Effect of the catechol-O-methyltransferase inhibitor entacapone on the steady-state pharmacokinetics and pharmacodynamics of warfarin

Abstract

Aims

To investigate the influence of a multiple-dose regimen with the catechol-O-methyltransferase inhibitor entacapone on the pharmacokinetics and pharmacodynamics of warfarin.

Methods

In a randomized, double-blind, two-way cross-over study, 12 healthy subjects (gender ratio 1 : 1) received treatment for 1 week with either entacapone 200 mg four times daily or placebo during individually optimized treatment with warfarin (INR 1.4–1.8). The effect of entacapone on the steady-state pharmacokinetics of both R- and S-warfarin was determined and, in addition, INR values were measured. The key pharmacokinetic variables were AUC_ss, C_max and t_max.

Results

Entacapone increased the exposure to R-warfarin by 18% (90% CI: 111, 126%), and caused a 13% (6, 19%) increase in INR values. No effect was seen on the pharmacokinetics of the pharmacologically more potent S-enantiomer. Safety and tolerability variables did not show any difference between the treatment phases.

Conclusions

In healthy subjects, entacapone displays a slight pharmacokinetic interaction with R-warfarin but, based on the lack of a clinically relevant pharmacodynamic interaction, it appears that it can also be used safely in Parkinson's disease patients who are receiving warfarin.

Introduction

The mainstay for the treatment of Parkinson's disease (PD) is combined oral therapy with the dopamine precursor levodopa and a dopa decarboxylase inhibitor such as carbidopa or benserazide. Although this combination is an improvement on levodopa alone, delivery of dopamine to the striatum is still low and variable. Improvements in levodopa delivery are desirable, particularly with regard to optimizing the delivery of levodopa to the brain. The pulsatile dopaminergic stimulation that occurs with levodopa alone may be associated with the development of drug-related motor complications [1–4]. In recent years, catechol-O-methyltransferase (COMT) inhibitors have been added to the therapeutic armamentarium in PD [5–8]. Inhibition of the levodopa catabolizing enzyme COMT results in greater availability of levodopa to the brain. In addition, smoother delivery occurs with no increase in maximum plasma concentration, but an increase in the plasma half-life [9]. This tendency towards more even and continuous dopaminergic stimulation of the striatal cells is closer to the physiological situation and may represent an advantage in minimizing the risk of motor complications developing later in the disease [4, 10].

Two COMT inhibitors, entacapone and tolcapone, have been licensed by regulatory authorities in many countries. These drugs result in prolonged duration of benefit from levodopa therapy [6, 8]. However, tolcapone has been withdrawn from most markets because of its potential to elicit serious liver toxicity [11]. Despite extensive clinical experience with entacapone in fluctuating and nonfluctuating PD [12–15], no hepatotoxicity has been observed with this drug. Furthermore, some preclinical studies have shown important differences in the toxicological profile of these two COMT inhibitors [16, 17].

Anecdotal reports on tolcapone have also indicated that the drug interacted with the oral anticoagulant warfarin, possibly due to an inhibitory effect on cytochrome P450 (CYP) 2C9 [18]. Warfarin is widely used for the treatment of various thrombotic and embolic diseases, including deep vein thrombosis, rheumatic heart disease, atrial fibrillation, as a prophylactic following insertion of artificial heart valves, and to treat transient ischaemic attacks. A significant number of PD patients are likely to require treatment with warfarin or a related anticoagulant at some point during their illness. Warfarin has three intrinsic properties that makes it susceptible to drug–drug interactions. It has a very narrow therapeutic window, is highly bound to plasma proteins and is stereoselectively metabolized by cytochrome P450 [19, 20]. Warfarin is a racemic mixture of its R- and S-enantiomers, but most of the pharmacological activity resides with the S-enantiomer [21]. CYP3A4 and CYP1A2 primarily metabolize the R-enantiomer, whereas the S-enantiomer is mainly metabolized by CYP2C9 [20]. Because S-warfarin is more potent than R-warfarin, it has the greater potential for causing drug–drug interactions if its pharmacokinetics are affected by other drugs [22].

