BackgroundPerampanel (PER) is a new antiepileptic drug (AED) with a novel mechanism of action. Investigations of the efficacy and safety of PER in pediatric and adult patients have increased recently. Although the clinical usefulness and pharmacokinetics of PER have been investigated in adolescent and adult populations, similar studies have not been performed in children. Patients and methodsWe retrospectively reviewed the medical records of patients treated with PER for more than 6 months in the Department of Pediatrics, Hiroshima University Hospital, between September 2016 and November 2018. We obtained demographic and clinical data including age, sex, epilepsy type, seizure type, seizure frequency before and after treatment initiation, adverse events, reasons for discontinuing PER treatment, doses at evaluation points, serum concentrations, concomitant AEDs, intellectual status, and epilepsy etiology. Seizure types and epilepsy syndromes were classified according to the criteria of the International League Against Epilepsy. ResultsThe study included 44 patients (22 males) between the ages of 6 months and 16 years. Of those, 10 patients discontinued PER therapy. The 50% response rate was 52.3% in patients treated with PER, and four patients achieved complete seizure control. Perampanel was highly effective in patients with generalized and focal epilepsy (50% responder rates, 52.9% and 50.0%, respectively). Favorable response rates were observed for tonic–clonic, focal nonmotor, and absence seizures with 50% response rates of 54.5%, 50.0%, and 66.7%, respectively. The 50% responder rate was 31.3 for epileptic spasms (ES). Treatment-emergent adverse events (TEAEs) included somnolence (n = 8), irritability (n = 2), ataxia (n = 2), and one case each of dizziness, compulsiveness, and enuresis. Serum concentrations of PER were compared in patients taking concomitant enzyme-inducing antiepileptic drugs (EIAEDs; carbamazepine, phenytoin, and phenobarbital) and those taking concomitant non-EIAEDs. Serum PER concentrations were correlated with dose per body weight in both groups (EIAED: r = 0.765, p = 0.00000212; non-EIAED: r = 0.71, p = 0.0000158). The mean concentration-to-dose (CD) ratio was 2398.4 ng mL−1 mg−1 kg−1 (range: 800–4524.7) in the non-EIAED group and 693.7 ng mL−1 mg−1 kg−1 (range: 344–1309.7) in the EIAED group. Serum PER levels were lower in the EIAED group than in the non-EIAED group. All patients with serum PER concentrations above 400 ng/mL experienced somnolence. ConclusionsPerampanel is effective against various types of seizures, including ES, in pediatric patients with refractory epilepsy. Furthermore, PER has good tolerability when the dose is adjusted based on serum concentrations. The PER CD ratio was lower in pediatric patients than in adolescents and adults; therefore, clinicians must consider the CD ratio when treating children with PER.