Amenamevir

Pharmaceuticals and Medical Device Agency approval summary: Amenamevir for the treatment of herpes zoster

Naoko SHOJI, Keiji TANESE, Ayano SASAKI, Taishi HORIUCHI, Yuji UTSUNO, Koichi FUKUDA, Yukiko HOSHINO, Shinichi NODA, Hirofumi MINAMI,
Wataru ASAKURA, Amenamevir Review Team
Office of New Drug IV, Pharmaceuticals and Medical Device Agency, Tokyo, Japan

ABSTRACT

In July 2017, Japan’s Ministry of Health, Labor and Welfare issued a marketing authorization valid throughout Japan for N-(2,6-dimethylphenyl)-N-(2-{[4-(1,2,4-oxadiazol-3-yl)phenyl]amino}-2-oxoethyl)-1,1-dioxothiane-4-car- boxamide (amenamevir) for the first time worldwide. The decision was based on the favorable opinion of the Pharmaceuticals and Medical Device Agency (PMDA) recommending a marketing authorization of amenamevir for treatment of herpes zoster (HZ). Amenamevir has a different action mechanism from previously approved synthetic nucleoside compounds for the treatment of HZ including acyclovir, valacyclovir and famciclovir. The usual adult dose is 400 mg amenamevir p.o. once daily for 7 days. The benefit is its ability to cure HZ as well as preventing postherpetic neuralgia. The most common side-effects are increase of urine N-acetyl-b-D-glu- cosaminidase and a1-microglobulin levels. However, based on the detailed evaluation of the submitted clinical studies, there seems to be no serious safety concerns about amenamevir regarding the kidney of both renally normal and impaired patients. The objective of this article is to summarize the scientific review of the applica- tion. The detailed scientific assessment report and product information, including the summary of product char- acteristics, are available on the PMDA website (www.pmda.go.jp/PmdaSearch/iyakuSearch/).

Key words: amenamevir, helicase–primase inhibitor, herpes zoster, pharmaceuticals and medical device agency, renal failure.

INTRODUCTION

Varicella zoster virus (VZV) causes two distinct diseases: varicella and herpes zoster (HZ). In primary infection, VZV enters through the respiratory tract and conjunctiva. During the incubation period of approximately 2 weeks, VZV replicates at the entry site and regional lymph nodes. Thereafter, the virus disseminates to other organs including the skin to cause vari- cella. Although varicella resolves within a couple of weeks, VZV remains latent in the sensory ganglia. HZ onsets by the reacti- vation of latent VZV. The major factors thought to influence its reactivation are overfatigue, aging, immunosuppression, malig- nancies and autoimmune diseases. HZ is characterized by a rash and pain limited to the skin innervated by a single, infected sensory ganglion. Although the eruption resolves within a cou- ple of weeks, persistent pain in the affected lesion termed pos- therpetic neuralgia (PHN) occurs in some instances.1

N-(2,6-dimethylphenyl)-N-(2-{[4-(1,2,4-oxadiazol-3-yl)phenyl] amino}-2-oxoethyl)-1,1-dioxothiane-4-carboxamide (amenamevir) (Fig. 1) is a herpesvirus helicase–primase complex inhibitor discovered by Astellas Pharma (Tokyo, Japan).2 The clinical development of amenamevir was initially conducted by Astellas Pharma Global Development and Astellas Pharma Eur- ope in the USA and Europe. However, development was sus- pended as the healthy participants in a phase I study (15L-CL- 019) experienced serious adverse events (AE) including throm- bocytopenia, for which a causal relationship with amenamevir could not be excluded. Later, Maruho (Osaka, Japan) acquired a license for its development both inside and outside of Japan. Based on the safety information obtained from non-clinical and regional clinical studies conducted by Astellas Pharma, and the linical studies data of HZ patients,3 Maruho submitted an appli- cation to the Japanese Ministry of Health, Labor and Welfare for the marketing authorization of amenamevir. The Pharmaceuti- cals and Medical Device Agency (PMDA) conducted the scien- tific review. A marketing authorization was issued on July 3, 2017. The approved therapeutic indication was: “Treatment of Herpes zoster”.

Figure 1. Chemical structure of N-(2,6-dimethylphenyl)-N-(2- {[4-(1,2,4-oxadiazol-3-yl)phenyl]amino}-2-oxoethyl)-1,1-dioxothi- ane-4-carboxamide (amenamevir).

