Questions that arrive through the contact form, answered here so the next person does not have to ask. Each answer cites the clause. Ask your own at the bottom.
When does the PQR have to be finished?
No regulation sets a number of days after the period ends. EU GMP 1.10 says the review should normally be conducted and documented annually, and 1.11 says the QP and the MAH must ensure it is performed in a timely manner. The deadline is the one in your own procedure, and that is what an inspector holds you to. Most sites set three to six months after the review period closes; some set the due date per product group so the workload is spread across the year.
What gets cited is a review that is late against its own procedure with no documented reason, or a procedure that quietly allows twelve months so nothing is ever late. If a report will miss the date, record the reason, the new date and the QA approval before the original date passes.
Yes. EU GMP 1.10 allows reviews to be grouped by product type, for example solid dosage forms, liquid dosage forms or sterile products, where scientifically justified. The justification is the part that gets read: the products should share a process, equipment train and critical attributes closely enough that a trend in one would be visible and meaningful across the group.
Grouping should not hide a product. Each product in a group still needs its own batch list, its own attribute trends and its own conclusion where the data differs. A group review that averages ten products into one control chart has lost the trend it exists to find.
We only made four batches this year. Do we still trend?
You still review. Trending with four points is not statistically meaningful and the report should say so rather than draw a control chart with limits derived from four results. What you can do is plot the four batches against the previous periods' data, compare each to the specification and the historical mean, and state whether anything looks different. ICH Q7 2.50 notes that the review should be done for products with few batches too; the depth is proportionate, the obligation is not.
Set a rule in your procedure for the minimum batch count below which capability indices are not calculated, and for what replaces them. Then apply it consistently. Inspectors accept a proportionate approach; they do not accept a Cpk of 3.4 computed from three batches presented as evidence of control.
We import a product from a third country. Who writes the PQR and what does our QP need from it?
The manufacturer writes it, because it holds the manufacturing and testing data, and EU GMP 1.10 applies to the product wherever it is made. The EU importer and the MAH must evaluate it, because 1.11 names the MAH and Chapter 7 makes the contract giver responsible for assessing the acceptor's records and results. The quality agreement with the third-country site has to say who authors, what the EU side supplies (complaints, variations, post-marketing commitments), when the report is due and how the MAH's evaluation gets back to the certifying QP.
The QP certifying imported batches relies under Annex 16 on the third-country site's audited GMP compliance and on the pharmaceutical quality system. The PQR is the cross-batch record that keeps that reliance justified: it confirms the process is consistent, the active substance supply chain is traceable, the changes made are the ones approved in the dossier and the agreements are current. A QP who cannot show they have read the PQR, or who has certified for a year against a review that is overdue, has a gap an inspector will ask about.
Our contract manufacturer writes the PQR. Is that enough?
Only if someone at the marketing authorisation holder evaluates it and documents that evaluation. EU GMP 1.11 puts the responsibility on the MAH to evaluate the review and satisfy itself that the assessment is correct, and Chapter 7 requires the technical agreement to define who does what. A CMO report filed unread at the MAH is one of the most common findings in this area.
The technical agreement should say who compiles which sections, which data the MAH supplies (complaints, variations, stability held by the MAH), the timeline for the draft and the MAH's evaluation, and who signs. The MAH's evaluation can be short, but it has to exist, be dated and reach its own conclusion.
Every result is within specification. What trend are we meant to find?
A trend is a change in the data over time, not a failure. EU GMP 1.10 asks the review to highlight trends specifically so that a drift is seen before it produces an out-of-specification result. Results that are all within specification but moving steadily toward one limit, or whose spread has doubled since the last period, are exactly the trend the paragraph is about, and Annex 15 ongoing process verification expects the relevant process trends to be evaluated throughout the lifecycle.
The tools are a control chart with limits derived from the data rather than the specification, a small set of run rules, and a capability index per attribute. The output is a sentence per attribute: stable, drifting, more variable, or too few data to say. A table of results with 'all within specification' underneath is the finding, not the trend evaluation.
EU GMP 1.10 asks the manufacturer to evaluate the results of the review and assess whether corrective and preventive action or revalidation is needed, and says the reasons for such actions should be documented and completed in a timely manner. ICH Q10 3.2.4 lists the outcome of product quality reviews as an input to management review. The review does not have to be read aloud at a meeting, but its conclusions and actions have to reach the people who own resources, and the record has to show that they did.
EU GMP does not name the signatory. It says the manufacturer should conduct the review and, for the MAH and manufacturer, that the reviews are evaluated. Annex 16 requires the QP to take account of the product quality review when certifying batches, and most sites therefore have the QP approve or at least formally review it. What matters to an inspector is that the person who signs had the authority to commit the actions and had actually read the body, not just the conclusion.
For APIs, ICH Q7 2.50 puts the review under the quality unit. Whoever signs, 1.11 makes the QP, together with the MAH, responsible for the review being timely and accurate, so a QP who does not sign still needs a documented route by which the PQR conclusion reaches them before the next certification.
