Research Use Only Is Not a Regulatory Shield: What the FDA’s 2026 Peptide Warning Letters Actually Say
The FDA published five warning letters to research-peptide sellers on 1 September 2026. All five rest on website review alone, and the agency states directly that research-use-only labelling did not displace the evidence of intended human use.
On 1 September 2026 the FDA published five warning letters dated 24 August 2026 to Royal Peptides, NuScience Peptides, Peak Performance Peptides, Peptide Partners and TXP Innovations trading as Tex Peptides, classifying named products including tirzepatide, semaglutide, retatrutide, PT-141 and tesamorelin as unapproved new drugs. Each letter is based solely on a review of the seller's own website. The agency addresses the research-use-only disclaimer explicitly and concludes that the website evidence establishes the products are intended to be drugs for human use.
On 1 September 2026 the FDA published five warning letters, all dated 24 August, to United States sellers of research peptides: Royal Peptides, NuScience Peptides, Peak Performance Peptides, Peptide Partners and TXP Innovations trading as Tex Peptides.12345 Each letter reaches the same conclusion by the same route: the agency reviewed the seller’s own website, found statements establishing that the products were intended for human use, and classified them as unapproved new drugs. The disclaimer on the label did not change that conclusion, and the agency says so explicitly. This article is regulatory analysis of public enforcement documents. It is not legal advice, and Condor Research supplies reference materials for laboratory research use only.
What exactly did the FDA publish?
Five letters issued from the Center for Drug Evaluation and Research, each dated 24 August 2026 and posted on 1 September, each based on a review of the recipient’s website conducted in July 2026.1 None of them describes an inspection, a product test, a seizure or an adverse event. The entire evidentiary basis is what the company published about itself. Recipients were given fifteen business days to respond in writing.
The products named across the five letters overlap heavily: tirzepatide, semaglutide, retatrutide, survodutide, mazdutide, SS-31 (elamipretide), PT-141, tesamorelin, and in several cases bacteriostatic water. The Royal Peptides letter also names a tesamorelin and ipamorelin blend sold as BIMORELIN.1
What legal theory is the agency actually using?
One theory, applied consistently. A product is a drug under section 201(g)(1) of the Federal Food, Drug, and Cosmetic Act if it is intended for use in the diagnosis, cure, mitigation, treatment or prevention of disease, or intended to affect the structure or any function of the body. If it is a drug and it has not been approved, it is an unapproved new drug under section 505(a), and shipping it in interstate commerce violates section 301(d).1
The second prong of that definition does most of the work and is the one sellers routinely miss. A product does not have to be presented as treating a disease to be a drug. Claiming that it affects the structure or function of the body is sufficient. “Promotes sustainable weight loss” is a structure-function claim. So is most of the language that appears on peptide product pages.
Worth noting for accuracy: the agency’s own database categorises these letters as “Unapproved New Drugs/Misbranded”, but in the letters we read in full the misbranding theory is not developed and no charge under section 502(f)(1) appears. The substance of these five letters is the unapproved new drug argument.
How does the FDA decide what a product is “intended” for?
Intended use is inferred from objective evidence, not from what the seller declares it to be. The letters cite 21 CFR 201.128, the regulation that defines the term, and then simply quote the seller’s website back at them.6 In the Royal Peptides letter the cited material includes product pages describing tirzepatide as acting at the GLP-1 and GIP receptors, and a page stating that tirzepatide “has led to up to 22.5% body weight reduction in participants over 72 weeks”, compared favourably with semaglutide by its brand names.1
The evidence was not a customer testimonial or a private message. It was the seller’s own published copy, read in July and quoted back in August.
What kinds of pages did the agency quote?
