One in three people with epilepsy has drug-resistant disease — defined as failure of two appropriate antiseizure medication trials — and that proportion hasn't meaningfully changed despite the introduction of more than a dozen new ASMs since 1990. The drugs available share similar mechanisms: sodium channel blockers, GABA potentiators, calcium channel modulators. They're working through the same limited set of pathways. What's genuinely different in 2026 is the gene therapy pipeline. For several Mendelian epilepsy syndromes caused by single gene mutations, early-phase gene therapy data has been sufficiently promising to warrant Phase 3 planning — and that's a moment the epilepsy field has been building toward for years.
This article is for informational purposes only and does not constitute medical advice. Clinical trial eligibility and availability vary. Always consult a qualified healthcare professional before making any medical decisions or considering participation in a clinical trial.
Summary
Approximately 30% of people with epilepsy have drug-resistant seizures — seizures that persist despite trying two or more appropriate antiseizure medications. For this population, and for children with severe genetic epilepsies like Dravet syndrome and tuberous sclerosis complex (TSC), 2026 clinical research offers targeted approaches that didn't exist a decade ago: fenfluramine reducing convulsive seizures by 62% vs. placebo in Phase 3, cannabidiol approved for three severe childhood epilepsies, ETX101 gene therapy showing early promise in Dravet syndrome, and closed-loop responsive neurostimulation achieving 75% median seizure reduction at 9 years of follow-up.
Fenfluramine (Fintepla) for Dravet Syndrome and Lennox-Gastaut
Dravet syndrome is caused primarily by SCN1A haploinsufficiency — loss of function in one allele encoding the Nav1.1 sodium channel, expressed predominantly in inhibitory GABAergic interneurons. The result is impaired inhibitory interneuron function, leading to the paradox of hyperexcitability from lost inhibition. Patients experience prolonged febrile seizures in the first year of life, followed by a lifetime of multiple refractory seizure types and developmental impairment.
Fenfluramine (Fintepla, UCB) was originally a weight-loss drug withdrawn from the market in 1997 due to cardiac valvulopathy at the high doses used. At doses reformulated for epilepsy (0.1–0.7 mg/kg/day — roughly 10-fold lower than the weight-loss doses), the cardiac effects have not been observed across trial populations, and cardiac monitoring remains a standard protocol requirement. Fenfluramine's antiseizure mechanism likely involves serotonin-mediated modulation of sigma-1 receptors and possibly direct interaction with SCN1A-expressing interneurons.
The Phase 3 STUDIO trial in Dravet syndrome showed fenfluramine reduced monthly convulsive seizure frequency by 62.3% from baseline versus 11.1% for placebo — a highly statistically significant and clinically striking result. FDA approved it for Dravet syndrome in June 2020 and for Lennox-Gastaut syndrome (LGS) in July 2022. Current 2026 trials are evaluating combination strategies with existing ASMs and long-term cardiac surveillance protocols in children with prolonged use.
Cannabidiol (Epidiolex): Mechanism and Expansion
Pharmaceutical-grade cannabidiol (Epidiolex, Jazz Pharmaceuticals) is FDA-approved for seizures associated with Dravet syndrome, Lennox-Gastaut syndrome, and tuberous sclerosis complex. Unlike THC, CBD is non-intoxicating. Its mechanism in epilepsy is not fully established — the leading candidates include GPR55 receptor antagonism, modulation of voltage-gated sodium channels, and effects on adenosine signaling pathways — and the fact that the precise mechanism remains debated doesn't diminish its clinical utility.
The GWPCARE clinical trial series (GWPCARE1–6) established the evidence base: median seizure reduction of 40–50% versus placebo across these three syndromes. A critical drug interaction to understand: CBD inhibits CYP2C19, significantly increasing serum levels of clobazam (a common co-medication in Dravet and LGS). This interaction likely contributes to some of the efficacy observed in combination regimens but also contributes to somnolence.
Current 2026 research is investigating CBD formulations and dosing strategies in CDKL5 deficiency disorder (CDD) and PCDH19 epilepsy — conditions that share phenotypic features with Dravet syndrome and where no targeted therapies currently exist.
