Perampanel in the treatment of partial seizures: Time to onset and duration of most common adverse events from pooled Phase III and extension studies Academic research paper on "Clinical medicine"

ELSEVIER

Contents lists available at ScienceDirect

Epilepsy & Behavior

journal homepage: www.elsevier.com/locate/yebeh

Perampanel in the treatment of partial seizures: Time to onset and duration of most common adverse events from pooled Phase III and extension studies

David Ko a'*, Haichen Yang b, Betsy Williamsc, Dongyuan Xing d, Antonio Laurenza b

a Keck Medical Center ofUSC, 1520 San Pablo Street, Los Angeles, CA 90033, USA b Eisai Neuroscience and General Medicine PCU, Eisai Inc., 155 Tice Blvd., Woodcliff Lake, NJ 07677, USA c Eisai Medical and Scientific Affairs, Eisai Inc., 100 Tice Blvd., Woodcliff Lake, NJ 07677, USA d Formerly Eisai Medical and Scientific Affairs, Eisai Inc., 100 Tice Blvd., Woodcliff Lake, NJ 07677, USA

ARTICLE INFO ABSTRACT

Perampanel (PER) is a novel noncompetitive AMPA-receptor antagonist approved in over 40 countries for treatment of partial seizures. The safety and tolerability of PER have been well-documented in three double-blind, randomized, placebo (PBO)-controlled Phase III studies and an open-label extension (OLE). This post hoc analysis evaluated the occurrence and characteristics of the most common treatment-emergent adverse events (TEAEs) associated with PER. Results from the Phase III studies were pooled; post hoc analyses on the double-blind phase and up to 1 year of the OLE were performed on the four most common TEAEs for which incidence was higher for PER than PBO. The four most common TEAEs were dizziness, somnolence, fatigue, and irritability. For most subjects in the Phase III double-blind studies, these TEAEs were observed during 6-week titration and were mild or moderate in severity. For severe AEs, no dose-response relationship was observed. Patients in the PBO group during Phase III (who therefore received their first PER treatment during OLE) experienced these TEAEs with incidence and timing similar to that of PER-treated patients in Phase III. The first onset of these TEAEs occurred during the early weeks of PER conversion in the OLE. After 6 months and up to 1 year of PER treatment, low to no incidence of the first onset of the four TEAEs was observed. Post hoc analyses of data from pooled Phase III studies provide greater insight into occurrence/duration of TEAEs. Phase III double-blind and OLE data showed that dizziness, somnolence, fatigue, and irritability were the most common TEAEs reported by patients taking PER. Additionally, these results suggest consistency between studies in patient responses to onset of these TEAEs. Although concomitant antiepileptic drugs (AEDs) might be predicted to affect development of TEAEs in patients taking PER, an effect was not observed in this analysis. The low incidence of TEAEs in these studies provides additional support for long-term PER treatment.

© 2015 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/4.0/).

CrossMark

Article history: Received 25 February 2015 Revised 11 May 2015 Accepted 12 May 2015 Available online 4 June 2015

Keywords: Antiepileptic drugs Epilepsy Adverse events

1. Introduction

Perampanel (PER) is approved in the United States for the adjunctive treatment of partial seizures, with or without secondarily generalized seizures, in patients 12 years of age or older with epilepsy [1]. It is approved in the European Union, in Canada (for patients 18 years of age or older), and, to date, has been approved in more than 40 countries worldwide [2-4]. Perampanel has a novel mechanism of action, being

Abbreviations: AED, antiepileptic drug; DB, double blind; MedDRA, Medical Dictionary for Regulatory Activities; MTD, maximum tolerated dose; OLE, open-label extension; PBO, placebo; PER, perampanel; TEAE, treatment-emergent adverse event.

» Corresponding author. Tel.: +1 323 442 5710.

E-mail addresses: dko@usc.edu (D. Ko), haichen_yang@eisai.com (H. Yang), betsy_williams@eisai.com (B. Williams), dyxing2010@gmail.com (D. Xing), antonio_laurenza@eisai.com (A. Laurenza).

a selective, noncompetitive AMPA receptor antagonist. There were three randomized, double-blind, placebo (PBO)-controlled Phase III studies that demonstrated significant decreases in seizure frequency with PER at 4-, 8-, and 12-mg daily doses compared to PBO [5-7].

