Setmelanotide

A ring-shaped peptide, eight amino acids long, injected under the skin once a day. It is an approved prescription medicine in both the United States (Imcivree, since 25 November 2020) and the European Union (since 16 July 2021), and the list of things it is approved for is short and specific: obesity following damage to the hypothalamus from a brain tumour, injury or its treatment, in people aged 4 and over; and obesity caused by Bardet-Biedl syndrome or by faults in both copies of the POMC, PCSK1 or leptin-receptor gene, in people aged 2 and over. Its own approved label carries a Limitations of Use section stating that it is not indicated for ordinary obesity or for obesity from other genetic syndromes, because it would not be expected to work there. The sequence field is deliberately left empty. The label's own chemical name is acetyl-L-arginyl-L-cysteinyl-D-alanyl-L-histidinyl-D-phenylalanyl-L-arginyl-L-tryptophanyl-L-cysteinamide cyclic (2->8)-disulfide: it contains two amino acids in their mirror-image D form, its ends are capped with an acetyl group and an amide, and the chain is closed into a ring by a sulphur bridge between the second and eighth building blocks. A plain one-letter string would describe a different, straight-chain molecule. PubChem gives C49H68N18O9S2 and 1,117.3 g/mol (CID 11993702, CAS 920014-72-8), which matches the approved product information exactly.

Sold as

Injection under the skin once a day, self-administered or given by a carer, stepped up from a low starting amount over the first weeks. Younger children are dosed by body weight, and the amount is reduced in severe kidney impairment.

Legal status in the EU

An approved medicine across the European Union. Imcivree holds a centrally authorised marketing authorisation valid throughout the EU from 16 July 2021 (EU/1/21/1564, Rhythm Pharmaceuticals Netherlands), and the European Commission's Union Register lists its full indication: obesity and the control of hunger in adults and children aged 4 and over with acquired hypothalamic obesity, and in adults and children aged 2 and over with genetically confirmed Bardet-Biedl syndrome or genetically confirmed loss-of-function faults in both copies of POMC (including PCSK1) or the leptin receptor. It carries EU orphan designations for each of those conditions (POMC deficiency 14 July 2016, leptin-receptor deficiency 19 November 2018, Bardet-Biedl syndrome 21 August 2019, acquired hypothalamic obesity 13 October 2023), and its EU product information was last updated on 27 May 2026. In the United States it has been approved since 25 November 2020 (NDA 213793), with the Bardet-Biedl indication added 16 June 2022, the extension to children aged 2 to 6 on 20 December 2024, and acquired hypothalamic obesity on 19 March 2026. Funding is a separate question from authorisation: England's NICE funds it for the POMC and leptin-receptor conditions, and for Bardet-Biedl syndrome only in people who start treatment between the ages of 6 and 17. It is NOT prohibited in sport — the WADA 2026 Prohibited List does not name setmelanotide, Imcivree, melanocortin or melanotan anywhere in its text, and section S0 cannot capture it because S0 applies only to substances with no current approval by any governmental regulatory health authority.

It switches on MC4R, a receptor on nerve cells in the part of the brain that decides whether the body has had enough to eat, with about 20-fold less activity at the two related receptors MC3R and MC1R. In every condition it is licensed for, that signal is failing for a different reason: the hormone that normally pushes the switch is not being made (POMC and PCSK1 deficiency), the leptin signal that triggers its release cannot be received (leptin-receptor deficiency), the cellular machinery around it is faulty (Bardet-Biedl syndrome), or the brain region containing it has been physically damaged (acquired hypothalamic obesity). The drug pushes the switch directly, which the label says reduces food intake and raises energy use. Activity at MC1R, the receptor pigment cells use, is why most people who take it get darker skin without any sun. Levels peak about 8 hours after an injection, half a dose is gone in about 11 hours, and a steady level is reached within two days, so it has to be taken every day. About 39 per cent of a dose leaves unchanged in urine, which is why the dose is cut in severe kidney impairment, though it is unchanged in mild and moderate impairment.

Part of the bodyWhat happensHow it is knownKind of evidence

Hunger, in the brain

This is the thing it is built to do. MC4R is a switch on nerve cells in the part of the brain that decides whether you have eaten enough. In the conditions this drug is licensed for, that switch is either not being pushed — because the hormone that pushes it is missing or its factory is damaged — or the machinery around it is broken. Setmelanotide pushes it directly, and the constant, starvation-like drive to eat that defines these conditions eases. In the randomised trial in people with a damaged hypothalamus, the daily worst-hunger score on a 0-to-10 scale fell 2.27 points against 1.44 on a dummy injection over a year — a real gap, but a much smaller one than the weight numbers imply.

Measured with a daily diary score in 110 people aged 12 and over inside a randomised, dummy-controlled, 52-week trial, and in smaller uncontrolled groups in the other trials — 14 people with Bardet-Biedl syndrome and 16 people with the single-gene faults.

Measured in people

Source [01]Source [02]

Body weight, and how quickly it comes back

Weight falls, and it falls a long way in the people the drug is licensed for. It also comes straight back when the injections stop. The clearest demonstration of that is a deliberate trick built into the two single-gene trials: after ten weeks of open treatment, participants were secretly switched to a dummy injection for four weeks without knowing. They regained an average of 5.5 kg and 5.0 kg in those four weeks, and their hunger scores went back up; both reversed when the drug was restarted. This is a treatment that has to be continued, not a cure.

Measured in a randomised, dummy-controlled 52-week trial in 142 people with obesity after hypothalamic damage, and in four smaller trials — 44 people with Bardet-Biedl syndrome, 10 with POMC or PCSK1 deficiency, 11 with leptin-receptor deficiency and 12 children aged 2 to 5. The blinded withdrawal was done in the 16 people who had already lost weight in the single-gene trials.

