In Vitro Evaluation of CYP‐Mediated Metabolism of Fezolinetant and Pharmacokinetic Interaction Between Fezolinetant and Fluvoxamine in Healthy Postmenopausal Smokers and Nonsmokers

Abstract

Fezolinetant is an oral, nonhormonal, neurokinin 3 receptor antagonist treatment option for moderate to severe vasomotor symptoms associated with menopause. An in vitro study using human recombinant cytochrome P450 (CYP) enzymes and human liver microsomes showed that fezolinetant is metabolized to its major but inactive metabolite, ES259564, predominantly through CYP1A2, with minor contributions from CYP2C9 and CYP2C19. The clinical impact of CYP1A2 inhibition and induction on single‐dose pharmacokinetics of fezolinetant was assessed in an open‐label, single‐sequence, phase 1 study in healthy postmenopausal women, where the impact of fluvoxamine, a strong CYP1A2 inhibitor, and smoking, a moderate CYP1A2 inducer, were evaluated. In total, 18 participants, 9 of whom were smokers, were enrolled. Fezolinetant pharmacokinetics were evaluated after a single 30‐mg dose on Day 1 and Day 7. Fluvoxamine 50 mg was administered as a single dose on Days 3 and 10 and twice daily from Days 4 to 9. Fluvoxamine increased geometric mean ratio of fezolinetant maximum plasma concentrations (C_max) and area under the curve from time of dosing extrapolated to infinity (AUC_inf) to 182% and 939%, respectively, while ES259564 C_max decreased to 20.1% with no significant change in AUC. In smokers versus nonsmokers, when fezolinetant was administered alone, fezolinetant C_max and AUC_inf decreased to 71.7% and 48.3%, respectively, while ES259564 C_max increased to 130.2% and AUC_inf decreased to 81.8%. A single oral 30‐mg dose of fezolinetant was considered safe and well tolerated when co‐administered with fluvoxamine in healthy postmenopausal women.

Introduction

The menopausal transition is marked by fluctuations in hormone levels as ovarian function declines, resulting in a variety of physiologic changes and clinical symptoms. Among them, vasomotor symptoms (VMS), also known as hot flashes or hot flushes, are the most common complaint among women during menopause. About 80% of American women experience VMS during the menopausal transition. ¹ The duration of VMS from the time of onset is typically 7.4 years, ² but symptoms commonly persist for nearly a decade. ³ It is now understood that VMS associated with menopause are highly linked to the function of kisspeptin–neurokinin B (NKB)–dynorphin (KNDy) neurons. ⁴ The signaling of KNDy neurons is balanced by the negative feedback from estrogen/estrogen receptor alpha signaling and positive stimulation from NKB/neurokinin 3 (NK3) signaling.

Fezolinetant is an oral, nonhormonal NK3 receptor antagonist treatment option for moderate‐to‐severe VMS and is approved in many countries, including the United States, Europe, and Australia at a dose of 45 mg once daily. ⁵ , ⁶ , ⁷ , ⁸ Fezolinetant blocks the binding of NKB to the KNDy neurons to restore normal sensitivity of the thermoregulatory center in the hypothalamus. ⁹ , ¹⁰ Results from phase 2 trials have demonstrated a rapid and substantial reduction in VMS frequency and severity by fezolinetant. ⁹ , ¹⁰ Similar efficacy and safety were observed for once‐ and twice‐daily doses of fezolinetant in the phase 2b dose ranging study. ⁹ Fezolinetant efficacy, safety, and tolerability have been further characterized in phase 3 clinical trials, SKYLIGHT 1 and 2 (NCT04003155 and NCT04003142), which were 12‐week, randomized, placebo‐controlled trials of fezolinetant 30 and 45 mg once daily followed by a 40‐week active treatment extension period, ¹¹ , ¹² and a 52‐week safety study (SKYLIGHT 4; NCT04003389). ¹³

Fezolinetant is well absorbed and extensively metabolized, primarily to the oxidated metabolite ES259564. An in vivo metabolic profiling study based on mass balance samples suggested only fezolinetant and ES259564 are detected in human plasma. Although other metabolites were detected in human urine and feces, contributions of these metabolic pathways are minor (<10% in total). The major elimination pathway of fezolinetant was identified as a metabolism to ES259564. ES259564 is approximately 20‐fold less potent at the NK3 receptor than fezolinetant and is thus considered pharmacologically inactive. The metabolite‐to‐parent ratio ranges from 0.7 to 1.8. The metabolite is mainly excreted into urine, while fezolinetant urinary elimination is limited (1.1% of administered dose). ⁵

Fezolinetant pharmacokinetics (PK) have been well characterized in phase 1 studies in healthy male and female subjects. ⁵ , ¹⁴ When administered orally in healthy subjects, fezolinetant is rapidly absorbed, reaching maximum plasma concentrations (C_max) at a median time of 1.5‐2.0 h after dosing. Maximum plasma concentrations and total exposure (AUC) increased linearly across the 20‐ to 60‐mg dose range. In healthy women, the observed half‐life (t_1/2) was 4.2‐4.8 h at doses of 20‐60 mg, ¹⁴ and steady‐state trough concentrations were achieved by Day 2 following multiple once daily fezolinetant administration. ⁵

