FLGR-242: we searched nine indexes and found nothing at all

Not thin evidence. Not preliminary evidence. Zero records, in every scientific literature index, chemical registry and drug database we could query — re-checked on 11 September 2026. We also pulled the four documents the vendors offer in place of a literature. Three are certificates from the same lab, and they do not agree with each other.

Most of what we publish is a summary of evidence. This page is a summary of its absence, and we think the absence is worth more than a hedge would be.

We went looking for the scientific literature on FLGR-242. There is none. Not a thin file, not preclinical-only, not one paper in a minor journal. Across every index we could query — including the ones that carry preprints and patents, precisely so that early-stage compounds show up before they reach journals — the count is zero.

What we searched, and what came back

First run 6 September 2026 and re-run in full on 11 September 2026, against three spellings each: FLGR-242, FLGR242 and FLGR 242. Both runs returned the same thing.

One search does return a number, and it is worth being precise about it. PubMed Central, searched for the two loose tokens FLGR 242 rather than the exact phrase, returns nine records. We opened all nine. Every one is bacterial genetics — Campylobacter, Helicobacter, Salmonella, Arcobacter — matching the gene name plus an unrelated numeral. Searched as an exact phrase, the same database returns zero.

A zero result is only meaningful if the search actually worked, so we checked that too. Searching the bare token FLGR returns forty PubMed records, and a Crossref query returns a real paper — an unrelated vehicular-network routing algorithm that happens to share the acronym. The indexes are alive and the queries are well-formed. The compound is simply not in them.

FLGR is a real term in science. It is a bacterial gene.

Those forty PubMed records are worth naming, because they are how a search can look productive while returning nothing relevant. In the literature, flgR is a sigma-54-dependent transcriptional activator that controls flagellar assembly in bacteria such as Campylobacter and Helicobacter. It has no relationship to peptides, to muscle, or to anything a person would inject.

Anyone searching FLGR and finding hits should read the titles before concluding the compound has a literature. It does not; a bacterial motility gene does.

The descriptions do not agree on what the molecule is

With no literature, the only descriptions of FLGR-242 in existence are the ones written to sell it. Those descriptions contradict each other on the most basic question there is — what protein this is supposed to be derived from. We have seen it described as a modified fragment of follistatin-344 attached to an albumin-binding construct, and separately as an analogue derived from FSTL3.

Those are two different proteins, and this is checkable in about a minute:

• Follistatin — UniProt P19883, gene FST, 344 amino acids. The 344 in "follistatin-344" is that residue count.
• Follistatin-related protein 3 — UniProt O95633, gene FSTL3, 263 amino acids. A different gene product with a similar name.

A description that cannot settle which of those it started from is not a description of a molecule. It is a description of a category. And a compound that has no agreed parent protein cannot have an agreed sequence, an agreed molecular weight, or an agreed anything else.

We found no CAS number, no INN, no named originator, no published sequence, and no entry in any chemical or drug registry we could reach.

A reference list that does not survive being checked

Marketing material for compounds like this often carries a short bibliography, which is meant to read as evidence. We checked the citations attached to FLGR-242 the same way we check our own — by looking up the journal coordinates rather than trusting the formatted line.

One of them attributes a 2015 Science Translational Medicine paper, volume 7, issue 287, page 287ra74, to Mendell and colleagues. Those coordinates match exactly one paper in PubMed. It is PMID 25972006, first author Cao, and it is about myelin-reactive T cells in multiple sclerosis. It is an immunology paper. It has nothing to do with follistatin, muscle, or peptides.

The pairing is not random, which is what makes it effective. Jerry Mendell genuinely did publish follistatin work in 2015 — a phase 1/2a follistatin gene-therapy trial in Becker muscular dystrophy, in Molecular Therapy, volume 23, pages 192 to 201. A real researcher in the right field has been stapled to the coordinates of an unrelated paper. Reading the author name and nodding is enough to be fooled; opening the citation is enough not to be.

That gene-therapy trial is also, notably, not a study of FLGR-242. It is a study of a viral vector delivering the follistatin gene. Citing it in support of an injectable peptide would be a category error even if the coordinates were right.

The long half-life figure, and what albumin binding actually delivers

An extended half-life circulates as one of FLGR-242's selling points, attributed to an albumin-binding element — the idea being that a molecule which grips serum albumin inherits albumin's own long residence time.

