Investigation of mutual pharmacokinetic interactions between macitentan, a novel endothelin receptor antagonist, and sildenafil in healthy subjects

Abstract

Aim

To study the mutual pharmacokinetic interactions between macitentan, an endothelin receptor antagonist, and sildenafil in healthy male subjects.

Methods

In this open-label, randomized, three way crossover study, 12 healthy male subjects received the following oral treatments: A) a loading dose of 30 mg macitentan on day 1 followed by 10 mg once daily for 3 days, B) sildenafil 20 mg three times a day for 3 days and a single 20 mg dose on day 4 and C) both treatments A and B concomitantly. Plasma concentration−time profiles of macitentan and its active metabolite ACT-132577 (treatments A and C) and sildenafil and its N-desmethyl metabolite (treatments B and C) were determined on day 4 and analyzed non-compartmentally.

Results

The pharmacokinetics of macitentan were not affected by sildenafil. In the presence of sildenafil C_max and AUC_τ of the metabolite ACT-132577 decreased with geometric mean ratios (90% confidence interval (CI)) of 0.82 (0.76, 0.89) and 0.85 (90% CI 0.80, 0.91), respectively. In the presence of macitentan, plasma concentrations of sildenafil were higher than during treatment with sildenafil alone, resulting in increased C_max and AUC_τ values. The respective geometric mean ratios were 1.26 (90% CI 1.07, 1.48) and 1.15 (90% CI 0.94, 1.41). The pharmacokinetics of N-desmethylsildenafil were not affected by macitentan. All treatments were well tolerated.

Conclusion

A minor, not clinically relevant, pharmacokinetic interaction was observed between macitentan and sildenafil. Based on these results, no dose adjustment of either compound appears necessary during concomitant treatment with macitentan and sildenafil.

What Is Already Known about This Subject

• Macitentan, a novel endothelin receptor antagonist, was recently approved for the long term treatment of pulmonary arterial hypertension (PAH).

• PAH patients may be concomitantly treated with sildenafil, an approved treatment for PAH, and macitentan. As both share the cytochrome P450 (CYP) 3A4 metabolic pathway, pharmacokinetic interactions might occur.

What This Study Adds

• Based on the results of this study, no dose adjustment of either macitentan or sildenafil is necessary during concomitant treatment.

• Results also indicate that macitentan, at clinically relevant concentrations, does not affect CYP3A4 activity.

Introduction

Pulmonary arterial hypertension (PAH) is characterized by a progressive elevation of pulmonary artery pressure and pulmonary vascular resistance, ultimately producing right ventricular failure and death. This serious condition can occur in the absence of a demonstrable cause (idiopathic or familial) or as a complication of congenital heart disease, systemic conditions such as connective tissue disease, particularly scleroderma, HIV infection or as result of the use of exogenous stimuli such as appetite suppressant drugs [1–4]. At present, specific targeted treatment options include endothelin receptor antagonists (ERAs), inhibitors of phosphodiesterase type 5 (PDE5-I), prostacyclin agonists and lung-heart transplantation [5–7]. Available drugs to treat PAH have positive effects but they do not provide a cure and in many patients the disease will still progress. Thus, new treatment strategies including combination therapy are being explored. Combining an ERA with a PDE5-I appears an attractive therapeutic option to address several pathophysiological mechanisms that are involved in PAH [8] as these classes of compounds are orally available and have different mechanisms of action, aiming at different pathophysiological pathways. A number of clinical studies have been performed exploring such a combination and preliminary results are positive [9–11].

Macitentan is a novel ERA, which was shown to reduce the risk of morbidity and mortality in PAH patients in a long term, event-driven phase III study and which was recently approved for the long term treatment of PAH by the United States Food and Drug Administration, Health Canada's Therapeutic Products Directorate and the European Commission [12–14]. In preclinical models, macitentan exhibited sustained receptor binding and enhanced tissue penetration in comparison with other ERAs [15,16].

