Pharmacokinetics and Safety Evaluation of Maribavir in Healthy Japanese and Matched White Participants: A Phase 1, Open‐Label Study

This phase I study compared pharmacokinetics and safety of maribavir in Japanese and White participants, and evaluated dose proportionality in Japanese participants. Under fasting conditions, 12 healthy adult participants of Japanese descent and 12 matched White participants received a single 400‐mg dose of maribavir. Japanese participants received 2 further doses of maribavir: 200 mg and 800 mg, or 800 mg and 200 mg, separated by a ≥72‐hour washout period. Serial blood samples were collected up to 24 hours after dosing for pharmacokinetic assessments. Following the 400‐mg dose, the geometric mean ratios (90% confidence interval) of Japanese versus White participants were 110% (91.7%–133%) for maximum plasma concentration, 122% (96.8%–155%) for area under the plasma concentration–time curve (AUC) from time of dosing to the last measurable concentration, and 125% (98.0%–160%) for AUC extrapolated to infinity. In Japanese participants, maribavir AUC extrapolated to infinity and AUC from time of dosing to the last measurable concentration increased in a dose‐proportional fashion over 200–800 mg; maximum plasma concentration increased less than dose proportionally. Seven participants reported treatment‐emergent adverse events (TEAEs; Japanese participants, 400 mg: 2 [16.7%], 200 mg: 1 [8.3%]; White participants, 400 mg: 4 [33.3%]), all mild and most commonly dysgeusia. No serious TEAEs or TEAEs leading to discontinuation were reported. This study demonstrated higher maribavir systemic exposure in Japanese than White participants and similar safety outcomes. This difference in exposure is not considered clinically important and its significance remains to be determined.

Cytomegalovirus (CMV) infection is a common posttransplant complication and is associated with increased morbidity and mortality. [1][2][3][4] Maribavir (5,6dichloro-2-(isopropylamino)-1, β-l-ribofuranosyl-1-Hbenzimidazole) is an orally bioavailable anti-CMV drug that inhibits UL97 kinase and CMV DNA synthesis. [5][6][7][8] A phase III, open-label study (SOLSTICE; NCT02931539) in transplant recipients with CMV infection refractory to prior treatment with or without resistance demonstrated that maribavir (400 mg twice daily) was superior to conventional therapies (valganciclovir/ganciclovir, foscarnet, and cidofovir) for the clearance of CMV infection at week 8, and viremia clearance plus symptom control maintained from week 8 through week 16. 9 In addition, maribavir was associated with lower rates of neutropenia versus valganciclovir/ganciclovir and acute kidney injury versus foscarnet, and fewer patients discontin-ued maribavir treatment compared with conventional therapies. [9][10][11] Following oral administration, maribavir is rapidly absorbed, with mean peak plasma concentrations (C max ) around 1-3 hours after dosing, and eliminated with a half-life of ≈5-7 hours. 12 Maribavir has time-independent pharmacokinetics (PK), and its accumulation ratio is close to 1 following twice-daily dosing. 13 Studies in Western populations showed that maribavir exposure (C max and area under the plasma concentration-time curve [AUC]) increases approximately dose proportionally following single oral doses of 50-1600 mg. [12][13][14] While coadministration of maribavir with food had no effect on AUC, its C max decreased by 28%, indicating that maribavir can be dosed with or without food. 12 Maribavir is primarily metabolized in the liver by cytochrome P450 (CYP) 3A4, and to a minor extent by CYP1A2, to form its primary metabolite VP44469 (also identified as M4). 8,12,15 Maribavir administered at the dose of 400 mg has a urinary recovery of <2% of the dose as unchanged maribavir, while 30%-40% of the dose is recovered as VP44469. 14 In plasma, the AUC of VP44469 represents <20% of the AUC of maribavir. However, only the parent drug contributes to the pharmacological activity of maribavir, as VP44469 is pharmacologically inactive. 14 In vitro, maribavir is a substrate of P-glycoprotein (P-gp) and does not affect most CYPs, uridine 5 -diphosphoglucuronosyltransferase, and transporters at clinically relevant concentrations; therefore, there is a low risk of drug-drug interaction. 16 Inhibitors of CYP3A4 and/or P-gp may increase maribavir systemic exposure, 16,17 but dose adjustment of maribavir is not necessary based on accumulated safety/tolerability data of doses up to 1200 mg twice daily (3 times the clinical dose of 400 mg twice daily). 11 Potent inducers of CYP3A4 and P-gp have been shown to reduce maribavir exposure, 16 thus requiring a maribavir dose increase. In addition, maribavir has been shown to increase the exposure of immunosuppressants, such as tacrolimus. 18 Therefore, the blood concentration of concomitantly administered immunosuppressants should be monitored at initiation, administration, and discontinuation of maribavir. Maribavir PK does not vary significantly by age, sex, ethnicity, renal impairment, and hepatic impairment. [19][20][21] Maribavir PK, efficacy, and safety have been extensively evaluated in Western populations. In a population PK analysis using pooled PK data of maribavir in healthy participants and transplant recipients with CMV infection, race was not identified as a significant covariate. However, PK and safety data in Asian populations are very limited. 21 Thus, this phase I study compared the PK and safety of a single oral dose of maribavir in healthy adult participants of Japanese descent with matched non-Hispanic White participants. Data obtained in this study are used to support the clinical evaluation of maribavir in Japanese transplant recipients with CMV infections.

