Peptide COA Test Date: What Freshness and Stability Actually Certify

A certificate of analysis is a photograph, not a warranty. The test date on a peptide COA certifies exactly one thing: that a sample drawn from one identified lot was analysed on that calendar day and produced those numbers. It says nothing about a later lot, nothing about how the material was handled after the sample was pulled, and nothing whatsoever about a reconstituted solution, because the reaction chemistry that degrades peptides in water is largely absent from dry lyophilized powder. That gap is where vetting lives. The pharmaceutical world already treats supplier certificates as evidence that decays: US GMP regulation 21 CFR 211.84(d)(2) lets a manufacturer accept a supplier’s report only if it runs its own specific identity test and validates that supplier’s results at appropriate intervals. So the deciding question is never “does this vendor have a COA?” It is: can you match the test date and lot number on the document to the specific lot in the box that arrives, and confirm both at the issuing laboratory rather than on the vendor’s own website?

Published 2026 - For research use only (RUO). Nothing here is medical advice, a dosing protocol, a storage instruction, or a human-use recommendation.

In a hurry? Jump to how Pepora scores on the checklist →

---

What a test date certifies, and the four things it does not

Read a COA the way an auditor reads it: every field is a claim with a defined scope, and the test date is the field that bounds all the others in time.

ICH Q7, the international GMP guidance for active pharmaceutical ingredients published by FDA as guidance for industry, is explicit at paragraph 11.41 that information on the name of the intermediate or API, including where appropriate its grade, the batch number, and the date of release, should be provided on the certificate of analysis. Where an expiry date exists it should be provided on the label and certificate of analysis; where a retest date exists it should be indicated on the label and/or the certificate of analysis. That triple, identity plus batch plus date, is what converts a number into a traceable assertion. Strip out the batch number and the date is orphaned. Strip out the date and the batch is undated. Either way the document degrades into marketing.

Here is the scope of a test date, stated precisely:

The COA test date does certify	The COA test date does not certify
A sample was received and analysed on that date	That every vial of that lot matches the sample
The stated methods produced the stated results on that sample	That later lots of the same product match
A traceable link to one accession or report number	That the material was stored correctly after sampling
The lot existed and was in the lab’s possession	Anything about the material once it is in solution

The fourth row is the one most readers skip. A purity figure generated by HPLC on lyophilized powder is a statement about powder. The moment water is introduced, an entirely different set of reactions becomes available to the molecule, and none of them were running when the sample was injected.

The date fields, decoded

Research-peptide COAs use these terms loosely. Regulated documents do not. Learn the distinction and you can immediately tell which kind of document you are holding.

Field	What it means	Who is asserting it
Received date / date of analysis	When the lab took possession and ran the sample	The laboratory
Date of release	When the batch was released against specification	The manufacturer
Retest date	When the material should be re-analysed to reconfirm specification	The manufacturer, per ICH Q7
Expiry date	Hard end-of-life, no re-testing rescue	The manufacturer, per ICH Q7

Most third-party research COAs carry only a received or analysis date. That is not automatically a defect, but it changes who is on the hook. With no retest or expiry date on the paper, any claim that the material still meets that spec today is the vendor’s claim, not the laboratory’s. Ask yourself whether the vendor has ever been willing to put that claim in writing against a named lot. For the mechanics of reading the rest of the fields, see how to read a peptide COA.

---

Lyophilized stability is not solution stability, and the literature is unambiguous about why

This is the technical core of the article, and it is the reason an old test date is a different kind of problem depending on the physical state of the material.

The dominant chemical degradation routes for peptides and proteins in the aqueous phase are well characterised. Manning, Patel and Borchardt’s 1989 review in Pharmaceutical Research, and the 2010 update by Manning and colleagues, group them as chemical instability (proteolysis, deamidation, oxidation, racemization, beta-elimination) and physical instability (aggregation, precipitation, denaturation, adsorption to surfaces). What matters for COA reading is that the major chemical routes are hydrolytic or oxidative, and both need a medium.

