Carbetocin (intranasal)
Intranasal carbetocin, a long-acting oxytocin receptor agonist
Written by Aaron CuhaReviewed Sep 2026
Also known as: LV-101, ACP-101, Carbetocin nasal spray
A failed programme, reported as one. Both phase 3 trials in Prader-Willi syndrome missed their primary endpoint, the confirmatory trial returned a treatment difference of 0.3 points on hyperphagia with P = 0.79 and a point estimate favouring placebo, and the open-label extension was terminated by the sponsor for futility.
Overview
This entry exists because intranasal carbetocin is frequently cited as evidence that oxytocin-pathway drugs work on behaviour. The trials it is cited from failed.
Carbetocin is an oxytocin analogue with greater oxytocin receptor selectivity and a longer half-life than oxytocin itself. The intranasal formulation was developed for hyperphagia in Prader-Willi syndrome, on the hypothesis that the oxytocinergic system is one of the dysfunctional circuits in that condition. The injectable carbetocin used for postpartum haemorrhage is a different product with a different route and is not what these trials studied.
The programme ran in three stages and each one is public.
A phase 2 trial, NCT01968187, enrolled 38 participants and was published in 2018, sponsored by Ferring Pharmaceuticals.
The first phase 3, CARE-PWS, NCT03649477, sponsored by Levo Therapeutics, randomised 130 participants aged 7 to 18 to 9.6 mg, 3.2 mg or placebo three times daily over an 8-week placebo-controlled period. Its primary endpoints were hyperphagia on the HQ-CT scale and obsessive-compulsive symptoms on CY-BOCS, for 9.6 mg against placebo. The paper's own words: the primary endpoints showed numeric improvements which were not statistically significant. Only the 3.2 mg arm, tested as a secondary comparison, showed nominally significant improvements. Enrolment was stopped early because of the COVID-19 pandemic.
The confirmatory phase 3, COMPASS PWS, NCT06173531, sponsored by ACADIA Pharmaceuticals, was designed specifically to confirm that 3.2 mg signal. It randomised 175 participants 1:1 to carbetocin 3.2 mg three times daily (n=85) or placebo (n=90) for 12 weeks. The least squares mean change from baseline at week 12 on HQ-CT was -4.8 on carbetocin and -5.1 on placebo. The treatment difference was 0.3, 95% confidence interval -1.8 to 2.4, P = 0.79. The published paper states there was no separation between carbetocin and placebo for any secondary or exploratory endpoint.
The open-label extension, NCT06420297, enrolled 168 participants and was terminated. The sponsor's stated reason, retrieved verbatim from the registry, is that the primary efficacy endpoint was not met in the parent study and that continued administration was not considered likely to provide significant clinical benefit.
Mechanism of action
The oxytocin receptor agonism is established pharmacology. Carbetocin binds the oxytocin receptor with greater selectivity and a longer duration than oxytocin, which is the whole point of the molecule. What was proposed rather than demonstrated is the step after that: that agonising the oxytocin receptor would reduce hyperphagia in Prader-Willi syndrome. Two adequately conducted phase 3 trials tested that proposition and it did not hold. The mechanism is intact; the clinical hypothesis built on it failed.
Human evidence
Two phase 3 randomised placebo-controlled trials, 305 randomised participants between them, and a terminated open-label extension of 168. The evidence quality is high. The result is negative.
- COMPASS PWS (NCT06173531, ACADIA): 175 randomised, carbetocin 3.2 mg three times daily (n=85) versus placebo (n=90), 12 weeks. HQ-CT change -4.8 versus -5.1, treatment difference 0.3, 95% CI -1.8 to 2.4, P = 0.79. No separation on any secondary or exploratory endpoint (PMID 42486751).
- CARE-PWS (NCT03649477, Levo): 130 randomised aged 7 to 18, three arms, 8-week placebo-controlled period. Both primary endpoints, HQ-CT and CY-BOCS at 9.6 mg versus placebo, showed numeric improvement that was not statistically significant. Enrolment stopped early for COVID-19 (PMID 36633570).