Study Objectives and End Points
This study compared the PK profile of maribavir (400mg single dose) in healthy adults of Japanese descent and matched non-Hispanic White participants. Primary end points were C max , AUC from the time of dosing to the last measurable concentration (AUC last ), and AUC extrapolated to infinity (AUC 0-inf ).
The dose proportionality of maribavir PK was assessed in the Japanese population. In addition, the safety of maribavir, following single oral doses of 200, 400, and 800 mg in healthy adult participants of Japanese descent, and a single 400-mg oral dose of maribavir in non-Hispanic White participants was evaluated. The PK profile of VP44469 in plasma was evaluated in both Japanese and White populations at all dose levels.

Study Design
This phase I, open-label, randomized, crossover, partially fixed-sequence, single-center study was conducted at PPD Clinical Research Unit, Las Vegas, Nevada, between August and November 2020. Prior to study initiation, the study protocol, informed consent form, and all participant recruitment information were approved by the Institutional Review Board (Advarra, Columbia, Maryland). This study was conducted in accordance with the Integrated Addendum to International Council for Harmonisation E6 (R1): Guideline for Good Clinical Practice E6 (R2), the US Code of Federal Regulations (2016), the European Union Directives, and the principles of the Declaration of Helsinki (1996). Written informed consent was collected from all participants prior to study procedures. A standard COVID-19 polymerase chain reaction (PCR) testing protocol and social distancing measures were followed during the conduct of the study. The PCR results were available within 36 hours, and negative PCR results were required before randomization or study dosing.
A total of 24 Japanese and non-Hispanic White adults were planned for enrollment. All participants were required to be healthy, as determined by the principal investigator on the basis of screening evaluations, aged 18-55 years, have a body mass index between 18.5 and 28.0 kg/m 2 , and a body weight >45 kg. Individuals of Japanese descent were matched with non-Hispanic White participants of similar age (±10 years), sex, and body mass index (±15%). For inclusion, participants of Japanese descent had to have been born in Japan, not lived outside of Japan for >10 years, and have both parents and all 4 grandparents of Japanese origin. Non-Hispanic White participants had to have both parents and all 4 grandparents of non-Hispanic White origin. Participants were excluded from the study if they had  a history of disease that could affect the action, absorption, or systemic disposition of maribavir; a known history of alcohol or other substance abuse within the past year; a positive HIV, hepatitis B surface antibody, or hepatitis C virus antibody screen; used tobacco or other nicotine-containing products in any form in the 30 days prior to receiving the first dose of maribavir; or had ingested known CYP3A modulators within the 7 days before the study.
The study design is presented in Figure 1. Participants were screened up to 28 days prior to the first dose of maribavir and admitted to the Clinical Research Unit on day −1. Japanese participants received 3 doses of maribavir in 3 treatment periods separated by a ≥72-hour washout period between doses. On day 1 of treatment period 1, all participants received a single dose of 400-mg maribavir. For Non-Hispanic White participants, there was only treatment period 1, and they were discharged on day 2. In treatment periods 2 and 3, Japanese participants were randomly assigned to receive either a single dose of 200-mg maribavir in treatment period 2 followed by a single dose of 800-mg maribavir in treatment period 3, or a single dose of 800mg maribavir in treatment period 2 followed by a single dose of 200-mg maribavir in treatment period 3. Maribavir doses were given under fasting conditions (10 hours before and 4 hours after dose). All participants received a posttreatment follow-up telephone call 7 (±4) days after their last dose of maribavir.

