Melanotan-1 (Afamelanotide)

Chemical identity

The duplication here is cosmetic. PubChem carries the molecule twice: CID 16197727 under Afamelanotide, and CID 16154396 under the proprietary name of the authorised implant product. Both entries return molecular formula C78H111N21O19, molecular weight 1646.8, and the same InChIKey, UAHFGYDRQSXQEB-LEBBXHLNSA-N. One substance, two names. The FDA Global Substance Registration System resolves it under UNII QW68W3J66U with CAS 75921-69-6, ChEMBL2108301, ChEBI 136034, DrugBank DB04931, ATC code D02BB02 and INN number 9010. Its names array lists MELANOTAN, MELANOTAN I and MELANOTAN-1 among the registered names, together with the development codes CUV1647 and CUV-1647. EF-157 sits in the codes array as a USAN designation, not a registered name.

The sequence verifies to Ac-Ser-Tyr-Ser-Nle-Glu-His-D-Phe-Arg-Trp-Gly-Lys-Pro-Val-NH2: thirteen residues, acetylated at the N-terminus and amidated at the C-terminus. It departs from human α-melanocyte-stimulating hormone at two positions, with norleucine replacing methionine at position 4 and D-phenylalanine replacing L-phenylalanine at position 7. Sawyer and colleagues described in 1980 how those two substitutions were arrived at. Heat-alkali treatment of α-MSH produced prolonged activity, and gas chromatography of the hydrolysed product attributed that to partial racemisation at the methionine-4 and phenylalanine-7 positions.

The FDA prescribing information describes the drug substance as afamelanotide acetate, C78H111N21O19 multiplied by between three and four molecules of acetic acid, and gives 1646.85 as the molecular weight of the anhydrous free base. A 16 mg implant is stated to contain 16 mg of afamelanotide equivalent to 18 mg of afamelanotide acetate, which fixes the ratio between the two conventions at roughly nine to eight. Weight figures therefore need a form attached to them. Any figure quoted without specifying salt form and hydration state is ambiguous by about a tenth.

Two supply chains for one molecule

The European Medicines Agency authorised the 16 mg afamelanotide implant on 22 December 2014 for prevention of phototoxicity in adult patients with erythropoietic protoporphyria, under exceptional circumstances — a category the agency applies when comprehensive efficacy and safety data cannot be supplied, here because of the rarity of the condition. The FDA approved the same product on 8 October 2019 under NDA 210797, with the indication worded as increasing pain-free light exposure in adult patients with a history of phototoxic reactions from erythropoietic protoporphyria. One supplement was approved in August 2024. Neither authorisation mentions cosmetic pigmentation.

What was approved is a manufactured object with a fixed release profile. The authorised implant is a solid bioresorbable rod roughly 1.7 cm long and 1.45 mm in diameter, containing the peptide dispersed in poly(DL-lactide-co-glycolide), stored refrigerated at 2 to 8 degrees Celsius and protected from light. The label restricts administration to a trained health professional using a specified implantation device, and the product is dispensed as a single sterile rod in a glass vial rather than as a powder for reconstitution. Material sold online under the melanotan-1 name is a lyophilised powder in a vial.

The trials behind both authorisations administered the implant, and that governs how the literature transfers. Human data on the free peptide given by injection come from a single university programme in Arizona that ran between the late 1980s and 2004, and involve fewer than seventy volunteers in total. The two authorisations checked for this record, the European and the American, both name erythropoietic protoporphyria and no other condition, and both restrict administration to trained clinicians.

What the pigmentation studies measured

Levine and colleagues randomised 28 healthy white men to ten subcutaneous injections of a purified NDP-MSH preparation or saline over twelve days, and followed them for seven weeks, with all subjects using a high-potency sunscreen throughout. Skin darkening was quantified by serial chromaticity measurement. The 1991 report in JAMA describes a significant parabolic curve of darkening in the poor-tanning group with Fitzpatrick types I or II and in the good-tanning group with types III or IV, no darkening in the saline group, and peak change one to three weeks after dosing ended.

Pigment chemistry was addressed in 2000 by Dorr and colleagues, who gave seven volunteers with skin types III or IV 0.16 mg/kg subcutaneously on ten weekdays across two weeks and took shave biopsies at baseline and day 21. Eumelanin was measured as its permanganate oxidation product PTCA by HPLC. Forearm PTCA rose by a mean 98 percent with a standard deviation of 25.4 percent across seven subjects, and forehead PTCA by 49 percent across three. Phaeomelanin did not change significantly, and the eumelanin to phaeomelanin ratio moved from 51:1 to 86:1.

