Mechanism of Action
Perampanel is a first-in class selective, non-competitive antagonist of the ionotropic α-amino-3-hydroxy-5- methyl-4-isoxazolepropionic acid (AMPA) glutamate receptor on post-synaptic neurons. Glutamate is the primary excitatory neurotransmitter in the central nervous system and is implicated in a number of neurological disorders caused by neuronal overexcitation. Activation of AMPA receptors by glutamate is thought to be responsible for most fast excitatory synaptic transmission in the brain. In in vitro studies, perampanel did not compete with AMPA for binding to the AMPA receptor, but perampanel binding was displaced by non-competitive AMPA receptor antagonists, indicating that perampanel is a non-competitive AMPA receptor antagonist. In vitro, perampanel inhibited AMPA-induced (but not NMDA-induced) increase in intracellular calcium. In vivo, perampanel significantly prolonged seizure latency in an AMPA-induced seizure model.
The precise mechanism by which perampanel exerts its anti-epileptic effects in humans remains to be fully elucidated.
Pharmacodynamic effects
A pharmacokinetic-pharmacodynamic (efficacy) analysis was performed based on the pooled data from the 3 efficacy trials for partial-onset seizures. In addition, a pharmacokinetic pharmacodynamic (efficacy) analysis was performed in one efficacy trial for primary generalized tonic-clonic seizures. In both analyses, perampanel exposure is correlated with reduction in seizure frequency.
Psychomotor performance. Single and multiple doses of 8 mg and 12 mg impaired psychomotor performance in healthy volunteers in a dose-related manner. The effects of perampanel on complex tasks such as driving ability were additive or supra-additive to the impairment effects of alcohol. Psychomotor performance testing returned to baseline within 2 weeks of cessation of perampanel dosing.
Cognitive function. In a healthy volunteer study to assess the effects of perampanel on alertness and memory using a standard battery of assessments, no effects of perampanel were found following single and multiple doses of perampanel up to 12 mg/day.
In a placebo-controlled study conducted in adolescent patients, no significant changes in cognition relative to placebo as measured by Cognitive Drug Research (CDR) System Global Cognition Score were observed for perampanel. In the open label extension, no significant changes were observed in global CDR system score after 52 weeks of perampanel treatment
In an open-label uncontrolled study conducted in paediatric patients, no clinically important changes in cognition relative to baseline as measured by ABNAS were observed following adjunctive perampanel therapy.
Alertness and Mood. Levels of alertness (arousal) decreased in a dose-related manner in healthy subjects dosed with perampanel from 4 to 12 mg/day. Mood deteriorated following dosing of 12 mg/day only; the changes in mood were small and reflected a general lowering of alertness. Multiple dosing of perampanel 12 mg/day also enhanced the effects of alcohol on vigilance and alertness, and increased levels of anger, confusion and depression as assessed using the Profile of Mood State 5-point rating scale.
Cardiac Electrophysiology. Perampanel did not prolong the QTc interval when administered in daily doses up to 12 mg/day, and did not have a dose-related or clinically important effect on QRS duration.
Clinical efficacy and safety
Partial-Onset Seizures
The efficacy of perampanel in partial-onset seizures was established in three adjunctive therapy 19-week, randomized, double-blind, placebo-controlled, multicenter trials in adult and adolescent patients. Patients had partial-onset seizures with or without secondary generalization and were not adequately controlled with one to three concomitant AEDs. During a 6-week baseline period, patients were required to have more than five seizures with no seizure-free period exceeding 25 days. In these three trials, patients had a mean duration of epilepsy of approximately 21.06 years. Between 85.3% and 89.1% of patients were taking two to three concomitant AEDs with or without concurrent vagal nerve stimulation.
Two studies (studies 304 and 305) compared doses of Perampanel 8 and 12 mg/day with placebo and the third study (study 306) compared doses of Perampanel 2, 4 and 8 mg/day with placebo. In all three trials, following a 6-week Baseline Phase to establish baseline seizure frequency prior to randomization, patients were randomized and titrated to the randomized dose. During the Titration Phase in all three trials, treatment was initiated at 2 mg/day and increased in weekly increments of 2 mg/day to the target dose.
