AOD-9604 for Joints: The Cartilage Evidence

Almost everything written about AOD-9604 concerns fat loss. That is what the molecule was designed for, and it is what our AOD-9604 data sheet is mostly about. Over the last few years a second pitch has grown alongside it: that the same peptide belongs in a knee, for cartilage. That claim is younger and much thinner than the fat-loss one, and it is easier to check because there is so little to check. Traced to its source, the joint case rests on a single animal study in rabbits. There is no completed human trial of AOD-9604 in knee osteoarthritis anywhere in the public registries, and the one registered human trial of the molecule in its modern form is about nerve pain in the lower back, not cartilage.^{[1, 3]}

What AOD-9604 actually is

AOD-9604 is a synthetic fragment of human growth hormone rather than the hormone itself. Doping-control chemists describe it as the C-terminal fragment spanning amino acids 177 to 191, with an extra tyrosine residue at the front. The original work at Monash University in Australia, around the turn of the century, isolated that tail end because it appeared to carry the fat-metabolism activity without the blood-sugar effects of the full hormone, in obese mice and rats. That lineage is why joints entered the conversation at all. Growth hormone signalling is bound up with cartilage biology, so a fragment of it is at least a plausible thing to point at a joint. Plausible is where an argument starts, not where it ends. HGH fragment 176-191 versus AOD-9604 untangles the two names that get confused most often.^{[2, 8]}

The one joint study worth reading

In 2015 two rehabilitation-medicine researchers at the Catholic University of Daegu, Dong Rak Kwon and Gi Young Park, published a controlled experiment in the Annals of Clinical and Laboratory Science. They induced knee osteoarthritis in 32 mature New Zealand white rabbits using collagenase, then gave weekly ultrasound-guided injections in the knee joint over four to seven weeks. There were four arms: saline, hyaluronic acid alone, AOD9604 alone, and AOD9604 combined with hyaluronic acid. The readouts were gross morphology, histopathology, and how long the animals limped. Cartilage damage scores were significantly worse in the saline arm than in any of the three active arms. The combination arm scored significantly better than either hyaluronic acid alone or AOD9604 alone, and the limping period in the combination arm was significantly shorter than in the other three. The authors concluded that the combination was more effective than either component by itself in this model. That is a real result, properly controlled, with a saline comparator arm. It is also the whole joint literature.^[1]

What that study can and cannot tell you

Three things about it are worth holding onto. First, it is a rabbit study using a chemically induced model. A collagenase injection in a joint produces cartilage breakdown quickly and reproducibly, which makes it a good screening tool and a poor stand-in for human osteoarthritis, which develops over decades under mechanical load. Effects that look clean in that model routinely fail to carry over. Second, the standout arm was the combination, not AOD-9604 alone. Hyaluronic acid is already a widely used joint injection, so the interesting finding is about what AOD9604 added on top of an existing option, in rabbits, over eight weeks. Vendors describing AOD-9604 as a cartilage compound in its own right are quoting the wrong arm of the study. Third, it measured cartilage appearance and limping over eight weeks, which is not long enough to say anything about durability, and there is no equivalent measurement in a person at all. How to read a peptide study covers why the gap between an animal model and a knee matters so much.^[1]

The human trial that exists is not about knees

The molecule did not stop being developed. Under the code LAT8881, the Australian company Lateral Pharma took it into a registered human study, NCT05298306, a two-part proof-of-concept trial. Part A was an ascending-dose safety run in healthy volunteers; Part B a placebo-controlled crossover in patients with lumbar radicular pain, the nerve pain commonly called sciatica. It enrolled 26 people, was listed as Phase 1, and completed in June 2023. That is a genuine registered human trial and worth knowing about. It is also not a knee trial, not a cartilage trial, and not an osteoarthritis trial: its endpoint was pain, in a different condition, by a different route. The registry record is the only part of that programme we can verify, and it says nothing about joints.^[3]

What we searched for and did not find

Because the joint claim is repeated so confidently, we went looking for the study behind it. ClinicalTrials.gov returns no records at all for AOD9604 or AOD-9604 as an intervention. PubMed returns 23 records, searched on 26 July 2026: mostly doping-control analytical chemistry, several obesity drug-development round-ups, a handful of the original rodent metabolic work, and exactly one on joints, the rabbit study above. None reports a completed human efficacy trial in any indication, including obesity, the indication the compound was actually developed for. Europe PMC returns 24 and adds nothing new. One more thing is worth flagging, because readers will meet it. A 2020 human paper on intra-articular AOD9604 in knee osteoarthritis, supposedly in a journal of joint surgery, circulates in vendor copy and in AI-written summaries. We searched for it by title, by author, and by keyword in both PubMed and Europe PMC. No such record exists in either database. We are not citing it, and neither should anyone else.^[9]

An ingredient listing is not an approval

The most common piece of misdirection around this compound is regulatory. The FDA has not approved AOD-9604 for any use, in joints or anywhere else. What did happen is that in June 2012 the Australian company Calzada Limited told the stock exchange that its subsidiary had obtained a conditional GRAS determination for AOD9604, GRAS standing for Generally Recognized As Safe. Its own chief executive called it a self-affirmed conditional GRAS status. A company-convened panel of experts made that determination, it concerned adding a very small amount of the peptide to foods and drinks in the United States, and full status was contingent on publishing the existing safety data in a peer-reviewed journal. So the GRAS story is a food-ingredient safety judgement about trace amounts in food and drink, reached by a panel the company assembled. It is not a regulator's finding that the peptide does anything for a knee. Those are not the same claim. The agency's own position is less flattering. AOD-9604 appears on the FDA list of bulk drug substances that may present significant safety risks in compounding, under the heading of substances nominated but withdrawn. The entry cites possible immunogenicity risk for certain routes, complexities around peptide-related impurities, and no, or only limited, safety-related information. On that basis the agency says it lacks enough information to know whether the drug would cause harm in humans, and adds that serious adverse events have been identified that may be associated with AOD-9604, though causality is not clear.^[4]