Measured in people

Source [01]Source [04]

Skin, hair and moles

It darkens you, and this is not a rare side effect — it is the expected consequence of how the drug works. MC4R is one of a family of five receptors, and the drug also touches MC1R, the switch pigment cells use to make melanin. In the randomised trial, 58 per cent of people on the drug developed darker skin against 10 per cent on the dummy injection; in the open-label trials the figure ran to 78 and 83 per cent. Existing moles darken and new ones appear: 15 per cent against 6 per cent in the randomised trial. Hair colour changes have been reported too. The label says the pigment change reverses when the drug is stopped, and it requires a full-body skin examination before starting and repeated during treatment, so that a mole that is changing for a different reason is not missed.

Counted in 94 people on the drug and 48 on a dummy injection in the randomised trial, and in a further 43, 27 and 12 people in the open-label trials where there was nothing to compare against.

Measured in people

Source [01]Source [03]

The stomach and gut

Feeling sick is the commonest thing that happens after the pigment change, and unlike the pigment change it does have a large placebo share. In the randomised trial 55 per cent of people on the drug felt sick against 25 per cent on the dummy injection, and 38 per cent vomited against 19 per cent. Constipation ran 12 per cent against 6, diarrhoea was common in the open-label trials, and abdominal pain affected a third of the people in the single-gene trials. Despite all of that, roughly the same proportion of people stopped the drug in each arm of the randomised trial — 10.6 per cent on setmelanotide and 12.5 per cent on the dummy injection.

Counted in 94 people on the drug and 48 on a dummy injection over 52 to 60 weeks, and in the open-label trials of 43, 27 and 12 people.

Measured in people

Source [01]Source [03]

Erections, in men and boys

The same receptor family that this drug pushes for hunger also drives sexual arousal — it is the receptor the approved desire drug bremelanotide works on — so spontaneous erections happen. In the randomised trial they were reported by 7 per cent of people on the drug against 4 per cent on the dummy injection. In the open-label trials, where everyone knew what they were taking, the figures were much higher: 25 per cent of the 20 males with Bardet-Biedl syndrome and 23 per cent of the 13 males in the single-gene trials. The label tells people this can happen and tells anyone with an erection lasting longer than four hours to go to hospital.

Counted in the randomised trial of 142 people, and separately among the 20 and 13 male participants of the two open-label programmes, where there was no comparison group.

Measured in people

Source [01]

The adrenal glands and the body's salt balance

This one applies only to people whose obesity comes from a damaged hypothalamus, and it exists because those people often already have a damaged pituitary gland underneath it. Serious episodes of the adrenal glands failing to keep up were reported by 5 per cent of people on the drug and by nobody on the dummy injection. Among people who also had diabetes insipidus — a condition where the body cannot hold on to water — sodium levels went too low in 6 per cent on the drug against 2 per cent on the dummy injection, and too high in 5 per cent against 4 per cent. Both warnings were added to the US label in March 2026, at the same time as the indication that created them.

Measured inside the randomised, dummy-controlled trial of 142 people with acquired hypothalamic obesity, over 56 to 60 weeks.

Measured in people

Source [01]Source [03]

The immune system

Some people make antibodies against the drug, and rather more make antibodies against the natural hormone it imitates. Across five trials in which people were treated for at least a year, 4 of 245 — 1.6 per cent — had antibodies against setmelanotide itself, and 17 of 245 — 6.9 per cent — had antibodies against alpha-MSH, the body's own version of the signal. That second figure reached 6 of 21 in people with leptin-receptor deficiency. The label states plainly that there is not enough information to say what any of these antibodies do to how well the drug works or how safe it is.

Blood measured in 245 people across five trials, each with at least 52 weeks of exposure.

Measured in people

Source [01]

How long it stays in you

It is a once-a-day injection and it behaves like one. Levels in the blood peak about 8 hours after the injection, half of a dose is gone in about 11 hours, and the amount in the body settles to a steady level within two days of starting. About 39 per cent of a dose leaves unchanged in urine, which is why the dose is cut in severe kidney impairment, where exposure runs roughly 86 to 96 per cent higher; in mild and moderate impairment the label sets the same dose as for normal kidneys. Children carry more of it for the same dose than adults do, which is why the dose is set by age and by body weight.

Measured in dedicated pharmacology studies submitted with the US application and modelled from 109 adults with normal kidney function; paediatric figures come from population modelling rather than direct measurement in every age band.

Measured in people

Source [01]

Mood

This is the most uncertain part of the record and the label treats it seriously. In the two open-label trials in people with the single-gene faults, 26 per cent of the 27 participants reported depression or low mood and 11 per cent reported thoughts of suicide. Those trials had no comparison group at all, so nobody knows how much of that belongs to the drug and how much to living with a severe genetic disease. In the randomised trial, depression did not appear among the reactions that were both common and more frequent than on the dummy injection. The label nonetheless carries a standing warning, tells doctors to watch for new or worsening depression, and says to consider stopping the drug if suicidal thoughts appear.

Counted in 27 people across two open-label trials with no comparison group, plus one serious report of suicidal thinking in the Bardet-Biedl trial that its investigators did not consider drug-related. The randomised trial's table of reactions occurring in 5 per cent or more and more often than on placebo does not include depression.