A separate question is what happens when the peptide and ultraviolet light are combined. Dorr and colleagues reported three phase 1 trials in 2004. In the first, four subjects received 0.08 mg/kg per day subcutaneously for ten days and four received saline before the neck was irradiated at three times the minimal erythema dose of UV-B; three of the four treated subjects tanned and those three had 47 percent fewer sunburn cells at the irradiated site. In the third, eight subjects were randomised to sunlight or sunlight plus peptide, and the controls required 50 percent more exposure time for equivalent tanning.

The protoporphyria trials, and where the sources disagree

Langendonk and colleagues published two multicentre, randomised, double-blind, placebo-controlled implant trials together in the New England Journal of Medicine in 2015. The European trial, NCT00979745, gave five implants over 270 days; the American trial, NCT01605136, gave three over 180 days. The published account reports 74 patients in the European trial and 94 in the American, with a primary endpoint of hours of direct sunlight exposure without pain. Median pain-free time was 69.4 hours against 40.8 on placebo in the American trial and 6.0 hours against 0.8 in the European.

Regulatory review of the American trial produced different figures from the publication, and the registry sides with the paper. The posted results for NCT01605136 give the same primary outcome as the journal report, a median of 69.4 hours against 40.8, on the same analysis population of 46 against 43. The FDA label, reviewing the study as CUV039 under that registration, states 93 subjects enrolled, 48 on afamelanotide and 45 on vehicle, and medians of 64.1 against 40.5. Registry and label agree that 93 were enrolled; paper and registry both analysed 89. The label appears to report across all enrolled subjects, and that difference in analysis population, rather than a data error, is the likely source of the median gap. A smaller version of the same split sits in the European trial, where the paper prints a placebo median of 0.8 hours while label and registry both give 0.75.

Registered work extends past protoporphyria without resolving much. ClinicalTrials.gov holds 23 interventional studies of afamelanotide, of which twelve have posted results. Lim and colleagues randomised patients with non-segmental vitiligo to narrowband UV-B alone or with monthly implants and reported 48.64 percent repigmentation against 33.26 percent at day 168. A phase III trial in polymorphic light eruption, NCT04704713, posted a median pruritus score of 0.25 against 0.47 on placebo across 31 patients with no statistical analysis attached. A 200-patient phase III vitiligo trial, NCT06109649, remains active without results.

Pharmacokinetics belong to the formulation

Ugwu and colleagues ran a three-route crossover in 1997 in three male volunteers, comparing intravenous and oral doses of 0.16 mg/kg with subcutaneous doses of 0.08 to 0.21 mg/kg, ten doses over two weeks, with plasma measured by radioimmunoassay. Subcutaneous delivery was fully bioavailable against intravenous. Oral dosing produced no detectable plasma concentration at all. Beta-phase half-life after subcutaneous dosing ranged from 0.8 to 1.7 hours, clearance from 0.12 to 0.19 litres per kilogram per hour, and urinary recovery was 3.9 percent of dose or less.

Implant pharmacokinetics look nothing like that. The FDA label reports a single subcutaneous implant in twelve healthy adults giving a mean maximum concentration of 3.7 nanograms per millilitre with a standard deviation of 1.3, an area under the curve of 138.9 hour-nanograms per millilitre, a median time to maximum of 36 hours, and an apparent half-life of about fifteen hours. In nine of twelve subjects the last measurable concentration was at 96 hours. The apparent half-life is release-limited, so it describes the polymer matrix.

Handling chemistry has one published study and it is not a shelf-life study. Chawathe and Sharma subjected afamelanotide to acidic, basic, neutral and oxidative stress, ultraviolet exposure and 60 degrees Celsius, and reported fourteen degradation products separated by reversed-phase HPLC and characterised by high-resolution tandem mass spectrometry, with truncation, methylation, deacetylation and oxidation among the pathways. Forced degradation maps how a molecule falls apart under stress. It does not establish how long a reconstituted solution holds at a given temperature, and no study reporting that was located.

