Home › The Protocol Vault › AOD-9604

AOD-9604

HGH Fragment 176-191

Metabolic & Fat LossInjectableNasal🧪 Theoretical

AOD-9604 (HGH Fragment 176-191) is a metabolic & fat loss research compound. Modified C-terminal fragment of GH — stimulates lipolysis and inhibits lipogenesis without hitting IGF-1 or blood sugar.

Research & educational use only. The information below summarizes published research and mechanisms. It is not medical advice or a recommendation for human use. The protocol that uses it — dosing, sequence and what to retest — is inside Skool ($10/mo).

AOD-9604 quick facts

Reported research dosing (Injectable)250mcg-500mcg
RouteSubq
Cycle length8-12 Weeks
Frequency1x Daily AM · 5 On 2 Off or Daily
Half-life~30 min
FormsInjectable, Nasal
Evidence levelAnimal + limited human (modest)
Other forms availableNasal, Oral — dosed differently
Coach Cam’s take

Fat-loss lever without the GH/IGF baggage. Manage expectations — it's a nudge, not magic.

How AOD-9604 works

Modified C-terminal fragment of GH — stimulates lipolysis and inhibits lipogenesis without hitting IGF-1 or blood sugar.

Proposed benefits

Lipolysis / fat loss without GH- or blood-sugar-related side effects.

Where to get AOD-9604

AOD-9604 is sold in 3 forms. They are not interchangeable — the dose and the route differ, so pick the one this page describes unless you know why you want another.

AOD-9604 reconstitution calculator

Research reconstitution calculator

For research reconstitution math — 100 units = 1 mL on a U-100 syringe. Enter the vial size and acetic acid/bac water to convert a research amount into syringe units.
U-100 syringe
—
Enter the vial size to calculate

Bacteriostatic water is the diluent — sterile water with 0.9% benzyl alcohol, which is what lets a vial be drawn from more than once. It does not come with the vial, and unlike the compound it is bought again every time.

Need bacteriostatic water? Get it at AminoWell USA (my company) → Code CAMERON.

The evidence for AOD-9604

Graded by what exists behind each claim.

✅ Clinically validated

📊 Correlative data

🧪 Theoretical / extrapolated

How to read these tiers: they say how much human evidence exists, not how well something works — and ✗ flags harm, never a disappointing trial. How the evidence tiers work →

What AOD-9604 actually does

Start with the name, because the name on this site's card is not quite this molecule. The card's aka is ‘HGH Fragment 176-191’. The anti-doping laboratory that developed the urine assay describes AOD9604 as “a peptide consisting of the C-terminal fragment of human growth hormone from amino acids 177-191 with an additional tyrosine residue at the N-terminus” Cox 2015, and the peer-reviewed account of its human trials calls it a hexadecapeptide — sixteen residues Stier 2013. hGH 176-191 is also sixteen residues. So the two names describe sixteen-residue peptides that differ in their first residue: one begins with the natural growth-hormone residue, the other with a tyrosine that was put there. Identical length is exactly why the substitution passes unnoticed in a market that treats the two names as synonyms.

Why a tyrosine. Tyrosine is the residue you add to a peptide when you need to radioiodinate it for a binding or pharmacokinetic assay. An N-terminal tyrosine on a fragment that does not otherwise have one is a laboratory handle, and it is a reasonable inference — stated here as inference, not as a published claim — that the molecule sold today carries a modification that was originally there to make the molecule measurable.

What the fragment is supposed to do, and what it deliberately does not. Intact human growth hormone has a lipolytic action and a diabetogenic one, and they are separable in principle because they are carried by different parts of the molecule. AOD9604 is the lipolytic C-terminal domain without the receptor-binding regions that drive IGF-1 production and insulin resistance. Ng 2000 tested exactly that claim: chronic AOD9604 in obese Zucker rats showed no adverse effect on insulin sensitivity as measured by euglycemic clamp, unlike chronic intact hGH. That is a real, specific, method-named negative result and it is the strongest evidence the compound has.

