AOD-9604 Dosage: How It's Used

The short answer on AOD-9604 dosing

Most “dosage” pages for a research peptide have to start by admitting that no one ever ran a proper human trial, so there’s no official dose to report. AOD-9604 is the unusual case where the opposite is true — and it ends up in the same place for a sharper reason.

AOD-9604 (“Anti-Obesity Drug 9604”) was actually built and tested as a prescription weight-loss medicine. It went through a full human development program: several trials, more than 900 people, all the way to a pivotal Phase 2b study. So a real clinical dose record exists. The problem is what that record says. The most-cited early result came from a small, modest signal, the larger study didn’t separate from placebo, and development was abandoned in 2007. There is no dose that was shown to work well enough to standardize around.

There’s a second problem that almost every online “AOD-9604 dosage guide” quietly skips: the doses the human program used were oral tablets, while the product sold and injected today is a subcutaneous injection. The microgram-per-day injection schedules you’ll find on forums and vendor pages were never the basis of any human trial. They are numbers written for a route the drug’s human program never validated, applied to a molecule that failed.

So this page does not give you a per-injection number, a frequency, a titration ladder, or reconstitution math. Those things would read as a self-injection recipe for an unapproved, unregulated injectable — exactly the thing that isn’t safe to put on a public page. Instead, here is how dosing is genuinely decided, why there is no real “standard” for AOD-9604, and why copying a figure from a website is the wrong starting point.

How dosing is actually determined

For any legitimately prescribed medicine, a dose isn’t a fixed internet figure — it’s a clinical decision a licensed prescriber makes for one specific person and then adjusts over time. It depends on the goal of treatment, the person’s age, weight, health history and lab work, how they respond, and what monitoring shows. The same number can be right for one patient and wrong for another, which is the entire reason a prescriber is in the loop.

Crucially, dosing is paired with oversight: a baseline, defined checkpoints, objective measures, and a willingness to change course or stop. A dose with no evaluation behind it and no monitoring around it isn’t really a dose in any medical sense — it’s a guess you’re running on yourself.

Note: A precise-looking microgram figure is one of the most persuasive things on a peptide vendor page, because precision looks like science. But a number is only meaningful if it was tied to a measured effect, in the form and route it’s being used, in a verified product. With AOD-9604, none of those three conditions hold.

Why there’s no “standard” AOD-9604 dose

The honest reason there’s no standard dose is that the molecule failed the test it was designed to pass.

AOD-9604 is a 16-amino-acid fragment of human growth hormone — the lipolytic “tail” of the GH molecule — engineered to trigger fat metabolism without the rest of growth hormone’s effects (no IGF-1 rise, no glucose disturbance). Designed at Monash University and developed by Metabolic Pharmaceuticals in Australia, it was carried through six human clinical trials. An early 12-week study produced a small signal: roughly a couple of kilograms of weight loss above placebo. But the pivotal Phase 2b OPTIONS study, which enrolled more than 500 obese adults over 24 weeks, did not separate from placebo on its primary weight-loss endpoint. The detailed results were never published in the peer-reviewed literature, and the program was discontinued in 2007.

Two features of that record matter specifically for the dosing question:

The effect was not dose-dependent. Higher doses did not produce more weight loss — and in at least one comparison a higher-dose group did worse than a lower one. That directly contradicts the instinct behind every escalating internet protocol. With most drugs you can at least reason about climbing toward a stronger effect. Here, the data give you nothing to climb toward; there’s no clean dose-response curve, and “more” was never shown to mean “more results.”

The effect didn’t survive a controlled test. The most rigorous study — the one with the diet-and-exercise structure that exposes whether a drug adds anything beyond lifestyle — is the one that showed no clear benefit. Early, smaller signals that vanish under tighter control are exactly what a drug that doesn’t really work looks like.

Put together: you cannot have a “right dose” of a compound that has no demonstrated effect to dose toward. The dosing question is, in a real sense, already answered — and the answer is that the doses that were tested didn’t work.

The route problem: the trial doses weren’t injections

This is the part most “dosage” content gets wrong, and it’s the single most important thing to understand before trusting any number.

AOD-9604’s human studies used oral dosing — tablets, taken by mouth — supported by a couple of intravenous pilot studies. The famous early “weight loss above placebo” figure came from an oral dose of about a milligram a day. The obesity efficacy program was an oral-tablet program, full stop.

The product sold on the gray market is a lyophilized powder reconstituted and injected subcutaneously. So when a vendor page presents a fixed daily microgram injection as “the clinically studied dose,” it’s quietly performing a swap: taking an oral trial number and relabeling it as an injection schedule. Those are not interchangeable.

