Safety evaluation of perampanel as monotherapy or first adjunctive therapy in patients with epilepsy

ABSTRACT

Introduction

There is a need for anti-seizure medications (ASMs) that are well tolerated and effective as monotherapy or first adjunctive therapy to reduce the need for adjunctive ASMs to treat newly diagnosed epilepsy, and to reduce the number of concomitant ASMs in patients with refractory epilepsy. Although the pivotal trials of perampanel evaluated its adjunctive use in patients with refractory seizures, open-label/real-world studies support its use in first/second-line settings.

Areas covered

This paper reviews the pharmacology, efficacy, and safety/tolerability of perampanel, focusing on its use as monotherapy or first adjunctive therapy. The safety of perampanel in special populations and its safety/tolerability compared with that of other ASMs is also discussed.

Expert opinion

Perampanel is a favorable candidate for initial or first adjunctive therapy due to its favorable efficacy and safety/tolerability as monotherapy and adjunctive therapy, its long half-life and ease of use, and its limited drug–drug interactions. The proposed mitigation strategies for managing the risk of serious psychiatric adverse events are appropriate patient selection, use of low doses, and slow titration. The growing body of evidence might shift current treatment strategies toward the early use of perampanel and its use at a low dose (4 mg/day).

KEYWORDS:

• Anti-seizure medications
• first-line
• first adjunctive therapy
• monotherapy
• perampanel
• safety
• second-line

1. Introduction

Approximately 50% of patients with epilepsy fail to achieve seizure control with first-line anti-seizure medication (ASM); therefore, an alternative monotherapy or concomitant ASMs are commonly prescribed with the aim of improving seizure control [1]. However, adjunctive treatment has been associated with an increased frequency of adverse events (AEs) due to the combination of different AE profiles and potential pharmacokinetic (PK)/pharmacodynamic interactions between the ASMs [2,3]. Thus, ASMs that are well tolerated and effective as monotherapy or first adjunctive therapy are needed for reducing the need for adjunctive ASMs to treat newly diagnosed epilepsy and reducing the number of concomitant ASMs to treat refractory epilepsy. Further considerations when choosing first-line or first adjunctive ASMs include compatibility with other ASMs, convenient route of administration, and long half-life [3,4].

Perampanel, a selective, noncompetitive α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor antagonist is a once-daily oral ASM [5–7] (Box 1). Perampanel was initially approved as adjunctive therapy for focal onset seizures (FOS) and later for generalized tonic-clonic seizures (GTCS) in the US and EU, based on pivotal Phase III randomized controlled trials (RCTs) in patients with refractory FOS (Studies 304 [NCT00699972], 305 [NCT00699582], and 306 [NCT00700310]) or GTCS (Study 332 [NCT01393743]) [8–11]. Perampanel monotherapy for FOS later gained approval in the US, based on efficacy and safety/tolerability data extrapolated from pivotal trials of adjunctive perampanel, and in Japan, based on the open-label Phase III Study 342 (NCT03201900) [12,13]. Notably, Study 342 was the first study in which perampanel was administered as first-line monotherapy [12]. Perampanel has been on the market for 10 years (initial approval was in 2012) and thus has a well-established safety profile and acceptable tolerability. As of 22 July 2018, there had been an estimated ~45 million patient-days of perampanel exposure since product launch [14].

There are increasing open-label/real-world data on perampanel monotherapy and first adjunctive therapy across different seizure types and age groups [12,15–26] that complement data from RCTs. Perampanel use with fewer previous/concomitant ASMs was associated with improved seizure control and reduced likelihood of AEs, suggesting an optimized therapeutic benefit when taken early in the treatment pathway [26]. In line with this growing body of evidence, experts justified perampanel use as a first adjunctive therapy in the 2021 Italian consensus, based on its efficacy, tolerability, retention rate, limited drug–drug interactions, and ease of use [4].

This review article summarizes the safety/tolerability of perampanel monotherapy and first adjunctive therapy, and provides an expert opinion on perampanel use in first/second-line settings.

2. Mechanism of action and pharmacology

Perampanel is the only currently approved ASM that acts by selectively blocking AMPA receptors [4–7] (Figure 1). Perampanel is rapidly absorbed after oral administration and is ~95% bound to plasma proteins [5,6]. It is primarily metabolized by cytochrome P450 3A4/5 and has a half-life of ~105 hours [5,6].

