AOD-9604 is unusual among the peptides sold for research: somebody actually ran placebo-controlled human trials of it. Its Australian developer, Metabolic Pharmaceuticals, tested it as an obesity drug in the early 2000s, and in 2013 two of its staff co-authored a paper pooling the safety data from all six trials. That paper is the side-effect record. This page sets out what it contains, trial by trial, and the one thing about it that most summaries leave out: none of the six trials used the route by which AOD-9604 is now sold and discussed. It reports what the record says; it does not reassure or warn, and a decision about use or about a symptom belongs with a clinician. The dose figures that circulate, and why the indexed literature contains more anti-doping chemistry than clinical work, are on our AOD-9604 dosage page. How brands appear on this site is on our disclosure page.
The source, and who wrote it
The paper is 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; full text). The corresponding author's address was a Berlin firm, analyze & realize; the other two gave their address as Metabolic Pharmaceuticals in Port Melbourne. That does not make the numbers wrong, but it is the context a reader needs: this is the developer's own summary of its own programme, published after the programme had ended, in a small open-access journal. None of the six trials was ever registered on ClinicalTrials.gov — the registry returns no record under either spelling — so there is no independent registry entry to check the pooled figures against.
The paper says "approximately 900 adult subjects" took part. The trial-by-trial counts it prints add up to about 893 randomised, with 250 people on AOD-9604 in the 12-week study alone.
Six trials, two routes — neither of them an injection under the skin
| Trial | People | Route and dose | Length |
|---|---|---|---|
| METAOD001 | 15 healthy men | Single IV infusion, 25–400 µg/kg (growth hormone as a comparator) | Single doses, 7-day washouts |
| METAOD002 | 23 people, BMI ≥35 | Single IV infusion, 25, 50, 100 µg/kg | Single doses, 7-day washouts |
| METAOD003 | 17 obese men | Oral capsules, 9, 27, 54 mg | Single doses, 2-week washouts |
| METAOD004 | 36 obese men | Oral capsules, 9, 27 or 54 mg daily | 7 days |
| METAOD005 | 300, BMI ≥35 | Oral capsules, 1, 5, 10, 20 or 30 mg daily | 12 weeks |
| METAOD006 | 502, BMI 30–45 | Oral tablets, 0.25, 0.5 or 1 mg daily | 24 weeks |
The research market sells AOD-9604 as a lyophilised powder for injection under the skin, and the figure that circulates is about 300 micrograms a day. That route appears nowhere in the table. The intravenous studies were single infusions over twenty minutes in a clinic. The long studies were pills — and a sixteen-residue peptide swallowed as a tablet is largely digested before it reaches the blood, which is why the oral doses run to tens of milligrams. So the safety record describes what happened when people swallowed AOD-9604 or received it once into a vein. How a daily subcutaneous injection of research-grade material behaves over weeks is not something the trials can answer, and nothing since has tested it.
What the trials recorded
The short studies. In METAOD001, twelve of fifteen men reported 29 adverse events between them; headache was the commonest (six times), followed by fatigue, "hypoglycaemia unspecified" and dizziness. None was severe, and they were spread across AOD-9604, growth hormone and placebo periods. In METAOD002, sixteen of 23 people (69.6%) reported headache at some point; three events were severe (one after 50 µg/kg of AOD-9604, two after placebo), and one — a feeling of chest tightness — was judged possibly related to the drug. Five of 23 reported mild or moderate euphoria during AOD-9604 periods and none during placebo, the one symptom in the short studies that appeared only on the drug.
The oral dose-escalation. In METAOD003, all seventeen men reported something — 97 events in total — and two serious events occurred at 54 mg: diarrhoea judged possibly related, and a bronchial pneumonia judged unrelated. In the seven-day METAOD004, every one of 36 men reported at least one event (207 in total), and the authors wrote that those on 54 mg "experienced a greater number of headaches, diarrhea and flatulence". The discussion section returns to it as the only dose-related finding: the 54 mg dose "was associated with an increased incidence of GI-related AEs."
The 12-week trial (METAOD005). During a two-week placebo run-in, 70% of people already reported at least one adverse event, mostly headache. After active treatment began, 88.9% did, distributed similarly across the five dose groups and placebo; headache (42.6%), diarrhoea (9.0%) and infections were the commonest. Five people had a serious adverse event, and all five were cancers or growths on AOD-9604: a basal cell carcinoma, a lipoma and a squamous cell carcinoma in the 20 mg group, a breast cancer at 5 mg and a melanoma at 10 mg. The investigator judged none related. The authors' reasoning was that IGF-1, the growth-hormone signal linked to cancer risk, did not rise in any group. The arithmetic a reader should hold alongside that: 250 people were on AOD-9604 and 50 on placebo, the trial lasted twelve weeks, and neither fact supports a conclusion in either direction.