One small study

Source [01]Source [05]

Approved on one properly controlled trial and four very small ones, and the gap between them is the whole story. The controlled trial (TRANSCEND, NCT05774756) randomised 143 people with obesity following hypothalamic damage — 47 per cent adults, 23 per cent under 12 — and found that over 52 weeks BMI fell 15.84 per cent against a 2.55 per cent RISE on a dummy injection, a placebo-adjusted difference of 18.40 percentage points (95% CI 14.85 to 21.94, p<0.0001). 75.8 per cent of the treated group lost at least 5 per cent of their BMI against 9.7 per cent on the dummy injection, and 50.1 per cent lost at least 15 per cent against 2.2 per cent. The New England Journal report of the same trial covers its 120-person pivotal cohort and gives -16.51 per cent against +3.32 per cent; both figures are correct and describe different analysis sets. Hunger, the ranked secondary measure, moved far less: -2.27 points against -1.44 on a 0-to-10 diary scale (p=0.04), and 61.0 versus 45.0 per cent when counted as responders, with overlapping confidence intervals. The other four pivotal trials are the reason this must not be read as a competitor to the GLP-1 injections. The Bardet-Biedl trial randomised 52 people to 14 blinded weeks against a dummy injection, and which figure is quoted decides what it shows: across everyone randomised, body weight changed -2.4 versus -0.3 per cent, a difference that did not reach statistical significance (-2.1, 95% CI -4.6 to 0.4, p=0.052), while a prespecified exploratory analysis of the 44 people who had Bardet-Biedl syndrome itself gave BMI -4.6 versus -0.1 per cent (difference -4.5, 95% CI -6.5 to -2.5). Everyone then went on the drug, and the headline 52-week figure of -7.9 per cent has no comparison group at all. The POMC/PCSK1 and leptin-receptor trials were open-label and single-arm, with 10 and 11 people in the efficacy analysis, and tested their primary endpoint against an assumed 5 per cent background rate rather than a control arm: 8 of 10 and 5 of 11 reached 10 per cent weight loss at a year. What makes those two convincing instead is a blinded four-week swap to placebo built inside them — participants regained 5.5 kg and 5.0 kg and their hunger scores worsened, both reversing on restarting. The trial that took the licence down to age 2 enrolled 12 children. Eight people with Alstrom syndrome were randomised inside the Bardet-Biedl trial and the published result in that group was inconclusive; Alstrom syndrome is not in the licence on either side of the Atlantic. Outside the licensed conditions the drug has been tested and did little: a randomised, double-blind 28-day study found 3.07 kg lost in ordinary obesity against 0.90 kg GAINED on a dummy infusion, and 3.48 kg against 0.85 kg in carriers of one faulty MC4R gene, a difference that did not reach significance (p=0.09). Serious adverse events were recorded in 26 of 94 people on the drug (28 per cent) against 3 of 48 on the dummy injection (6 per cent), and 2 of the 95 assigned to setmelanotide died against none of the 48 on placebo, with no cause stated in the trial registry or in the approved label. It is graded as an approved medicine because it is approved by both the US and EU regulators with a public product information sheet; one of its five pivotal trials was large and randomised, one was randomised and placebo-controlled for only its first 14 weeks, and three were open-label single-arm trials of 10, 11 and 12 people. The sequence field is null because the molecule is a cyclic octapeptide containing D-alanine and D-phenylalanine with an acetylated N-terminus, a C-terminal amide and a disulfide ring, none of which the standard one-letter code can express.

The drug itself is fast and the effect is slow. Levels peak about 8 hours after an injection, half a dose is gone within about 11 hours, and the amount in the body reaches a steady level within two days — so it has to be taken every day. Getting to the full dose takes 2 to 12 weeks of stepping up, deliberately, because the sickness is worst early. Weight then falls over months rather than weeks: the trials measured their main result at 52 weeks, and the year-long curves show the fall continuing throughout rather than levelling off early. Going the other way is quick: in the two trials that secretly withdrew the drug, participants regained about 5 kg in four weeks and their hunger scores went back up, then both reversed on restarting. Skin darkening appears during treatment and the label says it reverses after stopping.

What was measuredWhat happenedIn whomHow it is known

Body mass index, in obesity after damage to the hypothalamus

Fell 15.84 per cent against a 2.55 per cent RISE on a dummy injection over 52 weeks — a difference of 18.40 percentage points (95 per cent confidence interval 14.85 to 21.94, p<0.0001). The published paper reports the same trial's pre-specified pivotal group slightly differently: −16.51 per cent against +3.32 per cent. Both figures are correct and describe different analysis sets of one trial.

142 adults and children aged 4 to 66 with obesity following a hypothalamic tumour, lesion or injury; 94 on setmelanotide and 48 on a dummy injection, 52 weeks after an escalation period. Nearly half were adults, a quarter were under 12. The published analysis covers the pivotal subgroup of 120.

Measured in people

Source [01]Source [02]Source [03]

How many people lost a meaningful amount, in hypothalamic obesity

At least 5 per cent of their BMI: 75.8 per cent on setmelanotide against 9.7 per cent on a dummy injection. At least 10 per cent: 61.0 against 4.7. At least 15 per cent: 50.1 against 2.2. Half the treated group therefore lost at least 15 per cent of their BMI, and one in fifty of the dummy-injection group did.

The same 142 people, 52 weeks. The 10 and 15 per cent thresholds were not corrected for multiple testing, which the label states.

Measured in people

Source [01]

Hunger, on a 0-to-10 daily diary score

Fell 2.27 points against 1.44 on a dummy injection (difference 0.82, 95 per cent confidence interval 0.02 to 1.62, p=0.04). Counted as responders instead of averages, in the trial's 120-person pivotal group, the picture is weaker: 61.0 per cent of the treated group had a fall of at least 2 points against 45.0 per cent on the dummy injection, and those two ranges overlap. Everyone started around 6.7 to 7.1 out of 10.

110 people aged 12 and over in the randomised trial who could report their own hunger — 74 on setmelanotide and 36 on a dummy injection — over 52 weeks. A quarter of the scores at week 52 were missing and were filled in statistically.

Measured in people

Source [01]Source [03]

Body mass index in Bardet-Biedl syndrome, in the blinded part of the trial

Fell 4.6 per cent against 0.1 per cent on a dummy injection over 14 weeks — a difference of 4.5 percentage points (95 per cent confidence interval 2.5 to 6.5). This 14-week comparison is the only placebo-controlled evidence that exists for this condition; the headline 52-week figure below has no comparison group.