All antiepileptic drugs (AEDs) are associated with risk for adverse events (AEs); thus, minimizing AEs is an important consideration in their successful use [8,9]. The nature, distribution, and severity of these AEs vary widely by drug, by patient characteristics, and by interactions with other drugs, among other factors [8,9].

This article describes the post hoc analyses of the occurrence and nature of AEs associated with PER therapy in patients with partial seizures observed in the double-blind phase of three Phase III studies and during the first 52 weeks of an open-label extension (OLE). Analysis of the pooled Phase III data focused on the frequency, severity, time to the first onset, and duration of the four most common treatment-emergent adverse

http://dx.doi.org/10.1016/j.yebeh.2015.05.020

1525-5050/© 2015 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

events (TEAEs) observed. The relationship of these TEAEs to dosing during titration and maintenance in the three Phase III studies was also determined. Analysis of OLE data focused on the time to the first onset of the TEAEs over the course of up to 52 weeks of PER treatment.

2. Methods

2.1. Phase III studies

All three Phase III studies [5-7] included subjects 12 years of age or older with refractory partial seizures despite treatment with one to three AEDs. The use of concomitant AEDs at baseline for PER and PBO treatment groups is summarized in Table 1. The studies were multinational, multicenter, randomized, double-blind, and PBO-controlled. They all included an initial 6-week baseline period. This was followed by a 19-week, double-blind phase (6 weeks of titration and 13 weeks of maintenance). In two of the three studies, subjects were randomized in equal proportion to once daily PER 8 mg, PER 12 mg, or PBO, while in the third study, subjects were equally randomized to once daily PER 2 mg, 4 mg, or 8 mg, or PBO. After completing the double-blind phase, subjects were eligible to immediately enter the open-label extension study, or proceed into a 4-week follow-up period in which no PER treatment was provided. In all three studies, regardless of patients' baseline AEDs, PER dosing was initiated at 2 mg daily and increased at a rate of 2 mg per week during titration until the randomized target dose was reached. If a patient experienced tolerability issues, the investigator had the authority to reduce the dose by 2 mg, but more than a single 2-mg dose reduction was discouraged. Subsequent uptitration was permitted if tolerability issues resolved [5-7].

Assessment of safety and tolerability included recording of AEs, previous and concomitant medication use, discontinuations, clinical laboratory parameters, vital signs, electrocardiography studies, physical and neurologic examinations, and administration of both photosensitivity and withdrawal questionnaires. Treatment-emergent adverse events were recorded for each subject at every study visit, regardless of possible relationship to the study drug [5-7].

22. Open-label extension

Enrollment in the OLE was available exclusively to subjects who had completed the double-blind phase of a Phase III study. The ongoing OLE is comprised of three stages: (1) a 16-week, blinded conversion period; (2) a 256-week, open-label maintenance period; and (3) a 4-week follow-up period [10]. Data presented herein represent results from up to the first 52 weeks of PER treatment.

Table 1

Use of concomitant antiepileptic drugs at baseline, from pooled Phase III study data.

Antiepileptic drug Placebo group (N = 442) n (%) Total perampanel group (N = 1038) n (%)

Carbamazepine 143 (32.4) 348 (33.5)

Lamotrigine 125 (28.3) 333 (32.1)

Levetiracetam 125 (28.3) 310 (29.9)

Oxcarbazepine 88 (19.9) 182 (17.5)

Topiramate 90 (20.4) 204 (19.7)

Valproic acid 140 (31.7) 338 (32.6)

CYP enzyme inducing (EIAEDs) 255 (57.7) 610 (58.8)

1 AED 27 (6.1) 66 (6.4)

2 AEDs 131 (29.6) 313 (30.2)

3 AEDs 97 (21.9) 231 (22.3)

Non-EIAEDs 187 (42.3) 427 (41.1)

1 AED 33 (7.5) 80 (7.7)

2 AEDs 87 (19.7) 220 (21.2)

3 AEDs 67 (15.2) 127 (12.2)

1 AED 60 (13.6) 146 (14.1)

2 AEDs 218 (49.3) 533 (51.3)

3 AEDs 164(37.1) 358 (34.5)

All subjects who had attained a once daily, 12-mg dose of PER during their Phase III study remained on that dose upon entry into the OLE. Subjects who previously received PBO or whose dose of PER was

< 12 mg/day remained blinded to their Phase III dose, which was uptitrated during the conversion period in 2-mg increments every 2 weeks to an individualized maximum tolerated dose (MTD) of

< 12 mg/day; this titration rate was half as fast as the 2-mg/week titration of the Phase III studies. Subjects remained on their MTD of PER unless the investigator determined that additional dose titration was needed to address tolerability or efficacy issues. Discontinuation, also at the investigator's discretion, was automatic for subjects unable to tolerate a 2-mg daily dose. Concomitant AED regimens from the Phase III studies were maintained at entry into the OLE but could be adjusted at the investigator's discretion. Safety evaluations in the OLE were similar to those in the Phase III studies [10].

