Summarize cases of perampanel use in established SE, refractory SE, and super-refractory SE (SRSE).

Novel treatments are needed to control treatment-resistant SE. Perampanel is not currently licensed for SE; as there are limited clinical data assessing the effectiveness of perampanel in SE treatment, case reports are valuable for guiding treatment strategies.

Medical records were retrospectively reviewed for perampanel administration in patients with SE treated at five European hospitals between 2011-2015. Cases were classified according to the 2015 International League Against Epilepsy definition of SE. Clinical characteristics, treatment parameters, treatment outcomes, and safety outcomes were included in the data collection.

Overall, 3.9% (n=52/1319) of patients experiencing SE received perampanel. Median latency from SE onset to perampanel initiation was 10 days. Prior to perampanel, benzodiazepines had failed in all patients receiving them (n=51/51); patients had failed a median of five other antiepileptic drugs (AEDs). Median initial perampanel dose was 6 mg/day; median maximum perampanel dose was 10 mg/day. Perampanel was the last drug added in 32/52 (61.5%) patients, with response attributed to perampanel in 19/52 (36.5%) patients. A greater proportion of perampanel non-responders had SRSE (51.5% [n=17/33]) vs perampanel responders (31.6% [n=6/19]) and failed a higher mean number of AEDs before initiating perampanel (5.9 vs 5.1, respectively). Most commonly reported adverse effects (AEs) during perampanel treatment were dizziness (n=1 [1.9%]) and somnolence (n=1 [1.9%]). No serious AEs or AEs leading to perampanel discontinuation were documented.

Across the SE cases reported here, perampanel was administered at greater initial doses than what is routinely administered to patients with epilepsy. The rate of seizure cessation attributed to perampanel treatment (36.5%) in this refractory and heterogeneous population represents a notable response. These data should be confirmed in a larger patient population.

Funding: Medical writing support, under the direction of the authors, was funded by Eisai Inc.