In vitro, macitentan showed potential to inhibit murine permeability glycoprotein (P-gp), breast cancer resistance protein (BCRP) and solute carrier organic anion transporter family member 1B1 and 1B3 (SLCO1B1, and SLCO1B3). Further, in vitro, only moderate inhibition of cytochrome P450 (CYP) isoenzymes 3A4 and 2C19 was observed [17]. However, additional in vitro and in vivo studies, performed at clinically relevant concentrations, did not confirm these findings [18–20]. Hepatic uptake of macitentan is mostly driven by passive diffusion and is not dependent on organic anion-transporting polypeptide (OATP) transport unlike other ERAs such as bosentan [19,21]. Macitentan is metabolized predominantly by CYP3A4 with minor contributions of CYP2C8, CYP2C9, and CYP2C19 [20]. Clinical drug−drug interaction studies indicated moderate effects on macitentan exposure in the presence of the strong CYP3A4 inhibitor ketoconazole and CYP3A4 inducer rifampicin [19,22].

Pharmacokinetic studies in healthy subjects indicated that macitentan is slowly absorbed and eliminated with a terminal half-life (t_1/2) of about 16 h [23,24]. One metabolite, ACT-132577, was identified, which is also an ERA albeit with lower affinity for endothelin receptors than macitentan. This metabolite is formed via CYP3A4, has a t_1/2 of about 48 h and accumulated approximately 8.5-fold upon multiple dosing [22,24]. Measurement of urinary 6β-hydroxycortisol : cortisol ratio upon multiple dosing with macitentan indicated a small, not statistically significant, increase in this ratio when compared with placebo [24]. As the urinary 6β-hydroxycortisol : cortisol ratio is a non-invasive test for evaluating CYP3A4-mediated drug inhibition or induction, the increase in this ratio suggests a potential for macitentan to induce CYP3A4 mildly [24,25].

Sildenafil is an inhibitor of cyclic guanine monophosphate (cGMP) specific PDE5 in the smooth muscle of the pulmonary vasculature, where PDE5 is responsible for degradation of cGMP. Through this mechanism sildenafil increases cGMP within pulmonary vascular smooth muscle cells, resulting in relaxation. Sildenafil is rapidly absorbed after oral administration and C_max is reached within 30 to 120 min following oral dosing and it is metabolized predominantly via CYP3A4 (major route) and CYP2C9 (minor route) isoenzymes. The major circulating active metabolite results from N-desmethylation of sildenafil and is, itself, further metabolized. In healthy subjects, plasma concentrations of this metabolite are approximately 40% of those of sildenafil. Both sildenafil and the active metabolite have terminal half-lives of about 4 h [26].

It is likely that a large number of PAH patients will be treated with both macitentan and sildenafil. Both compounds are substrates of CYP3A4 and therefore could, theoretically, interact by competition for the same binding site. Further, based on the observations in the multiple dose ascending study, macitentan could impact on the pharmacokinetics of sildenafil through induction of CYP3A4. Also, other mechanisms of interaction, such as inhibition of drug transporter proteins, are possible as was recently shown for sildenafil [27,28]. Therefore, the present study was conducted to investigate the mutual pharmacokinetic interactions between these two compounds.

Methods

The study (AC-055-106) followed the principles of the Declaration of Helsinki and Good Clinical Practice and the protocol was approved by an independent Ethics Committee (Ethik-Kommission der Medizinischen Universität Wien und des Allgemeinen Krankenhauses der Stadt Wien AKH, Vienna, Austria). Written informed consent was obtained from all subjects prior to study start.

Study design and subjects

This was a single centre, open label, randomized, three way crossover study, in which healthy male subjects received the following oral treatments A) a loading dose of 30 mg macitentan on day 1 followed by 10 mg once daily for 3 days, B) sildenafil 20 mg three times daily for 3 days and a single 20 mg dose on day 4 and C) a loading dose of 30 mg macitentan on day 1 followed by 10 mg once daily for 3 days and sildenafil 20 mg three times daily for 3 days and a single 20 mg dose on day 4. Both macitentan and sildenafil are indicated for chronic use and, therefore, their mutual pharmacokinetic interactions were investigated under steady-state conditions. Because of the long t_1/2 of ACT-132577, a loading dose of macitentan was administered to reduce the time to steady-state from about 8 days to 4 days. Subjects were assigned to one of six possible treatment sequences and treatments were separated by a washout phase of at least 10 days. Selection of the doses and dosing regimens of macitentan (10 mg once daily) and sildenafil (20 mg three times daily) were based on approved dosing recommendations [26,29].

