Expiry Date vs Retest Date: What Dating on Research Material Means

An expiration date and a retest date answer different questions, and the difference is not cosmetic. An expiration date says the material must not be used after that day. A retest date says the material should be examined again on that day — and if it still complies with its specification, it can continue in use, with a fresh date assigned. One is a terminal boundary; the other is a checkpoint that can be passed repeatedly. Active substances and bulk research material are far more often dated the second way ^[1][3].

That changes what a label is telling you, and what your own records have to carry. A container marked with a retest date makes no open-ended claim about the powder inside. It makes a claim about a batch, in a particular closure, under a particular storage condition — a claim that matures into an obligation to test rather than an obligation to discard. What follows is where that convention comes from, what evidence it rests on, and what a receiving record should capture so the date still means something six months into your own inventory.

What is the difference between a retest date and an expiration date?

Both terms are defined, and the definitions are short. ICH Q1A(R2) defines a re-test date as the date after which samples of the substance should be examined to confirm the material is still in compliance with the specification and thus suitable for use, and an expiration date as the date after which a batch must not be used ^[1]. ICH Q7 puts the first even more plainly: the date when a material should be re-examined to ensure that it is still suitable for use ^[3]. Note what neither says. Neither claims the material is fine on one day and ruined on the next; both are administrative boundaries derived from stability data, not physical events.

The renewable part is what people most often get wrong. ICH Q1A(R2) is explicit that a batch of drug substance can be re-tested multiple times, with a different portion used after each re-test, for as long as it continues to comply with the specification ^[1]. A retest date is not a softer expiry date. It is a different instrument: it moves the decision from the calendar to the analysis, and moves the cost and the record-keeping onto whoever holds the material.

Why research material usually carries a retest date

ICH Q7, the GMP guide for active pharmaceutical ingredients, settles the convention in one sentence. Section 11.6 states that an API expiry or retest date should be based on an evaluation of data derived from stability studies, and that common practice is to use a retest date, not an expiration date ^[3]. The logic: a bulk substance in a protective closure changes slowly and measurably, and it is an input to be verified rather than a finished article to be handed over. If the question is whether the batch still meets its specification, that question has an answer available — test it.

The same guide anticipates the material travelling: when an intermediate is transferred outside the manufacturer's material management system and a date is assigned, supporting stability information — published data or test results — should be available ^[3]. That is the clause a buyer is leaning on, and knowing it exists makes asking for the basis of a stated date a normal question rather than an awkward one.

There is an honest caveat, written into the same guideline that defines the retest period. ICH Q1A(R2) notes that for most biotechnological and biological substances known to be labile, it is more appropriate to establish a shelf life than a re-test period ^[1]. Synthetic peptides sit across that line: chemically defined and usually supplied as a lyophilised solid, which puts them nearer a small-molecule substance than a protein therapeutic, yet with degradation chemistry that is specifically peptide chemistry. Reviews of solid-state behaviour describe deamidation, peptide-bond hydrolysis, oxidation, the Maillard reaction, beta-elimination and aggregation as the operative routes, with residual moisture, temperature, excipients and the amorphous or crystalline state of the solid as the controlling variables ^[4]. A retest date on peptide material is defensible, not automatic — and on its own it says nothing about which of those degradation pathways the stability study actually watched.

How is a retest date established from stability data?

The date is the output of a defined study, not an estimate. ICH Q1A(R2) asks for at least three primary batches, with long-term testing covering a minimum of twelve months at submission and continuing far enough to cover the period proposed. For the general case the long-term condition is 25 °C ± 2 °C at 60% RH ± 5% RH, or 30 °C ± 2 °C at 65% RH ± 5% RH; the accelerated condition is 40 °C ± 2 °C at 75% RH ± 5% RH for six months. Long-term testing runs every three months through the first year, every six months through the second, annually thereafter. For a drug substance the definition of significant change is blunt: failure to meet its specification ^[1].

Waiting for real-time data to cover the whole period would take as long as the period itself, so ICH Q1E governs how far a declared retest period may run beyond the data supporting it ^[2]. Three things set the allowance: whether significant change appeared at the accelerated condition, how much change and variability the long-term data show, and whether a statistical analysis was actually performed.

