry 145:1478, 1988.

von Sloltke, LL; Greenblatt, DJ; Cotreau-Bibbo, MM; Duan, SX; Harmatz, IS; and Shader, Rl. Inhibition of desipramine hydroxylation in vitro by serotonin-reuptake-inhibitor antidepressants, and by quinidine and ketoconazole: A model system to predict drug interactions in vivo. J. PharmacoL E.rp. Ther. 268:1278-1283, 1994.

von Mloltke, LL; Greenblatt, DJ; Cotreau-Bibbo, MM; Harmatz, JS; and Shader, Rl. Inhibitors of alprazolam metabolism in vitro: Effect of serotonin-reuptake-inhibitor antidepressants, ketoconazole and quinidine. Br. J. Clin. Pharmacol, 1994b.

Weber, WW. The Acetylator Genes and Drug Response. New York: Oxford University Press, 1987.

 

Go to Home Page