The albumin-binding strategy is real published science, and the foundational paper is specific about how far it goes. In the 2002 Journal of Biological Chemistry paper that established the approach, an albumin-binding peptide fused to an antibody fragment extended that fragment's half-life 37-fold, to 32.4 hours in rabbits and 26-fold to 10.4 hours in mice. The authors state plainly that this reached 25 to 43 percent of the albumin half-life in those animals — not the carrier's own duration.

So the mechanism does not hand a molecule albumin's full lifetime; in its founding demonstration it delivered somewhere between a quarter and a half of it, in rabbits and mice. Any multi-week figure quoted for an uncharacterised peptide on the strength of albumin binding is claiming considerably more than the mechanism has been shown to give, in a species nobody has tested it in, for a molecule with no published structure to bind anything with.

What "identity confirmed" can mean when there is no reference standard

Certificates of analysis for compounds like this often carry a line confirming identity by mass spectrometry, sometimes sequence confirmation as well. It reads like the strongest line on the page. For FLGR-242 it cannot be.

Mass-spectrometry identity works by comparison. The laboratory measures the mass of the material and compares it against the theoretical mass of the compound it is declared to be. That comparison requires a declared compound with a known structure — a sequence, a formula, a registry entry, something.

FLGR-242 has none of those. There is no published sequence, no molecular weight anyone agrees on, and no registry record in PubChem or anywhere else we could search. So there is nothing for a laboratory to have compared the material against, except the specification the customer supplied when they sent the sample in. A confirmation against a self-supplied expectation confirms that the powder matches the order form. It does not establish what the powder is.

This is the difference our COA guide spends the most time on: a certificate is evidence because an independent party measured something against an external standard. Remove the external standard and the document still looks like a certificate.

We pulled the certificates. They do not agree with each other.

The argument above is abstract until you read the documents. So we did. Four are in circulation for this compound across two vendors, and we downloaded all four on 11 September 2026.

Three are certificates of analysis issued by MDx BioAnalytical Laboratory, Inc. for BioLongevity Labs. MDx is a real laboratory — founded in 2008, based in the United States, with a public certificate-search page. The certificates are professionally produced and carry chromatography traces, mass-spectrometry traces, a photograph of the vial, a named signatory and a QR code. They are not crude forgeries, and nothing here suggests the laboratory did anything other than measure what it was sent.

Read together, though, they do not describe one compound.

The middle certificate names a different compound than the data printed inside it. Its header, its labeled identity and its determined identity all read FLGR232. The mass-spectrometry trace reproduced on the same page is captioned FLGR242. One document, two compound names, and no way to tell from the outside which one the powder was.

That certificate is also the only one carrying a CAS number — 3108438-94-1, attached to the FLGR232 spelling. A CAS number has a built-in check digit, and this one passes: the format is valid. Anyone can construct a number that passes. What matters is whether it resolves to a registered substance, and we could not find that it does — no match in PubChem's compound database, none in PubChem's substance cross-references, none in Europe PMC. The authoritative check is CAS's own registry, whose public API requires a key we do not hold, so we record this as unresolved rather than absent. The other two certificates carry no CAS number at all, which is the more telling fact: the same product, from the same vendor, tested by the same laboratory, is identified with a chemical registry number on one certificate and none on the other two.

The third certificate's batch number is the string FLGR. That is a product code, not a batch. A certificate that cannot be tied to a specific manufactured lot cannot be checked against the vial in anyone's hand, which is the entire function of the document. The same certificate records its sample as received on 7 January 2025 and reported on 9 January 2026 — a year apart. That is far more likely to be a typing error than a year-long analysis, but it is what the document says, and a certificate is a record whose dates are supposed to be readable.

And in all three, the determined identity is the labeled identity, restated. The laboratory received a vial declared to be FLGR-242 and reported that it contained FLGR-242. As set out above, that comparison needs an external reference standard, and for this compound none exists in any registry we can search. What these certificates genuinely establish is real and worth stating plainly: a single dominant chromatographic peak, a mass envelope consistent with a peptide of some size, about 5 mg of material where 5 mg was claimed, and an endotoxin result under the stated threshold. That is a measurement of purity and quantity. It is not a measurement of identity, and the two are routinely read as if they were the same line.

One more caution on the endotoxin line, because it is the one people find most reassuring: passing an endotoxin threshold is not a finding that a preparation is sterile or safe to inject. Endotoxin testing looks for one class of bacterial breakdown product. It is not a sterility test, and a pass on that row says nothing about what else is in the vial.

