FDA has been sending warning letters to research-peptide sellers, and it just moved past the GLP-1s

All six letters came from the same office — CDER's Office of Unapproved Drugs and Labeling Compliance — and all six carry the same signature, Captain Tina Smith, its director.

The code names are the sellers' own. FDA quotes them back and, in most of the letters, supplies the translation: Prime Sciences' "GLP1-R" is retatrutide, "GLP1-S" is semaglutide, "GLP1-T" is tirzepatide. Gram Peptides sold retatrutide under the name "GLP-1-R peptide" and tirzepatide as "GLP-2 peptide" — numbering that does not correspond to anything pharmacological.

Every one of the six pleads a single count: unapproved new drug, under sections 505(a) and 301(d) of the Food, Drug and Cosmetic Act, reached through the drug definition in 201(g)(1) and the new-drug definition in 201(p). No misbranding charge appears in any of the six letter bodies. That is worth stating precisely, because FDA's own portal labels the Wholesale Peptide letter "Unapproved New Drugs/Misbranded" while the letter itself contains no misbranding allegation at all — no 502, no reference to adequate directions for use. The word appears only in the index.

This is the part worth reading if you buy peptides online, because the agency has now said the same thing six times since March, which makes it a position rather than an opinion.

FDA does not argue with the disclaimer. It does not claim the label is fraudulent or that the seller is lying about their intentions. It makes a narrower and more durable move: the disclaimer loses to the rest of the website. The construction is identical from letter to letter, with only the quoted disclaimer changing. To Pink Pony Peptides: "Despite statements on your product labeling marketing your products for 'laboratory research purposes only' and '[n]ot for human consumption,' evidence obtained from your website establishes that your products are intended to be drugs for human use."

To Gram Peptides, the same sentence with "Research Use Only" and "not intended for human consumption, medical use, or veterinary use" swapped in. To Mile High Compounds, with "in vitro research and educational purposes only." To PekCura Labs, whose disclaimer FDA quotes with four of its own "[sic]" markers, preserving the seller's capitalisation errors in the record.

In the Wholesale Peptide letter the same holding survives but has been demoted to a footnote below the signature block, and slightly rephrased — evidence obtained "from your product labeling, including your website." Same doctrine, less prominence.

The legal machinery underneath is 21 CFR 201.128, the intended-use regulation. Intended use is established by the objective evidence of what a seller means the product for, not by what the label disclaims. A disclaimer is one piece of evidence. Marketing copy, dosing information and the way products are sold together are others, and they can outweigh it.

The most concrete thing in these letters, and the one least likely to be understood by buyers, concerns the water.

Peptides sold as lyophilised powder have to be reconstituted before injection. Sellers routinely offer bacteriostatic water alongside them. FDA's position is that this pairing makes the water itself a drug. From the Mile High Compounds letter: "Your firm offers 'BAC Water' for sale alongside peptide products, which are drugs intended for injection and require reconstitution, including the above-mentioned products. The sale of these products together demonstrates that you intend for your 'BAC Water' to be used in combination for injection. Therefore, your 'BAC Water' is a drug."

The same paragraph, with the product name changed, appears verbatim in the Pink Pony, Gram and PekCura letters. Five of the six carry some version of the holding. It is absent only from the Wholesale Peptide letter, which names no water and no injection equipment.

Prime Sciences got the sharpest version, and it is the fifth. That firm sold a "Reconstitution Kit" containing BAC water and a syringe, and FDA treated the syringe as the tell: "The sale of a kit containing the 'BAC water' and a means of injection (a syringe) with these peptide drugs demonstrates that you intend for the 'BAC water' to be used in combination for injection."

The practical reading for a buyer: the store's own product bundling is evidence about what the store intends. A research reagent does not ship with a means of injecting it into a person.

Two of the letters show the agency reaching past the storefront.

Mile High Compounds linked from its site to a community forum hosted on Skool. FDA reviewed both, from January to March 2026, and drew evidence not only from the shop pages but from documents circulating in the forum — a file called "The Ultimate Mile High Compounds Peptide Cheat Sheet" and a set of per-compound sheets. What those documents contained was dosing guidance written for people. FDA quotes it directly: for "GLP-3 RT (Retatrutide)," uses listed as weight loss and addiction recovery, dosing "Start: 1mg - 2mg weekly. Titrate up to: 12mg weekly." Parallel entries for tirzepatide and semaglutide. None of that is research-reagent documentation, and none of it was on the part of the site a casual visitor would see first.

PekCura Labs went the other way and it also failed. That seller wrote in careful clinical prose and cited the literature — Lincoff 2023, Marso 2016 and others — while almost never naming the molecules, using code names and doses instead. FDA treated the scientific framing itself as evidence of intended use. Writing like a pharmaceutical company about a product you say is not a pharmaceutical is not a defence.

Less than you might expect.

As of FDA's own portal, current to July 23, 2026, no response letter and no closeout letter has been posted for any of the six. That is worth stating carefully: FDA posts responses selectively, and its standard notice warns that matters described in a warning letter may have been the subject of later interaction that changed the regulatory status. Absence of a posted closeout is not proof that nobody replied. What it does mean is that FDA has not publicly recorded any of these six as resolved.

None of the six has been the subject of a publicly announced seizure or injunction, which is the escalation the letters themselves threaten. And no newer peptide-seller letter has appeared: the most recent comparable activity from the same FDA office, posted July 7, was a 17-letter batch of which 14 went to ketamine sellers, not peptides.

We attempted to check what each seller's website looks like today and could not complete it — several are now behind login gates or automated bot verification. We are not going to characterise what has changed on pages we could not properly read. What we can say is that none of the six appears to have exited.

Two FDA tracks pointed in opposite directions this summer, and it is worth being precise about why that is not actually a contradiction.

On June 17 the agency told a seller that offering Prostamax and gonadorelin to the public was distribution of unapproved new drugs. On July 23 and 24 its Pharmacy Compounding Advisory Committee voted to recommend BPC-157, KPV, TB-500, MOTS-c, epitalon and semax for the 503A bulks list, rejecting only emideltide.

These are different questions. The advisory vote is about whether a licensed pharmacy may compound a substance from bulk against a prescription for a specific patient. The warning letters are about websites selling vials to anyone with a card, under a label that says the contents are not for human use. Nothing the committee recommended would make the second thing legal. If anything the two moves fit together: the argument several committee members made for a legal compounding channel was precisely that it would draw demand away from the grey market the letters are aimed at.

What has not changed is the bottom line for a buyer. None of these substances is an approved drug. None is on the 503A positive list today. FDA has issued no final determination. For what the compounding pathways actually are, see our compounded versus branded explainer; and the seller-side signal that matters most is the one FDA acted on in every letter above: a vendor that labels product "research use only" while marketing it for human use is not operating inside any compounding pathway at all.