The short answer on melanotan II and melanoma
There is no controlled human evidence that melanotan II causes melanoma. The entire melanoma literature on it is case reports and small clinic series: people who, while using it, developed crops of new melanocytic naevi, marked darkening of naevi they already had, or a melanoma. A case report establishes that something happened in one person in a particular sequence. It cannot establish how often that happens, whether it happens more often than in comparable non-users, or whether the drug caused it, because there is no denominator and nothing to compare against.
That is not a clean bill of health, and reading it as one is the commonest error made about this compound in either direction. Nothing has ruled the association out. No cohort has been followed, no case-control study run, and no registry captures exposure, so the question is not merely unanswered but unanswerable with the data that exist. What can be said without hedging is that a plausible concern has real published cases attached to it, that no epidemiology capable of measuring it exists, and that the undisputed effects stand regardless: nausea, flushing, spontaneous erections, and a pigment change that makes a person's moles harder to monitor.
What melanotan II is and where it came from
Melanotan II is a synthetic cyclic analogue of alpha-melanocyte-stimulating hormone, developed at the University of Arizona in the 1980s and early 1990s. The goal was photoprotective tanning: trigger the pigment response pharmacologically so the tan arrives without the ultraviolet exposure that normally produces it. Human work stayed small. A pilot phase 1 study in a modest number of volunteers reported skin darkening after repeated subcutaneous administration, along with nausea and spontaneous penile erections. That erectile signal became its own programme and produced bremelanotide, approved by the FDA in 2019 for hypoactive sexual desire disorder in premenopausal women on the strength of two randomised placebo-controlled trials.
Melanotan II itself never advanced past that exploratory stage. It has no completed controlled efficacy trial for tanning, no marketing application, and no approval in any jurisdiction. What circulates under the name today is also not traceably the material from those studies. It is sold by unregulated online sellers, usually as a lyophilised powder with research-use-only labelling, sometimes as a nasal spray, with certificates of analysis that carry no independent verification. The gap between the peptide of a 1990s pharmacology paper and the contents of an anonymous vial is itself part of the safety question, and nobody has measured it.
How a melanocortin agonist darkens skin
Pigment cells carry the melanocortin-1 receptor, which raises intracellular cyclic AMP when alpha-MSH binds it. That signal drives the transcription factor MITF, which drives tyrosinase and the rest of the pigment machinery, shifting production from reddish-yellow pheomelanin toward brown-black eumelanin. People carrying loss-of-function variants of this receptor make relatively more pheomelanin, freckle rather than tan, and sit at higher baseline melanoma risk. That relationship is what the tanning-peptide idea was built on: if weak signalling tracks with poor tanning and higher risk, strong signalling ought to do the opposite.
Melanotan II is not selective. It is a potent agonist across the melanocortin receptor family, hitting the receptors governing appetite and sexual function as well as the pigment receptor. Nor is its pigment effect cosmetically targeted. A systemic melanogenic agonist darkens whatever is already pigment-competent: sun-exposed skin, but also freckles, areolae, scars and existing naevi. Users describe moles going darker as a normal part of the response. That matters for everything below, because the drug's intended action and the phenomenon that raises the melanoma question are the same phenomenon.
The adverse effects that are not in dispute
Set the cancer question aside and much about melanotan II is uncontroversial. Nausea, vomiting and flushing shortly after administration, yawning and stretching, appetite suppression, and spontaneous erections in men appear repeatedly across the published human experience. They are mechanistically coherent central melanocortin effects rather than impurities, and they were visible in the earliest formal study. Beyond that tier sit isolated serious reports, including rhabdomyolysis after injection, typically as single cases. Those are signal generation, not incidence: they show an event is possible in temporal association with exposure and say nothing about frequency.