Given that the major elimination pathway of fezolinetant is metabolism to form ES259564, three in vitro assessments were conducted to assess (1) the in vitro metabolic activity to form ES259564 in recombinant human cytochrome P450s (CYPs), (2) the inhibitory effect of CYP isozyme‐specific inhibitors on ES259564 formation in human liver microsomes, and (3) the correlation of ES259564 formation with CYP isozyme‐specific activities in 16 different single‐donor human liver microsomes. Based on the results of these in vitro assessments, a clinical study was designed to evaluate the effect of concomitant administration of fluvoxamine, a strong CYP1A2 inhibitor, on the PK of a single oral 30‐mg dose of fezolinetant in healthy postmenopausal female subjects. Additionally, the study evaluated the effect of smoking, a CYP1A2 inducer, on the PK of fezolinetant and ES259564.

Methods

In Vitro Studies

Three in vitro studies were conducted to identify responsible CYP enzymes. First, recombinant CYP enzymes were used to evaluate formation of major metabolite, ES259564. Second, impact of CYP isoform‐specific inhibitors was evaluated in human pooled liver microsomes. In addition, 16 different single‐donor human liver microsomes (n = 16 donors, 1.0 mg protein/mL) were used to evaluate the correlation between the rate of ES259564 formation and CYP isoform‐specific substrate metabolic activities. In the preliminary assay with recombinant CYP enzymes and human liver microsomes, fezolinetant was highly stable and it was not possible to precisely determine the rate of degradation due to the low level of total metabolism. Therefore, the in vitro studies were conducted by evaluating the formation of ES259564.

Chemicals and Reagents

Fezolinetant, ES259564, and ES246567 (used as an internal standard) were synthetized at Onyx Scientific Limited (Sunderland, UK). The following copy DNA (cDNA)‐expressing recombinant CYP enzymes and control microsomes with P450 reductase from the baculovirus‐infected insect cell system were purchased from Corning Incorporated (NY): CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9 CYP2C19, CYP2D6, CYP2E1, CYP2J2, CYP3A4, CYP3A5, and CYP4F2. Pooled human liver microsomes (mixed sex, 50 donor pool) and a bank of human liver microsomes (Reaction Phenotyping Kit, version 8) were purchased by XenoTech, LLC (Kansas City, KS), and a bank of human liver microsomes were characterized for the following enzymatic activities: phenacetin O‐dealkylation (CYP1A2), coumarin 7‐hydroxylation (CYP2A6), bupropion hydroxylation (CYP2B6), amodiaquine N‐dealkylation (CYP2C8), diclofenac 4ʹ‐hydroxylation (CYP2C9), S‐mephenytoin 4ʹ‐hydroxylation (CYP2C19), dextromethorphan O‐demethylation (CYP2D6), chlorzoxazone 6‐hydroxylation (CYP2E1), testosterone 6β‐hydroxylation (CYP3A4/5), and lauric acid 12‐hydroxylation (CYP4A11).

Metabolic Reactions and Analytical Procedures

Fezolinetant (final concentration of 3 µmol/L) was incubated with 12 kinds of recombinant human CYPs (60 pmol P450/mL), pooled human liver microsomes (1.0 mg protein/mL), or single‐donor human microsomes (n = 16 donors, 1.0 mg protein/mL) at 37°C for 15 min in 50 mmol/L potassium phosphate buffer (pH 7.4) in the presence of an nicotinamide adenine dinucleotide phosphate (NADPH)‐generating system cofactor solution with a final concentration of NADP⁺ (1.3 mmol/L), glucose‐6‐phosphate (3.3 mmol/L), magnesium chloride (3.3 mmol/L), and glucose‐6‐phosphate dehydrogenase (0.4 U/mL) in polypropylene tubes, at a final incubation volume of 0.5 mL. Each experiment was performed in duplicate. Following pre‐incubation of the reaction mixtures (excluding fezolinetant) for 5 min, reactions were started by adding 0.005 mL of fezolinetant dissolved in DMSO/methanol (30:70, v/v).

In addition, the following typical CYP isoform inhibitors (final concentration) were added to assess their inhibitory impact in pooled human liver microsomes: furafylline (20 µmol/L) for CYP1A2, sulfaphenazole (3 µmol/L) for CYP2C9, S‐(+)‐N‐3‐benzylnirvanol (5 µmol/L) for CYP2C19, and quinidine (1 µmol/L) for CYP2D6. ¹⁵

After incubation for 15 min, the incubation mixture was mixed with the stop solution, which contained methanol and internal standard (ES246567, analog of fezolinetant, final concentration of 100 ng/mL) to terminate the metabolic reaction. The mixture was centrifuged (10,000 g, 4°C, 5 min) and the supernatant was collected and filtered through Ultrafree‐MC (0.45 µm, PTFE; Merck KGaA, Darmstadt, Germany) by centrifugation (5000 g, 4°C, 5 min). The filtrate was placed in an HPLC vial and analyzed by liquid chromatography‐tandem mass spectrometry (Shimadzu LC‐30A or LC‐10Avp HPLC system, Shimadzu Corporation, Kyoto, Japan; API 4000 or 4000QTRAP mass spectrometer, AB Sciex LLC, Framingham, MA) to determine the concentration of ES259564 using electrospray ionization in positive ion mode with mass transitions (m/z) 358.6 → 123.2 (fezolinetant, only monitoring), 374.8 → 123.2 (ES259564), and 405.9 →123.2 (ES246567, internal standard) on Kinetex C18 column (50 × 3.0 mm i.d., 2.6 µm, Phenomenex Inc., Torrance, CA) with a gradient of 10 mmol/L ammonium acetate solution/acetonitrile as the mobile phase. The method was validated over a range of 5‐1000 nmol/L for ES259564.

