Seven ways a certificate of analysis misleads

Representative certificates, lot numbers that match nothing, in-house tests dressed as independent, purity without identity, stale dates and purity without net content: each pattern, and what a good certificate shows instead.

By Touchstone PeptidesSeptember 22, 20269 minute read

A certificate of analysis misleads when it is genuine but describes different material, hides who ran the test, answers the easy question and skips the hard one, or has aged out of relevance. None of the seven patterns below needs a forged document. Each is paired with what a good certificate shows instead; how to read a certificate of analysis explains what each line measures, and this piece covers the lines that are missing or misplaced.

1. The representative certificate

A "representative", "typical" or "sample" certificate describes a product, not a lot. It may be a real report from some earlier batch, reused for every batch since, or a composite of the figures the supplier expects. Either way it is a brochure with a chromatogram on it.

The regulatory baseline is that a certificate belongs to a batch. ICH Q7, the good-manufacturing-practice guidance for active ingredients, says "Authentic certificates of analysis should be issued for each batch of intermediate or API on request", and that "the batch number, and the date of release should be provided on the certificate of analysis". A certificate with no batch or lot number cannot be authentic in that sense because there is nothing for it to be authentic about.

What a good certificate shows instead: a lot number in the header, a date of analysis, and figures that change from lot to lot. If every certificate a supplier has ever shown carries the same purity to the same decimal, you are reading one document with several dates.

2. A lot number that matches nothing

A lot number is the key that connects the paper to the material. FDA's definition in 21 CFR 210.3 is exact: a lot number is "any distinctive combination of letters, numbers, or symbols, or any combination of them, from which the complete history of the manufacture, processing, packing, holding, and distribution of a batch or lot of drug product or other material can be determined". ICH Q7 puts it more briefly: a batch number is a unique code "from which the production and distribution history can be determined".

The pattern that misleads is a certificate whose lot number appears nowhere else: not on the vial, not on the packing slip, not in any index you can search. A second version is a lot number that appears on the vial but was assigned by a reseller, so it connects to the reseller's stock record and not to the manufacturer's batch. The existing piece on what a lot number is for covers the mechanics; the short version is that the failure usually happens one tier upstream, when material is repackaged and a number is reassigned.

Instead: one string that appears on the vial, the packing slip and the certificate, traceable to the manufacturer's batch, and published somewhere it can be looked up without asking.

3. In-house presented as independent

A manufacturer's release certificate is a normal and necessary document. It is also issued by the party that loses money if the number is wrong. The misleading pattern is a certificate that omits who ran the test, or carries a supplier's branding over data generated by the supplier itself, and is then described as "third-party tested".

ICH Q7 is specific about attribution. A manufacturer's certificate "should be dated and signed by authorized personnel of the quality unit(s) and should show the name, address, and telephone number of the original manufacturer". Where a repacker, agent or broker issues a new certificate, it "should show the name, address and telephone number of the laboratory that performed the analysis" and "contain a reference to the name and address of the original manufacturer and to the original batch certificate, a copy of which should be attached". A certificate silent on who did the work has removed the one fact that decides whether it is independent.

FDA's own manufacturing rules show the standard for relying on someone else's certificate, at 21 CFR 211.84(d)(2):

In lieu of such testing by the manufacturer, a report of analysis may be accepted from the supplier of a component, provided that at least one specific identity test is conducted on such component by the manufacturer, and provided that the manufacturer establishes the reliability of the supplier's analyses through appropriate validation of the supplier's test results at appropriate intervals.21 CFR 211.84(d)(2)

The principle transfers to a buyer of a research vial: an in-house certificate is accepted on the strength of periodic independent checks, not on its own.

The independent version: the laboratory's name and address on the report itself; where the supplier commissioned an independent test, a second document under the same lot number from a named laboratory, and a statement of who drew the sample. How to check the named laboratory is the subject of how to verify a peptide lab report, and the reasoning behind a second test is in why we pay for a second laboratory.

4. Purity without identity

An HPLC purity figure answers one question: what fraction of the detector signal fell under the main peak. It does not say what the main peak is. A pure sample of the wrong peptide, a truncated sequence, or a near-relative with one residue substituted can each give a clean chromatogram.

Identity is a separate measurement. Mass spectrometry compares the observed mass of the main component with the mass calculated from the sequence; ICH Q2(R2) describes this kind of confirmation as an orthogonal procedure comparison, where the figure from one method is checked against "a second, well-characterized analytical procedure that ideally applies a different measurement principle". Q2(R2) also observes that when "a single procedure is not considered specific or sufficiently selective, an additional procedure should be used to ensure adequate discrimination". HPLC alone is that single procedure. FDA's rule for manufacturers, 21 CFR 211.84(d)(1), states the minimum: "At least one test shall be conducted to verify the identity of each component".

