Sourcing, purity and testing for research peptides
A certificate of analysis answers some questions well and others not at all. Here is how to read one, what third-party testing adds, and what the published studies of online peptides found.
Most peptides discussed outside a prescription are sold labelled "for research use only." That label means the product is not an FDA-approved drug, has not been made under drug manufacturing rules, and is not intended for people. What is in the vial is therefore a question the buyer has to answer from documents and testing. This page explains what those documents can and cannot tell you. It is not advice on where to buy anything.
Four different questions
People ask "is it pure?" as though that were one question. It is at least four.
- Identity. Is the peptide in the vial the one on the label?
- Purity. Of the peptide material present, how much is the intended sequence and how much is related impurities, such as truncated or modified chains?
- Content. How many milligrams of the peptide are actually in the vial, against what the label says?
- Safety contaminants. Bacterial endotoxin, microbial contamination, residual solvents, heavy metals.
A typical certificate of analysis answers the first two. It often does not answer the third and fourth, which are the ones most relevant to anyone injecting the product.
What each test shows
| Test | Answers | Does not answer |
|---|---|---|
| HPLC (high-performance liquid chromatography) | Purity, as the main peak's share of the total area at a UV wavelength, often 214 or 220 nm | Content in milligrams; impurities that hide under the main peak or do not absorb UV; endotoxin |
| Mass spectrometry | Identity: whether the measured molecular mass matches the expected sequence | How much is there; how pure it is (unless combined with LC as LC-MS) |
| Net peptide content (amino acid analysis, quantitative assay) | How much of the powder by weight is peptide, as opposed to water, salts and counter-ions | Purity of that peptide |
| Endotoxin (LAL or recombinant assay) | Bacterial endotoxin, reported in EU per mg | Live bacteria; other contaminants |
| Sterility | Whether viable microorganisms grow | Endotoxin left by dead bacteria |
Two points catch people out. First, "99% purity" by HPLC says nothing about how many milligrams are in the vial. A lyophilised peptide is part peptide and part water, salts and counter-ions such as acetate or TFA left from synthesis, so a 10 mg vial of 99 percent pure peptide may contain noticeably less than 10 mg of peptide. Second, a sterile product can still carry endotoxin, because endotoxin comes from bacterial cell walls and survives the bacteria. The purity testing glossary and certificate of analysis glossary define the terms.
Reading a certificate of analysis
- Lot number. Does it match the lot on the vial? A certificate for a different batch describes a different product.
- Who ran it. In-house testing by the seller, or an independent lab? Is the lab named, and is it accredited (ISO/IEC 17025 is the usual standard for testing labs)?
- Date. Peptides degrade. A certificate from long before the vial reached you describes the powder then, not now.
- The raw data. A chromatogram and a mass spectrum, not just a number. A certificate that reports "99.5%" with no trace is a claim, not a result.
- What is missing. No content, no endotoxin, no sterility is common. It means those questions are unanswered, not that the answers are good.
What published testing has found
| Study | Year, venue | Samples | Method | Findings |
|---|---|---|---|---|
| Ashraf et al. | 2024, J Med Internet Res (peer reviewed) | Semaglutide test purchases from 6 online sellers without prescription: 3 vials delivered, 3 pens never arrived | LC-MS, sterility, endotoxin, packaging inspection | Content 28.56% to 38.69% above label; no peptide-like impurities identified; no viable microbes; endotoxin in every sample (2.16 to 8.95 EU/mg); purity reported by the authors as 7.7% to 14.37% against 99% on the labels |
| Mendias and Awan | 2026, Preprints.org (not peer reviewed) | 6,441 samples across 14 compounds, from a publicly available independent testing dataset | Purity, measured abundance and endotoxin against two acceptance standards | 41.6% to 71.1% failed basic quality criteria depending on the standard applied; measurable endotoxin in 15% |
The Ashraf study is small and looked at products sold as semaglutide without a prescription, not at products labelled for research. Its purity and content figures are hard to reconcile without the full paper, and we report them as the authors stated them. The Mendias and Awan preprint is much larger and covers the compounds most people ask about, including BPC-157, TB-500, tirzepatide and retatrutide, but it has not been through peer review and relies on a public dataset of independent tests, which may not be a random sample of what is sold. Neither tells you about any individual product. Together they show that label claims and measured content often diverge, and that endotoxin is a recurring finding.
Where licensed supply fits
There are three legal channels for a peptide medicine in the US. An FDA-approved product, made under current good manufacturing practice. A 503A compounded preparation, made by a licensed pharmacy for an individual prescription, from a bulk substance that is an approved drug component, has a USP monograph, or appears on the FDA's 503A bulks list. Or a 503B outsourcing facility, held to manufacturing standards. Products labelled "research use only" sit outside all three. Several of the most discussed research peptides are currently under review for the 503A list; see the PCAC vote and peptide policy. The FDA's stated concerns for these substances include immunogenicity, peptide-related impurities and difficulty characterising the active ingredient, which are the same issues a certificate of analysis is meant to address.
A clinician who prescribes, and a pharmacy that dispenses, take on responsibility for those questions. That is the practical difference a licensed provider makes. The compounded and research chemical glossary entries explain the categories.
Keeping records
Whatever the source, the vial in front of you has a lot number, a date and a document. Record them. If a reaction happens, the first question anyone will ask is which batch. Peptide AI's vial tracker stores the lot, the reconstitution date and the certificate against each vial, and the dose log ties every dose back to the vial it came from.
Questions people ask
What does a certificate of analysis show?
Usually identity, by mass spectrometry, and purity, by HPLC. Many certificates do not report net peptide content, endotoxin or sterility, which are separate tests.
Does 99 percent purity mean the vial contains the labelled amount?
No. HPLC purity is the share of the peptide material that is the intended sequence. A lyophilised powder also contains water, salts and counter-ions, so content in milligrams needs a separate quantitative test.
What is endotoxin and why does it matter?
Endotoxin is a component of bacterial cell walls that can cause fever and inflammation when injected. It survives after bacteria die, so a product can be sterile and still contain it. A 2024 JMIR study found endotoxin in every semaglutide sample bought online without prescription.
How often do research peptides fail quality testing?
A 2026 preprint, not yet peer reviewed, analysed 6,441 samples across 14 compounds and found 41.6 to 71.1 percent failed basic quality criteria depending on the standard used, with measurable endotoxin in 15 percent.
Is "research use only" a legal way to buy peptides for personal use?
The label means the product is not an FDA-approved drug and is not intended for human use. Legal routes to a peptide medicine in the US are an FDA-approved product, a 503A pharmacy prescription from an eligible bulk substance, or a 503B outsourcing facility.
Sources
- Multifactor quality and safety analysis of semaglutide products sold by online sellers without a prescription, Ashraf et al., Journal of Medical Internet Research, 2024
- Evaluation of research grade peptides marketed directly to consumers reveals extensive variability in purity and measured abundance (preprint, not peer reviewed), Mendias and Awan, Preprints.org, 24 April 2026
- Certain bulk drug substances for use in compounding that may present significant safety risks, U.S. Food and Drug Administration, page updated 22 April 2026
- Bulk drug substances used in compounding under section 503A of the FD&C Act, U.S. Food and Drug Administration
- Compounding and the FDA: questions and answers, U.S. Food and Drug Administration
- July 23-24, 2026: Meeting of the Pharmacy Compounding Advisory Committee, U.S. Food and Drug Administration