The study subjects were divided into two groups for this post hoc analysis: those who received PER during both the double-blind Phase III studies and the OLE (DB-PER, includes the 19 weeks of double-blind Phase III studies), and those who received PBO during Phase III and converted to PER during the OLE (DB-PBO, starts with the 16 weeks of conversion in the OLE). Results from the OLE are reported here as "time to first onset" for TEAEs, and were analyzed over a period of up to 52 weeks of PER treatment.

23. Statistical methods

A TEAE was defined as an adverse event that began either on or after the first dose date and up to 30 days after the last dose date of study drug, or that began before the first dose date and increased in severity during the treatment period. In the analysis of overall frequency, a subject who had two or more AEs with the same Medical Dictionary for Regulatory Activities (MedDRA) preferred term was counted only once. Adverse event frequency was also stratified by AE severity. In the computation of duration, if AE outcome was unrecovered, recovering, or unknown, the AE ending date was imputed with the last dose date plus 30 days. The analysis on duration of AEs was limited to the first AE if the subject had more than one AE.

Analysis by dose level was based on the dose groups at the end of the study. Four subjects who did not receive PER at the end of the study were not shown in figures and tables due to small n values, but were included in the total PER group. Because this was a post hoc analysis, p values are not provided.

3. Results

Safety results from all three Phase III studies were pooled, resulting in a safety population of 1480 study subjects (PER, n = 1038; PBO, n = 442). Of the 1264 study subjects who completed the double-blind Phase III studies, 1218 (96.4%) elected to enter the OLE. Of these, 380 were DB-PBO subjects, and 838 were DB-PER subjects. The safety population for the OLE was comprised of 1186 study subjects [10].

3.1. Pooled Phase III study data

Table 2 illustrates TEAEs occurring in > 5% of subjects in the safety analysis set who received PER during the double-blind Phase III studies and has been reported previously [11,12]. Very common TEAEs were defined as those occurring in > 10% of patients in any treatment group and included dizziness, somnolence, headache, fatigue, irritability, and fall. Of these, the four TEAEs occurring with greater frequency in the total PER group than in the PBO group were dizziness (28.1% vs 9.0%), somnolence (14.5% vs 7.2%), fatigue (8.5% vs 4.8%), and irritability (7.0% vs 2.9%). Headache, the third most very common TEAE in the PER group, occurred with approximately the same frequency in the PER and PBO groups (11.4% and 11.3%, respectively). As shown in Fig. 1, severity of the four TEAEs was mild or moderate in most cases.

Treatment-emergent adverse events occurring in >5% of safety population receiving perampanel by decreasing frequency, from pooled Phase III study data.

Four very common TEAEs occurring with greater frequency in the total PER group than in the PBO group were dizziness, somnolence, fatigue, and irritability (noted in bold). Post hoc analyses of the occurrence and nature of these 4 TEAEs were evaluated. Sources: [11,12].

MedDRA preferred termb Placebo (N = 442) n (%) Perampanela

2 mg/day (N = 180) n(%) 4 mg/day (N = 172) n(%) 8 mg/day (N = 431) n(%) 12 mg/day (N = 255) n(%) Total (N = 1038) n(%)

Subjects with any TEAE 294 (66.5) 111 (61.7) 111 (64.5) 350 (81.2) 227 (89.0) 799 (77.0)

Dizziness 40 (9.0) 18(10.0) 28(16.3) 137 (31.8) 109 (42.7) 292(28.1)

Somnolence 32 (7.2) 22 (12.2) 16(9.3) 67 (15.5) 45 (17.6) 150(14.5)

Headache 50 (11.3) 16 (8.9) 19 (11.0) 49 (11.4) 34(13.3) 118 (11.4)