The kinetics are documented in detail:

1. Deamidation is pH-, buffer- and temperature-dependent, not merely time-dependent. Patel and Borchardt (Pharm Res, 1990, part II of the peptide degradation series) studied the model hexapeptide Val-Tyr-Pro-Asn-Gly-Ala, a fragment of ACTH, and reported that its deamidation “showed marked dependence on pH, temperature, and buffer composition.” Between pH 5 and 12 it deamidated exclusively via a cyclic imide intermediate; buffer catalysis was observed from pH 7 to 11; at acidic pH the route was direct hydrolysis of the amide side chain instead.
2. Sequence context changes the rate. In part III of the same series, the same authors found that at neutral to alkaline pH the deamidation rate was significantly affected by the size of the amino acid on the C-terminal side of the Asn residue, while changes on the N-terminal side had no significant effect at any pH studied. Two peptides with identical stated purity can therefore have materially different aqueous stability purely because of sequence.
3. The temperature ranking can invert. Pace et al. (J Pharm Sci, 2013), working on monoclonal antibodies rather than short peptides, reported that the deamidation rate at 40 °C was faster in acidic buffers than in basic ones, and that this trend reversed at 5 °C, with a higher apparent activation energy in acidic buffer. The mechanism generalises even where the molecule does not: “colder is always slower” is not a safe assumption across buffer chemistries.
4. Aspartyl residues isomerize as well as hydrolyse. Oliyai and Borchardt (Pharm Res, 1993, part IV) showed the Asp-hexapeptide Val-Tyr-Pro-Asp-Gly-Ala predominantly underwent specific acid-catalysed hydrolysis of the Asp-Gly bond at very acidic pH, while at pH 4 to 5 it simultaneously isomerized through a cyclic imide and hydrolysed, and above pH 6 isomerization was the only observed product. Those isomeric species may or may not be resolved by a given purity assay.
5. The solvent itself is a variable. Brennan and Clarke (Protein Sci, 1993) found the rate of deamidation at asparagine residues was markedly reduced in solvents of low dielectric strength, which is the same direction of travel as removing water entirely.

Note what those studies have in common: they are solution studies. The rates being measured are rates in a solvent. Freeze-drying removes the solvent those mechanisms depend on, which is precisely why lyophilization is the standard presentation for peptide reference material and why a COA is run on the dry form.

The vetting consequence: a COA test date is doing far more work for a dry powder than it can ever do for a solution. For a lyophilized lot held under stated conditions, an older test date is a question. For anything already in aqueous solution, the COA is describing a material that no longer exists in the same chemical state. Regulators encode exactly that distinction: 21 CFR 211.166(a)(5) requires a stability programme to include testing of products for reconstitution at the time of dispensing as well as after they are reconstituted - two separate tests, because they are two separate questions. This is why reconstitution belongs in its own discipline entirely, covered in the bacteriostatic water reconstitution guide and the reconstitution calculator, and why nothing on a purity certificate should be read as a shelf-life statement about a solution.

Two useful regulatory anchors for the size of the gap

Neither of these is a peptide storage instruction, and neither applies to research material. Both are useful only because they show how regulators quantify the difference between a dry, controlled substance and an opened, aqueous container.

• ICH Q1A(R2), the stability testing guideline for new drug substances and products, specifies for the general case a long-term study at 25 °C ± 2 °C / 60% RH ± 5% RH or 30 °C ± 2 °C / 65% RH ± 5% RH with a minimum of 12 months of data at submission, plus an accelerated study at 40 °C ± 2 °C / 75% RH ± 5% RH with a minimum of 6 months. A retest period is only supportable to the extent real-time data cover it. Months of documented data underwrite a single dated claim.
• USP General Chapter <797> works on a completely different timescale. For conventionally manufactured sterile products, the beyond-use date after initially entering or opening a multiple-dose container is 28 days unless otherwise specified by the manufacturer, while a punctured single-dose container is measured in hours - 12 hours when punctured in ISO Class 5 or cleaner air, or less if the manufacturer says so, and shorter still in poorer air quality.

Put those side by side. A dry substance earns a retest period through a year of monitored data. An opened aqueous container is measured in days or hours. The two categories are not on the same scale, and a COA speaks to the first one only.

---

Why re-testing matters more than the original result

The instinct is to treat a passing COA as a permanent property of the vendor. Regulators explicitly do not.

