Xenon Pharmaceuticals submitted the azetukalner NDA to the US Food and Drug Administration on 17 September 2026. The filing covers focal seizures in epilepsy. The same announcement disclosed a voluntary pause on new enrolment in the company’s depression studies. Xenon said the pause followed an analysis of neuropsychiatric adverse events in those studies. The release published no event rates, severity grades or adverse event terms. Executives supplied some of that detail on an investor call the same day.
What the azetukalner NDA covers
The company issued its 17 September announcement after the US market closed on Thursday 17 September 2026. The azetukalner NDA seeks approval as an adjunctive treatment for focal seizures. It rests on two studies, the Phase 2b X-TOLE trial and the Phase 3 X-TOLE2 trial. Xenon described both as global, randomised, double-blind and placebo-controlled. It puts the whole epilepsy programme at more than 1,500 patient-years of safety and exposure data.
The release then turned to psychiatry. Xenon said it had “implemented a voluntary pause on enrollment for new patients”. The pause covers its ongoing studies in major depressive disorder and bipolar depression. Participants already randomised continue on study, as do those in the associated open-label extensions.
Xenon described the pause as a precautionary measure taken in consultation with its Data Safety Monitoring Board. It said the pause is expected to be temporary. It also said the action does not affect its ongoing epilepsy studies. The Phase 3 X-TOLE3 and X-ACKT epilepsy studies continue to enrol.
Enrolment in X-NOVA2, the Phase 3 depression study, has closed at about 360 patients. That is roughly 80 per cent of its initial 450-patient target. Xenon said the study remains powered to show a clinically meaningful change on its primary endpoint. Topline data are now expected in the first quarter of 2027.
Ian Mortimer, President and Chief Executive Officer, called the submission “a significant milestone for Xenon”. Chris Kenney, Chief Medical Officer, said the company remains “very confident in the product profile of azetukalner” in epilepsy. For psychiatry, he said Xenon wants to evaluate whether modifying the dose regimen could improve tolerability. Both remarks come from the company announcement.
What the azetukalner NDA release and call disclosed
The release
One sentence carried the safety analysis behind the pause. Xenon said the events “are consistent with the known safety and tolerability profile of azetukalner and its mechanism”. It added that such events “had not previously been seen in the Phase 2 X-NOVA clinical study”. X-NOVA was the company’s Phase 2 trial in major depressive disorder.
The two clauses read awkwardly together on the page. The release asserts consistency with a known profile, yet concedes the events are new to depression patients. The release itself did not explain the difference.
The investor call
Executives addressed that point on an investor call later on 17 September, as reported by BioSpace and Reuters. Kenney said X-NOVA had shown a “benign” safety profile, with fewer adverse events than the epilepsy trials. Executives said the much larger Phase 3 depression population had made rare events noticeable.
The events described on the call included confusion, speech difficulties and problems with motor coordination. BioSpace reported the terms as confusion, aphasia and ataxia. Executives said a “very small number” of cases fell under the broad term of psychosis. Kenney said the rate of psychosis was “hovering around 1% or under”.
Kenney said the events were mostly mild or moderate. He described them as short in duration and reversible, without long-term sequelae. Xenon is evaluating changes to dosing, which it believes may be contributing given the dose-dependent nature of the events.
Psychosis in the epilepsy trials behind the azetukalner NDA
The new events arose in the depression studies, not in the epilepsy programme. Psychosis has, however, appeared in the epilepsy dataset at a low level. BioSpace reported one psychosis case in each of X-TOLE and X-TOLE2. Analysts noted the side effect had also been seen in healthy volunteers in an early Phase 1 study.
X-TOLE and X-TOLE2 are the two studies supporting the azetukalner NDA now before the FDA. Those individual cases therefore sit within the submitted epilepsy dataset. LSDN draws no conclusion about how reviewers will weigh them.
No regulator has acted. The FDA has not placed a clinical hold, and the pause was the company’s own decision. Mortimer said the FDA has been notified of the psychiatry plans. A voluntary pause binds only the sponsor.
A remark from eight days earlier is relevant here. Chief Financial Officer Tucker Kelly spoke at the Wells Fargo healthcare conference on 9 September 2026. He said X-TOLE2 had not been enriched for patients with depression. Nor did it use the psychiatric screening and assessment of a depression trial. The epilepsy dataset was therefore not designed to characterise risk in a depressed population.
Kelly also called azetukalner “the only potassium channel modulator” in epilepsy. He said adverse event rates at the 10 mg dose were very comparable to placebo in the Phase 2b study. He confirmed the company would file that month, which it did.
What the first Kv7 opener’s label recorded
Azetukalner is not the first Kv7 potassium channel opener licensed for epilepsy. Ezogabine, sold as Potiga, reached the US market and was later discontinued. Its FDA prescribing information carried a Neuro-Psychiatric Symptoms warning.
