How to Read a Peptide Certificate of Analysis

What the document is

A certificate of analysis, usually abbreviated COA, is a summary report tying analytical results to a defined batch of material. It is not a certification in the regulatory sense and it is not issued by an authority. It is a statement by whoever performed or commissioned the testing that the named tests were run on the named lot and produced the stated results.

That framing sets the standard for reading one. The value of a COA rests entirely on traceability — whether the results on the page can be tied to the vial in your hand — and on transparency about method. A well-constructed COA makes both easy to check. A weak one presents conclusions without the information needed to evaluate them.

Identification fields

The header should name the product, the lot or batch code, the fill size, and the date the material was released. The lot code is the single most important field on the document, because it is the link between the report and the physical container. It should match the code printed or applied to the vial label exactly, character for character.

A report that names only the product and not a lot describes a category, not a batch. It may be an accurate description of some material the supplier once tested, but it cannot be checked against a vial. The same applies to a report where the lot field is present but blank, or where the lot code on the paperwork and on the label differ in any character.

• Product name and any alternate designation used on the label.
• Lot or batch code, matching the vial.
• Fill size and physical form, for example lyophilized powder.
• Date of manufacture or release, and the date of analysis.
• Identity of the laboratory that performed the testing.

The results table

The body of a COA is a table of tests, each with a method, a specification and a result. The specification is the acceptance criterion set before testing; the result is what was measured. Both should be present. A table with results and no specifications gives no indication of what would have constituted a failure, and a table with specifications and no results is not a report at all.

Appearance is usually the first row: a visual description such as white to off-white lyophilized powder. It sounds trivial and is not. A deviation in colour or in cake structure is often the earliest visible sign of a problem with a lyophilization cycle or with storage, and it costs nothing to check against your own vial.

Purity by HPLC follows, and should carry its wavelength and method conditions. Identity by mass spectrometry reports an observed mass against the theoretical mass calculated from the sequence; the two should agree within the instrument's stated tolerance, and the report should give both numbers rather than only the word conforms. Water content by Karl Fischer, peptide content, counter-ion content and bacterial endotoxin each appear when they have been tested, and each has its own method reference.

Signature, date and laboratory

A COA should identify who performed the analysis and who approved the release, with dates. An unsigned report, or one signed only by an unnamed quality department, is weaker evidence than one naming an analyst and an approver.

Where testing was performed by an outside laboratory, the report should name that laboratory. Third-party testing is meaningful specifically because the testing party is identifiable and independent; an unnamed third party is an assertion, not a check. If a supplier states that outside testing was performed, it is reasonable to ask for the laboratory's own report rather than a supplier-prepared summary of it.

Common ambiguities

Several entries recur on peptide certificates and are routinely over-read.

The word conforms in place of a value. Conforms means the analyst judged the result acceptable against a specification you may not be able to see. A numeric result is always more informative.

A purity figure with no wavelength or gradient. As covered in the article on HPLC purity, area-percent is method-dependent, and an unqualified figure cannot be compared to one produced under different conditions.

Sterile as a description. Sterility is a defined test with a defined protocol and incubation period. Endotoxin testing is not a sterility test, and a low endotoxin result does not establish that a preparation is sterile.

Research grade, pharmaceutical grade or similar phrases. These are marketing terms, not defined analytical categories, and they have no fixed specification behind them.

A report date long before the release date, or no date at all. Analytical results describe the material at the moment of testing. The interval between analysis and release, and the storage conditions during it, are part of the picture.

Checking a certificate against a vial

The practical routine takes a minute. Read the lot code from the vial label and confirm it appears on the report. Confirm the product name and fill size match. Compare the appearance description on the report against the cake in the vial. Confirm the report carries a date and a signature, and note which laboratory produced it.

Where a supplier maintains a lot lookup, entering the code should return the record for that lot rather than a general product page. A lookup that returns a result for any string entered is not verifying anything. Meridian's lot tool returns no result unless a signed report for that exact code is on file.

The limits of the document

A certificate describes the tests listed on it and nothing else. Tests not on the report were not performed. The results describe a sample drawn from a batch, which is a statement about the batch to the extent that sampling was representative. And every result is anchored to the date of analysis: it says nothing about temperature excursions in transit, about how long a vial sat at room temperature before it was placed in a freezer, or about degradation after reconstitution.

Read that way — as a dated, scoped, traceable record rather than a guarantee — a certificate of analysis is one of the more useful documents in a laboratory supply chain.