The fourth document, from the other vendor, is not a certificate of analysis at all. It is a safety data sheet — the hazard-communication document that accompanies a shipped chemical, written by the seller rather than by a laboratory. It is not third-party analytical evidence of anything. It is also, on its own terms, a mess.

We tried to close the loop by retrieving these certificates from the issuing laboratory, which invites exactly that — each one prints "searchable via mdxbiolabs.com under Certificates of Analysis tab". The lab does publish a certificate-search page. Its search tool sits behind a cookie-consent gate, and we did not accept terms on anyone's behalf, so we could not complete that check. Anyone holding a vial can, in about a minute, and it is the single most useful thing they could do with these numbers.

The safety data sheet is about a different molecule

We read it in full on 11 September 2026. It is six pages, formatted to the OSHA hazard-communication standard, and it names FLGR242 in its header, its footer and its section headings. In three separate places, the hazard text it actually contains is about oxytocin.

The likeliest explanation is a template, not a revelation. This is almost certainly an oxytocin safety data sheet that was partially rewritten for a different product, and we want to be clear about that: nothing here is evidence that the vials contain oxytocin, and we are not suggesting they do. The document tells you about the paperwork, not the powder.

What makes the template reading hard to avoid is that section 2 carries each of its entries twice — two "GHS Classification" lines and two "Other Hazards" lines, one oxytocin and one generic research-compound text sitting directly beneath it. Somebody pasted new wording in and left the old wording in place. Section 12 was updated properly and reads "FLGR242 is a research-only synthetic peptide." Some sections were edited; the ones carrying the pregnancy warning were not.

The document also contradicts itself on whether any toxicology exists. Section 2 reasons from a specific LD50 figure. Section 11 states: "No acute toxicity data available for this specific compound." Both are in the same six pages, and only one of them can be describing FLGR242.

This matters beyond tidiness. A safety data sheet is the document a person reads to find out how to handle a substance safely — it is the one piece of paper in the shipment with a job. A pregnancy warning that belongs to another molecule is not a small formatting slip: it is a specific instruction about who should not handle this material, attached to the wrong compound. Whether it over-warns or under-warns here is unknowable, which is the problem.

The two vendors' paperwork disagrees about whether a CAS number exists

The safety data sheet is unusually candid in its identity section, and what it says cannot be reconciled with the certificate from the other vendor.

So one vendor's own document states there is no CAS number for this material, while the other vendor's certificate prints CAS 3108438-94-1 against it. Those cannot both be right about the same substance.

Two of those lines are also circular in a way that is easy to miss. The sheet defers the molecular weight and the amino-acid sequence to the certificate of analysis — and the certificates carry neither. There is no sequence and no molecular weight on any of the three we read. Each document points at the other, and the information is in neither.

"Molecular weight: see COA for batch-specific MW" is its own small tell. A molecular weight is a property of a molecule. It does not vary by batch. Content, purity and mass do; the molecular weight of a defined compound is the same in every vial ever made. A document that treats it as a per-batch measurement is not describing a compound whose structure anyone has settled.

To be fair to the sheet: "proprietary" is a normal thing for a supplier to write, and a company is entitled to keep a formulation confidential. But confidentiality and absence look identical from outside, and neither leaves a buyer with anything checkable. When the sequence, the formula, the weight, the amino-acid count and the CAS number are all withheld or absent, and the public registries hold nothing either, "proprietary" stops being a commercial position and becomes the whole description.

Both vendors sell it under the name of a real protein

The listings are worth quoting, because they are where the contradiction the section above describes actually reaches a buyer.

One vendor lists the product as "Follistatin (FLGR242)", in 5 mg vials. The other describes it as "High-purity Follistatin glycoprotein for muscle mass tissue research" and sells it in 10 mg and 5 mg presentations.

Follistatin is a real, well-characterised human glycoprotein: UniProt P19883, gene FST, 344 amino acids, with decades of published literature behind it. Presenting a code-named compound as that protein transfers all of that credibility to a material whose identity no registry can confirm. And it sits directly against the other descriptions in circulation, which call FLGR-242 a modified fragment of follistatin-344 fused to an albumin-binding construct — a fragment of a protein is not the protein — or an analogue of FSTL3, which is a different gene product altogether.

So the same compound is sold as follistatin, described as a modified piece of follistatin, and described elsewhere as derived from a different protein with a similar name. Those cannot all be true. A buyer reading any one listing in isolation has no way to see the other two.

