SOP: Annual Product Review / Product Quality Review (APR/PQR)

This is a ready-to-use SOP for the Annual Product Review (APR) / Product Quality Review (PQR) cycle. Replace every <<FILL: ...>> placeholder with your own specifics, set your document numbers and dates, and route it through your normal document control, review, and approval. A worked filled specimen follows the template. This document is educational reference, not legal or regulatory advice; confirm every cited regulation against the current published source before you rely on it.

Document control header

1. Purpose

This procedure defines how <<FILL: COMPANY NAME>> conducts the annual retrospective review of each commercial product’s manufacturing, testing, and quality event history, so that the organization has a documented, evidence-based answer to whether the specifications, the manufacturing process, or the in-process controls need to change. It satisfies 21 CFR 211.180(e) for products marketed in the United States and EU GMP Part I, Chapter 1, section 1.10 for products marketed in the European Union, and is designed to satisfy both from a single review where a product is marketed in both regions.

2. Scope

This procedure applies to every commercial drug, biologic, and combination product manufactured or marketed by <<FILL: COMPANY NAME>> at the sites listed in the header, whether manufactured in house or by a contract manufacturing organization (CMO) or contract testing laboratory. It covers the full review cycle: scheduling, data collection, trending, signal disposition, CAPA linkage, cross-check against the prior period’s actions, consolidation of contract-site contributions, and QA approval. It does not cover the underlying investigation of any individual deviation or OOS, which is governed by <<FILL: SOP-ID for deviation management>> and <<FILL: SOP-ID for OOS investigation>>, nor the CAPA program itself, governed by <<FILL: SOP-ID for CAPA>>. It does not cover the continuous, within-year monitoring performed under continued process verification, governed by <<FILL: SOP-ID for CPV>>, though the two programs must not reach contradictory conclusions on the same data.

3. Responsibilities

4. Definitions

• Annual Product Review (APR): the US term, under 21 CFR 211.180(e), for the written, at-least-annual review of a representative number of batches plus complaints, recalls, returns, salvage, and investigations, evaluated to determine whether specifications, process, or controls need to change.
• Product Quality Review (PQR): the EU term, under EU GMP Part I, Chapter 1, section 1.10, for the equivalent review, with an explicitly longer list of required inputs (stability, marketing authorization variation status, equipment and utility qualification status, technical agreement currency).
• Review period: the fixed twelve-month (or otherwise justified) window a given review covers, defined so it neither overlaps nor gaps against the prior period.
• Data cutoff: the date by which all records for the period must be closed in the source systems before the data pull is finalized. Records opened in the period but still open at cutoff are listed as open, not omitted.
• Signal: a statistically or practically meaningful pattern surfaced by trending, for example a run-rule violation on a control chart, a capability index below an internal threshold, a climbing categorical rate, or a stability station drifting toward a limit, whether or not any individual result failed specification.
• Disposition: the documented outcome assigned to a signal, either “justified acceptable” with a written technical rationale, or “escalated” with a CAPA reference.
• Change creep: the cumulative drift of a process away from its validated center produced by a series of individually approved, individually minor changes.
• Product grouping: the documented basis for reviewing multiple strengths or presentations of a product together as one review versus reviewing them separately, based on whether they share a process and a specification logic tight enough to make pooled trending meaningful.

5. Procedure

5.1 Schedule and scope the review

1. Maintain a master schedule of every product requiring a review, its review period, and its due date, reviewed at least annually by the review author / coordinator.
2. For each cycle, confirm in writing: the review period (calendar year, anniversary of approval, or a justified campaign window), the product definition (which strengths, presentations, and manufacturing sites are grouped into one review versus reviewed separately, per section 5.2), and the data cutoff date.
3. A product approved mid-year receives a short first review, from the date of first commercial batch to the standard period end, and then enters the standard annual cycle.

5.2 Confirm the product grouping

1. For a multi-strength or multi-presentation product, confirm and document whether the strengths or presentations share the same core manufacturing process, the same critical parameters, the same specification logic, and the same equipment train.
2. Where they do, review them together as one product, trending each strength or presentation individually within the report and rolling up to one product-level conclusion.
3. Where they do not, for example different dosage forms, different formulations, or different process descriptions in the registration dossier, review them as separate products and cross-reference the two reports through the shared active substance.
4. Record the grouping decision and its basis in the review record. A grouping decision made silently, or one that pools data from genuinely different processes onto one chart, is a defensibility gap.

