This is the first of several Cleaning Memos dealing with issues related to cleaning processes and cleaning validation for campaigns. We’ll start with some definitions and clarifications. A campaign is a series of batches of the same product manufactured consecutively (that is, one batch immediately followed by another batch). Those batches within a campaign may or may not be followed by a cleaning step. In some cases that “between batches” cleaning may be what is sometimes called “minor cleaning”. Other terms used include “batch to batch cleaning”, “inter-batch cleaning”, and “intra-campaign cleaning” (although I am not a fan of the latter two because of the possible confusion of what “inter” and “intra” mean). After the final batch of a campaign the cleaning that is done is commonly called “changeover cleaning” (or perhaps “major” cleaning). That changeover may be a change to a different product or may be a change to a new campaign of the same product. I have also seen the terms “Type A” cleaning used for the minor cleaning and “Type B” cleaning used for the changeover cleaning, which is a distinction I like because it prevents individual interpretation of what is major, minor, inter-batch, intra-campaign, or changeover by forcing the reader to carefully look at the definition section of the relevant document (so in the interest of clarity I will generally use the terms “Type A” and “Type B” for the rest of this Cleaning Memo).
In many cases the Type A cleaning is a simplified process which is not individually validated. For example, for solid oral dose drug products the Type A cleaning may just involve vacuuming (preferably with a HEPA filtered vacuum system). The rationale here is that a small amount of product from batch “n” left on the equipment may not affect the quality or safety of batch “n+1” (made next). Of course there may be situations where a more stringent cleaning is required, such as due to the development of certain degradation products in the drug product left on the equipment between batches for an extended time. Note that this possible situation may be addressed by a more stringent cleaning, or it may be addressed by sampling the equipment surfaces at the end of batch “n” and then again immediately before manufacture of batch “n+1”, looking at the difference in the impurity profile.
Another example of this Type A cleaning is for liquid oral dose drug products. For those drug products, the Type A cleaning could be a simple water flush, followed by adequate draining to avoid a significant change in the API concentration in the next batch (due to dilution with water). If adequate draining is not possible, drying may be needed, but that drying may create additional issues (it is one thing to dry the equipment after a Type B cleaning, where the equipment is expected to be visually clean; for Type A cleaning that added drying step at elevated temperatures may require additional study to confirm no possible adverse effects on the next batch). For a water rinse with no drying, a possible concern is microbial growth in the water left in the equipment between batches. Assuming the water is at least Purified Water quality, this concern can be addressed by minimizing the time interval between the end of Type A cleaning and the beginning of manufacture of the next batch.
A third example of Type A cleaning is for ointments, where the Type A cleaning might be physical removal of excess drug product using a plastic spatula or scraper. This may or may not be followed by processing and then discarding a fixed portion of the next batch to minimize comingling.
Note in all these situations of Type A cleaning, the equipment may or may not be visually clean. Particularly for the situation with vacuuming of solid products, it is not likely that all equipment surfaces will be visually clean (as would be expected following Type B cleaning). However, it is still prudent to conduct a visual examination of the equipment following the Type A cleaning. The expected outcome is that there should not be gross levels of product left on the equipment surfaces. I realize that “gross levels” is not a well-defined term; however, I suspect that determining a “gross level” would be similar to viewing good artwork – you can’t exactly define it, but you know it when you see it. For training purposes, I generally suggest preparing equipment or coupons that illustrate what might be unacceptable (as well as acceptable) for product on surfaces that are not expected to visually clean.
We will now consider situations where the cleaning between batches in a campaign (Type A) and the cleaning after the last batch of the campaign (Type B) are more or less the same. One situation where this may occur is in biotech fermentation. If I am producing a bulk protein active, for the fermentation step I probably don’t say that the products of the fermentation of one batch don’t affect the quality of the next fermentation batch (of the same product). Even if I did a WFI flush, I likely would not proceed directly with a fermentation of a second batch. The concern would not be batch comingling or safety concerns from carryover of prior products. The major concern is likely to be the fact that any remaining fermentation products may interfere with the subsequent processing steps such that the purity and/or yield of that second batch would be significantly and adversely affected.
This is consistent with the fact that biotech companies (at least the major players) emphasize the criticality of these processes in order to produce consistent products. For that reason, the cleaning typically done after any fermentation batch in a campaign is the same as the cleaning done after the last batch in the campaign. That is Type A cleaning and Type B cleaning are the same. That said, there may be some additional things that are done at the end of the cleaning after the last batch in a campaign. For example, after the last batch in a campaign, prior to changeover to a different product, it is possible to consider changeover of soft parts such as gaskets or membranes. The gasket may be replaced because of concern about product getting trapped within the gasketed area. In that situation, my recomendation is that after the gaskets are replaced (and the equipment reassembled), the equipment is cleaned again using the same process used at changeover. The rationale for the additional cleaning step is that residues removed from the gasketed area may have been deposited on equipment product contact surfaces.
In this example with biotech fermentation, the Type A cleaning process and the Type B cleaning process are exactly the same. In fact, if an additional cleaning is done after gasket change step, I would make that cleaning process exactly the same also. The reason for doing that is to simplify my cleaning “recipes”, thus avoiding the possibility that the wrong recipe were used for a given step. I should also point out that some companies have evaluated whether a change of gaskets is needed at the time of changeover from one product to a different product; such an evaluation may mean that a second cleaning step is not needed. That said, if there are any equipment reconfigurations in the change from one product to a different product, it may be prudent to clean the equipment again before beginning manufacture of that different product.
I should also clarify that in this biotech situation where the Type A cleaning is exactly the same as the Type B cleaning, if it is required that the equipment be visually clean after the Type B cleaning, then it should be an expectation that the equipment be visually clean after the Type A cleaning. (Note that in this biotech example, I may not choose to use the terminology of Type A cleaning and Type B cleaning. I would just say that that my cleaning process would be used both between batches of the same product in a campaign and after the final batch of the campaign.)
Make sure to come back for the next few months as we explore how to determine campaign length and how to validate for a campaign (including validating the maximum campaign length).