Expiry Date vs Retest Date

Expiry Date vs Retest Date is the distinction between a labelled end-of-use date and a scheduled quality re-evaluation date before further use. This is an operational summary, not a quotation or a universal legal definition. In this article, “expiry” and “expiration” are alternative terms. The practical question is whether a date marks a use limit or triggers reassessment of the material’s suitability.

For drug products, ICH Q1A(R2), Glossary: Expiration date gives a narrower definition: the date on the container label identifies the time before which the batch is expected to remain within its approved shelf life specification under defined storage conditions, and after which it must not be used. This is an ICH guidance definition for drug products, not a rule to apply indiscriminately to every manufacturing material.

The storage qualification matters: the printed date is not an unconditional promise of quality regardless of handling. A temperature excursion raises a separate question about whether the conditions supporting that date were maintained.

A restriction on use is also distinct from physical disposal. Crossing a calendar boundary does not establish that material changes instantaneously, nor does it mean destruction must occur at midnight. Material eligibility, scientific condition and authorised disposition are separate decisions. Likewise, raw-material retest dating signals a reassessment point rather than, by itself, an instruction to discard.

With the opening definitions in mind, distinguish durations from calendar events. This table is practical quality terminology, not a single legal classification applicable across sectors.

For new drug substances and products, ICH Q1A(R2), Glossary: Formal stability studies associates a drug substance with a re-test period and a drug product with shelf life. This guidance distinction helps identify what the stability program supports; it does not make every material’s dating convention interchangeable.

A duration becomes a calendar date only when tied to an explicitly documented reference event. Record both the interval and its origin rather than storing an unexplained calculated date. Receiving or releasing a lot does not, by itself, establish a new stability-supported origin. Keep the decision timestamp in the electronic release record distinct from the dating basis; do not assume manufacture, receipt or release is the universal starting point.

Use this practical classification sequence before selecting a dating rule. A chemical name alone does not establish the material’s role, and a supplier’s generic database field does not establish what its date means.

1. Identify intended use and market. Record what the material will be used to make, its destination jurisdiction and the applicable product framework.
2. Classify its role: finished drug product, API or drug substance, or excipient or other component. For the U.S. drug-manufacturing component branch, consult the current text of 21 CFR 211.87 to check scope.
3. Identify the assigned date type. Determine whether the supplier means expiration, retest or another limit; preserve that meaning in material master data rather than mapping every date into one generic field.
4. Consult the applicable evidence and instructions. Match the lot, specification, storage conditions and supplier documentation to the proposed use before deciding which assessment is appropriate.

For an excipient carrying a supplier retest date, distinguish the supplier’s dating basis from the receiving manufacturer’s suitability decision. Resolve ambiguous terminology against specifications, storage history and supplier documentation—not by automatically borrowing an API procedure. Treat conflicting date descriptions as a receiving discrepancy for documented resolution.

Keep food, supplements, devices and non-pharmaceutical chemicals outside this pharmaceutical rule map. Similar vocabulary does not make the cited pharmaceutical provisions applicable: check each sector’s own product and jurisdictional framework.

Start with the product category and applicable jurisdiction, not the date field on a supplier document. The map below separates binding U.S. drug-product requirements from ICH recommendations for drug substances and APIs. For context, see the broader guides to 21 CFR 211 and ICH Q1A stability guidance.

Binding U.S. drug-product requirements

For covered U.S. drug products, 21 CFR 211.137(a) requires an expiration date determined by appropriate stability testing described in § 211.166. This is a drug-product requirement; an API supplier’s retest date is not a substitute for that expiration date.

Under 21 CFR 211.137(b), expiration dates must relate to any labelled storage conditions, as determined by stability studies. For a drug product reconstituted at dispensing, 21 CFR 211.137(c) requires expiration information for both the reconstituted and unreconstituted product. That requirement does not itself specify a universal post-reconstitution interval.

ICH recommendations for substances and APIs

ICH Q7 11.61 recommends basing either an API expiry date or retest date on evaluation of stability-study data. It describes retest dating as common practice rather than recommending that every API must carry a retest date.

For drug substances, ICH Q1A(R2) 2.1.10 recommends deriving the retest period from stability information and displaying a retest date on the container label if appropriate. Keep that qualification when translating guidance into specifications or supplier instructions.

As a practical mapping control, document which framework supports each material’s dating convention. Do not apply these drug-substance recommendations to every excipient, food ingredient or chemical merely because its certificate contains a retest field.

For U.S. drug products subject to Part 211, 21 CFR 211.166(a) requires a written testing program to assess stability characteristics. Its results must be used to determine appropriate storage conditions and expiration dates. Operationally, the stability program should make the connection between study results and the assigned date traceable, rather than leaving a date field as an unsupported conclusion.