Sport status, side effects and supply

AOD-9604 is named on the WADA 2026 Prohibited List under section S2.2.3, growth hormone fragments, alongside hGH 176-191. Everything in the S2 class is prohibited at all times, in and out of competition, and is a non-Specified substance, the category carrying the heavier default sanction. Are peptides banned in sports covers the wider list. On side effects the picture is thin in exactly the place it matters. A 2026 review in Frontiers in Endocrinology, which sorts performance peptides by evidence tier down to a complete absence of human studies, groups the reported problems across this class as hormonal and metabolic disturbances, fluid retention, joint and muscle aches, and injection-site reactions. A 2026 orthopaedic review in JAAOS Global Research and Reviews puts it plainly for this family: the preclinical work is interesting and clinical trials are lacking. On top of that sits the ordinary gray-market problem: vials sold for laboratory research use only carry no assurance of purity, endotoxin load, or label accuracy, and a fragment peptide is easy to get wrong. Gray-market peptides covers what that means in practice.^{[5, 6, 7]}

Where this leaves the joint question

The cartilage case for AOD-9604 consists of one well-run rabbit experiment in which the peptide's best showing came alongside an injection that already exists. Everything else said about knees is extrapolation from that, from the growth hormone family tree, or from nothing at all. That is not the same as calling the idea worthless. A controlled animal result is exactly the reason to run a human trial, and the molecule has an unusually long safety history for a gray-market peptide. It is simply not evidence about a human knee, because that study has not been run. If you are dealing with joint pain, the useful next step is a clinician who can look at the actual joint. Osteoarthritis has options with real human evidence behind them, and a peptide with one rabbit study is not competing with those on equal terms. Our peptides for joint pain guide compares the class at the symptom level.

References & sources

1. Kwon DR, Park GY. Effect of Intra-articular Injection of AOD9604 with or without Hyaluronic Acid in Rabbit Osteoarthritis Model. Ann Clin Lab Sci. 2015;45(4):426-432. Thirty-two mature New Zealand white rabbits, collagenase-induced knee osteoarthritis, four arms (saline; hyaluronic acid; AOD9604; AOD9604 plus hyaluronic acid), weekly injections over four to seven weeks. Morphological and histopathological scores were significantly higher in the saline group than in the three active groups and significantly lower in the combination group than in the hyaluronic acid or AOD9604 groups; the lameness period was significantly shorter in the combination group. PMID 26275694.
2. Cox HD, Smeal SJ, Hughes CM, Cox JE, Eichner D. Detection and in vitro metabolism of AOD9604. Drug Test Anal. 2015;7(1):31-38. Describes AOD9604 as the C-terminal fragment of human growth hormone from amino acids 177 to 191 with an additional tyrosine residue at the N-terminus, notes it is banned by WADA, and reports it was identified in confiscated vials in the USA. PMID 25208511.
3. ClinicalTrials.gov NCT05298306. Lateral Pharma Pty Ltd. A Two-part Proof-of-Concept Study Assessing the Safety and Efficacy of LAT8881 in Lumbar Radicular Pain. Part A ascending intravenous doses in healthy volunteers; Part B placebo-controlled crossover in patients with lumbar radicular pain. Phase 1, 26 participants enrolled, completed 16 June 2023. Condition listed as lumbar radiculopathy, not osteoarthritis.
4. US Food and Drug Administration. Certain Bulk Drug Substances for Use in Compounding That May Present Significant Safety Risks. Content current as of 22 April 2026. AOD-9604 is listed under bulk drug substances nominated but withdrawn, citing possible immunogenicity risk for certain routes, complexities around peptide-related impurities and active ingredient characterization, no or only limited safety-related information, and serious adverse events that may be associated with AOD-9604 though causality is not clear.
5. World Anti-Doping Agency. 2026 Prohibited List, in force 1 January 2026. Section S2.2.3 names growth hormone fragments including AOD-9604 and hGH 176-191; the S2 class is prohibited at all times and its substances are non-Specified.
6. Dominikowski A, Rekos Z, Olejarz M, Szczepanek-Parulska E, Domin R, Ruchala M. The emerging landscape of performance-enhancing peptides modulating GH-IGF1 axis: bridging the gap between clinical evidence and patient self-administration. Front Endocrinol (Lausanne). 2026;17:1822475. Narrative review including AOD9604 (hGH 176-191); stratifies peptides into evidence tiers down to a complete absence of human studies and summarises class-level adverse effects. PMID 42395176.
7. Rahman OF, Lee SJ, Seeds WA. Therapeutic Peptides in Orthopaedics: Applications, Challenges, and Future Directions. J Am Acad Orthop Surg Glob Res Rev. 2026;10(1):e25.00236. Reviews AOD-9604 among growth hormone secretagogues in an orthopaedic context and states that although preclinical studies are promising, there is a current lack of clinical trials. PMID 41490200.
8. Ng FM, Sun J, Sharma L, Libinaka R, Jiang WJ, Gianello R. Metabolic studies of a synthetic lipolytic domain (AOD9604) of human growth hormone. Horm Res. 2000;53(6):274-278. The originating Monash University work on the lipolytic domain, conducted in obese Zucker rats. PMID 11146367.
9. PubMed record count run for this article on 26 July 2026: the query AOD9604 OR AOD-9604 returns 23 records, of which one concerns joints (the rabbit osteoarthritis model above) and none reports a completed human efficacy trial.