NON-CLINICAL EVALUATION

The in vitro anti-VZV activity of amenamevir and acyclovir (ACV) was evaluated by exposure to human embryonic fibrob- last (HEF) cells infected with several VZV strains. Amenamevir showed a 50% effective concentration (EC50) of <0.1 lmol/L in all analyzed strains including the ACV low-sensitivity strain, Kanno-Br (Table 1). Meanwhile, the drug concentration re- quired for 50% reduction in cell viability against the host HEF cells was ≥200 lmol/L. Amenamevir showed inhibitory effects against the helicase, primase and DNA-dependent ATPase activities of the herpes simplex virus (HSV)-1 helicase–primase complex. It inhibited the helicase activity and primase activity of the recombinant HSV-1 helicase–primase complex of wild-type HSV-1 strain KOS at the concentrations ≥0.1 lmol/L and ≥0.03 lmol/L, respectively. DNA-dependent ATPase activity was also inhib- ited at a mean 50% inhibitory concentration (IC50) of 0.078 lmol/L. Amenamevir inhibited the DNA replication of VZV, which was confirmed by quantifying the viral DNA content in VZV virus CaQu strain-infected HEF. After 3 days of expo- sure, amenamevir reduced the quantity of VZV DNA in a con- centration-dependent manner, with an IC50 of 0.057 lmol/L. The frequency of amenamevir low-sensitivity or ACV low- sensitivity HSV variants was evaluated by counting the plaque numbers of the African green monkey kidney cell line, Vero, infected with HSV-1 or HSV-2 strains under drug exposure. The frequency of amenamevir low-sensitivity variants was signifi- cantly less than that of ACV low-sensitivity variants (Table S1). The frequency of development of drug less-sensitive HSV vari- ants was evaluated by counting plaques of Vero cells infected with HSV-1 strains (KOS, WT-51, CI-25 and CI-116) and HSV-2 strains (G, Lyon, Kondo and CI-5243) under the long-term drug exposure. The viral titer increased within 168 h in all eight strains under ACV exposure, but amenamevir maintained the viral titer below the detection limit for 168 h in seven of eight strains (except for CI-25) and for 120 h in all eight strains. The in vivo antiviral activity of amenamevir was evaluated using the mouse skin HSV-1 infection model, as there is no estab- lished HZ animal model due to the high species-specificity of VZV. HSV-1 causes herpetiform eruptions on the skin innervated by a single ganglion upon primary infection in mice. Amenamevir showed both preventive and treatment effects on this mouse model in a dose-dependent manner.5,6 Although the in vivo antivi- ral activity of amenamevir was evaluated only against HSV types, the PMDA agreed with the applicant’s opinion that amenamevir also exerts antiviral activity against VZV by inhibiting the VZV heli- case–primase complex. This was based on the results of in vitro anti-VZV activity and evidence that HSV and VZV have high homology in their helicase–primase complex genes as they belong to the same Alphaherpesvirinae subfamily. In the pharmacokinetic (PK) evaluation, amenamevir showed dose-dependent absorption in mice and dogs. Distribution was observed in the liver, small intestine, Harderian gland, kidneys and large intestine of mice, and was mainly excreted in the feces. It was also able to bind melanin reversibly, to cross the placenta and to be excreted in milk. In humans, the plasma protein binding rate was 75.0–75.3% and the erythrocyte trans- fer rate was 46.7–49.5%. Metabolization is mainly done by cytochrome P450 (CYP)3A4. PRECLINICAL TOXICITY The single-dose toxicity study determined the approximate lethal dose as >500 mg/kg in mice and >750 mg/kg in dogs. Repeated-dose toxicity study showed hepatotoxicity in dog were 2.6-fold higher than those produced by the maximum recommended clinical dose in humans (400 mg/day).

Genotoxicity was not found in the bacterial reverse mutation assay, the mammalian chromosome aberration assay or the mouse bone marrow micronucleus assay. The evidence of car- cinogenicity was not found even at the highest tested dose (250 mg/kg per day) in mice and rats; the corresponding plasma exposure value (plasma concentration–time area under developmental toxicity in mice and rabbits also showed no treatment-related abnormal findings in embryos and fetuses even at the highest tested dose; the corresponding plasma exposure values (AUC0–24) were 5.3- and 9.3-fold higher, respectively, compared with the maximum recommended clinical dose in humans. Pharmacological safety study studies did not show noteworthy effects on the central nervous, car- diovascular and respiratory systems.

