Melanotan II: The Documented Safety Signals

The documented safety concerns with Melanotan II are dermatological, gastrointestinal, urological and — most consequentially — diagnostic: eruptive and changing melanocytic naevi, melanoma diagnosed in users, nausea and facial flushing, spontaneous penile erection extending in some reports to priapism, and skin darkening that occurs without ultraviolet exposure and can therefore obscure the visual changes by which melanoma is normally caught early 2345. Every one of those signals comes from case reports and narrative reviews. There is no controlled safety programme. The compound was taken as far as a small pilot study in the 1990s and no further 1, and the evidence base reflects that.

What the compound is

Melanotan II is a synthetic cyclic analogue of alpha-melanocyte-stimulating hormone (alpha-MSH), the endogenous peptide that drives pigmentation. Cyclisation confers resistance to enzymatic degradation and potency well above the natural hormone, which is why the original pharmacological literature described it as a superpotent melanotropic peptide 1. It is also — and this is the fact from which most of the rest follows — a non-selective agonist, activating across the melanocortin receptor family rather than at one subtype.

The melanocortin system is not a pigmentation system with incidental other roles. It is five G-protein-coupled receptors distributed across skin, brain, adrenal cortex, immune cells and exocrine tissue, participating between them in pigmentation, energy balance, sexual function, inflammation and cardiovascular regulation. An agonist that does not discriminate between them produces effects in several domains at once. The tanning, the nausea and the spontaneous erections are not separate phenomena that happen to co-occur. They are one pharmacology observed in different tissues.

How the pigmentation happens

MC1R agonism on the melanocyte raises intracellular cyclic AMP, which activates protein kinase A, which phosphorylates the transcription factor CREB, raising expression of microphthalmia-associated transcription factor and in turn of tyrosinase and the other enzymes of melanin synthesis. The result is a shift in the ratio of the two melanins a melanocyte can make — away from red-yellow phaeomelanin, toward brown-black eumelanin — with an increase in total output. Eumelanin is the darker of the two, and the visible consequence is a tan.

The important feature of that pathway is where it begins. Physiological tanning starts with ultraviolet-induced DNA damage in keratinocytes; the damage response drives p53-dependent transcription of pro-opiomelanocortin, cleaved to alpha-MSH, which then acts on the melanocyte. A pharmacological MC1R agonist enters the cascade at its final step and bypasses everything upstream. Pigmentation increases with no ultraviolet exposure and no preceding damage signal. That decoupling is the property the compound is sought for, and also the property that creates the surveillance problem below.

The comparison that matters: afamelanotide

A closely related alpha-MSH analogue took the other path. Afamelanotide is a linear analogue with far greater selectivity for MC1R, developed conventionally: preclinical characterisation, dose-ranging work, controlled trials against defined endpoints, regulatory review. Its indication is erythropoietic protoporphyria, a rare inherited disorder in which accumulated protoporphyrin makes ordinary sunlight acutely painful. It is approved in several jurisdictions for that condition, prescribed by specialists, and covered by post-marketing pharmacovigilance.

The two share a pharmacological ancestor and diverge on two axes. The first is selectivity: MC1R selectivity narrows afamelanotide's effects toward pigmentation, while non-selectivity spreads Melanotan II's across the receptor family. The second is the development pathway, and it may matter more. Regulated development means afamelanotide's adverse events are collected systematically, against a known denominator. Melanotan II has neither pathway nor denominator: nobody knows how many people have been exposed, so no associated adverse event can be expressed as a rate. The point is not that one molecule is good and the other bad. It is that one has been characterised and the other has not.

The clinical record

The human literature on Melanotan II is thin and old. The principal citable work is a pilot phase-I study published in 1996, which evaluated the cyclic peptide in a small number of subjects and confirmed measurable pigmentary change together with the acute effects — nausea, flushing, spontaneous erection — that recur throughout the later literature 1. It was designed to demonstrate pharmacological activity in humans. That is what it did.

Be exact about what a pilot phase-I study can support. It establishes that a compound is active in humans and gives a first impression of acute tolerability. It cannot detect adverse events rarer than roughly one in the number of subjects enrolled, cannot characterise anything emerging over years, and cannot address carcinogenicity — which, for a melanocyte-stimulating compound, is the question that matters most. Development as a licensed medicine did not continue. Everything since known about longer-term effects comes from clinicians writing up patients who arrived in their clinics.

The documented adverse signals

• Eruptive melanocytic naevi — new moles in unusual numbers (case reports).
• Change in existing naevi: darkening, enlargement, altered outline (case reports).
• Melanoma in users, including melanoma in situ (case reports and small series).
• Nausea, sometimes with vomiting (pilot study; review literature).
• Facial flushing (pilot study; review literature).
• Spontaneous penile erection, extending in some reports to priapism (case reports).
• Yawning, drowsiness, transient blood pressure change (review literature; less consistent).
• Degraded visual melanoma surveillance — mechanistic, not an adverse event in the conventional sense.

The best-characterised dermatological signal is eruptive melanocytic naevi. Case reports describe new naevi appearing in numbers and at a pace unusual for the patient's age, and pre-existing naevi darkening, enlarging or changing outline, within weeks to months of exposure beginning 3. Mechanistically this is coherent rather than surprising: melanocortin agonism is a stimulus to melanocytes, and naevi are melanocytic lesions. Visible change in melanocytic lesions is the expected result of the compound's own pharmacology.