Figure 1. Mechanism of action of perampanel.

AMPA, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid

Perampanel has no apparent drug–drug interactions with other ASMs that worsen tolerability, although its apparent clearance is increased by enzyme-inducing ASMs (EIASMs; including carbamazepine, oxcarbazepine, and phenytoin), which decreases perampanel exposure [4–6,27]. Therefore, patients receiving concomitant EIASMs may require a higher perampanel dose to achieve similar efficacy to patients receiving non-EIASMs [28,29].

Perampanel has few drug–drug interactions apart from the effect of concomitant EIASMs on its exposure [5,6]. Although using 12-mg/day perampanel with contraceptives containing levonorgestrel may render these contraceptives less effective, perampanel does not reduce the exposure of these contraceptives when co-administered at 4 or 8 mg/day [4,5].

3. Clinical efficacy

In pivotal RCTs (Studies 304, 305, and 306), adjunctive perampanel demonstrated efficacy in improving FOS control with doses of ≥4 mg/day [8–10,30]. In a pooled analysis of these trials, reduction in FOS was significantly higher with perampanel 4–12 mg/day than placebo (median, 23.3–28.8% vs 12.8%, respectively; all P < 0.01) [31]. The 50% responder and seizure-freedom rates with perampanel 4–12 mg/day were 28.5–35.3% and 3.5–4.4%, respectively, which were significantly higher vs the placebo group (all P < 0.05) [31]. Perampanel 8 mg/day was efficacious in patients with GTCS in Study 332, with median GTCS reduction from baseline of 76.5%, 50% responder rate of 64.2% (GTCS only), and seizure-freedom rate of 23.5% (all seizures) [11].

RCT data showed reduction in seizure frequency with perampanel across multiple seizure types (FOS, FBTCS, GTCS), with greater reductions observed in FBTCS vs FOS [11,31,32]. This supports use of perampanel as a broad-spectrum agent in patients with unclear or multiple seizure types [32].

The effectiveness of perampanel monotherapy and adjunctive therapy has been demonstrated in open-label/real-world studies, with seizure-freedom rates of up to 76% reported in patients with FOS [12,15–25]. Discontinuations due to inadequate therapeutic effect or lack of efficacy were reported in 4.9–6.7% of patients receiving perampanel as a first-line monotherapy [12,18,19].

4. Safety evaluation

4.1. Safety/tolerability of perampanel adjunctive therapy in RCTs

In the pooled analysis of pivotal FOS trials (n = 1038 perampanel-treated patients) [31], treatment-emergent AEs (TEAEs) occurred in 77.0% of patients receiving adjunctive perampanel (vs 66.5% with placebo); however, fewer TEAEs were reported with lower perampanel doses (61.7% and 64.5% with 2 and 4 mg/day, respectively) than with higher doses (81.2% and 89.0% with 8 and 12 mg/day, respectively). Serious TEAEs were reported in 5.5% with adjunctive perampanel vs 5.0% with placebo; serious psychiatric TEAEs were reported in 1.2% of patients receiving perampanel, the most common being aggression.

In patients with GTCS (n = 81 perampanel-treated patients) [11], TEAEs occurred in 82.7% with adjunctive perampanel 8 mg/day vs 72.0% with placebo. Serious TEAEs were reported in 7.4% with perampanel and 8.5% with placebo; serious psychiatric TEAEs were reported in 2.5% of patients receiving perampanel (one event of suicide ideation and one of suicide attempt).

The most common TEAEs with adjunctive perampanel (≥10% of patients in any treatment group) were dizziness, somnolence, headache, fatigue, and irritability in patients with FOS/GTCS, plus falls in patients with FOS [11,31]. The incidence of TEAEs leading to treatment discontinuation was ~10% in perampanel-treated patients (FOS: 9.5% with perampanel vs 4.8% with placebo, most common: dizziness, convulsion, and somnolence; GTCS: 11.1% with perampanel [most common: dizziness and vomiting] vs 6.1% with placebo) [11,31].

