AOD-9604 is a laboratory-made peptide (a short chain of amino acids) cut from human growth hormone. It was created in Australia in the late 1990s as a candidate anti-obesity drug: the idea was to keep the fat-burning part of growth hormone's action while shedding its unwanted effects — a rise in IGF-1 and worse blood-sugar control. It worked as planned in obese rats and mice. In humans it went through six clinical trials involving almost 900 people: it proved well tolerated, but in the largest, 24-week trial it did not beat placebo on weight loss, and the programme was closed in 2007. In this review we explain in plain terms what it is, how it was meant to work, and how strong — or rather how clearly negative — the evidence for its efficacy in humans is.
What AOD-9604 is
Human growth hormone (GH or hGH for short) is a protein built from 191 amino acids. From the 1970s onwards researchers at Monash University in Melbourne (Professor Frank Ng's group) tested whether individual pieces of that protein had actions of their own. It turned out that the final stretch of the chain — amino acids 177 to 191 — carries the effect on fat tissue: it curbs fat storage and stimulates fat release from cells. AOD-9604 is exactly that fragment with one extra amino acid (tyrosine) added at the start to make the molecule more stable. The name AOD stands for “anti-obesity drug”. The same compound is sold under the name “HGH Fragment 176-191” — these are not two peptides but two names for one molecule, which we cover in our piece on fragment 176-191.
How it was meant to work
Whole growth hormone burns fat, but along the way it acts on its own “switch” on cells (the GH receptor), which makes the liver produce IGF-1 — a growth factor that drives tissue growth — and makes the body handle sugar less well. The designers of AOD-9604 wanted to keep only the fat effect. In cell studies the fragment did not bind the growth-hormone receptor at all and did not make cells divide — so it could not act like the full hormone (Heffernan et al., 2001). The most studied lead for the mechanism is the beta3 receptor — a “switch” on fat cells that turns on fat burning. Mouse studies, however, showed something more complicated: AOD-9604 increased the number of those receptors in fat tissue, yet in mice lacking the beta3 receptor it still acutely raised energy expenditure and fat oxidation. The authors stated plainly that the action is not transmitted directly through the beta3 receptor, although a larger number of them may sensitise the tissue to fat burning. The exact mechanism has never been pinned down.
Structure and origin
AOD-9604 is a hexadecapeptide (16 amino acids) with the sequence Tyr-hGH 177-191, made synthetically. Two cysteines in the chain form a disulphide bridge, so the molecule has the shape of a small loop, similar to that region of the natural hormone. It was developed by the Australian company Metabolic Pharmaceuticals, which funded all of the clinical trials. That is worth remembering when reading the results: both papers summarising safety (Stier 2013, Moré 2014) were written by people connected with the manufacturer, and the journal that published them is not indexed in PubMed.
What was studied — cells and animals
The preclinical data are consistent. In obese Zucker rats, oral AOD-9604 given for 19 days cut body-weight gain by more than half (15.8 g versus 35.6 g in controls), increased the lipolytic activity of fat tissue and — unlike whole growth hormone — did not impair insulin sensitivity, which was checked with the metabolic clamp method (Ng et al., 2000). In genetically obese ob/ob mice, 14 days of treatment increased fat oxidation and blood glycerol (a marker of fat breakdown), without the hyperglycaemia caused by the full hormone (Heffernan et al., 2001). Toxicology commissioned by the manufacturer — the Ames test, a chromosomal aberration test, a micronucleus test, six months of dosing rats and nine months of dosing monkeys — found no genotoxicity and no organ toxicity; oral doses were well absorbed by pigs but broken down quickly (Moré and Kenley, 2014).
Joints are a separate thread. A Korean study of 32 rabbits with chemically induced knee osteoarthritis found that weekly intra-articular injections of AOD-9604, alone or with hyaluronic acid, improved cartilage scores and shortened the period of lameness; the combination worked better than either component alone (Kwon and Park, 2015). This is the only paper on the subject, done in animals, with no follow-up in humans — and yet it is cited as evidence of “joint regeneration”.
Human data — six trials and their outcome
Between 2001 and 2006 six randomised, double-blind, placebo-controlled trials were run in Australia, enrolling a total of 893 adults, most of them obese (Stier et al., 2013). The first three were single-dose studies (intravenous, 15 healthy and 23 obese men; oral, 17 obese men) and the fourth lasted seven days. The two that matter are the last ones: a 12-week phase IIb trial of 300 people with a BMI of at least 35 (capsules, 1–30 mg daily) and a 24-week trial of 502 people with a BMI of 30–45 (tablets, 0.25–1 mg daily), with a four-week placebo run-in. Both were designed to measure weight loss.
What came out of them? The paper by Stier et al. discusses safety only: AOD-9604 did not change IGF-1 concentrations, did not worsen oral glucose tolerance test results, did not induce antibodies and caused no withdrawal or serious adverse event judged related to the drug. The most common complaints — headache, diarrhoea, upper respiratory infections — were just as frequent with placebo. Efficacy results are not reported in that paper. They have to be sought elsewhere: according to a 2026 review (Dominikowski et al.), which cites the manufacturer's stock-exchange announcement of February 2007, the 24-week trial of 502 people failed to show a significant advantage over placebo on the primary endpoint, weight loss, and the company announced the end of the programme because the results “did not support commercial viability”. The review's authors stress that the conclusion of no efficacy rests on sponsor-reported data rather than a peer-reviewed publication — the full results of the METAOD005 and METAOD006 trials never appeared in a scientific journal.