5.3 Request and pull the data

1. Issue a data request to each function named in section 3, and to each contract manufacturing organization or contract testing laboratory involved, stating the product, the review period, the data cutoff date, and the required level of detail (batch level or result level, never a pre-digested summary).
2. Query the validated source systems directly wherever the review author has access; retain the query, the extract date, and the extracting analyst’s identity.
3. For CMO or CDMO contributions, confirm the data arrives at the level of detail the quality agreement requires. A summary stating “all batches conforming, no significant deviations” is not acceptable; request the underlying batch-level and result-level data so it can be trended, not just read.

5.4 Reconcile the data set

1. Confirm the batch list ties to the production schedule and to disposition records: batches made equals batches released plus rejected plus reworked plus on hold, with no orphans.
2. Confirm the OOS count, the deviation count, and the complaint count in the review reconcile against the source system of record for the same period, batch for batch.
3. Document, with a written reason, any record excluded from the quantitative trending.
4. Do not proceed to trending until reconciliation is complete. An unreconciled data set produces trending that cannot be defended.

5.5 Trend the data

1. For every critical quality attribute and critical in-process control, build a control chart (individuals/moving range, or X-bar/R where subgroups exist) using a pre-defined run-rule set, and calculate the applicable capability index (Cpk and/or Ppk) against the registered specification.
2. Apply the pre-defined out-of-trend (OOT) rule for the product; an OOT definition created after seeing the data is not trending.
3. For categorical data (deviations, OOS, complaints, returns), trend counts and rates by category against at least the prior one to two periods, normalized to a denominator that removes volume as a confound.
4. Consolidate stability data (annual commitment batches and any active studies) across stations and time points, per <<FILL: SOP-ID for stability program>>, and project any adverse trend against remaining shelf life.
5. Review the full set of change controls approved in the period as a group against the same period’s batch data, to detect cumulative drift (“change creep”) that no single change control’s own impact assessment would show.

5.6 Build the signal register and disposition every signal

1. For every pattern surfaced in section 5.5 that constitutes a signal (as defined in section 4), enter it in the signal register with the specific evidence that triggered it.
2. For each signal, determine the disposition: either a documented, evidence-based technical justification that the signal reflects known, bounded variability and not an assignable or worsening cause (“justified acceptable”), or escalation.
3. Escalate any signal without an adequate justification, or where the evidence points to an assignable or worsening cause, by opening a CAPA in the existing CAPA program per <<FILL: SOP-ID for CAPA>>. Record the CAPA reference number in the signal register. The review does not create or maintain a separate action-tracking mechanism.
4. A signal register with no dispositions, or with every signal defaulting to the same outcome, is itself a finding; the reviewer named in section 3 checks for this pattern before forwarding for approval.

5.7 Review the prior period’s actions

1. Retrieve the signal register and any CAPAs opened from the prior period’s review.
2. Confirm each prior action closed, and confirm an effectiveness check was performed and its outcome recorded.
3. Where a prior action remains open, or where its effectiveness was never verified, record this as a finding in the current review and escalate per the CAPA program.

5.8 Consolidate contract manufacturing contributions

1. Incorporate each CMO or contract testing laboratory’s site-level data contribution into the product-level trending; do not substitute a site-level contribution for the product-level conclusion.
2. Where the product is made at more than one site, trend and disposition signals at the product level across all sites, noting any site-specific pattern.
3. Confirm the quality agreement covering each contract site remains current, and record its status and reference in the review.

5.9 Confirm the EU-list cross-checks

For every review, whether the product is EU-marketed or not, confirm and record the status of: the stability program (per 5.5), marketing authorization variations filed, granted, or refused in the period, the qualification and requalification status of equipment and utilities touching the product, and the currency of technical and quality agreements with any contract site. Writing to this fuller list, rather than the US bare minimum, is what makes one review satisfy both regulatory regimes.

5.10 Write the conclusion

1. State explicitly, in plain terms, whether the specifications, the manufacturing process, or the controls need to change, and why.
2. Where the answer is yes, reference the CAPA numbers or the variation filing that follows from that conclusion.
3. Where the answer is no, state the basis, referencing the signal register’s justified-acceptable entries and the trending that supports the conclusion. A boilerplate “no changes warranted” pasted in without a supporting basis is not an acceptable conclusion.

