How to read a peptide certificate of analysis (COA)
Short answer
A certificate of analysis (COA) lists the tests run on a sample from one batch of a material, the result of each test, and the limit each result had to meet. To read one, check that it names the product, the lot (batch) number, the lab, the method, the dates, the results with their limits, and a signature or report number. Then check that the lot number matches your vial and note whether the lab is independent of the seller. A COA that lacks these basics tells you very little.
What is a certificate of analysis?
A certificate of analysis (COA) is a document that lists the tests performed on a sample, the results, and the acceptance criteria applied. Acceptance criteria are the limits a result must meet. The World Health Organization (WHO) describes a COA this way and adds that one is usually prepared for each batch 2.
A batch, also called a lot, is a specific quantity of material made so that it is expected to be uniform 1. So a COA is not a general statement about a product. It describes one batch, and strictly one sample of it. Lab reports written under the laboratory standard ISO/IEC 17025 carry a statement that the results relate only to the items tested 3.
The references used in this guide were written for pharmaceutical ingredients and for testing laboratories. This guide uses them only as a benchmark for what complete paperwork looks like.
What fields should a complete COA have?
Three public references agree closely on the core fields: the ICH Q7 guideline for pharmaceutical ingredients 1, the WHO model certificate of analysis 2, and the reporting requirements of ISO/IEC 17025 3.
| Field | What it tells you | What a weak one looks like |
|---|---|---|
| Product name | Which material was tested 1, 2. | A vague or missing name. |
| Lot (batch) number | Which batch the results belong to. A batch number is a unique code that identifies one batch 1. | No lot number, or one that differs from the number on your vial. |
| Dates | When the sample was received, when the work was done and when the report was issued 2, 3. Some COAs also give a retest date, the date when a material should be re-examined to confirm it is still suitable 1. | No date, or a date that does not fit the batch you hold. |
| Lab name and address | Who did the testing 2, 3. | No lab named, or only a logo. |
| Method | How each test was done. A report should identify the method used 2, 3. | A bare claim such as "Purity: 99%" with no method. |
| Results | What was found, given as numbers where the test produces numbers 1, 2. | Only "Pass" where a number is expected. |
| Acceptance criteria | The limits each result was compared against 1, 2. | Results with no limits beside them. |
| Signature and report ID | Who approved the report, and a unique number that identifies it 2, 3. | No signature, no report number, or unnumbered pages. |
For a peptide (a short chain of amino acids), the results section should answer more than one question. A paper by United States Pharmacopeia (USP) scientists groups peptide tests into identity (is it the right molecule?), purity (how much is impurity?) and strength (how much peptide is present?) 6. One purity number does not answer all three. ICH guideline Q6A states that identification by a single chromatographic retention time is not regarded as specific 5. In plain terms, a peak that appears at the expected time on a purity test does not by itself prove which molecule it is.
What is the difference between an in-house COA and a third-party lab report?
An in-house COA is issued by the company that made or sells the material. A third-party report is issued by a separate laboratory. The difference is independence. The International Laboratory Accreditation Cooperation (ILAC) notes that customers increasingly rely on independent evidence rather than simply accepting a supplier's word 4.
In-house COAs are normal. ICH Q7 expects a manufacturer to issue an authentic COA for each batch on request 1. But the same guideline tells manufacturers to check the reliability of their suppliers' certificates at regular intervals 1. A supplier's certificate is a claim to be verified.
Two cautions apply to third-party reports.
First, a lab reports on the sample it received. Under ISO/IEC 17025, a lab must identify any information that the customer supplied, and it is not responsible for that information 3. So check whether the report says that the product name or lot number was supplied by the customer.
Second, "third-party" does not mean "accredited". Accreditation is a formal assessment of a lab's technical competence, based on ISO/IEC 17025 4. ISO 9001 certification is different. It does not specifically evaluate a lab's technical competence 4. Accreditation also covers specific tests only. These are listed in the lab's scope of accreditation 4.
When a reseller issues a new certificate, both ICH Q7 and WHO say it should show the name and address of the lab that did the analysis 1, 2.
What are the signs of a weak COA?
- No lot number. The COA cannot be tied to any batch.
- The lot number does not match the vial. Results relate only to the items tested 3. A COA for another lot says nothing direct about yours.
- No lab named. You cannot tell who did the work or whether they were independent.
- No method. A number without a method cannot be interpreted or repeated.
- No acceptance criteria. A COA should list each test with its limits and the numerical result 1.
- One COA reused for every batch. A COA should be issued for each batch 1, 2.
- An old date. A COA describes a sample at the time it was tested. Look for a retest or expiry date 1, 2, and check that the dates fit the batch you hold.