- Phase 2 (NCT01968187, Ferring): 38 participants, COMPLETED, published 2018 as a positive result (PMID 29925684).
- Open-label extension (NCT06420297, ACADIA): 168 participants, TERMINATED, sponsor citing the parent study's failed primary endpoint.
- Safety was not the problem. The most common adverse event in CARE-PWS was mild to moderate flushing; in COMPASS, headache in 7.2% and pyrexia in 6.0% of carbetocin participants.
What this does not tell you: The limits here run the opposite direction from most entries on this site: this is well-conducted evidence and the question is whether it was given a fair chance. CARE-PWS stopped enrolling early because of the pandemic, which cost it power. The 3.2 mg comparison it produced was a secondary analysis in a trial that had already missed its primaries, so it needed confirmation. COMPASS PWS was designed to provide exactly that confirmation, was adequately powered, and returned a point estimate favouring placebo. Nothing in the record suggests a dose, duration or population that was left untested and might have worked.
Reading the research record
This page exists because of how this programme gets cited rather than because of what it found.
The CARE-PWS paper is titled Intranasal Carbetocin Reduces Hyperphagia, Anxiousness, and Distress in Prader-Willi Syndrome. Read only the title and you would conclude the drug worked. Read the results and the primary endpoints were not met. The title describes a nominally significant secondary comparison in a dose arm that was not the primary comparison, in a trial that stopped enrolling early.
That is how a failed trial becomes a citation. Titles travel; results sections do not.
Then the confirmatory trial ran, which is exactly what the field is supposed to do, and it returned P = 0.79 with placebo ahead on the point estimate. Then the sponsor terminated its own open-label extension and said in the registry why: the parent study missed its primary endpoint and continued dosing was not considered likely to provide significant clinical benefit.
The correct summary of intranasal carbetocin is that the oxytocin hypothesis for hyperphagia in Prader-Willi syndrome was tested properly and did not hold. That is a useful result. It is not a promising one, and anyone presenting this programme as evidence that oxytocin-pathway agonists improve behaviour is citing the title of a trial that missed.
A cognitive endpoint is a measured one: a validated test of working memory, processing speed, attention, executive function or episodic recall, administered before and after, against a control group who did not know which arm they were in. None of the compounds in this batch has ever been tested that way. Rating scales of fatigue, global clinical impression, anxiety or hyperphagia are not cognitive endpoints, and neither is a hormone level. This matters because the marketing for several of these compounds describes focus, clarity and mental energy, and the underlying studies measured none of those things. CY-BOCS, which CARE-PWS used as a co-primary, measures obsessive-compulsive symptoms rather than cognitive performance, and it is the nearest thing to a cognitive-adjacent endpoint anywhere in this batch. It failed.
The evidence, charted
Fig. 1 · evidence composition
4of 4 citations (100%) are in people
Every citation cited here is Human work; no other study type is cited on this page. This count is of our own citation list and understates any literature larger than the sources we cite.
Fig. 2 · evidence over time
Evidence spans 3 distinct years, 2018 to 2026, counted from the citation list on this page.
Fig. 3 · legal status at a glance
US
Not approved
UK
Not approved
AU
Not approved
CA
Not approved
Not approved in any of the four jurisdictions shown. A jurisdiction's classification is a regulatory fact, not a verdict on the science; see Legal status below for the full text and any notes.
Fig. 4 · dose response
No human dose response curve exists
We draw this figure where the data supports it. For this compound in humans it does not, so the panel stays empty rather than borrowing an animal curve and implying it transfers.
Key studies & citations
- Human2026
Carbetocin Nasal Spray for the Treatment of Hyperphagia in Prader-Willi Syndrome: Results From the Randomized, Placebo-Controlled, Phase 3 Compass PWS Study
The decisive trial. 175 participants randomised 1:1 to carbetocin 3.2 mg three times daily (n=85) or placebo (n=90) over 12 weeks. Least squares mean change from baseline at week 12 on HQ-CT was -4.8 (SE 0.8) on carbetocin and -5.1 (SE 0.8) on placebo; treatment difference 0.3, 95% CI -1.8 to 2.4, P = 0.79. The authors state there was no separation between carbetocin and placebo for any secondary or exploratory endpoint. Most frequent treatment-emergent events on carbetocin were headache (6 participants, 7.2%) and pyrexia (5, 6.0%). The authors' own conclusion is that carbetocin nasal spray did not demonstrate efficacy compared with placebo. Four of the ten authors are employees and stakeholders of ACADIA Pharmaceuticals, the sponsor.