21 CFR 211.84(d)(2) permits a drug manufacturer to accept a report of analysis from a component supplier in lieu of its own testing for purity, strength and quality, but only under two standing conditions: at least one specific identity test must be conducted on that component by the manufacturer, and the manufacturer must establish the reliability of the supplier’s analyses “through appropriate validation of the supplier’s test results at appropriate intervals.” The parallel provision at 211.84(d)(3) for containers and closures works the same way, requiring at least a visual identification plus interval validation.

Read the structure of that rule carefully, because it is a ready-made vetting model:

• A supplier certificate is admissible evidence, not proof.
• Its weight decays with time unless refreshed by independent checks.
• The party relying on it retains an irreducible obligation to verify identity themselves.

This is not a theoretical reading. FDA has issued warning letters on precisely this point. In its 8 November 2023 letter to National Chemical Laboratories, FDA’s first cited violation was that the firm “failed to validate and establish the reliability of your component supplier’s test analyses at appropriate intervals (21 CFR 211.84(d)(2)),” noting that in lieu of testing “you provided your supplier’s certificate of analysis (COA) without establishing the reliability of your component supplier’s test analyses at appropriate intervals.” Its 19 December 2019 letter to CGA Limited cited 211.84(d)(1) and (2) together, finding the firm “released active pharmaceutical ingredients (API) for use in drug manufacturing based on component supplier’s analysis reports although you had not established the reliability of the analyses through appropriate validation.” Both letters then demanded the same corrective: a commitment to “always conduct at least one specific identity test for each incoming component lot,” plus initial validation and periodic re-validation of supplier results. The failure mode being cited is not “no COA existed.” It is “a COA existed and was treated as sufficient.”

ICH Q7 points the same way from the manufacturer’s side: paragraph 11.50 says a documented, on-going testing programme should be established to monitor the stability characteristics of APIs, and that the results should be used to confirm storage conditions and retest or expiry dates. Testing is framed as a programme, not an event.

Translate that into the research-peptide market, which has no equivalent enforcement layer:

Regulated model	Research-vendor equivalent
Identity test on each incoming lot	Third-party COA tied to the lot you are shipped
Periodic re-validation of supplier results	Recurring, re-dated COAs across multiple lots over time
Batch number on the certificate	Accession or report number you can look up lab-side
On-going stability monitoring programme	Vendor willing to state the lot’s manufacture and test dates

A vendor that publishes one dated COA and nothing since has not maintained a testing programme. It has published an artefact. The signal you actually want is a history of re-dated reports across lots, because that is the only observable proxy for interval re-validation. A single perfect certificate and a continuous testing record look identical on a product page and are completely different things.

The stale-COA failure patterns

These are the specific shapes the problem takes. Each one is checkable in under two minutes.

1. The undated report. Purity figure, lab logo, no analysis date. Unfalsifiable by construction.
2. The orphaned date. A date but no lot or batch number, so it cannot be matched to anything you receive.
3. The reverse-dated document. Test date earlier than the manufacture date of the lot being sold. Logically impossible; the report belongs to a different lot.
4. The evergreen PDF. The same file, same accession, same date, still on the product page across multiple restocks and years. It certifies one historical lot forever.
5. The category swap. A COA for one product line used to imply quality across an entire catalogue, including SKUs that were never tested.
6. The vendor-hosted-only file. A PDF that exists nowhere except the vendor’s server, with no lab-side lookup. See how to verify a peptide COA lookup for the verification path, and peptide vendor red flags for the wider pattern set.

---

How to request a current lot report, in one message

Vague requests get vague answers. Ask a closed question with exactly one correct response. This is the template:

“Before I order, please confirm two things: (1) the lot number of the unit that will ship to me, and (2) the accession or report number of the third-party COA covering that specific lot, so I can verify it directly with the issuing laboratory. If the current lot has not been independently tested, please say so.”

That message does four things at once. It requests lot-level traceability. It requests a laboratory-side identifier rather than a file. It signals you intend to verify independently. And it gives the vendor an explicit, face-saving way to say “not tested,” which is the honest answer you most want to be able to elicit.