In view of these considerations, the influence of multiple-dose entacapone on the steady-state pharmacokinetics and pharmacodynamics of warfarin was investigated in healthy male and female subjects.

Methods

Subjects

Eight healthy female and six healthy male subjects, aged between 24 and 46 years, and who were within 15% of their ideal body weight participated in this study. Ethics Committee approval was obtained from the Landesärztekammer Bavaria, München, Germany, and all subjects gave their written informed consent before any screening procedures were performed. The entire study was conducted in accordance with the principles of the Declaration of Helsinki, as amended, and also in accordance with the International Conference on Harmonization Guidelines for Good Clinical Practice. Additionally, the German Drug Laws relating to the proper conduct of clinical drug trials were adhered to. Non-smoking subjects were selected who were healthy on the basis of medical history, physical examination, and clinical laboratory determinations (haematology, biochemistry, urinalysis, drug screen). Females underwent a pregnancy test and practised adequate contraception throughout the study. Vitamin K₁ intake had to be controlled carefully in this study because of its antagonistic effect on warfarin. Therefore, during study periods, subjects received standardized meals that contained known quantities of this vitamin, present in, for example, leafy green vegetables and tomatoes.

Design

This was a randomized, double-blind, two-way, cross-over study conducted in 12 healthy male and female subjects. The study consisted of four consecutive phases: a screening phase, a warfarin dose finding and stabilization phase, an interaction phase and a recovery phase (Table 1). The warfarin dose-finding and stabilization phase was designed to produce steady-state conditions for warfarin, such that the prothrombin time (PT), expressed as the International Normalized Ratio (INR) became stable at between 1.4 and 1.8 at which time subjects were receiving a constant dose of racemic warfarin (given in the morning) for at least 5 days. Stable anticoagulation status was taken as three consecutive daily INR values between 1.4 and 1.8, not differing by more than 0.20. The value of 0.20 corresponds to 11–14% of the target INR and is approximately 50% of the value considered to be of at least borderline clinical importance [23].

Table 1.

Study design.

	Pre-study examination	Warfarin dose-finding and stabilization			Interaction (cross-over)		Recovery
Period	Pre-start	Days 1+2	Days 3+4	Days* 5–21	7 days	7 days	2–4 days
Treatment	None	8 mg racemic	5 mg racemic	Variable racemic	Racemic warfarin	Racemic warfarin	10 mg vitamin K1
		warfarin	warfarin	warfarin	+placebo orentacapone200 mg fourtimes daily	+entacapone200 mg four timesdaily or placebo	once daily

^*Length of period varied depending on how long each subject took to reach steady state.

Steady-state conditions were attained by using a loading dose of 8 mg warfarin (Marevan® 1 mg tablets) on days 1 and 2, followed by 5 mg on days 3 and 4. This was followed by a period of between 5 and 21 days during which the investigator varied the warfarin dosage until the required INRs were established. Failure to achieve the specified INR within 21 days, or three measurements of an INR greater than 2.5, resulted in removal of the subject from the study. Subjects entered the interaction phase, which consisted of two 1 week, randomized, double-blind, cross-over periods, with no wash-out, between dosing with entacapone (200 mg Comtan®, four times daily at 08.00 h, 12.00 h, 16.00 h, and 20.00 h) and placebo. In the recovery phase, all entacapone and warfarin treatment was stopped, and, in order to reverse the anticoagulant effects of warfarin, subjects received phytomenadion (10 mg, once daily), until INR values returned to the normal reference range. Within 7 days after the recovery phase, routine clinical and laboratory tests were carried out. Because of the potential risk of bleeding during warfarin treatment, the subjects stayed in the clinical unit for the first 4 days and throughout the interaction phase of the study. Furthermore, the target endpoint chosen for the steady-state warfarin treatment, an INR of 1.4–1.8, was considered to represent a subtherapeutic level of anticoagulation, such that the risk to subjects was reduced to the lowest practicable level, while retaining sufficient anticoagulation to allow meaningful assessment of any interaction with entacapone.

