Introduction

The thienodiazepine etizolam shows pharmacological effects similar to those of diazepam [21]. Etizolam is used in the treatment of psychiatric disorders such as general anxiety disorder [6] and panic disorder [17]. Unpublished data from the manufacturer (Mitsubishi Pharma 2003, unpublished data in package insert) suggest that etizolam is metabolized mainly by hydroxylation of the methyl group (1′-hydroxylation) and that of the ethyl group.

Cytochrome P450 (CYP) 3A4 is the major enzyme catalyzing the metabolism of several benzodiazepines, e.g., alprazolam [25, 26], brotizolam [20], midazolam [14], and triazolam [22]. This enzyme is also partly involved in the metabolism of diazepam [1, 13], flunitrazepam [13], and quazepam [12]. Unpublished in vitro data suggest that CYP3A4 catalyzes the 1′-hydroxylation of etizolam (Mitsubishi Pharma 2003, unpublished data in package insert). Our recent study suggests that itraconazole, an inhibitor of CYP3A4, inhibits the metabolism of etizolam in humans, providing in vivo evidence that CYP3A4 is at least partly involved in etizolam metabolism [3].

It has been shown that carbamazepine induces CYP3A4 activity [16, 24, 27]. In fact, carbamazepine induces the metabolism of alprazolam [8] and midazolam [4]. Therefore, it is expected that carbamazepine induces the metabolism of etizolam as well.

To clarify this point, the authors examined the effect of carbamazepine on the single oral dose pharmacokinetics of etizolam.

Subjects and methods

The subjects were 11 healthy male volunteers. They were all Japanese. The mean±SD age was 33±9 years, and that of body weight was 73±13 kg. No subject had taken any drug or food that affects CYP3A4 activity for at least 1 week before the study. The study protocol was approved by the ethics committees of Yamagata University School of Medicine and Hirosaki University School of Medicine, and each subject gave his written informed consent before the study.

The study was conducted in a double-blind, randomized, crossover manner, with at least a 4-week washout period. The subjects were randomly allocated to one of the two treatment sequences: placebo-carbamazepine or carbamazepine-placebo. Carbamazepine 200 mg as the tablet formulation (Tegretol, Novartis Pharma, Tokyo, Japan) or matched placebo was given orally at 2100 hours for 6 days. After an overnight fast, at 0900 hours of the 6th day, a single oral 1-mg dose of etizolam (Depas, Mitsubishi Pharma, Osaka, Japan) as the tablet formulation was given with a cup of tap water. No food was allowed until 3 h after etizolam dosing. Blood samples (10 ml each) for etizolam measurement were taken before and at 0.25, 0.5, 1, 2, 3, 4, 6, 8, 10, 12, and 24 h after etizolam dosing. Blood samples for carbamazepine measurement were taken at 0900 hours on the 6th day. At times of blood samplings, psychomotor function was assessed using the digit symbol substitution test (DSST) adapted from the Wechsler Adult Intelligence Scale [23] in 90 s, and the Stanford Sleepiness Scale (SSS) [11].

Plasma concentration of etizolam was measured by means of the high-performance liquid chromatography (HPLC) method of Hikida et al. [10]. The lowest limit of detection [signal/noise ratio (S/N)=3] was 0.3 ng/ml, and that of quantification (S/N=6) was 0.6 ng/ml. The coefficients of variation (CV) at the concentrations of 1, 5, and 20 ng/ml were all less than 2.8%. Plasma concentration of carbamazepine was measured using the same HPLC method as for etizolam with some modifications. The lowest limit of detection was 0.05 μg/ml, and that of quantification was 0.1 μg/ml. The CV at the concentrations of 0.5, 2.5, and 5.0 μg/ml were all less than 0.8%.

The elimination rate constant (k) was estimated by the linear regression analysis of the terminal log-linear plasma concentration–time curve. The elimination half-life (t1/2) was calculated by 0.693/k. The area under the plasma concentration–time curve (AUC) from 0 h to 24 h [AUC(0–24)] was calculated using the trapezoidal rule. The AUC from 0 h to infinity [AUC(0–∞)], or total AUC, was calculated by AUC(0–24)+C24/k, in which C24 is the plasma concentration at 24 h. The peak plasma concentration (Cmax) and time to Cmax (tmax) were determined graphically. For the DSST and SSS, the area under the score-time curve from 0 h to 24 h [AUSC(0–24)] was calculated.

The paired t test was used to examine whether there were significant differences between the placebo and carbamazepine phases in the mean plasma concentrations, and in pharmacokinetic and pharmacodynamic parameters of etizolam. A P value of 0.05 or less was considered statistically significant. In the bioequivalence statistics, the 90% confidence interval (CI) of the carbamazepine/placebo ratio of each parameter was calculated and compared with the bioequivalence range of 0.80–1.25 [18].

