Long-term efficacy and safety of adjunctive perampanel in pediatric patients aged 4–19 years with epilepsy: a real-world study

This study determined the 24-month outcomes of perampanel treatment in children and adolescents with epilepsy. The percentage of ≥ 50% responders was 47.3% (139/294) at 12 months and 49.0% (144/294) at 24 months. A 100% reduction in seizures for more than 12 months was observed in 12.2% (36/294). Discontinuation occurred in 39.8% (117/294). The most common reason for discontinuation was adverse events (29.1%, 34/117). Baseline seizure frequency was higher in children aged < 12 years than in patients aged ≥ 12 years; however, the percentage of seizure reduction and ≥ 50% responders did not significantly differ between the two groups. The rate of early discontinuation was higher (p < 0.001) and the duration of perampanel treatment was shorter in children aged < 12 years (p = 0.001). Most children aged < 12 years discontinued PER due to inadequate effectiveness, while adverse event was the most common reason in patients aged ≥ 12 years (p = 0.045). Only slow titration was significantly associated with ≥ 50% of responders. In conclusion, this study showed that perampanel can be utilized effectively and safely for a prolonged period in pediatric patients aged 4 to < 12 years, as well as in patients aged 12 years and older.

Statistical analyses.Seizure outcomes were analyzed using both observed and imputed data.For imputed data, the last observation carried forward (LOCF) analysis and intention to treat (ITT) analysis were performed.Wilcoxon rank-sum test was used to compare continuous variables, and chi-squared test or Fisher's exact test was used to compare categorical variables.Our modeling logit link was used to deal with longitudinal binary outcome measurements.Logistic regression was performed to identify the factors associated with ≥ 50% seizure reduction at 24 months.Furthermore, univariate and multivariate analyses were used to find an association with good seizure outcomes.Kaplan Meier plot and log-rank test were performed to evaluate the risk of discontinuation between age subgroups.Data are expressed as a median (interquartile range) or number (%), as appropriate.A p-value < 0.05 was considered statistically significant.All statistical analyses were performed using the SAS software, version 9.4 (SAS Institute Inc., Cary, NC, USA.).
Ethical approval.This study was supported by the Severance Hospital Yonsei University and approved by the Institutional Review Board of Severance Children's Hospital (IRB No. 4-2016-0684).The need for informed consent was waived by the Severance Hospital Yonsei University and approved by the Institutional Review Board of Severance Children's Hospital due to the study being retrospective in nature.
All methods were carried out in accordance with relevant guidelines and regulations.
Efficacy outcomes related to PER therapy.The percentage of ≥ 50% responders increased from 41.7% at baseline to 73% at 24 months.The percentage of ≥ 90% responders was 34.0% at 24 months.A 100% reduction in seizures was observed in 36 patients (12.2%) at 24 months (Fig. 1).Among 132 patients who had epileptic surgery before PER initiation, nine patients achieved a 100% reduction in seizures with PER.
When the LOCF imputation method was used, the percentage of ≥ 50% responders was 47.3% at 12 months and 49.0% at 24 months.The ≥ 90% responder rates slightly reduced from 20.1% at 6 months to 16.3% at 24 months (Fig. 1).
Overall, the median percentage of seizure frequency reduced by 40.9% at 24 months from the baseline when the LOCF imputation method was used.Seizure frequency reduced significantly at 24 months compared to the baseline (p < 0.001).
Comparison between age groups.We compared children aged < 12 years with patients aged ≥ 12 years (Table 2).The percentage of seizure reduction and ≥ 50% responder rates did not significantly differ between the two groups during the follow-up period.However, the rate of early discontinuation was higher (p < 0.001), and the duration of PER treatment was shorter in children aged < 12 years (p = 0.001) (Fig. 2).
The most common reason for discontinuation was also different.Most children aged < 12 years discontinued PER due to inadequate effectiveness, while adverse event was the most common reason for PER discontinuation in patients aged ≥ 12 years (p = 0.045) (Table 2).
Adverse events and safety.One hundred twenty-six patients (42.9%) experienced adverse events.The most common adverse event was drowsiness (34 patients, 11.6%), followed by aggressive behavior (33 patients, 11.3%) and gait disturbance (30 patients, 10.2%).The most common adverse events leading to discontinuation were aggressive behavior and lethargy (Fig. 3).Other cases with adverse events were temporary or had minimal effects on daily activities.Most of these adverse events were resolved spontaneously or after reducing the dose or slowing titration.

Factors associated with ≥ 50% reduction in seizures.
We investigated the factors associated with ≥ 50% seizure reduction at 24 months.Higher maximum and maintenance dose, slow titration, and etiology were factors associated with ≥ 50% of responders in univariate analysis.However, in multivariate analysis, only slow titration was significantly associated with ≥ 50% seizure reduction.Patients with good responders used the PER treatment longer (p < 0.001).(Table 3).