In correlation analysis, incubation mixtures with 16 different single‐donor human liver microsomes and a pooled sample were prepared. The correlation coefficient between ES259564 formation and each CYP marker activity, as given by the supplier, was calculated. For assessment of the correlativity, Pearson's product‐moment correlation coefficient was calculated to test the statistical significance of correlation using SAS release 9.2 (SAS Institute Inc., Cary, NC) and EXSUS version 7.7 (EPS Corporation, Tokyo, Japan). The significant levels were 5% for both sides. The CYP marker activities were described in “In Vitro Studies.”

Clinical Study Design

The clinical study was conducted at the site of Parexel International GmbH, Early Phase Clinical Unit Berlin Kliniken Westend Haus 31 (Berlin, Germany) in accordance with the ethical principles that have their origin in the Declaration of Helsinki, Good Clinical Practice, International Council for Harmonisation guidelines, and applicable laws and regulations. This study was approved by an independent ethics committee (Landesamt für Gesundheit und Soziales [State Office of Health and Social Affairs] Ethics Committee of the Land Berlin), and all subjects provided written informed consent.

This was a phase 1, open‐label, single‐sequence study conducted at one clinical unit in Germany in 2018 (EudraCT Number: 2017‐003656‐24). The primary objective of the study was to assess the effect of fluvoxamine and smoking on the PK of fezolinetant and its metabolite ES259564. A total of 18 subjects were planned to be enrolled, 8 of whom were planned to be smokers and 8 of whom were planned to be nonsmokers, with at least 16 subjects planned to be completed. No formal sample size calculation was performed. However, based on data from the previous clinical trial, ¹⁴ the intrasubject coefficient of variation (CV) for pharmacokinetic parameters AUC_inf and C_max of fezolinetant was estimated to be between 13% and 18%, and the intersubject CV for the same parameters was estimated to be between 17% and 29%. Assuming the underlying intrasubject variability was similar to 18%, and an observed ratio between fezolinetant with fluvoxamine and fezolinetant alone of 400%, the 90% CI was to lie within (352, 455) with >80% probability. Assuming the underlying intersubject variability was similar to 29%, and an observed ratio between smokers and nonsmokers of 50%, the 90% CI was to lie within (38, 66) with >80% probability.

After giving written informed consent, subjects were screened for up to 22 days prior to study drug administration on Day 1. The end‐of‐study visit occurred 5‐9 days after discharge from the clinical unit or early withdrawal from the study. On Day 1, a single oral dose of 30 mg (2 × 15‐mg capsules) fezolinetant was administered in the morning after an overnight fast of 10 h followed by pharmacokinetic blood sampling. On Day 3 in the evening and on Day 10 in the morning, a single oral dose of 50‐mg fluvoxamine (Fevarin®) was administered. On Days 4‐9, oral doses of 50‐mg fluvoxamine were administered twice daily with a 12‐h interval. Fluvoxamine was administered under standardized fed conditions except for the morning dose on Day 7. On Day 7 (fifth day of fluvoxamine dosing), a single oral dose of 30‐mg fezolinetant was administered simultaneously with the morning dose of fluvoxamine under fasted conditions, followed by pharmacokinetic blood sampling. Restrictions on previous and concomitant medications other than the study drugs, exercise, smoking, and dietary and fluid intake were employed to reduce confounding factors.

Study Population

Eighteen healthy postmenopausal female subjects aged >40 to ≤65 years, with a body weight of at least 50 kg and a body mass index of 18.5‐30.0 kg/m², were enrolled in the study. “Postmenopausal” was defined as subjects who had spontaneous amenorrhea for ≥12 months and follicle stimulating hormone (FSH) >40 IU/L or had bilateral oophorectomy ≥6 weeks before the screening visit (with or without hysterectomy). Smokers were defined as having smoked ≥10 cigarettes per day in the 4 months prior to screening and had cotinine levels of 1000 ng/mL or more at screening and upon admission to the clinical unit. Smokers were instructed to continue smoking in designated areas as per usual routine during the study. Nonsmokers were defined as having not smoked in the past 6 months prior to screening and had cotinine levels <200 ng/mL at screening and upon admission to the clinical unit. Other exclusion criteria included any deviation from the normal range of blood pressure or pulse rate, any clinically significant abnormalities in electrocardiogram or laboratory tests, concurrent or previous drug allergy, any clinically significant cardiovascular, respiratory, hepatic, renal or gastrointestinal disease, and excessive alcohol use.