Results

During the placebo phase, plasma etizolam concentrations were measurable for up to 24 h in all subjects. However, during the carbamazepine phase, plasma etizolam concentrations were no longer measurable at 24 h in four subjects. Carbamazepine treatment significantly (P<0.05–0.001) decreased plasma etizolam concentrations at all time points except at 0.25 h (Fig. 1).

Fig. 1
figure 1

Mean plasma concentration-time data of etizolam after a single oral 1-mg dose of etizolam during the treatment with placebo (open circle) or carbamazepine (filled circle). The bracketed bars indicate standard deviations. *P<0.05; **P<0.01; ***P<0.001

Full size image

Carbamazepine significantly decreased the Cmax (P<0.05), total AUC (P<0.001), and elimination t1/2 (P<0.01) of etizolam (Table 1). The greatest individual decreases in these parameters were 39, 59, and 76%, respectively, which were observed in three separate subjects. The 90% CIs of the carbamazepine/placebo ratio of the total AUC and elimination t1/2 were not contained in the bioequivalence range of 0.80–1.25.

Table 1 Pharmacokinetic and pharmacodynamic parameters of etizolam after a single oral 1-mg dose of etizolam. Values are means±SD. Cmax peak plasma concentration, tmax time to Cmax, AUC area under the plasma concentration-time curve, t1/2 half-life, AUSC area under the score-time curve, DSST digit symbol substitution test, SSS stanford sleepiness scale
Full size table

The AUSC (0–24) of DSST and SSS were not significantly different between the two phases (Table 1). The mean (±SD) plasma concentration of carbamazepine at 0900 hours was 4.3±1.5 μg/ml.

Discussion

In the present study, the number of subjects was not decided beforehand by a power calculation. However, a post-hoc power calculation revealed that the actual number of subjects (n=11) was sufficient to detect even a small change caused by carbamazepine. For instance, the power to detect a change of 19.5 ng h/ml in the total AUC (10% of the mean value of the placebo phase) at the significance level of 0.05 was calculated as 0.89 [28].

Alprazolam [25, 26] and midazolam [14] are representative substrates of CYP3A4. Furukori et al. [8] have shown that carbamazepine treatment markedly decreased the total AUC and elimination t1/2 of alprazolam in healthy volunteers. Backman et al. [4] have reported that the Cmax, total AUC, and elimination t1/2 of midazolam were markedly decreased in epileptic patients taking carbamazepine in comparison with healthy volunteers. These studies indicating that carbamazepine induces the metabolism of CYP3A4 substrates prompted us to examine the effect of carbamazepine on etizolam metabolism.

The results of this study were largely in line with those of the studies on alprazolam and midazolam. Carbamazepine treatment significantly decreased plasma etizolam concentrations at almost all time points. As to the pharmacokinetic parameters, carbamazepine significantly decreased the Cmax, total AUC, and elimination t1/2 of etizolam. The bioequivalence statistics showed significant effects of carbamazepine on the total AUC and elimination t1/2 of etizolam. These results strongly suggest that carbamazepine induces the metabolism of etizolam. This metabolic induction may be mediated by CYP3A4, although other CYP enzyme(s) may also be involved. For instance, it has been reported that CYP1A2 is also inducible by carbamazepine [15].

To take the decrease in total AUC as an index of metabolic induction by carbamazepine, the value of etizolam (42%) was smaller than that of alprazolam (58%) [8] and much smaller than that of midazolam (94%) [4]. This relative extent may be partly explained by the differences in doses of carbamazepine (200 mg versus 300 mg versus 700–900 mg/day) and duration of treatment (6 days versus 8 days versus at least 2 months).

As mentioned above, carbamazepine decreased plasma etizolam concentrations, but this was not accompanied by significant changes in the pharmacodynamic parameters. The possibility that this negative result is ascribable to the sedative effect of carbamazepine itself [2] cannot be excluded, though the baseline scores of the DSST and SSS were not significantly different between the placebo and carbamazepine phases (data not shown). Another possibility is that the formation of 1′-hydroxyetizolam, which is suggested to be equipotent to the parent compound [7], was increased in the carbamazepine phase, and increased sedation by this active metabolite nullified the pharmacodynamic consequence of decreased etizolam.

In the present study, carbamazepine treatment of a rather short period of 6 days was used for an ethical reason. It has been suggested that the enzyme-inducing effect of carbamazepine is maximal at 3–5 weeks [5]. Therefore, the possibility that carbamazepine treatment over a longer period than in this study further induces etizolam metabolism, leading to significant pharmacodynamic changes, remains.

The elimination t1/2 of etizolam in this study (11.1 h) was much longer than that in one study conducted in Caucasians (3.4 h) [7]. This difference does not appear to be explained by CYP3A4 activity, since another study has suggested that there is no clear difference in CYP3A4 activity between Japanese and Caucasians [19]. Other CYP enzyme(s), showing clear interethnic differences, e.g., CYP2C19 [9], may be involved in the difference in elimination t1/2 mentioned above.

In conclusion, the present study suggests that carbamazepine induces the metabolism of etizolam.