The receptor question, and the answer is stranger than the marketing. The obvious candidate mechanism is the beta-3 adrenergic receptor, the main lipolytic receptor on a fat cell. Heffernan 2001 tested it properly, in beta-3-AR knockout mice. Both hGH and AOD9604 raised beta-3-AR RNA expression in obese mice, restoring it toward lean levels. But in knockout animals, long-term treatment with either failed to produce the weight change and the increase in lipolysis seen in wild-type controls — while in an acute experiment, AOD9604 still increased energy expenditure and fat oxidation in the knockouts. The authors' own conclusion: the lipolytic actions are not mediated directly through the beta-3-AR, although both compounds increase its expression. So the chronic effect needs a receptor the acute effect does not, and twenty-five years later nobody has named the acute one.

That is the honest mechanistic position. A sixteen-residue fragment with a demonstrated acute metabolic effect in an animal lacking the obvious receptor, a chronic effect that requires that receptor to be present, and no identified binding site of its own. Every vendor page states the mechanism with more confidence than the primary literature does.

Cell, rodent, human — and where it stops

Step one, rats, and by mouth. Ng 2000 gave obese Zucker rats an oral dose of 500 µg/kg/day for 19 days. Body weight gain was reduced by over 50% — 15.8 ± 0.6 g against 35.6 ± 0.8 g in controls — with increased lipolytic activity in adipose tissue and no adverse effect on insulin sensitivity by clamp. Read the endpoint precisely: these animals still gained weight. The drug slowed the gain; it did not produce loss.

Step two, mice, by injection, with a receptor knockout arm. Heffernan 2001: 14 days of chronic intraperitoneal administration reduced body weight and body fat in obese mice, correlating with increased beta-3-AR expression; the effect disappeared in beta-3-AR knockouts on chronic dosing but the acute energy-expenditure effect survived.

Step three, and this is the part the market has never been told: there were six human trials, they have names, and only their safety was ever published. Stier 2013 is the single peer-reviewed account of the entire clinical program, and it lists the studies by protocol code:

METAOD001 — phase I, dose-escalation, intravenous, 15 healthy males, 25 to 400 µg/kg. METAOD002 — phase IIa, intravenous, 23 obese males with BMI 35 or above, 25 to 100 µg/kg. METAOD003 — phase IIa, oral capsules, 17 obese males, 9 to 54 mg. METAOD004 — phase IIa, oral capsules, 36 obese males, 7-day multiple dose, 9 to 54 mg. METAOD005 — phase IIb, oral capsules, 300 obese adults, 12 weeks, 1 to 30 mg. METAOD006 — phase IIb, oral tablets, 502 obese adults, 24 weeks, 0.25 to 1 mg.

Now the three things that follow from that list, and each of them contradicts something the market says.

(1) Not one of the six used the subcutaneous route. Two were intravenous and four were oral. This site's card — and every vendor — describes a subcutaneous injection. The route people use has never been studied in a registered trial of this compound.

(2) The oral doses were in milligrams, and the injected doses were in micrograms per kilogram. METAOD005 ran 1 to 30 mg by mouth; the retail subcutaneous dose is 250 to 500 µg. The intravenous phase I went to 400 µg/kg, which in an 80 kg adult is 32 mg — sixty to a hundred and thirty times the retail dose, by a route with complete bioavailability.

(3) The efficacy results of METAOD005 and METAOD006 were never published in a peer-reviewed journal. Stier 2013 reports the phase IIb objective as “to assess the efficacy (reduction in body weight), safety and tolerability” and then reports safety only. What exists publicly about the outcome is a company announcement — Metabolic Pharmaceuticals' 2004–2007 updates on the phase 2b program, listed under Sources and carrying no citation chip here because a press release is not evidence. An 802-participant phase IIb program across two studies that produced no efficacy publication and no marketing application is a program that did not meet its endpoint. That is the honest reading, and it is confirmed by what happened next: the compound was repositioned as a nutraceutical ingredient rather than pursued as an anti-obesity drug.