Pharmacokinetic Assessments
Blood samples for PK analysis of plasma maribavir and VP44469 were collected before dosing, and at 0.25, 0.5, 1, 1.5, 2, 3, 4, 5, 6, 8, 12, 16, 20, and 24 hours after maribavir administration on day 1 of all treatment periods. Plasma concentrations of maribavir and VP44469 were determined at Celerion Laboratories (Lincoln, Nebraska) using a validated liquid chromatographytandem mass spectrometry method. The assays were designed in accordance with the Food and Drug Administration and European Medicines Agency guidelines. Briefly, maribavir (internal standard [IS]: d 7 -maribavir) and VP44469 (IS: 13C-labeled VP44469) were first extracted from the plasma using liquid-liquid extraction. The extracted samples were then analyzed using a JASCO PU-980 (JASCO, Easton, Maryland) or equivalent pump and a LEAP Technologies CTC PAL (LEAP Technologies, Carrboro, North Carolina) or equivalent autosampler. Reversed phase chromatographic retention was performed using an ACE column (C18, 2.1 × 50 mm, 5 μm; Advanced Chromatography Technologies, Mansfield, Massachusetts) using 4:6:90:0.1 acetonitrile:methanol:ultrapure water:formic acid and 40:60:0.1 acetonitrile:methanol:formic acid as the mobile phase(s). The flow rate was 0.4 mL/min, and the column was maintained at ambient temperature. A triple quadrupole mass spectrometer (API 4000TM; AB SCIEX, Framingham, Massachusetts) with an electron spray ionization source was used in the multiple reaction monitoring mode, with negative ions monitored for maribavir, and positive ions monitored for VP44469. The multiple reaction monitoring transitions were m/z 374.0 to 2420 for maribavir (IS: 381.0 to 249.0) and m/z 334.0 to 202.0 for VP44469 (IS: 340.0-208.1). For the maribavir assay, the linear range was between 50 and 25,000 ng/mL and for the VP44469 assay between 5 and 2500 ng/mL. Quality control (QC) samples for the maribavir assay of 150, 1100, 7500, 18,000, and 100,000 ng/mL had a percentage bias of −3.3%, −0.9%, 0.3%, −1.7%, and −2.2%, respectively. For the VP44469 assay, QC samples of 15, 110, 750, 1800, and 5000 ng/mL had a bias of 2.0%, 5.5%, 3.7%, 2.2%, and 3.2%, respectively. The interday accuracy for maribavir was −3.3% to −0.3% and between-run precision ranged from 1.8% to 4.8%. The interday accuracy for VP44469 was 2.0% to 5.5%, and between-run precision ranged from 2.6% to 5.2%. Both study and QC samples were stored at −20°C. Quantitation was determined using a weighted linear regression analysis (1/concentration 2 ) of peak area ratios of the analyte and IS. Computer application programs used to acquire and derive data for this study included AB SCIEX Analyst 1.7.1 (or later versions) and Watson LIMS 7.6 SP1 (or later versions; Thermo Scientific, Waltham, Massachusetts), respectively.
PK primary end point parameters were C max , AUC last (calculated using the linear-up/log-down method), and AUC 0-inf . Additional PK parameters included time to maximum observed plasma concentration during a dosing interval (t max ), terminal half-life (t 1/2 ), apparent total body clearance following oral administration (CL/F), and the delay between the time of dosing and time of drug appearance in plasma in the employed sampling scheme (t lag ). AUC and C max normalized to 400 mg were calculated for Japanese participants following 200-and 800-mg doses. PK parameters were determined by noncompartmental methods using Phoenix WinNonlin 8.0 (Certara, Princeton, New Jersey).

Safety Assessments
Safety was assessed using the number, severity, seriousness, and causality of treatment-emergent adverse events (TEAEs) coded using the Medical Dictionary for Regulatory Activities version 23. Changes in electrocardiograms; vital signs including weight, blood pressure, heart rate, respiratory rate, and body temperature; and clinical laboratory results including hematology, biochemistry, and urinalysis were analyzed as the change from baseline at postbaseline time points; for each variable, the last value collected before the first dose of maribavir in that treatment period was used as the baseline for analyses.