Based on pooled data from RCTs (n = 1119 perampanel-treated patients), hostility/aggression-related TEAEs were reported in 12.2% of perampanel-treated patients with FOS/GTCS, including irritability (7.3%), aggression (1.6%), and anger (1.1%) [33]. Homicidal ideation and/or threat was reported in 0.1% of 4368 perampanel-treated patients in Phase II/III trials and their open-label extensions for FOS, and non-epilepsy studies [34].

In a post hoc analysis of low-dose perampanel (4 mg/day; n = 1376), TEAEs were reported in 30.5% of patients, serious TEAEs in 1.2% of patients, and TEAEs leading to discontinuation in 2.0% of patients, suggesting perampanel 4 mg/day is generally well tolerated [30].

4.2. Safety/tolerability of perampanel monotherapy or early adjunctive therapy in open-label/real-world studies

Data from four Eisai-sponsored trials that included patients taking perampanel monotherapy (Study 342 and the real-world Phase IV Studies 504 [NCT02736162] and 506 [PROVE; NCT03208660]) or first adjunctive perampanel (the open-label Phase IV Study 412 [NCT02726074]) were selected by the authors for review in this section [12,15,16,20]. Additionally, a PubMed literature search was conducted in January 2022 to identify other relevant studies of perampanel monotherapy/first adjunctive therapy. Shortlisted articles included the terms ‘perampanel’ AND ‘monotherapy’ OR ‘first’ AND ‘adjunctive’ OR ‘add-on.’ Six additional articles were identified and included, which described studies that were considered relevant in terms of clinical safety and included ≥20 patients receiving perampanel monotherapy (two studies; Toledano Delgado retrospective study and Chinvarun retrospective study) or first adjunctive therapy (four studies; PERADON, Santamarina retrospective study, Rea retrospective study, and COM-PER study) [18,19,22–25].

The most common AEs reported with perampanel in these open-label/real-world studies are summarized in Table 1. The incidence of TEAEs ranged from 20.0–75.3% with perampanel monotherapy and 38.1–75.5% with first adjunctive perampanel (including patients taking perampanel second adjunctive therapy in PERADON); serious TEAEs were reported in 0.0–10.1% and 0.0–7.8% of patients, respectively (where reported), and the incidence of TEAEs leading to discontinuation were 6.7–16.3% and 9.5–22.2%, respectively [12,15–25].

Table 1. Summary of study length, TEAEs, and most common TEAEs (occurring in ≥5% of patients) across studies that included patients treated with perampanel monotherapy or early adjunctive therapy (Safety Analysis Sets, where specified).

