Is retatrutide fake? What independent testing has actually found

You've found a retatrutide listing, the price looks plausible, and somewhere in the back of your mind is every headline about fake pens. Fair. The useful question is narrower than "is it fake?": when independent labs have actually tested products sold as retatrutide, what did they find? Short answer: the published data so far points less at the wrong molecule and more at the wrong amount. Here is what each source says, and what it means for checking a batch.

The regulatory position, first

Retatrutide is an investigational compound. It has no marketing authorisation in the UK. The MHRA's own words, from its 24 July 2026 consumer warning, are blunt: "It has not been authorised for use in the UK, meaning anyone selling this product is doing so illegally." The same notice says such products "haven't been tested for safety, quality or effectiveness and there's no guarantee that they even contain retatrutide."

On 28 May 2026 the MHRA's Criminal Enforcement Unit raided a country estate near Northampton and recovered a stock of unlicensed medicines, including retatrutide and tirzepatide, in what the agency described as its largest ever seizure of such products. The property was believed to be a facility to manufacture, assemble and distribute them. That is the backdrop: illicit manufacturing and assembly operations exist, and the regulator is actively shutting them down.

What follows is about the measurement question the MHRA raises, does the product contain what the label says, and what independent testing has answered so far. Biovanta supplies research compounds for in-vitro laboratory research only.

What the testing found, source by source

Two sources, two very different shapes. One is tiny and rigorous; the other is large and aggregated. Neither on its own describes "the market", and both say so in their own way.

The peer-reviewed study: right molecule, wrong amount

The Australian analysis is small, and the authors are the first to say it: "only three samples were analysed, and the findings may not reflect the broader illicit peptide market." What makes it worth reading is that it separates the three questions a buyer conflates.

Identity. All three samples matched the expected molecular weight of retatrutide (4730.477 Da). None of them was a different compound wearing a retatrutide label.

Quantity. This is where it went wrong. One vial held about half the labelled amount; the other two held well over it, at 165% and 190%. The authors' summary: measured content "ranged from approximately half to almost double" the label. An overfill is as much a failure of labelling as an underfill; both mean the number on the vial is not the number in it.

Impurities. Arsenic, cadmium, chromium, nickel and mercury were below the lab's quantitation limits. Lead, copper and zinc were detected at low levels, and the authors judged every detected element to be substantially below the relevant ICH Q3D exposure thresholds.

What was not tested. Sterility and endotoxin, which the paper flags as a gap for future work. Purity by HPLC, in the sense of related-peptide impurities, is not reported either.

The authors' own reading is the one to take away: although attention has focused on counterfeit or adulterated products, inaccurate content "alone may represent an important source of risk." That matches what purity versus quantity predicts: a sample can be the right molecule, clean, and still nowhere near its label.

The aggregate database: how to read a pass rate

Finnrick is a public database that collects lab certificates for peptide samples and scores them. On 27 October 2026 its retatrutide page listed 200 tests: 133 pass, 65 fail and 2 incomplete, the earliest dated 17 December 2024 and the latest 21 October 2026. The page states its scoring bar as purity of 99.5% or more and measured content within ±35% of the label.

Three things to keep in mind before reading 65 fails as "a third of retatrutide is fake":

• A fail is not an identity fail. Under that bar, a sample can fail on purity, on content, or both. The pass/fail count alone does not tell you how many were the wrong molecule.
• The purity bar is strict. 99.5% is a high line. Our own retatrutide batch analysed in August 2026, measured independently at 98.86%, would not clear it on purity. That is not a defence of anyone, ours included; it is a reminder that "fail" depends on where the line is drawn.
• Samples are not random. Results come from whoever had a sample tested. That can skew either way, and it is why a database count is a picture of submitted samples, not a survey of the market.

What the database is good for is the per-test record: the date, the lab and the certificate for each sample, which is the level at which a result means something.

What this means for checking a batch

The published data says the realistic risk is a sample that is the right molecule in the wrong amount, sometimes with unknown purity and almost always with no endotoxin data. So the check has to cover all three numbers, for the batch in your hand:

1. Ask for a report on this batch, not this compound. A report describes one synthesis. The batch code on the report should match the batch on the carton. How to read a certificate of analysis walks through every field.
2. Find the identity line. A mass spectrometry result matching retatrutide's molecular mass. How HPLC-MS works, and what it can and cannot tell you.
3. Find the quantity line, in mg or mg/ml. This is the number that caught every failure in the peer-reviewed study. A report with only a purity percentage has not answered it. Why purity is not quantity.
4. Verify the report at the lab. A PDF can be borrowed or edited. The lab's own verification page cannot. The five-minute Janoshik check.
5. If in doubt, test it yourself. Independent testing is available to anyone. How to get a peptide tested from the UK.

If a seller cannot show you a verifiable report with identity and quantity for the batch they are shipping, that is your answer.

Our own report

We publish ours the same way we ask you to read anyone else's. The product is listed on the store as BV-RETA. The batch analysed by Janoshik on 27 August 2026 (R40-20260203) measured 98.86% purity. The current report, its batch number and its verification key are on the testing page, so you can open the lab's original rather than take our transcription. That report covers purity and content; it is one batch, and it does not cover endotoxin. Products are supplied strictly for in-vitro laboratory research.