Introduction
Perampanel [2-(2-oxo-1-phenyl-5-pyridin-2-yl-1,2-dihydropyridin-3-yl) benzonitrile] (PER) is a novel antiepileptic drug (AED) with a reported mechanism of action that works as a noncompetitive and selective AMPA (amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid) receptor antagonist. It has been demonstrated to be useful in the treatment of both focal and generalized seizures. Reference French, Krauss and Biton4–Reference French, Krauss and Bibbiani8 PER was approved for use in Canada in April 2013.
Several post marketing reports on the use of PER, mainly from Europe, are available. Reference Steinhoff, Hamer and Trinka1,Reference Shah, Reuber and Goulding2
Dizziness, fatigue, headache, somnolence, and neuropsychiatric side effects, including irritability, anger, emotional liability, suicidal ideation, aggression, psychosis, and depression, all have been reported as potential side effects of this drug in these reports. These adverse side effects led to discontinuation of the medication in 15%–40% of the patients. Reference Steinhoff, Hamer and Trinka1,Reference Shah, Reuber and Goulding2
Most of these reports had a titration rate of 2 mg/1–2 weeks. PER has a long half-life (approximately 105 hours) indicating that the new steady-state serum level following each dose increment will be achieved in 17–21 days. Reference Greenwood and Valdes9 A too rapid upward dose titration could result in a patient experiencing dose-related adverse effects, and withdrawing treatment, caused by a dose higher than required for seizure control.
A 2016 study suggested that a slower titration rate might be associated with lower side effects rates based on a limited number of patients. Reference Villanueva, Garcés and López-González10
In Halifax at the QEII Health Science Centre, we prescribed PER for 124 patients between September 2013 and November 2016. All patients who were started on the drug had a titration rate of 2 mg/day and increased by 2 mg every 4 weeks.
Our hypothesis for this study was whether a slower titration of PER leads to a higher retention rate at 6 months.
Method and Statistical Analysis
All patients attending the adult epilepsy clinic of one epileptologist (RMS) who were prescribed PER with a starting dose of 2 mg titrated by 2 mg every 4 weeks between September 2013 and November 2016 and had at least one follow-up visit at 6 months were identified by review of the patients’ charts.
Clinical data abstracted from the patients’ charts included demographic data (gender and age when PER was started), seizure data (age of onset, seizure types, and frequency of each type before starting the medication and during scheduled follow-up visits) and medications data (previous and concomitant drugs). The highest dose the patient received and any reported adverse effects were documented. No presumed seizure etiology data was collected.
Exclusion criteria were titration of PER faster than 2 mg/4 weeks; medication noncompliance or no follow-up visit.
Patients whose seizure frequency could not be assessed and/or had other therapeutic interventions less than 6 months after PER was started were included for the retention analysis but excluded from the efficacy analysis.
The retrospective chart review did not permit accurate assessment of the frequency of myoclonic jerks or focal aware seizures. These seizure types were excluded from efficacy analysis.
Efficacy of PER treatment was assessed in two ways: (a) the Engel classification and (b) 50% responder rate.
The Engel classification was originally proposed to describe the postoperative outcome of epilepsy surgery and has been extensively used for this purpose (Table 1). Reference Engel12 Although the Engel classification is not the standard measure of efficacy in most nonsurgical epilepsy research papers, this outcome classification provides, at a glance, information with respect to clinically meaningful improvement. The more traditional 50% responder rate provides data attesting to a measurable effect of the study drug on seizure frequency and is useful for regulatory drug trials. However, in the individual patient, a seizure frequency of, for example, 20/month that is reduced to 10/month does not usually equate to any truly clinically meaningful benefit for the patient.
Table 1: Engel classification
Descriptive statistics including mean (± standard deviation), median (interquartile range), counts, and frequency were used, where appropriate, to summarize baseline clinical characteristics. Event rates and percentage change were reported based on documented cases with 95% confidence interval (CI). All analyses were performed using SAS STAT software version 9.4 (SAS, Cary, NC, USA).
This study was approved by Nova Scotia Health Authority Research Ethics Board.
Results
Between September 2013 and November 2016, 124 patients were prescribed PER; 102 patients met the entry criteria and were included for further analysis. Most of the excluded patients were due to either noncompliance, other medication changes or failure to attend follow-up visits after 6 months (Figure 1).
Figure 1: Reasons of exclusion of 22 patients.
