The FDA Peptide Vote of July 2026: What the Evidence Shows

On July 23 and 24, 2026, an FDA advisory committee meets to weigh seven peptides that compounding pharmacies want cleared to make. The agency’s own reviewers had already recommended no, in writing, on all seven, before the vote.

People are calling this the FDA peptide ban 2026. What’s up for a vote is narrower: whether compounding pharmacies can make these seven from bulk powder. Nobody is outlawing the molecules. A no just takes them off the pharmacy menu.

Semax raises a separate question about what a favorable vote actually permits, which I work through in is semax legal after the FDA panel vote.

A peptide seller won’t grade the human evidence for you. I will. For each of the seven, I read what’s actually been tested in patients and how well it held up: real trials, where people took the peptide and someone measured what happened.

For every peptide under review, the human evidence for its specific use is absent, tiny, decades old, uncontrolled, or from a single country that never got replicated. One of the marquee compounds rides on a nine-patient eye study of a different molecule.

That’s bad news for anyone selling these as clinically proven. It doesn’t make them worthless either. Thin evidence means we don’t know yet, not that they’ve failed.

Update, July 24, 2026: Both days are done. The committee recommended six of the seven peptides for the 503A Bulks List and turned down one. On day one it went against the FDA’s own reviewers and backed BPC-157, KPV, and TB-500 (each 8-6, with one abstention) plus MOTS-c (7-5, two abstentions). On day two it backed semax (8-5, one abstention) and epitalon, reported as 7-4 with one abstention by most outlets covering the meeting and 7-5 by one specialist publication. DSIP, which appears on the agenda under the name emideltide, is the only one the committee declined, 6-7 with one abstention. Every one of these is a recommendation, not an approval. It does not by itself make any peptide available to prescribe or compound. FDA now decides whether to accept the recommendations, and if it does, each change still has to go through a proposed rule, a public comment period, and a final rule, a process that usually runs 12 to 24 months. FDA hasn’t posted the official meeting record yet, so these tallies come from reporters who were in the room. Nothing about the votes changes the safety evidence below.

What the Committee Is Actually Voting On

Section 503A of the federal drug law lets a licensed pharmacist or physician compound a medication from a raw bulk substance only under set conditions. The substance needs a USP monograph, or it has to be a component of an FDA-approved drug, or it has to sit on a special list called the 503A Bulks List. None of these seven peptides meets the first two conditions. The Bulks List is the only remaining door, and that door is what the July vote is about.

FDA weighs each nominated substance against four things, set by a 2019 final rule: how well the substance is physically and chemically characterized, its safety, the evidence that it works or the lack of it, and how long it has been used in compounding. The agency balances those one substance at a time. If you want the mechanics of how compounding rules treat these peptides, I laid them out in the companion explainer on FDA peptide compounding.

The original nominations, filed in 2015, came from Wells Pharmacy Network and from LDT Health Solutions on behalf of the International Peptide Society. Both later withdrew them. FDA reviewed the peptides anyway, and its briefing documents say the agency is electing to proceed with the presentations. And the committee’s vote is advisory. The agency usually follows it, but it makes the final call itself.

If FDA keeps them off the list, a compounding pharmacy can’t legally make them for you. That closes one of the few above-board routes people have had. The alternative, which I covered in the earlier peptide reclassification piece, is a gray market with far less oversight.

How I’m Grading These Seven

I’m using a plain scale. Strong means several good human trials point the same way. Moderate means at least one solid controlled trial. Weak means the human data exist but they’re small, old, uncontrolled, or come from a single country with no outside replication. Preliminary means the evidence is animal, cell-culture, or observational, with no adequate human trial for the advertised use.

Animal data and a plausible mechanism are reasons to study a molecule. They aren’t reasons to inject it. The whole question is what happened when these went into actual patients.

