PDF: Cannabidiol in Dravet Syndrome
(2017)Objective
Cannabidiol - To assess the efficacy and safety of cannabidiol as an adjunctive therapy for drug-resistant convulsive seizures in children and young adults with Dravet syndrome.
Study Summary
Intervention
Oral cannabidiol 20 mg/kg/day vs. placebo, in addition to standard antiepileptic therapy, administered over a 14-week period.
Inclusion Criteria
Children and young adults (2–18 years) with confirmed Dravet syndrome and ≥4 convulsive seizures in a 28-day baseline period while on a stable antiepileptic drug regimen.
Study Design
Arms: Cannabidiol vs Placebo (both with background AEDs)
Patients per Arm: Cannabidiol: 61; Placebo: 59
Outcome
Bottom Line
Among patients with Dravet syndrome, cannabidiol resulted in a greater reduction in convulsive-seizure frequency than placebo (median reduction 38.9% vs 13.3%, P=0.01) and was associated with higher rates of adverse events including somnolence, elevated liver enzymes, diarrhea, vomiting, and fatigue.
Major Points
- First randomized, double-blind, placebo-controlled trial of cannabidiol for Dravet syndrome
- 120 children and young adults (ages 2-18) with drug-resistant Dravet syndrome were randomized 1:1 to cannabidiol 20 mg/kg/day or placebo
- Trial consisted of 4-week baseline, 14-week treatment period (2-week titration + 12-week maintenance), 10-day taper, and 4-week safety follow-up
- Median convulsive seizure frequency decreased from 12.4 to 5.9 per month with cannabidiol vs 14.9 to 14.1 with placebo
- Adjusted median difference in seizure reduction: -22.8 percentage points (95% CI, -41.1 to -5.4; P=0.01)
- 43% of cannabidiol patients achieved ≥50% seizure reduction vs 27% with placebo (OR 2.00, P=0.08)
- 5% of cannabidiol patients became seizure-free vs 0% with placebo (P=0.08)
- 62% of caregivers reported overall condition improvement with cannabidiol vs 34% with placebo (P=0.02)
- Total seizures (all types) significantly reduced with cannabidiol (P=0.03), but no significant reduction in nonconvulsive seizures alone
- More withdrawals in cannabidiol group (9/61, 15%) vs placebo (3/59, 5%)
- Common adverse events: diarrhea (31%), somnolence (36%), decreased appetite (28%), vomiting (15%), fatigue (20%)
- Elevated liver enzymes occurred in 12 cannabidiol patients vs 1 placebo patient, all taking valproate
- Three patients discontinued due to elevated aminotransferases in cannabidiol group
- 90% of patients completed treatment period; 105/108 entered open-label extension
Study Design
- Study Type
- Multinational, multicenter, randomized, double-blind, placebo-controlled trial
- Randomization
- Yes
- Blinding
- Double-blind (patients, caregivers, investigators). Matching placebo solution identical except for absence of cannabidiol.
- Sample Size
- 120
- Follow-up
- 14-week treatment period plus 10-day taper plus 4-week safety follow-up
- Centers
- 23
- Countries
- United States, Europe
Primary Outcome
Definition: Percentage change per 28 days from 4-week baseline period in convulsive-seizure frequency during 14-week treatment period
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| Median change -13.3% (IQR -52.5 to 20.2). Median frequency decreased from 14.9 to 14.1 seizures per month. | Median change -38.9% (IQR -69.5 to -4.8). Median frequency decreased from 12.4 to 5.9 seizures per month. | - (Adjusted median difference -22.8 percentage points (95% CI -41.1 to -5.4)) | 0.01 |
Limitations & Criticisms
- Partially subjective primary endpoint (caregiver-reported seizures) could be affected by unblinding due to side effects
- Higher rate of adverse events in cannabidiol group may have unblinded some patients/caregivers
- Differences in palatability between active treatment and placebo could have affected blinding
- Short treatment duration (14 weeks) limits assessment of long-term efficacy and safety
- No data on nonconvulsive seizure reduction, which was not significant - unclear if this represents lack of efficacy or difficulty in counting by parents in developmentally delayed children
- Higher withdrawal rate in cannabidiol group (15% vs 5%) may affect interpretation
- Drug interactions with concomitant medications (especially valproate and clobazam) complicate safety profile
- Applicability of Columbia Suicide Severity Rating Scale unclear in population with cognitive impairment
- Study population highly selected with severe, treatment-resistant epilepsy - may not generalize to all Dravet syndrome patients
- No active comparator - only compared to placebo
- No P-value adjustments for multiple secondary endpoints
- Median baseline seizure frequency higher in placebo group (14.9 vs 12.4), though not statistically significant
Citation
N Engl J Med 2017;376:2011-20