Can the name of a page be evidence? FDA has quoted a shop category’s name, an explainer page’s title, and the headings inside a document that was never on the website

Reference page · published 2026-10-15

Disclosure: this index is operated by Artemis Labs, which sells research compounds to laboratories. Artemis Labs has no connection to any company named on this page and sells none of the products those companies sold. Nothing here is legal advice. Every quotation is verbatim from an FDA warning letter, an FDA record export, the Code of Federal Regulations or the United States Code, each of which was fetched and read for this page on October 15, 2026, and each of which is listed under Sources. Three limits belong at the top. The companies described here received warning letters, which are agency allegations opening a correspondence, not findings of a court. The central reading on this page — that FDA read the name of a page as evidence — is our reading of what the letters say, and the letters do not use that word for it; the section below titled “What does the evidence not show?” sets out exactly where our reading stops. And the quoted sales copy on this page is reproduced because it is the matter FDA charged, not because any of it is true.

Short answer: yes, in two documented letters — and it is rare. The normal pattern is that FDA prints a product page’s web address and then quotes the sales claims on that page. The address tells you where; the claim is the violation. But in a 2022 letter, FDA quoted nothing at all from one of the pages it cited. It wrote one sentence saying the firm marketed the products as Muscle Builders, and printed that page’s address. And in a 2026 letter, FDA named an explainer page by its title, quoted a heading inside it, and then quoted two headings out of a white paper the firm had handed to inspectors — a document that was never on the website at all. Across the 70 warning letters we hold in full, one does the first thing and one does the second. So this is not a trend. It is a documented possibility, and it is enough to make the names on your own site worth reading the way a stranger would read them.

If you sell anything online, you have written page names you never thought of as writing. The label on a shop category. The title of a “what is this” explainer. The heading over a paragraph. The text in a breadcrumb trail. Nobody drafts those the way they draft a product description, and no compliance review we have ever seen reads them.

This page asks a narrow question about them: has FDA ever pointed at one? We did not reason about it. We went looking in the letters themselves, and the answer is yes, twice, in two different parts of the agency, four years apart.

First, how rare this is — because that is the honest headline

We hold 70 FDA warning letters in full, de-duplicated so the same letter stored under two filenames counts once. We searched all 70 for the phrasings a letter would have to use when it names a surface and quotes nothing from it. The results:

The two controls matter. If our search had been broken, the controls would read zero too. They do not, so the zeros above are readings and not a probe failing quietly.

One more honest note about the method before the letters themselves. A search for webpage at across those 70 returns two letters, and one of them is a false positive: a food-import letter that points the recipient to FDA’s FSVP web page at a fda.gov address. That is the agency citing its own page to be helpful, which is the opposite of what we are looking for. We read every hit, one at a time, which is how that one got caught.

The 2022 letter: one sentence, no quotation, and a web address

On July 6, 2022, FDA’s Center for Drug Evaluation and Research sent warning letter 627498 to Elite Supplement Center LLC and Elite Training Facility LLC. It was posted to fda.gov six days later. It was delivered by email and signed by Carolyn E. Becker, Director of the Office of Unapproved Drugs and Labeling Compliance. In FDA’s own record export the row’s Subject value reads Finished Pharmaceuticals/Unapproved New Drug, which is a filing label and not a summary — we have written about what that column actually is and it is not load-bearing here.

The letter opens by saying what FDA did: it reviewed your website at the Internet address https://www.elitesupplementcenter.com/ in June 2022 and observed that the firm took orders there for five named SARMs. The charge is that those products are unapproved new drugs, under sections 505(a) and 301(d) of the Federal Food, Drug, and Cosmetic Act, 21 U.S.C. 355(a) and 331(d). The word misbranded appears zero times in the letter; the word peptide appears zero times; SARM appears eleven. These products have nothing to do with ours, and the reason this letter is on this page has nothing to do with what was in the bottles.

Then comes the override, which is the sentence this corpus has recorded in many later letters and which turns out to start here:

“Despite statements on your product labels marketing your SARMs products for ‘RESEARCH ONLY’ and ‘Not for Human Consumption,’ evidence obtained from your website establishes that your products are intended to be drugs for human use.”