44 people aged 6 and over with a clinical diagnosis of Bardet-Biedl syndrome, 22 on setmelanotide and 22 on a dummy injection, 14 weeks.

Measured in people

Source [01]Source [06]

Serious harm, and deaths

Serious side effects of any kind were recorded in 26 of 94 people on setmelanotide — 28 per cent — against 3 of 48 on the dummy injection, 6 per cent. Two of the 95 people assigned to setmelanotide died and none of the 48 on the dummy injection did. The trial registry states that its all-cause mortality figure covers people on study-drug bridging visits beyond the main comparison period, and neither the registry nor the approved label says what either person died of.

The 143 people randomised in the trial in acquired hypothalamic obesity, over 56 to 60 weeks.

Measured in people

Source [02]Source [03]

Body mass index in Bardet-Biedl syndrome after a full year

Fell 7.9 per cent on average, with 61.3 per cent of people losing at least 5 per cent and 38.7 per cent losing at least 10 per cent — against no comparison group at all, because everyone was on the drug by then. The nearest thing to a control is a separate published analysis that matched 29 trial participants to 58 people from an international Bardet-Biedl registry who were not treated: 58.6 per cent of the treated group met a combined weight target against 6.9 per cent of the matched untreated group. That is an indirect comparison between a trial and a registry, not a randomised one, and a correction to it has since been published.

31 people aged 6 and over with Bardet-Biedl syndrome, 52 weeks from the start of setmelanotide; the five who left early were counted as having changed 0 per cent. The matched comparison used 29 trial participants and 58 registry controls.

One small study

Source [01]Source [15]Source [16]

Weight in POMC or PCSK1 deficiency

8 of 10 people lost at least 10 per cent of their body weight in a year, and the average loss was 23.1 per cent — from 118.7 kg to 89.8 kg. There was no comparison group: the result was tested against an assumption that 5 per cent of untreated people would do the same. What makes it convincing instead is the blinded withdrawal built into the trial — when the drug was silently swapped for a dummy injection for four weeks, participants regained 5.5 kg on average.

10 people aged 6 and over with obesity from a fault in both copies of the POMC or PCSK1 gene, one year, open-label. A further 4 enrolled but had not completed a year at the analysis cut-off.

One small study

Source [01]Source [04]Source [17]

Weight in leptin-receptor deficiency

5 of 11 people — 46 per cent — lost at least 10 per cent of their body weight in a year, and the average loss was 9.7 per cent. Again there was no comparison group and the test was against an assumed 5 per cent background rate. During the blinded four-week withdrawal, participants regained 5.0 kg on average. This is roughly half the effect seen in POMC deficiency, in the same trial design.

11 people aged 6 and over with obesity from a fault in both copies of the leptin-receptor gene, one year, open-label. A further 2 enrolled but had not completed a year at the analysis cut-off.

One small study

Source [01]Source [18]

Body mass index in children aged 2 to 5

Ten of the 12 children reached the trial's target — a drop of at least 0.2 on the age-and-sex-adjusted BMI score — and average BMI fell 18 per cent over a year. Broken down, it was 33.8 per cent in the three children with POMC deficiency, 13.1 per cent in the four with leptin-receptor deficiency and 9.7 per cent in the five with Bardet-Biedl syndrome. There was no comparison group and there were 12 children in total, so the ranges around each of those numbers are wide.

12 children aged 2 to 5 with obesity from POMC deficiency, leptin-receptor deficiency or Bardet-Biedl syndrome, 52 weeks, open-label. No child with PCSK1 deficiency enrolled, although they were eligible.

One small study

Source [01]Source [07]Source [19]

Weight in people with ordinary obesity, and in people carrying one faulty MC4R gene

Small. Over 28 days, people with ordinary obesity lost 3.07 kg on setmelanotide against a 0.90 kg GAIN on a dummy infusion — a difference of 3.97 kg. People carrying one faulty copy of the MC4R gene, which is far commoner than the conditions the drug is licensed for, lost 3.48 kg against 0.85 kg, and that difference did not reach statistical significance (p=0.09). In other words, having the gene fault made them no better off than ordinary obesity did, and neither group came near the 15 to 25 per cent falls seen in the licensed conditions.

A randomised, double-blind, 28-day phase 1b study: 6 carriers of a faulty MC4R gene on the drug and 2 on a dummy infusion, plus 5 people with ordinary obesity on the drug and 3 on a dummy infusion, all by continuous infusion under the skin at 0.01 mg per kg per day.

One small study

Source [01]Source [08]

Weight in Alström syndrome

Nothing could be concluded. Eight people with Alström syndrome — a different rare disease that also disrupts the same signalling pathway — were enrolled in the Bardet-Biedl trial and randomised alongside them. The published paper states the results in that group were inconclusive, and Alström syndrome is not in the approved licence on either side of the Atlantic. This is the most useful negative result in the whole programme: the same drug, the same trial, a neighbouring disease, and no answer.

8 people with Alström syndrome inside a trial that randomised 52 people in total, of whom 44 had Bardet-Biedl syndrome.

