AOD-9604: safe in trials, but the weight loss never showed up

AOD-9604 was safe in six placebo-controlled human trials covering about 900 adults, and it did not work as a weight-loss drug [5,6]. The largest trial, 24 weeks in more than 500 people, missed on weight loss, and the developer abandoned the obesity programme in 2007 [6]. The rodent data that launched it, slower weight gain and more fat burning in obese animals, were real [1–3], but the effect never came across into people. AOD-9604 is the rare peptide whose human evidence is good enough to answer the question, and the answer is that it does not work.

What it is

Human growth hormone is a 191-amino-acid protein. Researchers at Monash University in Australia argued that its fat-burning activity sits in the tail end of the molecule, residues 177 to 191, separate from its growth-promoting effects [1]. AOD-9604 is that fragment with an extra tyrosine added at the start, giving 16 amino acids: YLRIVQCRSVEGSCGF, with a ring formed between its two cysteines [5].

The appeal was straightforward. Growth hormone breaks down fat but also raises IGF-1 and can worsen glucose tolerance [5]. A fragment that kept the first effect and dropped the rest would be a cleaner weight-loss drug. In lab tests, AOD-9604 did not bind the growth hormone receptor and did not make receptor-bearing cells multiply, unlike full growth hormone [3]. How it does work is still unclear. The proposed mechanism is more breakdown of stored fat and less fat formation [1,2], and mouse experiments suggest the effect is not driven directly by the beta-3 adrenergic receptor, a major fat-burning switch [4].

An Australian company, Metabolic Pharmaceuticals, took it into human trials for obesity [6].

The human evidence

The human data come from six randomised, double-blind, placebo-controlled trials run between 2001 and 2006 [5]. The only peer-reviewed account of them, published in 2013, is a safety paper; it doesn't report the weight results.

What the trials found on safety

Across the programme, the paper reports no treatment-related serious adverse events or withdrawals, no rise in IGF-1, no worsening of glucose tolerance and no antibodies against the peptide [5]. That last point matters because the body can learn to neutralise injected or absorbed peptides. The FDA later raised this same risk of immune reactions (immunogenicity) for some routes of administration [8].

Headache was the most common complaint in almost every study. In the 12-week trial, there were more cancer diagnoses than expected, including several skin cancers; the authors judged them unrelated to the drug because none occurred in the top-dose group and the trial was run in Australia, which has high skin cancer rates [5]. That is the authors' own judgement; the underlying data have not been analysed outside the sponsor.

In the 24-week trial, 78.7% of participants had at least one adverse event once active treatment began, but the rate was highest on placebo (83.2%) and lowest on the 0.25 mg dose (75.6%), mostly colds, headache, back pain and diarrhoea [5]. Most of those are the background noise of any six-month trial.

Five people in one intravenous study reported mild or moderate euphoria during AOD-9604 dosing and none during placebo [5].

What the trials found on weight

This is where the story falls apart. According to a 2010 review in Clinical Pharmacology & Therapeutics, the 12-week trial showed an average loss of 2.6 kg on 1 mg a day against 0.8 kg on placebo, but a tenfold higher dose (10 mg) produced less weight loss, not more [6]. A drug that works less well at higher doses is a warning sign. The larger 24-week trial then failed to produce significant weight loss, and development for obesity was stopped in 2007 [6].

The weight data from those trials were never published in full in a peer-reviewed journal, so we can't check the numbers ourselves. What is published supports one clear conclusion: in people, at the doses tested, AOD-9604 was well tolerated and did not work well enough as a weight-loss drug to continue.

Every human trial used oral or intravenous dosing [5]. We found no published controlled trial of injection under the skin, which is how it is commonly promoted today.

