Hey,
GLP-1s changed the weight-loss conversation faster than almost anything in recent memory. Semaglutide (Ozempic) and Tirzepatide (Mounjaro) went from specialist prescriptions to dinner-table topics in about two years. That shift has produced a second wave of questions I see rather consistently now, and they are generally more specific than "how do I lose weight."
They sound like this. How do I lose fat without losing my appetite entirely? What do I do about the last stubborn layer around my midsection? What happens when I come off the GLP-1, or when I can't afford it, or when the nausea isn't worth it? And under all of it: is there something that targets fat directly, without rewiring my hunger?
The peptide market has an answer ready for that last question, and it has been selling it for years. It is a small fragment of human growth hormone, marketed as the fat-burning piece of GH with none of the downsides of GH. It has a name that sounds like a lab code because it is one: AOD, for "anti-obesity drug."
What makes it worth an issue is that someone did the thing this newsletter keeps asking for. They took it into proper human trials, with real sample sizes, a placebo arm, and a pre-specified endpoint. Then they published what happened. That almost never happens in this corner of the market, so it is worth talking about.
Let's get into it.
🔬 The Lead: AOD-9604, the Peptide That Got Its Trial
Where it came from. AOD-9604 was developed in the 1990s at Monash University in Australia by researchers studying how growth hormone regulates fat. Full GH does many things, including raising IGF-1, affecting blood sugar, and driving growth. The idea was to isolate the part of the molecule tied to fat metabolism and leave the rest behind. The result is a 16-amino-acid synthetic peptide based on the tail end of human growth hormone with a tyrosine (amino-acid) added to stabilize it.
The pitch, then and now. In animal work it looked strong. Obese mice lost fat, lean mass appeared to hold, and the fragment did not seem to raise IGF-1 or disturb glucose the way full GH does. That profile is still the core of every vendor page today: fat loss without the GH baggage. The mouse results were real. The question was always whether they would carry over to humans.
The part almost nobody tells you. An Australian company, Metabolic Pharmaceuticals, licensed the compound and ran it through a clinical program. A 12-week Phase 2a trial looked promising: treated participants lost about 2.6 kg versus 0.8 kg on placebo. That was a small study, roughly two dozen people per arm, but it was encouraging enough to justify a larger one.
The larger one is the result that matters. It was a 24-week Phase 2b trial in 536 people with obesity, randomized across three oral doses or placebo. None of the dose arms produced statistically significant weight loss compared with placebo. As a result the company ended AOD-9604's development as an obesity drug in March 2007.
This distinction is the center of the issue. Compounds like BPC-157 and TB-500 have a missing-evidence problem: nobody ran the trials. AOD-9604 has the opposite problem. Someone ran the trial, and the answer was no. A well-powered, placebo-controlled null result is not an absence of evidence. It is evidence of absence, at least for the question it asked. When the small early study and the large later study disagree, the large one usually wins, and here it was clear.
What the trials did show. Across roughly six trials and about 900 participants, AOD-9604's safety profile looked broadly similar to placebo. That is a real finding and deserves credit. It comes with two caveats that vendor pages skip:
Short duration. These were short trials, not long-term safety data.
Different route. The large obesity trials tested the peptide taken by mouth. The gray-market product is almost always injected. Safety data from one route does not automatically transfer to another.
The GRAS confusion. You will often see AOD-9604 described as "GRAS," generally recognized as safe, as though that were an FDA endorsement. It is not. The GRAS status was self-affirmed by an industry expert panel for limited use as a food ingredient. It says nothing about injecting the compound, and nothing about whether it works.
Where it stands now. In December 2024, the FDA's Pharmacy Compounding Advisory Committee voted against adding AOD-9604 to the 503A bulks list. That leaves no current lawful compounding pathway in the U.S. It was not among the seven peptides the panel reconsidered this July. Australia's TGA made it a prescription-only substance in 2015. It also sits on WADA's prohibited list, under growth hormone fragments, so any tested athlete should treat it as off-limits.
The honest summary. AOD-9604 is not a mystery compound with untapped potential. It is a compound whose central claim was tested in humans, at a meaningful scale, and did not hold up. The mouse data was real. The small early trial was real. The large trial that should have settled the question did settle it, just not in the direction the marketing needs. Anyone selling it today as a fat-loss tool is selling the 2001 result and leaving out the 2007 reality.
RQS: 33/100 (Weak Evidence)
⚖️Claim vs. Reality: "Sleeping on Your Back Is the Healthiest Position"
The claim: Back sleeping is the gold standard. Your spine stays neutral, your weight is spread evenly, your face isn't pressed into a pillow, and nothing is twisted. Side sleeping supposedly scrunches your shoulders and creases your skin. Stomach sleeping is treated as close to a sin. If you want to sleep "correctly," you train yourself onto your back.