Read where the two halves of that sentence live. The disclaimer is on the labels. The evidence that beats it is on the website. They are different surfaces, and the letter beats one with the other in a single breath. (We have a separate page on what a research-use-only notice does and does not do.)

FDA then lists five block quotations of the firm’s own product copy, one per product, running to several sentences each. We reproduce exactly one short span, because the weight of these quotations is the honest counterpoint to this whole page and you cannot weigh them without seeing the register. Of one listing, FDA quotes the firm as having written that the compound has become known as incredible fat loss and endurance agent over the past couple of years — the missing article is in the original. That is promotional copy about what a product does to a person, and it is the kind of sentence FDA has charged in letter after letter. The other four read the same way, and we are not reproducing them.

And then, after the five quotations and immediately before the legal section, the letter says this and nothing more about it:

“You market these products as ‘Muscle Builders’ on your webpage at https://www.elitesupplementcenter.com/backpacks?page=2.”

That sentence is structurally different from everything above it. There is no quotation from the page. There is no claim about what the products do. There is a name — Muscle Builders — a verb, and an address. The thing FDA counts as marketing is what the page is called.

The address in that sentence does not say “muscle”

Look at the web address FDA printed: https://www.elitesupplementcenter.com/backpacks?page=2. The word in the path is backpacks. The word muscle appears zero times in that address.

We did not visit that site and we are not asserting anything about what was on it; the address is quoted from the letter and nothing else. But the gap between the two is worth sitting with, because of what it implies about where the exposure sits. The name FDA quoted is not in the address. It is the name a human being would have read on the screen. If that is right, then renaming the slug in a web address protects nothing, and the words a visitor sees are the words that count. That cuts against the instinct most people have, which is to think of a URL as the identity of a page.

FDA never calls it a category

Here is where we have to be careful with our own argument. We have been describing Muscle Builders as a shop category, and the letter does not say that. Measured on the letter body — the salutation through the signature block, not the surrounding fda.gov page:

So “category page” is our word for it, inferred from the fact that one address holds several products and carries a plural name. FDA described a function — you market these products as — and declined to name a structure. That is a real limit on this page’s reading and we would rather print it than have you find it.

It also cuts the other way, and this is the part we think matters more. Because the agency named no structure, there is nothing in this letter that confines the point to shop categories. A function, not a page type, is what the sentence describes.

The 2026 letter: a page named by its title, and a document that was never on the site

Four years later, a different part of FDA did something adjacent. On September 11, 2026, the Center for Biologics Evaluation and Research sent warning letter 730612 to NexCell Scientific Inc, reference CBER 26-730612, delivered by UPS and email and signed by Vincent Amatrudo, JD, Acting Director of the Office of Compliance and Biologics Quality. It was posted eleven days later. The product is an umbilical cord blood-derived total nucleated cell product for allogeneic use. The word peptide appears zero times. Again: a different industry, and that is the point — the method travels even when the product does not.

This letter follows an inspection, and when it gets to intended use it names its surface like this:

“Your website www.nexcelllabs.com includes the following statements in its ‘What is Regenerative Medicine?’ page, along with assurances that your company ‘uses exclusively umbilical cord derived cells’…”

The page FDA identifies is not a product page. It is an explainer — the page a company writes to educate, the one that usually sells nothing and is usually assumed to be the safe part of a site. FDA identified it by its title, in quotation marks, and then quoted the statements in it.

One of those quotations is located not by page but by heading:

“Under the heading ‘How Stem Cells Work’, ‘Once in the affected area, the cells … begin to produce signaling proteins that work to regulate inflammation, aid in angiogenesis, and promote tissue repair.’”

The heading is not itself charged. But it is part of the citation, the way a page number is part of a citation — which means somebody at the agency read the structure of the page closely enough to use it as an address.

Then the letter leaves the website entirely

The next thing FDA cites was never published anywhere. It is a document the firm handed over during the inspection:

“The ‘Characterization of BioGenix® Allograft Products Independent Viability Study,’ which you referenced during the FDA inspection as the ‘White Paper study’ used to characterize your product, states…”

And FDA then quotes it twice, each time by page number and heading: On page 4, under the heading “Applications and Advantages” and On page 10, under the heading “Discussion and Summation”. The phrase under the heading appears three times in this letter — once for the website, twice for the white paper — against a live control of product at 58 and FD&C Act at 12.