What the animal and lab work suggests

The animal data are why AOD-9604 attracted attention in the first place. All of the following were in rodents or rabbits, not people:

• In obese Zucker rats, 19 days of oral AOD-9604 at 500 µg/kg roughly halved weight gain (15.8 g against 35.6 g) without harming insulin sensitivity [1].
• In obese mice given a closely related version orally (called AOD-9401 in that paper) for 30 days, weight gain slowed from day 16 with no change in food intake, and fat tissue from rodents and people broke down more fat when exposed to it in a dish [2].
• In obese mice on a continuous infusion for 14 days, both growth hormone and AOD-9604 cut weight gain and raised fat burning, but only growth hormone raised blood sugar [3].
• In mice lacking the beta-3 adrenergic receptor, longer-term treatment lost its weight effect, though a single dose still raised energy use [4].
• In rabbits with chemically induced knee arthritis, injections of AOD-9604 into the joint improved cartilage scores and shortened lameness, especially combined with hyaluronic acid [7].

The knee study is the source of much of the recent interest in AOD-9604 for joints. It is a single study in 32 rabbits, and there is no human trial of joint injection [7].

Most of the rodent studies measured slower weight gain in genetically obese animals that were putting on weight fast [1–3]. That is not the same as weight loss in adults who have stopped gaining, and the gap may help explain why the effect didn't translate.

Safety and side effects

From the published human trials, AOD-9604 is well tolerated at the oral doses tested, with a side-effect profile the authors called indistinguishable from placebo [5]. Headache was the commonest complaint across studies. One limitation applies to all of it: the programme was manufacturer-sponsored and reported years after the fact, and the longest exposure was 24 weeks.

The FDA's own assessment, written for its compounding review, said it had found no or only limited safety information, flagged a risk of immune reactions and impurities, and noted serious adverse events that may be associated with AOD-9604, though causality was not clear [8].

Legal and regulatory status

A 2013 anti-doping lab study found AOD-9604 does not interfere with the standard WADA test for growth hormone isoforms, so it has to be detected by other methods [10].

What is still open

• The full weight-loss results of the two long human trials, which were never published in a journal [5,6].
• Whether injection, as opposed to oral dosing, behaves differently in people. No controlled human trial has tested it.
• Whether the joint findings in rabbits apply to people [7].
• How it works, given that it doesn't act on the growth hormone receptor [3,4].
• Whether the euphoria reported in one small intravenous study is real or chance [5].

What this means

AOD-9604 is not an under-researched peptide. It is the opposite: a sponsored programme that ran six placebo-controlled human trials, and the human data are strong enough to answer the question the rodent work raised. The answer is no. An oral dose of 1 mg a day produced 2.6 kg of loss against 0.8 kg on placebo over 12 weeks [6], and the far larger 24-week trial with more than 500 people missed its weight endpoint outright [6], which is why Metabolic Pharmaceuticals stopped development in 2007. When a higher dose performs worse than a lower one, as the 10 mg arm did in that 12-week trial, you are looking at noise rather than a dose response [6]. If you are buying AOD-9604 for fat loss, you are paying for the use its own developer abandoned.

That last point is the one most coverage gets wrong in the other direction. Every human trial dosed AOD-9604 by mouth or by vein; not one tested injection under the skin, which is how it is sold now [5]. So the product in circulation is being used in a way that has never been measured in people, and the safety record belongs to a route nobody is using. The "well tolerated" line is real, and it applies to oral dosing up to 1 mg a day, not to a vial someone reconstitutes at home.

There is one thread worth pulling. The only study pointing at a new use is the rabbit knee work, where injected AOD-9604 improved cartilage scores and shortened lameness [7]. That is a single study in 32 animals, and no human joint trial exists [7]. We would not back it for joints today, but if any sponsor funds AOD-9604 again, that is the trial to fund, not another weight-loss study the 2007 data already answered.

For anyone weighing it up anyway: it is not approved in the US or the UK, it is banned at all times in sport under WADA S2 [9], and the FDA has flagged it for immune reactions and impurities [8]. The risk you can actually control is not the peptide but the vial. Purity, sterility, endotoxin and heavy metals all have to be shown to you before you know what was sent, and anything that arrives without that evidence should be treated as unknown contents.

Educational content only — not medical advice. Many peptides discussed on HPR are not approved for human use. Talk to a qualified clinician before making any decision about your health.