The rest of the indexed human-relevant literature is not about weight at all. Kwon 2015 injected AOD9604 with and without hyaluronic acid into a rabbit knee osteoarthritis model — an intra-articular indication with no relationship to fat loss. Vanhee 2014 is a Belgian regulatory laboratory identifying AOD9604 in seized pharmaceutical preparations of unknown composition. Cox 2015 is a doping-control assay. And Dominikowski 2026 places it in the wider review of GH-axis performance peptides and the gap between clinical evidence and patient self-administration. There is no modern efficacy literature because there was never a modern efficacy program.

AOD-9604 pharmacokinetics — how much of it actually gets in

There is no published human half-life for AOD9604, and this site's card gives about 30 minutes. That figure is not in Stier 2013, which reports the trials' safety rather than their pharmacokinetics, and it is not in the indexed literature. What is published is more useful than a half-life anyway.

What degrades it, measured. Cox 2015 incubated AOD9604 in serum and in urine and identified six potential metabolites. Quantification in serum identified a single metabolite, consisting of amino acids CRSVEGSCG, which is significantly more stable than the other metabolites or the parent compound. Two conclusions fall out of that sentence. First, the parent peptide is short-lived in serum — that is what ‘more stable than the parent’ means, and it is the closest thing to a half-life statement that exists. Second, the surviving fragment retains two cysteines, which in the growth-hormone C-terminal region form a disulfide loop; a constrained ring resists exopeptidase attack from both ends, which is exactly why that piece outlives the linear tails.

The assay numbers, because they bound what is detectable. The validated urine method has a limit of detection of 50 pg/mL with recovery of 62% and precision better than 20% Cox 2015. AOD9604 is banned by the World Anti-Doping Agency, and the reason that laboratory built the assay around the stable metabolite rather than the parent is that the parent does not stay around long enough to catch.

The oral barrier, and the surprise is that there may not be one. Ordinarily a sixteen-residue peptide taken by mouth is destroyed by gastric acid and pancreatic proteases before it can be absorbed, and this site's default position on oral peptides is scepticism. AOD9604 is the awkward case: four of its six registered human trials were oral, including both phase IIb studies Stier 2013, and the rat efficacy work that started the program was oral too Ng 2000. Whoever designed those trials believed enough of an oral dose survived to be worth testing at 1 to 30 mg. The milligram-scale oral doses against microgram-scale intravenous doses in the same program imply an oral bioavailability in the low percent, which is exactly what you would expect — and it means the honest comparison for the 250–500 µg subcutaneous dose in circulation is the intravenous arm, not the oral one.

Do that arithmetic once, because it reframes the whole page. METAOD002 gave obese men 25 to 100 µg/kg intravenously Stier 2013. For an 80 kg adult that is 2 to 8 mg per dose, straight into a vein. The retail subcutaneous dose is 0.25 to 0.5 mg. Even assuming subcutaneous delivery is fully bioavailable, the community dose is four to thirty-two times smaller than the smallest dose tested in a phase IIa obesity trial — and that trial's program did not go on to publish an efficacy result at any dose.

What would have to be true, and how you would know it was not

Four predictions with a marker, a direction and a window. The first two are the compound's own selling points made falsifiable; the third is the one that cuts against it and it is the most likely outcome.

1. IGF-1 should NOT rise, and that is the whole claim. The entire pharmacological argument for a fragment rather than whole growth hormone is that it separates the lipolytic action from the somatotropic one. Draw IGF-1 at baseline and at 6 weeks. A rise means the vial does not contain what the label says — the obvious contaminant in this market is a growth-hormone secretagogue or hGH itself, and both raise IGF-1 reliably. This is a cheap identity test for the product, not just a test of the drug.

2. Fasting glucose and fasting insulin should not worsen. Ng 2000 measured insulin sensitivity by euglycemic clamp in chronically treated rats and found it unimpaired, which is the specific advantage claimed over intact hGH. Draw Fasting Glucose and Fasting Insulin at baseline and 6 weeks, and calculate HOMA-IR from them. Deterioration would contradict the one differentiating claim the molecule has.