Study Objectives

The primary objective was to evaluate the effect of fluvoxamine on plasma fezolinetant PK parameters after a single oral dose of fezolinetant, including C_max and area under the plasma concentration–time curve from the time of dosing and extrapolated to infinity (AUC_inf). The secondary objectives were to evaluate the effect of smoking on plasma fezolinetant PK parameters, with and without concomitant administration of fluvoxamine. Safety and tolerability were also evaluated. An exploratory objective was to evaluate the effect of fluvoxamine and the effect of smoking on the PK of the metabolite ES259564.

Sample Collection and Analysis

Blood sampling for plasma concentrations of fezolinetant and ES259564 were collected predose and 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12, 16, 24, 32, 40, 48, and 96 h postdose on Day 1 and Day 7. On days when fluvoxamine was administered alone, the blood samples were collected prior to fluvoxamine dosing. Blood samples were collected into tubes containing lithium heparin via a peripherally placed intravenous cannula or by direct venipuncture in a suitable forearm vein. Plasma was separated and shipped to SGS Life Science (Wavre, Belgium), in which concentrations of fezolinetant and ES259564 were measured using a validated ultra‐high performance liquid chromatography‐tandem mass spectrometry (UHPLC‐MS/MS) method. In brief, fezolinetant, ES259564 and the internal standard (ES246567) were extracted from 50 µL of plasma by solid‐phase extraction and separated by reversed‐phase liquid chromatography on a phenylhexyl column (ACE Ultra‐Core SuperPhenylHexyl chromatographic column) using a gradient of water/methanol, with 2 mmol/L ammonium acetate and 0.05% trifluoroacetic acid as the mobile phase. An API4000 (AB/MDS SCIEX, Nieuwerkerk a/d IJssel, The Netherlands) with a turbo ion spray interface in positive ion mode was used for the detection of mass transitions (m/z) 359.0 → 123.0 (fezolinetant), 375.0 → 123.0 (ES259564), and 406.0 → 123.0 (ES246567, internal standard). The method was validated over a range of 1‐1000 ng/mL for both fezolinetant and ES259564, and inter‐run accuracy and precision were within acceptance criteria per FDA, EMA, and ICH M10 requirements.

Pharmacokinetic Analysis

Pharmacokinetic parameters were calculated using Phoenix (Certara L.P., Princeton, NJ). Actual sampling times were used in the calculations. The following PK parameters were obtained for the study: C_max, time of C_max (t_max), AUC_inf, and terminal elimination half‐life (t_1/2). The primary PK endpoints for statistical evaluation were C_max and AUC_inf.

Statistical Analysis

All data processing, summarizing, and analysis were performed using SAS (Version 9.3, SAS Institute Inc., North Carolina). Demographic and baseline characteristics and pharmacokinetic parameters were summarized using descriptive statistics for continuous endpoints, and frequency and percentage for categorical endpoints.

The pharmacokinetic analysis set consisted of all subjects who took at least one dose of study drug and for which concentration data were available to facilitate derivation of at least one primary pharmacokinetic parameter. To assess the effect of fluvoxamine on the PK of fezolinetant, a mixed effects analysis of variance (ANOVA) model with fixed effect for treatment (fezolinetant + fluvoxamine and fezolinetant alone) and subject as a random effect was fitted on natural logarithm‐transformed C_max and AUC_inf (AUC_last would have been used in case AUC_inf could not be used). The model was fitted separately for all subjects, smokers and nonsmokers.

To assess the effect of smoking on the PK of fezolinetant alone and in combination with fluvoxamine, an ANOVA model with fixed effect for smoking status (smoker and nonsmoker) was fitted separately for the two treatments (fezolinetant alone and fezolinetant + fluvoxamine).

Within the ANOVA, the least‐squares (LS) mean differences between fezolinetant in combination with fluvoxamine and fezolinetant alone, or differences between smoker and nonsmoker along with 90% confidence interval (CI) for the differences were estimated. The LS means for C_max and AUC_inf were back‐transformed to produce the geometric LS means and presented with the number of subjects for each treatment or smoking status. The geometric LS mean ratios and their corresponding 90% CIs for each pharmacokinetic parameter were expressed as percentages. These analyses were also performed to assess the effect of fluvoxamine or smoking on the PK of ES259564.

Safety and Tolerability Assessment

Safety was assessed based on nature, frequency, and severity of adverse events (AEs), vital signs (supine blood pressure and pulse), and clinical laboratory tests (hematology, biochemistry, and urinalysis). AEs were collected from the time of receipt of the informed consent until the end of the study. Safety data were reported for all subjects who took at least one dose of study drug.

Results

In Vitro CYP Identification Involved in the Metabolism of Fezolinetant

As shown in Figure 1a, after incubation with recombinant human CYP enzymes, the metabolic activity to form ES259564 was observed with CYP1A2, CYP2C9, CYP2C19, and CYP2D6. ES259564 was not detected after incubation with the other isoforms. The contribution of CYP isoforms was assessed by human pooled liver microsomes with CYP isoform‐specific inhibitors and correlation analysis in individual human liver microsomes. In human pooled liver microsomes with CYP isoform‐specific inhibitors for CYP1A2, CYP2C9, CYP2C19, or CYP2D6, the relative inhibition ratios of ES259564 formation were highest in the presence of furafylline (CYP1A2) with an inhibition ratio of 83.3%, followed by sulfaphenazole (CYP2C9) and S‐(+)‐N‐3‐benzylnirvanol (CYP2C19). The inhibition ratio of CYP2C9 was 17.5% and CYP2C19 was 22.0%. Quinidine (CYP2D6) did not inhibit metabolic activity (Figure 1b).