CLINICAL, PHARMACOLOGICAL, MASS BALANCE AND BIOPHARMACEUTICAL STUDIES

The PK parameters of maximum drug concentration (Cmax), area under the plasma concentration–time curve from time zero to infinity (AUCinf), half-life (t1/2) and time to peak drug concentration (tmax) were calculated from the plasma drug con- centration–time data in phase I studies (15L-CL-001, -003 and -006).

The plasma PK after a single dose was evaluated in a dose- escalation study (15L-CL-001) conducted in Japan. Sequential cohorts of healthy participants were randomized to receive a single capsule or placebo under fasting conditions (Table 2). The bioavailability of amenamevir in capsule and tablet form, and food-effect bioavailability of the tablets were evaluated in a three-period cross-over study (15L-CL-006) conducted out- side Japan. Healthy participants received a single capsule or tablet of 800 mg amenamevir under fasting conditions, or a single tablet of 800 mg amenamevir under fed conditions of high-fat diet (Table S2). The AUCinf almost doubled and the Cmax also increased (1.55-fold) under fed conditions. Based on these results, the study drug was administered in the fed state in subsequent studies.
The plasma PKs after multiple doses was evaluated in a mul- tiple ascending dose study (15L-CL-003) conducted in Japan. Non-elderly (aged 20–44 years) and elderly (aged 65–79 years) healthy male participants were randomized to receive 300 mg/ day or 600 mg/day capsule under fed conditions for 7 days (Table S3). The mean AUCinf and Cmax at day 7 was slightly lower or almost the same compared with day 1. In participants receiving the amenamevir 300 mg/day, there were no consis- tent differences in the amenamevir PKs between nonelderly and elderly participants. In contrast, both parameters increased slightly higher in elderly participants after receiving single and multiple amenamevir doses at 600 mg/day.8 Based on the PK data of HZ patients (621 sampling points in 223 patients) in a phase II study (15L-CL-221), population PK analysis was per- formed using NONMEM version 6 software (ICON Development Solutions, Dublin, Ireland). The Cmax and plasma concentration– time area under the curve over a dosing interval (AUCtau) were estimated as 1.26 lg/mL and 14.6 mg·h/mL for 200 mg ame- namevir, and 1.94 lg/mL and 22.9 mg·h/mL for 400 mg ame- namevir. The plasma PK in moderate liver dysfunction patients (Child–Pugh classification B) showed no abnormal findings.

Mass balance was investigated following a single p.o. dose of 200 mg 14C-amenamevir in non-Japanese healthy subjects. The excretion rates of radioactivity into the urine and feces up to 168 h post-dose were 20.6% and 74.6%, respectively. Potential interactions between 400 mg amenamevir and concomitant medications were evaluated in studies 15L-CL- 008, -009, -010, -018, M522101-EU21, -EU22, -EU23, -EU24 and -EU25.9 The exposure to amenamevir decreased with con- comitant administration of rifampicin (CYP3A4 inducer) and cyclosporin, and increased with ketoconazole and ritonavir (strong CYP3A4 inhibitors). The exposure to midazolam (CYP3A4 substrate) and bupropion (CYP2B6 substrate) decreased with concomitant amenamevir administration.

CLINICAL EFFICACY
Phase II study

Study 15L-CL-221 was a valacyclovir (VACV)-controlled, ran- domized, four-armed, parallel-group, phase II study that com- pared 100, 200 and 400 mg amenamevir once daily or 1000 mg VACV three times daily for 7 days to treat HZ (1:1:1:1 ratio). The primary objective was to determine an effective, safe and tolera- ble dose of amenamevir for the treatment of HZ. The study enrolled 20–80-year-old male and non-pregnant female sub- jects, clinically diagnosed with localized HZ presenting within 72 h after rash onset. The assigned study drug was adminis- trated p.o. in the fed state. The primary efficacy analysis variable was the proportion of patient with cessation of new lesion forma- tion by day 4. A total of 403 patients were randomized and trea- ted (400 mg amenamevir group, n = 97; 200 mg group, n = 102; 100 mg group, n = 102; and VACV group, n = 102). Among them, 107 patients whose medical records were missing and 10 patients in whom VZV was undetected by virus identification test were excluded. Subsequently, 286 participants (400 mg amenamevir group, n = 66; 200 mg group, n = 76; 100 mg group, n = 73; and VACV group, n = 71) were defined as the full analysis set (FAS) and the populations for the primary analyses.