- An edited or cropped image. The WHO model puts an identification number and "page X of Y" on every page so that each page is recognized as part of the certificate 2. ISO/IEC 17025 requires any change to an issued report to be clearly identified 3. A cropped screenshot shows neither.
How can you check a COA yourself?
- Compare the lot number on the COA with the label on the vial.
- Find the lab's name and address on the report. Confirm that the lab exists.
- If the report claims accreditation, look up the lab in the accreditation body's directory. Check that the scope of accreditation covers the test reported 4. WHO says an accredited lab should refer to its accreditation on the COA when it relates to that analysis 2.
- Note the report's identification number. Ask the lab whether it can confirm the report.
- Ask for the full document with all pages, not a screenshot.
No single check proves that a COA is genuine. Each one makes a weak COA easier to spot.
Key facts
- A COA lists the tests performed on a sample, the results obtained and the acceptance criteria applied, and is usually prepared for each batch 2.
- ICH Q7 says a COA should give the material's name, its batch number and its release date, and should list each test with its acceptance limits and numerical results 1.
- ICH Q7 says certificates should be dated and signed by authorized personnel of the quality unit 1.
- ICH Q7 defines a batch number (or lot number) as a unique combination of numbers, letters or symbols that identifies a batch and from which its production and distribution history can be determined 1.
- A report under ISO/IEC 17025:2017 is expected to include the lab's name and address, a unique identification, the method, the relevant dates, the results and the person who authorized it 3.
- A report under ISO/IEC 17025:2017 includes a statement that the results relate only to the items tested 3.
- ISO 9001 certification does not specifically evaluate a laboratory's technical competence, while accreditation based on ISO/IEC 17025 is designed to do so 4.
- ICH Q6A says identification solely by a single chromatographic retention time is not regarded as specific 5.
Common questions
Is a "lot" the same as a "batch"?
Yes. ICH Q7 treats the two terms as the same thing 1.
Does a COA prove what is in my vial?
No. A COA reports on the sample that was tested, and the results relate only to that sample 3. It is relevant to your vial only if the lot number matches, and even then it is evidence rather than proof.
Is a third-party report always better than an in-house COA?
It is more independent, and independence matters 4. But a third-party report still depends on which sample was sent, how competent the lab is, and whether the report is complete.
Sources
- Q7 Good Manufacturing Practice Guidance for Active Pharmaceutical Ingredients: Guidance for Industry (ICH Q7), U.S. Food and Drug Administration, September 2016. https://www.fda.gov/media/71518/download — Used for the contents of a COA (section 11.4), checking the reliability of supplier certificates (section 7.31), and the glossary definitions of batch, batch number, acceptance criteria and retest date.
- Model certificate of analysis, Annex 4, WHO Technical Report Series No. 1010, World Health Organization, 2018. https://cdn.who.int/media/docs/default-source/medicines/norms-and-standards/guidelines/quality-control/who-model-certificate-of-analysis.pdf — Used for the description of a COA, the list of items it should include, page identification, certificates issued by repackers or traders, and the reference to accreditation.
- ISO/IEC 17025 Crosswalk – Reporting the Results, NIST Weights and Measures, U.S. National Institute of Standards and Technology, February 2018. https://www.nist.gov/document/7-8-17025-crosswalk-reporting-results-20180201pdf — Used for a public summary of what ISO/IEC 17025:2017 clause 7.8 requires in a report (lab name and address, unique identification, method, dates, results, authorizing person, results relate only to items tested, customer-supplied information, amendments). The standard itself is sold by ISO and was not opened.
- Why use an Accredited Laboratory?, International Laboratory Accreditation Cooperation (ILAC), copy hosted by NIST. https://www.nist.gov/document/ilacwhyuseanal2015pdf — Used for what accreditation assesses, how it differs from ISO 9001 certification, the scope of accreditation, accreditation body directories, and the value of independent evidence.
- ICH Topic Q6A, Specifications: Test Procedures and Acceptance Criteria for New Drug Substances and New Drug Products: Chemical Substances (CPMP/ICH/367/96), European Medicines Agency, May 2000. https://www.ema.europa.eu/en/documents/scientific-guideline/ich-q-6-test-procedures-and-acceptance-criteria-new-drug-substances-and-new-drug-products-chemical-substances-step-5_en.pdf — Used for the statement that a single chromatographic retention time is not a specific identity test.
- McCarthy D, Han Y, Carrick K, et al. "Reference Standards to Support Quality of Synthetic Peptide Therapeutics." Pharmaceutical Research 40, 1317–1328 (2023), Springer. https://pmc.ncbi.nlm.nih.gov/articles/PMC10338602/ — Used for the grouping of peptide tests into identity, purity and strength.