The mean age of the patients was 40.25 (SD = 14.51) ranging between 18 and 72 years. The median number of past AED was 6 (range 3–20) (SD = 2.84), and the median number of concomitant AEDs was 2 (range = 1–5, SD = 0.92).
Tables 2 and 3 summarize the baseline demographics and clinical data of the patients.
Table 2: Patients baseline demographics
Table 3: Patients baseline clinical data
Std dev=standard deviation.
Retention Rate Analysis
Retention rate at 6 months was 78.4% (80 patients) (Figure 2). Of those patients who were still taking the drug at 6 months, 50 (62.5%) were still taking PER at their last clinic visit up to 37 months from starting the drug. For the other 30 patients, 18 discontinued after 6 months and 12 patients did not have a follow-up after the 6-month visit.
Figure 2: Retention rate at 6 months based on epilepsy type.
There were 22 patients (21.57%) who developed adverse effects that led to discontinuation before 6 months. These patients were included in the retention analysis but excluded from the efficacy analysis.
The adverse effects are detailed as follows: 15 patients (14.7%) developed new onset or worsened neuropsychiatric symptoms including irritability, anger, emotional liability, and depression, 3 patients (2.9%) developed dizziness, 3 patients (2.9%) developed fatigue or somnolence, and 1 patient (1%) with worsened seizures (Figure 3).
Figure 3: Analysis of the side effects that led to discontinuation.
Efficacy Analysis
All patients who were included for the efficacy analysis at 6 months had a baseline frequency of at least 6 seizures per 6 months (minimum of one seizure per month) but the majority had more (mean seizure frequency per month = 9.25; median = 6). The only exception is one patient who had a lower baseline seizure frequency (3 per year) and this patient did not achieve Engel class I because they had one seizure in the first 6 months of PER treatment.
The efficacy analyses for the Engel outcome are shown in Figures 4 and 5 and the 50% responder rate in Table 4.
Figure 4: Efficacy in GGE at 6 months using Engel classification.
Figure 5: Efficacy in focal epilepsy at 6 months using Engel classification.
Table 4: 50% responder rate at the 6-month visit for all the patients who were still taking the drug and not excluded from the efficacy analysis
The 16 patients and the reason(s) for exclusion from the efficacy analysis are depicted in Figure 6. Note that all these patients were included for the retention rate analysis.
Figure 6: Patients excluded from efficacy analysis but included for retention rate analysis.
Efficacy in genetic generalized epilepsy at 6 Months
Among the 11 patients with genetic generalized epilepsy (GGE), 7 (64%) achieved an Engel class I outcome, 1 patient (9%) with Engel class II, 2 patients (18%) with Engel class III, and 1 patient (9%) with Engel class IV (Figure 5).
Efficacy in Focal Epilepsy at 6 Months
Among 52 patients with focal epilepsy, the efficacy analysis revealed that 16 patients (30.7%) achieved Engel class I, 3 patients (5.8%) with Engel class II, 11 patients (21.2%) with Engel class III, and 22 patients (42.3%) with Engel class IV (Figure 6).
Focal to Bilateral Tonic–Clonic Seizures at 6 Months
Of the 68 patients with focal epilepsy, 18 patients experienced focal to bilateral Tonic–Clonic (TC) seizures prePER with a frequency of at least 1 per 6 months. At the 6-month visit, 12/18 (66.7%) of these patients had their TC seizures abolished, 3 patients had significant reduction of focal to bilateral seizures, and 3 had no or minimal reduction of TC seizures.
The Optimum Dose
All patients started PER at a dose of 2 mg and titrated to the maximum tolerable dose if seizure freedom was not achieved. Only 14 patients reached a maximum dose of 12 mg/day. The optimum dose (defined as the optimum ratio of efficacy to adverse effects) was determined for some patients according to patients’ and neurologist’s overall impression but this was difficult to assess reliably in this retrospective review. Table 5 shows the recorded optimum dose where Table 6 shows the highest dose the patient received.
Table 5: Reported optimum dose of perampanel
Table 6: The highest dose the patient received
The Increasing the Dose after 6 Months
Although 28 patients had their PER dose gradually increased after the 6-month visit to 10–12 mg, only 3 patients had further benefit from increasing the dose without side effects. In total, 25/28 patients had their doses reduced back to 6–8 mg or PER was tapered gradually to discontinuation.
Discussion
This study reviews the experience of PER treatment in 102 patients with refractory epilepsy from a tertiary epilepsy care center with patient management supervised by one epileptologist.