	Monotherapy						Early adjunctive therapy
	Study 342 (FREEDOM; prospective)^a,b [12]
	Perampanel 4 mg/day (N = 89)	Perampanel 4 and/or 8 mg/day (N = 89)	Study 504 (retrospective)^a,c,d [15] (N = 60)	Study 506 (PROVE; retrospective)^a,c,d [16,17] (n = 47)	Toledano Delgado retrospective study^c,d [18] (N = 98)	Chinvarun retrospective study^d [19] (N = 39)	Study 412 (prospective)^b,e [20] (N = 102)	PERADON (prospective)^f [22] (N = 113)	Santamarina retrospective study^d,e [23] (N = 149)	Rea retrospective study^d,e [24] (N = 27)	COM-PER study (retrospective)^d,e [25] (n = 21)
Study length	32 weeks^g		Up to 38 months^h	2 years	>1 year^i	1 year	36 weeks	1 year	1 year	6 months	1 year
TEAEs, n (%)	57 (64.0)	67 (75.3)	12 (20.0)	17 (36.2)	45 (45.9)	16 (41.0)	77 (75.5)	NR	72 (48.3)	18 (66.7)	8 (38.1)
Treatment-related TEAEs, n (%)	38 (42.7)	47 (52.8)	NR	NR	NR	NR	NR	34 (30.1)	NR	NR	NR
Serious TEAEs, n (%)	9 (10.1)	9 (10.1)	0 (0.0)	2 (4.3)	NR	0 (0.0)	8 (7.8)	0 (0.0)^j	NR	NR	NR
Most common TEAEs reported in ≥5% patients in any group, n (%)
Dizziness	20 (22.5)	28 (31.5)	3 (5.0)	6 (12.8)	8 (8.1)	11 (28.2)	51 (50.0)	8 (7.1)^j	23 (15.4)	0 (0.0)	5 (23.8)
Nasopharyngitis	11 (12.4)	13 (14.6)	NR	0 (0.0)	NR	0 (0.0)	1 (1.0)	0 (0.0)^j	NR	0 (0.0)	NR
Somnolence	11 (12.4)	12 (13.5)	1 (1.7)	2 (4.3)	10 (10.2)	4 (10.3)	10 (9.8)	2 (1.8)^j	21 (14.1)	0 (0.0)	0 (0.0)
Headache	10 (11.2)	10 (11.2)	1 (1.7)	1 (2.1)	NR	0 (0.0)	9 (8.8)	0 (0.0)^j	6 (4.0)	0 (0.0)	NR
Epilepsy	5 (5.6)	5 (5.6)	NR	0 (0.0)	NR	0 (0.0)	NR^k	0 (0.0)^j	NR	0 (0.0)	NR
Irritability	2 (2.2)	3 (3.4)	2 (3.3)	0 (0.0)	17 (17.3)	0 (0.0)	2 (2.0)	9 (8.0)^j	21 (14.1)	4 (14.8)	3 (14.3)
Gait disturbance	NR	NR	NR	3 (6.4)	NR	0 (0.0)	1 (1.0)	0 (0.0)^j	NR	0 (0.0)	NR
Depression	NR	NR	NR	1 (2.1)	10 (10.2)	1 (2.6)	1 (1.0)	1 (0.9)^j	NR	0 (0.0)	NR
Anxiety	0 (0.0)	2 (2.2)	NR	NR	1 (1.0)	0 (0.0)	0 (0.0)	2 (1.8)^j	10 (6.7)	2 (7.4)	NR
Ataxia	NR	NR	NR	NR	NR	6 (15.4)	0 (0.0)	0 (0.0)^j	4 (2.7)	4 (14.8)	NR
Fatigue	NR	NR	NR	NR	NR	0 (0.0)	2 (2.0)	NR^j,l	9 (6.0)	0 (0.0)	NR
Vertigo	NR	NR	NR	NR	NR	0 (0.0)	0 (0.0)	0 (0.0)^j	NR	5 (18.5)	NR
Aggression	0 (0.0)	0 (0.0)	NR	NR	NR	0 (0.0)	0 (0.0)	4 (3.5)^j	NR	4 (14.8)	NR
Suicidal thoughts	NR	NR	NR	NR	NR	0 (0.0)	1 (1.0)	0 (0.0)^j	NR	2 (7.4)	NR

Note that there is a possibility that patients may have been included in more than one of the studies presented here

^aA patient with ≥2 TEAEs is counted only once for that event; ^bIncludes the core study data only; ^cIncludes patients receiving primary or secondary monotherapy; ^dNote that not all TEAEs that occurred during perampanel treatment may have been recorded in retrospective studies due to the nature of these studies; ^eIncludes patients receiving perampanel as first adjunctive therapy; ^fIncludes patients receiving perampanel as first or second adjunctive therapy; ^gFor patients who were up-titrated to 8 mg/day, study length was 30 weeks plus the time in which they received 4 mg/day; ^hMedian duration of perampanel monotherapy of 6.1 months at cutoff; ⁱMedian duration of exposure to perampanel monotherapy of 14 months; ^jTreatment-related TEAEs; ^kSeizure TEAEs were reported in 3.9% of patients; ^lTiredness was reported in 0.9% of patients

NR, not reported; TEAE, treatment-emergent adverse event

The incidence of psychiatric TEAEs was generally comparable with perampanel monotherapy/first adjunctive therapy (7.1–37.0% in open-label/real-world studies [18,20,22,24]) and perampanel adjunctive therapy (21.0–24.4% in real-world/post-approval studies [16,26,35] and 24.3% based on pooled RCT data [35]). No changes in cognition were observed with perampanel based on a systematic review including data from an RCT/open-label extension and real-world studies [36].