The proportions of patients with cessation of new lesion forma- tion by day 4 were 90.9% (60/66), 85.5% (65/76), 87.7% (64/73)
and 87.3% (62/71), in 400 mg, 200 mg and 100 mg amenamevir and VACV groups, respectively.

Phase III study

Study M522101-J01 was a randomized, double-blind, three- armed, parallel-group, phase III study that compared 200 or 400 mg amenamevir once daily or 1000 mg VACV three times daily for 7 days to treat HZ (1:1:1 ratio).3 The primary objective was to verify the effectiveness and safety of 200 and 400 mg amenamevir for the treatment of HZ. The study drug administra- tion schedule, patient enrollment criteria and primary efficacy analysis variable were the same as those in study 15L-CL-221. Secondary end-points included days to cessation of new lesion formation, days to complete crusting and healing, and days to resolution of pain. The primary statistical analyses were performed according to the Mantel–Haenszel method, in which the non-inferiority margin was set to 10%. The populations for the the time of registration, received the study drugs and had any efficacy variable measured, defined as FAS.

A total of 750 patients were randomized and treated (400 mg amenamevir group, n = 249; 200 mg group, n = 252; and VACV group, n = 249). Among them, 14 patients in whom HSV-1 was detected and one patient with no efficacy assess- ment data were excluded. Finally, the FAS consisted of 735 participants (400 mg amenamevir group, n = 243; 200 mg group, n = 247; and VACV group, n = 245). The efficacy evalu- ation verified non-inferiority of 400 mg amenamevir, but not of 200 mg, to VACV in the proportion of patients with cessation of new lesion formation by day 4. Other efficacy end-points were almost the same among three treatment groups (Table 3). The incidence rates of PHN, which was defined as persistent regional pain not less than 91 days after taking the study drugs, were 1.0% (2/193), 1.9% (4/209) and 1.0% (2/206) in the 400 mg and 200 mg amenamevir, and VACV groups respectively, suggesting that amenamevir and VACV have a similar PHN preventive effect.

CLINICAL SAFETY

The overall incidence of any AE in the M522101-J01 study was 46.6% (116/249), 45.6% (115/252) and 45.4% (113/249) in the
400 mg and 200 mg amenamevir, and VACV groups, respec- tively. The most frequent AE was increased urine N-acetyl-b-D- glucosaminidase (NAG) and a1-microglobulin (a1-MG) levels. The proportions of patients who experienced drug-related AE and precise AE with a frequency of at least 2% in any group were reported previously.3 Further safety evaluations were per- formed on a safety population comprised of 1046 patients who were randomized and treated for HZ in the M522101-J01 and 15L-CL-221 studies and 737 patients for herpes simplex (HS) in the M522101-J11 and M522101-J12 studies (Table 4). None of †Mantel–Haenszel method, adjusted by age (<65 and ≥65 years) and time from the onset of rash to the start of treatment (≤24 h, >24 to ≤48 h and >48 to ≤72 h).‡Median value. §Cox hazard ratio, adjusted by time from the onset of rash to the start of treatment (≤24 h, >24 to ≤48 h and >48 to ≤72 h). CI, confidence interval; VACV, valacyclovir.the treatment groups showed apparent difference in frequency. Further evaluation was performed on its effect on the kidneys, thrombocytopenia, cardiovascular system and elderly patients.

EFFECT ON RENAL FUNCTION

As subjects in a phase I study (15L-CL-002) experienced urinal crystal-like material assumed to be related to amenamevir, urine NAG, a1-MG and renal function parameters were evaluated in a phase III study (M522101-J01) for HZ and two studies (M522101-J11 and M522101-J12) for HS. The corrected values of NAG and a1-MG based on urine creatinine levels were ele- vated in 2.8–3.6% and 9.2–11.0% of patients in the amenamevir groups, respectively, which were relatively higher than those in the VACV group (NAG, 2.8%; a1-MG, 5.6%). However, those elevations were transient in most cases and recovered without additional treatment. Furthermore, clinical symptoms suggesting proximal renal tubular dysfunction were not detected.