Fatigue 21 (4.8) 8 (4.4) 13 (7.6) 36 (8.4) 31 (12.2) 88 (8.5)

Irritability 13 (2.9) 7 (3.9) 7(4.1) 29 (6.7) 30(11.8) 73 (7.0)

Nausea 20 (4.5) 4 (2.2) 5 (2.9) 25 (5.8) 20 (7.8) 54 (5.2)

Fall 15 (3.4) 2 (1.1) 3(1.7) 22 (5.1) 26 (10.2) 53 (5.1)

A treatment-emergent adverse event (TEAE) is defined as an adverse event that either begins on or after the first dose date and up to 30 days after the last dose date of study drug, or begins before the first dose date and increases in severity during the treatment period. A subject with two or more adverse events with the same preferred term is counted only once for that preferred term.

a Subjects treated during the double-blind study. Dose groups are based on the actual (randomized) treatment groups.

b Medical Dictionary for Regulatory Activities (MedDRA) preferred terms are sorted in descending order of frequency in the total perampanel column.

The incidence rates for severe dizziness (2.7% vs 0.2%), somnolence (0.6% vs 0.2%), fatigue (0.5% vs 0.0%), and irritability (0.4% vs 0.0%) were low in the total PER and PBO populations, respectively. In the 6-mg group, severe dizziness was reported by eleven subjects (13.8%), and no subjects reported severe fatigue. In the 8-mg group, severe dizziness or fatigue was reported by eight (2.1%) and two (0.5%) subjects, respectively. Two subjects (5.4%) from the 10-mg PER group reported severe dizziness, and one subject (2.7%) reported severe fatigue, while no subject in the 12-mg PER group reported a severe case of dizziness or fatigue. Severe somnolence was reported by three subjects (3.8%)

in the 6-mg group, two subjects (0.5%) in the 8-mg group, no subjects in the 10-mg group, and one subject (0.6%) in the 12-mg group. Severe irritability was reported by one subject (0.5%) in the 4-mg group, one (0.3%) in the 8-mg group, and two (1.3%) in the 12-mg group, but no subjects in the 2-mg, 6-mg, or 10-mg groups.

Forty-three subjects (4.1%) in the PER safety population discontinued treatment due to these four TEAEs, compared to six subjects (1.4%) in the PBO group. Dizziness was the most common event leading to discontinuations in the PER group (22 [2.1%]), and it occurred more frequently with increasing PER dose, thus, exhibiting a dose response.

Fig. 1. Occurrences of treatment-emergent adverse events of interest by severity rating, from pooled Phase III study data. TNote: There were 4 patients in the PER 0-mg/day group; 2 of these patients experienced mild dizziness and are included in the PER total. PER, perampanel.

Discontinuation rates for the three other TEAEs were much lower (somnolence, 10 [1.0%]; fatigue, 7 [0.7%]; and irritability, 4 [0.4%]), with only discontinuations due to somnolence and irritability exhibiting a dose response with increasing PER dose. Study drug interruption or dose reduction resulting from these four TEAEs occurred in 136 (13.1%) PER subjects and seven (1.6%) PBO subjects. The majority of drug interruptions or dose reductions in PER subjects resulted from dizziness (85 [8.2%]), whereas somnolence (30 [2.9%]), fatigue (12 [1.2%]), and irritability (9 [0.9%]) resulted in fewer dose adjustments. For each of the four TEAEs, more patients receiving 8 and 12 mg of PER required drug interruption or dose reduction compared to those receiving 2 mg and 4 mg (based on actual treatment groups).

Time to the first onset of each of the four TEAEs for the pooled PER and PBO groups is shown in Fig. 2. Although new onset of these TEAEs occurred throughout the 19-week, double-blind phase, most events emerged during the first 6 weeks (the titration phase). This trend for early emergence was true for both the PER and PBO groups, although TEAE rates were higher among subjects receiving PER. Of all cases of

first-onset dizziness in the PER group, 83.2% occurred during the titration phase, with a peak at Week 3. This phase also included 87.3% of first-onset somnolence, 72.7% of first-onset fatigue, and 56.2% of first-onset irritability. The peaks for all three of these TEAEs occurred at Week 1.