Figure 1.

Metabolism of fezolinetant in cDNA‐expressed CYP enzymes (a), in human pooled liver microsomes with CYP isoform‐specific inhibitors (b). cDNA, complementary DNA. Data are the mean of duplicate experiments

The results of the correlation analysis are shown in Table 1. The metabolic activity of ES259564 formation in human liver microsomes was strongly correlated with the activity of phenacetin O‐dealkylation (P < .001), a marker activity of CYP1A2. Weaker correlations were observed with the activities of diclofenac 4ʹ‐hydroxylation (P < .05) and lauric acid 12‐hydroxylation (P < .05), marker activities of CYP2C9 and CYP4A11, respectively. Taken together, the formation of ES259564 from fezolinetant in human liver microsomes is primarily mediated by CYP1A2, and to a lesser extent by CYP2C9 and CYP2C19.

Table 1.

Correlation Between Activities of ES259564 Formation and the CYP Isoform‐Specific Activities in 16 Single‐Donor Human Liver Microsomes.

Metabolites	Pearson Product‐Moment Correlations (r)
	CYP1A2	CYP2A6	CYP2B6	CYP2C8	CYP2C9	CYP2C19	CYP2D6	CYP2E1	CYP3A4/5	CYP4A11
	0.8076	0.1911	−0.1053	0.4803	0.5412	0.2857	0.3788	−0.1308	0.1107	0.5279
ES259564	P < .001	NS	NS	NS	P < .05	NS	NS	NS	NS	P < .05

CYP1A2, phenacetin O‐dealkylation; CYP2A6, coumarin 7‐hydroxylation; CYP2B6, bupropion hydroxylation; CYP2C8, amodiaquine N‐dealkylation; CYP2C9, diclofenac 4’‐hydroxylation; CYP2C19, S‐mephenytoin 4’‐hydroxylation; CYP2D6, dextromethorphan O‐demethylation; CYP2E1, chlorzoxazone 6‐hydroxylation; CYP3A4/5, testosterone 6β‐hydroxylation; CYP4A11, lauric acid 12‐hydroxylation; NS, no significant difference.

Subject Disposition in Clinical Interaction Study

A total of 18 subjects, of whom 9 (50.0%) subjects were smokers and 9 (50.0%) subjects were nonsmokers, were enrolled and completed the study. All 18 subjects completed all study procedures and were included into both the safety and PK analysis set.

All subjects were white, postmenopausal females ranging in age from 45 to 64 years, with a mean of 54.5 years. Weight ranged from 54.4 to 92.6 kg, with a mean of 65.5 kg, and BMI ranged from 18.6 to 27.5 kg/m², with a mean of 23.4 kg/m². No obvious demographic differences were observed between smokers and nonsmokers.

Effect of Fluvoxamine on Plasma PK of Fezolinetant and ES259564

The plasma concentration–time profile of fezolinetant and ES259564 after fezolinetant administration alone (Day 1) and after co‐administration with fluvoxamine (Day 7) are shown in Figure 2. Pharmacokinetic parameters of fezolinetant and ES259564 in the presence or absence of fluvoxamine and statistical analysis are summarized in Table 2. When administered alone, fezolinetant was rapidly absorbed with a median t_max of 1.00 h. After reaching C_max, fezolinetant concentrations appeared to decline in a near‐monophasic manner with a mean t_1/2 of 4.01 h. After the co‐administration with fluvoxamine, the mean fezolinetant exposure, as measured by C_max and AUC_inf, increased relative to fezolinetant administered alone. Consistent with the increase in exposure after co‐administration with fluvoxamine, t_1/2 increased relative to fezolinetant administration alone. Based on the geometric LS mean ratio, the presence of fluvoxamine increased to 182% for C_max and 939% for AUC_inf. These effects of fluvoxamine on fezolinetant exposure are the same between smokers and nonsmokers. When fezolinetant was administered alone, ES259564 formed rapidly with a median t_max of 1.50 h. Co‐administration with fluvoxamine delayed the t_max of ES259564, and the geometric LS mean ratio of C_max was reduced to geometric LS mean ratio of 20.10% with no significant change in AUC.

Figure 2.

Mean (SD) plasma concentration–time profile of fezolinetant (a) and ES259564 (b) after fezolinetant administration alone and after co‐administration with fluvoxamine in all subjects (i.e., smokers and nonsmokers).

Table 2.

Effect of Fluvoxamine Co‐Administration on Fezolinetant and ES259564 Pharmacokinetics: Summary of Fezolinetant and ES259564 Plasma Pharmacokinetic Parameters and Statistical Assessment after Fezolinetant Administration Alone and after Co‐administration with Fluvoxamine in All Subjects.