Patients experiencing intolerable adverse events could remain on the same dose or have their dose reduced to the previously tolerated dose. In all three trials, the Titration Phase was followed by a Maintenance Phase that lasted 13 weeks, during which patients were to remain on a stable dose of Perampanel.
The pooled 50% responder rates were placebo 19%, 4 mg 29%, 8 mg 35% and 12 mg 35%. A statistically significant effect on the reduction in 28-day seizure frequency (Baseline to Treatment Phase) as compared to the placebo group was observed with Perampanel treatment at doses of 4 mg/day (Study 306), 8 mg/day (Studies 304, 305 and 306), and 12 mg/day (Studies 304 and 305). The 50% responder rates in the 4 mg, 8 mg and 12 mg groups were respectively 23.0%, 31.5%, and 30.0% in combination with enzyme-inducing anti-epileptic medicinal products and were 33.3%, 46.5% and 50.0% when perampanel was given in combination with non-enzyme inducing anti-epileptic medicinal products. These studies show that once-daily administration of perampanel at doses of 4 mg to 12 mg was significantly more efficacious than placebo as adjunctive treatment in this population.
Data from placebo-controlled studies demonstrate that improvement in seizure control is observed with a once-daily Perampanel dose of 4 mg and this benefit is enhanced as the dose is increased to 8 mg/day. No efficacy benefit was observed at the dose of 12 mg as compared to the dose of 8 mg in the overall population. Benefit at the dose of 12 mg was observed in some patients who tolerate the dose of 8 mg and when the clinical response to that dose was insufficient. A clinically meaningful reduction in seizure frequency relative to placebo was achieved as early as the second week of dosing when patients reached a daily dose of 4 mg. 1.7 to 5.8% of the patients on perampanel in the clinical studies became seizure-free during the 3 month maintenance period compared with 0% -1.0% on placebo.
Open-label Extension Study
Ninety-seven percent of the patients who completed the randomized trials in patients with partial-onset seizures were enrolled in the open label extension study (n=1,186). Patients from the randomized trial were converted to perampanel over 16 weeks followed by a long-term maintenance period (≥1 year). The mean average daily dose was 10.05 mg.
Primary Generalised Tonic-Clonic Seizures
Perampanel as adjunctive therapy in patients 12 years of age and older with idiopathic generalized epilepsy experiencing primary generalized tonic-clonic seizures was established in a multicenter, randomised, double-blind, placebo-controlled study (Study 332). Eligible patients on a stable dose of 1 to 3 AEDs experiencing at least 3 primary generalized tonic-clonic seizures during the 8-week baseline period were randomised to either Perampanel or placebo. The population included 164 patients (Perampanel n=82, placebo n=82). Patients were titrated over four weeks to a target dose of 8 mg per day or the highest tolerated dose and treated for an additional 13 weeks on the last dose level achieved at the end of the titration period. The total treatment period was 17 weeks. Study drug was given once per day. The 50% primary generalized tonic-clonic seizures responder rate during the Maintenance Period was significantly higher in the perampanel group (58.0%) than in the placebo group (35.8%), P=0.0059. The 50% responder rate was 22.2% in combination with enzyme inducing anti-epileptic medicinal products and was 69.4% when perampanel was given in combination with non-enzyme-inducing anti-epileptic medicinal products. The number of perampanel subjects taking enzyme inducing anti-epileptic medicinal products was small (n = 9). The median percent change in primary generalized tonic-clonic seizure frequency per 28 days during the Titration and Maintenance Periods (combined) relative to Pre-randomisation was greater with perampanel (-76.5%) than with placebo (-38.4%), P<0.0001. During the 3 months maintenance period, 30.9% (25/81) of the patients on perampanel in the clinical studies became free of PGTC seizures compared with 12.3% (10/81) on placebo.