What to expect instead: an HPLC purity and a mass-spectrometry identity on the same page, with the observed and calculated masses both printed so the reader can see the match rather than take it. For a BPC-157 vial, that means an observed mass sitting next to the calculated 1419.5.

5. Stale dates, and what an expiry on a lyophilized peptide means

Three dates matter on a certificate: manufacture, analysis, and either an expiry or a retest date. The misleading patterns are a certificate with no date of analysis at all, an analysis older than the material's own retest date, or a "manufacture" date that has been reset to the date a reseller received the stock.

The two forward-looking dates are not the same thing, and Q7 defines both. An expiry date is "the date placed on the container/labels of an API designating the time during which the API is expected to remain within established shelf life specifications if stored under defined conditions and after which it should not be used". A retest date is "the date when a material should be re-examined to ensure that it is still suitable for use". Q7 adds that for an active ingredient, "Common practice is to use a retest date, not an expiration date", and that either "should be based on an evaluation of data derived from stability studies".

For a dry, sealed, lyophilized peptide the more informative date is therefore the retest date: it is a re-examination date set from stability data, and after it the honest position is "not known without a new test", not "known to be degraded". A certificate that prints an expiry with no stability basis, or a retest date that resets every time the material changes hands, has turned a data-driven date into a marketing one. The chemistry behind the dates is covered in storage and stability.

The honest dates: manufacture date, analysis date and retest date in that order, all earlier than today, with the retest date carried through from the manufacturer's certificate rather than assigned by the reseller.

6. Purity without net content

A certificate that prints purity and stops has answered the easy question. Purity is a ratio within a chromatogram; it says nothing about how much of the solid in the vial is peptide. Bachem's technical FAQ defines the missing number: "The net peptide content (NPC) is the fraction of peptidic material (i.e. the requested peptide and the peptidic impurities) relative to counter-ions and residual water." The remainder is trifluoroacetate or acetate and water, and it can be a substantial fraction of the mass. A certificate without net peptide content, water content and the counter-ion leaves the arithmetic to you; the consequences are worked through in net peptide content.

Instead: net peptide content with its method, water by Karl Fischer, and the counter-ion named.

7. One batch number on different fills

A certificate can also hide quantity by putting one batch number on vials of different label mass, a 5 mg and a 10 mg, say, each described by the same certificate. The bulk peptide may indeed be one batch. The fills are not. FDA's definition allows a lot to be "a specific identified portion of a batch, having uniform character and quality within specified limits", and a fill of a given mass is such a portion, with its own filling record and its own fill checks. A certificate for the bulk material is silent on whether either fill hit its target, and a certificate that quotes a fill mass without saying which fill it measured is quoting a number that applies to at most one of the vials. Fill weight and label accuracy covers what the label mass does and does not promise.

For blends the problem compounds: a single lot number now covers several syntheses, and a single purity figure for a mixture is not a number at all, which is why what a blend certificate should say asks for each component reported separately, as for the CJC-1295 with ipamorelin vial, a two-peptide co-lyophilized blend in the catalog.

Instead: either a lot number per fill or a fill statement that names which fill it measured.

What a good certificate shows, in one table

Pattern	The tell	What a good certificate shows instead
Representative certificate	No lot number; identical figures on every "lot"	Lot number, analysis date, figures that vary by lot
Lot number matches nothing	Not on the vial, packing slip or any index	One string on all three, traceable to the manufacturer's batch
In-house as independent	No laboratory name or address	Named laboratory on the report; a second report under the same lot number
Purity without identity	HPLC only	HPLC purity and mass-spectrometry identity, observed and calculated masses printed
Stale or reset dates	No analysis date; retest date newer than the manufacturer's	Manufacture, analysis and retest dates in order, carried from the manufacturer
Purity without net content	No NPC, water or counter-ion	NPC with its method, water, counter-ion named
One batch number on different fills	One certificate over several label masses	A lot number per fill, or a fill statement per label mass

None of these patterns is what FDA cited in its 2024 to 2026 letters to research-peptide sellers; that evidence came from the websites, as what FDA reads before calling a research product a drug sets out.

Sources

For laboratory research use only. Not a drug, not a supplement, and nothing here is a claim about what any of this material does in a person or an animal.

Two things before you come in.