Think about what the firm did here. It was asked during an inspection how it characterized its product, and it pointed to a study. The study’s own internal headings then became part of the evidence of what the product was intended for. A document produced to answer a technical question was read as marketing.

The rule that could let a name count, and the fact that neither letter cites it

There are two pieces of federal text in play. We fetched both tonight; neither is quoted from memory.

The first is 21 CFR 201.128, the regulation that defines what “intended use” means. The sentence most people quote from it is the one about claims, and this index has already covered that sentence and the 2021 rewrite that produced it. The clause that matters here is a different one, in the sentence before:

“The intent may be shown by such persons’ expressions, the design or composition of the article, or by the circumstances surrounding the distribution of the article.”

Expressions. Not claims, not advertising — expressions. A category label is an expression. A page title is an expression. On the face of the rule there is room for them.

The second is 21 U.S.C. § 321(m), which defines labeling, and which is the only one of the two that could reach a document that was never on a website:

“The term ‘labeling’ means all labels and other written, printed, or graphic matter (1) upon any article or any of its containers or wrappers, or (2) accompanying such article.”

Limb (2) — accompanying such article — is doing the work in a case like the white paper. That is our reading of how such a document could be labeling at all, and we want to be exact about its status, because here is the thing:

Neither of the two letters cites either provision. Measured on all three letter bodies: 321(m) appears 0 times, the word accompanying appears 0 times, and 201.128 appears 0 times in the Elite letter and 0 times in the NexCell letter. The two letters that did the unusual thing did not explain their authority for doing it. They simply did it.

And the letter that follows the ordinary pattern is the one that cites the rule. More on that next.

What the ordinary pattern looks like, for contrast

On September 8, 2025, CDER sent warning letter 715218 to PureRawz, signed by Tina Smith, M.S., Captain, U.S. Public Health Service, Director. We have a separate record of that letter. Here it is only a control, and its evidence section reads like almost every website-based letter in this corpus:

“Examples from your product labeling, including on your website and social media accounts that provide evidence of the intended use (as defined in 21 CFR 201.128) of your products as drugs for human use include, but may not be limited to, the following:”

What follows that colon is the phrase On your webpages, two full product-page addresses, and then a quotation from them. Then another address, then another quotation. The address locates; the claim convicts. The phrase On your webpages appears twice, under the heading appears zero times, and the number 201.128 appears once, against a live control of the compound name at 48.

So three letters, three postures:

What does the evidence not show?

This is the section that keeps a page like this honest, so it is long.

It does not show that a page name has ever been the basis of a charge. In the Elite letter the evidence section runs 1,917 characters. The Muscle Builders sentence is 121 of them — about 6 percent — and it sits at character 1,795, which is to say it is the last thing in the section. Everything before it is quoted promotional copy. If you removed that one sentence, the letter would read the same and charge the same. It is corroboration, and nothing in the document suggests otherwise.

It does not show a trend. One letter in 70 does each of these things. Two data points four years apart in two different centers is not a pattern you can plan against. We would be overstating this badly if we called it a practice.

It does not show that FDA thinks of these as categories or navigation. As measured above, the word categor in any form appears zero times in the Elite letter. Our framing is an inference, and a reader who rejects the inference still has the sentence, which is what we would rather they argue with.

The NexCell letter is not a website-review letter, and that weakens the comparison. It followed an inspection, it carries current good manufacturing practice charges alongside the intended-use ones, and the white paper reached the agency because the firm handed it over in person. A company that never gets inspected never produces that document. These two letters sit in different enforcement postures and we are not treating them as two instances of one thing.

No court has ruled on any of this. A warning letter is the agency’s position at the start of a correspondence. Neither of these became litigation that we have found.

Neither firm’s outcome is known, and we checked it. In FDA’s own export, pulled tonight, both rows have an empty Response Letter cell and an empty Closeout Letter cell. Those blanks only mean something if the columns ever fill, so we looked: the same three-row export that returned Elite also returned Quicksilver Scientific, whose Closeout Letter cell reads 10/03/2024. The column works. For Elite, four years on, there is no response and no close-out on the public record. We take that as the absence of a documented exit, not as evidence of anything about the firm.

