Dravet syndrome: 2026 so far

The months covered added information beyond seizure counts, including status epilepticus.

In the months covered, fenfluramine evidence extended beyond just counting seizures. Pooled analyses from randomized trials supported lower convulsive seizure frequency versus placebo in most subgroups; the reduction was statistically significant in children aged 4 and older, but not in the under-4 subgroup. Retrospective routine-care data also linked fenfluramine exposure to fewer status epilepticus episodes (status epilepticus is a prolonged seizure emergency). Real-world follow-up linked fenfluramine to fewer seizures over time and fewer seizure-related urgent visits among people who stayed on treatment. One cohort comparison found that fenfluramine-containing regimens were associated with lower seizure burden at follow-up than other regimens. Across these reports, differences in function or cognitive/behavior outcomes were not shown consistently. Also, observational studies cannot prove that fenfluramine caused the changes seen. In practical terms, the January 2026 to September 2026 evidence kept pointing toward fewer convulsive seizures and possibly fewer prolonged seizure emergencies for some people, while still requiring attention to monitoring and side effects.

Research covered in: April 2026, May 2026, June 2026, August 2026, September 2026

Real-world follow-up in the months covered showed a wide range of responses.

From January 2026 to September 2026, reports on several non-fenfluramine options added more real-world follow-up. Many people had meaningful seizure reductions. At the same time, responses ranged from large improvements to little change, and side effects were common in some settings. Cannabidiol (pharmaceutical-grade CBD) was repeatedly linked to average seizure reductions in pooled placebo-controlled trial summaries and to sustained responder rates in routine-care cohorts. Side effects and treatment discontinuations were reported. Interaction-related risks were also flagged when CBD was combined with valproate or clobazam. Perampanel had mixed Dravet-specific evidence across the months covered. One infant/toddler drug-resistant epilepsy cohort was not mainly Dravet. A later prospective real-world Dravet study reported common seizure reductions and some seizure freedom, with relatively few discontinuations. Device-based approaches included VNS (vagus nerve stimulation, an implanted nerve stimulator) and RNS (responsive neurostimulation, an implanted device that detects and responds to brain signals). These were described in retrospective cohorts or case series with many responders, but uneven outcomes and limited ability to predict who benefits. Overall, the available reports suggest several established medicines and devices can reduce seizures for many people, but the tradeoffs in benefit and tolerability can be very different from one person to the next.

Research covered in: January 2026, May 2026, June 2026, July 2026, September 2026

Human results in the months covered were mixed, and safety uncertainty was prominent.

In March, a large phase 3 (a large late-stage clinical trial) add-on study of soticlestat showed seizure reductions favoring soticlestat over placebo. However, the trial's main planned statistical test did not clearly pass. Several secondary measures still moved in a favorable direction, including caregiver/clinician impressions and seizure intensity/duration ratings. Also in March, an open-label (everyone knows they are receiving the drug) program of zorevunersen reported substantial convulsive seizure reductions during longer follow-up. The reports also emphasized procedure-related effects and serious events, including reported SUDEP deaths (SUDEP is sudden unexpected death in epilepsy) and one death due to malnutrition. Without a control group, treatment effects cannot be separated from other influences over time. Later months added observational signals for cenobamate in very small Dravet series and in a larger mixed-DEE cohort (DEE is developmental and epileptic encephalopathy, epilepsy with developmental impacts). In that mixed cohort, Dravet showed limited sustained benefit, while meaningful adverse events were reported across the overall study population, keeping syndrome-specific benefit and safety uncertain. Taken together, the months covered show that newer approaches were being tested, but the most direct human evidence was either statistically mixed (soticlestat) or uncontrolled with important safety events (zorevunersen).

Research covered in: March 2026, June 2026, August 2026

Measurement in the months covered leaned more toward behavior, sleep, and participation.

In the months covered, research more often highlighted that Dravet syndrome affects more than seizures, and that these areas need to be measured directly. New tools and frameworks included a highly repeatable caregiver-administered D-DAND interview (a structured interview about development and behavior). Caregiver co-design work also prioritized behavior, independence, and social participation for home tracking. Several surveys linked caregiver burden and distress to behavioral challenges, complex routines, and navigating systems. In one study, empowerment scores were associated with lower distress. Smaller studies also described early timing of behavior and social-emotional concerns in SCN1A-positive young children, and variable communication profiles, often with expressive language more affected than understanding. Overall, January 2026 to September 2026 reports put more emphasis on tracking and describing day-to-day functioning and caregiver impact, not only seizure frequency.

Research covered in: January 2026, February 2026, April 2026, July 2026, August 2026, September 2026

The months covered put more focus on status epilepticus, SUDEP estimates, and rare complications.

So far in 2026, monitoring-related work added more structure around high-impact risks. It also highlighted uncertainty about how well research datasets match everyday risk. An expert pathway proposed structured cardiology follow-up around fenfluramine treatment. Retrospective data linked fenfluramine exposure to fewer status epilepticus episodes. In the same dataset, stiripentol exposure was linked to shorter status epilepticus duration. Separately, a Bayesian meta-analysis (a statistical method that combines results across studies) using randomized trials and extension studies estimated mortality and SUDEP rates for Dravet syndrome. It cautioned that trial-based data may underestimate real-world risk. Rare case reports described reversible pulmonary arterial hypertension (high blood pressure in the lung arteries) during off-label fenfluramine treatment in a very young child, and severe ARDS episodes (acute respiratory distress syndrome, sudden severe lung failure) during febrile illness/status epilepticus in two children without known chronic lung disease. In effect, the January 2026 to September 2026 reports brought more explicit attention to how severe seizure emergencies and rare complications are monitored, while leaving the true frequency and predictors uncertain.

Research covered in: April 2026, June 2026, August 2026, September 2026