AOD-9604: What the Evidence Actually Shows

AOD-9604 is the most clinically-tested compound in this site's five-newest-additions batch, and that turns out to be a genuinely two-sided story: real safety data from real placebo-controlled human trials, alongside a pivotal efficacy trial that failed badly enough to end the drug's development entirely.

What it actually is

AOD-9604 is a synthetic fragment of the C-terminal end of human growth hormone — specifically the Tyr-hGH(177-191) region, the segment researchers identified as carrying hGH's fat-reducing ("lipolytic") activity without the rest of the hormone's growth-promoting effects. The founding mechanism paper, from a Monash University team, tested it directly against full-length hGH in obese mice: both reduced body weight gain and increased fat oxidation, but unlike hGH, AOD9604 did not induce high blood sugar and did not bind the growth-hormone receptor or drive cell proliferation in a receptor-binding assay¹. That is the entire design logic of the molecule — isolate the fat-burning signal, leave the growth-driving signal behind — and it is a real, published finding, not marketing language.

The safety data is real, and it's better than most peptides on this site can claim

A pooled summary covering six randomized, double-blind, placebo-controlled human trials of AOD-9604 reports a genuinely clean profile: no anti-AOD9604 antibodies detected in any patient tested for immunogenicity, no effect on serum IGF-1 levels (confirming the drug doesn't act through the IGF-1 pathway the way full hGH does), no negative effect on oral glucose tolerance, and no withdrawal or serious adverse event tied to the drug across all six trials. The paper's own conclusion: "a very good safety and tolerability profile indistinguishable from placebo"². Read that plainly: this is a stronger, better-controlled human safety record than IGF-1 LR3, Kisspeptin, or 5-Amino-1MQ have — three compounds this same research batch found have zero or near-zero human safety data of any kind.

The efficacy trial is where it fell apart

Safety is not the same question as "does it work," and on that second question AOD-9604 has a real, documented failure. The compound's developer, Metabolic Pharmaceuticals, ran a Phase IIb obesity trial and disclosed the result via a stock-exchange announcement in 2007: "trial results do not support commercial viability of obesity project: programme is terminated"³. That is worth naming honestly as what it is — a negative pivotal trial disclosed through a financial filing rather than a peer-reviewed publication, which is common for abandoned drug candidates but is itself a real transparency gap. No peer-reviewed report of that trial's actual weight-loss numbers was located for this article.

A route mismatch worth naming directly

Here is a detail that matters for anyone shopping today's market rather than reading a 2007 press release: the completed Phase IIb obesity trial dosed AOD-9604 as an oral tablet — 1 mg by mouth, once daily, for 12 weeks³. That's genuinely unusual for a peptide fragment (most don't survive oral administration well), and it's also not what compounding sellers primarily sell today. A live check of one such vendor's own catalog lists AOD-9604 as an injectable vial first, with an oral troche as a secondary option. So the one human efficacy trial this compound has ever completed — the one that failed — tested a route of administration that isn't what most of today's market actually sells. Whatever the injectable version does in a person has not itself been tested against placebo in any located study.

A separate investigated use, with even less behind it

AOD-9604 also shows up in a completely different research context: intra-articular injection for osteoarthritis, on the theory that its lipolytic mechanism might have local cartilage benefits distinct from its systemic fat-burning one. The only study located is in rabbits — collagenase-induced knee osteoarthritis, AOD9604 injected with or without hyaluronic acid, weekly for 4-7 weeks. The combination reduced cartilage degeneration scores and shortened lameness compared to saline or hyaluronic acid alone⁴. That is a real, positive preclinical finding, and it is exactly that: preclinical, in rabbits, for a joint indication with zero human data of any kind — not the same claim as the obesity research, and not evidence either use works in a person.

The regulatory picture: nominated, flagged, then withdrawn

AOD-9604 does not appear anywhere in FDA's current 503A Bulks List — checked directly against the May 14, 2026 update, no match in Category 1, 2, or 3⁵. It does appear on FDA's companion page describing substances with documented significant safety concerns, specifically in that page's section for nominations that were later withdrawn. FDA's own stated language there: compounded AOD-9604 "may pose significant risk for immunogenicity for certain routes of administration," the agency has "identified no, or only limited, safety-related information," and — the sharper line — "FDA has also identified serious adverse events that may be associated with AOD-9604, though causality is not clear"⁶. Both facts belong in the same sentence: the nomination was withdrawn (which is why AOD-9604 isn't on the active list today), but FDA's own historical safety-concern language about it is real and still published. DailyMed returns zero results, confirming no FDA-approved product exists under any name⁷; openFDA's FAERS database returns zero adverse-event reports under every naming variant tried⁸ — a null result that reflects thin real-world reporting, not a demonstrated all-clear.

What this means if you're considering it

None of the 31 providers this site's own reviews cover sell AOD-9604 as a primary, boarded product — this site's own identity-layer research confirms it is not tagged to any reviewed seller. If you're weighing AOD-9604 specifically for its safety record, that record is genuinely better-supported than most compounds this site covers outside its three boarded molecules. If you're weighing it for weight loss specifically, the one trial built to answer that question failed, was never published in a peer-reviewed journal, and tested a different route of administration than what's commonly sold today. Our IGF-1 LR3 evidence review covers a different GH-axis-adjacent compound with the opposite problem — no human trial of any kind, safety or efficacy — which makes AOD-9604's actual, if negative, trial record the more informative of the two rather than the more reassuring one. This site's reconstitution calculator does the vial-to-syringe arithmetic for the injectable form the market actually sells; it cannot confirm that form has ever been tested against placebo, because the trial that exists tested a tablet.