A Contamination Control Strategy is a cyclical process designed to prompt the manufacturers to identify and resolve risk. With the 2020 draft update to Annex 1, David Keen from Ecolab looks at the interesting specific impacts of the changes on the role of a well-designed CCS
Those familiar with EU regulations know that they come with a glossary. When the three words 'contamination control strategy' are used together, one would assume the glossary is the place to go for a clear definition.
In the previous 2017 draft of Annex 1, Contamination Control Strategy (CCS) was mentioned and put in context but not included in the glossary. It has since been added into the current version (12), which leads us to ask why?
So, does this matter and what does this tell us? The three words clearly go together, and when combined you have a clear direction on what you are supposed to do. Both the Annex scoping section (1) and principle section (2) of the 2017 draft Annex and now the version 12, 2020, give a useful amount of detail on why you are required to have a CCS in place. In essence it is a cyclical process designed to prompt the manufacturers to identify, evaluate and control the risk of contamination to the quality of their product and ultimately protect the end user.
When something goes wrong, the site should be able to prevent it happening again
Why has this requirement been added to the draft? One only has to look at the most recent list of drug product recalls, regulatory citations or even deaths caused by the contamination of sterile products, to understand that this problem is real and current. The pharmaceutical industry is well versed in the myriad of contamination causes and effects, but a long view may cause you to conclude that manufacturers are apparently helpless to prevent these issues from reoccurring.
Regulations and current GMPs are updated when best practices are identified, when technological advances push changes, or even when mistakes have been made. The human psychological make up appears helpless in predicting when things may go wrong; the brain is too hard wired to form habits and can become blind to the signals. Humans arguably learn best from trial and error. Unfortunately, this way of learning, when working with sterile products is far from safe.
The requirement to have a robust CCS was therefore a critical addition to the 2017 draft Annex, and has even greater prominence in the 2020 v.12 draft.
Throughout the draft Annex v.12, areas have been identified that should be covered by a site's CCS.
Under Section 2 Principle, you are informed that a functioning CCS will act as a 'health check' on your control and monitoring functions. It informs the reader, that to collect data on the status of manufacturing operations is not enough, rather, you must evaluate that data and look for trends that could be an indication of a looming loss of control and, more importantly, prevent it from happening.
Lines 67‐68 of the draft text ask you take that trend data and feed it back into your site's CCS, making this a living document that adjusts and could become a primary document detailing how a site is minimising risks to product quality. Section 2.6 has been added to further ensure the site's adopt a life cycle approach to the CCS and that current risk mitigation steps feed into it.
Lines 75‐117 detail the fourteen areas in question (incorrectly numbered i to xv, as iii is missing in the draft), which essentially cover all aspects of manufacturing operations. This should not really be a surprise given that a site wide strategy should consider everything. It also pulls in the site's existing Quality metrics and deviation reporting into the feedback loop for the CCS.
When something goes wrong, the site should be able to prevent it happening again (CAPA) and adjust its risk profile in the CCS accordingly. The section's end statement reminds that for a product to remain sterile, terminal sterilisation and the sterility tests are not the only considerations (Section 2.7). The sterility test itself is statistically irrelevant, at best only capable of detecting gross contamination.
In Section 3 Pharmaceutical Quality System (PQS) there is a fleeting mention of CCS (section 3.1 Part iv) in that the site's risk management process should be used to generate and maintain it.
Section 4, Premises 4.3 discusses the need to control human interventions in Restricted Access Barrier Systems (RABS). There is nothing illogical about this, as humans are the major source of microbial contamination. This section requires a site to review interventions and remove the need for them if possible. Where they cannot be avoided, by placing them under the remit of the CCS, a site can show it has given these potentially high‐risk activities some thought.
GMPs are updated when best practices are identified, with technological advances, or even when mistakes are made
Section 4.12 asks you to consider using separate entry and exit change rooms for personnel should the risk of cross contamination be high. Again, the decision to use one or multiple change rooms is driven by the CCS and risk. It goes on to guide that non‐routine items required within the cleanrooms should be subject to a specific risk assessment and that any mitigation measures, be based on this assessment. The method of transferring non‐routine items, normally via disinfection, needs to be documented in accordance with the CCS. An important point is that it requires sites to perform cleaning AND disinfection as part of the material transfer process. Disinfection will not work on a dirty surface, hence the requirement to consider cleaning first. This will be more appropriate to non‐routine items, as routine items will ordinarily be triple wrapped and pre‐sterilised.
Section 4.16 is an addition to the draft where the CCS is tied to another parameter. There is a requirement for a site to document controlling pressure regimes for either isolators or the cleanrooms themselves, within the CCS. The pressure differentials and alarm parameters should be justified and documented.