The more serious reports concern melanoma diagnosed in people using the compound. These exist in the peer-reviewed dermatological literature as individual cases and small series: melanoma associated with the use of melanotan-II 2, and melanoma in situ arising in a patient with a documented exposure history 5. A narrative review of the risks of unregulated alpha-MSH analogue use collates these alongside the wider signal set 4. The number of published cases is small in absolute terms, and — absent any denominator — uninterpretable as a rate.

The acute systemic effects are the most consistently reported and, in isolation, the least serious. Nausea and facial flushing appear in the pilot human work 1 and throughout the subsequent literature 4. Spontaneous erection is reported reliably enough to be understood as a predictable consequence of central MC4R agonism. Its severe form, priapism, is a urological emergency: sustained erection produces ischaemia in the erectile tissue and can cause permanent damage if unrelieved.

The surveillance problem

This signal deserves separate treatment, because it is the least intuitive and arguably the most consequential. Early melanoma detection is overwhelmingly visual. Clinicians and patients look for a lesion that is asymmetric, irregularly bordered, variegated in colour, larger than its neighbours, or — the single strongest criterion — changing. Every one of those is a comparison: lesion against surrounding skin, lesion against its neighbours, or lesion against its own previous appearance.

A compound that darkens the whole skin surface while darkening and multiplying naevi degrades all of those comparisons at once. Background contrast falls, so a lesion stands out less. Variegation is harder to judge against uniformly deepened pigment. Change, the criterion carrying most weight, becomes ambiguous: a naevus that has darkened may be responding to melanocortin stimulation or may be evolving, and visual examination alone cannot separate the two. New naevi in numbers give both patient and clinician more lesions to track and a ready benign explanation for each.

What case reports can and cannot establish

A case report is a description of what happened to one patient. Its evidential function is precise and limited: it demonstrates temporal association, documents a phenomenon in enough clinical detail for others to recognise it, and generates hypotheses that other designs can test. It does not establish causation. It has no control group, no denominator, and no protection against the possibility that the association is coincidental.

The confounding here is specific and should be stated plainly. People who seek a tanning compound are plausibly enriched, as a group, for melanoma risk factors independent of it: sun-seeking behaviour, sunbed use, high cumulative ultraviolet exposure, and the fair phototypes that make a tan hard to achieve and therefore desirable to pursue. A population at elevated baseline risk produces melanoma cases whether or not any given exposure contributes. Published cases also carry ascertainment bias, since a melanoma in a user of an unlicensed peptide is more publishable than one with no unusual history. None of that argues the signal is spurious. It argues that the case literature alone cannot quantify it.

Having stated the limitation, the corresponding fact must be stated with equal directness. There is no controlled safety data on this compound: no cohort study, no exposure registry, no long-term follow-up, no post-marketing surveillance — because none of the structures that generate those things applies to a compound distributed outside medical regulation. In that situation the case-report and review literature is not merely the weakest available evidence. It is the only evidence that exists. And its direction is uniform: no published reports of naevi regressing, no series describing reassuring long-term outcomes, no counterbalancing literature of any kind.

Two errors are available here and both should be refused. The first is to dismiss the case literature because case reports cannot prove causation — which treats an evidential limitation as a refutation, and awards the benefit of the doubt to a compound never tested. The second is to describe that literature as though it demonstrated the compound causes melanoma, which it does not. The accurate position sits between, and it is not a comfortable midpoint: a plausible mechanism, a consistent and unfavourable set of clinical observations, a confounder that cannot be excluded, and no controlled data capable of settling the question.

What would be required to answer the question

1. A denominator — how many people have been exposed, and for how long. For an unlicensed compound that number does not exist and cannot be reconstructed.
2. A comparison group matched on baseline risk: skin phototype, sunbed history, cumulative ultraviolet exposure.
3. Prospective naevus documentation. Serial dermoscopic imaging separates melanocortin-driven darkening from genuine lesion evolution; retrospective description cannot.
4. Long-term follow-up. Melanoma latency runs to years or decades, and case reports capture only those who have already presented.
5. Systematic adverse-event collection, so uncommon events accumulate into a signal rather than depending on one clinician choosing to publish.

None of those five exists for Melanotan II. All five exist, in some form, for the approved analogue described earlier. That is the whole of the difference between a characterised risk profile and an uncharacterised one, and it is a difference in knowledge before it is a difference in pharmacology.

How to read this evidence base

Melanotan II and afamelanotide began from the same idea: that alpha-MSH signalling could be reproduced with a synthetic analogue. One was narrowed toward a single receptor subtype, tested against defined endpoints, submitted for regulatory review, and is now used under specialist supervision. The other was not narrowed, was not tested beyond a pilot study, and reaches users through channels that generate no data at all.

The consequence is asymmetric knowledge, not only asymmetric risk. For the developed compound the adverse events are enumerated, monitored and expressible as rates. For the undeveloped one they are whatever clinicians happened to write up. That is not a claim that the second is more dangerous by some measured factor, since no such factor can be computed. It is a claim that the question has never been asked in a form capable of answering it, and that the evidence which does exist leans one way.