Slow Titration of Cannabidiol Add-On in Drug-Resistant Epilepsies Can Improve Safety With Maintained Efficacy in an Open-Label Study.

To assess adverse events (AEs) and efficacy of add-on cannabidiol (CBD) with a slower titration protocol in pediatric clinical practice. We conducted a prospective, open-label, multicenter study in seven French reference centers for rare epilepsies. Patients had slow titration to reach a target dose of 10 mg/kg/day within at least 1 month and then gradually increased to a maximum dose of 20 mg/kg/day. We analyzed AEs and efficacy at M1 (month 1), M2, and M6, comparing two sets of subgroups: Dravet syndrome (DS) vs. Lennox-Gastaut (LGS) and patients with clobazam (CLB+) vs. patients without (CLB-). One hundred and twenty-five patients were enrolled (62 LGS, 48 DS, 5 Tuberous sclerosis, and 10 other etiologies). Median concomitant antiepileptic drugs (AEDs) was three (25th percentile: 3, 75th percentile: 4). Patients received a dose of 10 (10-12), 14 (10-20), and 15.5 mg/kg/day (10-20) at M1, M2, and M6, respectively. Twenty-six patients discontinued CBD, 19 due to lack of efficacy, 2 due to AEs, 4 for both, and 1 had a sudden unexpected death in epilepsy. AEs were reported in 61 patients (48.8%), mainly somnolence ( = 26), asthenia ( = 20), and behavior disorders ( = 16). Abnormal transaminases (≥3 times) were reported in 11 patients receiving both valproate and clobazam. AEs were significantly higher at M2 ( = 0.03) and increased with the number of AEDs ( = 0.03). At M6, total seizure frequency change from baseline was -41% ± 37.5% (mean ± standard deviation), and 28 patients (37.8%) had a reduction ≥50%. AE and efficacy did not differ between DS vs. LGS and CLB+ vs. CLB- patients. A slower titration of CBD dose delivered better tolerance with comparable efficacy to previous trials. Concomitant CLB did not increase efficacy rates but in a few cases increased AEs. This slow titration scheme should help guide clinicians prescribing CBD and allow patients to benefit from its potential efficacy.