Pharmacokinetics

Subjects fasted from at least 10 h before until 5 h after warfarin administration. On days 7 and 14 of the interaction phase, 9 ml blood samples were taken from an indwelling venous catheter at 0 (pre-dose), 0.5, 1, 2, 3, 4, 6, 8, 12, and 24 h, for analysis of the R- and S-enantiomers of warfarin. Blood samples were centrifuged at 1600 g at 4° C for 10 min, and plasma was subsequently frozen at −20° C pending analysis. A stereoselective h.p.l.c. method with a lower limit of quantification of 50 ng ml⁻¹ was used to measure the plasma concentrations of the R- and S-enantiomers of warfarin [24]. The performance of the analytical method was monitored by simultaneous analysis of independently prepared quality control samples containing three different concentrations (0.05, 0.25, and 0.60 µg ml⁻¹). The interassay coefficient of variation was always <5.3% and the inaccuracy of the method ranged from 1.2% to 3.9%. Pharmacokinetic parameters were determined for both enantiomers of warfarin. The maximum plasma concentration (C_max) and the time of its occurrence (t_max) after dosing were read directly from the concentration-time data. The area under the concentration-time curve for a dosing interval at steady state (AUC_ss) was calculated using the linear trapezoidal rule. C_min (trough plasma concentration) and C_av (average plasma concentration over the 24 h period) were also calculated. Peak-trough fluctuation (PTF) was estimated from the equation (C_max−C_min)/C_av.

Pharmacodynamics

Standard haematology parameters, coagulation factors (thromboplastin time, fibrinogen, factors II, VII, IX and X) and occult faecal blood smearing were measured regularly. PT, expressed as INR, was measured daily and used to assess any pharmacodynamic interaction between entacapone and warfarin. Data analysis was based on the mean of the last three values for PT obtained from each phase of the study.

Statistical methods

The extent of the interaction between warfarin and entacapone was assessed statistically by standard methods used for bioequivalence studies [25]. Analysis of variance (anova) was used to estimate any effect of treatment with entacapone on AUC_ss and C_max, relative to placebo, after first logarithmically transforming the data to counter any group effects (such as gender and sequence of dosing). anova was also performed to test for any effects of period (i.e. day during the interaction phase) and sequence of administration of placebo and entacapone. P<0.10 was taken as indicating significance for period, sequence or group effects, while all other parameters were regarded as significant if P<0.05. Any effects on t_max were tested using Koch's nonparametric sequential test [26]. Differences between AUC_ss and C_max were calculated as ratios of geometric means, and for t_max, as median differences. Ninety percent confidence intervals were also calculated. A lack of interaction between entacapone and warfarin was assumed if the 90% confidence intervals for the ratios of AUC_ss and C_max for entacapone compared with placebo during the interaction phase fell within the range 80, 125%. A change in INR of 25% when entacapone was given concomitantly with warfarin was considered to be of at least borderline clinical importance [23].

Tolerability and safety

Adverse events were assessed by spontaneous reports, observations, and questioning on each day of hospitalization and at the ambulatory visits. The intensity of any adverse events was rated on a three-point scale (mild, moderate, and severe), and the investigator assessed their association with drug intake before breaking the study code. Sitting blood pressure and pulse rate were measured and 12-lead ECGs were recorded at screening, on the last day of the warfarin dose titration and stabilization phase, on the last day (day 7) of each interaction period, and at the post-study examination.

Results

Subject withdrawals

Out of 14 subjects who were enrolled, 12 completed the study and were included in the analysis. Of the two discontinuations (both females), one achieved INR values that were consistently above 2.5 during the stabilization phase, and one requested to leave the study for personal reasons.