Peptide	Marketed for	Best human evidence	Grade	FDA staff position
BPC-157	tissue and gut healing	one 2005 conference abstract (ulcerative colitis), not significant	preliminary to weak	recommend not adding
KPV	wounds, inflammation	none in patients; human data are cell-culture only	preliminary	recommend not adding
TB-500	wound healing, recovery	none; the cited trial is a different molecule	no human evidence	recommend not adding
MOTS-c	obesity, osteoporosis	none given as a drug; human data are exercise observations	preliminary	recommend not adding
DSIP (Emideltide)	opioid withdrawal, insomnia, narcolepsy	small, old, mostly uncontrolled; best controlled trial near-null	weak	recommend not adding
Semax	stroke recovery, migraine	Russian, non-randomized, 110 patients	weak	recommend not adding
Epitalon	insomnia, anti-aging	none for sleep; a cell-culture telomerase result	preliminary	recommend not adding

Four of these have no dedicated write-up on this site yet, so I’ll take them one at a time. The last three already have their own deep dives, and I’ll keep those short and point you to them.

What TB-500 Actually Is

Start with TB-500.

Sellers describe TB-500 as thymosin beta-4, a natural healing protein. The FDA’s briefing document calls it a synthetic 7-amino-acid fragment, residues 17 through 23 of thymosin beta-4, capped with an acetyl group. Authentic thymosin beta-4 is 43 amino acids. TB-500 is a small piece of the protein, sold under the whole protein’s reputation.

The problem shows up in what gets cited to sell it. The human trial they point to used authentic thymosin beta-4, the full-length version, as an eye drop for severe dry eye. It was a nine-patient trial, and the drug beat placebo (PMID 25826322). The study is real. It also has nothing to do with TB-500, or with healing an injury by injection. It’s a different molecule tested for a different problem.

There’s no human trial of TB-500 itself for wound healing. The FDA also flagged that TB-500 is “not physically and chemically well characterized,” which means nobody supplied the data on purity, aggregation, or contamination you’d want before injecting it.

When the committee reviews TB-500, it’s weighing a poorly defined fragment, with no human efficacy data for its claimed use, and a nine-person eye study of a different molecule standing in for proof. The reviewers recommended against adding it. On that evidence, it’s the correct read.

KPV: Mechanism Without Patients

KPV is the tail end of a hormone called alpha-MSH, and it carries a believable anti-inflammatory mechanism. Added to human airway cells in the lab, it quiets several inflammation signals (PMID 22837805). That single lab result is where the human evidence stops. No patients have been treated with KPV in a proper trial, for wound healing or anything else. The rest of the data are in mice.

KPV is the cleanest example on this list of a molecule that is biologically interesting and clinically untested. FDA’s reviewers reached the same verdict.

MOTS-c: Exercise Biology, Mostly Mouse Data

MOTS-c is a peptide your own mitochondria make, and exercise raises it. That’s the pitch, a molecule that looks like a workout in chemical form. In mice, giving MOTS-c improved physical performance, even late in life (PMID 33473109). In humans, the only finding is that exercise raises the MOTS-c your body already makes.

Nobody has finished a trial showing that injecting MOTS-c treats obesity, and there’s nothing at all for osteoporosis, the other use under review. Two authors of the main paper hold shares in a company developing these peptides. Weigh that against the enthusiasm. A metabolic trial is underway. Until it reports, MOTS-c is a mouse result with a good story.

DSIP (Emideltide): A Sleep-and-Withdrawal Peptide With Almost No Controlled Data

This is the one closest to my own work. DSIP, delta-sleep-inducing peptide, which FDA lists as Emideltide, is pitched for opioid withdrawal, insomnia, and narcolepsy. It appears to work through the body’s own enkephalin system rather than binding opioid receptors directly.

The human evidence is thin and old. The withdrawal data come from a 1983 open-label study with no comparison group, in which most of the patients were reported to improve (PMID 6328354). Open-label means everyone knew they were getting the drug, which is the setup where hope does half the work. For insomnia, the best-controlled study, 16 patients and double-blind, found a slightly shorter time to fall asleep and concluded the peptide probably wasn’t of much therapeutic use (PMID 1299794). FDA’s own review calls the controlled data “at best some weak effects”.

