Short answer: A peptide COA usually answers two questions: is this the right molecule (mass spectrometry), and how much of what is in the vial is that molecule versus other peptide-like impurities (HPLC purity, as a percent of peak area). Neither tells you the vial is sterile or low in endotoxin, and HPLC purity is not the same as how many milligrams of peptide you actually have. For injectables, look for a batch number that matches your vial, the testing lab's name, the actual chromatogram and mass spectrum, and separate endotoxin and sterility results. In its 2026 reviews, FDA said public COAs for TB-500 lacked tests for impurities, aggregates, microbial quality and endotoxin.
Online, a certificate of analysis (COA) is often treated as proof that a peptide is "pharmaceutical grade." It is not. A COA is a test report: useful when it is real and complete, and easy to misread or fake when it is not. This guide explains each section in plain words and shows what FDA says is usually missing. It does not recommend any seller or lab.
A COA is a document from a testing lab that lists tests run on one specific batch (lot) of a product, the method used, the acceptance limit and the result. In regulated settings it matters: FDA's rules for pharmacy compounding under section 503A say bulk ingredients "must be accompanied by a valid certificate of analysis" and come from an FDA-registered establishment. For a product sold as "research use only," no one checks the COA but you. See what that label means.
High-performance liquid chromatography (HPLC) pushes a dissolved sample through a column. Different molecules come out at different times and show up as peaks on a chart called a chromatogram. "Purity 99%" usually means the main peak makes up 99% of the total peak area seen by the detector.
What that does tell you: among the things the method can see, most is one compound.
What it does not tell you:
For a sense of how strict regulated standards are: in FDA's guidance for generic synthetic peptides (such as generic teriparatide or liraglutide), FDA staff explain that any new impurity above 0.5% is not acceptable, and new impurities between 0.10% and 0.5% must be identified, characterized and justified, including for immune risk. A research COA reporting "98% purity" with no breakdown of the other 2% gives you none of that information.
Look for: the actual chromatogram (not just a number), the main peak's retention time, the method (column, wavelength), and whether impurity peaks are listed.
Mass spectrometry (MS) measures the mass of molecules. For identity, the COA should list the expected mass for the peptide's sequence and the observed mass, and they should match within the method's accuracy. Peptides often show several peaks for the same molecule carrying different charges; a real spectrum will show them.
What MS does not tell you: how much of the sample is that molecule, or whether it is sterile. And a matching mass cannot always tell apart two peptides with the same atoms arranged differently.
Look for: the spectrum itself, expected vs observed values, and the instrument type. "Identity: conforms" with no data is not evidence.
Some COAs list "net peptide content" or an assay result. This estimates what share of the powder's weight is actually peptide, after accounting for water, salts and counter-ions. It answers the question purity does not: if the vial says 5 mg, is there about 5 mg of peptide? If a COA lists only HPLC purity, you do not know the answer. That matters for your math, because the concentration you calculate when you reconstitute assumes the label amount is right.
Endotoxins are fragments of bacterial cell walls. In an injection they can cause fever, which is why FDA groups endotoxin and pyrogen testing together. The standard test is the bacterial endotoxins test in USP chapter <85>, which uses a reagent from horseshoe crab blood (Limulus amebocyte lysate, LAL) or a recombinant version. FDA has accepted LAL testing in place of the older rabbit fever test for decades.
Results are reported in endotoxin units (EU), often per milligram or per milliliter. A number on its own means little. FDA's endotoxin guidance says the right limit for an injectable is calculated with the methods in USP <85>, based on dose and route, rather than taken from an old table. For a peptide with no approved product, no official limit exists. So check that the COA states the limit it tested against, and how it was set.
FDA's 2026 reviews flagged this gap directly. For TB-500, FDA wrote that specific tests "such as tests for impurities, aggregates, microbiological quality, and bacterial endotoxin" were not available in the public domain or in the COAs it reviewed. For BPC-157, it listed microbial quality (bioburden, bacterial endotoxins) among the data that were "lacking or deemed to be inadequate."
Endotoxin and sterility are different. Sterility testing checks for living microorganisms, while the endotoxin test looks for bacterial debris. A lyophilized powder sold for "research" may never have been made or tested as a sterile product. If the COA has no sterility result, assume it was not tested.
Even a complete, honest COA describes the batch on the day it was tested. It cannot tell you how the vial was stored or shipped since, or whether your vial is from that batch. Once you mix it, stability depends on the product and the diluent; see how long reconstituted peptides last and bacteriostatic vs sterile water.
Save the COA, a photo of the vial label and its lot number. In Dose Track you can attach each vial to its compound, note its source and lot, log its reconstitution, and see every dose drawn from it. If a batch ever seems off, you will know exactly which doses came from it. For multi-peptide routines, see how to track a peptide stack.
Questions about the app? See the help center or join the Dose Track community.
Every number in this article comes from one of these sources, checked on October 22, 2026. This article is for education. It is not medical or legal advice, and nothing here is a dose recommendation.