Cenobamate May Offer Benefits in Rare Childhood Epilepsy

TOPLINEAmong children with developmental and epileptic encephalopathies (DEEs), cenobamate was associated with sustained seizure control, high response and retention rates, and an acceptable tolerability profile, with efficacy comparable between Lennox-Gastaut syndrome (LGS) and other DEEs; the use of concomitant sodium channel blockers was independently associated with a higher risk for adverse events.METHODOLOGYResearchers conducted a retrospective multicentre cohort study across 17 Spanish hospitals to evaluate the effectiveness, treatment retention, and safety profile of cenobamate in children with DEEs.The study included 152 children with a clinical diagnosis of DEE (median age, 12.0 years; 59.6% boys) who received "off-label cenobamate" and had at least 3 months of posttreatment follow-up; of these, 27.6% had LGS, 58.6% had unspecified DEE, and 14% had other syndromes including DEE with spike-wave activation in sleep, Dravet syndrome, early infantile DEE, epilepsy with migrating focal seizures, and infantile spasms.Cenobamate was initiated at a median dose of 12.5 mg/d (0.33 mg/kg/d), with mean maintenance doses rising to 1.80, 2.78, and 3.85 mg/kg/d at 3, 6, and 12 months, respectively. Overall, 72.2% of patients underwent titration that was slower than label, 18.1% underwent label-adherent titration, and 4.1% underwent faster-than-label titration.Primary outcomes were retention, response (a reduction of ≥ 50% in seizure frequency), and seizure freedom (the absence of seizures between consecutive visits) at 3, 6, and 12 months, assessed by intention-to-treat (ITT) and per-protocol analyses, with the latter limited to patients who remained on cenobamate.TAKEAWAYTreatment retention remained high at 88%, 90%, and 93% at 3, 6, and 12 months, respectively. In the ITT analysis, responder rates increased from 64% at 3 months to 79% at 12 months, and seizure freedom rates increased from 8% to 18%; in the per-protocol analysis, the corresponding rates increased from 72% to 85% and from 9% to 19%, respectively. Overall, 53.9% of patients achieved at least one documented seizure-free interval during treatment.Children with structural aetiologies had higher responder rates at 12 months than those with non-structural DEEs (93% vs 68%; P = .04), although aetiology was not independently associated with response in adjusted analysis. Responder rates were similar in those with LGS and other cases of DEEs (79% in both groups).Responder rates were highest for tonic seizures (81%), followed by bilateral tonic-clonic seizures (76%) and absences (54%). Overall, 10.5% of patients experienced treatment-emergent seizure worsening, which was more common in Dravet syndrome (33.3%) and cases with predominantly generalised EEG patterns.The use of concomitant sodium channel blockers was independently associated with adverse events overall (odds ratio [OR], 2.10) and somnolence specifically (OR, 5.84; P < .05 for both). Overall, 10.5% of children discontinued treatment due to adverse events, primarily driven by severe somnolence, irritability or aggressiveness, and suicidal ideation.IN PRACTICE"Our findings support a potential role for CNB [cenobamate] across the DEE spectrum, with sustained seizure benefit and acceptable tolerability observed and response rates comparable to those reported in adult focal epilepsy," the authors wrote.SOURCEThis study was led by Ángel Aledo-Serrano and Adrián Valls-Carbó, Clinical Neurosciences Institute, Blua Sanitas Valdebebas University Hospital, Madrid, Spain. It was published online on August 28, 2026, in Epilepsia.LIMITATIONSThe retrospective study design without any control group may have introduced selection bias. Outcomes relied on caregiver-reported seizure diaries and clinician-rated Clinical Global Impression, lacking video-EEG or neuropsychological validation. The sample size in subgroups was small.DISCLOSURESThis study was supported by an unrestricted grant from Angelini Pharma. Several authors declared receiving financial support for medical writing, manuscript submission, and the journal's rapid service fees from different pharmaceutical companies.This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.