Leif Bertilsson

A common novel CYP2C19 gene variant causes ultrarapid drug metabolism relevant for the drug response to proton pump inhibitors and antidepressants

Many drugs, including proton pump inhibitors and certain antidepressants, are metabolized by the polymorphic cytochrome P450 (CYP) 2C19 enzyme. A significant portion of extensive metabolizers do not reach appropriate drug levels, and our objective was to investigate any genetic background.

Geographical/interracial differences in polymorphic drug oxidation. Current state of knowledge of cytochromes P450 (CYP) 2D6 and 2C19.

SummaryThe isoenzymes which catalyse the polymorphic hydroxylations of debrisoquine/sparteine and S-mephenytoin are cytochromes P450 2D6 and P450 2C19 (CYP2D6 and CYP2C19), respectively. CYP2D6 is involved in the stereospecific metabolism of several important groups of drugs, for example antiarrhythmics, antidepressants and neuroleptics.About 7% of Caucasians but only 1 % of Orientals are poor metabolisers (PMs) of debrisoquine. The most common mutated allele CYP2D6B in Caucasian PMs is almost absent from their Oriental counterparts. On the other hand, the mean activity of CYP2D6 in Oriental extensive metabolisers (EMs) is lower than that in Caucasian EMs. This is due to the frequent distribution of a partially deficient CYP2D6 allele causing a Pro34 → Ser amino acid exchange in as many as 50% of Oriental alleles. This is the molecular genetic basis for the slower metabolism of antidepressants and neuroleptics observed in Oriental compared with Caucasian people, and consequently for the lower dosages of these drugs used.While CYP2D6 catalyses the metabolism of lipophilic bases only, CYP2C19 is involved in the metabolism of acids (e.g. S-mephenytoin), bases (e.g. imipramine and omeprazole) and neutral drugs (e.g. diazepam). About 3% of Caucasians and 12 to 22% of Orientals are PMs of S-mephenytoin. Polymerase chain reaction-based genotyping techniques recently became available for the two CYP2C19 mutated alleles m1 and m2, which cause no enzyme to be expressed, m1 accounts for about 80% of the mutations responsible for the PM phenotypes in Caucasian, Oriental and Black people.Diazepam is partially demethylated by CYP2C19, and the high frequency of mutated alleles in Orientals is probably the reason why such populations have a slower metabolism and are treated with lower doses of diazepam than Caucasians. Omeprazole is to a major extent hydroxylated by CYP2C19, and there is an approximately 10-fold difference in oral clearance between EMs and PMs of S-mephenytoin.The separation of Caucasians from Orientals is fairly recent in the evolutionary process (40 000 to 60 000 years ago); the separation of Black from Caucasian/ Oriental people occurred much earlier, about 150 000 years ago. As pronounced differences have been found between Caucasians and Orientals in the CYP2D6 and CYP2C19 enzymes, it might be expected that Black people will show even greater differences in this respect. Some studies have been performed with Black participants, but the picture is not clear. The mean CYP2D6 activity in Black EMs seems to be lower than that in Caucasian EMs and similar to that of Oriental EMs. The incidence of poor metabolism of S-mephenytoin in Black people is 4 to 7%, i.e. between that of Caucasians and Orientals. The major mutation m1 which causes the PM phenotype is the same in these 3 races, and is thus a very ancient one.

Pronounced differences between native Chinese and Swedish populations in the polymorphic hydroxylations of debrisoquin and S-mephenytoin

The frequency of poor metabolizers of debrisoquin was low and similar in four different native Chinese nationalities. In a total sample of 695 Chinese subjects, only seven (1.01%) had a urinary ratio between debrisoquin and 4‐hydroxydebrisoquin >12.6, which is the antimode between poor metabolizers and extensive metabolizers in white populations. This is significantly lower than the 6.82% found in 1011 white Swedish healthy subjects (p < 0.0001). Admixture analysis indicated the occurrence of two distributions within extensive metabolizers among both Chinese and white subjects. The mean of the distribution of metabolic ratios among Chinese extensive metabolizers was shifted toward higher values compared with Swedish extensive metabolizers (p < 0.01). The frequency of poor metabolizers of S‐mephenytoin was higher in 137 Chinese (14.6%) than in 488 Swedish (3.3%) subjects (p < 0.0001). Our findings imply that drugs metabolized by these two polymorphic hydroxylases should be prescribed in different dosages to Chinese and white subjects.

