Abstracts

Rationale: While pivotal trial data are available to demonstrate the efficacy and safety of perampanel, trial populations are often narrowly selected and may not represent real world populations. A systematic review of published observational studies and case reports was performed to understand how perampanel performs in the broader epilepsy population. Methods: Search methods followed Cochrane Collaboration and United Kingdom National Institute for Health and Clinical Excellence guidelines. Main inclusion criteria were English language case reports or observational studies with efficacy or safety information reported. Key exclusion criteria were regulatory trial data and reviews, animal studies, pharmacology or molecular studies. Studies were identified through a search of PubMed and the Centre for Reviews and Dissemination database. Results: Nine observational studies and 10 case reports were identified (see Table 1). Treatment responder rates (50%) were reported for 7 of 9 studies: values ranged from 14.8% (refractory epilepsy) to 50% (general epilepsy population); responder rates for complex partial seizures were 48 to 57%; and, for generalized tonic-clonic seizures (GTCS) 57%. Seizure freedom in reports ranged from 0 (n=47) to 15% (n=281). Discontinuation rates due to adverse events were reported in 4 studies and ranged from 29.7% to 67%. In 10 case reports of 12 cases, 6 of 12 patients with epilepsy linked conditions had successful perampanel treatment reported: 2 patients with Lafora disease had reduced GTCS and reduced myoclonus with perampanel treatment; perampanel 2 mg/day reduced seizures and did not trigger symptoms in a patient with variegate porphyria; 2 patients had improvement in myoclonus associated with stroke (n=1) and hypoxic injury (n=1); 1 patient with glioblastoma multiforme and temozolomide treatment became seizure free with treatment. Six patients had adverse effects of perampanel reported: 1 patient with Tourette's syndrome had behavioral problems, however, seizures were controlled; 3 patients were reported with suicidal ideation, 2 patients had decreased serum concentrations of phenytoin, phenobarbital or rufinamide reported. Conclusions: In our systematic review, "real world" efficacy and safety for perampanel in observational studies are comparable to controlled trial values. Case reports explore treatment in a variety of patients, reporting successful treatment as well as concerns both consistent (suicidal ideation; trials showed no difference from placebo) and inconsistent (serum level changes in other drugs) with extensive regulatory trial data. Funding: Eisai Inc. funded the research that formed the basis for this abstract.