Effectiveness and Safety of Perampanel in Patients with Lennox-Gastaut Syndrome: A Single-arm Meta-analysis with Meta-regression and Sensitivity Analyses
Amanda Rabelo1, Yasmim Nunes2, Julia Oliveira3, Carolina Corrêa4, Gabriela Carolino5, Daniel de Castro1
1Faculty of Medicine of Bauru - USP, 2AFYA Faculty of Medical Sciences of Palmas, 3Pontifical Catholic University of Minas Gerais, 4Faculty of Medical Sciences of Minas Gerais - FCMMG, 5Faculty of Medical and Health Sciences FCMS/JF
Objective:
To evaluate the efficacy and safety of perampanel in patients with LGS.
Background:
Lennox-Gastaut syndrome (LGS) is characterized by multiple types of drug-resistant seizures that begin before 18 years of age, including at least one tonic seizure, and are commonly associated with cognitive and behavioral impairments. Perampanel, a non-competitive antagonist of AMPA-type glutamate receptors, has emerged as a potential therapeutic strategy to disrupt the underlying excitatory mechanisms of this syndrome.
Design/Methods:
Inclusion was limited to randomized controlled trials (RCTs) or observational studies involving patients with confirmed LGS who received perampanel. A systematic review and meta-analysis were conducted in accordance with PRISMA and Cochrane guidelines. PubMed, Embase, and Cochrane Central databases were searched up to July 2025. Pooled analyses were performed using a single-proportion meta-analysis with 95% confidence intervals (CI). Statistical analysis was conducted using OpenMeta Analyst Version 3.1 (PROSPERO CRD420251078418).
Results:
14 studies comprising 330 patients were included. The pooled ≥50% responder rate was 46.4% (95% CI 33.6 to 59.2; p<0.001), ≥75% responder rate 23.3% (7.3 to 39.4; p=0.002), seizure freedom 8.2% (1.6 to 14.8; p=0.008), and seizure aggravation 6.0% (0.5 to 11.5; p=0.044). Reported adverse events included somnolence (13.7%), irritability (11.3%), aggression (10.7%), dizziness (5.3%), agitation (3.9%), and psychiatric symptoms (17.1%). Heterogeneity was substantial; sensitivity analysis was robust; no temporal effect on response.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.