In some cases a product may have been on the market without sufficient premarket process validation. In these cases, it may
be possible to validate, in some measure, the adequacy of the process by examination of accumulated test data on the product
and records of the manufacturing procedures used. Retrospective validation can also be useful to augment initial premarket
prospective validation for new products or changed processes. In such cases, preliminary prospective validation should have
been sufficient to warrant product marketing. As additional data is gathered on production lots, such data can be used to
build confidence in the adequacy of the process.
From EU GMP Guide Annex 15:
Validation of such processes should be based on historical data. The steps involved require the preparation of a specific
protocol and the reporting of the results of the data review, leading to a conclusion and a recommendation.The source of data
for this validation should include, but not be limited to batch processing and packaging records, process control charts,
maintenance log books, records of personnel changes, process capability studies, finished product data, including trend cards
and storage stability results.
Although not specifically stated in the FDA PAR or the Q7A PQR, and only mentioned briefly in the EU PQR, the PAR/PQR is actually
an analysis to identify adverse trends and is one of the tools for preventive action as defined in the corrective and preventive
action (CAPA) system. FDA's quality systems guideline defines preventive actions as "Action taken to eliminate the cause of a potential discrepancy or other undesirable situation to prevent such an occurrence."(7)
Frequency and procedures for PAR/PQR
FDA and EU require an annual frequency for the PAR/PQR, which is stated in all three GMP regulations and the guidance document.
FDA does not allow the extension of the review frequency beyond an annual basis, regardless of the number of batches produced
in the preceding 12-month period. FDA expressed the concern that "Potential problems with product quality standards could
go undetected and thereby delay recognition of a need to revise specifications or manufacturing or control procedures"(8).
Table III: Items for PAR/PQR review.
The EU PQR requires that reviews performed in previous periods be taken into account as part of the current review. This has
been an expectation of FDA, as indicated by the many FDA 483 observations, despite no mentioning of this in FDA GMPs or other
guidance documents. Although this requirement is also not mentioned in the Q7A, the PQR for active pharmaceutical ingredients
should include the trending of PQR results from previous periods because this is required by the EU and expected by FDA.
The EU PQR states that "Quality reviews may be grouped by product type (e.g., solid dosage forms, liquid dosage forms, sterile
products) where scientifically justified." FDA prohibits the grouping of different products despite their production using
similar processes, or any other similar approach, because the uniqueness of each production process and product specification
may result in different manufacturing outcomes (9).
The EU PQR allowance for grouping product types opens up the possibility to consolidate the review for a number of drug products
and take a "modular approach" to the information gathering and review process. The modular approach involves using currently
established systems, to the greatest extent possible, to provide summary data and information across a product type to satisfy
the requirements of the PQR. For example, to meet the requirement to review starting and packaging materials, the modular
approach involves querying established systems to provide summary data for all raw materials within a product type and then
performing a consolidated review, identifying any trends and specific actions for the product type. The same approach could
be used for marketing authorization (MA) variations, change control, stability, product returns, complaints and recalls, equipment
qualification, and the technical agreement reviews that are required by the EU PQR. The PQR data mining starts with a product
type review with individual products highlighted as necessary to support any trends and/or identify any product or process
improvements. It is important that this approach is scientifically justified and that this justification is documented. Several
criteria that may be relevant in providing such justification include:
- The PQR topic under review crosses individual product boundaries.
- There are established systems with reporting functionality by product type.
- The trending is more relevant across product types versus individual products.
- The individual product details are not lost during product type reviews.
- There is a concentration of product types at a given manufacturing site.