Layered on top is the product. This is a self-injected or self-sprayed preparation reconstituted outside any sterile environment, with no assured identity, purity, endotoxin limit or sterility, and no channel through which a user's adverse event reaches a regulator. The pharmacovigilance systems that would ordinarily generate a safety signal for a marketed drug never see this compound, because nobody prescribes it and no manufacturer is accountable for it.
What the melanocytic-lesion case literature actually says
The published lesion literature describes a recognisable pattern: eruptive melanocytic naevi, meaning crops of new moles appearing over weeks to months of use; darkening and sometimes enlargement of naevi already present; lesions excised during use and reported as dysplastic; and, less often, melanoma diagnosed in a person who had been using melanotan II, occasionally within months of starting. The reports come from dermatology services in several countries where use became visible enough for clinicians to notice, including the United Kingdom, the Netherlands, Scandinavia and Australia. That similar observations arose independently in different health systems is why the concern is treated as a live question rather than dismissed as internet folklore.
Be precise about what that establishes and what it does not. It establishes that users have presented with new and changing pigmented lesions, and that some of those lesions were histologically confirmed melanoma. It does not establish incidence, because nobody counts how many people use the compound; it does not establish relative risk, because there is no comparison group; and it does not show that melanoma occurs more often in users than in non-users of the same skin type and sun-exposure history. In individual cases it cannot separate four stories: that the drug initiated a malignancy, that it accelerated a lesion already on that path, that it made an existing lesion visible sooner, or that it was simply present while an unrelated melanoma developed.
Formal causality assessment on these cases is weak on nearly every axis. Latency between first exposure and diagnosis is short and inconsistent, which cuts against a straightforward initiation story. Dose and cumulative exposure are unknown, because the products are unregulated and self-administered. And the users described are frequently fair-skinned people with a history of sunbed use and prior sunburn: the population already carrying the highest baseline melanoma risk before any peptide enters the picture. Several reports note concurrent ultraviolet exposure, since pairing injections with sun or sunbeds is common practice.
Why case reports cannot settle causation here
The largest problem is the missing denominator. Some unknown number of people have used melanotan II, and no registry, prescription record or sales figure would let anyone estimate it. Without that number a pile of case reports has no interpretable meaning as a rate. Twenty published melanoma cases among a hundred users would be alarming; the same twenty among several hundred thousand users would be unremarkable against background incidence. The literature cannot tell you which situation you are in.
The second problem is confounding running in the obvious direction. People who seek out a tanning drug are, almost by definition, people who want to be tan, and those are disproportionately people who tan poorly, burn easily and use ultraviolet sources. That phenotype and that behaviour are established melanoma risk factors on their own. Any observed excess among users would have to be separated from an excess that would have been there anyway, and no published analysis has attempted that adjustment because none has had the data to try.
The third is ascertainment. Someone injecting a compound known to darken moles, who has read that it is linked to melanoma, is more likely to present for a skin check, and the clinician knows the exposure before examining the lesion. That raises both detection of genuine lesions and the chance the encounter is published, and publication bias runs one way: no journal prints a report of a user whose moles did not change. Run the standard causal considerations across this evidence and most come back empty. Strength is unmeasured; biological gradient is unmeasured, since neither dose nor duration is known; temporality is suggestive but variable. Only plausibility and coherence score well.
The mechanistic argument runs in both directions
The protective argument is genuine and was the original rationale. Eumelanin absorbs and scatters ultraviolet radiation more effectively than pheomelanin, and melanocortin-1 receptor signalling does more than change pigment ratios: in experimental systems, raising cyclic AMP through this receptor enhances nucleotide excision repair and reduces ultraviolet-induced DNA lesions. Its epidemiological counterpart is the elevated melanoma risk in carriers of weak receptor variants. On that reading a receptor agonist should lower risk, not raise it.
