Why a certificate of analysis does not make a vial safe

Sellers of unlicensed peptides often answer safety questions with a document: a certificate of analysis showing a purity figure, sometimes a "third-party tested" badge. This page explains what such a certificate measures, the long list of things it does not, and why no piece of paper can turn a hand-filled vial into a medicine.

What a certificate of analysis is

A certificate of analysis is a document reporting the results of tests run on a sample. For a peptide it typically shows one figure from HPLC, which separates a sample into its components and estimates what proportion of the peptide-like material is the intended compound. It may include a mass spectrometry trace confirming that a molecule of the expected weight is present.

In licensed manufacturing, a certificate of analysis is one document among many, generated inside a quality system and tied to a specific, traceable batch. On its own, outside that system, it is a sheet of paper with a number on it.

Five things a purity figure is not

It is not identity. HPLC purity tells you what fraction of what was detected matches the main peak. It does not, by itself, prove that the main peak is retatrutide rather than a similar peptide or an incomplete chain. The 2024 semaglutide study found a single clean peak in each sample and still classed every sample as substandard and falsified, because of what else was in the vial.^[1]

It is not the correct amount. A vial can be 99% pure and contain twice the labelled milligrams, or half. The Queensland test found nearly double the label; the Melbourne series found roughly as many samples over the label as under it.^[4][5] Purity and quantity are different measurements, and a purity-only certificate says nothing about strength.

It is not sterility. Sterility testing is a separate microbiological test that takes days. An HPLC run does not detect bacteria or fungi. A product can be chemically pure and still infect the tissue it is injected into.

It is not endotoxin. Endotoxin is tested with a specific assay that has nothing to do with HPLC. Every online semaglutide vial in the 2024 study contained it, at up to 8.95 endotoxin units per milligram, whatever its paperwork said.^[1]

It is not your vial. Even if all of the above were tested honestly, they were tested on a sample that is not the one you have.

Batch versus vial

In licensed manufacture, a batch is a defined quantity made in one validated run, with every vial filled by the same process, and testing is done on samples drawn from that batch under rules about how many and from where. The certificate describes the batch because the process guarantees the vials are alike.

A hand-filling operation has no such guarantee. Tens of thousands of empty pens next to raw powder, as the MHRA found in Northampton, means each pen is only as accurate as the person filling it.^[6] A certificate for "the batch" is describing a sample from a process that has no controls on consistency. Two pens from the same box can differ, and the Melbourne testing suggests they do.^[5]

Copied, reused and forged certificates

Certificate reuse and forgery is a known problem in the unlicensed peptide trade. A document can be copied from a different product, edited, or attached to batches it never described. There is no register to check it against and no inspector who has seen the laboratory.

The 2024 study is the clearest published example. Sellers' documentation claimed 99% purity; laboratory testing found 7.7%, 9.0% and 14.4%. When the researchers applied a standard checklist for inspecting medicines, the packaging and labelling failed 13 to 14 of 22 criteria, or 59% to 64%.^[1] The paperwork was not just optimistic; it bore no relation to the product.

"Third-party tested"

The phrase sounds like independent oversight. It usually means the seller sent a sample of its choosing to a laboratory of its choosing and published the result it received. The Melbourne chemist described businesses seeking testing precisely so they could give customers "assurance through a certificate", while her own results showed the peptide was typically about a tenth of what was in a vial.^[5]

Independent testing done by journalists and public-health researchers, rather than by sellers, is what produced the findings on the testing page. They are not reassuring.

No public reference standard

To measure how much of a substance is in a sample, a laboratory needs a reference standard: a certified, characterised quantity of the pure substance to compare against. For licensed medicines these are published in pharmacopoeias and supplied by official bodies.

Retatrutide is an investigational drug that has not been licensed anywhere.^[2] There is no pharmacopoeial monograph and no public reference standard. Any outside laboratory testing a product "as retatrutide" is working from whatever standard it could obtain, which may itself be an unlicensed product. That does not make such testing worthless, but it means even an honest certificate carries an uncertainty that the seller will not mention.

What GMP actually involves

Good Manufacturing Practice is the legally enforced quality system under which licensed medicines are made. In the UK the MHRA licenses manufacturers, inspects their sites, and can suspend or revoke a licence.^[3] For a sterile injectable, GMP means, among other things:

• validated clean rooms with controlled air quality, monitored continuously;
• sterile filling processes that have been proven to work, with operators trained and tested;
• starting materials of known origin, tested before use;
• every batch tested for identity, strength, purity, sterility and endotoxin before release, by a named qualified person who is personally accountable;
• full documentation so that any vial can be traced back through every step;
• regular unannounced inspections and a recall system if something goes wrong.

A certificate of analysis is one output of that system. On its own it is not the system. A warehouse with empty pens and raw powder is not the system either, however many certificates it emails out.^[6]

Why this page is not a checklist

It would be easy to turn the above into "what to look for in a good certificate". That would be a mistake, and this site does not do it. There is no version of an unlicensed, hand-filled injectable that becomes safe because its paperwork improves. The Victorian liver-injury cases involved products that were presumably sold with the usual assurances.^[7] A certificate did not protect those people, and it would not have, because the things that hurt them are not the things a certificate measures.