Platelet MAO activity and monoamine metabolites in cerebrospinal fluid in depressed and suicidal patients and in healthy controls

Platelet monoamine oxidase (MAO) activity and cerebrospinal fluid (CSF) concentrations of 5-hydroxyindoleacetic acid (5HIAA), homovanillic acid (HVA), and 4-hydroxy-3-methoxyphenylglycol (HMPG) were simultaneously measured in 20 currently depressed patients, 11 recovered depressed patients, 15 nondepressed suicide attempters, and 42 healthy control subjects. Both 5HIAA and HVA were positively and significantly correlated to platelet MAO activity in the healthy subjects, but not in any of the patient groups. Suicide attempters had significantly lower CSF 5HIAA than nonsuicidal patients.

CSF monoamine metabolites in melancholia

ABSTRACT– The neurotransmitter metabolites 5‐hydroxyindoleacetic acid (5‐HIAA), homovanillic acid (HVA) and 4‐hydroxy‐3‐methoxy‐phenyl glycol (HMPG) in cerebrospinal fluid (CSF) were measured by mass fragmentography in 83 patients with melancholia (diagnosed by the Newcastle Inventory and the Research Diagnostic Criteria), and 66 healthy volunteer controls. After adjustment by analysis of covariance for differences between the subject groups in body height, age and sex distribution, significantly (P < 0.001) lower concentrations of 5‐HIAA and HVA were found in the melancholia patients than in the controls. HMPG did not differ between the groups. The differences could not be accounted for by differences in timing or examination techniques, and not by previously administered drugs (all patients were drug‐free at the examination, but a minority had taken small amounts of psychotropic drugs prior to the wash‐out period). The differences persisted after excluding the suicidal patients. There were no clear‐cut differences between unipolar and bipolar patients. It is suggested that the reduced concentrations of 5‐HIAA and HVA in the melancholic patients may be due to altered serotonin and/or dopamine functions in the central nervous system, which may be connected with an increased vulnerability to certain types of affective illness.

10‐hydroxylation of nortriptyline in white persons with 0, 1, 2, 3, and 13 functional CYP2D6 genes

To investigate the disposition and effects of nortriptyline and its major metabolite 10‐hydroxynortriptyline in panels of white subjects with different CYP2D6 genotypes, including those with duplicated and multiduplicated CYP2D6*2 genes and to evaluate the contribution of the number of functional CYP2D6 alleles to the metabolism of nortriptyline, used here as a model drug for CYP2D6 substrates.

Fluvoxamine inhibition and carbamazepine induction of the metabolism of clozapine : evidence from a therapeutic drug monitoring service

Therapeutic drug monitoring data for clozapine were used to study interactions with other drugs. The distribution of the ratio concentration/dose (C/D) of clozapine was compared in four matched groups—patients simultaneously treated with benzodiazepines, patients on drugs that inhibit the cytochrome P450 enzyme CYP2D6, patients taking carbamazepine, and those not taking any of these drugs. No difference was seen among the monotherapy, CYP2D6, and benzodiazepine groups. Patients on carbamazepine had a mean 50% lower C/D than the monotherapy group (p < 0.001), indicating that carbamazepine is an inducer of the metabolism of clozapine. The C/D was inversely correlated to the daily dose of carbamazepine. Intraindividual comparisons in eight patients, with analyses both on and off carbamazepine, confirmed a substantial decrease of the clozapine concentration when carbamazepine was introduced. Four patients treated with clozapine were concomitantly given the antidepressant fluvoxamine. Three of them exhibited a much higher C/D ratio when on fluvoxamine compared with the monotherapy group. Two had their clozapine levels analyzed when on and off fluvoxamine. The dose-normalized clozapine concentration increased by a factor of 5–10 when fluvoxamine was added. We conclude that carbamazepine causes decreased clozapine plasma levels, while fluvoxamine increases the levels. The pathways are not known with certainty, but CYP1A2 may be of major importance for the metabolism of clozapine, since fluvoxamine is a potent inhibitor of this enzyme. A recent panel study suggests that determination of CYP1A2 activity with the caffeine test may be very useful for the dosing of clozapine. The induction of clozapine metabolism by carbamazepine might be partly mediated by CYP3A4.

