In July 2026 an FDA advisory committee voted to recommend six peptides for compounding — overruling the agency’s own scientists, who had recommended against all seven. Here is what that actually means, what the evidence shows compound by compound, and the problem with oral peptides that no amount of formulation has yet solved.
Peptides are the fastest-growing category in supplements, and among the most expensive. A month's supply of the popular ones runs to seventy or eighty pounds. Some of the marketing is careful. A good deal of it is not.
I have spent a while reading the underlying papers rather than the summaries. The short version: one of these has real human evidence, one is a licensed medicine that people confuse with the rest, and everything else is animal work. That is not the same as saying they do nothing. It means nobody has checked.
The regulatory story is being told badly, and the detail matters if you are deciding whether to spend money.
September 2023. The FDA moved nineteen widely used peptides into Category 2 of its 503A bulk substances list — substances that "may present significant safety risks." The stated reasons were immunogenicity, manufacturing impurities, and insufficient human clinical data. The practical effect was that compounding pharmacies could no longer supply them.
April 2026. Twelve were removed from Category 2 — BPC-157, TB-500, KPV, MOTS-c, Semax, Epitalon, DSIP, injectable GHK-Cu, LL-37, DiHexa, PEG-MGF and Melanotan II.
April 2026 was not an approval. They were not moved to Category 1 — the list of substances actually permitted in compounding. They were in neither list. And they had never been in Category 1 to begin with, so nothing was restored.
July 2026. The Pharmacy Compounding Advisory Committee met over two days and voted to recommend six of the seven peptides it reviewed for the 503A list: BPC-157, KPV and TB-500 each at 8–6 with one abstention, MOTS-c at 7–5, and Semax and Epitalon on similarly narrow margins. Only emideltide was rejected.
That last vote is the reason this article exists, and almost every summary of it has buried the most important detail.
FDA career staff spent months reviewing these substances and concluded that none of the seven met the agency’s evidentiary standard for inclusion. They recommended against every one.
The committee voted the other way six times out of seven. Observers reported an audible reaction in the room when the first tally was read, because a PCAC panel voting against FDA staff’s written recommendation is close to unprecedented. The committee’s composition had drawn scrutiny before the meeting opened, with reporting noting industry ties among members.
One voting member, explaining a no vote on KPV, put it as plainly as anyone could: “I’m voting on something here, but I don’t know what that something is.” He described it as a black box. That is a committee member, on the record, during the vote.
And nothing is legal today that was not legal yesterday. The votes are advisory. The FDA can accept or reject them, and if it accepts, it must run formal notice-and-comment rulemaking — a proposed rule, a public comment period of sixty to ninety days, then a final rule. Legal analysts put the realistic timeline at eight to twenty-four months, which means 2027 at the earliest. Until then, anything sold as BPC-157, KPV, TB-500, MOTS-c, Semax or Epitalon remains an unapproved substance with no regulatory oversight of purity or dose.
Two further things worth knowing. The April removals followed nominators withdrawing their nominations and sustained political pressure, including from the current HHS Secretary, alongside a lawsuit brought under the Administrative Procedure Act. And across the whole sequence, no new evidence was produced. Nothing about these compounds became better supported between 2023 and now. The regulatory weather changed; the science did not.
There is a reasonable argument on the other side, and I will give it fairly: the 2023 restrictions pushed people who had been getting these under physician supervision toward an unregulated grey market, where purity and dose are unverified. Bringing them into a regulated pathway is a genuine public health argument, and it is why this is not simply a story about a committee ignoring its scientists.
None of it applies in the UK anyway. These are not licensed medicines here and not authorised as food supplements. Products sold online are frequently labelled “research use only” — a legal position, not a quality claim.
The most popular by a distance, sold for gut healing, tendon repair and almost everything else. The healing effects come from rat and mouse studies, a large proportion from a single research group over many years.
There are no published human randomised trials. Not few — none. That is a remarkable fact for a compound this widely used, and it is the first thing anyone selling it should tell you. Prohibited in sport by WADA. Oral forms face an additional obstacle: it is a peptide, and peptides are what digestion is for.
A three-amino-acid fragment of alpha-MSH, with genuine anti-inflammatory activity in mouse colitis models. The mechanism is real and the biology is interesting.
No human trials. And note the gap between what was studied and what is sold: those are disease models — chemically induced colitis — while the products are marketed for daily wellness in people who are well. Different proposition, different evidence required.
TB-500 is related to thymosin beta-4 but not identical, which muddies the literature attributed to it. MOTS-c has genuinely interesting mitochondrial biology in animals and essentially no human data. Semax and Selank are approved in Russia, where nearly all the research was done, with little independent Western replication.
