AOD-9604 side effects have never been measured in a human trial

Nobody has published a human trial that measured AOD-9604’s side effects. Not a small one, not an open-label one, not even a case series. Everything sold under that name rests on a single sentence in a drug-development tracking journal and a handful of rat and mouse studies that are, at this point, old enough to vote. That gap is the actual story behind a query as specific as “AOD-9604 side effects.”

What the human record on AOD-9604 contains

Here is the entire public human record: by February 2002, AOD-9604’s developer, Metabolic, had Phase IIa trials of the peptide underway for obesity, according to a 2004 entry in a drug-development pipeline journal^[1]. The entry doesn’t give a participant count, a dosing schedule, a duration, an efficacy result, or a single adverse event. It reports a status, nothing about what the trial found.

No later peer-reviewed report of a completed AOD-9604 human trial turns up in PubMed, Europe PMC, or Crossref. Not an efficacy result, not a safety paper, not a case report of someone who took it and had a reaction. The trail that started in a 2002 pipeline note simply goes cold.

What we could not verify: any human dose, any human adverse event, and any human safety outcome for AOD-9604. The entire citable human safety record is one clause about a trial phase.

Because that clause is what exists, there is no citable human basis for any specific AOD-9604 side effect. Not injection-site pain, not water retention, not blood sugar changes. Anyone stating one as established fact is extrapolating from somewhere else, usually from full-length growth hormone or from forum reports, neither of which is AOD-9604 data.

The mechanism: an hGH fragment that isn’t quite what the marketing says

AOD-9604 is a synthetic fragment of human growth hormone, and the fragment’s own history complicates the fat-burning pitch built around it. In 1993, Wu and Ng identified the synthetic C-terminal sequence of hGH, residues 177 to 191, as reproducing the intact hormone’s antilipogenic activity in rats^[2]. Antilipogenic means it blocks new fat formation. It is not the same thing as lipolytic, which means breaking down fat that’s already there, and the same 1993 study found no significant lipolytic effect from the fragment alone, measured by glycerol release from treated rats’ fat pads.

That distinction rarely survives contact with a sales page. AOD-9604 gets marketed as a fat-burner, when the paper the fragment is named after found it stops fat from being made rather than melting fat that exists.

A separate 2001 study by Heffernan and colleagues treated obese and lean mice with hGH, AOD9604, or saline for 14 days, then ran a receptor-binding assay in cells transfected with the human growth hormone receptor^[3]. AOD9604 did not compete for that receptor, and it did not trigger the cell proliferation that intact hGH produced in the same assay. That’s a real mechanistic distinction, and it is the reason researchers expected AOD-9604 to sidestep growth-hormone-receptor side effects such as insulin resistance. But a receptor-binding assay in cultured cells tests receptor binding in cultured cells. It is not a safety study, in an animal or in a person.

What the animal tolerability data show

The closest thing to a tolerability signal for AOD-9604 comes from obese Zucker rats. Ng and colleagues gave the rats a daily oral dose of AOD9604, 500 micrograms per kilogram, for 19 days^[4]. Treated animals gained 15.8 grams versus 35.6 grams in untreated controls, more than a 50 percent reduction, and their fat tissue showed increased lipolytic activity.

56% less body-weight gain than untreated controls in obese rats given oral AOD9604 for 19 days PMID 11146367

The same study measured insulin sensitivity with the euglycemic clamp technique, the standard method for that measurement, and found chronic AOD9604 treatment had no adverse effect on it. Chronic treatment with intact growth hormone in the same rat model did. That’s a genuinely favorable animal signal, and it is worth taking seriously as pharmacology.

It is not a toxicology study. Nineteen days in Zucker rats measures metabolic markers, not a systematic panel of adverse events, organ pathology, or long-term exposure. A rat that keeps normal insulin sensitivity for 19 days on an oral dose tells you something real about a mechanism. It tells you nothing about what happens to a person injecting AOD9604 for months, at a dose nobody in the published literature has tested.

Side effects and red flags that remain unanswered

What does exist is data about the broader category AOD-9604 gets sold within.

A 2026 review by Dominikowski and colleagues covers growth-hormone-axis peptides used off-label for performance and fat loss, grouping AOD9604 with GHRH analogues like tesamorelin and GH secretagogues like ipamorelin^[5]. Across that whole class, the review reports:

• Endocrine and metabolic disturbances, including prolactin and cortisol elevations, appetite changes, and dysglycemia
• Fluid retention syndromes
• Musculoskeletal symptoms, myalgia and arthralgia
• Injection-site reactions

The review does not say which of these, if any, trace back to AOD-9604 specifically rather than to the GHRH analogues or secretagogues sitting next to it on the list. That limitation belongs to the source, and repeating it here is more honest than smoothing it over.

A separate 2026 review by Hailu and colleagues describes general risks from product-quality uncertainty, gray-market access, and informal dosing across unregulated, self-administered research peptides^[6]. The review does not name AOD-9604 specifically, but every one of those risks would apply to any AOD-9604 product obtained outside a clinical trial, regardless of what the peptide itself does in a rat.

The practical risk sits in the vial itself: made by whoever made it, to a purity nobody checked.

Long-term safety, safety at the doses people self-administer, and safety in combination with other peptides have never been studied for AOD-9604 in any published human trial. That sentence would be true of most research peptides sold this way. It happens to be unusually, completely true of this one.

Where AOD-9604 sits next to a peptide that finished its trials

Compare that record to tesamorelin, a growth-hormone-releasing-hormone analogue built on a completely different engineering strategy: instead of clipping a fragment off growth hormone, it prompts the pituitary to release more of it. A 2026 systematic review and meta-analysis by Ditta and colleagues pooled four randomized, placebo-controlled trials covering 909 patients with HIV-associated lipodystrophy^[7]. Tesamorelin significantly reduced visceral adipose tissue, waist circumference, and trunk fat, and increased lean body mass, though with more growth-hormone-related adverse effects and a higher discontinuation rate than placebo.

AOD-9604 versus tesamorelin: the human evidence gap

Based on the citable literature retrieved for this article.

Tesamorelin also carries FDA approval for that indication, discontinuations and all. AOD-9604 has no equivalent trial and no equivalent approval for anything, in obesity or any other indication. One peptide has a documented human record, warts included, and the other has none at all. For a longer look at tesamorelin’s evidence base, see our separate deep dive.

The companion piece on AOD-9604’s benefits claims and the piece on its dosing run into the identical wall from different directions: a 2002 phase mention, then nothing.

What would change this

A single published human trial, aimed at safety or at efficacy, would do more to settle AOD-9604’s reputation than another decade of pipeline mentions. It would give the field an actual dose, an actual adverse-event table, and an actual population to check against. Until that trial exists, every specific side-effect claim attached to AOD-9604 beyond generic unregulated-peptide risk is inference: from rats, from mice, or from a different peptide entirely. Anyone telling you otherwise is quoting a source that doesn’t exist.

This article is for research and informational purposes only and is not intended to diagnose, treat, cure, or prevent any disease. The peptides discussed here are sold for research use only and are not for human consumption. Nothing in this article constitutes medical advice. Consult a qualified clinician before making changes to a health, training, or supplementation protocol.