Pharmacokinetics

The 24 h concentration-time profiles for the R- and S-enantiomers of warfarin, in the presence and absence of entacapone are shown in Figure 1 and Figure 2, respectively. The pharmacokinetic parameters of both enantiomers are shown in Table 2 and the ratios of the geometric means of the primary pharmacokinetic endpoints (AUC_ss, C_max and t_max) for entacapone compared with placebo treatment are summarized in Table 3.

Figure 1.

Plasma concentration-time profile of R-warfarin at steady-state conditions during combined treatment with entacapone 200 mg four times daily (□) and placebo four times daily (▪). Data are presented as mean±s.d. (n = 12).

Figure 2.

Plasma concentration-time profile of S-warfarin at steady-state conditions during combined treatment with entacapone 200 mg four times daily (□) and placebo four times daily (▪). Data are presented as mean±s.d. (n = 12).

Table 2.

Pharmacokinetic parameters of R- and S-warfarin.

	R-warfarin		S-warfarin
	+entacapone	+placebo	+entacapone	+placebo
Cmax (µg ml−1)	0.51±0.17	0.47±0.15	0.64±0.23	0.62±0.21
tmax (h)	1.0 (0.5–3.0)	1.0 (0.5–2.0)	1.0 (0.5–3.0)	1.0 (0.5–2.0)
Cmin (µg ml−1)	0.24±0.09	0.21±0.09	0.36±0.15	0.36±0.14
Cav (µg ml−1)	0.35±0.12	0.30±0.11	0.47±0.18	0.45±0.16
AUCss (µg ml−1 h)	8.4±2.8	7.2±2.6	11.4±4.4	10.8±3.9
PTF (%)	78±19	90±29	60±15	60±16

Data are presented as mean±s.d. (n = 12) for all variables, with the exception of t_max which is expressed as median (range).

Table 3.

Summary pharmacokinetics of R- and S-warfarin (n = 12 subjects).

	LSM entacapone	LSM placebo	Ratio entacapone/placebo(90% CI)
R-warfarin
Cmax (µg ml−1)	0.50	0.46	109 (102, 117)
tmax (h)a	1.0	1.0	+0.38b (0.00, 0.75)
AUCss (µg ml−1 h)	8.30	7.02	118 (111, 126)
S-warfarin
Cmax (µg ml−1)	0.61	0.60	102 (95, 109)
tmax (h)a	1.0	1.0	+0.25b (−0.25, + 0.50)
AUCss (µg ml−1 h)	10.9	10.4	105 (99, 110)

LSM = least squares means

^a= median value

^b= difference.

Tests for period, sequence, and gender yielded insignificant effects. AUC_ss for R-warfarin underwent a modest increase, from 7.2 to 8.4 µg ml⁻¹ h, an increase of 18% over the value for placebo. The 90% confidence interval for this change (111, 126%) lies slightly outside the predefined range (80, 125%), which would have been consistent with entacapone having no effect on warfarin pharmacokinetics, assuming standard bioequivalence criteria. The small changes seen for C_max and t_max all lie within the predefined ranges for bioequivalence, and therefore indicate that these two parameters for R-warfarin were not affected by entacapone to a significant degree.

The pharmacokinetic parameters of S-warfarin indicate that entacapone did not exert a significant effect on any of the steady-state parameters measured. In contrast to R-warfarin, all the measured confidence intervals are within the predefined values, leading to the conclusion that entacapone does not affect the steady-state kinetics of S-warfarin.

Pharmacodynamics

Measurements of PTT, expressed as INR, resulted in a mean±s.d. value of 1.43±0.29 for warfarin combined with placebo and 1.62±0.38 for warfarin combined with entacapone. Thus, the latter caused a mean change of 13% (90% confidence interval 6, 19). The small increase in INR associated with entacapone correlates with the modest increase in the mean plasma concentration of R-warfarin seen on combination with entacapone, suggesting no clinically significant pharmacodynamic interaction between entacapone and warfarin.

