Influence of ketoconazole on azimilide pharmacokinetics in healthy subjects

Abstract

Aim

To assess the influence of ketoconazole on azimilide pharmacokinetics.

Methods

A two-period randomized crossover study was conducted in healthy male and female subjects (19–45 years). Placebo or 200 mg ketoconazole were administered orally every 24 h for 29 days. On day 8, a single oral dose of 125 mg azimilide dihydrochloride was coadministered following an overnight fast. Blood samples were obtained prior to and for 22 days following azimilide dihydrochloride administration. The plasma protein binding of azimilide was also assessed at 6 h after dosing.

Results

Following ketoconazole administration, a 16% increase in azimilide AUC (90% confidence interval (CI) 112%, 120%), a 12% increase in C_max (95% CI 107%, 116%), a 13% increase in t_1/2,z (95% CI 107%, 120%) and a 14% decrease in CL_o (95% CI 82%, 90%) were observed.

Conclusions

The changes in azimilide pharmacokinetics following ketoconazole treatment are not clinically important since the 90% CI for the AUC fell within the prespecified range of 80–125%. Thus, no clinically important drug interactions are expected when azimilide dihydrochloride is coadministered with CYP3A4 inhibitors.

Introduction

Azimilide dihydrochloride (NE-10064) is a class III antiarrhythmic drug. Unlike conventional potassium channel blockers such as sotalol and dofetilide, which antagonize only slowly activating (I_Ks) components of the delayed rectifier potassium current of human atrial and ventricular myocytes, azimilide blocks both the slowly activating (I_Ks) and rapidly activating (I_Kr) components [1]. Azimilide dihydrochloride (75–125 mg day⁻¹) is currently being developed for use in prolonging the time to recurrence of atrial fibrillation/flutter and for decreasing the frequency of shocks in patients with an implantable cardioverting defibrillator.

Azimilide pharmacokinetics have been assessed following intravenous and oral administration [2]. Following intravenous administration, azimilide pharmacokinetic parameters include: a total clearance of 0.136 l h⁻¹ kg⁻¹, a renal clearance of 0.013 l h⁻¹ kg⁻¹, a steady-state volume of distribution of 12.9 l kg⁻¹, and a terminal exponential half-life of 71.4 h. These parameters indicate that azimilide has a low hepatic extration ratio clearance and a steady-state volume of distribution consistent with extensive tissue distribution. Owing to its low clearance and large volume of distribution, azimilide has a long terminal exponential half-life. Following oral administration, azimilide is completely absorbed with peak blood concentrations occurring at approximately 7 h. When azimilide is administered following a high fat meal, no change in the extent of absorption is observed, although peak blood concentrations decrease by 19%[3].

Azimilide is mainly eliminated by metabolism (Fig. 1), with renal clearance accounting for only approximately 10% of total clearance. Cleavage of the isocyanide bond leading to the formation of the carboxylic acid F-1292 [4], the major metabolite in plasma, accounts for about 35% of the total clearance of azimilide. Minor metabolites in plasma are desmethylazimilide, azimilide N-oxide and azimilide carboxylate. The first two possess 20% and 11%, respectively, of the antiarrythmic activity of the parent drug in vitro, whereas F-1292 and azimilide carboxylate are inactive [5]. Since the plasma concentrations of the desmethyl and N-oxide metabolites are only 10% of those of azimilide, they do not contribute measurably to antiarrthymic activity in vivo[6]. Metabolism by CYP3A4/5 and CYP1A1 accounts for 15% and 25%, respectively, of the total clearance of azimilide [4].

Figure 1.

The metabolism of azimilide in humans

Ketoconazole is a broad-spectrum, imidazole antifungal agent indicated for the treatment of topical, vaginal and systemic infections. Following oral administration, ketoconazole is about 76% absorbed with peak concentrations occurring at 2–3 h [7, 8]. Ketoconazole is primarily metabolically cleaned with a terminal exponential half-life of ∼8 hours. Ketoconazole is a potent reversible inhibitor of CYP3A4-mediated metabolism and p-glycoprotein transport and a weak inhibitor of CYP2C9- and CYP2D6-mediated metabolism [9–12]. The extent of inhibition of CYP3A4 is dependent upon both dose and duration of dosing, with 80% inhibition occurring after a single oral dose of 200 mg of ketoconazole [13]. Multiple oral doses of 200 and 400 mg daily resulted in 94 and 98% inhibition, respectively [14, 15].