Tentative dating is conditional. Under 21 CFR 211.166(b), accelerated studies combined with basic stability information on the components, drug product and container-closure system may support tentative expiration dates when full shelf-life studies are unavailable and are being conducted. If accelerated data project a date beyond actual shelf-life support, stability studies must continue, including drug-product testing at appropriate intervals, until the tentative date is verified or the appropriate expiration date is determined. This is a route to provisional support, not a substitute for completing the supporting studies.

For new drug substances, ICH Q1A(R2), drug-substance evaluation discusses limited, justified extrapolation of long-term real-time data to extend a re-test period at approval time. This guidance assumes the same degradation relationship continues beyond the observed data. That scientific and approval context is distinct from deciding whether one warehouse lot can be used after retesting; a passing lot result does not itself establish an extrapolated period.

As an evidence-management practice, connect each assigned date to the applicable material, storage conditions, package configuration, specification and study reference. Review whether the studied configuration and stability storage conditions match the intended application before relying on the conclusion. A passing result from one sampling event describes the sampled material at that assessment; a longitudinal stability program addresses change over time under defined conditions. Neither the numerical result nor the calendar date alone explains why continued quality is supported. Preserve the applicability assessment alongside the date-setting rationale.

These examples use YYYY-MM-DD dates and periods measured in calendar months. Each assumes an already justified dating period and a documented manufacturing-date origin. All intervals are hypothetical, not recommended defaults. The calculations translate an established period into a date; they do not supply the scientific justification for that period.

Case 1: API retest date and remaining quantity

Assume an API manufactured on 2025-02-15 has an authorised 18-calendar-month retest period. Adding that period gives a retest date of 2026-08-15. Record the date type alongside the date as a distinct lot attribute, rather than storing an unexplained calendar value.

Separately, assume 100.0 kg was received, 40.0 kg was consumed and 0.2 kg was sampled, with no other inventory movements. The book balance is 100.0 − 40.0 − 0.2 = 59.8 kg. This quantity reconciliation answers how much remains on the ledger—not whether that material is eligible for further use. The subsequent quality decision remains with authorised people; the balance does not replace raw-material retest assessment.

Case 2: Finished-product expiration date

Assume a covered finished drug product was manufactured on 2024-11-05 and has a justified 24-calendar-month shelf life. Its documented convention assigns an exact date rather than a month-only label. The calculation is 2024-11-05 + 24 calendar months = 2026-11-05.

This example uses the finished product’s own dating basis. A later retest result for an API used in that product is not an input to this calculation: do not infer a new finished-product expiry from that ingredient result. Keep the ingredient reassessment and finished-product dating decision separate.

Case 3: Blended API and the oldest batch

Assume an API blend contains batches manufactured on 2025-04-10 and 2025-06-10. Blending occurs on 2025-07-10, and an applicable, justified 12-calendar-month period is already established for this blend.

ICH Q7 8.47 recommends basing the blended batch’s expiry or retest date on the manufacturing date of the oldest tailings or batch in the blend. Here, the origin is therefore 2025-04-10. Adding 12 calendar months gives 2026-04-10, not 2026-07-10.

Using the blending date would incorrectly add three calendar months in this example. The arithmetic assumes the period applies to this blend; it does not establish blend stability.

A due retest date routes material into re-evaluation; it does not predetermine physical disposal. For APIs, the retest concept is anchored in ICH Q7, which distinguishes a retest date from an expiry date and expects re-evaluation against the specification before further use. For new drug substances, ICH Q1A(R2), section 3 adds specific recommendations: after the re-test period, a batch should be retested for compliance with the specification and then used immediately; and where compliance continues, different portions may be retested and used within the re-test period concept. These are ICH recommendations for drug substances, not a universal use window for every material. A documentary review route can support an assessment where the applicable authorised procedure justifies it, but it does not replace that testing recommendation for materials within Q1A's scope. A passing laboratory result is evidence for the decision, not itself a release approval.

Suggested operating sequence

The following is an industry-practice control model, not a verbatim regulatory workflow or a software capability statement:

1. Identify the affected lot, containers, remaining quantity and intended use. Establish exactly which material the assessment covers.
2. Review storage and packaging history, including excursions, opened containers, repacking and any loss of protective conditions.
3. Determine the assessment route under the applicable procedure: documentary review, sampling and testing, or a combination justified for the material.
4. Obtain the applicable results and supporting records. Confirm that samples and methods correspond to the lot and assessment scope.
5. Evaluate results against the approved specification alongside the history review; do not assess analytical results in isolation.
6. Document the authorised disposition, responsible approver, decision date and any limits on further use before treating the material as available.

Keep exceptions visibly unresolved

• Missing results: retain an unresolved status and obtain the missing evidence.
• Unacceptable results: escalate through the applicable investigation and out-of-specification handling process; do not infer approval from repeat testing.
• Storage or packaging excursions: assess their material-specific impact through the deviation process.
• Conflicting supplier instructions: seek clarification and quality review before deciding which instruction governs.

