

Xenon Pharmaceuticals hit pause on its phase 3 depression trials after patients experienced psychosis, a nightmare side effect for a drug designed to treat the mind. The stock cratered 25%, analysts are split, and the race to fix depression just got a lot more complicated.
Imagine you're developing a drug to treat depression, and it starts causing psychosis. That's not a minor speed bump. That's the car driving off a cliff.
Xenon Pharmaceuticals hit pause on enrollment in its phase 3 depression trials this week after patients experienced neuropsychiatric adverse events, including a small number of cases classified as psychosis. The company says the events were rare, mild to moderate, and reversible. Wall Street wasn't comforted. Shares dropped roughly 25% in after-hours trading.
The drug is called azetukalner, and until this week, it was one of the more promising new approaches to treating major depressive disorder (MDD) and bipolar depression. Now it's fighting for its life.
Xenon voluntarily paused new enrollment in two late-stage trials: X-NOVA2 (for major depression) and X-Ceed (for bipolar depression). Patients already in the studies will keep taking the drug, but no new participants are being added.
The side effects included confusion, speech difficulties, and motor coordination problems. A small subset of patients experienced full-blown psychosis. According to Reuters, the psychosis rate was hovering around 1% or under. That sounds tiny, but in a depression trial, any psychosis signal is a five-alarm fire.
Xenon's chief medical officer described the events as short-lived and reversible, with no long-term consequences reported. The company also stressed that these events are consistent with how the drug works. Which, if you think about it, is both reassuring and deeply concerning at the same time.
Azetukalner doesn't work like your typical antidepressant. It's not an SSRI or an SNRI. Instead, it opens Kv7 potassium channels in the brain. Think of these channels like pressure valves on overexcited neurons: when they're open, they calm things down by reducing electrical activity.

Bristol Myers Squibb just killed a $100 million drug program before its trial even finished, and the fallout goes way beyond one company's balance sheet. The world's only clinical-stage degrader-antibody conjugate is now dead, leaving an entire emerging drug class without proof it works in humans.


Join thousands of biotech professionals who start their day with our free, daily briefing.
The idea is elegant. Depression involves certain brain circuits that are too active, and azetukalner turns down the volume. The problem? Turn it down too far, and you get exactly what Xenon saw: confusion, coordination issues, and in rare cases, psychosis. It's like adjusting a thermostat; a few degrees too cold and you've got a whole different problem.
Notably, none of these psychosis events appeared in the earlier phase 2 study (called X-NOVA), which enrolled 168 patients. That's a red flag in itself. When side effects show up for the first time in phase 3, it usually means one of two things: the larger patient population is revealing risks that smaller trials missed, or the dosing regimen needs serious rethinking.
Xenon says it's exploring dose adjustments to improve tolerability, which suggests the company believes this is a dosing problem rather than a fundamental flaw in the drug's mechanism.
The investment community didn't reach a consensus on this one, and the divergence is telling.
Deutsche Bank downgraded Xenon from Buy to Hold, slashing its price target from $90 to $46. That's almost a 50% cut, which signals serious concern about the entire psychiatry franchise. The uncertainty around the safety signal, combined with likely development delays, was enough to pull the plug on their bullish thesis.
H.C. Wainwright took a different view, keeping a Buy rating with a $74 price target. Their argument: the pause is concerning but manageable, and the long-term opportunity for azetukalner remains intact if the dose can be dialed in.
So who's right? The honest answer is nobody knows yet. The path forward depends entirely on whether Xenon can find a dose that treats depression effectively without occasionally causing psychosis. That's a narrow therapeutic window, and threading it is notoriously difficult in CNS (central nervous system) drug development.
Xenon says it has notified the FDA and consulted its Data Safety Monitoring Board (an independent group that watches over patient safety during trials). The company is calling this a temporary, precautionary pause rather than a formal clinical hold.
That distinction matters. A voluntary pause means the company is getting ahead of the problem. A formal FDA clinical hold is much worse; it means the agency has decided the trial poses an unreasonable risk and must stop until the sponsor proves otherwise. If the FDA were to impose a hold, Xenon would need to demonstrate that its fixes adequately address the safety concern before enrollment could restart.
For now, Xenon is still in control of the timeline. But if additional psychosis events emerge in patients already enrolled, that could change quickly.
Azetukalner isn't competing in a vacuum. The depression drug landscape in 2025-2026 is packed with novel approaches, all chasing the same massive unmet need: the roughly one-third of patients who don't respond well to existing antidepressants.
Neurocrine launched its phase 3 program for osavampator in January 2025, with five studies enrolling. Johnson & Johnson is running a head-to-head phase 3 study of seltorexant, which blocks a sleep-related receptor. Compass Pathways has a psychedelic-derived therapy in pivotal trials for treatment-resistant depression. The field is moving fast, and a prolonged pause could cost Xenon its competitive position.
Xenon had been expecting topline data from X-NOVA2 in the first half of 2027. Any delay to that timeline gives rivals more room to establish themselves.
Xenon's situation is the classic tightrope of CNS drug development. The same mechanism that makes azetukalner potentially effective at calming overactive brain circuits can, at the wrong dose, push patients into dangerous territory. A 1% psychosis rate might sound small, but for a drug that millions of depression patients could eventually take, it translates to thousands of people experiencing a terrifying side effect.
The bull case rests on dose optimization: find the sweet spot where the drug works without causing neuropsychiatric harm. The bear case is simpler: maybe this mechanism just doesn't have a wide enough safety margin for a chronic condition like depression.
Xenon called the pause temporary. The market, with that 25% selloff, is betting it might be something more. We'll be watching closely to see which side is right.
Novartis killed its ALS drug lifonebart after a Phase 2 trial missed every endpoint, adding to a decades-long streak of industry failures against one of neuroscience's most stubborn diseases. But the company's next move tells a different story.