The purpose of this study was to assess the effect of ketoconazole on azimilide pharmacokinetics.

Methods

Study design

This was a two-period, double-blind, placebo-controlled, randomized, crossover study conducted in healthy, nonsmoking male and female subjects, 19–45 years of age. Health was assessed based on a physical examination, ECG, clinical chemistry, haematology and urinalysis. Subjects were asked not to take other drugs for at least 7 (nonprescription medicine) to 30 (inducers/inhibitors of drug metabolism) days prior to dosing. The study was conducted at PPD Development Clinics (Austin, TX, USA), and in accordance with the Declaration of Helsinki and its amendments. All subjects provided written informed consent following Institutional Review Board approval (Research Consultant's Review Committee, Austin, TX, USA).

Ketoconazole (200 mg) or placebo was administered orally every 24 h for 29 days. On day 8, a single oral dose of 125 mg azimilide dihydrochloride was administered at the same time as ketoconazole/placebo following an overnight fast. All doses were administered with 240 ml of water, and subjects were not allowed to eat for 4 h after dosing. Each period was separated by at least 38 days.

Blood samples (sodium heparin anticoagulated) were collected prior to and for 22 days after dosing at the following times: 1, 2, 4, 6, 8, 10, 12, 18, 24, 36, 48, 72, 96, 120, 144, 192, 240, 288, 336, 384, 432, 480 and 528 h. Blood samples were also obtained prior to and at 6 h after dosing for assessment of plasma protein binding of azimilide. Blood and plasma samples were frozen at −20 °C until analysis.

Analysis and plasma protein binding of azimilide

Azimilide blood concentrations were determined by a high-performance liquid chromatographic method with MS/MS detection [16]. The lower limit of quantitative from a 0.2-ml sample was 5 ng ml⁻¹, with a linear range from 5 to 800 ng ml⁻¹. The accuracy and precision over this range were 98–102% and less than 10%, respectively.

The extent of the binding of azimilide to plasma protein was determined at 37 °C using equilibrium dialysis [17]. Plasma was spiked with ¹⁴C-azimilide dihydrochloride (∼ 12 000 d.p.m.) and dialysed against pH 7.4 Kreb's buffer. The results were corrected for volume shifts [18]. The lower limit of quantitative azimilide was 50 d.p.m. (0.227 ng). Experiments were performed over the concentration range 12.7–173 ng ml⁻¹. The precision of the assay was less than 9%.

Data analysis

Pharmacokinetic parameters were determined by ‘noncompartmental’ methods [19, 20]. The maximum blood concentration (C_max) and time to maximum concentration (t_max) were determined from visual inspection of the data. Area under the blood concentration-time curve from time 0 until the last quantifiable blood concentration (AUC_tlast) was determined by the linear trapezoidal rule and was extrapolated to infinity by dividing the last observed quantifiable blood concentration by the terminal exponential rate constant (λ_z). The latter was determined from linear regression analysis of data points during the terminal exponential phase of the log blood concentration-time plot. The terminal exponential half-life (t_1/2,z) was calculated from the relationship t_1.2,z = ln2/λ_z. Oral clearance (CL_o) and terminal volume of distribution (V_z/F) were obtained using standard equations.

An analysis of variance for a two-treatment, two-period, randomized, crossover design was performed for each pharmacokinetic parameter. The 90% confidence interval (CI) for the ratio of azimilide AUC following ketoconazole to azimilide AUC following placebo was constructed. The absence of a drug-drug interaction was concluded if the 90% CI fell within the prespecified interval of 80–125%. For all other parameters, a 95% CI for the ratio of ketoconazole to placebo was estimated. AUC and C_max were log transformed prior to analysis. All other parameters were assessed for adherence to the assumptions of the analysis of variance model using the nontransformed and log-transformed data.

Results

Demographic characteristics of subjects participating in this study are provided in Table 1. Of the 22 who were enrolled, 21 completed the study. One subject was withdrawn due to noncompliance with the dosing regimen.

Table 1.