The duration of the first occurrence of the four TEAEs was highly variable in each PER dose group and the PBO group, ranging from 1 day to more than 3 months. For both dizziness and somnolence, the first occurrence had a longer median duration in each PER group than PBO, whereas the first occurrences of fatigue and irritability had median durations that were more variable across PER groups compared to PBO (Fig. 3). At PER doses of 12 mg, the first occurrences of dizziness and somnolence had longer median durations than at all lower doses, indicative of a dose-response relationship for the duration of the first occurrence of these two TEAEs. These results also show that the median duration of the first occurrence was shorter in the total PER population than in the PBO group for fatigue (46 days vs 77 days) and similar in the two treatment groups for irritability (56 days vs 61 days), but it was longer

Fig. 2. Time to the first onset of treatment-emergent adverse events of interest in perampanel (N = 1038) and placebo (N = 442) groups, from pooled Phase III study data. PER, perampanel; PBO, placebo; TEAE, treatment-emergent adverse event.

Fig. 3. Median duration of treatment-emergent adverse events of interest: Duration of the first occurrence per group, from pooled Phase III study data. Dose groups are based on last dose. *n = number of subjects (placebo + total perampanel). TNote: There were 2 patients in the perampanel 0-mg/day group who experienced dizziness; these patients are included in the perampanel total. PER, perampanel.

in the PER group for dizziness (18 days vs 9 days) and somnolence (43.5 days vs 25 days).

3.2. Open-label extension data

Fig. 4 compares the time to the first onset of TEAEs of interest in the DB-PBO group with that in the DB-PER group. Throughout 52 weeks of PER treatment, corresponding patterns of the first onset of TEAEs were generally consistent between the DB-PER and DB-PBO groups. For TEAEs occurring in > 1% of subjects in the first week during titration (dizziness, somnolence, and fatigue), the rate of the first onset in both DB-PER and DB-PBO groups tapered off at approximately Week 10 (4 weeks after titration). Given that the half-life of perampanel is approximately 105 h, this corresponds to the maintenance phase when steady-state pharmacokinetics has been achieved [1]. With regard to irritability, the rates of the first onset for both groups were < 1%, even during titration (with the exception of Week 5 in the DB-PBO group, when the rate increased to 1.3%). In the DB-PER group, an increase in the rate of first-onset dizziness and somnolence was seen at the beginning of Week 20, corresponding with the OLE conversion period, in which subjects were uptitrated to their individualized MTDs [10]. Specifically, the rate of the first onset for dizziness at Week 19 was 0.7%, increasing to a peak of 3% at Week 23, while the somnolence rate was 0.4% at Week 19, increasing to a peak of 1% at Week 22. After approximately 26 to 52 weeks of PER treatment, both the DB-PBO and DB-PER groups had < 1% of patients experiencing the first onset of these four events.

For the DB-PBO group, the first occurrences of the four TEAEs during PER treatment manifested primarily during the 16-week conversion period: dizziness, with a peak of 7.1% at Week 3; somnolence, with a peak of 2.6% at Week 3; fatigue, with a peak of 2.4% at Week 1; and irritability, with a peak of 1.3% at Week 5. This observation is similar to the results for the patients in the DB-PER group, who experienced the most and the highest first onset of events during their titration period in the double-blind Phase III studies followed by an increase in the rate of the first onset as subjects were uptitrated to their individualized MTDs in the OLE.

4. Discussion

The pooling of data from the three Phase III studies allows for a more comprehensive perspective on the onset and characteristics of TEAEs associated with the use of PER in patients with partial seizures than can be developed by examining the studies individually. Furthermore, the inclusion of up to 52 weeks of data from the OLE provides additional support to the double-blind Phase III data and allows insight into the long-term safety of PER treatment. These data reveal dizziness, somnolence, fatigue, and irritability to be the most common TEAEs occurring at a rate higher than that seen with PBO. They also show that the majority of TEAEs observed in all of these studies were mild or moderate in severity and largely occurred during the titration period, with a very low rate of AEs in the open-label phase.