Twelve healthy male subjects aged between 18 and 45 years and with a body mass index between 18 and 28 kg m⁻² were included in this study. Subjects were included after a medical examination showing no clinically significant abnormalities. Subjects could not participate if they smoked, had a prior history of drug or alcohol abuse, were allergic to any drug, were using any medication or had participated in another clinical trial during the 3 month period preceding the screening examination.

Based on the coefficient of variation (CV%) of the variable AUC_τ for macitentan (26.5%) and sildenafil (28.7%), the study had 75 and 63% power to reject the null hypothesis that one of the limits of the 90% confidence interval (CI) of the ratio of the geometric mean of treatment C vs. the geometric mean of treatment A or B, was outside the interval 0.80–1.25 [24,28].

Study conduct

The screening examination was performed within 2 weeks of study start. The subjects were hospitalized from the evening of day −1 (treatments B and C) or day 3 (treatment A) until the last blood sample for pharmacokinetic purposes had been withdrawn in each treatment period. Assessments on other study days were done on an ambulatory basis. On day 4 of each treatment period, subjects consumed a standardized light breakfast before study drug intake. On all other days, drug intake was done irrespective of food intake. Smoking and the drinking of alcoholic beverages or xanthine-containing beverages were not permitted during the time in the clinic but the intake of water was ad libitum. Tolerability and safety were monitored by recording of treatment discontinuations, adverse events, vital signs, ECG, physical examination and clinical laboratory testing. The end of study examination took place at least 10 days after the last blood sample had been withdrawn in treatment period 3.

Sampling and bioanalytics

During treatments A and C, on study days 1 to 3, a predose 2 ml blood sample for measurement of macitentan and its active metabolite ACT-132577 was collected prior to the morning drug administration. After administration of study drug in the morning of day 4, blood samples of 2 ml were collected into EDTA-containing tubes from an indwelling catheter in an antecubital vein just before and at 1, 3, 5, 6, 7, 8, 9, 10, 12, 16 and 24 h for measurement of macitentan and ACT-132577 (treatment A) or at 0.33, 0.67, 1, 1.5, 2, 2.5, 3, 4, 5, 6 and 8 h for measurement of sildenafil and N-desmethylsildenafil (treatment B). During treatment C, 2 ml blood samples were withdrawn at all time points listed above for treatments A and B. However, at time points 1, 3, 5, 6 and 8 h, which were common to both treatments, two 2 ml blood samples were withdrawn. After collection, the tubes were centrifuged for 10 min at 1500 g and 4°C, the plasma was separated and stored at −20°C pending analysis.

Plasma concentrations of macitentan, sildenafil and their respective metabolites ACT-132577 and N-desmethylsildenafil were determined using validated liquid chromatography coupled to tandem mass spectrometry methods (LC-MS/MS) [30]. The macitentan and ACT-132577 assay was linear in the concentration range 0.5–1000 ng ml⁻¹ and the limit of quantification was 1.0 ng ml⁻¹ for both analytes. The method to measure sildenafil and its metabolite was validated in the range of 5 to 750 ng ml⁻¹ for sildenafil and 2 to 300 ng ml⁻¹ for N-desmethylsildenafil and the limits of quantification were 5 and 2 ng ml⁻¹, respectively. The performance of the methods was monitored using quality control samples. Precision (%CV) was ≤6.8% for macitentan, ≤6.5% for ACT-132577, ≤3.1% for sildenafil and ≤7.6% for N-desmethylsildenafil, whereas bias was ≤2.3%, ≤3.3%, ≤12.3% and ≤9.8% for macitentan, ACT-132577, sildenafil and N-desmethylsildenafil, respectively.

Data analysis and statistics

The pharmacokinetic variables C_max, t_max and AUC_τ of macitentan and sildenafil and their respective metabolites were determined by non-compartmental methods. C_max and the time to C_max (t_max) were read directly from the concentration−time data whereas the area under the plasma concentration−time curve over one dosing interval (AUC_τ) was estimated with the linear trapezoidal rule. Log transformed C_max and AUC_τ values were analyzed by linear mixed effect models tested in advance for period and carry-over effects. A potential carry-over effect was considered when the corresponding compound was given in the preceding period. If neither a period nor a carry-over effect was present, only treatment was included in the mixed effect model. If a period and/or a carry-over effect was present, the corresponding factor was included in addition to treatment. Geometric mean ratios and 90% confidence limits were calculated from the corresponding back-transformed contrasts of the mixed effect models.

