Areté
← All peptides
GH fragment (fat loss)clinical

AOD-9604

AOD-9604 (GH fragment 176-191)

A snipped-off tail of the growth-hormone molecule, sold as fat loss without the growth. The pitch is clean — the lipolytic piece of GH with none of the GH or IGF-1 baggage. The problem is the part the marketing skips: when this was actually put in front of obese humans, it didn’t beat placebo.

Area
Growth & metabolic
Class
GH fragment (fat loss)
Standard dose
~300 mcg / day
Evidence
clinical

What it is

People run it for fat loss — specifically the promise of burning fat without raising growth hormone, IGF-1, or touching blood sugar. The community use is a daily fasted shot as a lean-out adjunct alongside diet and training. Worth saying plainly up front: the human obesity trials that tested exactly this largely failed to outperform placebo, so what people are buying is a mechanism story, not a results story.

The whole idea is selectivity by subtraction. Full-length growth hormone raises IGF-1, can worsen insulin sensitivity, and carries the GH side-effect load — AOD-9604 is just residues 176–191 from the C-terminal tail (with a tyrosine tacked on for stability), the slice where GH’s fat-burning activity was thought to live. In animals it raised fat oxidation and trimmed weight without the GH/IGF-1 effects, which is a genuinely elegant result. The catch is that elegant-in-mice didn’t carry to humans, and that gap is the entire honest story of this peptide.

Mechanism

It’s a fragment of the C-terminal lipolytic region of human growth hormone, engineered to keep the fat-mobilizing activity while losing the receptor binding that raises IGF-1 and disturbs glucose. In obese mice it stimulated lipolysis (fat breakdown), suppressed lipogenesis (new fat storage), and increased fat oxidation — and the proposed link is the β3-adrenergic pathway: it restored suppressed β3-adrenergic-receptor expression in obese mice toward lean levels. Notably, even in the animal work the lipolytic effect wasn’t mediated directly through the β3 receptor — in β3-knockout mice the weight and lipolysis effects disappeared, so the receptor is necessary but the mechanism is indirect and not fully pinned down. None of this should be read as confirmed in humans.

Standard dose

Standard dose~300 mcg / day (proposed — pending dosing review)community
TimingMorning, fasted — the common framing is to dose before fasted activity, though the human data don’t validate thiscommunity
Frequency / routeOnce daily, SubQ (often into subcutaneous fat); oral was the form actually tested in the failed human trialscommunity
CycleRun in blocks of several weeks to a few months; there is no evidence-based protocolcommunity

Reconstitution calculator

U-100 · 100u = 1 mL
mg
mL

= 200 units

Concentration
2.5 mg/mL
1 mg equals
40 units
Draw to
12 units
05010012u

Set the vial size and water to match your product — amounts vary by supplier. This is unit-conversion math, not medical advice or a dosing recommendation.

Pushing higher— going beyond the standard dosecommunity
There’s no real community drive to escalate this one, and for an unusual reason: the honest problem isn’t that low doses are too gentle, it’s that the human trials didn’t show a meaningful effect at any dose. The failed multicenter obesity study tested 1 mg through 30 mg orally and the higher doses did not do better — if anything the low dose looked best in early data that later didn’t hold up. Pushing the milligrams chases an effect the trials couldn’t find at the bottom or the top of the range.

Side effects & cautions

This is the one area where AOD-9604 genuinely delivers: across the human trials it was very well tolerated, with a side-effect profile reported as essentially indistinguishable from placebo. It had no effect on serum IGF-1 and no negative effect on carbohydrate metabolism — the clean part of the pitch is real. The blunt caveat is that an excellent safety record on a compound that didn’t beat placebo for its main use is faint praise: it’s safe partly because it doesn’t do very much. As always in this space, the market is unregulated — insist on a certificate of analysis before running anything.

Stacking

There’s no evidence-based stack. In community use people fold it into fat-loss routines — fasted training, sometimes alongside a GH secretagogue stack (CJC-1295/Ipamorelin) on the theory that AOD handles lipolysis while the secretagogues handle GH — but none of that pairing rests on human data, and AOD-9604’s own efficacy data are negative to begin with. The honest version: it’s run as a hopeful add-on to diet and training, not as a load-bearing part of any protocol.

Evidence & sources

Unusually, real human trials exist — six randomized, placebo-controlled studies in over 900 people — and the efficacy result was essentially negative: it did not produce clinically meaningful weight loss versus placebo, and development was halted in 2007 after a longer trial failed. So the evidence base is strong, but what it shows is that the marketing ran far ahead of the data. The clean safety/IGF-1 findings did hold up; the fat-loss promise did not.

  • Heffernan M et al. (2001)Animal / in-vitro
    Human GH and its lipolytic fragment (AOD9604) on lipid metabolism in obese mice and β3-AR knock-out mice
    Endocrinology — animal (β3-adrenergic mechanism)PMID 11713213
  • Heffernan M et al. (2001)Animal / in-vitro
    Increased fat oxidation and weight loss in obese mice from a modified C-terminal GH fragment
    Int J Obes Relat Metab Disord — animalPMID 11673763
  • Stier H et al. (2013)Human study
    Safety and tolerability of the hexadecapeptide AOD9604 in humans
    J Endocrinol Metab — pooled analysis of six human RCTs (safety; no IGF-1 effect)DOI 10.4021/jem157w

More in Growth & metabolic