		Geometric Mean (CV%)
Analyte	Parameter		Fezolinetant Alone (n  =  18)	Fezolinetant + Fluvoxamine (n  =  18)	Geometric Mean Ratio (%) a	90% CI of Ratio (%) a
Fezolinetant	Cmax (ng/mL)	Geometric LS mean	233	424	181.64	(161.79, 203.93)
		Arithmetic mean ± SD (CV%)	244 ± 72.6 (29.7)	432 ± 83.7 (19.4)
	tmax (h)	Median [min–max]	1.00 [0.983‐2.00]	1.50 [1.00‐4.00]	‐	‐
	AUCinf (ng•h/mL)	Geometric LS mean	941	8840	939.44	(788.06, 1119.91)
		Arithmetic mean ± SD (CV%)	1040 ± 473 (45.7)	10,100 ± 5360 (52.9)
	t1/2 (h)	Arithmetic mean ± SD (CV%)	4.01 ± 1.25 (31.2)	22.4 ± 10.8 (48.4)	‐	‐
ES259564
	Cmax (ng/mL)	Geometric LS mean	239	48	20.1	(17.06, 23.68)
		Arithmetic mean ± SD (CV%)	246 ± 61.7 (25.1)	52.1 ± 21.5 (41.3)
	tmax (h)	Median [min–max]	1.50 [1.00‐3.00]	12.0 [1.48‐23.9]	‐	‐
	AUCinf (ng•h/mL)	Geometric LS mean	1680	1810	107.97	(102.81, 113.39)
		Arithmetic mean ± SD (CV%)	1710 ± 358 (20.9)	1850 ± 400 (21.5)
	t1/2 (h)	Arithmetic mean ± SD (CV%)	7.01 ± 2.39 (34.1)	23.3 ± 10.3 (44.4)	‐	‐

AUC_inf, area under the curve from time of dosing extrapolated to infinity; CI, confidence interval; C_max, maximum plasma concentrations; CL/F, apparent total systemic clearance after extravascular dosing; CV, coefficient of variation; LS, least‐squares; t_1/2, terminal elimination half‐life; t_max, time of maximum concentration.

Values are presented as mean ± SD (CV%), except for t_max, which is presented as the median (range).

^a Ratios and confidence intervals were transformed back to raw scale and values are expressed as percentages

Effect of Smoking on Plasma PK of Fezolinetant and ES259564

Mean plasma concentrations in smokers and nonsmokers with or without fluvoxamine are presented in Figure 3. PK parameters presented by smoking status and statistical analysis are presented in Table 3. Smoking resulted in decreased fezolinetant exposure. In smokers, when fezolinetant was administered alone, fezolinetant C_max decreased to a geometric LS mean ratio of 71.7% while AUC_inf decreased by about half to a geometric LS mean ratio of 48.3%. A similar decrease was observed when fezolinetant was co‐administered with fluvoxamine.

Figure 3.

Mean (SD) plasma concentration–time profile of fezolinetant (a) and ES259564 (b) after fezolinetant administration alone and after co‐administration with fluvoxamine in all subjects and by smoking status.

Table 3.

Effect of Smoking on Fezolinetant and ES259564 Pharmacokinetics: Summary of Fezolinetant and ES259564 Plasma Pharmacokinetic Parameters and Statistical Assessment after Fezolinetant Administration Alone and after Co‐Administration with Fluvoxamine by Smoking Status.

		Geometric mean (CV%)
			Fezolinetant Alone		Fezolinetant + Fluvoxamine
Analyte	Parameter		Nonsmoker (n = 9)	Smoker (n = 9)	Nonsmoker (n = 9)	Smoker (n = 9)	Geometric Mean Ratio (%) a	90% CI of Ratio (%) a
Fezolinetant	Cmax (ng/mL)	Geometric LS mean	276	198	450	399	71.74	(57.04, 90.22)
		Arithmetic mean ± SD (CV%)	282 ± 63.2 (22.4)	207 ± 63.9 (30.9)	458 ± 84.2 (18.4)	407 ± 79.5 (19.6)
	tmax (h)	Median [min–max]	1.00 [0.983‐2.00]	1.00 [1.00‐1.50]	1.50 [1.00‐4.00]	1.50 [1.48‐4.00]
	AUCinf (ng•h/mL)	Geometric LS mean	1350	654	12,000	6490	48.29	(39.14, 59.58)
		Arithmetic mean ± SD (CV%)	1400 ± 391 (27.9)	670 ± 154 (23.0)	13,200 ± 5530 (41.9)	7060 ± 3030 (42.9)
	t1/2 (h)	Arithmetic mean ± SD (CV%)	4.79 ± 1.03 (21.5)	3.23 ± 0.948 (29.3)	28.8 ± 10.6 (36.9)	16.0 ± 6.64 (41.6)
ES259564
	Cmax (ng/mL)	Geometric LS mean	209	273	38.6	59.7	130.20	(109.43, 154.92)
		Arithmetic mean ± SD (CV%)	211 ± 25.2 (12.0)	281 ± 68.2 (24.2)	41.0 ± 14.1 (34.3)	63.2 ± 22.6 (35.7)
	tmax (h)	Median [min–max]	1.50 [1.00‐3.00]	1.50 [1.00‐3.00]	8.00 [1.50‐23.9]	12.0 [1.48‐12.0]
	AUCinf (ng•h/mL)	Geometric LS mean	1860	1520	1880	1750	81.84	(70.34, 95.22)
		Arithmetic mean ± SD (CV%)	1880 ± 347 (18.4)	1540 ± 293 (19.0)	1920 ± 418 (21.7)	1780 ± 392 (21.9)
	t1/2 (h)	Arithmetic mean ± SD (CV%)	6.41 ± 1.09 (16.9)	7.61 ± 3.19 (41.9)	30.1 ± 10.0 (33.4)	16.5 ± 4.71 (28.5)

AUC_inf, area under the curve from time of dosing extrapolated to infinity; CL/F, ratio of apparent clearance to bioavailability; C_max, maximum plasma concentrations; CV, coefficient of variation; t_1/2, terminal elimination half‐life; t_max, time of maximum concentration.