This is the detail that deserves attention from anyone who publishes alongside a catalogue. In the Royal Peptides letter the citations are not confined to product pages. The agency also quotes the company’s editorial content, including URLs of the form tirzepatide-vs-retatrutide-weight-loss-peptides and retatrutide-vs-tirzepatide, using benefit lists and weight-loss percentages from those comparison articles as evidence of intended use.1
A blog post is labeling when it functions as promotion. The distinction that matters is not the URL pattern or the word “research” in the site architecture. It is whether the text tells a reader what the product will do for them. A sentence reporting what a published trial measured in its participants, attributed and dated, is a description of evidence. A bulleted list headed “Benefits” is a promise. The first is journalism about a drug. The second is an intended-use statement, and the FDA reads it that way.
What did the “research use only” disclaimer accomplish?
Nothing, and the agency addressed it directly rather than ignoring it. In a footnote to the Royal Peptides letter the FDA writes that despite statements marketing the products “for research use only” and “not for human or animal consumption”, evidence obtained from the website establishes that the products are intended to be drugs for human use.1 The NuScience letter carries the same reasoning against a differently worded disclaimer covering laboratory, research and analytical use.2
This is not a new doctrine and it is not specific to peptides. A disclaimer is one piece of evidence about intent, weighed against all the others. When everything else on the site points at human injection, a line of small print pointing the other way is treated as what it is. We set out the underlying legal architecture in why a research use only label is not a legal status, and the practical meaning of the category in what research use only actually means.
Why did bacteriostatic water matter?
Because of what it was sold next to. The same footnote records that the company marketed bacteriostatic water alongside a “peptide guide” and a “peptide calculator”, and concludes that those resources collectively provide the means to prepare an injectable drug for human administration.1 No sterility failure is alleged. No contamination is described. The finding concerns what the combination of products and tools tells you about who is expected to use them and how.
That reasoning generalises. A reconstitution calculator that outputs units on an insulin syringe is not an analytical instrument. A storage guide written for a person injecting at home is not a laboratory protocol. Each of these is defensible in isolation and indicative in combination, and the agency assembles the combination.
7 warning letters to research-peptide sellers are identifiable in 2026: Gram Peptides in March, Wholesale Peptide in June, and these five in August. The count is a floor, not a total.
Is this a new enforcement posture?
It is an escalation in pace rather than a change in theory. The same argument appears in the letter to Gram Peptides in March 2026 and to Wholesale Peptide in June.78 What changed in August is that five letters issued on a single day, which suggests a co-ordinated review of the sector rather than a response to individual complaints.
One comparison worth avoiding: the large batch of letters the FDA sent in September 2025 concerned compounding pharmacies advertising copies of approved GLP-1 medicines. Different recipients, different legal problem, different market. Adding the two counts together produces a number that describes nothing.
What follows practically for anyone selling research materials?
Five things the letters support directly. Assume every public page is labeling, including comparison articles and guides. Treat structure-function language as equivalent to disease claims, since section 201(g)(1) does. Recognise that accessories and calculators contribute to the picture of intended use and are read together with the catalogue. Understand that a disclaimer is evidence, not a shield, and that its weight depends entirely on whether the rest of the site agrees with it. And note what is absent from these letters: no product was tested, no certificate of analysis was questioned, and no quality finding appears anywhere. Analytical quality is a different axis, and it offers no protection on this one.
We hold ourselves to the same reading. Condor publishes research content alongside a catalogue, which puts us in the same structural position as the recipients of these letters, and the distinction we rely on is the one described above: reporting what studies measured, in whom, with what limitations, rather than telling a reader what a compound will do for them. Our writing on the gray market and on how this class is regulated in 2026 sets out that position at greater length.
What is established, and what is not
Established: five letters exist, dated 24 August 2026 and published on 1 September; they rest on website review alone; they invoke sections 201(g)(1), 201(p), 505(a) and 301(d); and the agency states expressly that research-use-only labelling did not displace the evidence of intended human use.
Not established: any finding about the identity, purity or safety of the products themselves, since none was tested. Not established: what happens next, since a warning letter is not an adjudication and the recipients had fifteen business days to respond. Unresolved: whether the agency develops the misbranding theory its own categorisation implies. What would change the assessment: litigation testing the intended-use inference, or a formal policy statement on research-use-only marketing, neither of which exists today.