The 24-week trial (METAOD006). This is the best evidence the record has. After treatment began, 78.7% reported at least one adverse event — 83.2% on placebo and 75.6% on the lowest dose — with colds and nasopharyngitis, headache, back pain and diarrhoea the commonest. The paper's Table 2 counts people with a serious adverse event:
| Group | People | With a serious adverse event |
|---|---|---|
| Placebo | 125 | 8 |
| 0.25 mg a day | 127 | 3 |
| 0.5 mg a day | 125 | 1 |
| 1 mg a day | 125 | 6 |
| Total | 502 | 18 (3.6%) |
Most were injuries and procedural complications. The authors reported no changes in laboratory values, vital signs or ECGs, and no difference from placebo in glucose tolerance at 12 or 24 weeks.
The growth-hormone questions the trials were designed to answer
AOD-9604 was built as a fragment of growth hormone (residues 177 to 191, with a tyrosine added) that would keep the hormone's effect on fat and drop its effects on blood sugar and growth. The trials therefore checked three things specifically. IGF-1 did not change in any group in either long study (mean changes of 1.76 and 1.24 nmol/L at 12 and 24 weeks in METAOD006, P = 0.51 and 0.76 across groups). Glucose tolerance showed no significant change from placebo. Antibodies against AOD-9604 were looked for at baseline and at 4, 8 and 12 weeks in METAOD005 and in a subset of METAOD006, and none were found in anyone.
Each of those is a genuine negative result, and each comes with the same limitation: it was measured after oral dosing. An absent antibody response to a peptide that is mostly digested in the gut says little about the immune response to the same peptide injected under the skin, which is the scenario FDA's reviewers flag for injected research peptides — including the aggregation and synthesis impurities that come with material made outside a regulated supply chain. AOD-9604 is one of the sixteen peptides FDA lists as nominated for compounding and then withdrawn, as covered on our Category 2 list page, and it was not among the peptides FDA's reviewers evaluated in briefing documents for the July 2026 compounding advisory committee.
What has been reported since
Nothing that adds to the trials. An openFDA query of FDA's adverse-event reporting system on 2026-09-27 (data exported 2026-09-23) returned no report naming AOD-9604 or AOD9604 as a product. That is an absence of reports, not an absence of harm: compounders and research-market buyers rarely file them, and a product sold "for research use only" has no manufacturer obliged to. The indexed literature contains no case report of harm from AOD-9604 that this census found. What it contains instead, as the dosage page counts, is detection chemistry — methods for finding the compound in athletes' urine. How the side-effect record for the category compares is on our peptide side effects overview, and the wider pattern of lists borrowed from other molecules is visible on our BPC-157 side-effects page.
Where the online side-effect lists come from
Most lists of "AOD-9604 side effects" name headache, injection-site redness, nausea and fatigue. Headache and gastrointestinal upset do trace to the trials — though at similar rates on placebo. Injection-site reactions cannot trace to the trials, because no trial injected it under the skin; they are either inferred from how peptide injections generally go, or anecdote. There is no FDA-approved product containing AOD-9604, no label and no registered trial, so no list can cite a labelled adverse-reaction section either.
What the record does not contain
A trial of subcutaneous injection at any dose; any human data beyond 24 weeks; a published efficacy paper from the obesity programme; any trial run by anyone other than the developer; any adverse-event report in FDA's database; any analysis of the purity, content or aggregation of research-market vials; and any study in women before menopause, in pregnancy, in adolescents or in people with diabetes, since the trials enrolled otherwise healthy adults with obesity. AOD-9604 has no approval for human use anywhere. Anyone weighing its use, or wondering whether a symptom is related to it, should take the question to a clinician who can see their history.
Sources and dates
Read 2026-09-27: Stier, Vos and Kenley 2013, full text (PDF from the journal), results and discussion sections and Table 2, DOI checked by content negotiation at doi.org; ClinicalTrials.gov API v2 searches for AOD9604 and AOD-9604 (zero studies each); openFDA drug event endpoint for both spellings (no matches); FDA's 503A bulks list page as summarised on our Category 2 page. Corrections go to the contact page.