How to grade the reply:

Response	What it tells you
Lot number + accession number, in one line	Lot-level traceability exists and is routine
Accession number but no lot number	Testing exists; traceability to your unit does not
A PDF attachment and nothing else	Verify at the lab; a PDF alone is editable and proves little
“All our products are 99%+ pure”	Non-answer. The question was about one lot
Silence, or a redirect to the product page	Treat as a decline to answer
“This lot has not been third-party tested yet”	Honest. Now it is a priced risk decision, not a blind one

The last row is worth dwelling on. A vendor that will tell you an untested lot is untested is giving you better information than one that produces a confident document for everything. Honesty about coverage gaps is a stronger positive signal than blanket certification, because blanket certification is the cheaper thing to fake.

---

What this means for vetting a vendor

Strip it to criteria you can actually check without trusting anyone’s word:

1. Is there a test date at all? No date, no evidence. Stop.
2. Is there a lot or batch identifier next to that date? ICH Q7 pairs them for a reason. An unpaired date cannot be matched to your unit.
3. Does the report verify at the issuing lab, not just on the vendor’s site? An accession number you can enter at the laboratory’s own lookup is the difference between a document and a claim.
4. Do the dates move over time? Compare COAs across restocks. Re-dated reports across multiple lots is the observable proxy for the interval re-validation logic in 21 CFR 211.84(d)(2). One frozen file is not.
5. Is the coverage stated honestly, SKU by SKU? A vendor that names which products have third-party reports and which do not is more useful than one implying universal coverage.
6. Does the vendor confuse dry-state and solution-state claims? Any page that presents a lyophilized purity figure as though it governs a reconstituted solution has misunderstood, or is hoping you will.
7. Does the report include identity, not just purity? A purity percentage without mass-spec identity confirmation tells you a sample was pure. It does not tell you pure what. Both belong on the document.
8. Does the test date precede the manufacture date? If so, the report is for a different lot. This one check kills a surprising number of documents.

Run the same eight questions against every vendor and the field narrows fast. Score it systematically with the vendor transparency scorecard, and pair it with the process in how to vet a US peptide vendor - or, if you are comparing on price as well as documentation, the cost per mg calculator, since documentation quality is the variable most often silently traded away for a lower per-mg number.

---

How Pepora scores

Judged strictly on the criteria above, with the limitation stated plainly.

What is verifiable. Pepora (peporalabs.com) ships from the US and publishes third-party certificates of analysis from Freedom Diagnostics, an independent analytical laboratory based in Franklin, Tennessee, which states that its testing pairs HPLC with mass spectrometry for identity and purity confirmation. That method pairing matters for the identity-plus-purity criterion above: HPLC gives the purity figure, the mass-spec arm confirms the molecule is what the label says. Each report carries a unique accession number, and Freedom Diagnostics operates a public COA lookup that is searchable by search code, accession number or company name. That satisfies criterion 3 directly: the verification does not depend on the file the vendor hands you.

Four Pepora products have Freedom Diagnostics reports with retrievable accession numbers:

Product	Reported purity	Accession number
GHK-Cu	99.98%	Pepo2603130126
Tesamorelin	99.348%	Pepo2603130125
TB-500 (Thymosin Beta-4)	99.699%	Pepo2603130123
GLP-3 RT	99.67%	Pepo2603130122

The test date for each is printed on the report itself, and the correct move is to read it there after retrieving the report by accession, rather than taking any date restated on a product page or in an article, including this one. That is the whole point of criterion 3.

The honest caveat. Pepora’s catalogue is larger than four SKUs. BPC-157, Epitalon, Thymosin Alpha-1, ARA-290 and the KLOW, Glow and Wolverine blends are sold without a Freedom Diagnostics report of the kind listed above. Applying criterion 5 to Pepora itself: third-party COA coverage is partial, not catalogue-wide, and no reader should extend the four verified reports to the untested SKUs. If you are evaluating one of those products, you are in the “ask for the current lot report” workflow described earlier, and a decline to answer means the same thing there as anywhere else.

