Drug-Resistant Epilepsy: Surgical Evaluation, Cenobamate, and the Pipeline Reshaping Care

Why Drug-Resistant Epilepsy Matters Now

Approximately 30% of people with epilepsy do not achieve sustained seizure freedom on antiseizure medication. This subgroup carries the bulk of epilepsy-related morbidity, including injury, cognitive decline, psychiatric comorbidity, and a 24-fold increase in the risk of sudden unexpected death in epilepsy (SUDEP) compared with the general population.

Despite this burden, real-world management of drug-resistant epilepsy frequently falls short of evidence-based standards. Median time from diagnosis to surgical evaluation exceeds 17 years in many large cohorts, even though randomised data have shown surgery’s superiority over continued medical therapy for two decades.

Three converging developments have reshaped the practical management of drug-resistant epilepsy. Cenobamate has produced seizure-freedom rates that surpass any modern adjunctive antiseizure medication. The Kv7-channel opener XEN1101 has shown a clean phase 2 efficacy signal. Neuromodulation evidence now spans three modalities with long-term follow-up. Together these advances make timely re-evaluation worthwhile in nearly every patient with persistent uncontrolled seizures.

Key Findings on Drug-Resistant Epilepsy

Consideration 1: The ILAE Definition Should Trigger Action, Not Description

The 2010 ILAE consensus defines drug-resistant epilepsy as failure of adequate trials of two tolerated, appropriately chosen, and appropriately used antiseizure medication schedules to achieve sustained seizure freedom. Each subsequent ASM trial in a patient meeting this threshold yields seizure freedom in fewer than 5% of cases on average.

The clinical implication is that the definition is operational rather than descriptive. Once a patient meets criteria, the probability of seizure freedom from another medication trial alone is low — comprehensive evaluation including surgical workup, dietary therapy, and neuromodulation candidacy becomes the appropriate next step rather than continued sequential ASM monotherapy or duotherapy.

Consideration 2: Surgical Evaluation Outperforms Continued Medical Therapy

Wiebe and colleagues randomised 80 patients with temporal lobe epilepsy to anteromesial temporal lobe resection versus continued medical therapy. At one year, 58% of surgical patients were free of disabling seizures versus 8% in the medical arm. The Engel ERSET trial extended this finding to early surgical referral after just two failed ASMs, with 11 of 15 surgical patients seizure-free at two years versus 0 of 23 medically managed patients.

Long-term durability is supported by de Tisi and colleagues, who followed 615 patients post-resection: 47% remained seizure-free at two years and 41% at five years. Despite this evidence, surgical referral remains delayed by a median of 17 to 20 years from diagnosis, representing a major care gap in real-world drug-resistant epilepsy management.

Consideration 3: Cenobamate Has Reset Adjunctive ASM Expectations

The Krauss 2020 phase 3 trial randomised 437 adults with focal-onset drug-resistant epilepsy on stable baseline ASMs to adjunctive cenobamate or placebo. During the 12-week maintenance phase, 28% of patients receiving 200 mg achieved 100% seizure reduction versus 9% on placebo. Median seizure-frequency reduction reached 55% at the 200 mg dose.

Long-term open-label extension data have shown sustained seizure-freedom rates approaching 20% in patients with histories of multiple prior ASM failures — a benchmark not previously achieved by any modern adjunctive agent. Drug-induced reactions including DRESS led to a slow titration protocol, and CYP2C19-mediated interactions require attention to concurrent ASM dosing.

Consideration 4: XEN1101 Demonstrates Clean Phase 2 Efficacy in a Novel Class

The X-TOLE phase 2b trial randomised 325 adults with focal-onset drug-resistant epilepsy to XEN1101 (azetukalner) at 10 mg, 15 mg, or 20 mg daily versus placebo. The 20 mg arm produced a median 52.8% reduction in monthly seizure frequency versus 18.2% on placebo, with a clear dose response across all three active arms.

XEN1101 is a selective Kv7.2/7.3 potassium-channel opener — the same target hit by retigabine but with improved selectivity and pharmacokinetics. Phase 3 trials in focal-onset and generalised tonic-clonic seizures are underway. If confirmed, this would represent the first new mechanism added to first-line drug-resistant epilepsy management since perampanel.

Consideration 5: Neuromodulation Provides Evidence-Based Palliation

Three neuromodulation modalities have randomised support. Vagus nerve stimulation has the longest evidence base, with approximately 43% of patients achieving at least 50% seizure reduction at three years in long-term follow-up. Responsive neurocortical stimulation (RNS) produced a 37.9% seizure reduction versus 17.3% with sham at 12 weeks in the pivotal Heck trial.

Deep brain stimulation of the anterior thalamic nucleus (DBS-ANT) showed 40.4% reduction at three months in the Fisher pivotal trial, with long-term Salanova data demonstrating 69% median reduction at five years. Across all three modalities, seizure-freedom rates remain low (typically 5–15%), positioning neuromodulation as durable palliation rather than cure within the drug-resistant epilepsy treatment hierarchy.

What the Evidence Does Not Show

Despite three decades of progress, several practical questions in drug-resistant epilepsy remain unanswered. Each gap below identifies a specific evidence limitation rather than a generic call for more research.

Evidence Grade & Bottom Line