Evidence suggests that incidence of TEAEs can be dependent on titration schedule. In Study 412, lower incidence of overall TEAEs was observed with slow vs fast titration of perampanel (70.5% with titration at ≥2-week intervals vs 82.9% at <2-week intervals) [20]. This was consistent with a real-world study showing that slow vs fast titration of perampanel (at ≥3–4-week vs 1- or 2-week intervals) was associated with lower incidence of TEAEs (P = 0.006) and numerically lower incidence of psychiatric TEAEs [37].

4.3. Safety in special populations

4.3.1. Pregnant women

There are limited data available for safety of perampanel in pregnancy [38,39]. As of August 2018, 96 pregnancies were reported to have had perampanel exposure: 43 were known to result in normal live births, 28 did not reach full-term, 18 were lost to follow-up, and seven were ongoing at data cutoff [38]. Preclinical data suggest perampanel exposure during pregnancy could potentially lead to diverticulum of the intestine, delays in aspects of physical development, and pregnancy loss [38]. Thus, pregnant women taking perampanel should be carefully monitored (including drug-level monitoring, if available, since dose adjustments may be needed to maintain consistent exposure) and encouraged to enroll in a pregnancy registry to allow more data to be collected in this population [5,38].

4.3.2. Pediatric patients

The incidence of TEAEs with adjunctive perampanel in patients aged 4 to <12 years in Study 311 (NCT02849626) was 89% (most common: somnolence, 26%; nasopharyngitis, 19%; dizziness, irritability, and pyrexia, 13% each) [40]. The incidence of TEAEs and most common types of TEAEs reported in this population were generally comparable with those reported in adolescent and adult patients in RCTs with adjunctive perampanel 4–12 mg/day (total TEAEs, 64.5–89.0%; dizziness, 16.3–42.7%; somnolence, 9.3–17.6%; headache, 11.0–13.3%; fatigue, 7.6–14.8%; irritability, 4.1–11.8%) [11,31]. Interim data from Study 506 also supported use of perampanel monotherapy or adjunctive therapy in patients aged 1 to <18 years with a safety/tolerability profile consistent with that of the overall study population [41].

4.3.3. Older patients

In the small cohort of older patients (n = 28; aged ≥65 years) in pivotal FOS trials, the incidence of TEAEs was similar vs adults aged <65 years; however, falls, dizziness, fatigue, and balance disorders were more common in older patients [42]. In line with the evidence discussed in section 4.2, slow titration is often recommended when initiating a new ASM to reduce the risk of AEs and improve tolerability [20,37,43]. According to the prescribing information, slow perampanel titration is also advised in older patients to reduce the risk of TEAEs that are more common in this population [5,42]. In real-world studies, the safety/tolerability of perampanel in older patients (aged >60 or ≥65 years) was generally in line with that of younger patients and/or the overall study population assessed (total incidence of AEs: 34.8–80.0% in older patients vs 38.5–67.6% in younger patients/overall study population) [19,35,44–46].

4.4. Safety after overdosage

Post-marketing cases of accidental and intentional perampanel overdoses have been reported; the highest reported overdose was 300 mg [5,6]. AEs reported after an overdose include somnolence, stupor, coma, psychiatric or behavioral reactions, altered mental status, and dizziness or gait disturbances; all patients recovered from their TEAEs without sequelae. There is no specific treatment for perampanel overdose and standard supportive treatment for overdose management should be used [5,6].

5. Comparison with the safety/tolerability of other ASMs

The most common TEAEs reported in the prescribing information for perampanel and other commonly used ASMs, including brivaracetam, cenobamate, eslicarbazepine acetate, lacosamide, lamotrigine, and levetiracetam, are described in Table 2; there is some overlap in the most common TEAEs across these ASMs, although direct comparisons cannot be made with these data.

Table 2. Summary of the most common TEAEs (in ≥10% of patients for any ASM and at a higher frequency in the treatment group vs placebo) reported in clinical trials of adjunctive oral^a PER, BRV, CNB, ESL, LCM, LMT, and LEV in adults^b with FOS,^c as described in the US prescribing information for each ASM.