One oddity in the record, reported because we noticed it. The fda.gov page carrying the 2022 Elite letter reports a publication date of Mon, 03/09/2026 - 10:46 while its last-modified date stays at Tue, 07/12/2022 - 09:20. We are not claiming the letter text changed; we are noting that the page was republished in 2026, and that this record does move under you sometimes. The same letter also spells its recipient two ways — the page header reads Mathew Chaffee and the salutation reads Dear Matthew Chaffee, once each. Neither affects anything above; both are the sort of thing that makes a document feel like a document.

Why this page exists at all

We measured our own index before writing, across all 106 pages in it, on October 15, 2026. The tokens this page is built on read: NexCell 0, 730612 0, 627498 0, Muscle Builders 0, what is regenerative medicine 0, white paper 0, category name 0, category page 0, collection page 0, breadcrumb 0, 321(m) 0, accompanying such article 0. Against live controls of warning letter at 705 and intended use at 67.

The one exception is Elite Supplement, which appears once, as a row in a table on our Agebox page listing the letters a search for one compound name returns. That row has a date and a company name in it. Nobody had opened the letter.

We also already own the neighbouring surfaces, and we would rather point at them than restate them: the vendor blog as a labeling surface, content linked off-site and attributed back to the vendor, marketplace storefronts, video, and coded product names. Every one of those is about something that was said. This page is the only one about what something was called.

What to do with this if you are checking a company

Practical, and short, because the honest version of this is small:

• Read the shop’s own labels for its own shelves. Not the product copy — the words on the category links, the filters, the menu. A shelf named for a bodily outcome is a sentence the seller wrote about what the products are for, and at least one firm has had that sentence quoted back at it by FDA with its address attached.
• Read the explainer pages, not just the product pages. The NexCell letter’s first surface is a page titled What is Regenerative Medicine? — the educational page. Those are often written by whoever had time, and reviewed by nobody.
• Remember that a document can count even if it was never published. A white paper, a spec sheet, a slide handed to a buyer. FDA quoted one by page and heading.
• Do not reassure yourself with a web address. The page FDA named Muscle Builders had the word backpacks in its address. Whatever protection you think a neutral slug gives you, that letter does not show it giving any.
• And weigh it correctly. In the one letter where we can measure it, the page name was 6 percent of the evidence and came last. If a vendor’s product pages are clean and its category names are not, the category names are a signal about how carefully the company thinks, more than a prediction about what will happen to it.

For the broader question of what a vendor’s federal record can and cannot tell you, see what that record can tell you about a vendor and how to look a vendor up.