Results

All 12 subjects (age range 20–33 years, body weight range 70.0–86.0 kg) completed the study according to protocol and were evaluable for pharmacokinetics. Trough concentrations of macitentan and its metabolite over time in the absence of sildenafil are shown in Figure 1. Those in the presence of sildenafil followed a similar profile (data not shown). For both compounds, steady-state conditions were reached by day 4. The plasma concentration−time profiles of macitentan and ACT-132577 in the presence and absence of sildenafil are shown in Figure 2 and the corresponding pharmacokinetic variables presented in Table 1. Figure 3 and Table 2 show the plasma concentration−time profiles and derived pharmacokinetic variables of sildenafil and its N-desmethyl metabolite, respectively. In the presence of sildenafil, no effect on the exposure to macitentan was seen whereas the plasma concentrations of ACT-132577 were lower when compared with macitentan administration alone. Concomitant administration of macitentan had no effect on N-desmethylsildenafil concentrations but those of sildenafil were about 20% higher. Statistical analysis revealed that the geometric mean ratios and their 90% CI were within the referenced bioequivalence limits of 0.80 to 1.25 for the variables C_max and AUC_τ for macitentan and N-desmethylsildenafil and for AUC_τ of ACT-132577. In contrast, either the lower (C_max ACT-132577) or upper (C_max and AUC_τ sildenafil) limit of the 90% CI was outside these bioequivalence limits (Table 3). No significant effect on t_max of any compound was observed in this study.

Figure 1.

Mean (±SD) trough plasma concentration vs. time profiles of macitentan and its metabolite ACT-132577 in the absence of sildenafil (n = 12). , macitentan; , ACT-132577

Figure 2.

Mean (±SD) plasma concentration−time profiles of macitentan (A) and its metabolite ACT-132577 (B) on day 4 following administration of a 30 mg loading dose of macitentan followed by a 10 mg dose for 3 days in the absence (treatment A, ○) and presence (treatment C, ▄) of sildenafil (n = 12). , macitentan; , macitentan + sildenafil

Table 1.

Plasma pharmacokinetic variables of macitentan and ACT-132577 in the presence (treatment C) or absence (treatment A) of sildenafil

Treatment	tmax (h)	Cmax (ng ml−1)	AUCτ (ng ml−1 h)
Macitentan
A	5.0 (3.0–10.0)	421 (346, 511)	6526 (5451, 7813)
C	6.0 (3.0–10.0)	414 (357, 481)	6946 (5829, 8278)
ACT-132577
A	10.0 (5.0–12.0)	909 (808, 1023)	18643 (16783, 20710)
C	10.0 (0.0–16.0)	747 (676, 827)	15925 (14396, 17616)

Data are geometric means (and 95% CI) or for t_max the median (and range), n = 12.

Figure 3.

Mean (±SD) plasma concentration−time profiles of sildenafil (A) and its N-desmethyl metabolite (B) on day 4 following administration of 20 mg three times daily for 3 days and a single 20 mg dose on day 4 in the absence (treatment B, Δ) and presence (treatment C, ▄) of macitentan (n = 12). , sildenafil; , macitentan + sildenafil

Table 2.

Plasma pharmacokinetic variables of sildenafil and N-desmethylsildenafil in the presence (treatment C) or absence (treatment B) of macitentan

Treatment	tmax (h)	Cmax (ng ml−1)	AUCτ (ng ml−1 h)
Sildenafil
B	1.5 (1.0–3.0)	60.9 (52.7, 70.3)	214 (180, 256)
C	1.5 (0.7-2.5)	76.7 (62.1, 94.6)	247 (190, 322)
N-desmethylsildenafil
B	1.8 (1.5-3.0)	19.9 (16.4, 24.0)	68.3 (57.7, 80.8)
C	2.0 (0.67-2.5)	21.8 (18.0, 26.3)	73.6 (56.3, 96.2)

Data are geometric means (and 95% CI) or for t_max the median (and range), n = 12.

Table 3.