Subject demographics

	Age (years)	Body weight (kg)	Gender	Race
Mean	30.5	80.6	21 M/1 F	African-American: 1
% CV	26	11		Caucasian: 15
Minimum	19	64.3		Hispanic: 5
Maximum	45	95.7		Multiple: 1

Mean azimilide blood concentration-time profiles following single-dose oral administration of 125 mg azimilide dihydrochloride and pretreatment with either ketoconazole or placebo are illustrated in Figure 2. Azimilide pharmacokinetic parameters are listed in Table 2. Plots of individual AUC and C_max values with and without ketoconazole pretreatment are illustrated in Figures 3 and 4. These results indicate that intersubject variability for each parameter is very similar between the two treatment groups (20% or less depending on the parameter) and that pretreatment with ketoconazole decreases azimilide oral clearance (14%), and increases azimilide C_max (12%), AUC (16%) and t_1/2,z (13%). No change in t_max, V_z/F or plasma protein binding was observed.

Figure 2.

Mean azimilide blood concentration-time profiles after single-dose oral administration of 125 mg azimilide dihydrochloride to healthy subjects alone or with ketoconazole. Placebo (•), ketoconazole (○)

Table 2.

The pharmacokinetics of azimilide given alone and with ketoconazole

	Least-squares geometric or arithmetic mean or mediana
Pharmacokinetic parameter	Ketoconazole	Placebo	Ketoconazole/placebo (%)	90% CI (%)
AUC (ng h−1 ml−1)	12 639.82	10 898.21	115.98	(111.71, 120.42)95% CI (%)
Cmax (ng ml−1)	152.43	136.40	111.76	(107.48, 116.20)
tmax (h)	8.00	8.00	100.00	(89.44, 111.80)
t1/2,Z (h)	93.57	83.01	112.73	(106.74, 119.68)
CLo (l h−1 kg−1)	0.11	0.12	86.21	(82.38, 90.22)
VZ/F (l kg−1)	14.14	14.67	96.40	(89.27, 104.10)
Fb	96.35	96.68	99.66	(99.24, 100.08)

AUC is the area under the blood concentration-time curve from time zero to infinity; C_max is the maximum blood concentration; t_max is the time corresponding to the occurrence of the maximum blood concentration (C_max); t_1/2,Z is the terminal exponential half-life; Cl_o is the oral clearance; V_Z/F is the terminal volume of distribution; F_b is the fraction bound to plasma protein.

^at_1/2,Z is a least-squares arithmetic mean; t_max is a median; all other parameters are least-squares geometric means.

Figure 3.

Individual data for the effect of ketoconazole on azimilide AUC

Figure 4.

Individual data for the effect of ketoconazole on azimilide C_max

Discussion

Azimilide is being developed for the treatment of ventricular and atrial arrhythmias where coadministration of drugs occurs frequently. Since azimilide is metabolized in part by CYP3A4, the influence of inhibition of this enzyme on azimilide pharmacokinetics were assessed using ketoconazole as a model inhibitor.

In subjects administered placebo, the median peak concentration of azimilide was at 08.00 h, the mean oral clearance 0.12 l h⁻¹ kg⁻¹, the mean volume of distribution 14.67 l kg⁻¹, and the mean terminal half-life 83.01 h. These results are similar to those previously observed following single-dose oral administration of azimilide dihydrochloride [2].

In a previous clinical study, CYP3A4-mediated metabolism accounted for about 15% of azimilide total clearance [4]. In this study, a 14% decrease in azimilide oral clearance was observed following pretreatment with ketoconazole. Since azimilide is not a p-glycoprotein substrate [21], this decrease in oral clearance reflects inhibition of CYP3A4-mediated metabolism [4, 22–24]. This change in CYP3A4-mediated metabolism corresponds to greater than 90% inhibition and is in good agreement with the percentage inhibition previously reported for ketoconazole upon multiple doses [14, 15]. Inhibition appears to be mediated systemically since azimilide is completely absorbed following oral administration [2] and a similar decrease in oral clearance (14%) and increase in the terminal exponential half-life (12%) was observed upon coadministration with ketoconazole, with no change in the volume of distribution. The increase in azimilide concentration was small and are not considered to be clinically important, since the 90% CI for the AUC fell within the prespecified range of 80–125%. In addition, the pharmacologically active metabolites of azimilide are primarily formed via CYP3A4. Thus, CYP3A4 inhibition would decrease their formation and their corresponding plasma concentrations (normally about 10% or less of that of azimilide), reducing further any minimal contribution to the overall cardiovascular response associated with azimilide administration.

In conclusion, administration of ketoconazole caused a 14% decrease in azimilide total clearance. However, since the 90% CI for AUC was contained within the prespecified interval of 0.8–1.25, no clinically important drug interactions are expected when azimilide dihydrochloride is coadministered with CYP3A4 inhibitors.