Values are presented as mean ± SD (CV%), except for t_max, which is presented as the median (range).

^a Ratios and confidence intervals were transformed back to raw scale and values are expressed as percentages

For ES259564, the C_max geometric LS means were higher for smokers relative to nonsmokers following both fezolinetant co‐administration with fluvoxamine and fezolinetant administration alone. The AUC_inf was slightly lower for smokers relative to nonsmokers. When fezolinetant was administered alone, ES259564 formed rapidly in smokers (median t_max of 1.50 h) and individual values showed minimal variation (range: 1.00‐3.00 h). Relative to nonsmokers, ES259564 C_max increased (geometric LS mean ratio of 130.2%) and AUC_inf decreased for smokers (81.8%). The impact of smoking was of the same magnitude when fezolinetant was administered in combination with fluvoxamine.

Safety

There were no serious AEs, or other significant treatment‐emergent AEs (TEAEs) that led to withdrawal of treatment during the conduct of the study.

Overall, 18 TEAEs were reported for 8 (44.4%) subjects enrolled in this study. TEAE incidence following co‐administration of fezolinetant, and fluvoxamine showed a minimal increase compared to the TEAE incidence following administration of fezolinetant alone.

The most commonly reported TEAEs were dizziness (reported for one subject after receiving fluvoxamine alone and two subjects after receiving fezolinetant + fluvoxamine) and orthostatic intolerance (reported for one subject after receiving fluvoxamine alone and one subject after receiving fezolinetant + fluvoxamine) within the system organ class (SOC) nervous system disorders, and erythema (reported for two [11.1%] subjects after receiving fluvoxamine alone) within the SOC skin and subcutaneous tissue disorders. No other TEAE was reported for >1 subject. All TEAEs reported were considered mild in severity, except for one TEAE of catheter site hematoma and one TEAE of orthostatic intolerance, which were considered moderate in severity. Only one TEAE reported for 1 (5.6%) subject after receiving fluvoxamine alone was considered by the investigator to be drug related.

Discussion

CYP contribution in the formation of ES259564, a major metabolite of fezolinetant, was evaluated in vitro by a study using human recombinant CYP enzymes and human liver microsomes. The results support a significant contribution of CYP1A2 with a minor contribution of CYP2C9 and CYP2C19 in fezolinetant metabolism. The metabolism of fezolinetant was possibly affected mostly by CYP1A2 metabolic activity.

In the clinical study, plasma concentrations of fezolinetant and ES259564 were measured to determine the effect of a CYP1A2 strong inhibitor, fluvoxamine, and a CYP1A2 inducer, smoking. Co‐administration of fezolinetant with fluvoxamine resulted in approximately 9‐fold increase in total exposure to fezolinetant due to a decrease in metabolic clearance resulting from fluvoxamine‐mediated inhibition of CYP1A2. In addition to the increase in exposure, the half‐life increased by almost 5.6‐fold, from 4.1 to 22.4 h. The magnitude of exposure increase (geometric LS mean ratio of 939% in total exposure and 182% in peak exposure) was similar between smokers and nonsmokers, suggesting that despite induction of CYP1A2 by smoking, the inhibitory effect of fluvoxamine on CYP1A2 was similar between smokers and nonsmokers. Fluvoxamine inhibits not only CYP1A2 but all CYP enzymes that can metabolize fezolinetant (CYP2C9 and CYP2C19). ¹⁶ The observed magnitude of increase in total exposure to fezolinetant might reflect the inhibition of multiple metabolic pathways by fluvoxamine.

Co‐administration of fezolinetant with fluvoxamine resulted in a large decrease in the peak exposure to the metabolite, ES259564, to 20%, with a shift in t_max from 1.50 to 12.0 h, suggesting a slower rate of formation of the metabolite due to inhibition of CYP1A2. A large increase in half‐life from 7.01 to 23.3 h and minimal increase in total exposure to geometric LS mean of 108% were observed. Given the fact that the main elimination of ES259564 is urinary excretion, the impact of co‐administration of fluvoxamine may have limited impact on its elimination rate. The half‐life of ES259564 after co‐administration with fluvoxamine was similar to that of fezolinetant (22.4 h for fezolinetant and 23.3 h for ES259564), indicating that the effect of fluvoxamine on ES259564 is primarily on the rate of metabolite formation rather than its elimination. This can be further supported by the fact that there was minimum change in the AUC.