Other subtypes of idiopathic generalised seizure
The efficacy and safety of perampanel in patients with myoclonic seizures have not been established. The available data are insufficient to reach any conclusions. The efficacy of perampanel in the treatment of absence seizures has not been demonstrated. In Study 332, in patients with PGTC seizures who also had concomitant myoclonic seizures, freedom from seizures was achieved in 16.7 % (4/24) on perampanel compared to 13.0 % (3/23) in those on placebo. In patients with concomitant absence seizures, freedom from seizures was achieved in 22.2% (6/27) on perampanel compared to 12.1% (4/33) on placebo. Freedom from all seizures was achieved in 23.5% (19/81) of patients on perampanel compared to 4.9% (4/81) of patients on placebo.
Open-label extension phase
Of the 140 patients who completed the Study 332 114 patients (81.4%) had entered the Extension phase. Patients from the randomised trial were converted to perampanel over 6 weeks followed by a long-term maintenance period (≥ 1 year). In the Extension Phase, 73.7% (84/114) of patients have a modal daily perampanel dose of greater than 4 to 8 mg/day and 16.7% (19/114) had a modal daily dose of greater than 8 to 12 mg/day. A decrease in PGTC seizure frequency of at least 50% was seen in 65.9% (29/44) of patients after 1 year of treatment during the Extension Phase (relative to their pre-perampanel baseline seizure frequency). These data were consistent with those for percent change in seizure frequency and showed that the PGTC 50% responder rate was generally stable across time from about week 26 through the end of year 2. Similar results were seen when all seizures and absence vs. myoclonic seizures were evaluated over time.
Conversion to monotherapy
In a retrospective study of clinical practice, 51 patients with epilepsy who received perampanel as adjunctive treatment converted to perampanel monotherapy. The majority of these patients had a history of partial-onset seizures. Of these, 14 patients (27%) reverted to adjunctive therapy in the following months. Thirty four (34) patients were followed up for at least 6 months and, of these, 24 patients (71%) remained on perampanel monotherapy for at least 6 months. Ten (10) patients were followed up for at least 18 months and, of these, 3 patients (30%) remained on perampanel monotherapy for at least 18 months.
Pediatric population
The European Medicines Agency has deferred the obligation to submit the results of studies with perampanel in one or more subsets of the pediatric population in treatment-resistant epilepsies (localization-related and age-related epilepsy syndromes) (see section 4.2 for information on adolescent and paediatric use).
The three pivotal double-blind placebo-controlled phase 3 studies included 143 adolescents between the ages of 12 and 18. The results in these adolescents were similar to those seen in the adult population.
Study 332 included 22 adolescents between the ages of 12 and 18. The results in these adolescents were similar to those seen in the adult population.
A 19 week, randomised, double-blind, placebo-controlled study with an open-label extension phase (Study 235) was performed to assess the short-term effects on cognition of Perampanel (target dose range of 8 to 12 mg once daily) as adjunctive therapy in 133 (Perampanel n = 85, placebo n = 48) adolescent patients, aged 12 to less than 18 years old, with inadequately controlled partial-onset seizures. Cognitive function was assessed by the Cognitive Drug Research (CDR) System Global Cognition t-Score, which is a composite score derived from 5 domains testing Power of Attention, Continuity of Attention, Quality of Episodic Secondary Memory, Quality of Working Memory, and Speed of Memory. The mean change (SD) from baseline to end of double-blind treatment (19 weeks) in CDR System Global Cognition t-Score was 1.1 (7.14) in the placebo group and (minus) -1.0 (8.86) in the perampanel group, with the difference between the treatment groups in LS means (95% CI) = (minus) -2.2 (-5.2, 0.8). There was no statistically significant difference between the treatment groups (p = 0.145). CDR System Global Cognition t-Score for placebo and perampanel were 41.2 (10.7) and 40.8 (13.0), respectively at the baseline. For patients with perampanel in the open-label extension (n = 112), the mean change (SD) from baseline to end of open-label treatment (52 weeks) in CDR System Global Cognition t-Score was (minus) -1.0 (9.91). This was not statistically significant (p = 0.96). After up to 52 weeks of treatment with perampanel (n = 114), no effect on bone growth was observed. No effects on weight, height and sexual development were seen following up to 104 weeks of treatment (n = 114).
