AOD-9604

Growth hormone does two quite different things: it drives growth through IGF-1, and it mobilises fat. Work at Monash University in the 1980s and 1990s localised the metabolic activity to the C-terminal end of the molecule, and showed that a synthetic peptide of residues 177-191 reproduced the antilipogenic effect [3] and reduced weight gain and fat mass in obese mice [4]. AOD-9604 is that fragment (in the 176-191 numbering, with a tyrosine added at the N terminus) developed as a drug candidate.

What it does in animals. Thirty days of oral dosing in ob/ob mice slowed weight gain from day 16 onwards without changing food intake, cut lipogenic activity and raised lipolytic activity in fat tissue, and raised energy expenditure and fat oxidation acutely [1]. Fourteen days by injection reduced body weight and body fat in obese mice, alongside a rise in beta-3 adrenergic receptor expression [2].

How it might work. The obvious guess was the beta-3 adrenergic receptor, the main lipolytic receptor on fat cells, whose expression both growth hormone and AOD-9604 restore in obese mice. But in beta-3 knockout mice, chronic treatment no longer reduced weight or increased lipolysis, while an acute dose still raised energy expenditure and fat oxidation — so the effect is not exerted directly through that receptor [2]. A parallel study in Zucker fatty rats mapped the same molecular and cellular actions on lipid metabolism [5]. An earlier study found the fragment altered glucose transport in rat fat cells [6], and the closely related C-terminal sequences have a hyperglycaemic action of their own [7].

Human evidence. There is none in the published literature. The only human material ever exposed to it in these papers is isolated adipose tissue, where it increased lipolysis in a dish [1]. AOD-9604 was taken into commercial development for obesity by Metabolic Pharmaceuticals and appears in mid-2000s reviews of obesity drugs in clinical development [8], but no trial results were published, and the programme did not produce an approved drug. A 2026 review of peptides used for physique purposes lists it among compounds with no supporting human trial data [9].

It has since been tried for cartilage: injected into the knee joints of rabbits with induced osteoarthritis, alone or with hyaluronic acid, it improved cartilage scores and shortened lameness [10]. Again, rabbits.

The rationale is separation of function. The growth-promoting effects of growth hormone run through the growth hormone receptor and IGF-1; the lipolytic effects were mapped to a distinct C-terminal domain [3]. A peptide of that domain should therefore mobilise fat without raising IGF-1, without the insulin resistance, and without the growth signalling that makes growth hormone itself unattractive as a weight-loss drug.

In animals that separation holds up. What is missing is the receptor: after 30 years there is still no identified binding site that explains the effect, and the beta-3 knockout work rules out the most obvious candidate as a direct mechanism [2]. Without a receptor and without human data, the mechanism remains a description of what happens in mouse fat rather than an explanation.

AOD-9604 is a 16-residue peptide: growth hormone residues 177-191 (176-191 in the other numbering convention) with an N-terminal tyrosine, and the same disulfide-closed loop the parent hormone has [3]. It has been given by mouth and by injection in animals [1]. Detection methods for it exist in doping control [11].

There is no published human safety data for AOD-9604 — no trial, no pharmacokinetics, no adverse event reporting. The animal studies report no toxicity, but they were not designed as safety studies [1][2].

One thing worth noting from the older literature: synthetic peptides related to growth hormone's C-terminal sequence have a hyperglycaemic action of their own in animals [7], and the fragment changes glucose transport in rat fat cells [6]. Whether that matters in people is unknown, because nobody has looked. A 2026 review groups AOD-9604 with the peptides sold for physique purposes whose composition and dosing in unregulated supply is itself a risk [9].

No human interaction data exist. In rabbits, combining intra-articular AOD-9604 with hyaluronic acid worked better than either alone for cartilage scores [10]. No interactions with other compounds on this site have been documented.