AOD-9604 study findings: what has been measured

Has AOD-9604 been tested in people?

Yes. An Australian company, Metabolic Pharmaceuticals Limited, ran a human obesity program. On 21 February 2007, it described its final trial, called OPTIONS, to the Australian Securities Exchange. OPTIONS enrolled 536 people at 16 sites across Australia. Adults took tablets for 24 weeks. Before that came a four-week run-in, a warm-up stage with a diet and exercise programme supervised by dietitians. The trial was double-blind and placebo-controlled. A placebo is a dummy tablet with no active drug. The company announcement gives the weight-loss results. A later journal paper, from 2013, reports safety observations.

Double-blind means neither the people in the trial nor the staff knew who got AOD-9604 and who got placebo. That stops hopes and expectations from tilting the result.

During the four-week run-in, everyone took placebo. That stage was single-blind: only the participants did not know what they were taking. At week 0, people were randomised, meaning put into groups by chance. They left the study at week 28 after final checks.

What did the AOD-9604 obesity trial find?

The company said OPTIONS did not meet its primary endpoint, the main result a trial is designed to test. Here that was weight loss compared with placebo after 12 weeks of treatment. The company said weight loss compared with placebo, at 12 or 24 weeks, was too small to reach statistical significance. Statistical significance means a result is unlikely to be down to chance alone. After allowing for the diet and exercise programme, the extra weight loss was under 1 kg in every treatment group. The company then stopped developing AOD-9604 for obesity. These figures come from a company statement, not a peer-reviewed paper.

The trial design picked out one subgroup before any results were known: women who lost little weight during the four-week run-in. The company said this subgroup showed effect sizes similar to the earlier trial, and that the full group did not respond in a consistent way. The company also suggested that large weight loss in the placebo group may have drowned out any drug effect.

A subgroup result from a trial that missed its main goal is a lead for more research. It is not a finding that the compound works in that group. No later trial in that subgroup was found in the sources read.

What did the earlier AOD-9604 trial report?

The same 2007 announcement describes an earlier phase 2b trial with no formal diet and exercise programme. After 12 weeks, all five AOD-9604 treatment groups lost more weight on average than the placebo group. The best group still fell short of statistical significance in the main analysis. Its p-value was 0.1, and 0.05 was needed; a smaller p-value means a result is less likely to be chance. That group did reach significance among women. For both sexes, the response was bimodal, which means people fell into two separate clusters. These weight-loss claims rest on the company's own account, not a journal paper.

OPTIONS was set up to test whether the earlier result would hold up in conditions closer to a phase 3 trial. Phase 3 is the large, final stage of testing before a regulator reviews a drug. The company said the result did not hold up across the whole group.

Which trial registry records exist for AOD-9604?

A trial registry is a public list where researchers record a trial's plan. ClinicalTrials.gov, the US registry, returned no records for AOD-9604 or AOD9604 on 16 October 2026. The Australian New Zealand Clinical Trials Registry (ANZCTR) has record ACTRN12605000067673. It names OPTIONS and the sponsor code METAOD006. That record lists a target of 360 people and no final count. The company's 2007 announcement said 536 enrolled. The 2013 safety paper said 534 enrolled and 502 were randomised. All three point to the same study, but the counts do not agree. A target number is not the number who actually took part.

The ANZCTR record was last updated on 14 September 2007. Its results and publications fields hold no uploaded or automatically found documents. An empty field does not prove that no paper exists: the 2013 safety paper can be read at its journal. The sources checked do not explain why the enrolled counts differ.

Who took part in the AOD-9604 obesity trial, and who was left out?

The registry lists otherwise healthy adults aged 18 to 65. Their body mass index (BMI), a measure of weight for height, was 30 to 45. Their waist was at least 102 cm for men or 95 cm for women. People were left out if they took regular medicine, or had diabetes or high blood pressure. Also left out: anyone with hypersensitivity, a strong reaction, to growth-hormone-related products, anyone with severe or many allergies, and anyone dieting or on a weight-loss programme in the three months before screening. The company gives the same age and BMI limits, uses greater-than wording for waist size, and reports about equal numbers of men and women.

A list of who was left out shows where the data are missing. It is not a safety finding about those groups. People with diabetes or high blood pressure, and children, were outside these records, so the trial says nothing about AOD-9604 in them.

Evidence limits

Does a mouse result show a human weight result?

The mouse work is still good science. It asked a clear question about how the peptide acts and recorded the answer, and how AOD-9604 works reports it. A 2026 review in Sports Medicine found that many unapproved peptides show good results in animal studies, while careful human safety data are scarce. The side-effects page covers the safety record.