This study provides a “real world” review of the efficacy and the safety of PER employing a slow titration schedule in a highly medically refractory epilepsy population.
The 6-month retention rate was 78%, achieved using a slow PER titration of 2 mg/month to an initial plateau of 6 mg/day. While study methodologic differences do not allow a strict comparison to other reports, the retention rate in the current study compares favorably to reported retention rates of 60%–71% with more rapid titration. Reference Steinhoff, Hamer and Trinka1,Reference Shah, Reuber and Goulding2
In the present study, adverse effects leading to discontinuation occurred in 22 out of 102 patients (21.57%) in less than 6 months. In comparison to other reported clinical experience from the UK and Ireland, the frequency of adverse effects leading to discontinuation was found to be 26.8% (83 out of 310 patients) with titration rate of 2 mg/2 weeks. Reference Shah, Reuber and Goulding2 Other reports from USA and South Korea showed discontinuation rates of 23% and 28.7%, respectively. Reference Dong and Jeeyoung3,Reference Singh, Shah and Luciano11
Similar to most reports, the most frequent adverse effect leading to discontinuation of PER in this study was found to be neuropsychiatric side effects including irritability, anger, emotional lability, or depression. Reference Shah, Reuber and Goulding2,Reference Datta, Xu, Sachedina, Boelman, Huh and Connolly13
The efficacy of PER was perhaps higher than expected in an epilepsy population with a median past AED exposure of six. Many reports have demonstrated the extremely low probability of seizure control after a failure of three different regimens. Reference Kwan and Brodie14,Reference Chen, Brodie, Liew and Kwan15
In this study, patients with generalized epilepsy demonstrated a trend toward a better response (64% Engel class I) than those with focal epilepsy (30.7% Engel class I).
The efficacy analysis was not remarkably different compared to a Spanish study that showed significant reduction from the baseline seizure frequency by median change of 33%. In that study, the effect was greater for secondarily generalized seizures (focal to bilateral TC), which had seizure reduction by 75%. This study suggested that PER has better efficacy outcome in the secondary generalized seizure group. Reference Villanueva, Garcés and López-González10 In our study, 83.3% of the patients with focal to bilateral TC seizures had significant reduction of those seizures.
The dose of 6 mg daily was found to be the optimum dose for the majority of patients (32.47%) followed by 8 mg (14.3%). In total, 14 patients reached the maximum dose (12 mg once daily) but only 3 found that this was their optimum dose (Note: data was not collected on the potential interaction of PER dose with co-treatment with enzyme inducing or inhibiting AEDs). This observation suggests that doses higher than 8 mg are associated with a higher risk of adverse effects without meaningful improvement in seizure control as only 3 of 28 patients (10.7%) who had a dose increase up to 10–12 mg/day had demonstrable benefit.
Some patients had an optimum response at low PER doses (two patients taking 2 mg/day; four patients on 4 mg/day). This low dose response is another potential advantage of a slower titration as patients may notice a beneficial effect of a low dose prior to reaching the planned initial dose plateau (e.g. 6 mg/day).
This is the first observational adult study of PER in a clinical setting in Canada. One retrospective Canadian study of PER tolerability and adverse events in a pediatric and adolescent population has been published. Reference Datta, Xu, Sachedina, Boelman, Huh and Connolly13
Limitations of this study include that it is a retrospective study that lacks a comparative group using a faster titration that would allow comparison for the retention rate and efficacy. No attempt was made to determine response rates as a function of epilepsy etiology because the number of patients in each etiologic category would likely have been too small to allow any reliable conclusion.
Another limitation is not collecting data regarding the intellectual state of the patients (more specifically intellectual disability).
Conclusion
Prescribing PER to patients with refractory epilepsy, whether generalized or focal, using a slow titration schedule appears to be more tolerable and leads to a higher retention rate compared to faster titration.
A prospective randomized study with parallel “fast” and “slow” titration schedules would give a definite answer to this hypothesis.
Acknowledgements
We would like to acknowledge Ms. Susan Rahey for her help in data collection and editorial assistance in preparation of this paper.
Conflict of Interest
Dr. Sadler has received speakers’ honoraria from Eisai Canada. Dr. Basheikh has no conflicts of interest to declare.
Statement of Authorship
MB: Conceptualization, methodology, investigation, writing – original draft. RMS: Conceptualization, methodology, resources, writing – review & editing, supervision.