Most common TEAEs, %	PER [5]		BRV^d [47]		CNB [48]		ESL [49]		LCM [50]		LMT [51]		LEV [52]
	PER 4, 8, or 12 mg/day (n = 858)	Placebo (n = 442)	BRV 50, 100, or 200 mg/day (n = 803)	Placebo (n = 459)	CNB 100, 200, or 400 mg/day (n = 442)	Placebo (n = 216)	ESL 800 or 1200 mg/day (n = 825)	Placebo (n = 426)	LCM 200, 400, or 600 mg/day^e (n = 944)	Placebo (n = 364)	LMT 150–500 mg/day (n = 711)	Placebo (n = 419)	LEV 1000, 2000, or 3000 mg/day (n = 769)	Placebo (n = 439)
Dizziness	16–43	9	12	7	18–33	15	20–28	9	31	8	38	13	9	4
Somnolence	9–18	7	16^f	8^f	19–37	11	11–18	8	7	5	14	7	15	8
Headache	11–13	11	–	–	10–12	9	13–15	9	13	9	29	19	14	13
Irritability	4–12	3	3	1	0–2	0	–	–	–	–	3	2	–	–
Fatigue	8–12	5	9	4	12–24	7	4–7	4	9	6	–	–	–	–
Falls	2–10	3	–	–	–	–	1–3	1	–	–	–	–	–	–
Ataxia	1–8	0	≤3^g	≤1^g	2–6	2	4–6	2	8	2	22	6	1	3
Nausea	3–8	5	≤5^h	≤3^h	6–9	3	10–16	5	11	4	19	10	–	–
Blurred vision	1–4	1	–	–	2–4	0	5–6	1	8	3	16	5	–	–
Infection	≥3–4^i	≥3^i	–	–	≥2–5^j	≥0–3^j	≥2^k	≥1^k	–	–	–	–	13	8
Vomiting	2–4	3	≤5^h	≤3^h	2–5	0	6–10	3	9	3	9	4	–	–
Diplopia	1–3	1	–	–	6–15	2	9–11	2	11	2	28	7	2	1
Asthenia	1–2	1	–	–	0–3	1	2–3	2	2	1	–	–	15	9
Rhinitis	–	–	–	–	–	–	–	–	–	–	14	9	4	3
Pharyngitis	–	–	–	–	0–2	0	–	–	–	–	10	9	6	4
Rash	–	–	–	–	≥0–2^l	≥0^l	1–3	1	–	–	10	5	–	–

^aOral administration not specified for BRV in the prescribing information; ^bIncluded adults and adolescents ≥12 years of age for trials of perampanel; ^cPatients were described as having epilepsy in trials of LMT; ^dOnly TEAEs with a frequency that was >2% higher with BRV than with placebo were listed; ^e600 mg/day is 1.5-times greater than the maximum recommended dose; ^fReported as somnolence and sedation; ^gReported as cerebellar coordination and balance disturbances, which include ataxia, balance disorder, coordination abnormal, and nystagmus; ^hReported as nausea or vomiting; ⁱReported as upper respiratory tract infection; ^jReported as nasopharyngitis, pharyngitis, or urinary tract infection; ^kReported as urinary tract infection; ^lReported as rash papular

ASM, anti-seizure medication; BRV, brivaracetam; CNB, cenobamate; ESL, eslicarbazepine acetate; FOS, focal-onset seizures; LCM, lacosamide; LMT, lamotrigine; LEV, levetiracetam; PER, perampanel; TEAE, treatment-emergent adverse event

Systematic reviews and meta-analyses generally found that adjunctive perampanel was associated with a higher frequency of TEAEs than adjunctive brivaracetam and levetiracetam, and a similar number of TEAEs as adjunctive cenobamate, eslicarbazepine acetate, and lamotrigine [53–55]. However, one systematic review and indirect treatment comparison of adjunctive perampanel and brivaracetam found both ASMs generally had similar TEAE profiles [56]. Dizziness was reported more commonly with perampanel than brivaracetam or lacosamide and occurred at a similar frequency with eslicarbazepine acetate and levetiracetam; somnolence rates were similar for all these ASMs [53,56,57]. Perampanel was generally associated with similar or lower rates of TEAEs leading to discontinuation compared with that of brivaracetam, cenobamate, eslicarbazepine acetate, lacosamide, lamotrigine, and levetiracetam [53,54,56,58], but high-dose perampanel (12 mg/day) was associated with higher discontinuation rates than high-dose levetiracetam [55,58]. A systematic review of observational studies found weighted mean incidences of anger and aggression were similar for perampanel, brivaracetam, levetiracetam, and topiramate but the incidence of irritability was higher for perampanel than for brivaracetam and topiramate [59]. No evidence of increased risk of suicidality was found with perampanel, brivaracetam, cannabidiol, cenobamate, or eslicarbazepine acetate in a meta-analysis of Phase II/III RCTs [60].