Dose-group analyses performed on TEAE severity ratings and duration of the first occurrence provide insight into tolerability, particularly with higher PER dose groups. The observed discontinuations due to TEAEs or drug interruption or dose reductions in patients receiving 8-mg and 12-mg PER (based on actual treatment groups) may help explain the slightly higher rates of mild, moderate, or severe TEAEs in the subjects receiving 6-mg and 10-mg PER. Specifically, patients who received 6-mg and 10-mg PER (based on last dose) were patients who were not able to stay in their randomized 8-mg or 12-mg dose groups. It is also important to note that these were small n values (80 and 37 patients received 6-mg and 10-mg PER, respectively); thus, the results should be interpreted with caution. For duration of the first occurrence of the TEAEs of interest, the median durations for the 6-mg and 10-mg dose groups were not notably different from those of the 8-mg and 12-mg PER groups. Nonetheless, a majority of subjects who experienced dizziness, somnolence, fatigue, or irritability continued the study, although some at reduced doses. The data presented support the notion that these TEAEs are manageable. In a real-world setting where the weekly or biweekly clinical study titration schedule of PER is not followed, these numbers may be more favorable. Given the long half-life of PER, a slower titration schedule, such as that recommended for special populations, may be more appropriate; thus, individual dosing should be adjusted based on tolerability.

The timing of AE occurrence is another important factor to consider in addition to dose-group analyses. The first onset of most of the TEAEs

A) Dizziness

5 4 3 2 1 0

DB-PBO (blinded conversion to individualized PER MTD, weeks 1-16) DB-PER (titration to randomized dose, weeks 1-6)

I DB-PER (uptitrated to individualized MTD, weeks 20-36)

i kv I -■-DB-PER (DB Phase] -*-DB-PER (Extension Phase) -♦-DB-PBO (Extension Ftiase}

Time on PER Treatment

B) Somnolence

DB PBO (blinded conve-sion to individualized PER MTO, weeks 1-16) DB-PER (titration to randomized dose, weeks 1-6)

DB-PER (uptitrated to individualized MTD, weeks 20-36)

^ -B-DB PER (DB Phase)

DB PER (Extension Phase) DB-PBO (Extension Phase)

V/////////////////////^

Time on PER Treatment

C) Fatig

DB-PBO (blinded conversion to individualized PER MTD, weeks 1-16) DB-PER (titration to randomized dose, weeks 1-6}

DB-PER (uptitrated to individualized IMTD, weeks 20-36)

-DB-PER (DB Phase) -DB-PER (Extension Phase} DB-PBO (Extension Phase)

-Si ^ ^ ^ # Jjp ^r # # ^ ^ ^ # Jfp ^

Time on PER Treatment

D) Irritability

DB-PBO (blinded conversion to individualized PER MTD, weeks 1-16) DB-PER (trtration to randomized dose, weeks 1-6}

DB-PER (uptitrated to individualized MTD, weeks 20-36)

-DB-PER (DB Phase) -DB-PER (Extension Phase} DB-PBO (Extension Phsse)

///^v////////////////^

Time on PER Treatment

Fig. 4. Time to the first onset of treatment-emergent adverse events of interest for up to 52 weeks of perampanel treatment, from pooled Phase III study and open-label extension (OLE) data (DB-PER, N = 838; DB-PBO, N = 378). DB-PBO = patients on placebo during the double-blind (DB) phase who converted to perampanel treatment during the OLE. Blindedconver-sion to MTD for patients in the DB-PBO group occurred during Weeks 1-16. DB-PER = patients on PER treatment during the DB phase (shown as Weeks 1-19) and OLE (Weeks 20-52). Titration to randomized dose for patients in the DB-PER group occurred during Weeks 1-6, followed by 13-week maintenance in the DB phase, and blinded titration to MTD during Weeks 20-36. DB, double-blind; MTD, maximum tolerated dose; OLE, open-label extension; PBO, placebo; PER, perampanel.

occurred during initial PER treatment, with peak first onset occurring within the first 6 weeks. This early-onset pattern was similar in the double-blind Phase III studies, in which the titration rate was 2 mg every week to randomized dose for 6 weeks, and in the OLE, in which the titration rate was 2 mg every 2 weeks to MTD for 16 weeks. In addition to titration schedules, there were differences in doses between the double-blind Phase III and OLE studies; specifically, in the double-blind phase, patients were randomized to lower PER doses, including 2 mg and 4 mg, whereas in the OLE, patients entered the study on their final dose of PER or PBO and were uptitrated to their individualized MTDs of PER, which were < 12 mg per day.