Safety and tolerability

The incidence of adverse events was very similar in subjects who received warfarin plus placebo or warfarin plus entacapone. In total, seven of the 12 subjects reported 12 adverse events when taking warfarin plus entacapone, while seven out of 12 reported 14 adverse events when taking warfarin plus placebo. The most frequently experienced adverse events were gingival bleeding and headache, which were each reported eight times. Gingival bleeding occurred three times during the warfarin dose-finding phase, four times during the interaction phase with placebo, and once during the interaction phase with entacapone. Similarly, headache did not appear to be associated with entacapone and warfarin, as it occurred four times during the warfarin dose-finding and stabilization phase, but not during any interaction phase. Other events that were reported three times or more were fatigue, (twice during the dose-finding and stabilization phase, and once during the entacapone interaction phase). Pyuria was reported once during the dose-finding and stabilization phase, once in the entacapone interaction phase, and once during the recovery phase. Most adverse events were of mild intensity and all but one resolved completely. Pyuria continued in one subject after the study, but, apart from a short interruption, had been present before the study and was considered unrelated to treatment. No serious adverse events were reported during the study. Coagulation factors decreased during warfarin treatment to a similar extent in the entacapone and placebo phase. All occult blood smearing tests were negative for all subjects throughout the study and there was no pattern of abnormal vital signs.

Discussion

The objectives of this study were to determine whether the pharmacokinetics and/or the pharmacodynamics of warfarin were altered to a clinically significant extent by entacapone. In view of the racemic nature of warfarin and established examples of stereoselective inhibition of its metabolism by certain drugs [27–29], it was decided to assay both of its enantiomers. Their pharmacokinetics when warfarin was given in combination with placebo were in good agreement with data reported in the literature [30, 31].

The primary endpoints in the present study were the steady-state pharmacokinetic parameters for each of the enantiomers of warfarin when given with placebo or in combination with entacapone. The multiple-dose design for both warfarin and entacapone ensured that the study was conducted under clinically relevant conditions [32]. Entacapone 200 mg given four times daily corresponds to regimens used in PD therapy. A washout period was not considered necessary in view of the short elimination half-life of entacapone [18, 33]. Steady-state conditions for warfarin were confirmed by three consecutive INR values which did not differ by more than 0.2. The study demonstrated a small increase in the mean AUC_ss for the R- but not for the S-enantiomer of warfarin. The increase for R-warfarin was slightly outside the prespecified range for bioequivalence. Changes in other pharmacokinetic parameters for R-warfarin (C_max and t_max) were not detectable. None of the pharmacokinetic parameters of the S-enantiomer of warfarin was affected by coadministration of entacapone since all variables fell within the bioequivalence ranges. The data for the small effect of entacapone on the pharmacokinetics of R-warfarin, the less active isomer, are consistent with the small increase in PT that was observed. Other biochemical measurements of coagulation factors failed to reveal any significant alterations associated with entacapone.

The mechanism by which entacapone influences the pharmacokinetics of R-warfarin remains unknown. Theoretically, it may compete for binding to one or more of the enzymes that metabolize R-warfarin. Although CYP3A4 is the major enzyme involved, there is also a contribution from CYP1A2 and carbonyl reductase [22]. Entacapone which is itself glucuronidated, has thus far not been shown to interact with the CYP-mediated metabolism of other drugs.

The present findings support the notion that interpretation of data from clinical interaction studies with racemic warfarin should be based on analyses of the separate enantiomers, because the differences in metabolism, potency and kinetics between R- and S-warfarin may lead to misinterpretation of pharmacodynamic data [21].

In conclusion, in healthy subjects there is a small pharmacokinetic interaction between entacapone and R-warfarin, but not S-warfarin. Given that the R- is less potent than the S-enantiomer, the effect of entacapone is probably not clinically significant. In patients with Parkinson's disease, intensified monitoring of INR upon initiating combined treatment with warfarin and entacapone is not likely to be warranted.