How this was checked. The Royal Peptides letter was read in full from fda.gov on 13 September 2026, and every quotation and the footnote reasoning attributed to it were taken from that text rather than from coverage. The other four letters of 24 August 2026 and the earlier 2026 letters to Gram Peptides and Wholesale Peptide were confirmed live at their fda.gov URLs on the same date. Where a characterisation applies only to the letters read in full rather than to all five, the article says so. Version 1.0, first published 13 September 2026.
Condor Research supplies characterised reference materials for laboratory research use only: not for human or veterinary use, not for diagnostic or therapeutic application, and not for any food or cosmetic purpose. This article is analysis of public regulatory documents and is not legal advice. Anyone with questions about their own compliance position should take advice from a qualified regulatory lawyer in the relevant jurisdiction.
Condor Research · Scientific desk
Atrio Sciences s.r.o., IČO 57 669 651, Nitra (SK) · info@condorresearch.com
- Five warning letters dated 24 August 2026 were published by the FDA on 1 September 2026, all issued from the Center for Drug Evaluation and Research.
- Each letter rests entirely on a review of the recipient's website conducted in July 2026; no inspection, product testing, contamination finding or adverse event appears in any of them.
- The legal theory is a single one: section 201(g)(1) defines the products as drugs, section 201(p) as new drugs, section 505(a) requires approval, and section 301(d) is violated by interstate shipment.
- The structure-or-function prong of the drug definition does most of the work: a product need not be presented as treating disease to be a drug.
- FDA's own database labels the letters Unapproved New Drugs/Misbranded, but in the letters read in full no misbranding charge under 502(f)(1) is developed.
- The cited evidence includes not only product pages but the sellers' comparison articles, with benefit lists and weight-loss percentages quoted as evidence of intended use.
- In a footnote, the agency states that despite research-use-only and not-for-human-consumption statements, website evidence establishes the products are intended to be drugs for human use.
- Selling bacteriostatic water alongside a peptide guide and a peptide calculator was cited as collectively providing the means to prepare an injectable drug for human administration.
- At least seven letters of this kind are identifiable in 2026: Gram Peptides in March, Wholesale Peptide in June and these five in August.
- These letters should not be conflated with the September 2025 batch sent to compounding pharmacies advertising copies of approved GLP-1 medicines, which involved different recipients and a different legal problem.
Does a research use only label protect a seller from FDA action?
No. In these letters the FDA addresses the disclaimer directly, stating that despite statements marketing the products for research use only and not for human or animal consumption, evidence obtained from the website establishes that the products are intended to be drugs for human use. A disclaimer is one piece of evidence about intent, weighed against everything else the seller publishes. When the rest of the site points at human injection, the disclaimer carries little weight.
What evidence did the FDA use to establish intended use?
The sellers' own published content. In the letter read in full for this article, the agency quotes product pages describing receptor activity and comparison articles listing benefits and weight-loss percentages, citing 21 CFR 201.128 as the governing definition of intended use. No testimonials, private messages, inspections or laboratory results appear as evidence in that letter.
Can a blog post count as drug labeling?
Yes, when it functions as promotion. The Royal Peptides letter quotes the company's own comparison articles as evidence of intended use. The operative distinction is not the section of the site but the nature of the statement: reporting what a published study measured, with attribution and limitations, describes evidence, while a list headed benefits tells a reader what the product will do for them, which is an intended-use statement.
Why was bacteriostatic water mentioned?
Not because of any sterility or contamination finding. The agency noted that bacteriostatic water was marketed alongside a peptide guide and a peptide calculator, and concluded that those resources together provide the means to prepare an injectable drug for human administration. The reasoning concerns what the combination reveals about intended users, not the quality of the water.
Is this a new FDA position on research peptides?
The legal theory is long-standing and appears in earlier 2026 letters to Gram Peptides in March and Wholesale Peptide in June. What changed in August 2026 is the pace: five letters issued on the same day, which points to a co-ordinated review of the sector rather than case-by-case response.