For opioid withdrawal, we already have proven, standard treatments: buprenorphine, lofexidine, clonidine, methadone. Reaching for a gray-market vial of a 1980s research peptide instead means trading something that works for something barely tested. If the real problem is sleep, the treatment with the best long-term evidence is cognitive behavioral therapy for insomnia, which I put ahead of any injectable in my guide to sleep medications.

If a peptide caught your attention because something specific feels off, the better first move is an evaluation that finds the cause. You can book one.

BPC-157, Semax, and Epitalon: The Three With a Longer Paper Trail

These three already have detailed write-ups on the site, so here are the short grades with links to the full versions.

BPC-157 carries the most preclinical hype and the most striking animal data, mostly around tissue and gut repair. In humans it thins out fast. A 2025 peer-reviewed review found only three small pilot studies in people, none large or conclusive (PMID 40789979). For ulcerative colitis, the use FDA examined, the only study was a 2005 conference abstract whose result wasn’t statistically significant. The animal data look strong. In people, it has barely been tested. The full picture is in the BPC-157 deep dive.

Semax is a Russian medication with genuine use there, studied mostly for stroke recovery. The catch is that almost all the evidence is Russian, and the representative stroke study of 110 patients wasn’t randomized or placebo-controlled (PMID 29798983). It’s registered as a drug there, but no independent group outside Russia has replicated the results. I covered its cognitive and mood claims in the Semax and Selank article.

Epitalon is sold on anti-aging and sleep claims. Its best-known data come from a cell-culture experiment showing it can switch on telomerase in human fibroblasts (PMID 12937682). That’s a laboratory finding. It’s a long way from a better night’s sleep, and there’s no adequate human insomnia trial behind the sleep claim. The longevity numbers come from the developer’s own research group. More on it in the Epitalon write-up.

What the People Using Them Say

The anecdotal reports on X and Reddit are hard to ignore. Thousands of people describe less pain, better sleep, more energy. Anecdote isn’t proof, and a comment feed can’t separate a drug effect from placebo. But a signal this loud strongly suggests something is working here.

That’s why these deserve proper trials. A controlled study is what would show whether the peptide is doing the work, and how.

Where I Land on These

I’m more optimistic about these than the current research is. I think there’s a signal in all the noise.

Several have genuine potential. BPC-157’s animal healing data is strong, DSIP has a plausible mechanism, and MOTS-c’s tie to metabolism is worth a hard look. A few may well pan out once someone runs the trials.

Still, a promising mechanism is not a reason to put a patient on an unproven, gray-market injectable. The ruling makes that harder, not easier: keeping these off the compounding list closes the clean, regulated route a careful clinician would use. Even for the ones I believe in, I can’t recommend them cleanly right now. They’re promising but unproven, and off the table for now. I’d rather say that than oversell it.

What This Means If You’re Weighing Peptides

For all seven of these peptides, the human evidence for the advertised use runs from thin to absent. The peptide world reads that as FDA suppressing something that works. I read it as data that was never collected.

If a product’s best evidence is a 1983 case series or a fibroblast assay, that’s a reason for caution. And there’s what you’re actually buying from a gray market: an injectable whose purity and contamination nobody independent has checked. That’s a risk on its own, separate from whether the molecule works.

None of this means the biology is worthless or that these peptides will never earn a real evidence base. Some of them might. But “might, once it’s studied” is a different claim from “proven, buy it now,” and the gap between those two is where people get hurt.

When patients come to me interested in one of these, the useful conversation is usually about what sits underneath the search: fatigue, poor sleep, slow recovery, low drive. Those have real workups and treatments. That’s the part I can actually help with.

If any of these do become compoundable, here’s what I’d want in place before anyone injects one:

• A physician licensed in your state who has your diagnoses, medications and recent labs in front of them before prescribing.
• A licensed compounding pharmacy that follows USP 797 sterile compounding standards and can show a certificate of analysis for the batch.
• A written reason for trying it, and an agreed point at which you stop.
• Scheduled follow-up with the same physician, with monitoring that fits the peptide.