The data presented here show that a greater percentage of patients experienced a first onset of dizziness and somnolence during Weeks 1-10 in the DB-PBO group versus the DB-PER group (Fig. 4A and B). This could be because the DB-PBO group uptitrated toward 12 mg or MTD, while the DB-PER group would have reached their various assigned dose levels from 2 mg to 12 mg during this period. On the other hand, the percentages of patients experiencing fatigue and irritability were similar in the DB-PBO and DB-PER groups (Fig. 4C and D). Nonetheless, patients from both groups were uptitrated to their MTDs, which likely increased the rate of the first onset of TEAEs despite the slower titration rate during the OLE conversion period. For patients in the DB-PBO group, this occurred during Weeks 1 and 16. For patients in the DB-PER group, uptitration to MTD occurred from Weeks 20 to 36, at which point patients in the DB-PBO group were maintaining their steady MTD doses, leading to the seemingly high incidence of dizziness in the DB-PER extension phase compared to DB-PBO. Despite the opposing contributions of slower titration rate and uptitration to MTD, results from the double-blind Phase III studies and the OLE suggest consistency between studies in patient responses with regard to early-onset AEs. Treatment with PER resulted in very low to almost no incidence of the first onset of dizziness, somnolence, fatigue, or irritability after approximately 26 to 52 weeks of maintenance therapy. This is in line with safety findings from the OLE study, which concluded that in patients with partial seizures, adjunctive PER treatment at doses up to 12 mg/day had a good tolerability profile [10].

Because PER is approved for use as adjunctive therapy for partial seizures, other concomitantly prescribed AEDs may influence the etiology of the TEAEs observed in these studies. Eligible patients were required to be receiving 1 -3 AEDs upon study entry, and no dose adjustment, discontinuation, or change of concomitant AEDs was allowed during the Phase III studies [5-7]. However, during the OLE, concomitant AEDs were subject to adjustment, discontinuation, or change at the investigator's discretion [10]. Concomitant AEDs used in the Phase III studies are listed in Table 1. According to a systematic review of AEDs published by the Agency for Healthcare Research and Quality, dizziness is a common AE seen with treatment with carbamazepine, oxcarbazepine, and lamotrigine, while a separate meta-analysis of placebo-controlled AED studies observed coordination disturbance (of which dizziness is a common manifestation) as an AE frequently associated with topiramate therapy [8,9]. Psychiatric AEs, such as irritability, although widely associated with AED use, are notably more common with topiramate, levetiracetam, and carbamazepine than with other AEDs [8,9,13]. Somnolence has been observed as a feature of treatment with carbamazepine, lamotrigine, and levetiracetam, while the risk for fatigue is increased in patients taking valproic acid, lamotrigine, or car-bamazepine [8]. Although it is not possible to precisely determine the effect of concomitant AED use on the incidence of AEs in this analysis, it is important to note that the use of each AED was approximately equal in both PER and PBO groups, suggesting that the occurrence of AEs may be attributable to PER alone.

The implications of concomitant AEDs for AE risk extend beyond their individual AE risk profiles, as drug-drug interactions between PER and other AEDs also likely play a role. For patients whose concomitant AED regimen includes agents that are CYP enzyme inducing (EIAEDs), the recommended starting daily dose of PER is 4 mg, whereas

the starting dose for those not receiving EIAEDs is 2 mg [1]. Reductions in PER plasma levels range from 50% to 67% when it is taken with EIAEDs, including carbamazepine, oxcarbazepine, or phenytoin, and a pooled analysis of Phase III study data showed a relative reduction in the effect of PER when it is taken along with enzyme inducers [1]. Conversely, concomitant treatment with AEDs that are not enzyme inducing, such as levetiracetam, valproic acid, lamotrigine, zonisamide, clobazam, or clonazepam, showed no effect on PER clearance [1]. In the present analysis of onset and duration of AEs, patients were not stratified by the presence or absence of EIAEDs in their AED regimen. Nevertheless, it should be noted that EIAED use and the corresponding change in PER plasma concentration may contribute to the onset and duration of AEs.

5. Conclusions

Dizziness, somnolence, fatigue, and irritability were the most common AEs observed in three double-blind Phase III studies and in up to 52 weeks of the OLE study of PER for the treatment of partial seizures. These events were largely mild or moderate in severity; nonetheless, as supported by the data presented, individual dosing should be adjusted based on tolerability. Although the first onset of most AEs occurred during the titration period of the double-blind Phase III studies or the conversion period of the OLE, the frequency of AE occurrence was slightly different between these two periods. Although there is a slower titration rate in the OLE conversion period compared to that in the titration period of the double-blind Phase III studies, which represents a potential source of this divergence, dosing differences between the doubleblind studies and the OLE may also contribute. Overall, the incidence of the first onset of these AEs after 6 months and up to 1 year of treatment, which is low to nonexistent, provides further evidence that long-term treatment with PER is safe and well tolerated.