Importance of genetic factors in the regulation of diazepam metabolism: Relationship to S‐mephenytoin, but not debrisoquin, hydroxylation phenotype

Single oral 10 mg doses of diazepam and demethyldiazepam were given on different occasions to 16 healthy subjects. The subjects included four poor hydroxylators of debrisoquin and three poor hydroxylators of mephenytoin. There was a correlation between the total plasma clearance of diazepam and demethyldiazepam (rs = 0.83; p < 0.01). There was no relationship between benzodiazepine disposition and debrisoquin hydroxylation. Poor hydroxylators of mephenytoin had less than half the plasma clearance of both diazepam (p = 0.0008) and demethyldiazepam (p = 0.0001) compared with extensive hydroxylators of mephenytoin. The plasma half‐lives were longer in poor hydroxylators than they were in extensive hydroxylators of mephenytoin for both diazepam (88.3 ± SD 17.2 and 40.8 ± 14.0 hours; p = 0.0002) and demethyldiazepam (127.8 ± 23.0 and 59.0 ± 16.8 hours; p = 0.0001). There was no significant difference in volume of distribution of the benzodiazepines between the phenotypes. This study shows that the metabolism of both diazepam (mainly demethylation) and demethyldiazepam (mainly hydroxylation) is related to the mephenytoin, but not to the debrisoquin, hydroxylation phenotype.

Homicide, suicide and CSF 5‐HIAA

ABSTRACT– Concentrations of 5‐hydroxyindoleacetic acid (5‐HIAA), homovanillic acid (HVA), and 4‐hydroxy‐3‐methoxyphenyl glycol (HMPG) in lumbar spinal fluid were measured by mass fragmentography in 16 men convicted for criminal homicide, 22 men who had attempted suicide, and 39 healthy male control subjects. Those men who had killed a sexual partner, and those who had attempted suicide, had lower levels of the serotonin metabolite, 5‐HIAA in spinal fluid than the controls. It is suggested that low levels of 5‐HIAA in spinal fluid reflect a disorder of serotonin turnover, which makes the individual more prone to acts of violence in states of emotional turmoil.

Clinical pharmacokinetics and pharmacological effects of carbamazepine and carbamazepine-10,11-epoxide. An update

SummaryCarbamazepine is a first-line drug in the treatment of most forms of epilepsy and also the drug of first choice in trigeminal neuralgia. Furthermore, it is now frequently used in bipolar depression.Most oral formulations of carbamazepine are well absorbed with high bioavailability. The drug is 75% bound to plasma proteins. The degree of protein binding shows little variation between different subjects, and there is no need to monitor free rather than total plasma concentrations.Carbamazepine is metabolised in the liver by oxidation before excretion in the urine. A major metabolite is carbamazepine-10,11-epoxide which is further metabolised by hydration before excretion. This epoxide-diol pathway is induced during long term treatment with carbamazepine. Co-medication with phenytoin or phenobarbitone further induces this metabolic pathway. Some but not all studies indicate an increased metabolism of carbamazepine during pregnancy. The drug crosses the placenta, and the newborns who are exposed to the drug during fetal life eliminate the drug readily after birth. There seems to be no problem to nurse children during treatment with carbamazepine. Metabolism of carbamazepine is comparable in children and adults.Several studies have tried to establish a relationship between plasma carbamazepine and clinical effect in epilepsy, but very few of these are controlled. The best anticonvulsant effect seems to be obtained at plasma concentrations of 15 to 40 µmol/L and a similar optimal plasma concentration range was found in a controlled study in trigeminal neuralgia. Side effects are more frequent at higher plasma concentrations but are also seen within that range. In some patients, with pronounced fluctuation of plasma concentrations during the dosage interval, side effects may be avoided by more frequent dosing. Carbamazepine-10,11-epoxide is a potent anticonvulsant in animal models. During treatment with carbamazepine the plasma concentrations of this metabolite are usually 10 to 50% of those of the parent drug. It has not been possible to establish the relative contribution of the two compounds to the pharmacological effects. The epoxide has therefore been given to humans with the aim of determining the relative potency of the parent drug and its metabolite. After single oral doses of carbamazepine-10,11-epoxide to healthy subjects, the compound was rapidly absorbed. As a mean of 90% of the given dose was recovered in urine as trans-10,11-dihydroxy-10,11-dihydro-carbamazepine, a complete absorption of unchanged epoxide was shown. The mean plasma half-life of unchanged epoxide was 6.1 hours with a mean volume of distribution of 0.74 L/kg.Six patients with trigeminal neuralgia had their optimal carbamazepine dose replaced with carbamazepine-10,11-epoxide for 3 to 6 days. The study was single-blind and placebo controlled. When carbamazepine and the epoxide were given in similar doses, the pain control was comparable. The results show that during carbamazepine therapy, the contribution of the epoxide to the effect is considerable. No side effect was seen during the epoxide therapy. Further studies on the effect of carbamazepine-10,11-epoxide administration in epilepsy are indicated.