Epitalon is marketed for telomeres and longevity on a small, largely unreplicated Russian literature. Telomere claims in supplements have a poor record — the astragalus case is the cautionary one, where the single human trial found significance at the low dose but not the high dose.
Ipamorelin, CJC-1295 and sermorelin do measurably raise growth hormone. That part is not in doubt. What is far less established is that raising GH in someone whose GH is normal produces the advertised benefits. Prescription territory, prohibited in sport, and not something to self-administer.
Palmitoylethanolamide has real human evidence — multiple randomised trials and meta-analyses, mainly for chronic pain, with a plausible mechanism through PPAR-alpha. It is inexpensive, widely available, and has a good safety record. If the peptide category appeals to you, this is the one that has actually been tested in people, and it is usually the cheapest thing on the shelf.
Semaglutide and tirzepatide are peptides, which is why they get swept into these conversations. They are licensed medicines with very large trial programmes. They work. They also need a prescription, monitoring, and a proper conversation about muscle loss, gastrointestinal effects and what happens when you stop. Anything sold outside a prescribing route is not this, whatever the website implies.
Almost every oral peptide product runs into the same wall, and the industry's answer to it deserves scrutiny.
Peptides are chains of amino acids. Your digestive system exists to break those chains. So oral products increasingly include an absorption enhancer, most often salcaprozate sodium — SNAC — the technology behind oral semaglutide.
It is compound-specific. The study that established it, in Science Translational Medicine, found absorption occurs in the stomach, confined to the area immediately around the dissolving tablet, and states explicitly that the mechanism is compound specific. It was engineered for semaglutide. Assuming it transfers to a different peptide is the extrapolation the original paper warns against.
The ceiling is low. Even with SNAC, and after enormous pharmaceutical engineering, oral semaglutide achieves 0.4–1% bioavailability. A pharmaceutical tablet contains 300 mg of SNAC to get there. Reviews of the field describe these enhancers as producing "single-digit, highly variable" increases.
And there is a signal worth watching. A 2026 study in Journal of Controlled Release gave SNAC alone to rats for three weeks. It depleted two major fermenting bacterial families, cut faecal butyrate by 77%, and raised TNF-alpha by 70%. Rat study, associative, needs replication — but an ingredient added to a product sold for gut comfort showing butyrate depletion is the kind of irony worth noticing.
Here is where this connects to how I actually work, and it is the part that changes the decision.
The usual question is "should I take BPC-157?" It is the wrong question, because it has no reference to you. The better question is what your testing shows about the thing you are trying to fix.
Someone with gut symptoms considering a peptide for "gut healing." A stool test distinguishes between an actual pathogen, an overgrowth, genuine mucosal inflammation, low pancreatic enzyme output, and a microbiome that is simply under-fed. Those have different answers. Only one of them is even theoretically a tissue-repair problem, and it is not the common one.
Someone with a grumbling tendon considering a peptide for repair. Eccentric loading is the best-evidenced conservative treatment there is, it costs nothing, and most people have never done it properly.
Someone exhausted, considering a peptide for energy. Ferritin, thyroid function, B12 and inflammatory markers — most of it free from a GP. Correctable fatigue is missed far more often than it is exotic.
In each case the peptide is a refinement being asked to do the work of a diagnosis. Occasionally the testing comes back and the peptide still looks reasonable. More often it points somewhere cheaper and better supported, and that is a good outcome rather than a disappointing one.
If you are drawn to this category: PEA is the one with human trials, it is cheap, and it is a sensible place to start.
If you are considering one of the others: that is a personal experiment at premium prices. Legitimate, as long as you know that is what it is. Set a time limit and decide in advance what would count as it having worked, because without that every supplement appears to work slightly.
If you are considering injectables: that is a prescribing decision, not a supplement decision. Unregulated injectable peptides carry sterility and purity risks that capsules do not, and "research use only" is not a quality standard.
And if you compete in tested sport: BPC-157, TB-500, MOTS-c and the growth hormone secretagogues are all prohibited. A supplement label is not a defence at a hearing.
Nothing here is a claim that peptides do not work. It is a claim that for most of them, nobody knows — and that you are being asked to pay a premium to find out on your own.
I have no financial interest in any peptide product, which is worth saying in a field where most of the people writing about this do.
Educational content, not medical advice. Regulatory positions described are current as of August 2026 and are actively changing. The PCAC recommendations of 23–24 July 2026 are non-binding; the FDA has not acted on them at the time of writing, and a second committee meeting on further peptides is expected in early 2027. Peptides sold in the UK are not licensed medicines and not authorised food supplements. Always tell your GP what you are taking.
Before spending on a peptide, it is worth knowing what you are actually treating. That is a cheaper question and usually a more useful one.
Take the readiness assessment Book a discovery call