Long-term efficacy and safety of cannabidiol (CBD) in children with treatment-resistant epilepsy: Results from a state-based expanded access program

Epilepsy Behav. 2020 Nov:112:107474. doi: 10.1016/j.yebeh.2020.107474. Epub 2020 Sep 28.


Introduction: An intermediate-sized, multicenter, expanded-access study was opened in 2015 through the support of the State of Georgia. This study provided children with treatment-resistant epilepsy (TRE) access to plant-derived highly purified cannabidiol (CBD; Epidiolex® in the US; Epidyolex® in the EU; 100 mg/mL oral solution). These children had failed to achieve seizure freedom with available treatment options and were ineligible to participate in randomized controlled trials that only included patients with Lennox-Gastaut and Dravet syndromes.

Methods: Cannabidiol safety, changes in seizure type, frequency, and seizure-free days were evaluated for children aged 1-18 years (at time of consent) as an adjunctive treatment for 36 months. The study consisted of a two-month baseline period, a titration period, treatment period, and optional titration period, which occurred after ≥26 weeks of treatment. Cannabidiol treatment was administered up to a targeted dose of 25 mg/kg/day, with an optional secondary treatment up to 50 mg/kg/day. Daily seizure type, seizure frequency, and seizure-free days were recorded in a Web-based diary, and changes in these outcomes were recorded and analyzed for the duration of the study. The occurrence of adverse events (AEs) was also recorded.

Results: The median percentage change in seizures for 45 patients in Months 3, 6, 12, 18, 24, and 36 showed a statistically significant (p < 0.001) reduction in major seizures (ranging from 54 to 72% at various time points) and all seizures (61-70%) compared with baseline. A mean increase in seizure-free days per 28 days was >5 in all treatment periods after Month 2, and an average increase of 7.52 (p < 0.001) seizure-free days per 28 days was observed at the end of follow-up compared with baseline. All patients experienced ≥1 AE. Children who transitioned to the optional secondary treatment (high-dose group) reported more AEs before increasing their dose to >25.0 mg/kg/day compared with the low-dose group. However, the average rate of AEs was significantly lower after moving to a high-dose regimen (p = 0.004). Twelve children reported 20 serious AEs, none of which were considered related to CBD.

Conclusions: This study supports CBD as an adjunctive treatment for children with TRE. Treatment was well tolerated in doses up to 50 mg/kg/day. Patients who did not achieve desired results at a dose of ≤25.0 mg/kg/day reported more AEs when CBD dose increased to >25.0 mg/kg/day. Decreases in major seizure frequency and an increase in seizure-free days compared with baseline were reported during treatment. This supports the efficacy and tolerability of CBD for mixed seizure etiologies.

Keywords: Cannabidiol; Pediatric epilepsy; Treatment-resistant epilepsy.

Publication types

  • Multicenter Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Anticonvulsants / therapeutic use
  • Cannabidiol* / therapeutic use
  • Child
  • Child, Preschool
  • Epilepsies, Myoclonic* / drug therapy
  • Epilepsy* / drug therapy
  • Humans
  • Infant
  • Seizures / drug therapy


  • Anticonvulsants
  • Cannabidiol