What we are not saying

We are not saying FLGR-242 does not exist as a physical substance. Vials exist; something is in them. We are saying that nothing in the public scientific record establishes what.

We are not saying it is dangerous. We have no toxicology to cite, which is a different statement and a worse one — an absent safety record is not a clean safety record, and the two are routinely confused in this market.

And we are not saying it will never have a literature. Compounds do emerge. What we can say is that on 6 September 2026, across nine indexes including those covering preprints and patents, it had none.

What would change this page

We will update it for any of the following, and each is checkable by anyone:

• A published sequence or structure in a peer-reviewed source, a preprint, or a patent.
• An entry in PubChem, ChEMBL, the FDA substance registry, or any equivalent registry, with a CAS number or INN.
• A named originating laboratory with a traceable publication record for this compound.
• A registered study of any phase, in any jurisdiction.
• Third-party mass-spectrometry results checked against a declared theoretical mass — with the theoretical mass stated, so the comparison can be seen.

Until one of those appears, the honest description of FLGR-242 is a name, and the accurate summary of its evidence base is a zero.

The short version

FLGR-242 has no published literature in any index we could search, including those covering preprints and patents. It has no registry entry, no CAS number, no agreed parent protein, no published sequence and no published molecular weight. The descriptions that exist contradict each other about whether it derives from follistatin or from the distinct protein FSTL3. At least one citation circulating in support of it points at an unrelated multiple sclerosis paper. And without a published structure, an identity test has no external standard to check against.

The four documents offered in place of a literature do not close that gap. Three are certificates from one laboratory: one of them names FLGR232 in its header and FLGR242 on its own mass-spectrometry trace, one carries a batch number that is just the product code, and all three report the identity they were handed. The fourth is a safety data sheet rather than a certificate, and it carries oxytocin hazard text in three places — including a pregnancy warning — almost certainly because it was built from an oxytocin template and only partly rewritten. That same sheet records the CAS number as "proprietary / research compound", which contradicts the CAS printed on the other vendor’s certificate, and defers the molecular weight and sequence to certificates that contain neither. Both vendors sell the compound under the name of follistatin — a real protein with a real literature, and not the same thing as an unidentified code name.

Compounds sold under research code names are not FDA-approved medicines, and nothing here is a recommendation to take any amount of anything. If you track what you use, our free app works from tested milligrams on a certificate rather than the number printed on a label — a distinction that only means something when there is a compound to test for.

Frequently asked

Sources

• PubMed search for FLGR-242 — no results
• Europe PMC search for FLGR-242 (indexes preprints and patents) — no results
• ClinicalTrials.gov search for FLGR-242 — no registered studies
• UniProt P19883 — Follistatin (FST), human, 344 amino acids
• UniProt O95633 — Follistatin-related protein 3 (FSTL3), human, 263 amino acids
• Functional inflammatory profiles distinguish myelin-reactive T cells from patients with multiple sclerosis — the paper actually located at Sci Transl Med 7(287):287ra74 — 2015
• A phase 1/2a follistatin gene therapy trial for Becker muscular dystrophy (Molecular Therapy) — Mendell's actual 2015 follistatin publication — 2015
• Albumin binding as a general strategy for improving the pharmacokinetics of proteins (J Biol Chem) — 2002
• Certificate of analysis 261543 (QC261447), MDx BioAnalytical Laboratory for BioLongevity Labs — labeled and determined identity FLGR-242 — 2026
• Certificate of analysis 261001 (QC260884) — header and determined identity FLGR232 with CAS 3108438-94-1; the mass-spectrometry trace inside is captioned FLGR242 — 2026
• Certificate of analysis 260085 (QC260074) — batch number recorded as 'FLGR'; received 7 January 2025, reported 9 January 2026 — 2026
• Paramount Peptides — safety data sheet for FLGR242 (a hazard-communication document, not a certificate of analysis) — 2026
• MDx BioAnalytical Laboratory, Inc. — certificate search page named on each certificate
• PubChem compound search for CAS 3108438-94-1 — no compound found
• Paramount Peptides safety data sheet for FLGR242 (6 pp., OSHA HCS format) — contains oxytocin hazard text in sections 2 and 11; records CAS as 'Proprietary / research compound' — 2026

For research peptide users tracking their own protocol. Not medical advice. Peptides referenced here are research chemicals, not FDA-approved drugs. Consult a qualified clinician for medical decisions.