Discussion
The anti-seizure effect of PER was sustained during 24-month treatment in children and adolescents with drugresistant epilepsy.Seizure reduction rates of 100% and 50% responder rates were comparable to those observed in previous short-term studies 4,5 .Remarkably, fewer adverse events and higher percentage of discontinuation due to ineffectiveness were observed in pediatric patients compared to adolescents, showing that the optimal pattern of PER use may be different among these patients.
In our ITT analysis, the 50% responder rate was 35% after 24 months of PER treatment, and a 100% reduction in seizure was achieved in 10%.In the LOCF analysis, the 50% responder and 100% reduction rates were higher and reached 49% and 12%, respectively.Observed data analysis showed that the highest 50% responder and 100% reduction rates at 24 months were 73% and 21%, respectively.These numbers were comparable to previous studies on pediatric patients; in a study which included pediatric patients aged 4 to < 12 years, 50% responder rates were 47% and 65% for focal seizures and focal bilateral tonic-clonic seizures, respectively 4 .Another study reported 50% responder rates of 44% after 12-month treatment of PER in pediatric patients aged 4 years or older 5 .The 100% reduction rates ranged between 18 and 36% in these previous studies.The reported rates were higher than those in our data, as these studies were not performed in an ITT manner [4][5][6][7] ; however, similar sustainability was observed in our study.A long-term study that included adolescents reported that the overall anti-seizure effect of PER was maintained for 48 months 8,9 .
When children were compared with adolescents, the outcomes of seizure reduction were not different between the two groups in our study.Previous studies have reported similar responses to PER between adults and adolescents 8,10 .Also, high responder rates similar to those of adults have been reported in children 4,7     significant difference in efficacy has been observed in patients of different ages 5,7 .Our study extends these findings, confirming that the efficacy of PER is sustained for 24 months regardless of the patient's age group.
In our study, adverse events occurred in 43% of the patients.Previous short-term studies reported similar rates ranging between 16 and 40% in children and adolescents 5,7,11 .Common treatment-emergent adverse events (TEAEs) were drowsiness, aggressive behavior, and gait disturbance, and no unusual adverse events were observed in the study, different from adult studies.Previous studies reported higher TEAEs ranging between 74 and 100%, probably due to the study's prospective nature 4,6,12,13 ; however, most of the TEAEs were temporary and resolved after reducing doses or titration speed 5,6 .The overall incidence of adverse events leading to discontinuation was low 4 .These findings show that PER can be used safely in children and adolescents over a long-term period.
Notably, the group aged aged 4 to < 12 years had fewer adverse events compared to the group aged ≥ 12 years.This finding did not correlate well with the previous ones.One study reported that TEAEs occurred more frequently in patients aged 4 to < 7 years than in patients aged 7 to < 12 years 4 .Another study reported a similar incidence of TEAEs between patients aged 2 to < 7 years and 7 to < 12 years 6 .This discrepancy may have occurred due to the different patient populations and the different sizes of the studies.Our study included patients with severe drug-resistant epilepsy.Indeed, in the group aged 4 to < 12 years, 60% of the patients had Lennox-Gastaut syndrome.These patients with developmental epileptic encephalopathies may not have reported certain adverse events, such as dizziness or aggression.Generally, PER efficacy is not influenced by age, changes in body weight, or liver function 4 .The pharmacokinetic properties of PER in children aged ≤ 2 to < 12 years are independent of age and weight of age.More studies are warranted to confirm these findings.The overall discontinuation rate was 40%.The most common reason for discontinuation was adverse events (36.2%), followed by inadequate effect (33.0%) and aggravation (30.9%).
The group aged 4 to < 12 years had higher discontinuation rates than the group aged ≥ 12 years, and ineffectiveness was the major reason for discontinuation in the group aged 4 to < 12 years.Again, this difference could have been attributed to the high percentage of patients with Lennox-Gastaut syndrome in the group aged 4 to < 12 years.It would be more challenging to treat these patients with severe drug-resistant epilepsy than to treat simple focal onset seizures in adolescents.In an open-label study, the most common primary reason for discontinuation was adverse events in children aged 2 to < 12 years, although adverse events led to discontinuation in a few patients (8%) 4 .We may have reduced such adverse events with slow titration.Slow titration of PER at a slow interval of 2 weeks or longer can reduce adverse events and discontinuation rate 14 .
This study had some limitations.There was selection bias.The study included patients with severe drugresistant epilepsy and several comorbidities; this could have underestimated the anti-seizure effect of PER.Additionally, this was a retrospective study.However, it should also be highlighted that this was a real-world study performed in typical clinical situations.
In conclusion, this study shows that PER can be utilized effectively and safely for a prolonged period in pediatric patients aged 4 to < 12 years, as well as in patients aged 12 years and older.

Figure 2 .
Figure 2. Kaplan-Meier curve for time to early treatment discontinuation.The rate of early discontinuation was higher (p < 0.001) and the duration of perampanel treatment was shorter in children aged <12 years (p = 0.001).Solid line indicates children aged <12 years (age subgroup 1) and dotted line indicates patients aged ≥12 years (age subgroup 2).

Table 2 .
Difference between children and adolescents.Data are expressed as number (percent) or median (interquartile range).ASMs, anti-seizure medications; DEE (Developmental and Epileptic Encephalopathy).