Read that table backwards and it becomes practical literacy. A twenty-four-month retest period on a room-temperature substance is commonly twelve months of real-time observation plus the maximum permitted extrapolation ^[2]. That is not a defect — it is the published rule, correctly applied — but the back half of the period is a projection supported by accelerated data, not something anyone watched happen. It also explains why a change of packaging voids the arithmetic: ICH Q7 requires fresh stability studies to justify an assigned expiration or retest date whenever material is repackaged into a different type of container from the one the manufacturer used ^[3].

What a retest date does not tell you

• Nothing about one container's journey. The date describes a batch held under a defined condition; it cannot describe an unrecorded temperature excursion in transit, because nobody measured that.
• Nothing outside the original closure. A date assigned to material in the manufacturer's packaging does not follow it into a different container type without new data ^[3].
• Nothing about attributes that were never in the specification. Compliance means compliance with the tests written down — a retest is a check against a document, not a general assurance.
• Nothing about the sample nobody took. ICH Q7 asks for a representative sample when a retest is performed ^[3], and a scoop from the top of a jar is not one.
• Nothing about a container that has been opened. Regulators treat the sealed container and the broached one as separate questions: EU guidance defines in-use stability testing as establishing, on its own data, how long a multidose container may be used once opened ^[5].

One limit deserves its own line, because it is the one that catches people: a date is not the only trigger for retesting. US cGMP requires components and closures to be retested or re-examined as necessary — for example after storage for long periods, or after exposure to air, heat or other conditions that might adversely affect them ^[6]. ICH Q7 says the same from the other direction, asking that materials be re-evaluated as appropriate to determine their suitability for use, with prolonged storage or exposure to heat or humidity as the examples ^[3]. An event and a date are independent triggers for the same action, and only one of them is printed on the label.

What should be recorded at receipt and when a container is opened?

The manufacturer's date is a statement about a sealed container held under stated conditions. Your records are the only thing connecting that statement to the object in your store. Two capture points do most of the work: arrival, and first opening.

1. The batch or lot number exactly as printed, including leading zeros and any suffix.
2. Which kind of date the label carries — expiry or retest — transcribed verbatim, not flattened into whichever field your log happens to have.
3. The stated storage condition, in the supplier's own words. A date without its condition is half a statement.
4. The date of manufacture and the date of release where the certificate gives them; they are not the same, and the declared period runs from whichever the manufacturer says it does.
5. The certificate of analysis identifier, and whether certificate and label agree. A disagreement is a finding in itself.
6. Condition on arrival: packaging integrity, the state of any temperature indicator or coolant, and the date and time received.

1. The date the container was first opened, and by whom.
2. Why it was opened and what was removed, so the remaining quantity stays reconstructable from the record alone.
3. The environment it was opened into, where that is controlled — the exposures driving solid-state degradation are largely moisture and air ^[4].
4. Where the remainder went, in what closure, and under what condition.
5. A local date for the opened container, derived from your own written rule and kept distinct from the manufacturer's sealed-container date.
6. For any subdivision into a second container, a new label carrying the material name or item code, the control number, the quantity, and the re-evaluation or retest date — the four items ICH Q7 asks for ^[3].

How the dating appears on a certificate of analysis

ICH Q7 sets a floor for what a certificate carries: the name of the intermediate or API, its grade where relevant, the batch number, and the date of release. That is a floor rather than a full account of the document — a field-by-field reading of a certificate of analysis sets out the identity, purity and method entries that sit around the dating. An assigned expiry date should appear on both the label and the certificate; an assigned retest date should appear on the label and/or the certificate ^[3]. That and/or causes avoidable confusion. A retest date may legitimately be printed on only one of the two, so a certificate without one is not automatically deficient, and neither is a label. What is not safe is assuming which document holds it and recording nothing.

Where the evidence thins out, better to say so than paper over it. Batch-specific published stability data for research peptides is sparse. Much of the dating that reaches a buyer is inherited from an upstream manufacturer's declaration rather than generated by whoever ships the container, and the underlying study — its batches, conditions and analytical methods — is rarely visible downstream. The general degradation chemistry is well described ^[4] and the framework for converting stability data into a date is public and specific ^[1][2], but the link between the two for an individual container usually is not. A supplier who cannot say which kind of date is printed, what condition it assumes, or where it came from has told you something useful about their documentation, and that answer belongs in the receiving record too.