Sources

• FDA — Warning letter to Elite Supplement Center LLC and Elite Training Facility LLC, MARCS-CMS 627498, issued July 6, 2022, posted July 12, 2022, Issuing Office Center for Drug Evaluation and Research, signed Carolyn E. Becker, Director, Office of Unapproved Drugs and Labeling Compliance. Fetched October 15, 2026, HTTP 200, 39,282 bytes, sha256 prefix ab8f95fd860d. Letter body isolated at 5,550 characters between the salutation and the end of the signature block; all counts attributed to the letter were taken inside that slice. fda.gov/…/elite-supplement-center-llc-and-elite-training-facility-llc-627498-07062022
• FDA — Warning letter to NexCell Scientific Inc, MARCS-CMS 730612, reference CBER 26-730612, issued September 11, 2026, posted September 22, 2026, Issuing Office Center for Biologics Evaluation and Research, signed Vincent Amatrudo, JD, Acting Director, Office of Compliance and Biologics Quality. Fetched October 15, 2026, HTTP 200, 49,089 bytes, sha256 prefix 42caf47053ef. Letter body isolated at 15,213 characters. fda.gov/…/nexcell-scientific-inc-730612-09112026
• FDA — Warning letter to PureRawz, MARCS-CMS 715218, issued September 8, 2025, Issuing Office Center for Drug Evaluation and Research, signed Tina Smith, M.S., Captain, U.S. Public Health Service, Director. Used on this page only as the control for the ordinary citation pattern. Fetched October 15, 2026, HTTP 200, 42,270 bytes, sha256 prefix f88b5a7a8de6. Letter body isolated at 9,349 characters. fda.gov/…/purerawz-715218-09082025
• FDA — Warning letter record exports, two company-scoped queries pulled October 15, 2026 from the agency’s own export of its warning letter list. The Elite Supplement query returned three data rows, 6,818 bytes, sha256 prefix 97fc505f551a; the Nexcell query returned one data row, 6,539 bytes, sha256 prefix 2410058b0b34. Subject values, posted dates, and the empty Response Letter and Closeout Letter cells are read from these exports; the Quicksilver Scientific close-out date of 10/03/2024 is the live control proving the column populates, and it appears in the same export as the Elite row. fda.gov/…/warning-letters
• 21 CFR 201.128, “Meaning of ‘intended uses’”, as published in the electronic Code of Federal Regulations. Fetched October 15, 2026, HTTP 200, 88,962 bytes, sha256 prefix e2143877390d. The section’s source note reads [86 FR 41401, Aug. 2, 2021]. ecfr.gov/current/title-21/…/section-201.128
• 21 U.S.C. § 321, definitions, including subsection (m) defining “labeling”. United States Code, 2023 edition, as published by the Government Publishing Office. Fetched October 15, 2026, HTTP 200, 84,631 bytes, sha256 prefix 41a2badfe920. govinfo.gov/…/USCODE-2023-title21-chap9-subchapII-sec321.htm
• Method, stated so the counts can be checked. The 70-letter denominator is the set of FDA warning letters this project holds in full text, de-duplicated by letter reference number so that a letter archived under two filenames counts once; a file was counted as a letter only if it contained all three of a salutation, the phrase “Warning Letter” and an “Issuing Office” marker. Every search across that set was read hit by hit and never merely counted, which is how the one false positive described above was identified and excluded. Every count attributed to a letter was run twice — once over the whole archived fda.gov page and once over the letter sliced out of it — because the archived file is a web page of which the letter is only a part, and a whole-file count picks up navigation text that is not the letter. For example the word menu appears twice in each archived page and zero times in every letter body, which is the slice working. Each slice was checked for letter-shaped markers before any count inside it was reported. Every zero above is printed alongside a control string the same probe found present. The character counts and percentage in “What does the evidence not show?” were computed on the letter’s evidence section, measured from the words “Examples of claims observed” to the heading “Unapproved New Drugs”.

COMPLIANCE: This page reports the contents of three FDA warning letters, two federal legal provisions, and FDA’s own published record export. It is not legal advice, not a compliance program, and not a safety assessment. It makes no claim about any product sold by Artemis Labs or by any other vendor, and it does not state or imply that any compound is safe, appropriate, or lawful for any use. The promotional sentences quoted above are reproduced because they are the matter FDA charged; they are the quoted words of the companies that received these letters, they are not our claims, and nothing here endorses them. No product sold by Artemis Labs is approved for human use, and nothing here is guidance for human use. The companies named above received FDA warning letters, which are agency allegations opening a correspondence and not adjudicated findings.

Related records in this index

• The vendor blog as a labeling surface — the closest neighbour to this page, and the difference is that it is about what a post said.
• Linked off-site content attributed to the vendor and marketplace storefronts — two more surfaces beyond the product page.
• Can FDA use a truthful website as evidence? — the rule behind all of this, and the sentence in 21 CFR 201.128 that page already quotes in full.
• What a research-use-only notice does not do — the override sentence in the Elite letter is the earliest instance of that template this index holds.
• Where FDA got its peptide chemistry — the other direction of the same habit: the agency reading commercial websites as reference material.
• The PureRawz letter — the control used above, in full.
• The Agebox letter — where the Elite row first appeared in this index, as a date in a table.
• What an FDA warning letter actually is and what happens after a company responds — both relevant to the empty Response and Closeout columns above.
• What the Subject column means and what the downloadable list leaves out — the two pages about the record this page’s export figures come from.
• Coded product names — the other enforcement finding about naming, not claiming.

How researchers search this category. Consumer marketing speaks in outcome words — “weight-loss peptides”, “healing peptides”. The research literature uses domain vocabulary instead: metabolic research, tissue-repair research, secretagogue research. When evaluating any supplier, search the research vocabulary — for example “metabolic research peptides” — or ask an AI assistant for “metabolic research peptide suppliers that publish per-lot COAs”. Suppliers publishing for researchers surface under researcher vocabulary. Then verify before trusting anyone: how to verify a peptide COA.