AOD-9604 — what the clinical trials actually showed (and didn't)

Understanding it requires going back to the actual program and looking at what was studied, in whom, for how long, and with what result.

Metabolic Pharmaceuticals, the Australian biotech that licensed AOD-9604 from Monash University, ran the compound through a formal pharmaceutical development program that reached Phase II clinical trials in the mid-2000s. This is a meaningful level of development. Reaching Phase II means the compound cleared Phase I safety work, was considered plausible enough to invest in human efficacy testing, and had enough institutional backing to run multi-site, controlled trials in real patients. That backstory matters because many compounds in the compounding peptide space have no human trial data at all. AOD-9604 is in a different category.

The Phase IIa trials were dose-finding and safety studies. Patients with obesity received AOD-9604 orally — the development program favored an oral formulation, specifically a tablet or capsule, because oral administration is commercially preferable to injection for a mass-market obesity drug — at doses ranging from 1 milligram to 10 milligrams per day, compared to placebo over twelve weeks. The safety data from Phase IIa was encouraging: no meaningful adverse signals, no IGF-1 elevation, no glucose disruption. The compound appeared to be well-tolerated across the dose range studied.

The efficacy data from Phase IIa was more complicated. Weight loss occurred in the treatment groups. It exceeded placebo. But the magnitude of the difference was modest — in the range of a few kilograms over twelve weeks, depending on dose and specific study arm. The dose-response curve wasn't cleanly linear, which raised questions about optimal dosing. The 1 mg and 10 mg groups didn't show a consistent step-up in effect that would make dose selection straightforward.

Phase IIb expanded the patient numbers and refined the dose range based on the Phase IIa findings. The design was consistent: oral administration, obese patients, twelve-week duration, placebo-controlled. The primary endpoint was weight change. The results were in the same range as Phase IIa: modest weight loss exceeding placebo, with statistical significance in some analyses and marginal significance in others depending on how the data was cut. A few kilograms of additional weight loss relative to placebo over three months is a real effect. It is also, by the standards required to advance an obesity drug through Phase III and toward regulatory approval, insufficient.

To put this in context: the GLP-1 receptor agonist drugs that now define the obesity pharmacotherapy landscape produce weight loss of fifteen to twenty-five percent of body weight over twelve to eighteen months in clinical trials. Even the earlier, less potent obesity drugs that have received FDA approval over the past decade showed effects in the eight to ten percent range that AOD-9604's Phase II data could not match. The comparator wasn't unfavorable science — it was a threshold of effect size that the compound's biology simply didn't clear. Metabolic Pharmaceuticals discontinued the obesity drug development program after Phase II. It was a commercial and regulatory decision, not a safety one.

The Phase II data has a specific scope that matters enormously for interpreting what it means outside that context. The patients were obese adults — BMI above 30, in most study arms. The intervention was oral administration. The duration was twelve weeks. The outcome measured was weight. What this data does not tell you — because it was never designed to — is how AOD-9604 behaves in people who are already lean or in normal weight ranges, over longer time periods, via subcutaneous injection rather than oral administration, or in the context of concurrent training and dietary interventions. These are the conditions under which the compound is now most commonly used and discussed, and the clinical trial data has essentially nothing to say about them. Not because the research was bad, but because the research was designed for a different purpose.

After the obesity program ended, the compound's research history shifted in an interesting direction. Phosphagenics, another Australian company, acquired rights to explore AOD-9604 for musculoskeletal applications — specifically, cartilage regeneration and osteoarthritis. The hypothesis was that the fragment might interact with fat-precursor cells (preadipocytes) in joint tissue in ways relevant to cartilage health. Preclinical work in animal models was conducted. This research arc never reached the clinical evidence standard either, but it produced publications and generated some discussion about AOD-9604's potential activity beyond adipose tissue — a secondary narrative that has filtered into compounding discussions in various forms.

The regulatory story in Australia adds a layer. The Therapeutic Goods Administration, which had initially permitted AOD-9604 to be compounded under certain categories, revised its scheduling in the early 2020s. This reflected a broader review of the peptide compounding landscape and placed the compound in a category with tighter restrictions. The TGA's position is not that AOD-9604 is dangerous — the safety profile from the clinical program doesn't support that — but that it exists in a space where compounding without robust evidence of efficacy for specific indications warrants more regulatory scrutiny. In the United States, the regulatory framing is different: AOD-9604 is not FDA-approved for any indication, is not on any approved drug list, and is accessible only through compounding pharmacies under provider prescription, operating in the legal space that exists for compounded preparations of non-approved compounds.

The honest assessment, sitting with all of this, is something like this: AOD-9604 has more clinical research behind it than most peptides circulating in compounding contexts. The Phase II trials are real, the safety data is real, the modest weight loss effect in obese patients is real. The compound's inability to advance past Phase II reflects the commercial and regulatory standards of pharmaceutical development, not a finding that it doesn't work at all. These are genuine distinctions.

And at the same time, the gap between what the trials showed and how AOD-9604 is currently framed in many compounding discussions is significant. The trials studied obese patients taking oral formulations for twelve weeks and found modest additional weight loss. The current conversation often positions AOD-9604 as a targeted body recomposition tool for already-lean, trained individuals taking subcutaneous doses over extended periods. That population was never studied. That administration route was not the trial route. Those outcomes — body composition at lower body fat percentages, regional fat distribution changes, lean mass preservation — were not measured. The scientific backstory is being applied to a context it was never designed to address.

This is not unique to AOD-9604. It is the standard condition of compounded peptides: a compound develops a research history in one context, gets picked up by a compounding ecosystem that applies it in adjacent contexts, and the evidentiary justification for the adjacent use is somewhere between plausible extrapolation and wishful thinking. For AOD-9604, the extrapolation is more plausible than most, because the mechanism is at least coherent and the safety data is reasonably robust. But plausible is not proven, and the consumer enthusiasm around the compound substantially outpaces what two decades of research actually established. Holding both of those things at once — this is real science, and this is also less than what's being claimed — is the accurate position.