• Oral and injectable doses don’t convert. A peptide taken by mouth has poor and variable absorption — most of it never reaches the bloodstream intact, which is part of why oral peptide drugs are hard to develop at all. An injected dose bypasses that entirely. There is no honest way to say “the trial used one milligram orally, so inject this much,” because the amount that actually reaches your system from each route is completely different. The number doesn’t carry across.
• The injected route was never the efficacy paradigm. The subcutaneous, microgram-twice-a-day, “loading phase then maintenance” pattern you’ll see online is a community convention assembled by analogy to other injectable peptides. It was not the design of any AOD-9604 efficacy trial. It’s a protocol for a use the human program never tested.

So an “AOD-9604 dosage” lifted from a forum is doubly detached from evidence: it’s a number for the wrong route, attached to a molecule whose right route already failed.

Why fixed internet protocols are unsafe

Even setting aside that the underlying drug failed and the route is wrong, the practical danger of a copied protocol comes down to what’s actually in the vial.

Gray-market AOD-9604 is an unregulated injectable of unknown concentration and purity. Independent context across the peptide market is consistent: vials are frequently underdosed, overdosed, degraded, mislabeled, or contaminated, and the molecule sold as “AOD-9604” isn’t always even the same modified fragment that was studied. A microgram figure assumes the vial contains exactly what the label claims, measured exactly the way you draw it. When the product is unverified, the most carefully copied number is still meaningless — the right dose of the wrong or contaminated product is still wrong.

A few specific traps worth naming:

• The “loading phase” borrowed from GH peptides doesn’t apply. AOD-9604 has a short half-life and, by design, doesn’t raise IGF-1, so there’s no growth-hormone-style accumulation or recomposition build-up to “load” toward and then “maintain.” A front-loaded cycling schedule is borrowed logic from a different class of compound.
• “More to feel it” backfires twice. The trial data show no clean dose-response, so pushing the dose adds cost and risk without a reason to expect more benefit — and on an unverified product, escalating into an unknown concentration is how people accidentally take far more, or far less, than they think.
• Nobody is monitoring. A legitimate dose comes with a baseline and follow-up. A schedule copied from a website comes with neither. “Just inject this and see how you feel” removes the only part of dosing that protects you.

What a legitimate process looks like — and the red flags

If you strip away the protocols, the responsible version of “how it’s used” is short:

A real provider doesn’t hand you a number off a chart. They evaluate the person, define what they’re actually treating, choose an approach individualized to that person, set objective checkpoints, and stay willing to change or stop based on what the checkpoints show. The dose is the least portable part of that — it’s specific to the patient and meaningless without the rest.

With AOD-9604 there’s also a hard legal wall around even that. The peptide isn’t on the FDA’s list of bulk drug substances that compounding pharmacies may use, and it was reviewed and effectively declined rather than swept into the 2026 reclassification many vendors point to. A provider can write the name on a script, but a 503A pharmacy generally has no lawful basis to fill it. That makes one red flag unusually clean here:

• A clinic confidently selling you a fixed AOD-9604 dose to inject in 2026 is operating outside the system, not inside it.
• “It became legal again with the 2026 reclassification” is false for this molecule specifically.
• A dosing chart sitting next to a “buy” button is marketing, not medicine.
• “Research use only” on a product meant to be injected into a person is a contradiction, not a loophole.

Where this leaves you

The reason there’s no AOD-9604 dosing protocol on this page isn’t caution for its own sake. It’s that an honest protocol would require three things the molecule doesn’t have: a demonstrated effect to dose toward, the route the numbers are written for, and a verified product to put in the syringe. AOD-9604 has none of them.

For the current US legal picture — why AOD-9604 sits outside the compounding pathway, and how that differs from the wellness peptides still under active review — see are peptides legal in the US? and the 2026 FDA peptide reclassification.

And if the actual goal is fat loss rather than this specific peptide, it’s worth being clear about the contrast. The whole appeal of AOD-9604 was that it was a real obesity drug candidate — and it’s the FDA-approved GLP-1 medicines that now hold what AOD-9604 was chasing: large published trials, meaningful average results, and dosing that a prescriber sets and titrates for the individual. That’s the validated, monitored dosing this molecule never earned. See semaglutide for weight loss for what that route involves.

This page is educational and not medical advice. It does not provide dosing instructions. Legal and regulatory status is current as of the date above and may change; any decision about treatment belongs with a licensed clinician.