Cenobamate linked to higher seizure control in hard-to-treat status epilepticus

Cenobamate, an antiseizure drug approved for drug-resistant focal epilepsy, was associated with high rates of seizure control in refractory status epilepticus — a life-threatening state of continuous or repeated seizures — in a multicentre cohort study published in Neurology [s1]. The signal is encouraging, but the study's own authors grade it as low-tier evidence, and the framing matters as much as the numbers [s1].

What status epilepticus is, and why it is hard

Status epilepticus is a neurological emergency. When it fails to respond to first- and second-line drugs it becomes refractory (RSE), and when it persists despite anaesthetic treatment it is termed super-refractory (SRSE) — stages that carry rising risks of brain injury and death. Options at that point are limited, so clinicians have been reaching for newer antiseizure medications off their usual label, cenobamate among them [s1].

What the study did

Researchers ran a multicentre retrospective cohort study drawing on a prospectively maintained registry, including adults with focal-onset RSE and SRSE treated between January 2022 and October 2025 [s1]. The primary outcome was sustained termination of status epilepticus — lasting more than 24 hours with electroclinical resolution — attributable to cenobamate [s1]. A secondary analysis used propensity-score matching (1:2) to compare cenobamate-exposed patients against matched, non-exposed controls from the same registry spanning 2011–2025, applying target-trial-emulation principles [s1].

Thirty-one patients received cenobamate (mean age 61 ± 21 years; 54·8% male), and 45·2% met criteria for the more severe super-refractory form [s1]. The drug was started a median of four days (IQR 2–12) after onset, after failure of a median of five other antiseizure medications (IQR 4–5), at a median starting dose of 50 mg per day (IQR 25–50) [s1].

What it found

Sustained resolution of status epilepticus occurred in 26 patients (83·9%), including all 17 of 17 with the refractory form and 64·3% of those with super-refractory disease [s1]. Where it worked, it worked reasonably quickly, with a median time to response of two days (IQR 1–6) [s1].

In the matched comparison — 28 cenobamate-exposed patients against 56 controls — resolution was higher with the drug: 82·1% versus 57·1% (p=0·02) [s1]. In a multivariable analysis, cenobamate exposure was independently associated with resolution, with an adjusted odds ratio of 5·9 (95% CI 1·2–30, p=0·03) [s1]. That confidence interval is wide, stretching from a small effect to a very large one, which is a direct consequence of the study's modest size.

On safety, no serious cenobamate-related adverse events occurred; mild events were seen in 51·6%, most commonly somnolence [s1]. In-hospital mortality was 22·6%, with no deaths attributed to the drug [s1].

Two details in the treatment pattern are worth drawing out. The drug was typically started late — a median of four days into the episode and only after five other medications had failed — so these were not early, easily treated cases but entrenched ones in which the usual tools had already been exhausted [s1]. And the doses used were modest, with a median starting dose of 50 mg per day, suggesting the observed responses did not require aggressive escalation [s1]. The relatively quick median response of two days, where it occurred, is the kind of timing that matters in an emergency where every additional day of seizures raises the risk of lasting harm [s1]. None of this proves cause and effect, but it sketches a plausible clinical picture that a trial could test. It is also consistent with the drug's known behaviour in drug-resistant focal epilepsy, where meaningful control has been reported at relatively low doses — the rationale the authors give for trying it in this setting at all [s1].

What it means

The crucial caveat is in the paper itself: it provides Class IV evidence — the weakest tier, reflecting an uncontrolled, retrospective design with no randomisation [s1]. Propensity matching narrows but cannot eliminate the differences between patients who were and were not given the drug, and clinicians may have chosen cenobamate for cases they judged more likely to turn around. An accompanying editorial asks directly whether this represents an efficacious therapy for refractory status epilepticus, treating the results as a promising lead that now needs prospective testing rather than a settled answer [s2].

So the honest reading is cautious optimism. In a desperate clinical situation with few good options, cenobamate was linked to sustained seizure control in most patients and to better outcomes than matched controls, without serious safety signals [s1]. But association is not proof, and a 31-patient cohort cannot establish that the drug caused the improvement. Whether cenobamate earns a formal place in treating status epilepticus will depend on the randomised trials this study is explicitly calling for.