The reality: Back sleeping has one real strength: it is the most neutral position for the spine. Stretch that into "healthiest overall" and it falls apart. The position that is best for your spine can be the worst for your airway, and your airway matters more.
When you lie flat on your back, gravity pulls your tongue and soft palate backward toward your throat. That is why back sleeping is the most common trigger for snoring and the position that worsens obstructive sleep apnea in most people who have it. For many people with sleep apnea, the problem gets substantially worse on their back and improves on their side. Given how many adults have undiagnosed sleep apnea, recommending back sleeping to everyone would be a mistake.
Reflux follows a similar pattern. A 2023 systematic review and meta-analysis found that sleeping on the left side reduced nighttime acid exposure compared with both the right side and the back. The stomach sits on the left side of the upper abdomen, so lying on your left keeps it below the esophagus. Lying flat on your back removes that advantage.
Pregnancy is the case where the stakes are highest. Later in pregnancy, obstetric guidance generally recommends going to sleep on your side rather than your back. Lying flat can compress major blood vessels, and observational studies have linked going to sleep on the back in late pregnancy with a higher risk of stillbirth. That is observational evidence, not a proven mechanism, but it is consistent enough that the guidance leans firmly toward side sleeping.
Stomach sleeping is less of a villain than its reputation suggests. It rotates your neck for hours and flattens the lower back's natural curve, which makes it the most likely position to cause neck discomfort. It also keeps the airway open better than back sleeping. It is a trade-off, not a sin.
There is also a practical problem with the whole debate: most people change position many times a night. The position you fall asleep in is not the one you spend eight hours in. You have less control over this than the "train yourself onto your back" advice assumes.
Where the claim has a real kernel, because it does. If you have no snoring, no sleep apnea, and no reflux, and you wake up without neck or shoulder pain, back sleeping is fine. For some people with certain shoulder or hip pain, it is the most comfortable option. Back sleeping is also the one position that is unambiguously correct for one group: infants. "Back to sleep" is one of the most effective public-health campaigns in history for reducing sudden infant death. That is a finding about babies, though, and it does not transfer to adults.
The honest version of the claim: There is no single best sleeping position. There is a best position for your specific body.
Default for most adults: side sleeping, with a pillow between the knees if your hips or lower back complain.
Snoring or sleep apnea: avoid your back. That is the clearest signal in this research.
Reflux: try your left side.
Late pregnancy: go to sleep on your side.
None of the above: back sleeping is fine.
The mistake is the same one that runs through this whole issue: taking something true for one outcome (spinal alignment) and generalizing it to everything. The same logic sells AOD-9604: a real mouse result became "fat loss without side effects," and the 536-person trial that disagreed got left out.
📊 Research Quality Score: Spotlight
AOD-9604 is a useful test of what the RQS actually measures, because it breaks the usual pattern.
Most weak scores in the catalog are weak because the human research is thin or missing: animal models, case reports, a handful of small open-label studies. AOD-9604's weakness is different. Look at the breakdown and its best dimension is Study Design: it had real randomized, placebo-controlled trials, which many catalog compounds cannot claim. The score is held down by everything around those trials:
Results: the trials did not produce a successful primary outcome.
Replication: essentially one developer ran all the meaningful studies.
Funding: the work was almost entirely developer-funded.
Publication: most results appeared in lower-tier journals or conference abstracts, with none of the primary outcomes in top-tier journals.
This is the distinction worth taking away. The RQS asks how much high-quality research stands behind the claims being made for a compound. For AOD-9604, the best research that exists stands against the main claim. A 33 here does not mean "we don't know yet." It means we do know something, and it isn't what the label says.
As always, the score is not a safety rating, not an efficacy rating, and not a statement about what is in any vendor's vial. It measures the research, and this research was unusually clear.
📡 On My Radar
The response window on last issue's warning letters has closed.
The five online vendors warned in the August 24 sweep had 15 business days to respond, which put the deadline in mid-September. Company responses are not usually made public quickly. What to watch for next is either a close-out letter from the FDA, which signals the issues were fixed, or quiet changes to those storefronts: removed claims, relabeled products, or bacteriostatic water pulled from the same page as the peptides.
GLP-1 compounding is narrowing, which pushes demand toward alternatives.
In April, the FDA proposed excluding semaglutide, tirzepatide, and liraglutide from the 503B bulks list, on the grounds that approved versions are available. Whatever you think of that, the likely downstream effect is predictable. People priced out of approved GLP-1s go looking for something else, and compounds like AOD-9604 are what the gray market will offer them. That is exactly why this issue's lead matters right now.
Hopefully that was helpful/interesting, see you next week.
-Emeka
You're receiving this because you signed up at peptideclear.com. No sponsors. No affiliate links. Just the research, translated.