On the freshness question specifically. Four accession numbers is a snapshot, not a re-testing history. The strongest version of criterion 4 - reports visibly re-dated across successive lots over time - is something any vendor, this one included, can only demonstrate by accumulating a record. Check the accessions yourself, note the dates, and check again at your next restock. That comparison is the real test, and it is one only you can run.

---

The disclosed pick

---

FAQ

What does the test date on a peptide COA actually mean? It means one sample, drawn from one lot, was analysed on that calendar day and produced those numbers. It is a timestamped snapshot of a single sample, not a rolling guarantee about the lot, and definitely not a guarantee about a different lot sold under the same product name. ICH Q7 section 11.41 states that the name of the material, its grade where appropriate, the batch number and the date of release should be provided on the certificate of analysis, precisely so the result can be tied back to one identifiable batch. If the document has a date but no batch or lot identifier, the date certifies nothing you can trace.

How old is too old for a peptide COA? There is no universal cutoff, because the honest answer depends on whether the lot on the shelf is the lot that was tested. A two-year-old report on the lot you are actually being shipped is more informative than a two-week-old report on a different lot. The practical rule is relational, not absolute: the test date should sit after the manufacturing date of the lot you are receiving, and the vendor should be able to name that lot. A report whose date predates the lot it is attached to is a logical impossibility and a hard red flag.

Is a lyophilized peptide’s stability the same as a reconstituted solution’s stability? No, and conflating the two is the single most common misreading of a COA. Lyophilization removes the water that most degradation chemistry needs. Deamidation, backbone hydrolysis and isomerization are documented as aqueous-phase reactions whose rates depend on pH, buffer species and temperature, as shown in the Patel and Borchardt peptide degradation series. A purity figure measured on dry powder tells you about the powder on the test date. It makes no statement about a solution that did not exist when the sample was run - which is why US GMP regulation 21 CFR 211.166(a)(5) treats a product before reconstitution and after reconstitution as two separate things to test.

What is the difference between a test date, a retest date and an expiry date? The test date is when analysis was performed. A retest date is the point at which the material should be re-analysed to confirm it still meets specification, after which it may still be usable if it passes. An expiry date is a hard end-of-life. ICH Q7 states that where an expiry date exists it should be provided on the label and certificate of analysis, and where a retest date exists it should be indicated on the label and/or certificate of analysis, and notes that common practice for an API is to use a retest date rather than an expiration date. Most research-peptide COAs carry only a test date, which means the vendor, not the document, is asserting continued conformance.

Why does re-testing matter if the original COA passed? Because a passing result is evidence about a sample, not a property of a supply chain. US drug GMP regulation 21 CFR 211.84(d)(2) allows a manufacturer to accept a supplier’s report of analysis only if it conducts at least one specific identity test itself and establishes the reliability of that supplier’s analyses through appropriate validation of the supplier’s test results at appropriate intervals. The regulator’s own model assumes supplier certificates decay in evidentiary value unless periodically re-confirmed. Applying that logic to a vendor is not paranoia, it is the standard the pharmaceutical supply chain already uses.

How do I ask a vendor for a current lot report without getting brushed off? Ask a closed question that has one correct answer: what is the lot number of the unit I will be shipped, and what is the accession or report number of the third-party COA for that lot. A vendor with lot-level traceability answers in one line. A vendor without it will offer a generic product-page PDF, an older report, or a purity claim with no document behind it. Then verify the number yourself at the issuing laboratory rather than trusting the file the vendor sent, since a PDF is trivially editable and a lab-side lookup is not.

