Safety, efficacy and outcome-related factors of perampanel over 12 months in a real-world setting: The FYDATA study

Real-world data of current antiepileptic drugs (AEDs) used to treat focal seizures is of importance to understand the efficacy and safety outside of the clinical trial setting. Here we report real-world data from a large series of patients treated with perampanel for 1year.FYDATA was a multicentre, retrospective, 1-year observational study assessing the efficacy and safety of adjuvant perampanel in patients ≥12 years of age with focal epilepsy in a real-world setting. At 12 months, the proportion of patients who were seizure free, median percentage seizure reduction, proportion of responders, retention rate and proportion of patients with adverse events (AEs) were assessed. Analyses were also performed to identify any patient-, medication- and disease-related factors associated with a large clinical response or carry a risk for AEs.A total of 464 patients were included in the study with a retention rate of 60.6% at 1year. The mean number of prior AEDs was 7.8. The median percentage reduction in overall seizures was 33.3% (75% for secondary generalised seizures) after 1year, with 7.2% of patients achieving seizure freedom. Furthermore, patients on non-enzyme-inducing AEDs were more likely to achieve seizure freedom, and logistic regression revealed that patients aged ≥65 years, those with epilepsy due to a vascular aetiology and those who had received fewer prior AEDs showed a better clinical response to perampanel. A total of 62.9% of the patients experienced AEs at 12 months; dizziness, somnolence and irritability were the most frequent AEs. Patients with prior psychiatric comorbidities (hyperactivity and personality disorder) were more likely to experience psychiatric AEs with perampanel, and slower titration schedules were associated with less AEs overall.Perampanel, for the treatment of focal epilepsy in a real-world setting in a refractory population, over 1year, demonstrates a similar efficacy and safety profile to that observed in clinical trials. Our results have implications for the optimisation of perampanel use in a clinical setting.