3. The prediction that cuts against it: body weight should not change measurably at 250–500 µg subcutaneously. The reasoning is the arithmetic above — the retail dose is a fraction of the smallest dose given intravenously in a phase IIa trial Stier 2013 — combined with the fact that an 802-participant phase IIb program at far higher oral doses produced no published efficacy result. Weigh fasted, same time of day, weekly for 12 weeks, with diet held constant. If weight falls on this compound at this dose with diet genuinely unchanged, that is a result that contradicts the published program and is worth recording carefully. The far more likely finding is no change, and a page that would not say so is not worth reading.

4. Nothing on a standard panel should move at all. No liver enzyme, no lipid, no thyroid marker, no CBC parameter has ever been reported to shift on this compound, and Stier 2013 reports the safety profile across all six trials as “indistinguishable from placebo”. That is a genuinely reassuring safety statement and it is simultaneously the observation that a compound doing nothing detectable and a compound doing nothing look the same on a blood panel. Both readings are consistent with the data, and only the weight endpoint separates them.

What nobody has tested yet

Nobody has published the efficacy results of METAOD005 and METAOD006. Two randomized, placebo-controlled phase IIb trials, 300 and 502 participants, 12 and 24 weeks, with reduction in body weight as a stated objective Stier 2013. The data exist. They have never appeared in a peer-reviewed journal. Publishing them — even now, even negative — would settle the entire question this compound is sold on, and it is the single largest piece of deliberately unavailable evidence in this catalog.

Nobody has measured subcutaneous bioavailability. The route everybody uses has no pharmacokinetic study of any kind. A crossover in twelve people — intravenous on one day, subcutaneous on another, plasma AOD9604 by the existing validated mass-spectrometric method Cox 2015 — would produce the number that lets anyone compare a retail dose to a trial dose. It has never been done.

Nobody has tested the compound at the doses that were trialed. The market runs 250–500 µg subcutaneously. The trials ran milligrams orally and micrograms per kilogram intravenously Stier 2013. No one has ever asked what a trial-equivalent subcutaneous dose would even be, let alone given it to anybody and measured a waistline.

Nobody has followed up the intra-articular finding. Kwon 2015 injected AOD9604 into rabbit knees in an osteoarthritis model, with and without hyaluronic acid. That is an entirely different indication from the one this peptide is sold for, it is the only positive non-metabolic result in the literature, and there is no human follow-up at all. If this molecule has a future, that paper is a more plausible starting point than fat loss.

AOD-9604 — its own safety story, not its class's

The safety record here is genuinely good, and the honest caveat is what it is a record of. Stier 2013 covers all six registered trials — 893 subjects in total across METAOD001 to METAOD006 — and states that AOD9604 “displayed a very good safety and tolerability profile indistinguishable from placebo”, and that “in none of the studies did a withdrawal or serious adverse event occur related to intake of AOD9604”. For a peptide sold in this market, having a peer-reviewed multi-study safety paper at all is unusual, and it should be said clearly.

What that record does not cover. Those trials ran up to 24 weeks, used intravenous and oral routes only, and were conducted with pharmaceutical-grade material under a company's quality system. None of the three conditions applies to a person injecting a reconstituted research vial subcutaneously for a year. The safety statement is about a drug program; the risk in front of the reader is about a supply chain.

And that supply chain has a documented problem. Vanhee 2014 is a case report from the Belgian federal medicines laboratory, identifying and characterizing peptide drugs in unknown pharmaceutical preparations seized by the authorities, with AOD9604 as the worked example. Cox 2015 notes the peptide is available on several internet websites and was recently identified in confiscated vials in the USA. Two national laboratories, on two continents, doing forensic identification on unlabelled vials of this compound. That is the actual risk profile of buying it.

It is a banned substance in sport. AOD9604 is prohibited by the World Anti-Doping Agency and there is a validated urine method for it with a 50 pg/mL limit of detection, built around a metabolite chosen specifically to extend the detection window Cox 2015. Any tested athlete should treat that as decisive.