Geometric mean ratios and 90% confidence intervals (CI) of C_max and AUC_τ comparing treatment C with treatments A (macitentan and ACT-132577) and B (sildenafil and N-desmethylsildenafil) (n = 12)

Variable	Lower 90% CI	Geometric mean ratio (treatment C : A)	Upper 90% CI
Macitentan
Cmax	0.92	0.99	1.06
AUCτ	1.01	1.06	1.12
ACT-132577
Cmax	0.76	0.82	0.89
AUCτ	0.80	0.85	0.91
Variable	Lower 90% CI	Geometric mean ratio (treatment C : B)	Upper 90% CI
Sildenafil
Cmax	1.07	1.26	1.48
AUCτ	0.94	1.15	1.41
N-desmethylsildenafil
Cmax	0.99	1.10	1.22
AUCτ	0.96	1.08	1.22

All three treatments were well tolerated. No serious adverse events occurred and no subject withdrew prematurely from the study. Of the 40 reported adverse events, 10 occurred during treatment A (macitentan alone), five during treatment B (sildenafil alone) and 25 during treatment C (macitentan + sildenafil). Mild to moderate headache was the most frequently reported adverse event (12 cases in total), which most notably occurred during treatment C (10 cases) whereas all other events, which included hypotension, increase in liver function tests and palpitations, were incidental, i.e. were reported only by one or two subjects per treatment. No pattern was observed to suggest a treatment-related effect on clinical laboratory variables, ECG and physical examination. When compared with baseline (predose assessment on day 1 of each treatment period), blood pressure decreased without effect on heart rate in all treatments. The maximum median decrease in systolic and diastolic blood pressure during these treatments was between 8.0 and 9.5 mmHg. The effect on diastolic blood pressure was more pronounced following concomitant administration of macitentan and sildenafil with a maximum median decrease of 15.5 mmHg, whereas the decrease in systolic blood pressure was similar to that observed during macitentan or sildenafil treatment alone.

Discussion

In this study, we investigated the mutual pharmacokinetic interactions between macitentan and sildenafil. Both compounds are substrates of CYP3A4 but have not been shown to inhibit or induce CYP isoenzymes at clinically relevant concentrations [18,19,24,26,31] and, thus, a pharmacokinetic interaction was not expected. Overall, the results of the current study confirm this hypothesis although some minor interactions were observed.

Concomitant sildenafil had no effect on macitentan concentrations and decreased those of ACT-132577 only slightly. Although ACT-132577 is a pharmacologically active metabolite, functional assays indicated a five-fold lower potency compared with macitentan [16]. Therefore, the observed small decrease in exposure (15% and 18% for AUC_τ and C_max, respectively) is not considered clinically relevant. A lack of effect of sildenafil on the pharmacokinetics of ambrisentan [32] was previously shown whereas this PDE5-I increased bosentan concentrations by 50% [28]. Sildenafil is known to be an inhibitor of OATP1B1 and OATP1B3 [27], which are involved in the hepatic uptake of substrate compounds [33]. Inhibition of these OATPs was shown to be the underlying mechanism for the sildenafil-induced increase in bosentan exposure [27,34]. Previous in vitro data and results from a drug-drug interaction study with ciclosporin in vivo indicated that macitentan and ACT-132577 are not substrates of OATP [19]. Therefore, the lack of an effect of sildenafil on macitentan concentrations would be consistent with these findings.

