19-21 February 2020
In our GMP News on the detailed analysis of the Annex 15 draft, we reported about the changes contained in the draft. Considerable changes have been made to Chapter 9 (Cleaning Validation) too: indeed, the number of sub items has doubled.
But what's really new? Is it (only) an update to the current state of the art? In the following, you'll find an estimating answer to both questions:
"Grouping of equipment" has become standard over the years as well as the fact that selection should be based on rationales. A new point is that a visual check for cleanliness alone isn't considered anymore as acceptance criterion. In the past, a few publications have been published on the topic "visual check for cleanliness" which underlined the sensitivity of this method and showed that this criterion is in some cases more sensitive than any other acceptance criterion usually used so far (1/1000 dose and 10 ppm). "Ongoing verification" after each batch for longer validation studies is a new requirement. The inclusion of the automation level within validation is useful and has hitherto been usual. Today, it is standard and the industry standard to perform a risk assessment before any cleaning validation - and not only for the validation of manual processes. Now, a worst case approach for manual cleaning is required which "automatically" results from this risk assessment.
A "new" requirement is that limits for the carry over of product residues should be based on a toxicological evaluation. You may wonder why "new" is cited in quotation marks. The concept directs to an EMA guideline which describes the PDE concept. Nevertheless, there are large uncertainties in the industry with regard to this concept. To consider potential cumulative effects of multiple equipment in the process equipment train regarding the results is already state of the art.
A standard currently used by the industry is the required observation of cleaning agents and "dirty and clean-hold times" in the context of cleaning validation. The observation of cleanability for campaign manufacture and the length of a campaign are also already implemented that way in the industry.
So far, the use of worst case products in the cleaning validation based on a rationale as well as the renewed observation for the inclusion of new products has been basically state of the art. This is also true for the observation of potential microbial and or endotoxin contamination. The observation of PDE values for the selection of worst case products is something new. The selection and location of samples now required and based on a rationale has also been an industry standard for a long time now. Usually, a risk assessement is performed.
The sampling techniques laid down (swabbing and rinsing) are "new" and depend on the sampling location. The swab material used shouldn't influence the result. Both are standard techniques which have been used for many years in the pharmaceutical and API area. Interestingly, If rinse methods are used, the sampling should be performed during the final rinse in the cleaning procedure. Sometimes, a deviating procedure is used in the industry and the "post-final-rinse" (also called "solvent rinse") is analysed - of course with pre-determined recovery rates. The use of "post-final-rinses" is justified with the argument that the final rinse is still part of the cleaning procedure. Yet, the current requirement for analysing the final rinse present a worst case scenario compared to the "post-final rinse" as higher values may be expected in the final rinse than in the post-final-rinse. Usually, recovery rates are already determined regarding "swabing". Nevertheless, no explicit recovery rates are mentioned in the draft.
The absence of text relating to 3 validation runs is really new. From now on, the number of validation runs should be set on the basis of risk. Regarding this point, the draft refers to a concept of PDA's Technical Report 29. However, the performance of validation verification for the production of investigational medicinal products or products which are only manufactured infrequently is already common practice. Today, the principles of cleaning verification are obviously based on the general principles of cleaning validation. Of course, where cleaning validation has shown to be ineffective or is not appropriate for some equipment, dedicated equipment should be used for each product.