Role of funding source

This study was funded by Eisai Inc., and the authors who are employees of Eisai were involved in the study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.

Editorial support was funded by Eisai Inc. and provided by Imprint Publication Science, New York, NY.

Acknowledgments

Dr. Ko: study concept and design, interpretation of data analyses, and critical revision of the manuscript for important intellectual content. Dr. Yang: acquisition, analysis and interpretation of data, and critical revision of the manuscript for important intellectual content. Dr. Williams: data analysis and interpretation and critical revision of the manuscript for important intellectual content. Ms. Xing: biostatistical data analysis. Dr. Laurenza: acquisition, analysis and interpretation of data, and critical revision of the manuscript for important intellectual content.

Disclosure of conflicts of interest

David Ko is a consultant to Supernus and Upsher-Smith and serves on the speaker bureaus for Eisai, Lundbeck, Supernus, and UCB.

Haichen Yang, Betsy Williams, and Antonio Laurenza are employees of Eisai Inc.

Dongyuan Xing is a former employee of Eisai Inc. References

[1] FYCOMPA (perampanel) US Prescribing Information. Woodcliff Lake, NJ: Eisai Inc.; 2014.

[2] FYCOMPA (perampanel) Product Monograph. Mississauga, Ontario, Canada: Eisai Limited; 2013.

[3] FYCOMPA (perampanel) Summary of Product Characteristics. Hatfield, Hertfordshire, UK: Eisai Europe Limited; 2012.

[4] Press release: Eisai submits applications for antiepilepsy agent Fycompa® simultaneously in Europe and U.S. seeking indication expansion as adjunctive treatment of primary generalized tonic-clonic seizures. http://www.eisai.com/news/enews201445pdf.pdf. Tokyo, Japan: Eisai Co., Ltd.; 2014.

[5] French jA, Krauss GL, Biton V, Squillacote D, Yang H, Laurenza A, et al. Adjunctive perampanel for refractory partial-onset seizures: randomized phase III study 304. Neurology 2012;79:589-96.

[6] French JA, Krauss GL, Steinhoff BJ, Squillacote D, Yang H, Kumar D, et al. Evaluation of adjunctive perampanel in patients with refractory partial-onset seizures: results of randomized global phase III study 305. Epilepsia 2013;54:117-25.

[7] Krauss GL, Serratosa JM, Villanueva V, Endziniene M, Hong Z, French J, et al. Randomized phase III study 306: adjunctive perampanel for refractory partial-onset seizures. Neurology 2012;78:1408-15.

[8] Talati R, Scholle JM, Phung OJ, Baker WL, Baker EL, Ashaye A, et al. Effectiveness and safety of antiepileptic medications in patients with epilepsy. Comparative

Effectiveness Review, Number 40. AHRQ Publication No. 11(12)-EHC082-EF. Rockville, MD: Agency for Healthcare Research and Quality; 2011.

[9] Perucca P, Gilliam FG. Adverse effects of antiepileptic drugs. Lancet Neurol 2012;11: 792-802.

[10] Krauss GL, Perucca E, Ben-Menachem E, Kwan P, Shih JJ, Squillacote D, et al. Perampanel, a selective, noncompetitive a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor antagonist, as adjunctive therapy for refractory partial-onset seizures: interim results from phase III, extension study 307. Epilepsia 2013;54:126-34.

[11] Steinhoff BJ, Ben-Menachem E, Ryvlin P, Shorvon S, Kramer L, Satlin A, et al. Efficacy and safety of adjunctive perampanel for the treatment of refractory partial seizures: a pooled analysis of three phase III studies. Epilepsia 2013;54:1481-9.

[12] Kramer LD, Satlin A, Krauss GL, French J, Perucca E, Ben-Menachem E, et al. Perampanel for adjunctive treatment of partial-onset seizures: a pooled dose-response analysis of phase III studies. Epilepsia 2014;55:423-31.

[13] Cramer JA, Mintzer S, Wheless J, Mattson RH. Adverse effects of antiepileptic drugs: a brief overview of important issues. Expert Rev Neurother 2010;10:885-91.