Macitentan slightly increased exposure to sildenafil (15% and 26% for AUC_τ and C_max, respectively) but had no effect on N-desmethylsildenafil. This small effect is not considered to be of clinical relevance because doses of up to 80 mg three times daily, i.e. four times the dose used in the present study, were well tolerated in patients with PAH [35,36] indicating that sildenafil has a wide therapeutic index. Furthermore, in a recent study it was shown that in the presence of bosentan the exposure to sildenafil decreased by 63% whereas in the presence of sildenafil the exposure to bosentan increased by 50% [28]. These pharmacokinetic interactions, which are more pronounced than those observed in the present study, have not led to a change in dosing regimen of either drug when used concomitantly [34,37]. Sildenafil is listed as a sensitive model substrate to study CYP3A4-mediated drug−drug interactions [38] and the present results thus indicate that macitentan is unlikely to affect to a clinically relevant extent the pharmacokinetics of other substrates of this enzyme. This is an important finding in view of the fact that CYP3A4 is involved in the metabolism of about 50% of marketed drugs [39]. Specific examples that are relevant for PAH patients are calcium channel blockers, anticoagulants and hormonal contraceptives. Warfarin is an anticoagulant drug with a narrow therapeutic index and a substrate of CYP3A4 but also of CYP2C9 and CYP1A2 [40]. New oral anticoagulants that selectively inhibit either thrombin or factor Xa have emerged on the market as they have more predictable anticoagulant effects without the need for routine monitoring [41,42]. However, the factor Xa inhibitors rivaroxaban and apixaban are substrates of CYP3A4 and PgP and use with strong inhibitors of these should be avoided [43]. As shown for all ERAs, macitentan displayed teratogenic effects in rats and rabbits and, therefore, reliable methods of contraception must be practiced by women of childbearing potential when treated with macitentan [20,29]. As numerous hormonal contraceptives are substrates of CYP3A4, it is relevant to understand the drug−drug interaction potential when given concomitantly with macitentan [29,44]. In addition, women with PAH should avoid becoming pregnant, as the physiological, cardiovascular and pulmonary changes that occur during pregnancy can exacerbate the condition. Based on the results of this study, it is unlikely that macitentan, at clinically relevant concentrations, is an inducer of CYP3A4 and the slight elevations of sildenafil concentrations would rather indicate a non-relevant inhibition of CYP3A4. It is therefore expected that macitentan does not interact with the pharmacokinetics of these drugs, and hence no dose adjustment would be required.

One of the shortcomings of this study was that the clinical relevance of the observed pharmacokinetic interactions could not be assessed in this study as it was performed in healthy subjects and no surrogate markers exist for the studied classes of compounds. As discussed above, the pharmacokinetic interactions observed in the present study are not considered to be of clinical relevance. Thus, no dose adjustment of either macitentan or sildenafil appears necessary when patients are treated with both drugs concomitantly. This is supported by results from the recently completed phase III study SERAPHIN, which investigated the effects of long term treatment with macitentan on morbidity and mortality. In this study, approximately 60% of the 742 PAH patients were on background treatment with a PDE5-I and consistent results were seen in patients on monotherapy and combination treatment [12,13]. Another shortcoming of the study is that in patients the pharmacokinetics of the investigated compounds may be different when compared with healthy subjects as was, for example, shown for bosentan [45]. Theoretically, this could have an influence on the study results. However, the pharmacokinetic interaction between bosentan and sildenafil was investigated in both healthy subjects [28] and PAH patients [46] and was similar in magnitude although in patients only the effect of bosentan on sildenafil pharmacokinetics was studied.

At the dosing regimens used in this study, administration of macitentan or sildenafil alone was well tolerated. With combined administration of both compounds more cases of headache were reported which in most subjects, however, were of mild intensity. Headache is a common side effect of both PDE5-Is [47] and ERAs [46,48]. It is unlikely that the increased incidence of headache can be explained by the pharmacokinetic interactions observed as these were limited in extent. Both compounds appeared to reduce blood pressure although the absence of a placebo treatment in this study precluded precise quantification of this decrease. Upon combined administration the decrease in blood pressure was more pronounced and this observation might well explain the increased incidence of headache, which is often associated with compounds that reduce blood pressure in healthy subjects [49]. It should be noted that in the phase III study SERAPHIN, which included a placebo control and was conducted in a large number of patients with a median treatment exposure of 115 weeks, this more pronounced decrease in blood pressure was not observed [12]. Therefore, the findings in the healthy subject group may be of limited relevance.

In conclusion, the mutual pharmacokinetic interactions between macitentan and sildenafil were characterized in this study and found to be of limited extent and clinical relevance. Based on these pharmacokinetic results, no dose adjustment of either compound appears necessary during concomitant treatment with macitentan and sildenafil. The findings of this study further indicate that macitentan is unlikely to be an inducer of CYP3A4 and, therefore, has a limited drug−drug interaction potential.

Competing Interests

All authors have completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author).

P.N. Sidharta and J. Dingemanse are current employees and P.L.M. van Giersbergen is a former employee of Actelion Pharmaceuticals Ltd, Allschwil, Switzerland. M. Wolzt was the principal investigator in this study and received funding from Actelion Pharmaceuticals Ltd to conduct the trial.