Receptor breadth, and the melanotan II problem

Haskell-Luevano and colleagues stated plainly in 1997 that NDP-MSH possesses nanomolar efficacies at every melanocortin receptor subtype except MC2R, and used that breadth as the starting point for a search for subtype-selective fragments. The IUPHAR Guide to Pharmacology curates human affinities of pIC50 10.0 at MC1R, pKi 8.9 at MC3R, pKi 8.5 to 8.8 at MC4R and pIC50 9.0 at MC5R. The FDA label uses the word predominantly, not selectively, and the curated spread is roughly one to one and a half log units. Selectivity for this molecule is asserted far more often than it is measured.

Most of the harm literature attached to the word melanotan describes a different molecule. Melanotan II is a cyclic seven-residue analogue with its own CAS number, and it is what the case reports characterise: eruptive naevi, a melanoma in a twenty-year-old reported by Hjuler and Lorentzen in 2014, rhabdomyolysis, priapism, oral mucosal lesions. The three published analytical characterisations of illicitly sold vials — Breindahl and colleagues in 2015, Mestria and colleagues in 2021, Deville and Charlier in 2024 — all examined melanotan II.

The findings from those analyses are worth stating even though they concern the other compound, because they describe the same market. Breindahl found 4.32 to 8.84 mg of peptide in vials from three shops all claiming 10 mg, with unknown impurities of 4.1 to 5.9 percent in two of them. Mestria identified melanotan II and bremelanotide in eight samples confiscated by police, an identification study that reported no content figures. Deville measured 30 percent purity in a single seized vial. Searches of PubMed and Europe PMC returned no equivalent measurement on any product sold as melanotan I.

What the safety record contains

Pooled trial safety in the FDA label covers 244 adults with erythropoietic protoporphyria across studies CUV029, CUV030 and CUV039, with 125 receiving implants and 119 receiving vehicle. Through month six, implant site reactions occurred in 21 percent against 10 percent, nausea in 19 against 14, skin hyperpigmentation in 4 against 0, and melanocytic naevus in 4 against 2. Postmarketing reports include serious hypersensitivity reactions and anaphylaxis. The label recommends a full body skin examination twice yearly, on the stated basis that darkening of pre-existing naevi and ephelides follows from the pharmacology.

Dorr and colleagues gave mice up to 2 mg/kg daily or weekly by topical application in DMSO or intraperitoneal injection for four or twelve weeks in 1988 and found no organ, haematologic or weight effects, with a slight rise in triglycerides and platelets in the twelve-week weekly group. Dawson and colleagues administered the analogue to pregnant rats in 1993 and reported no change in sex ratio, weight, morphology or histology. Lengweiler and colleagues assayed 26 patients exposed for up to six years and found no immunoreactivity in 23; the three with pre-existing immunoreactivity showed no change in titre.

Two gaps in the regulatory file are stated outright in the label. It records that carcinogenicity studies have not been conducted, while recording negative results in the Ames test, the in vitro mouse lymphoma assay and the in vivo mouse bone marrow micronucleus assay. It also states that the effect of renal or hepatic impairment on pharmacokinetics is unknown, that no drug interaction studies were conducted, and that safety and effectiveness in patients under eighteen have not been established.

What is not known

The published record is strong on one question and thin on the rest. Every modern trial administered a 16 mg controlled-release implant, to adults with erythropoietic protoporphyria, vitiligo, polymorphic light eruption and several smaller registered indications, so there is no contemporary characterisation of repeated subcutaneous injection of a reconstituted solution in healthy people; the human injection data are four publications reporting six small trials from Arizona between 1991 and 2004, the largest of which randomised 28 men. Carcinogenicity studies have not been conducted, and the FDA label pairs that absence with a recommendation for twice-yearly whole-body skin examination, so long-term neoplastic risk in exposed people is unmeasured rather than measured and found absent. No cohort study of skin cancer incidence among users exists. Renal and hepatic impairment effects on pharmacokinetics are stated by the label to be unknown, no drug interaction studies were conducted, safety and effectiveness under age eighteen are not established, and there are no human pregnancy or lactation data. Trial populations were 98 percent Caucasian with a mean age of 40, and only four of 125 treated subjects were 65 or older. Two further gaps sit in the identity layer: no published stability time-course exists for the peptide in solution, and no independent assay of any product sold under the melanotan I name has been published, so nominal content and purity in that supply chain are unverified.

Questions