5.11 Independent review and approval

1. The independent technical reviewer named in section 3 reviews the trending, the signal register, and the conclusion against the underlying data before QA review.
2. Quality Assurance reviews the complete report, confirms every signal has a disposition, every escalation has a CAPA reference, and the conclusion is supported by the data, then signs and dates the report.
3. Complete the review within <<FILL: number>> calendar days>> of the data cutoff date, per section 6.

6. Acceptance criteria

A review is acceptable when all of the following are true:

• The review period, product grouping, and data cutoff were fixed in writing before the data pull began.
• The data set reconciles against the source systems of record, with any exclusion documented and justified.
• Every critical quality attribute and critical in-process control was trended with a control chart and a capability index against a pre-defined run-rule and OOT logic.
• Every signal surfaced by trending has a documented disposition in the signal register, either justified acceptable with supporting evidence, or escalated with a CAPA reference.
• The prior period’s actions were checked for closure and effectiveness, with any gap recorded as a current finding.
• Contract manufacturing contributions arrived at batch or result level and were consolidated into a product-level view, not substituted by a site-level summary.
• The EU-list cross-checks (stability, MA variations, equipment/utility qualification, technical agreement currency) are addressed regardless of the product’s market.
• The conclusion explicitly answers whether specifications, process, or controls need to change, with a supporting basis.
• The report is signed, dated, and approved by QA within the defined completion window from the data cutoff.

7. References

21 CFR 211.180(e), Annual Product Review requirement (US). 21 CFR 211.192, Production record review / investigations (OOS, deviations). EU GMP Part I, Chapter 1, section 1.10, Product Quality Review requirement (EU). ICH Q10, Pharmaceutical Quality System (management review, periodic review of the PQS). ICH Q9(R1), Quality Risk Management. ICH Q1A(R2) and ICH Q1E, Stability testing and evaluation of stability data. FDA Process Validation: General Principles and Practices (2011), Stage 3, Continued Process Verification. FDA Guidance: Contract Manufacturing Arrangements for Drugs, Quality Agreements (2016). A revision of EU GMP Part I, Chapter 1 was under stakeholder consultation as of late 2025, proposing PQR clarifications on product grouping and thin or empty review periods; confirm its status before citing it as a current requirement.

Confirm the current version and clause numbers of each reference before issue.

8. Records generated

9. Revision history

10. Approvals

Filled specimen

The following shows the header and the section 5.1/5.2 scoping decisions completed for an illustrative solid oral dose product family, so you can see the level of detail an inspector expects. The company, product, and numbers are illustrative; replace them with your own.

Section 5.1 scheduling extract:

Section 5.2 product grouping decision: The 5 mg, 10 mg, and 20 mg strengths of Fictional Tab share one granulation and compression process, weight-adjusted per strength, one specification set scaled by dose, and one equipment train at Site A. Grouped as one product for this review; each strength trended individually in section 3 of the report and rolled up to one conclusion. Basis recorded per section 5.2.4.

Section 5.6 signal register extract (one row):

Section 5.11 approval extract: Independent technical review completed by K. Vance (Analytical SME, no role in the underlying data generation) on 2 Mar 2027. QA approval by A. Bello, QA Director, on 12 Mar 2027, 56 days after data cutoff, within the 60-day window.

Common inspection findings this SOP prevents

• A product review completed more than a defined, reasonable window after the review period closes, with no SOP-stated deadline to measure against.
• Multiple strengths or presentations of a product pooled into one trend chart with no documented basis for the grouping, hiding a real process difference.
• A CMO’s site-level summary treated as the product-level review, with no consolidation across sites.
• A signal register that exists but shows no disposition, or where every signal defaults to “justified acceptable” with no CAPA ever opened.
• Prior review actions carried forward year over year with no evidence they were ever verified effective.
• A conclusion stating no changes are needed with no supporting basis tied to the trending performed.

How to adapt this SOP

1. Set your document number, owner, and effective date in the header.
2. Point every <<FILL: SOP-ID ...>> cross-reference to your real deviation, OOS, CAPA, stability, and CPV procedures.
3. Set the completion window in section 5.11 and section 6 to a number your organization can actually meet, and hold to it.
4. Attach or reference your product grouping decisions as a living annex so section 5.2 does not have to be re-litigated every cycle.
5. Confirm every regulation in section 7 against the current published version before issue, including the status of the EU GMP Chapter 1 revision.