An open-label, uncontrolled study (Study 311) was performed to assess the exposure-efficacy relationship of perampanel as adjunctive therapy in 180 paediatric patients (aged 4 to 11 years old) with inadequately controlled partial-onset seizures or primary generalised tonic-clonic seizures. Patients were titrated over 11 weeks to a target dose of 8 mg/day or the maximum tolerated dose (not to exceed 12 mg/day) for patients not taking concomitant CYP3A-inducing antiepileptic drugs (carbamazepine, oxcarbazepine, eslicarbazepine and phenytoin) or 12 mg/day or the maximum tolerated dose (not to exceed 16 mg/day) for patients taking a concomitant CYP3A-inducing antiepileptic drug. Perampanel dose achieved at the end of titration was maintained for 12 weeks (for a total of 23 weeks of exposure) at the completion of the core study. Patients who entered into Extension Phase were treated for an additional 29 weeks for a total exposure duration of 52 weeks.
In patients with partial-onset seizures (n = 148 patients), the median change in seizure frequency per 28 days, the 50% or greater responder rate, and seizure-free rate following 23 weeks of perampanel treatment were -40.1%, 46.6% (n = 69/148), and 11.5% (n = 17/148), respectively, for total partial-onset seizures. The treatment effects on the median reduction in seizure frequency (Weeks 40-52: n = 108 patients, -69.4%), 50% responder rate (Weeks 40-52: 62.0%, n = 67/108), and seizure-free rate (Weeks 40-52: 13.0%, n = 14/108) were sustained following 52 weeks of perampanel treatment.
In a subset of partial-onset seizure patients with secondarily generalised seizures (n = 54 patients), the corresponding values were -58.7%, 64.8% (n = 35/54), and 18.5% (n = 10/54), respectively, for secondarily generalised tonic-clonic seizures. The treatment effects on the median reduction in seizure frequency (Weeks 40-52: n = 41 patients, -73.8%), 50% responder rate (Weeks 40-52: 80.5%, n = 33/41), and seizure-free rate (Weeks 40-52: 24.4%, n = 10/41) were sustained following 52 weeks of perampanel treatment.
In patients with primary generalised tonic-clonic seizures (n = 22 patients, with 19 patients aged 7-<12 years and 3 patients aged 4-<7 years), the median change in seizure frequency per 28 days, the 50% or greater responder rate, and seizure-free rate were -69.2%, 63.6% (n = 14/22), and 54.5% (n = 12/22), respectively. The treatment effects on the median reduction in seizure frequency (Weeks 40-52: n = 13 patients, -100.0%), 50% responder rate (Weeks 40-52: 61.5%, n = 8/13), and seizure-free rate (Weeks 40-52: 38.5%, n = 5/13) were sustained following 52 weeks of perampanel treatment. These results should be considered cautiously as the number of patients is very small. Similar results were obtained in a subset of patients with primary generalised tonic-clonic seizures of idiopathic generalised epilepsy (IGE) (n = 19 patients, with 17 patients aged 7-<12 years and 2 patients aged 4-<7 years; the corresponding values were -56.5%, 63.2% (n = 12/19), and 52.6% (n = 10/19), respectively. The treatment effects on the median reduction in seizure frequency (Weeks 40-52: n = 11 patients, -100.0%), 50% responder rate (Weeks 40-52: 54.5%, n = 6/11), and seizure-free rate (Weeks 40-52: 36.4%, n = 4/11) were sustained following 52 weeks of perampanel treatment. These results should be considered cautiously as the number of patients is very small.