Until then, a vial labeled “research use only” or “not for human consumption” hasn’t been through any of those checks.

Frequently Asked Questions

Is the FDA banning peptides?

It’s not a ban in the usual sense. The July 2026 committee voted on whether these seven peptides can be added to the 503A Bulks List, which is what lets a compounding pharmacy legally make them. Keeping them off the list restricts that route. It doesn’t criminalize the molecules or ongoing research.

Does a “no” mean these peptides don’t work?

No. FDA’s position rests on a mix of poor product characterization, safety uncertainty, and thin effectiveness data for the uses nominated. Absence of good evidence isn’t the same as evidence of absence.

Which peptides are under review?

Seven across two days: BPC-157, KPV, TB-500, and MOTS-c on July 23; Emideltide (DSIP), Semax, and Epitalon on July 24. A second group of peptides is reportedly slated for a later meeting.

Are these safe to buy from an online seller instead?

The sourcing is its own problem. These are injectable products whose purity and contamination often haven’t been independently verified, which FDA flagged directly for several of them. An unregulated online product isn’t safer just because FDA hasn’t listed it.

I take a peptide now. What should I do?

Talk with a physician who will look at why you started it and what you’re actually trying to fix. Most of the goals behind peptide use, whether better sleep, energy, recovery, or mood, have evaluations and treatments with real evidence behind them.

This article is for education and isn’t medical advice or a substitute for care from your own clinician. It does not recommend or offer any peptide. If you’re in crisis or thinking about harming yourself, call or text 988 for the Suicide and Crisis Lifeline. For help with substance use, the SAMHSA National Helpline is 1-800-662-4357.

References

1. Sosne G, Dunn SP, Kim C. Thymosin beta-4 significantly improves signs and symptoms of severe dry eye in a phase 2 randomized trial. Cornea. 2015;34(5):491-6. PMID 25826322.
2. Land SC. Inhibition of cellular and systemic inflammation cues in human bronchial epithelial cells by melanocortin-related peptides: mechanism of KPV action. Int J Physiol Pathophysiol Pharmacol. 2012;4(2):59-73. PMID 22837805.
3. Reynolds JC, Lai RW, Lee C, et al. MOTS-c is an exercise-induced mitochondrial-encoded regulator of age-dependent physical decline and muscle homeostasis. Nat Commun. 2021;12(1):470. PMID 33473109.
4. Dick P, Grandjean ME, Tissot R. Successful treatment of withdrawal symptoms with delta sleep-inducing peptide. Neuropsychobiology. 1983;10(4):205-8. PMID 6328354.
5. Bes F, Hofman W, Schuur J, Van Boxtel C. Effects of delta sleep-inducing peptide on sleep of chronic insomniac patients: a double-blind study. Neuropsychobiology. 1992;26(4):193-7. PMID 1299794.
6. McGuire FP, Martinez R, Lenz A, Skinner L, Cushman DM. Regeneration or risk? A narrative review of BPC-157 for musculoskeletal healing. Curr Rev Musculoskelet Med. 2025;18(12):611-619. PMID 40789979.
7. Gusev EI, Martynov MY, Kostenko EV, et al. The efficacy of semax in the treatment of patients at different stages of ischemic stroke. Zh Nevrol Psikhiatr Im S S Korsakova. 2018;118(3 Vyp 2):61-68. PMID 29798983.
8. Khavinson VKh, Bondarev IE, Butyugov AA. Epithalon peptide induces telomerase activity and telomere elongation in human somatic cells. Bull Exp Biol Med. 2003;135(6):590-2. PMID 12937682.
9. US Food and Drug Administration. July 23-24, 2026: Meeting of the Pharmacy Compounding Advisory Committee. Docket FDA-2025-N-6895.
10. US Food and Drug Administration. Pharmacy Compounding Advisory Committee briefing materials: introduction and evaluation criteria; BPC-157; TB-500; Emideltide (DSIP). July 2026.