By Kimberly Pargeon, MD

Assistant Professor of Clinical Neurology, Weill Cornell Medical College

Dr. Pargeon reports no financial relationships relevant to this field of study.

SYNOPSIS: In a double-blind study, 120 children and young adults with the Dravet syndrome and medically refractory seizures were assigned randomly to receive either cannabidiol or placebo, as well as their usual antiepileptic drugs/therapies. The primary finding was a significant decrease in convulsive seizure frequency during the 14-week treatment period for patients receiving cannabidiol compared to those receiving placebo.

SOURCE: Devinsky O, Cross JH, Laux L, et al. Trial of cannabidiol for drug-resistant seizures in the Dravet syndrome. N Engl J Med 2017;376:2011-2020.

Currently, there are no approved medical treatments for Dravet syndrome (DS), a childhood epileptic encephalopathy.1 Typically, DS is caused by mutations in the SCN1A gene, resulting in seizures starting at around age 6 months, usually in the context of high fever, often after initially normal development. Seizures can become medically refractory and children can show neurological regression, often with significant developmental delay, and are at risk for early death.2 Families can be frustrated by failed treatment with polypharmacy using current antiepileptic drugs (AEDs) leading to frequent side effects.1 Medical marijuana (MMJ) is a popular topic for the media and patients alike, particularly for those seeking “natural” treatments or for patients and families in desperate need of a “magic bullet” for devastating drug-resistant seizures. Unfortunately, until recently, available information for clinicians has been limited to case reports, case series, or inconclusive trials in varied populations.1 Based on preclinical data and anecdotal reports, a group of clinical investigators and GW Pharmaceuticals collaborated to investigate the safety and efficacy of cannabidiol (CBD) in childhood refractory epilepsies using Epidiolex (100 mg/mL CBD).1

This was a double-blind, placebo-controlled trial in which children and young adults with confirmed DS and medically refractory seizures were assigned randomly to receive CBD at 20 mg/kg/day or placebo in addition to their usual AEDs. The study was conducted at 23 centers in the United States and Europe, with 120 subjects undergoing randomization (61 CBD, 59 placebo). Subjects ranged in age from 2-18 years (mean 9.8 years) and 52% were male. All treatments were stable for four weeks prior to randomization and throughout the trial, which began with a four-week baseline period, during which baseline seizure frequency was determined (eligibility required four or more seizures). This was followed by a 14-week treatment period, during which CBD or placebo was escalated to 20 mg/kg/day over 14 days and then maintained for 12 weeks. This was followed by a 10-day taper and then a four-week follow-up period. After trial completion, subjects could enter an open-label study.

Ninety percent of subjects completed the treatment period (52/61 CBD vs. 56/59 placebo). Subjects previously had tried a median of 4.0 AEDs, most commonly clobazam, valproate, stiripentol, levetiracetam, and topiramate. Baseline convulsion frequency was a median of 13.0 seizures per month, the most common type being generalized tonic-clonic (78%). Nearly all subjects had some degree of developmental delay, with nearly half (48%) having “severe or profound” disability.

The primary endpoint was change in convulsive seizure frequency from baseline. There was a decrease from a median of 12.4 seizures/month at baseline to 5.9 seizures/month (-38.9%) in the CBD group, as compared to a decrease from 14.9 seizures/month at baseline to 14.1 seizures/month in the placebo group (-13.3%). The adjusted median difference was -22.8% (P = 0.01). For secondary endpoints, the percentage of subjects with at least a 50% reduction of convulsive seizures was 43% and 27% for the CBD and placebo groups, respectively (P = 0.08). Although the frequency for all seizure types in the CBD group was significantly reduced compared to the placebo group (P = 0.03), there was no significant reduction for non-convulsive seizures. Five percent of the CBD patients were seizure free for all seizure types, but none were seizure free in the placebo group (P = 0.08). Patients’ overall condition improved, measured by a caregiver scale, in 62% of the CBD group compared to 34% of the placebo group (P = 0.02). However, more side effects were reported in the CBD group (93% vs. 75%), most rated mild to moderate (89%). Most adverse events (75%) were attributed to the trial agent and included diarrhea, vomiting, fatigue, somnolence, pyrexia, and abnormal liver function tests. More trial withdrawals due to adverse events occurred in the CBD group (8 vs. 1).


Up to one-third of patients with epilepsy will be medically refractory, especially those with epileptic encephalopathies, such as DS, for which there is no approved medical treatment.1 Patients often are treated incompletely with multiple AEDs, leading to negative side effects. Although MMJ has gained recent popularity, until recently, we had only limited or anecdotal evidence for the efficacy of MMJ, with little empiric data related to safety and proper dosing. Although more than half of states and the District of Columbia have approved MMJ programs, the available forms and modes of administration vary.1 By using a randomized, controlled design, Devinsky et al finally have provided prospective empiric evidence supporting the specific use of CBD at 20 mg/kg in children and young adults with DS for significant reduction of convulsive seizures, which potentially could reduce the risk for convulsion-related complications, such as status epilepticus and sudden unexpected death in epilepsy. Future directions should focus on replicating results with DS and other epileptic encephalopathies, as well as for other refractory epilepsies and in other age groups.


  1. O’Connell BK, Gloss D, Devinsky O. Cannabinoids in treatment-resistant epilepsy: A review. Epilepsy Behav 2017;70:341-348.
  2. Berkovic SF. Cannabinoids for epilepsy – Real data, at last. N Engl J Med 2017;376:2075-2076.