Why development was stopped
The story of AOD-9604 is a textbook case of a drug that “works in mice”: reproducible animal data, good tolerability in humans — and no sign of the effect the molecule had been created for. After 2007 the manufacturer tried a different route: as the paper by Stier et al. mentions, an expert panel assessed the peptide as GRAS (“generally recognised as safe”) for use in foods in the USA. No regulatory agency has ever approved AOD-9604 for treating any disease.
Safety and the limits of the evidence
The good safety profile seen in the trials concerned tablets and capsules taken for at most 24 weeks under medical supervision, using a product made to pharmaceutical standards. It says nothing about injections of preparations of unknown composition. The US FDA placed AOD-9604 in category 2 of bulk substances for pharmacy compounding, meaning it “may present significant safety risks” — because of possible immunogenicity (antibody formation) after injection, peptide-related impurities and difficulty in confirming the composition; the agency also noted reports of serious adverse events of unclear causality (FDA list; as of April 2026 the nomination has since been withdrawn by the nominators). In the 12-week trial five serious adverse events were recorded — four malignancies (three skin cancers and a breast cancer) and a lipoma — all in the groups taking the peptide and none with placebo; the investigators judged them unrelated to treatment, pointing among other things to the lack of any dose relationship and the high incidence of skin cancer in Australia — but with 250 people taking the drug and 50 taking placebo, such a difference can be neither confirmed nor ruled out. The peptide is banned by the World Anti-Doping Agency; a method exists for detecting it in urine together with a stable metabolite (Cox et al., 2015). We deliberately give no methods of use and no doses.
The wider context — growth-hormone fragments and obesity drugs
AOD-9604 is the best-studied “lipolytic fragment” of growth hormone and one of the few research peptides for which a large, controlled human trial exists — and that trial came out negative. By comparison, whole growth hormone (somatropin) does reduce fat tissue, but at the cost of higher IGF-1 and a diabetes risk, while the drugs that genuinely cause weight loss in clinical trials work in an entirely different way — through incretin receptors, like semaglutide. The 2026 review places AOD-9604 in the tier of compounds with “limited or indirect human data”, and treats claims about “body recomposition” or joint repair as speculation extrapolated from animal work.
Summary
AOD-9604 is a 16-amino-acid fragment of growth hormone designed to burn fat without stimulating IGF-1. In rats and mice it did exactly that. In humans, six trials involving 893 people confirmed good tolerability and no effect on IGF-1 or glucose, but the largest of them — 24 weeks, 502 people — showed no advantage over placebo for weight loss, and the manufacturer closed the programme in 2007. The efficacy data were never fully published, the safety publications come from people connected with the company, and the joint research is limited to a single rabbit study. The state of the evidence is therefore unusually clear: a mechanism confirmed in animals, and efficacy in humans not confirmed by the very trial that was meant to demonstrate it.
Sources
- Stier H, Vos E, Kenley D. Safety and Tolerability of the Hexadecapeptide AOD9604 in Humans. Journal of Endocrinology and Metabolism. 2013;3(1–2):7–15. DOI: 10.4021/jem157w. doi.org/10.4021/jem157w
- Dominikowski A, Rękoś Z, Olejarz M, et al. The emerging landscape of performance-enhancing peptides modulating GH-IGF1 axis: bridging the gap between clinical evidence and patient self-administration. Frontiers in Endocrinology. 2026;17:1822475. PMID: 42395176. DOI: 10.3389/fendo.2026.1822475. pubmed.ncbi.nlm.nih.gov/42395176
- Ng FM, Sun J, Sharma L, et al. Metabolic studies of a synthetic lipolytic domain (AOD9604) of human growth hormone. Hormone Research. 2000;53(6):274–278. PMID: 11146367. DOI: 10.1159/000053183. pubmed.ncbi.nlm.nih.gov/11146367
- Heffernan MA, Thorburn AW, Fam B, et al. Increase of fat oxidation and weight loss in obese mice caused by chronic treatment with human growth hormone or a modified C-terminal fragment. International Journal of Obesity and Related Metabolic Disorders. 2001;25(10):1442–1449. PMID: 11673763. DOI: 10.1038/sj.ijo.0801740. pubmed.ncbi.nlm.nih.gov/11673763
- Heffernan M, Summers RJ, Thorburn A, et al. The effects of human GH and its lipolytic fragment (AOD9604) on lipid metabolism following chronic treatment in obese mice and beta(3)-AR knock-out mice. Endocrinology. 2001;142(12):5182–5189. PMID: 11713213. DOI: 10.1210/endo.142.12.8522. pubmed.ncbi.nlm.nih.gov/11713213
- Moré MI, Kenley D. Safety and Metabolism of AOD9604, a Novel Nutraceutical Ingredient for Improved Metabolic Health. Journal of Endocrinology and Metabolism. 2014;4(3):64–77. DOI: 10.14740/jem213w. doi.org/10.14740/jem213w
- Kwon DR, Park GY. Effect of Intra-articular Injection of AOD9604 with or without Hyaluronic Acid in Rabbit Osteoarthritis Model. Annals of Clinical and Laboratory Science. 2015;45(4):426–432. PMID: 26275694. pubmed.ncbi.nlm.nih.gov/26275694
- Cox HD, Smeal SJ, Hughes CM, et al. Detection and in vitro metabolism of AOD9604. Drug Testing and Analysis. 2015;7(1):31–38. PMID: 25208511. DOI: 10.1002/dta.1715. pubmed.ncbi.nlm.nih.gov/25208511
For in-vitro laboratory research only. It is not a human medicine and is not for treatment.