One small study

Source [01]Source [05]

Skin darkening is the dominant effect and it is the mechanism showing on the outside: 58 per cent of people against 10 per cent on a dummy injection in the randomised trial, and 63, 78 and 83 per cent in the three open-label trials, affecting skin, gums, nails, freckles and hair colour, with new or darkening moles in 15 per cent against 6 per cent. The label calls the pigment change reversible on stopping but requires a full-body skin examination before treatment and repeatedly during it. Feeling sick affected 55 per cent against 25 per cent on the dummy injection and vomiting 38 against 19; headache 37 against 31, dizziness 12 against 4, constipation 12 against 6, injection-site reactions in 21 of 94 against 11 of 48, the same rate in each group. Spontaneous erections were reported by 7 per cent against 4 per cent in the randomised trial and by about a quarter of male participants in the open-label trials, with the label telling anyone with an erection lasting over four hours to seek emergency care. Depression or low mood was reported by 26 per cent and suicidal thinking by 11 per cent of the 27 people in the two open-label single-gene trials, which had no comparison group; the label carries a standing warning and tells doctors to consider stopping. In people whose obesity follows hypothalamic damage there are two condition-specific risks added to the label in March 2026: serious acute adrenal insufficiency in 5 per cent against nobody on the dummy injection, and sodium going too low (6 versus 2 per cent) or too high (5 versus 4 per cent) in those who also have diabetes insipidus. Serious hypersensitivity reactions including anaphylaxis have occurred and are the only outright contraindication. Serious adverse events overall ran 28 per cent against 6 per cent, and two people in the setmelanotide arm died against none on placebo, with no cause given. It should be stopped when pregnancy is recognised and is not recommended while breastfeeding.

EffectWhat it feels likeHow often

The adrenal glands failing to keep upSerious

This only affects people whose obesity comes from hypothalamic damage, because that damage often takes the pituitary gland with it, and the pituitary is what tells the adrenal glands to make cortisol. Those people are usually already on replacement steroids. Losing weight fast can unmask a shortfall that was previously balanced, and an untreated adrenal crisis is a medical emergency. The label tells doctors to watch for it in anyone with known secondary adrenal insufficiency.

5 in 100 people with a damaged hypothalamus, against nobody on a dummy injection.

Source [01]

Severe allergic reactionsSerious

They generally happen within minutes to hours of an injection. This is the drug's only outright contraindication: anyone who has had a serious hypersensitivity reaction to setmelanotide or to anything else in the vial must not have it again.

Rare, but reported: one anaphylactic reaction among the 94 people treated in the randomised trial, and further cases after the drug went on sale.

Source [01]Source [03]

Depression and thoughts of suicideSerious

The drug acts on the brain, and the label carries a standing warning that drugs which do so may cause depression or suicidal thinking. Doctors are told to monitor for new or worsening mood changes and to consider stopping the drug if suicidal thoughts or persistent depression appear. Whether the rates above reflect the drug or the disease is genuinely unresolved — living with severe genetic obesity carries its own risk, and the randomised trial in hypothalamic obesity did not show depression more often on the drug than on the dummy injection.

Depression or low mood in about 26 in 100 people, and suicidal thoughts in about 11 in 100, in the two open-label trials of 27 people. Neither had a comparison group, so there is no placebo figure to set beside them.

Source [01]

Sodium going too low or too highSerious

Diabetes insipidus is a condition where the body cannot hold on to water, and it is common in people whose hypothalamus has been damaged. Changing how much someone eats and drinks changes their salt balance, and severe sodium swings cause confusion and seizures. The label tells doctors to check sodium when fluid intake changes and to adjust the medicines used for the water problem.

Among people who also have diabetes insipidus: too low in 6 in 100 on the drug against 2 in 100 on a dummy injection; too high in 5 in 100 against 4 in 100.

Source [01]

Darker skin, and new or darkening moles

This is the mechanism showing up on the outside, not an unrelated reaction: the drug also pushes the receptor pigment cells use, so melanin production rises whether or not you go in the sun. Skin, gums, nails, freckles and hair colour have all been reported. The label says the pigment change reverses after stopping, and requires a full-body skin check before starting and repeatedly during treatment — partly to track the drug's own effect, and partly so a mole changing for some other reason is not lost in it. A published case report describes a crop of unusual moles appearing on treatment.

58 in 100 people against 10 in 100 on a dummy injection in the randomised trial; 63, 78 and 83 in 100 in the three open-label trials, which had nothing to compare against. New or darkening moles, 15 in 100 against 6 in 100.

Source [01]Source [20]

Feeling sick and vomiting

Both are common on the dummy injection too, which is worth holding on to — roughly half of the sickness reported on this drug was reported by people who were not taking it. In the open-label trials sickness affected about a quarter to a half of participants. It did not drive people out of the randomised trial: about the same proportion stopped in each arm.

Feeling sick, 55 in 100 against 25 in 100 on a dummy injection. Vomiting, 38 in 100 against 19 in 100.

Source [01]

Reactions where the needle goes in

Redness, itching, hardening, bruising and pain at the site. It is a daily injection, so the sites accumulate. There is no gap over the dummy injection: the rate is the same in both groups, which is why this reaction does not appear in the label's table of reactions that were more frequent than placebo.

21 of 94 on the drug and 11 of 48 on a dummy injection in the randomised trial — about 22 in 100 in each group. In the open-label single-gene trials it was 96 in 100, with nothing to compare against.

Source [01]Source [03]

Spontaneous erections

Unwanted and unprompted erections, including in boys, because the receptor family being pushed is the one that drives sexual arousal as well as appetite. The label tells patients this may happen and to seek emergency care for an erection lasting more than four hours.

7 in 100 people against 4 in 100 on a dummy injection in the randomised trial; about 1 in 4 of the male participants in the open-label trials.

Source [01]

Headache, dizziness and constipation

The small stuff. Headache is the clearest example of why a comparison group matters: it looks alarming at 37 per cent until you notice that nearly a third of the people on an injection with nothing in it reported it too.

Headache 37 in 100 against 31 in 100 on a dummy injection; dizziness 12 against 4; constipation 12 against 6.

Source [01]

These are the amounts the studies below gave their participants, and they are here as facts about those studies. They are not a recommendation, not a starting point and not a range to pick from. What a person should take, if anything, is a question for a doctor who knows them.

TRANSCEND, the only large randomised placebo-controlled trial — obesity after damage to the hypothalamus, in adults and children aged 4 and over

Up to 3 mg injected under the skin once a day, reached by stepping up from a low starting dose. Children under 6 were dosed by body weight. A matching dummy injection was given to the comparison group.