A CCS is a cyclical process
In section 4.32 the Annex specifies the air speed band for unidirectional flow required for routine manufacture. The reference to the CCS here is for when higher or lower airspeeds than the specified band need to be used and justified. Table 3 in this section is aligned with these considerations for air speed. However, there is an Asterix against air velocity testing required for Grades B‐D. The table footnote directs you, that based on your CCS, you may want to perform this test where there is a particular risk to product quality in that particular area or room. It is clearly stated that this testing is a requirement for filling zones (e.g. when filling terminally sterilised products) and background to Grade A RABS.
It is right that isolators feature prominently in the Annex and in the CCS requirements, as their (poor) operation can lead to significant microbiological risks. Section 4.22 asks that the site methods for cleaning, disinfection or decontamination of an isolator are assessed as part of the CCS. It also covers, via a risk assessment documented in the CCS, the surrounding background cleanroom of the isolator. The higher the risk identified with the operation of the isolator, the higher the grade of cleanroom it should be in.
A known weakness in both isolator systems and RABS are the captive gloves. Should these be damaged, the barrier between humans and the critical area has been breached. There is equipment of various levels of sensitivity that can be employed to aid detection of this. However, CCS requirements do not specifically call out the testing frequency or sensitivity. What the Annex does ask a site to consider is the frequency of glove changes.
In Section 5 Equipment, 5.1 there is an expectation that equipment qualification is kept up to date. If equipment begins to move from the validated set points, the risk of contamination occurring will obviously increase. Here the 'validation lifecycle' approach ‐ an important part of the QMS, is being influenced and impacted by a CCS.
Section 6 Utilities also encompasses vacuum and cooling systems. The guidance is to perform a risk assessment for these critical systems to determine if any required interventions add benefit and to ensure any subsequent risks are mitigated. Section 6.23 includes a cleaning and disinfection requirement for vacuum and cooling systems. However it also states that the requirement for this should be determined by a risk assessment within the CCS. This has tempered this requirement and made it more understandable.
Moving on to Section 7 Personnel, 7.10 has been updated in response to changes in the technological approach to cleanroom management. As paper systems are replaced by electronic ones, handheld devices are being increasingly used, however not all devices are robust enough to be permitted. This new section allows these devices, provided they are supplied by the company solely for use in the cleanrooms and can survive the cleaning and disinfection process, to transfer into a cleanroom. The use and disinfection of these items must be documented within the CCS.
A welcome addition to Section 7 Personnel is 7.14 Part iv, which gives specific instructions about when to wear gloves in Grade D areas. They should be worn when there is a defined risk of contamination to the product or process. This is quite helpful as there are often debates on when to wear gloves in these lower grade areas.
The newly titled Section 8 Production and Specific Technologies that first appeared in the 2017 draft, has been revised further in v.12. Paragraph 8.8 under the subtitle Aseptic Preparation explains particularly well the philosophy behind the CCS and can be applied to the rest of the text contained within the Annex, and is worth quoting in full:
"8.8 The aseptic process should be clearly defined. The risks associated with the aseptic process, and any associated requirements, should be identified, assessed and appropriately controlled. The site's CCS should clearly define the acceptance criteria for these controls, requirements for monitoring and the review of their effectiveness. Methods and procedures to control these risks should be described and implemented. Accepted residual risks should be formally documented."
Section 8.9 goes on to give further helpful advice on the entire scope of this specific section. It states that the risk review does not end with sterile filtration but is applicable during the preparation of the aseptic environment, during all processing stages, and until the product is sealed in its final container. Section 8.82 advises the use of double filtration when performing final filter sterilisation. Naturally, the decision to use filtration over other forms of sterilisation and where to use 1 or 2 in‐line filters should be covered by the CCS.
Filters and the CCS are mentioned again in this section in association with campaign manufacture (starting at 8.96). Continuous or campaign filling when using filters, either for product or for utilities such as gasses, carries a different level of risk to that of batch to batch manufacturing. As such, campaign filling and the specific associated risks, including those associated with filtration, should be documented in the CCS. The reader should refer back to points i. through to iv. of section 8.84 for helpful pointers on what should be considered.
When Blow‐Fill‐Seal (BFS) is considered in a sub section of section 8, it directs that risks from the BFS process to the particular product should be taken into account (in the form of a risk assessment). This is further reinforced in section 8.107, whereby the design, control, and maintenance of the polymer storage, sampling and distribution systems and the capability of the extrusion system to provide appropriate sterility assurance for the moulded container (forming the final container) is assessed for risk of being a source of contamination. This is in addition to the monitoring of the filling machine itself.