Another phase I study (15L-CL-014) evaluated its effect on impaired kidneys by exposing amenamevir to renally impaired patients.10 The mean AUCinf increased in accordance with the renal dysfunction severity (Table 5). The AUCinf of the subject with the greatest renal impairment was 28.6 µg/h/mL, which was similar to the mean AUCinf of healthy individuals administered 1200 mg amenamevir (28.3 µg/h/mL) without sev- ere AE in another phase I study (15L-CL-004). The safety pop- ulation of 1046 patients in HZ studies did not show meaningful differences in the incidence rate of AE and drug-related AE between the moderate renal dysfunction (50 mL/min ≤ crea- tinine clearance [CLcr] < 80 mL/min) and normal renal function (CLcr ≤ 80 mL/min) group. Based on these results, there seems to be no serious concern in administering amenamevir to renally impaired patients. However, since the worst CLcr level in the patients enrolled in the 15L-CL-014 study was 16.8 mL/min, its safety in patients with very severe renal dys- function who are receiving dialysis has not been elucidated. Case (%). Detail of adverse events (AE) leading to discontinuation of the study: contact dermatitis in 400 mg group; headache and nausea in 200 mg group; neutropenia in the 100 mg group of herpes zoster safety population; and back pain in 200 mg group of herpes simplex safety population. Headache and nausea was drug-related AE but their grades were mild and recovered after termination of the study drug. VACV, valacyclovir. EFFECT ON PLATELET COUNT In a phase I repeated-dose administration study (15L-CL-019), serious thrombocytopenia occurred in one healthy study partici- pant. Platelet count decrease (13 9 103/µL) was observed on day 21, and it decreased further to below detectable levels by day 23. Other laboratory findings were normal. Together with other clinical findings, the condition was diagnosed as idiopathic thrombocytopenic purpura (ITP). The relevance of amenamevir was considered to be low, because concomitant ibuprofen could not be excluded as a causative drug, and detailed immunological analysis could not directly associate amenamevir and ITP. In a phase III study (M522101-J01), 5.3% of the 400 mg amenamevir group, 4.1% of the 200 mg group and 5.4% of the VACV group showed platelet counts lower than 14.0 9 104/µL, yet that sever- ity was mild. Further evaluation of the safety population could not find AE or AE leading to drug withdrawal related to thrombocy- topenia or decreased platelet count. Based on these results, there seems to be no serious concern of developing thrombocy- topenia by administration of amenamevir at present. EFFECT ON CARDIOVASCULAR SYSTEM In a phase I study (15L-CL-019), one healthy participant experi- enced serious chest tightness day after the final date (day 28) of repeated dose administration which was diagnosed as endocarditis based on the electrocardiogram findings. Although amenamevir was not excluded as a causative drug, the patient recovered without intervention. In further evaluation of the safety population, severity of most AE related to the cardiovas- cular system were mild or moderate, and amenamevir-related serious AE were not observed. The AE related to electrocardio- gram findings were observed in 13 patients, yet none of them had arrhythmias. The AE related to arrhythmia was observed in three patients, yet the severity was mild in all of them. A phase I study (M522101-J22) evaluating the effect of amenamevir on QT/QTc elongation did not show noteworthy abnormalities. Taken together, amenamevir seems not to elicit serious con- cerns affecting the cardiovascular system. SAFETY IN ELDERLY PATIENTS As HZ tends to occur in elderly subjects,11 its safety in elderly patients was evaluated. Although the incidence of AE tends to be high in elderly patients (>65 years), apparent differences were not observed in the severity of AE (Table S4). A risk management plan, which would perform safety evalu- ation and provide missing information, was submitted to address important potential risks. The PMDA required evalua- tion of: (i) the effectiveness of amenamevir in HZ patients who had previously been administrated ACV, VACV and famciclovir (FCV); (ii) safety in patients with impaired liver function; (iii) AE of the kidneys; (iv) safety in patients with impaired renal func- tion; and (v) efficacy and safety in elderly subjects.