An escalation period of up to 8 weeks followed by 52 weeks at the full dose; 56 to 60 weeks in total.

Source [01]Source [03]

The Bardet-Biedl and Alström syndrome trial — 14 weeks against a dummy injection, then a year in which everyone had the drug

3 mg under the skin once a day, titrated over the first two weeks of each period so that nobody could tell from the dose whether they had been on the dummy injection.

14 blinded weeks followed by 52 open weeks; 66 weeks in total.

Source [05]Source [06]

The two single-gene trials — POMC or PCSK1 deficiency, and leptin-receptor deficiency — each with a hidden four-week placebo swap inside it

Stepped up over 2 to 12 weeks to a personal maintenance dose, then 10 open weeks. Anyone who had lost at least 5 kg went into 8 blinded weeks: four on the drug and four on a dummy injection, without knowing the order. Then up to 32 more open weeks.

About one year in each trial.

Source [01]Source [17]Source [18]

VENTURE, the trial that took the licence down to age 2 — children aged 2 to 5 with the single-gene faults or Bardet-Biedl syndrome

0.5 mg under the skin once a day to start, raised by 0.5 mg every two weeks to a ceiling set by the child's weight — 0.5, 1, 1.5 or 2 mg. Open-label; every child had the drug.

8 weeks of stepping up followed by 44 weeks; 52 weeks in total.

Source [07]Source [19]

The 28-day study in ordinary obesity and in carriers of a single faulty MC4R gene — the one that tested whether this could be a general obesity drug

0.01 mg per kg of body weight per day, given as a continuous infusion under the skin rather than a daily injection, against a matching dummy infusion.

28 days.

Source [08]

What people report

• Why two of the 95 people assigned to setmelanotide in the largest trial died and none of the 48 on the dummy injection did — neither the trial registry nor the approved label states a cause, and the registry's own note says its mortality count reaches beyond the blinded comparison period.
• Whether it does anything useful in Alström syndrome — 8 people with it were randomised inside the Bardet-Biedl trial and the published result in that group was inconclusive, and no further trial has been run.
• Whether it helps people carrying one faulty MC4R gene rather than two faulty POMC, PCSK1 or LEPR genes — the only randomised test lasted 28 days and its result did not separate from the dummy infusion. The trial designed to answer this, EMANATE, randomised 296 people and has posted no results.
• How much of the 26 per cent depression and 11 per cent suicidal thinking seen in the single-gene trials belongs to the drug — those trials had no comparison group at all, and the disease itself carries that risk.
• Whether the benefit lasts beyond one year, because every controlled comparison in the programme ends at 14 or 52 weeks and everything after that was open-label.
• Whether it changes anything a person would notice besides weight and hunger — the label reports "general numeric improvements" in blood pressure, cholesterol and blood sugar and then says in as many words that the effects could not be accurately quantified.
• Whether the new and darkening moles it causes ever become dangerous, and what years of pushing pigment cells does to melanoma risk — the trials were far too small and too short to tell, and the requirement for repeated skin examinations exists precisely because nobody knows.
• Whether the antibodies people make against the drug, and the antibodies 1 in 4 people with leptin-receptor deficiency made against the body's own version of the hormone, do anything at all — the label states there is not enough information to say.
• What it does to a human pregnancy: there are no data in pregnant women, weight loss during pregnancy can itself harm a baby, and pregnant rabbits given it had more resorptions and post-implantation losses at the two higher of the three doses tested, alongside significant toxicity to the mothers.
• Whether it passes into human milk — it is present in rat milk, which the label says makes human milk likely, and it is therefore not recommended while breastfeeding on the basis of no measurement at all.
• What it does in anyone under 2, and whether people aged 65 and over respond differently — the label says the trials did not include anyone aged 65 or over.
• Whether the drug or the disease explains the very different responses between conditions: 23 per cent average weight loss in POMC deficiency, 10 per cent in leptin-receptor deficiency and 8 per cent in Bardet-Biedl syndrome, all in trials too small to compare with each other.
• How the effect compares with the drugs most people mean when they say weight-loss injection — no trial has ever compared setmelanotide with semaglutide or tirzepatide, and none is registered.
• Whether the hunger benefit is as large as the weight benefit — the average diary score moved 0.82 points more than on the dummy injection, and on a responder count the ranges for the two groups overlapped.
• Whether people who never start, or stop early, look like the people in the published patient surveys — every one of those studies asked people who were already on treatment, and both were funded by the manufacturer.
• How many people worldwide actually have these conditions, and how many of them are ever diagnosed — the drug's licence depends on a genetic test that most people with severe obesity never receive.
• Whether it is worth what it costs: England's health technology body funded it for the single-gene conditions but restricted Bardet-Biedl funding to people starting between the ages of 6 and 17, and Canada's assessment put the cost per quality-adjusted life year above two million Canadian dollars.

Does setmelanotide work?

For the specific conditions it is licensed for, yes, and by a lot. In the one large randomised trial — in people whose obesity followed damage to the hypothalamus — BMI fell about 16 per cent over a year while it rose about 3 per cent on a dummy injection, and half the treated group lost at least 15 per cent of their BMI against one in fifty on the dummy. In the rarer single-gene conditions the trials were tiny and mostly had no comparison group, but they contained a hidden four-week switch to a dummy injection, and people regained about 5 kg in those four weeks. Outside those conditions there is no evidence it does much at all.

Source [01]Source [02]

Who can get setmelanotide, and what is it approved for?

It is approved in the United States and across the European Union for four things and nothing else: obesity after the hypothalamus has been damaged by a tumour, injury or its treatment, in people aged 4 and over; and obesity caused by Bardet-Biedl syndrome or by faults in both copies of the POMC, PCSK1 or leptin-receptor gene, in people aged 2 and over. The genetic conditions have to be confirmed by a genetic test, and the label says explicitly that it is not indicated for ordinary obesity, for obesity from other genetic syndromes, or for people whose gene changes have been graded harmless.