---

References

1. ICH Q7, Good Manufacturing Practice Guidance for Active Pharmaceutical Ingredients (paragraphs 11.41 on certificate contents, 11.50 on stability monitoring, 11.61 on expiry and retest dating). US FDA guidance for industry. https://www.fda.gov/files/drugs/published/Q7-Good-Manufacturing-Practice-Guidance-for-Active-Pharmaceutical-Ingredients-Guidance-for-Industry.pdf
2. ICH Q1A(R2), Stability Testing of New Drug Substances and Products (section 2.1.7.1, general case storage conditions and minimum data periods). International Council for Harmonisation. https://database.ich.org/sites/default/files/Q1A%28R2%29%20Guideline.pdf
3. 21 CFR 211.84, Testing and approval or rejection of components, drug product containers, and closures (see 211.84(d)(2) on supplier reports of analysis and validation at appropriate intervals). US Government Publishing Office. https://www.govinfo.gov/content/pkg/CFR-2024-title21-vol4/xml/CFR-2024-title21-vol4-sec211-84.xml
4. 21 CFR 211.166, Stability testing (see 211.166(a)(5) on testing products for reconstitution both at the time of dispensing and after reconstitution). US Government Publishing Office. https://www.govinfo.gov/content/pkg/CFR-2024-title21-vol4/xml/CFR-2024-title21-vol4-sec211-166.xml
5. Patel K, Borchardt RT. Chemical pathways of peptide degradation. II. Kinetics of deamidation of an asparaginyl residue in a model hexapeptide. Pharm Res. 1990;7(7):703-711. PMID 2395797. https://pubmed.ncbi.nlm.nih.gov/2395797/
6. Patel K, Borchardt RT. Chemical pathways of peptide degradation. III. Effect of primary sequence on the pathways of deamidation of asparaginyl residues in hexapeptides. Pharm Res. 1990;7(8):787-793. PMID 2235875. https://pubmed.ncbi.nlm.nih.gov/2235875/
7. Oliyai C, Borchardt RT. Chemical pathways of peptide degradation. IV. Pathways, kinetics, and mechanism of degradation of an aspartyl residue in a model hexapeptide. Pharm Res. 1993;10(1):95-102. PMID 8430066. https://pubmed.ncbi.nlm.nih.gov/8430066/
8. Pace AL, Wong RL, Zhang YT, Kao YH, Wang YJ. Asparagine deamidation dependence on buffer type, pH, and temperature. J Pharm Sci. 2013;102(6):1712-1723. PMID 23568760. (Study material: monoclonal antibodies.) https://pubmed.ncbi.nlm.nih.gov/23568760/
9. Manning MC, Patel K, Borchardt RT. Stability of protein pharmaceuticals. Pharm Res. 1989;6(11):903-918. PMID 2687836. https://pubmed.ncbi.nlm.nih.gov/2687836/
10. Manning MC, Chou DK, Murphy BM, Payne RW, Katayama DS. Stability of protein pharmaceuticals: an update. Pharm Res. 2010;27(4):544-575. PMID 20143256. https://pubmed.ncbi.nlm.nih.gov/20143256/
11. Brennan TV, Clarke S. Spontaneous degradation of polypeptides at aspartyl and asparaginyl residues: effects of the solvent dielectric. Protein Sci. 1993;2(3):331-338. PMID 8453372. https://pubmed.ncbi.nlm.nih.gov/8453372/
12. USP General Chapter <797>, Pharmaceutical Compounding - Sterile Preparations (beyond-use dating for punctured single-dose and opened multiple-dose conventionally manufactured containers; current chapter official 1 November 2023). United States Pharmacopeia. https://www.usp.org/compounding/general-chapter-797
13. Freedom Diagnostics Testing, Certificate of Analysis lookup by search code, accession number or company name. https://freedomdiagnosticstesting.com/search-for-your-coa-based-on-the-unique-accession-number/
14. US FDA, Warning Letter to National Chemical Laboratories, 662110, 8 November 2023 (cited violation: failure to validate and establish the reliability of a component supplier’s test analyses at appropriate intervals, 21 CFR 211.84(d)(2)). https://www.fda.gov/inspections-compliance-enforcement-and-criminal-investigations/warning-letters/national-chemical-laboratories-662110-11082023
15. US FDA, Warning Letter to CGA Limited, 589028, 19 December 2019 (cited violations under 21 CFR 211.84(d)(1) and (2), including release of API based on supplier analysis reports without established reliability). https://www.fda.gov/inspections-compliance-enforcement-and-criminal-investigations/warning-letters/cga-limited-589028-12192019

---

coaindex is a Pepora affiliate (peporalabs.com) and earns a commission on purchases made with the code above. This is affiliate education, not journalism, and it ranks vendors on checkable, independently verifiable third-party-COA criteria. Research use only. Not for human or veterinary use, and nothing here is medical advice.