In placebo-controlled trials, that label recorded confusional state in 9 per cent of patients on Potiga. The placebo rate was 3 per cent. Psychosis appeared in 1 per cent against none on placebo. Hallucinations appeared in 2 per cent against under 1 per cent. The label called the effects dose related, and said they generally emerged within the first eight weeks.
About two thirds of patients with psychosis in the controlled trials had no prior psychiatric history. Half of those who stopped Potiga because of hallucinations or psychosis required hospitalisation. In the vast majority of patients, psychiatric symptoms resolved within seven days of stopping the drug. LSDN searched DailyMed on 18 September 2026 and found no current ezogabine label listed.
This is a different molecule and a separate dataset, and the two were never designed for comparison. It does, however, give documented content to the reference to “its mechanism” in the azetukalner NDA release.
Xenon has addressed the class’s other historical liabilities directly. Its April 2026 release on the X-TOLE2 data reported no retinal pigment epithelium or macular abnormalities. That finding covered the double-blind period. It also reported no notable weight gain, severe allergic rashes or notable cardiovascular events. That release did not address neuropsychiatric events, and LSDN draws no inference from the omission.
The X-TOLE2 topline gives the tolerability picture the azetukalner NDA package rests on. Dizziness affected 20.5 per cent across both azetukalner doses, against 3.2 per cent on placebo. Discontinuations for adverse events reached 14.5 per cent on 25 mg against 3.2 per cent on placebo. Serious adverse events were 5.6 per cent on 25 mg against 2.4 per cent.
A second Kv7 setback inside two weeks
The azetukalner NDA disclosure follows a separate Kv7 event. On 4 September 2026 the FDA placed a partial clinical hold on Biohaven’s opakalim. The hold followed a metabolite found in rodent studies. LSDN reported the opakalim partial clinical hold, and the SK Biopharmaceuticals licence that preceded it.
Two Kv7 assets have now met safety-linked interruptions within a fortnight. The causes are unrelated on the published record. William Blair reported that Xenon management does not believe the two situations are related. LSDN asserts no common mechanism.
UK exposure and the access question
Xenon’s release said nothing about European or UK filings. LSDN searched on 18 September 2026 and found no announced MHRA or EMA application for azetukalner. Kelly said on 9 September that Xenon does not intend to commercialise outside the US itself. He said the company is open to regional or global partners.
The epilepsy programme has UK ethics approval on record. The Health Research Authority records a favourable opinion for X-ACKT, the Phase 3 study in primary generalised tonic-clonic seizures. The East of England Cambridgeshire and Hertfordshire Research Ethics Committee gave that opinion on 1 September 2023. The record carries reference 23/EE/0082 and IRAS number 1007246.
LSDN found no HRA research summary for the paused depression studies, and draws no conclusion from that. A UK launch would land in a crowded tier. NICE technology appraisal TA753, published on 15 December 2021, recommends cenobamate as an add-on for focal onset seizures. It applies to adults whose epilepsy is not adequately controlled after at least two antiseizure medicines.
TA753 also requires that treatment is started by a healthcare professional with expertise in epilepsy. A new Kv7 opener would be arguing for space in that same second-line add-on tier. That is an LSDN observation about the pathway, not a NICE position on azetukalner.
Market reaction and analyst views
Accounts of the share move differ by measurement window. Reuters reported a fall of more than 25 per cent in after-hours trading on 17 September. BioSpace reported a 23 per cent fall to 44.27 US dollars in premarket trading on 18 September. That compared with a close of 57.35 US dollars on 17 September. LSDN publishes no single figure.
Analysts kept their epilepsy estimates. William Blair wrote on 18 September that the aftermarket move looked “overblown”. It cited psychosis incidence on the labels of established antidepressants. Jefferies put epilepsy sales above 2 billion US dollars. Both are projections, and neither is an LSDN estimate.
Kelly also gave pricing guidance at the September conference. He said Xenon aims to price at a premium to branded epilepsy drugs. He put those at about 15,400 and 18,000 US dollars a year. He described a sales force of more than 75 representatives, targeting roughly 7,000 to 8,000 physicians.
What to watch on the azetukalner NDA
The agency’s filing decision is the next milestone. Xenon’s release gave no expected action date, and LSDN found no acceptance announcement as at 18 September 2026. Any action date will appear in LSDN’s tracker of FDA approval decisions.
Jacqueline A. French is Professor in the Department of Neurology at NYU Grossman School of Medicine. She chaired the X-TOLE2 steering committee. In Xenon’s March 2026 topline announcement, she welcomed the X-TOLE2 data. She said they reinforce the value azetukalner and its KV7 mechanism “may bring to the treatment armamentarium for focal seizures”. Her assessment predates the neuropsychiatric disclosure.
Three questions follow for the azetukalner NDA review. The first is whether the agency seeks further neuropsychiatric data. The second is whether Xenon resumes psychiatry enrolment after the dosing changes Kenney described. The third is whether the company, or a partner, files in the UK or Europe.