Source [01]Source [09]

Setmelanotide vs semaglutide — which is better for weight loss?

They are not alternatives and no trial has ever compared them. Semaglutide is licensed for ordinary obesity in millions of people; setmelanotide is licensed for a handful of rare conditions in which the brain's fullness signal is specifically broken, and its own label says it would not be expected to work for common obesity. A search of the US trials registry on 3 August 2026 for studies naming both returned exactly one, a withdrawn surgery study that listed both only as background medication. If you do not have one of the named conditions, setmelanotide is not the drug being discussed.

Source [01]Source [21]

Does setmelanotide work for normal obesity?

Barely. It has actually been tested: in a randomised, double-blind 28-day study, people with ordinary obesity lost 3.07 kg on setmelanotide while the dummy-infusion group gained 0.90 kg. People carrying one faulty copy of the MC4R gene — a much commoner situation than the licensed conditions — lost 3.48 kg against 0.85 kg, a difference that did not reach statistical significance. Nothing there resembles the 15 to 25 per cent falls seen in the licensed conditions, and the approved label states the drug would not be expected to be effective in general obesity.

Source [01]Source [08]

What are the side effects of setmelanotide?

The commonest by far is your skin going darker — 58 in 100 people against 10 in 100 on a dummy injection — along with new or darkening moles, which is why a full skin examination is required before and during treatment. Feeling sick affected 55 in 100 against 25 in 100, and vomiting 38 against 19. Spontaneous erections happen in men and boys. In people whose obesity follows hypothalamic damage there are two further risks the label spells out: the adrenal glands failing to keep up, in 5 in 100 against nobody on the dummy injection, and sodium levels swinging in those who also have diabetes insipidus. Serious side effects of any kind were recorded in 28 per cent of the treated group against 6 per cent on the dummy injection.

Source [01]Source [03]

Is setmelanotide legal?

Yes, as a prescription medicine, in both the United States and the European Union. It was approved in the US on 25 November 2020 and in the EU on 16 July 2021, and the licence has been widened several times since — most recently in March 2026, when obesity after hypothalamic damage was added in the US. It is not a controlled substance and it is not banned in sport. Getting hold of it legally means a prescription from a specialist and, in practice, a genetic test or a documented hypothalamic injury.

Source [10]Source [11]

Can you buy setmelanotide?

Only on prescription, and only for the licensed conditions. Unlike several compounds in this register, it does not appear on the US regulator's list of substances nominated for pharmacies to compound, so there is no legitimate pharmacy route around the licence. It is also a difficult molecule to copy — a ring closed by a chemical bridge, with two mirror-image amino acids in it — which is not a guarantee that nothing is sold online under the name, only that anything so sold is unverified. This register does not link sellers and does not tell anyone how to use anything.

Source [01]Source [22]

How much does setmelanotide cost?

It is one of the most expensive medicines any health service buys. England's health technology body records a list price of £2,376.00 for a single 10 mg vial excluding VAT, with a confidential discount negotiated on top; the adult maintenance dose is 3 mg every day. A published review of the economic evaluations found Canada's assessment put the cost per quality-adjusted life year above two million Canadian dollars, far beyond what health systems normally pay, while England's assessments came out more favourably. England funds it for the single-gene conditions, and for Bardet-Biedl syndrome only in people who start treatment between the ages of 6 and 17.

Source [14]Source [23]Source [24]

Does setmelanotide make your skin darker?

Yes, in most people, and this is the mechanism rather than a side effect that came out of nowhere. The drug pushes a family of five related receptors, and one of them is the switch pigment cells use to make melanin — so skin, gums, nails, freckles and even hair colour can darken without any sun. It happened to 58 in 100 people in the randomised trial against 10 in 100 on a dummy injection, and to 78 and 83 in 100 in the open-label trials. The label says it reverses after stopping, and requires a full-body skin examination before treatment and repeatedly during it, because new and darkening moles also occur.

Source [01]

Is setmelanotide the same as melanotan or PT-141?

They are cousins, and the differences matter more than the resemblance. All three are small ring-shaped copies of the body's own pigment hormone and all three touch the same family of receptors, but setmelanotide is tuned towards the appetite receptor MC4R, is approved for obesity in rare genetic conditions, and has published trials behind it. PT-141 is approved in the United States only for low sexual desire in women. Melanotan II has never been approved anywhere and is sold for tanning with no controlled trial behind it. Skin darkening is the effect all three share, which is why it comes up on every one of these pages.

Source [01]Source [13]

Is setmelanotide banned in sport?

No. The 2026 World Anti-Doping Agency prohibited list does not name setmelanotide, Imcivree, melanocortin or melanotan anywhere in its text, and it does not belong to any of the listed classes. The catch-all category for unapproved substances cannot capture it either, because that category applies only to substances with no approval from any government health authority, and this one has been approved in the United States since 2020 and in the European Union since 2021.

Source [11]Source [12]

What happens if you stop taking setmelanotide?

The weight comes back, and quickly. Two of the trials tested this on purpose: after people had lost weight, they were secretly switched to a dummy injection for four weeks without knowing. They regained an average of 5.5 kg and 5.0 kg in that month, and their hunger scores went back up; both reversed when the drug was restarted. The skin darkening, by contrast, is described in the label as reversing after stopping. This is a treatment that has to be continued for as long as it is working.

Source [01]Source [04]

How long does setmelanotide take to work?

Months, not weeks, for the weight. Getting to the full dose alone takes between two and twelve weeks of stepping up, deliberately, because the sickness is worst early on. All the trials measured their main result at 52 weeks and the year-long curves keep falling rather than flattening early. The drug itself works far faster than that: it peaks in the blood about 8 hours after an injection and reaches a steady level in the body within two days.

