Optimal perampanel dose as the treatment for refractory partial-onset seizures: Network meta-analysis based on...

Intractable epilepsy with partial-onset seizures (POS), affecting more than 20% of individuals with epilepsy, is one of the most common neurological conditions worldwide. Perampanel (PER), an antiepileptic drug (AED), is widely used in the treatment of refractory POS. However, the optimal dose of PER remains undetermined. This network meta-analysis (NMA) aims to determine the optimal dose of PER by comparing and ranking the outcomes acquired from eligible randomized, double-blind, placebo-controlled trials, quantitatively analyzing the data.

Electronic databases such as Embase, MEDLINE, CENTRAL, and ClinicalTrials.gov were searched from database inception through May 30, 2024. This NMA includes and synthesizes data from four eligible randomized controlled trials and compares outcomes from four fixed doses of PER (i.e. 12 mg, 8 mg, 4 mg, and 2 mg) along with placebo (PBO). The outcomes assessed include the 50% responder rate, seizure-free rate, dropout rate due to treatment-emergent adverse events (TEAEs), retention rate, and the incidence of TEAEs and serious TEAEs (sTEAEs), using a modified intention-to-treat (MITT) approach. Two reviewers independently screened studies, extracted data from eligible studies and assessed the risk of bias (ROB) using the Cochrane ROB tool 2.0. Pairwise meta-analysis (PMA) and NMA were conducted using a random-effects model with ADDIS v1.16.8 software. The odds ratio (OR) and 95% confidence interval (CI) were calculated for PMA, and OR and 95% credible intervals (CrI) for NMA for dichotomous outcomes. Statistical heterogeneity and sensitivity analyses were performed, and inconsistencies were assessed using the node-split and inconsistency models. We evaluated the quality of evidence and strength of recommendations for clinical practice using the GRADE approach. This prospective study protocol is registered with PROSPERO (registration number: CRD42021247514).

A total of 281 citations were initially identified from the databases and manual searches, with 4 studies involving 2187 participants ultimately included in the systematic review and NMA. All included trials were assessed as having a low ROB according to the revised Cochrane ROB tool 2.0. The ORs and CrIs for the 50% responder rate showed that PER 12 mg compared to PER 4 mg, PER 2 mg, and PBO ranged from 1.96 (1.12 to 3.21) to 2.78 (1.74 to 4.08). For PER 8 mg, the ORs and CrIs compared to PER 4 mg, PER 2 mg, and PBO ranged from 1.66 (1.01 to 2.67) to 2.33 (1.57 to 3.39). For the seizure-free rate, the ORs and CrIs of PER 4 mg, PER 8 mg, and PER 12 mg compared to PBO ranged from 4.21 (1.18 to 15.35) to 5.90 (1.88 to 22.06), respectively. For TEAEs, the ORs and CrIs for PER 12 mg compared to PER 4 mg, PER 2 mg, and PBO ranged from 2.26 (1.30 to 4.00) to 3.07 (1.86 to 4.82). No significant differences were found in the incidence of sTEAEs across all intervention groups. The ORs and CrIs for the dropout rate showed that PER 12 mg compared to PER 8 mg, PBO, and PER 4 mg ranged from 2.16 (1.22 to 4.06) to 5.00 (2.26 to 14.19), and for PER 8 mg compared to PBO and PER 4 mg, the ORs and CrIs ranged from 2.31 (1.24 to 4.58) to 2.38 (1.06 to 6.22). For the retention rate, the ORs and CrIs of PER 12 mg compared to PER 4 mg, PBO, and PER 8 mg ranged from 0.43 (0.24 to 0.69) to 0.65 (0.43 to 0.97). The overall quality of evidence was deemed high. A strong recommendation is made to prioritize PER 8 mg as the first-line treatment, with PER 12 mg as a secondary option for refractory epilepsy with POS.

PER is safe and effective for the treatment of intractable epilepsy with POS. This NMA, which provides high-quality evidence, strongly recommends PER 8 mg as the first-line treatment option, with PER 12 mg as the second choice when necessary. These recommendations are likely applicable to the majority of patients with intractable epilepsy with POS; however, they may not apply to all patients, particularly those with unique clinical characteristics or in specific clinical circumstances.