Gene Therapy for Dravet Syndrome and TSC
Dravet syndrome's single-gene etiology — haploinsufficiency of SCN1A — makes it a compelling gene therapy target, but the approach requires creativity because delivering a full-length SCN1A gene exceeds the packaging capacity of standard AAV vectors. Multiple therapeutic strategies have been developed:
ETX101 (Encoded Therapeutics) — ENDEAVOR Phase 1/2
ETX101 uses an AAV9 vector to deliver an engineered transcription factor that upregulates expression from the intact (non-mutated) SCN1A allele — compensating for the haploinsufficiency without needing to deliver the full gene sequence. This transcription factor approach is elegant: rather than correcting the defective allele or delivering a replacement, it amplifies the signal from what's already functional. Phase 1/2 ENDEAVOR trial data in young children with Dravet syndrome has shown meaningful seizure reduction with an acceptable safety profile. FDA Breakthrough Therapy designation was granted, accelerating the regulatory path to Phase 3.
Antisense Oligonucleotides for TSC
Tuberous sclerosis complex (TSC) is caused by mutations in TSC1 or TSC2, leading to mTOR pathway hyperactivation and cortical tuber formation. While mTOR inhibitors (everolimus) reduce subependymal giant cell astrocytomas and some seizure types, they don't address the full neurological impact of the condition. ASO approaches targeting mTOR pathway components — designed to reduce pathway hyperactivation more selectively — are in early-phase development, with the goal of addressing seizures and potentially cognitive symptoms simultaneously.
Reducing Vigabatrin Exposure in TSC-Related Infantile Spasms
Vigabatrin is first-line for TSC-related infantile spasms but causes irreversible visual field defects with prolonged exposure. Trials of mTOR inhibitor and ACTH combination regimens are attempting to reduce vigabatrin exposure without sacrificing seizure control — a harm-reduction strategy that the TSC community has prioritized given the visual consequences of long-term vigabatrin use in infants.
Closed-Loop Neurostimulation: 9 Years of Improving Data
For drug-resistant focal epilepsy, the NeuroPace RNS System is unique in being the only closed-loop responsive neurostimulation device in clinical use. Electrodes implanted directly at identified seizure foci continuously record electrocorticography; when the system detects the pre-ictal pattern characteristic for that patient's seizures, it delivers a brief burst of electrical stimulation to abort the event before it propagates. The system learns — it stores years of intracranial EEG data and can be reprogrammed based on that individual's evolving seizure patterns.
The long-term follow-up data from the pivotal RNS trial is one of the more compelling outcome datasets in epilepsy neurostimulation: median seizure reduction of 75% at 9+ years, with responder rates continuing to improve over time rather than plateauing. This suggests ongoing neuroplastic changes beyond the acute stimulation effect. Active 2026 trials are evaluating RNS for primary generalized epilepsies (historically considered a contraindication), pediatric patients, and machine learning-enhanced detection algorithms.
Deep brain stimulation targeting the anterior nucleus of the thalamus (Medtronic Percept PC — ANT-DBS) received FDA approval in 2018. The SANTE trial showed 69% median seizure reduction at 5 years. Updated closed-loop DBS systems using on-device sensing are in Phase 2 trials — the ability to record field potentials from the thalamus and trigger stimulation based on biomarkers rather than fixed schedules is a meaningful technical advance over first-generation DBS for epilepsy.
Key Takeaways
- Fenfluramine at low doses reduces Dravet syndrome monthly convulsive seizures by 62.3% vs. 11.1% placebo in Phase 3 STUDIO — now FDA-approved for Dravet (2020) and Lennox-Gastaut (2022).
- Cannabidiol (Epidiolex) is FDA-approved for three severe childhood epilepsies; trials are expanding to CDKL5 deficiency disorder and PCDH19 epilepsy where no targeted therapies currently exist.
- ETX101 gene therapy for Dravet syndrome upregulates the intact SCN1A allele using an AAV9-delivered transcription factor — Phase 1/2 ENDEAVOR trial has FDA Breakthrough Therapy designation and is showing early efficacy signals.
- Responsive neurostimulation (RNS) achieves 75% median seizure reduction at 9+ years with continuing improvement — the long-term durability data distinguishes it from earlier neurostimulation results.
- Closed-loop ANT-DBS and machine learning-enhanced responsive DBS systems are in Phase 2 trials, expanding options for patients with generalized and focal drug-resistant epilepsy.