Fezolinetant exposure was decreased in smokers compared with nonsmokers. When fezolinetant was administered alone in smokers, fezolinetant total exposure was about half of the exposure in nonsmokers, which is explained by the induction of CYP1A2 by smoking. When fezolinetant was dosed with fluvoxamine, similar decreases were observed in both smokers and nonsmokers, suggesting a similar inhibitory effect of fluvoxamine in both smokers and nonsmokers. The increase in apparent total systemic clearance and decrease in half‐life were consistent with smoking increasing CYP1A2 activity.

When fezolinetant was administered alone in smokers, ES259564 peak exposure slightly increased to geometric LS mean of 130%, while total exposure slightly decreased to 81.8%. A faster rate of formation may contribute the slightly higher peak exposure, although the impact was limited.

CYP1A2 inducibility has been reported to have a high interindividual variability. In addition, CYP1A2 has genetic polymorphisms that may affect inducibility, occurring at different frequencies in different races. ¹⁷ All subjects enrolled in this study were White, and the frequency of the alleles associated with reduced inducibility is low in this population. It is possible that the study population led to a slightly overestimated smoking impact.

However, the number of cigarettes smoked per day and plasma or urine cotinine level also correlate with CYP1A2 induction activity, with some further variability. ¹⁸ , ¹⁹ , ²⁰ , ²¹ As per protocol, a “smoker” in this study was defined as a participant who smoked ≥10 cigarettes per day in the 4 months prior to screening, and had cotinine levels of 1000 ng/mL or more at screening and upon admission to the clinical unit. While the magnitude of CYP1A2 induction caused by smoking in each subject was not evaluated in this study, the degree of smoking seems adequate to evaluate the impact of CYP1A2 induction. Furthermore, the observed fezolinetant exposure decrease of about 50% by smoking is similar to the reported exposure decrease in other CYP1A2 sensitive substrates such as caffeine. ²² , ²³

In conclusion, the inhibition and induction of CYP1A2 affected the exposure to fezolinetant, with only a minor effect on ES259564 total exposure. A single dose of fezolinetant was well tolerated in healthy postmenopausal females when co‐administered with fluvoxamine, although fluvoxamine increased the fezolinetant exposure to a large extent. Due to the magnitude of increase in exposure with an index strong inhibitor of CYP1A2, the use of strong inhibitors of CYP1A2 is contraindicated with fezolinetant. Since smoking decreased the fezolinetant exposure by about 50%, participants in the large‐scale, phase 3 studies were stratified based on their smoking status (active smoker or nonsmoker [former/never]). ¹¹ , ¹² , ¹³ Phase 3 data demonstrated a lack of impact of smoking on fezolinetant efficacy. ²⁴ Analysis of pooled 12‐week data from SKYLIGHT 1 and 2 studies showed an LS mean difference in VMS frequency of −3.48 (95% CI: −5.19, −1.77) for fezolinetant 45 mg versus placebo in current smokers. For former/never smokers, an LS mean difference of −2.32 (95% CI: −3.08, −1.57) was obtained for the same comparison between groups. Therefore, the exposure changes observed by CYP1A2 induction were not regarded as clinically meaningful differences.

Conclusions

The results of the present study support the primary contribution of CYP1A2 in fezolinetant metabolism. A CYP1A2 strong inhibitor, fluvoxamine, increased fezolinetant total exposure by a large magnitude of about 9‐fold, while smoking, which induces CYP1A2, reduced exposure by approximately half. Since the major metabolite ES259564 was formed mainly by CYP1A2 and eliminated via urinary excretion, the rate of formation was affected by CYP1A2 inhibition and induction, but its elimination was not affected. A single dose of fezolinetant with co‐administration of fluvoxamine was well tolerated. Based on the increase in fezolinetant exposure when co‐administered with a strong CYP1A2 inhibitor, concomitant use in patients is contraindicated. Although its exposure is moderately reduced, no dose modification is recommended for smokers based on the lack of impact on fezolinetant efficacy.

Author Contributions

Jiayin Huang, Melanie Patton, Jace Nielsen, Angela Sinn, and Tomasz Wojtkowski made substantial contributions to the study design; Jiayin Huang, Melanie Patton, Angela Sinn, and Tomasz Wojtkowski acquired study data; Jiayin Huang, Jace Nielsen, Peter Bonate, and Tomasz Wojtkowski analyzed the study data; and Megumi Iwai, Jiayin Huang, Xuegong Wang, Mayuko Miyagawa, Jace Nielsen, and Peter Bonate interpreted the study data.

Conflicts of Interest

Megumi Iwai, Jace Nielsen, Mayuko Miyagawa, Melanie Patton, Peter Bonate, Xuegong Wang, and Tomasz Wojtkowski are employees of Astellas Pharma Global Development, Inc. and Astellas Pharma, Inc. and may hold stock or stock options.

Funding

This study was funded by Astellas Pharma, Inc. Medical writing support was provided by Sue Cooper from Envision Pharma Ltd. (Horsham, UK) and funded by the study sponsor.

Data Sharing

Researchers may request access to anonymized participant level data, trial level data and protocols from Astellas sponsored clinical trials at www.clinicalstudydatarequest.com. For the Astellas criteria on data sharing see: https://clinicalstudydatarequest.com/Study‐Sponsors/Study‐Sponsors‐Astellas.aspx