The concerning argument is equally coherent. Alpha-MSH is a growth and differentiation signal for melanocytes, not merely a pigment switch, and sustained supraphysiological melanocortin tone is not a state the pigment system evolved to occupy. The visible clinical consequence, naevi darkening and new naevi appearing, is direct evidence the drug acts on nevus melanocytes rather than only on background pigment, and melanoma cells themselves express the receptor. Neither direction has been tested in humans with a cancer endpoint, and realistically neither will be: a trial powered to detect a change in melanoma incidence would need thousands of participants followed for years, and no ethics committee would approve chronic dosing of an unapproved compound for a cosmetic indication.
The harm that does not depend on causation being proven
Skin cancer detection is built almost entirely on change. Self-examination, clinical review and dermoscopy all work by comparing a lesion against its own past and against its neighbours, which is why the advice is always about moles that are new, growing, changing colour or standing out from the rest. A drug that darkens most of a person's naevi at once and adds new ones degrades all of those comparisons together. That matters more than a small relative risk would, because melanoma outcome is dominated by thickness at diagnosis, so a modest delay in noticing a lesion moves prognosis considerably.
A related point deserves stating plainly, because the marketing gets it backwards: a drug-induced tan is not sunscreen. Even granting that induced eumelanin is genuinely photoprotective, the protection conferred by a tan of any origin is modest and far below what a sunscreen provides, and the confidence a tan encourages tends to run the other way. Many users deliberately combine injections with ultraviolet exposure to bring the colour out faster, which adds ultraviolet dose rather than replacing it.
Afamelanotide: the same biology through an approval pathway
Afamelanotide is the linear alpha-MSH analogue from the same academic lineage, historically called melanotan I. It is a different molecule, preferring the pigment receptor rather than agonising the whole family, and it is delivered as a controlled-release subcutaneous implant rather than injected at whatever interval a user chooses. Both differences matter, but neither is the main point. The main point is that afamelanotide was taken through development, trials and regulatory review, and melanotan II was not.
It was developed not for cosmetic tanning but for erythropoietic protoporphyria, a rare inherited disorder in which accumulated protoporphyrin causes severe burning pain within minutes of light exposure. The phase 3 programme, whose pooled European and United States results appeared in the New England Journal of Medicine in 2015, used time spent in direct sunlight without pain as the endpoint, and the implant increased that time relative to placebo. The absolute differences were modest, and blinding is imperfect in a trial of a drug that visibly tans people. The European Medicines Agency authorised it in 2014 under exceptional circumstances, and the FDA approved it in October 2019 as the first therapy indicated to increase pain-free light exposure in adults with this disease, under orphan drug designation.
The instructive detail sits in that label, which directs full-body skin examination at regular intervals, twice yearly, during treatment. Faced with the same unresolved question about chronic melanocortin stimulation and melanocyte proliferation, and with no demonstrated causal link either, regulators did not treat absence of proof as absence of risk. They approved a narrow indication in a disease with serious unmet need, required monitoring, and required ongoing safety data collection. That is a proportionate response to genuine uncertainty, and a useful benchmark for reading a compound sold for cosmetic use with no monitoring at all.
The regulatory status, stated plainly
Melanotan II is an unapproved new drug. There is no approved indication for it anywhere, no marketing application has been granted, and it cannot lawfully be marketed for tanning in the United States, the United Kingdom, the European Union or Australia. The FDA has issued warning letters to sellers of unapproved melanotan products, and the MHRA has stated repeatedly that melanotan is unlicensed and illegal to sell in the United Kingdom. The research-use-only labelling on what is sold is a commercial device, not a scientific status: it implies nothing about verified identity, purity, sterility or potency, and carries no lot traceability and no route by which an adverse event reaches a safety database.
That last consequence is the one to carry away. The absence of epidemiology on melanotan II and melanoma is not a coincidence or a funding oversight; it follows directly from how the compound is distributed. You cannot run a cohort study on an exposure that leaves no prescription record, no registry entry and no verifiable dose, and you cannot detect a safety signal in a system that never receives reports. The evidence gap is a structural feature of the grey market, and it will not close on its own.