6. Summary

Perampanel demonstrated efficacy for improving seizure control: in RCTs, highest seizure-freedom rates (up to 23.5%) were observed with adjunctive perampanel (8 mg/day) in patients with GTCS, whereas in open-label/real-world studies, highest seizure-freedom rates (up to 76%) were reported for perampanel monotherapy (4-mg/day median dose) in patients with FOS. Additionally, perampanel was generally well tolerated, with safety outcomes for perampanel as monotherapy or early adjunctive therapy consistent with those of adjunctive perampanel. The safety/tolerability of perampanel is generally consistent with that of other commonly used ASMs.

7. Expert opinion

7.1. Improvement over other therapies

Although pivotal studies included patients with refractory seizures receiving adjunctive perampanel [8–11], open-label/real-world studies support perampanel use in first/second-line settings. Seizure-freedom rates of up to 76% have been demonstrated with perampanel monotherapy/first adjunctive therapy in patients with FOS in open-label/real-world studies [12,15–25]. Additionally, perampanel use with fewer previous ASMs may be better tolerated based on a bivariate analysis of data pooled from 44 real-world studies (PERMIT) [26]. Furthermore, perampanel monotherapy or adjunctive therapy has demonstrated effectiveness at a low dose (with seizure-freedom rates of up to 63% reported specifically for 4-mg/day perampanel monotherapy in patients with FOS), which may lead to improved tolerability relative to that of higher doses, without compromising efficacy [4,12,30].

Perampanel is an oral tablet taken once daily and has a long half-life [5,6]; therefore, it has favorable PK properties. A long half-life of immediate-release ASMs is associated with reduced pill burden and may particularly benefit patients who are not fully adherent to prescribed medication by reducing the likelihood of seizures occurring after missed doses [61,62]. Furthermore, perampanel has few drug–drug interactions [4–6], allowing perampanel to be easily combined with other medications.

A systematic review and indirect treatment comparison found perampanel had greater efficacy than brivaracetam in patients taking concomitant levetiracetam, possibly because its mechanism of action is different to that of brivaracetam and levetiracetam [56]. Perampanel first adjunctive therapy was associated with higher adherence to treatment regimen and treatment persistence rates than carbamazepine, topiramate, or valproate in a retrospective observational cohort study of claims data in South Korea [63]. Additionally, lower healthcare resource utilization was reported with perampanel vs non-perampanel based ASM combinations in several studies [63–65].

7.2. Impact on current treatment strategies

Key aspects to consider when selecting a first-line ASM are favorable safety/tolerability profile, good efficacy with a low dose, and possibility for slow titration. ASMs that can be used at a low dose may provide benefit in terms of improved tolerability without compromising efficacy [2]. Based on evidence discussed in Sections 3 and 4.2, we believe perampanel is a favorable candidate for a first/second line of therapy.

The use of perampanel as first adjunctive therapy is supported by the 2021 Italian consensus statements, based on the optimal criteria to guide the choice of first adjunctive ASM [4]. It should be noted that many of these criteria also apply to optimal choice of first-line treatment [3]. These consensus statements considered aspects such as efficacy for both FOS and GTCS, tolerability (especially when used with fewer concomitant ASMs), favorable retention rate, and ease of use [4]. Additionally, perampanel fulfills recommendations for achieving a desirable safety profile suggested in Cramer et al. 2010: there is no evidence for the need of laboratory monitoring with perampanel (indicating a lack of organ toxicity), no lethal drug-related AEs have been identified, and AEs reported with perampanel (including psychiatric AEs) can usually be managed with dose adjustments and/or treatment discontinuation [2,4,11,31,35].

