A new kind of antipsychotic: what the Cobenfy trial found
Since the 1950s, every antipsychotic has blocked dopamine receptors. Cobenfy takes a different route.
Published 3 min read
This article explains a study published in The Lancet in 2024. Read the original paper
What the study found
- Symptoms fell further on xanomeline–trospium than on placebo, by 9.6 points on a standard 30-item scale.1
- Weight gain and movement side effects were similar to placebo. Nausea and other digestive effects were more common.1
- The trial lasted five weeks. Long-term effects, and comparisons with existing medicines, are still unknown.
The study at a glance
Original paper- Study
- EMERGENT-2
- Published in
- The Lancet, 2024
- Design
- Randomised, double-blind, placebo-controlled trial (phase 3)
- Participants
- 252 adults in hospital with worsening schizophrenia
- Where
- 22 sites in the United States
- Length
- 5 weeks
- Main result
- PANSS score fell 21.2 points on the drug and 11.6 on placebo: a difference of 9.6 points
- Funded by
- Karuna Therapeutics, the drug’s developer
- Evidence
- StrongHow evidence is rated
Clinical summaryFor health professionals
- One capsule, two drugs: xanomeline with trospium chloride, in three strengths: 50 mg/20 mg, 100 mg/20 mg and 125 mg/30 mg. By weight, xanomeline makes up 71–83% of the two (US label).
- Xanomeline: muscarinic M1 and M4 receptor agonist. It does not block dopamine D2 receptors, unlike every antipsychotic approved before it (Kaul 2024).
- Trospium: muscarinic antagonist that stays outside the brain. It reduces xanomeline’s side effects in the rest of the body (Kaul 2024).
- Approved in the US in September 2024 as Cobenfy, for adults with schizophrenia. In the trial it was called KarXT (FDA 2024; Kaul 2024).
- EMERGENT-2: 252 adults in hospital with acute psychosis and a PANSS score of 80 or higher, at 22 US sites; 5 weeks, against placebo (Kaul 2024).
- Trial dose, twice a day: xanomeline 50 mg with trospium 20 mg on days 1–2, then 100 mg with 20 mg; from day 8, optionally 125 mg with 30 mg (Kaul 2024).
- PANSS total fell 21.2 points on the drug and 11.6 on placebo, from about 98 at the start: a difference of 9.6 points at week 5 (Kaul 2024).
- More common than on placebo: constipation 21% vs 10%, nausea 19% vs 6%, indigestion 19% vs 8%, vomiting 14% vs 1%, high blood pressure 10% vs 1% (Kaul 2024).
- Similar to placebo: movement side effects 0% vs 0%, akathisia 1% vs 1%, weight gain 0% vs 1%, sleepiness 5% vs 4%. Stopped because of side effects: 7% vs 6% (Kaul 2024).
- US label: not for patients with urinary or gastric retention, moderate or severe kidney or liver disease, or untreated narrow-angle glaucoma. It can cause urinary retention and a faster heart rate (FDA 2024).
- Europe: not approved in the EU as of November 2026; the maker plans the UK first. EU trials: one recruiting adolescents aged 13–17 in Romania, against placebo; one on relapse prevention in adults, not yet recruiting in the EU. Both due to finish in 2029 (ClinicalTrials.gov).
Why a new mechanism matters
Every antipsychotic (opens in a new tab) approved before 2024 blocks the dopamine D2 receptor (opens in a new tab).1,2 Most block it strongly; clozapine blocks it only weakly.3 These medicines help many people with hallucinations (opens in a new tab) and delusions (opens in a new tab). But not everyone responds, and side effects such as weight gain, stiffness and restlessness lead many people to stop taking them.4
Cobenfy combines two drugs in one capsule, with more xanomeline than trospium.5 Xanomeline stimulates muscarinic receptors (opens in a new tab) in the brain, which influence dopamine signalling indirectly. Trospium blocks the same receptors in the rest of the body but barely enters the brain, so it eases side effects such as nausea without cancelling xanomeline’s effect.1
In the trial, the combination was still called KarXT.1 The US Food and Drug Administration approved it for adults with schizophrenia in September 2024, under the brand name Cobenfy.6
What the trial did
The trial, called EMERGENT-2, enrolled 252 adults who were in hospital with a recent worsening of symptoms, at 22 sites in the United States.1 Half took xanomeline–trospium twice a day; half took a placebo (opens in a new tab). Neither the participants nor the staff knew who was taking which: the trial was double-blind (opens in a new tab).
Symptoms were measured with the PANSS (opens in a new tab), a 30-item interview that trained clinicians use to score hallucinations, delusions, withdrawal and other symptoms.
What they found
After five weeks, symptoms had improved more on xanomeline–trospium than on placebo, by 9.6 points on the PANSS.1 Both positive symptoms (opens in a new tab), such as hallucinations, and negative symptoms (opens in a new tab), such as low motivation, improved more than on placebo.
The most common side effects were digestive: nausea, indigestion, constipation and vomiting. High blood pressure was reported in 10 in 100 people on the drug, against 1 in 100 on placebo. Weight gain and movement side effects, which are common with dopamine-blocking medicines, were similar to placebo. About as many people stopped because of side effects on the drug as on placebo: 7 in 100 against 6 in 100.1
The US prescribing information adds that Cobenfy can make it hard to pass urine and can speed up the heart. It is not prescribed to people who already have trouble emptying the bladder or the stomach, who have moderate or severe kidney or liver disease, or who have untreated narrow-angle glaucoma.6
In Europe
As of November 2026, Cobenfy is not approved in the European Union. Its maker has said the United Kingdom will be the first country in Europe to offer it, with a launch planned for 2026.7
In the EU, the maker has two trials in people with schizophrenia. One is recruiting teenagers aged 13 to 17 at hospitals in Romania, to test it against placebo.8 The other will test whether it prevents relapse in adults; its hospitals in Bulgaria, Czechia, Poland, Romania and Spain have not started recruiting yet.9 Both are due to finish in 2029.
The European Group for Research in Schizophrenia, with 15 university hospitals and research institutes including UMC Groningen and UPC KU Leuven, planned its own trial of the medicine’s effect on memory and thinking. It withdrew the trial when an ethics committee did not approve it.10
What we don’t know yet
- How well it works over years, not weeks.
- How it compares head to head with existing antipsychotics.
- Whether it helps people whose symptoms did not improve on other medicines.
_2D_SKELETAL.png){kind=link}