One only has to look at the most recent list of drug product recalls, regulatory citations or even deaths caused by the contamination of sterile products, to understand that this problem is real and current
Following on in Section 8 is the sub section for Lyophilisation (freeze drying). Again, the section demands that all the risks associated with this process are documented in the site's CCS, and the prevention of microbiological and particulate contamination is highlighted as a major consideration for this process. Section 8.112 also requires that the batch to batch sterilisation process be thought through in terms of risk and documented in the site CCS.
The next sub sections discuss Closed systems and Single use systems (DSUS), known to be a lower risk of external contamination compared to more traditional open manufacturing. These systems mitigate the known risks associated with open filling, either through deliberate design or an 'unintended' consequence of the technology itself. These risks, including some specific risks associated with single use systems, and their mitigation, need to be added to the site's CCS. Section 8.19 covers the main weakness in a closed system, that of the risk associated with performing aseptic connections. 8.122, for single use systems gives advice on the specific risks with this technology and how they should be assessed as part of the CCS.
Section 9 Viable and non‐viable environmental and process monitoring under the sub section 'General' outlines in 9.1 that the CCS is designed to prevent or minimise the risk from microbiological and particulate contamination. A repetition from early in the document, but one that helps reinforce the intent of a CCS.
Environmental monitoring should form a major part of the monitoring system for demonstrating the level of control in place for risks identified in the CCS. Rather than the individual results themselves, the more powerful tool of trending should be used to periodically review the in‐use risk assessments to determine if they need reevaluation. (paragraph 9.4).
This sub section also covers some of the physical work and analysis that can be carried out (not including EM sampling) in support of and to provide evidence for, ongoing risk assessment. Once the initial work has defined the level of risk and controls, the Annex asks you to use that data further. This paragraph urges you not just to consider EM as part of a batch to batch monitoring process, but to consider its use to determine effectiveness of other cleanroom control measures such as disinfection. It can be the case that the disinfection process is rarely reviewed or scrutinised, and usually a site's EM results alone are used to justify the effectiveness of the process. However inappropriate choice of disinfectant or poorly performed disinfection of surfaces can lead to increased levels of contamination.
Rather than the individual results themselves, the more powerful tool of trending should be used to periodically review the in‐use risk assessments to determine if they need reevaluation
Table 6 within the 'General' chapter contains the limits for airborne non‐viable particulates. It includes the universally accepted 0.5 μm/m3 limits and the more controversial 5.0 μm/m3 limits. Note 2 for this table gives a scientific explanation as to why there is still interest in documenting and trending 5.0 μm/m3 counts, again with reference to the site CCS. Simply put, the idea of monitoring and trending these larger particles is to look for evidence of a negative trend in the area. An example of this could be a piece of equipment that is becoming worn and is beginning to shed more non‐viable particulates. These particulates may be lost in the higher level of background noise at a 0.5 level, but may be more detectable at the higher particle size. Whatever your thoughts on the 5 micron issue, it is here to stay, at least in this version.
Paragraph 9.20 covers how a site should assess the risk of not being able to perform certain expected EM samples:
("In the case where contaminants are present due to the processes involved and would potentially damage the particle counter or present a hazard (e.g. live organisms, powdery products and radiation hazards").
As not performing monitoring could be regarded as critical, it is clearly a risk and needs to be mitigated. Paragraph 9.24 touches on the performance of sampling within the critical areas. Simply put, these sampling techniques should not have a detrimental impact on the manufacturing process itself. These methods should be assessed and documented within the CCS. The paragraph demonstrates where not performing sampling is a lower risk than sampling. To be absolutely clear, personally I think environmental monitoring is a very powerful tool when performed correctly but in some instances, its value is outweighed by the risk.
Continuing the theme of environmental monitoring and risk, paragraph 9.33 asks that personnel monitoring be based on scientific rationale. For that I would read a 'risk‐based approach' rather than a 'generic approach'. For example, a process where monitoring is performed after every critical intervention may be of more value than one performed only once on exit, regardless of the number of interventions. Again, the monitoring frequency for both scenarios should be justified within the CCS.
The final section to be associated with a site CCS within section 9 is paragraph 9.37 Part xii. Here there is a requirement that activities at the start and end of campaign filling (often different from routine activities and potentially of a higher risk) be covered by something like an aseptic process simulation.
A well thought out and implemented CCS should help a site reduce the risk of product quality defects. From this review there do not appear to be any new activities associated with the process, bar documentation of a strategy document(s) itself. Fundamentally, a well thought out CCS will demonstrate to a regulator inspector or auditor that a site is aware of its specific risks to products and how to mitigate them.
A site's ability to demonstrate this level of awareness of its potential problems goes a long way to demonstrating it is a competent GMP compliant entity.