RISK–BENEFIT ASSESSMENT

Based on the efficacy evaluation of phase III study showing similar efficacy between amenamevir and VACV, and the safety evaluation of all submitted clinical studies, the clinical benefits of amenamevir in treating HZ exceed the few and mostly mild AE. The p.o. administrated drugs approved for the treatment of HZ in Japan were ACV, VACV and FCV, all of which are syn- thetic nucleoside compounds. As these drugs are excreted renally, dose reduction based on the renal function is recom- mended. In contrast, the patient’s renal function does not have to be considered with amenamevir at this time.

CONCLUSION

In conclusion, the PMDA recommended the granting of a mar- keting authorization valid throughout Japan for amenamevir for treating HZ. The PMDA will review new information about ame- namevir on a regular basis. The most current information on this medicinal product is available on the PMDA website (www. pmda.go.jp/PmdaSearch/iyakuSearch/).

DISCLAIMER

The objective of this paper is to summarize the scientific review of the application dossier leading to regulatory approval in Japan, and it is not written for promotional purposes. The views expressed are independent work and do not necessarily represent the views and findings of the PMDA. The full scientific assessment report and product information, including the summary of product character- istics, are available on the PMDA website (https://www.pmda.go. jp). For the most current information on this marketing authoriza- tion, please refer to the PMDA website.

ACKNOWLEDGMENTS: The scientific assessment as summarized in this report is based on important contributions of the additional reviewer and experts belonging to the PMDA.

CONFLICT OF INTEREST: None declared.

REFERENCES

1 Urushibata O. Comprehensive Handbook of Clinical Dermatology, Volume 15. Tokyo: Nakayama Shoten, 2003; pp. 33–41.
2 Chono K, Katsumata K, Kontani T et al. ASP2151, a novel helicase- primase inhibitor, possesses antiviral activity against varicella-zoster virus and herpes simplex virus types 1 and 2. J Antimicrob Che- mother 2010; 65: 1733–1741.
3 Kawashima M, Nemoto O, Honda M et al. Amenamevir, a novel helicase-primase inhibitor, for treatment of herpes zoster: a random- ized, double-blind, valaciclovir-controlled phase 3 study. J Dermatol 2017; 44: 1219–1227.
4 Chono K, Katsumata K, Kontani T, Shiraki K, Suzuki H. Characteri- zation of virus strains resistant to the herpes virus helicase-primase inhibitor ASP2151 (Amenamevir). Biochem Pharmacol 2012; 84: 459–467.
5 Katsumata K, Chono K, Kato K et al. Pharmacokinetics and phar- macodynamics of ASP2151, a helicase-primase inhibitor, in a mur- ine model of herpes simplex virus infection. Antimicrob Agents Chemother 2013; 57: 1339–1346.
6 Katsumata K, Chono K, Sudo K, Shimizu Y, Kontani T, Suzuki H. Effect of ASP2151, a herpesvirus helicase-primase inhibitor, in a guinea pig model of genital herpes. Molecules 2011; 16: 7210– 7223.
7 Lenac Roviˇs T, Bailer SM, Pothineni VR et al. Comprehensive analy-
sis of varicella-zoster virus proteins using a new monoclonal anti- body collection. J Virol 2013; 87: 6943–6954.
8 Kusawake T, Keirns JJ, Kowalski D et al. Pharmacokinetics and safety of amenamevir in healthy subjects: analysis of four random- ized phase 1 studies. Adv Ther 2017; 34: 2625–2637.
9 Kusawake T, den Adel M, Groenendaal-van de Meent D et al. Phar- macokinetic evaluation of the interactions of Amenamevir (ASP2151) with ketoconazole, rifampicin, midazolam, and warfarin in healthy adults. Adv Ther 2017; 34: 2466–2480.
10 Kusawake T, Kowalski D, Takada A et al. The influence of hepatic and renal impairment on the pharmacokinetics of a treatment for herpes zoster, Amenamevir (ASP2151): phase 1, open-label, single- dose, parallel-group studies. Adv Ther 2017; 34: 2612–2624.
11 Kurokawa I, Kumano K, Murakawa K, Hyogo Prefectural PHN Study Group. Clinical correlates of prolonged pain in Japanese patients with acute herpes zoster. J Int Med Res 2002; 30: 56–65.