The theoretical risk that has not been excluded. The compound's claim is that it separates lipolysis from the growth-hormone receptor, and Ng 2000's clamp data supports the insulin half of that claim in rats. What has never been measured in a person is whether long-term dosing does anything to IGF-1 — the prediction section above proposes the draw — and the GH-axis review Dominikowski 2026 exists precisely because self-administration of this family has outrun the clinical evidence. The safety data says 24 weeks. Nobody has data on year three.

Sources read for this page

AOD-9604 — safety, predicted from mechanism

Predicted from mechanism, not from a human safety trial. How that reasoning works →

What the mechanism predicts

Derived from the molecule, not a trial.

What has actually been reported

How to reduce the risk

Same mechanism as the prediction.

What it does to your bloodwork

A fact about the assay.

What it overlaps with

Don't run this if

The honest unknown

Not medical advice. If you take prescription medication or have a diagnosed condition, check this with a pharmacist or doctor.

When to take it

Close to the tissue, and consistency beats the clock — fasted

A brief exposure starts a process that runs for days, so the hour you dose is a minor variable — missing days is the one that costs you. Where the target is local, dosing near the site is worth more than any timing choice.

The fasted flag on this one is real: food blunts absorption enough to matter.

From half-life and route, not a dosing trial.

AOD-9604 — interference & stacking

Predicted from mechanism, not from an interaction study. How mechanism-predicted claims are made →

What AOD-9604 moves on your bloodwork

Expected direction, not a measured one.

This class is where honest expectation-setting matters most: the markers above are how you find out whether anything happened, and for most of these compounds that question is genuinely open.

Everything on this page, in an order

This one is free and stays free. What Skool adds is the rest of the shelf — 278 compounds and 371 supplements with the protocol, the stack order and the bloodwork to run beside it.

Join Skool — $10/mo →

Bloodwork to run alongside AOD-9604

Baseline first, then again at 8–12 weeks.

MarkerWhat it’s watching for
HbA1c (Hemoglobin A1c)Where you started, so you can prove the change was real
Fasting InsulinMoves years before HbA1c does — the earliest signal you get
Lipid Panel (Cholesterol, HDL, LDL, Triglycerides)Rapid fat loss shifts triglycerides fast, in both directions
Comprehensive Metabolic Panel (CMP)Liver, kidney and electrolytes while intake is restricted

The Metabolic Health & Prediabetes panel covers these in one order — 8 markers, $81.45 with the discount applied.

Check results you already have → · All 103 markers A–Z

AOD-9604 — frequently asked questions

What is AOD-9604?

AOD-9604 (HGH Fragment 176-191) is a metabolic & fat loss research compound. Modified C-terminal fragment of GH — stimulates lipolysis and inhibits lipogenesis without hitting IGF-1 or blood sugar.

What dosing does the research reference for AOD-9604?

In the research literature, AOD-9604 is referenced in the 250mcg-500mcg range, 1x Daily AM · 5 On 2 Off or Daily. It is supplied as a lyophilized powder and reconstituted with acetic acid/bac water; the calculator above converts a research amount into syringe units. For research use only — not a recommendation for human use.

What is the half-life of AOD-9604?

AOD-9604 has an approximate half-life of ~30 min, which is part of what determines how often it's dosed.

What forms does AOD-9604 come in?

AOD-9604 is available as: Injectable, Nasal.

What's the evidence behind AOD-9604?

Current evidence level: Animal + limited human (modest). AOD-9604 is offered for research purposes only and is not an approved medicine.

AOD-9604 inside a finished plan

One arm of 1 Protocol Blueprint, free to read in full.

The Fat Loss Blueprint16 weeks · AOD-9604 runs as the direct-lipolysis arm

What AOD-9604 is used for

AOD-9604 appears under 1 goal in the goal router.

🔥 Lose fatDirect lipolysis & adrenergic drive

Where this goes next

The full protocol$10/mo

AOD-9604 is the direct-lipolysis arm of this plan. The page above is the free breakdown of one compound; the plan it belongs to — the dosing, the order to correct things in, the week-by-week schedule and what to retest — is a lesson inside Skool.

← Explore the full Protocol Vault

↑ Back to on this page