Statistical Analysis
All statistical analyses were done using SAS version 9.4 (SAS Institute, Cary, North Carolina). Individual plasma PK parameters for maribavir and VP44469 were summarized by cohort and maribavir dose using descriptive statistics. Differences of log-transformed PK parameters (C max , AUC last , and AUC 0-inf ) of maribavir and VP44469 between participants of Japanese descent and matched non-Hispanic White participants at the 400-mg dose were compared by an analysis of variance model; the geometric mean ratio and its 90% confidence interval (CI) were provided from the model. Dose proportionality in participants of Japanese descent was examined for maribavir PK parameters and assessed for C max , AUC last , and AUC 0-inf using the power model. The power model assumed a linear relationship between the natural log-transformed PK parameter and the natural log-transformed dose. Dose proportionality was assessed by estimating mean slope with the corresponding 2-sided 90%CI from the power model.

Participants
Overall, 24 participants, 12 of Japanese descent and 12 of non-Hispanic White descent, were enrolled in the study. All 24 participants completed the study and were included in the safety and PK analyses. Overall baseline characteristics were balanced across cohorts (Table S1).

Maribavir PK
The maribavir concentrations declined monoexponentially with time after reaching C max and were measurable up to 24 hours ( Figure 2). Maribavir PK parameters are presented in Table 1. At the 400-mg single dose, the geometric mean C max of maribavir in Japanese participants was 17.4 μg/mL, 10% higher than in non-Hispanic White participants (15.8 μg/mL). The geometric mean AUC 0-inf was 96.7 μg • h/mL for Japanese participants, 25% higher than for non-Hispanic White participants (77.3 μg • h/mL). Mean t 1/2 values were similar in Japanese (5.45 hours) and non-Hispanic White participants (4.81 hours) ( Table 1). The betweenparticipant variability in systemic exposure was moderate (<40%) and similar across Japanese and non-Hispanic White cohorts (Table 1).
When normalized to a 400-mg dose, there were overlaps in the distributions of maribavir C max , AUC last , and AUC 0-inf between Japanese participants and non-Hispanic White participants (Figure 3). In Japanese participants, maribavir AUC last , and AUC 0-inf were dose proportional over the dose range of 200-800 mg (90%CI of fold change when dose doubling included 2; Table 2). Maribavir C max increased in a less-than-doseproportional manner, with an estimated slope of 0.771, <1 (Table 2).

VP44469 PK
Similar to maribavir concentration-time profiles, VP44469 concentrations declined monoexponentially with time after reaching C max and were measurable up  AM,arithmetic mean;AUC 0-inf ,area under the plasma concentration-time curve extrapolated to infinity;AUC last ,area under the plasma concentrationtime curve from the time of dosing to the last measurable concentration; CI, confidence interval; CL/F, apparent total body clearance following oral administration; C max, maximum concentration; CV, coefficient of variation; GLSM, geometric least square means; GM, geometric mean; PK, pharmacokinetic; SD, standard deviation; t 1/2 , terminal half-life; t lag , the delay between the time of dosing and time of drug appearance in plasma in the employed sampling scheme; t max , time to maximum observed concentration. All PK parameters were assessed in plasma, determined by noncompartmental methods using Phoenix WinNonlin 8.0 (Certara, Princeton, New Jersey). a A linear mixed-effect model with treatment as fixed effect (Japanese 400 mg versus non-Hispanic White 400 mg) was used to fit to ln-transformed PK parameters and estimate all treatment differences and corresponding 2-sided 90%CI. The difference and 90%CI of difference were back-transformed and expressed as geometric least square mean ratios and 90%CI interval of the ratios. to 24 hours (Figure 4). VP44469 PK parameters are presented in Table 3. Following administration of 400mg maribavir, the geometric mean C max of VP44469 for Japanese participants was 1.38 μg/mL and for non-Hispanic White participants was 1.37 μg/mL, attained in both cohorts at a median time to maximum observed plasma concentration during a dosing interval of 2 hours after dosing. The geometric mean AUC 0-inf of VP44469 was 15.7 μg • h/mL for Japanese participants and 14.8 μg • h/mL for AUC 0-inf , area under the plasma concentration-time curve extrapolated to infinity; AUC last , area under the plasma concentration-time curve from the time of dosing to the last measurable concentration; CI, confidence interval; C max, maximum concentration; PK, pharmacokinetic; SE, standard error. Dose proportionality was calculated using the power model. a P value, testing slope = 1.  non-Hispanic White participants. Mean t 1/2 values of VP44469 were 7.34 hours for Japanese participants and 5.98 hours for non-Hispanic White participants who received 400-mg maribavir (Table 3).