Assign an owner to each exception and carry it through the appropriate investigation and disposition process. An incomplete assessment should remain distinguishable from an approval.

A lot-specific further-use decision is separate from changing the generally assigned retest period discussed above. If a revised date is justified and authorised, preserve the original date, rationale, decision scope and revision history rather than overwriting the earlier assignment. Make clear whether the decision covers this lot only or changes the broader dating basis.

For intermediates and APIs, ICH Q7 11.41 recommends different placement depending on the date type: an expiry date should appear on the label and Certificate of Analysis (CoA); a retest date should appear on the label and/or CoA. The distinction matters: “and” calls for both locations, while “and/or” allows either location or both. These are ICH guidance recommendations, not a universal labelling rule for every material.

As a practical handoff control, preserve the following at receipt and through transfers, relabelling and downstream document preparation:

• Date value and explicit date type, rather than an ambiguous “valid until” field.
• Original label text, manufacturer identity and manufacturer batch identity.
• Storage statement, document version and provenance: who issued the information and which source document it came from.

Keep these details linked as lot attributes. If the CoA and label disagree, request supplier clarification and document the authorised decision, its basis and any corrected documents. Do not silently choose the later date or convert a retest date into an expiry field.

For repackaging into a different container type, ICH Q7 17.50 recommends stability studies to justify assigned expiration or retest dates for the API or intermediate. Treat this as an evidence-applicability question: does the stability basis support the new container? It is not a blanket instruction to reset dates after every package-handling event. As a practical control, identify the original and replacement container types before carrying a supplier date onto replacement packaging.

Use five linked record entities to reconstruct why a date applied and who authorised the resulting disposition. This is a suggested implementation pattern, not a legally prescribed database schema.

Keep assignments versioned rather than overwriting a single date cell. Each assignment should identify its evidence basis; each assessment and decision should preserve the version actually considered. Keep source evidence distinct from calculated values and human decision records: a calculated date is not itself an approval.

For API manufacturing, ICH Q7 6.13 recommends retaining all production, control and distribution records for at least one year after the batch expiry date. For APIs with retest dates, it instead recommends at least three years after the batch is completely distributed. These are guidance recommendations with different triggering events—not one universal retention duration.

When complete distribution is the applicable trigger, create a separate distribution-completion event record linked to the batch and supporting distribution records. Retention calculations need that actual event. Do not substitute the first shipment date, retesting date or a reserve-sample schedule without assessing the applicable requirement; each represents a different operational purpose.

A practical division is for laboratory records to hold test results, inventory records to hold physical balances, and manufacturing records to hold planned and actual consumption. Across MES–LIMS interfaces and MES–ERP interfaces, use a shared lot identity and consistent date meanings. Keep assigned date, date type (expiry or retest), and disposition status separate: a passing result is not itself a material-use decision.

Design interfaces for ambiguity and sequence

• Flag missing date types for review rather than inferring expiry or retest from a field name.
• Preserve the original value and precision. Keep month-only dates identifiable; do not silently invent a day or shift a date-only value through timezone conversion.
• Identify duplicate updates and retain revision identifiers. Detect stale messages and out-of-order revisions before they overwrite the authorised version.

Recover the decision, not just the display

1. Identify conflicting versions, their origins and affected lots.
2. Route affected material or transactions for review under the site procedure.
3. Reconstruct the authorised decision from approval history and supporting records. For a late supplier revision, include already-consumed quantity and linked manufacturing records in the impact assessment.
4. Reconcile affected systems while preserving superseded values and the reason for correction.
5. Verify downstream acknowledgement and confirm that receiving systems reflect the reconciled decision.

Review queues, reconciliation reports and transaction checks are design options to evaluate and validate, supporting exception-based review. These controls support, rather than replace, authorised people’s decisions about material disposition and use.

Use this practical self-review during implementation and periodic review—not as a list of documented enforcement findings. Follow one lot from its source label and evidence basis through date assignment, assessment, disposition, consumption and retention trigger, using the distinctions covered above.

• Confirm product scope, intended use and explicit date meaning against the source label.
• Trace applicable evidence to the assigned date; reproduce arithmetic from the documented origin and units.
• Check preserved original values and revision history using the earlier records model and data-integrity controls.
• Locate complete assessment results and authorised disposition; distinguish that decision from laboratory completion.
• Reconcile downstream inventory and consumption records with the decision, then identify the defined retention trigger.

Rehearse exception recovery with an unresolved assessment and a late date revision. Trace affected transactions, confirm who owns the decision, and test whether reconciliation reaches every downstream record. Record gaps, owners and follow-up checks without treating a successful rehearsal as proof of compliance.