Clinical Therapeutics - Human2023
Intranasal Carbetocin Reduces Hyperphagia, Anxiousness, and Distress in Prader-Willi Syndrome: CARE-PWS Phase 3 Trial
The trial the optimistic title comes from, and the one that did not meet its primary endpoints. 130 participants aged 7 to 18 at 24 ambulatory clinics, randomised to 9.6 mg, 3.2 mg or placebo three times daily over an 8-week placebo-controlled period. Primary endpoints were HQ-CT hyperphagia and CY-BOCS obsessive-compulsive symptoms for 9.6 mg against placebo, and in the paper's own words they showed numeric improvements which were not statistically significant. The 3.2 mg arm showed nominally significant improvements on HQ-CT, PADQ and CGI-C, a secondary comparison. Enrolment was stopped prematurely because of the COVID-19 pandemic. Most common adverse event was mild to moderate flushing. Several authors are employees of Levo Therapeutics, the sponsor.
The Journal of Clinical Endocrinology and Metabolism - Human2018
Intranasal carbetocin reduces hyperphagia in individuals with Prader-Willi syndrome
The phase 2 study that launched the programme, a randomised controlled trial in 38 participants sponsored by Ferring Pharmaceuticals, registered as NCT01968187 and listed as COMPLETED. Its positive result is the origin of the whole development path, and it did not survive replication at phase 3 twice.
JCI Insight - Human2026
OLE Study of Carbetocin Nasal Spray for the Treatment of Hyperphagia in Prader-Willi Syndrome
The open-label extension, 168 participants actual, status TERMINATED. Sponsor ACADIA Pharmaceuticals. The registry's whyStopped field, verbatim: primary efficacy endpoint was not met in the parent study (ACP-101-302), and based on review of available data, continued administration of investigational product in ACP-101-303 was not considered likely to provide significant clinical benefit. A sponsor stopping its own extension study is the clearest statement available about what the programme found.
ClinicalTrials.gov, NCT06420297
Frequently asked questions
Did intranasal carbetocin work for Prader-Willi syndrome?
No. Both phase 3 trials missed their primary endpoint. The confirmatory trial, COMPASS PWS, compared carbetocin 3.2 mg three times daily against placebo in 175 participants over 12 weeks and found a treatment difference of 0.3 points on hyperphagia, 95% CI -1.8 to 2.4, P = 0.79, with no separation on any secondary or exploratory endpoint. The sponsor then terminated the open-label extension.
Why do people cite the CARE-PWS trial as positive?
Because of its title, which says carbetocin reduces hyperphagia, anxiousness and distress. The results section says the primary endpoints were not statistically significant, and the positive-sounding finding is a nominally significant secondary comparison in the 3.2 mg arm of a trial that stopped enrolling early because of the pandemic. The trial designed to confirm that comparison did not confirm it.
Is this the same as the carbetocin used in childbirth?
No. Injectable carbetocin for preventing postpartum haemorrhage is a separate product with a different route and a different indication, and it is approved in several countries. None of the Prader-Willi trial results apply to it, and none of its approvals apply to the nasal spray.
Was it unsafe?
Tolerability was not the issue. The most common adverse event in CARE-PWS was mild to moderate flushing; in COMPASS the most frequent events on carbetocin were headache in 7.2% and pyrexia in 6.0%. The drug was well tolerated and did not work.
Is intranasal carbetocin still in development?
There is no active path in this indication. The confirmatory phase 3 failed and the sponsor terminated the open-label extension study, stating in the registry that continued administration was not considered likely to provide significant clinical benefit.