Safety
Overall, 29.2% of participants reported at least 1 TEAE (Table S2). All TEAEs were considered by the investigator to be related to maribavir, were mild in severity, and all were considered resolved. No serious TEAEs or TEAEs leading to discontinuation occurred in the study. The most frequently reported TEAE in both Japanese and non-Hispanic White participants was dysgeusia. In the Japanese cohort, 2 (16.7%) participants had dysgeusia following 400 mg maribavir, 1 (8.3%) following 200 mg maribavir, and no TEAEs were reported following administration of 800 mg maribavir. In the non-Hispanic White cohort, 4 (33.3%) participants had dysgeusia following 400-mg All PK parameters were assessed in plasma, determined by noncompartmental methods using Phoenix WinNonlin 8.0 (Certara, Princeton, New Jersey). a A linear mixed-effect model with treatment as fixed effect was used to fit to ln-transformed PK parameters and estimate all treatment differences and corresponding 2-sided 90%CIs. The difference and 90%CI of difference were back-transformed and expressed as geometric least square mean ratios and 90%CI of the ratios.
maribavir. Somnolence was the only other reported TEAE, which occurred in 2 (8.3%) non-Hispanic White participants following administration of 400-mg maribavir.

Discussion
Following administration of a 400-mg maribavir oral dose, maribavir C max was ≈10% higher and AUC 0-inf 25% higher in Japanese participants compared with non-Hispanic White participants; however, there was substantial overlap in AUC 0-inf between the 2 cohorts. The slightly higher maribavir exposure in Japanese participants may be partially due to lower body weight (difference in weight-normalized apparent total body clearance following oral administration was smaller than that in AUC; Table 1) and is not considered to have clinical importance with respect to dosing adjustment given the comparable safety profile to non-Hispanic White participants in this study as well as accumulated safety and tolerability profiles of maribavir at doses up to 1200 mg twice daily 11 (3 times the recommended dose of 400 mg twice daily) in transplant recipients. There was also no apparent difference between the 2 cohorts with respect to VP44469 C max or AUC 0-inf . The reported data suggest that maribavir may be used without dose adjustment in Japanese populations.
Total systemic exposure to maribavir at doses of 200-800 mg increased in a dose-proportional fashion in Japanese participants, while peak exposure increased less than dose proportionally. These results were consistent with previously published data in healthy White participants and White transplant recipients in which C max increased in a less-than-dose-proportional manner. 12 Maribavir single oral doses of 200, 400, and 800 mg had similar safety outcomes between healthy adult participants of Japanese descent and non-Hispanic White participants. The most common TEAE in both cohorts was dysgeusia (taste disturbance), which was consistent with previous studies in both healthy participants 14 and transplant recipients. 9 The only other TEAE was somnolence, which occurred only in non-Hispanic White participants.
The study limitations include the single-dose study design and maribavir administration under the fasting conditions in healthy participants given that maribavir is indicated for the treatment of CMV infection/disease in transplant recipients. Single dosing of maribavir was chosen for this study, as maribavir has demonstrated linear and time-independent PK. No significant food effects have been previously observed on maribavir PK. 12 Thus, although maribavir dosing was given under fasting conditions in this study, dietary differences between populations are not expected to affect maribavir PK. Based on previous population PK analyses, transplant recipients tend to have ≈20% higher maribavir exposure than healthy participants. 12 This increase in exposure is likely attributed to the use of concurrent medications by transplant recipients that are known to increase exposure to maribavir. Given the accumulated data on absorption, distribution, metabolism, and excretion of maribavir, the drug disposition is expected to be similar between transplant recipients and healthy participants. 12 Therefore, PK results obtained in this study may be applicable to repeated doses of maribavir in transplant recipients.

Conclusion
In this Japanese PK bridging study, no differences that were considered potentially clinically important were observed in the PK of maribavir following a 400-mg dose in Japanese and non-Hispanic White participants. Maribavir safety in Japanese participants was similar to that observed in non-Hispanic White participants. Taken together, the results of this PK study suggest that the same maribavir dose as approved in White populations could be used in Japanese populations.

Funding Information
This study was funded by Takeda Development Center Americas, Inc.

Conflicts of Interest
I.S., K.I., and J.W. are employees of Takeda Development Center Americas, Inc. and Takeda stock owners. B.S. is an employee of qPharmetra LLC, which received funding from Takeda Development Center Americas, Inc. for study analysis and clinical study report development.

Data Availability Statement
The data sets, including the redacted study protocol, redacted statistical analysis plan, and individual participants' data supporting the results reported in this article will be made available within 3 months from initial request to researchers who provide a methodologically sound proposal. The data will be provided after its deidentification, in compliance with applicable privacy laws, data protection, and requirements for consent and anonymization.