7.3. How likely are physicians to prescribe perampanel?

Perampanel is likely to be prescribed due to its route of administration, long half-life, ease of use (i.e. one pill taken once daily), unique mechanism of action, and compatibility with other ASMs. Its unique mechanism of action may lead to improved efficacy with monotherapy (when ASMs targeting other mechanisms have failed) and adjunctive therapy (by targeting multiple mechanisms at once) [3,4,56].

Evidence suggests patients with a history of psychiatric events are predisposed to psychiatric AEs with perampanel. A post hoc analysis of Phase III FOS trials found an increased risk of psychiatric AEs in patients with vs without prior psychiatric history, particularly with higher perampanel doses (8 or 12 mg/day; this association were also observed in the placebo group) [66]. Similarly, in real-world studies, the frequency of psychiatric AEs was significantly higher in patients with vs without prior psychiatric history (P < 0.001) [26,37], whereas no significant effect of perampanel on psychiatric symptoms was observed in the overall study populations [23,67]. Therefore, suggested mitigation strategies that could be considered when prescribing perampanel include appropriate patient selection (e.g. patients without history of psychiatric events and/or who are earlier in their treatment pathway), low starting dose, and slow titration [22,37,66].

7.4. Data gaps

More data on timing of rare/serious AEs during perampanel treatment and the incidence of these AEs across age groups are needed to identify further risk mitigation strategies. Additionally, further data on fetal safety during perampanel use in pregnancy are needed.

7.5. Future implications

Based on the evidence discussed, perampanel shows many characteristics favoring first/second-line use. Previous and new emerging data on efficacy and safety/tolerability of perampanel as monotherapy and first adjunctive therapy, and its effectiveness across a broad spectrum of seizure types, long half-life, limited drug–drug interactions, and efficacy at a low dose (4 mg/day) might lead to its greater use in both settings.

Table

Box 1. Drug summary
Drug name	Perampanel
Development status	Approved by the US Food and Drug Administration and European Medicines Agency for focal-onset seizures and generalized tonic-clonic seizures [5,6]
Indications	In the US, perampanel is approved for the treatment of focal-onset seizures (adjunctive and monotherapy), with or without focal to bilateral tonic-clonic seizures, in patients with epilepsy aged ≥4 years, and as adjunctive treatment of generalized tonic-clonic seizures in patients with epilepsy aged ≥12 years [5] In the European Union, perampanel is approved as adjunctive treatment of focal-onset seizures, with or without focal to bilateral tonic-clonic seizures, in patients aged ≥4 years, and as adjunctive treatment of generalized tonic-clonic seizures in patients aged ≥7 years with idiopathic generalized epilepsy [6]
Mechanism of action	Selective, noncompetitive α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor antagonist [5,7]
Route of administration	Oral [5,6]
Chemical structure	[5]
Pivotal trials	Three randomized, double-blind, placebo-controlled Phase III trials in patients with treatment-resistant focal-onset seizures (Studies 304 [NCT00699972], 305 [NCT00699582], and 306 [NCT00700310]) and one randomized, double-blind, placebo-controlled Phase III trial in patients with treatment-resistant generalized tonic-clonic seizures (Study 332 [NCT01393743]) [8–11]

Declaration of interests

J Wheless has received grants support from Aquestive, Eisai, Greenwich, INSYS Inc., LivaNova, Mallinckrodt, Neurelis, NeuroPace, the Shainberg Foundation, and West; has served as a consultant for Aquestive, BioMarin, Eisai, Greenwich, Mallinckrodt, Neurelis; NeuroPace, Shire, Supernus, and Zogenix; and has received speaker bureau honoraria from BioMarin, Eisai, Greenwich, LivaNova, Mallinckrodt, and Supernus. N Chourasia have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Scientific accuracy review

Eisai provided a scientific accuracy review at the request of the journal editor.

Author contribution statement

All authors were involved in the decision to submit this article for publication, reviewed the manuscript, and approved the final version for submission.

Additional information

Funding

Eisai Inc. provided funding for medical writing support, under the direction of the authors, was provided by Rebekah Waters, PhD, on behalf of CMC AFFINITY, a division of IPG Health Medical Communications, funded by Eisai Inc., in accordance with Good Publication Practice (GPP3) guidelines.