Source [01]

1. [01]IMCIVREE (setmelanotide) injection, US prescribing information revised March 2026 — indications and limitations of use, warnings 5.1-5.6, adverse reaction tables 5-8 (skin hyperpigmentation 58% vs 10%, nausea 55% vs 25%), section 11 structure acetyl-L-arginyl-L-cysteinyl-D-alanyl-L-histidinyl-D-phenylalanyl-L-arginyl-L-tryptophanyl-L-cysteinamide cyclic (2->8)-disulfide, section 12 pharmacology and immunogenicity, section 14 tables 9-18 (2026)
2. [02]Miller and colleagues, setmelanotide for the treatment of acquired hypothalamic obesity (N Engl J Med 2026;395:138-150, TRANSCEND) — 120 participants in the pivotal cohort, BMI -16.5% vs +3.3% on placebo (p<0.001), hunger -2.73 vs -1.45 (p=0.009), serious adverse events 28% vs 8% (2026)
3. [03]TRANSCEND registry results (NCT05774756, 143 randomised, 142 dosed, pivotal cohort 120, results posted) — primary outcome, adverse events by arm, serious adverse events 26 of 94 vs 3 of 48, deaths 2 vs 0 (2026)
4. [04]Clement and colleagues, efficacy and safety of setmelanotide in individuals with severe obesity due to LEPR or POMC deficiency: single-arm, open-label, multicentre, phase 3 trials (Lancet Diabetes Endocrinol 2020;8:960-970) — 8 of 10 and 5 of 11 reached 10% weight loss (2020)
5. [05]Haqq and colleagues, setmelanotide in patients with Bardet-Biedl syndrome and Alstrom syndrome: randomised, double-blind, placebo-controlled phase 3 trial with an open-label period (Lancet Diabetes Endocrinol 2022;10:859-868) — 32.3% of trial patients aged 12+ reached 10% weight loss at 52 weeks, all of them BBS patients; results inconclusive in Alstrom syndrome — free full text (corrected: Lancet Diabetes Endocrinol 2023;11(2):e2) (2022)
6. [06]Bardet-Biedl and Alstrom syndrome trial (NCT03746522, 52 randomised — 44 with Bardet-Biedl and 8 with Alstrom, completed, results posted) (2018)
7. [07]Argente and colleagues, setmelanotide in patients aged 2-5 years with rare MC4R pathway-associated obesity (VENTURE), Lancet Diabetes Endocrinol 2025 — 12 children, 10 met the co-primary endpoint, mean BMI change -18% (2025)
8. [08]Collet and colleagues, evaluation of a melanocortin-4 receptor agonist (setmelanotide) in MC4R deficiency (Mol Metab 2017) — randomised double-blind 28-day phase 1b, table 1: -3.07 kg vs +0.90 kg on placebo in ordinary obesity, -3.48 kg vs -0.85 kg in MC4R heterozygotes (p=0.09) — free full text (2017)
9. [09]Union Register of medicinal products, European Commission — Imcivree, EU/1/21/1564, setmelanotide, Rhythm Pharmaceuticals Netherlands, with the full EU indication text (downloaded and searched 3 August 2026) (2026)
10. [10]Imcivree, European Medicines Agency medicine overview — marketing authorisation valid throughout the EU from 16 July 2021, orphan designations for all four indications, product information last updated 27 May 2026 (2026)
11. [11]Drugs@FDA, NDA 213793 — original approval 25 November 2020, efficacy supplements approved 16 June 2022, 20 December 2024 and 19 March 2026, sponsor Rhythm (2026)
12. [12]WADA World Anti-Doping Code International Standard, Prohibited List 2026 — full text downloaded and searched 3 August 2026: no occurrence of setmelanotide, Imcivree, melanocortin or melanotan; S0 covers only substances with no current approval by any governmental regulatory health authority (2026)
13. [13]PubChem CID 11993702, setmelanotide — C49H68N18O9S2, 1,117.3 g/mol, CAS 920014-72-8, IUPAC name showing the disulfide ring and the D-alanine and D-phenylalanine residues
14. [14]NICE highly specialised technologies guidance HST21, setmelanotide for treating obesity caused by LEPR or POMC deficiency (6 July 2022) — list price GBP 2,376.00 per 10 mg/ml vial excluding VAT, with a confidential commercial arrangement (2022)
15. [15]Setmelanotide in Bardet-Biedl syndrome: a 52-week comparison of phase 3 trial participants with a matched registry cohort (Obesity) — 58.6% vs 6.9% responders, propensity-score matched against the CRIBBS registry (2026)
16. [16]Correction to "Setmelanotide in Bardet-Biedl Syndrome: A 52-Week Comparison of Phase 3 Trial Participants With a Matched Registry Cohort" (Obesity) (2026)
17. [17]POMC deficiency obesity trial (NCT02896192, 15 enrolled, single-arm open-label, completed, results posted) (2017)
18. [18]LEPR deficiency obesity trial (NCT03287960, 15 enrolled, single-arm open-label, completed, results posted) (2018)
19. [19]VENTURE trial (NCT04966741, 12 enrolled, completed, results posted) (2022)
20. [20]Eruptive dysplastic melanocytic nevi induced by setmelanotide (International Journal of Dermatology) — a published case report (2025)
21. [21]ClinicalTrials.gov API v2 search for studies naming both setmelanotide and semaglutide: totalCount 1, a withdrawn study with 0 participants (checked 3 August 2026)
22. [22]FDA, bulk drug substances nominated for use in compounding under section 503A, downloaded 3 August 2026 — no occurrence of setmelanotide (2026)
23. [23]NICE highly specialised technologies guidance HST31, setmelanotide for treating obesity and hyperphagia in Bardet-Biedl syndrome — recommended only if treatment starts between ages 6 and 17 (2024)
24. [24]A systematic literature review of economic evaluations of setmelanotide (European Journal of Clinical Pharmacology) — Canadian ICERs above CAD 2 million per QALY (2026)