Fenfluramine Dravet Study
(2019)Objective
To evaluate the efficacy and safety of fenfluramine at two doses (0.2 and 0.7 mg/kg/day) versus placebo for convulsive seizure reduction in Dravet syndrome
Study Summary
• FDA boxed warning for valvular heart disease/PAH requiring REMS with echocardiographic monitoring; AEs: decreased appetite (23-38% vs 8%), somnolence (25-26% vs 11%), abnormal echo (18-23% vs 6%); stiripentol users excluded (separate study)
Intervention
Fenfluramine 0.2 mg/kg/day or 0.7 mg/kg/day oral solution vs placebo, as add-on to existing AEDs
Inclusion Criteria
Age 2-18, clinical Dravet syndrome diagnosis with SCN1A variant, >=6 convulsive seizures per 6-week baseline, 1-4 concomitant AEDs, stable regimen >=4 weeks, no stiripentol use
Study Design
Arms: Fenfluramine 0.7 mg/kg/day vs Fenfluramine 0.2 mg/kg/day vs Placebo
Patients per Arm: FFA 0.7: 40, FFA 0.2: 39, Placebo: 39
Outcome
• Responder rate (>=50%): FFA 0.2: 58%, FFA 0.7: 50%, Placebo: 33%; Seizure-free: 8% both FFA vs 0% placebo
• AEs: decreased appetite 23-38% vs 8%, abnormal echo 18-23% vs 6%, somnolence 25-26% vs 11%; 13% discontinuation in 0.7 mg group; FDA boxed warning for VHD/PAH, REMS required
Bottom Line
Fenfluramine at both 0.2 mg/kg/day and 0.7 mg/kg/day significantly reduced monthly convulsive seizure frequency compared to placebo in Dravet syndrome (median reductions of 70.0% and greater vs 31.7% for placebo). However, concerns about cardiac safety (historical association with valvular heart disease) led to an FDA boxed warning and mandatory REMS program with echocardiographic monitoring every 6 months.
Major Points
- Fenfluramine 0.7 mg/kg/day reduced convulsive seizure frequency by 62.3% vs placebo 1.2% (P<0.001). 0.2 mg/kg/day reduced by 32.4% (P=0.001).
- ≥50% responder rate: 68% (0.7 mg/kg) vs 40% (0.2 mg/kg) vs 5% (placebo).
- 119 patients (ages 2-18) with Dravet syndrome. 3-arm, 14-week treatment. 35 centers.
- Longest seizure-free interval: 22 days (0.7 mg/kg) vs 13 days (0.2 mg/kg) vs 3.5 days (placebo).
- Cardiac safety: no echocardiographic abnormalities. Regular cardiac monitoring required.
- Most common AEs: decreased appetite (24-35%), diarrhea (15-24%), fatigue, lethargy.
- Published Lancet 2019. Led to FDA approval of Fintepla for Dravet in 2020.
- Dose-response relationship: 0.7 mg/kg clearly superior to 0.2 mg/kg.
- Mechanism: serotonin release + sigma-1 receptor agonism (distinct from prior appetite suppressant use).
- Fenfluramine previously withdrawn for cardiac valve disease at higher doses — low-dose epilepsy use safe.
Study Design
- Study Type
- Phase 3, multicenter, randomized, double-blind, placebo-controlled, parallel-group trial
- Randomization
- Yes
- Blinding
- Double-blind
- Sample Size
- 119
- Follow-up
- 14 weeks (2-week titration + 12-week maintenance)
- Centers
- Multiple
- Countries
- United States, Europe, Australia
Primary Outcome
Definition: Change in monthly (per 28 days) convulsive seizure frequency vs baseline
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| Placebo: -31.7% median change from baseline | FFA 0.2 mg/kg/day: -70.0%; FFA 0.7 mg/kg/day: greater reduction (not explicitly stated) | - | 0.2 mg/kg vs placebo: p=0.043; 0.7 mg/kg vs placebo: p<0.001 |
Limitations & Criticisms
- Baseline seizure frequency imbalance: placebo group had substantially higher baseline (29.4/28 days) than fenfluramine groups (18.1 and 18.7), potentially biasing results
- Short treatment duration (14 weeks) insufficient to assess long-term efficacy and cardiac safety
- Small sample size (39-40 per arm) underpowered for rare adverse events
- Cardiac safety concerns: given fenfluramine's history causing VHD in obesity, short trial inadequate for long-term cardiac risk
- No dose-response: 0.2 mg/kg/day paradoxically showed greater seizure reduction and higher responder rate than 0.7 mg/kg/day
- Industry-funded: designed, conducted, and analyzed by Zogenix
- Exclusion of stiripentol users limits generalizability since stiripentol is a Dravet treatment mainstay
Citation
Lagae L et al. Lancet. 2019;394(10216):2243-2254. DOI: 10.1016/S0140-6736(19)32500-0