Comparative efficacy and safety of stiripentol, cannabidiol and fenfluramine as first-line add-on therapies for seizures in Dravet syndrome: A network meta-analysis

Epilepsia Open. 2024 Apr;9(2):689-703. doi: 10.1002/epi4.12923. Epub 2024 Mar 1.


Objectives: Stiripentol, fenfluramine, and cannabidiol are licensed add-on therapies to treat seizures in Dravet Syndrome (DS). There are no direct or indirect comparisons assessing their full licensed dose regimens, across different jurisdictions, as first-line add-on therapies in DS.

Methods: We conducted a systematic review and frequentist network meta-analysis (NMA) of randomized controlled trial (RCT) data for licensed add-on DS therapies. We compared the proportions of patients experiencing: reductions from baseline in monthly convulsive seizure frequency (MCSF) of ≥50% (clinically meaningful), ≥75% (profound), and 100% (seizure-free); serious adverse events (SAEs); discontinuations due to AEs.

Results: We identified relevant data from two placebo-controlled RCTs for each drug. Stiripentol 50 mg/kg/day and fenfluramine 0.7 mg/kg/day had similar efficacy in achieving ≥50% (clinically meaningful) and ≥75% (profound) reductions from baseline in MCSF (absolute risk difference [RD] for stiripentol versus fenfluramine 1% [95% confidence interval: -20% to 22%; p = 0.93] and 6% [-15% to 27%; p = 0.59], respectively), and both were statistically superior (p < 0.05) to licensed dose regimens of cannabidiol (10 or 20 mg/kg/day, with/irrespective of clobazam) for these outcomes. Stiripentol was statistically superior in achieving seizure-free intervals compared to fenfluramine (RD = 26% [CI: 8% to 44%; p < 0.01]) and licensed dose regimens of cannabidiol. There were no significant differences in the proportions of patients experiencing SAEs. The risk of discontinuations due to AEs was lower for stiripentol, although the stiripentol trials were shorter.

Significance: This NMA of RCT data indicates stiripentol, as a first-line add-on therapy in DS, is at least as effective as fenfluramine and both are more effective than cannabidiol in reducing convulsive seizures. No significant difference in the incidence of SAEs between the three add-on agents was observed, but stiripentol may have a lower risk of discontinuations due to AEs. These results may inform clinical decision-making and the continued development of guidelines for the treatment of people with DS.

Plain language summary: This study compared three drugs (stiripentol, fenfluramine, and cannabidiol) used alongside other medications for managing seizures in a severe type of epilepsy called DS. The study found that stiripentol and fenfluramine were similarly effective in reducing seizures and both were more effective than cannabidiol. Stiripentol was the best drug for stopping seizures completely based on the available clinical trial data. All three drugs had similar rates of serious side effects, but stiripentol had a lower chance of being stopped due to side effects. This information can help guide treatment choices for people with DS.

Keywords: Dravet syndrome; cannabidiol; fenfluramine; network meta‐analysis; stiripentol.

Publication types

  • Systematic Review
  • Meta-Analysis

MeSH terms

  • Anticonvulsants / therapeutic use
  • Cannabidiol* / therapeutic use
  • Dioxolanes*
  • Epilepsies, Myoclonic* / drug therapy
  • Fenfluramine / therapeutic use
  • Humans
  • Network Meta-Analysis
  • Randomized Controlled Trials as Topic
  • Seizures / drug therapy
  • Seizures / etiology


  • Cannabidiol
  • stiripentol
  • Anticonvulsants
  • Fenfluramine
  • Dioxolanes

Grants and funding