Guidance 105 – Defining Worst Case Conditions for Aseptic Process Simulations

First it is important to note that the definition of “worst case” does not mean execution of a media fill at processing failure points where media fill failure would likely occur. Instead, it is the expectation that “worst-case” conditions within the media fill are designed to be performed at the normal operating limits of the production process. Also, “worst-case” conditions should be considered and defined within the media fill simulation program for product holding times, process filling times, filling line speed, container sizes, interventions and personnel.

Members within the regulatory community agree that “media fill studies should closely simulate aseptic manufacturing operations incorporating, as appropriate, worst-case activities and conditions that provide a challenge to aseptic operations”. Practical ways to help determine appropriate process “worst-case” conditions include reviewing routine production batch records as well as observations performed as part of routine aseptic manufacturing operations.

Remember that as part of determining appropriate “worst-case” conditions for media fills the activity “should not be used to justify practices that pose unnecessary contamination risks”.

However when used in the proper context, “if under circumstances of the “worst case” challenge, acceptable results are achieved, then there is greater confidence in the reliability of the system under more normal situations”.

Media fill process design should also consider not only specific product filling line equipment and components, but also local facility (e.g. HVAC) and environmental characteristics (e.g. personnel traffic patterns, process flow, etc.) where the aseptic filling operations take place.

Consideration for a “worst-case” scenario for personnel traffic patterns for example may include accounting for routine entries and exits by operators into a filling suite if the facility design has a dramatic impact on the ability of the HVAC system to maintain proper air pressurization or as a result of entry and exits introduces undesired air patterns for certain aseptic filling line configurations. Depending upon specific area configurations, conditions may exist in adjacent areas to an aseptic filling operation that may have an influence on the air quality to exposed product if the aseptic filling process is not fully isolated from these areas (e.g. air flow patterns, isolators, blow-fill-seal, etc.).

There is however flexibility on how these various elements are actually emulated. There is a common understanding that media fills may not necessarily simulate the actual filling process in an absolute sense, but the design “should simulate the aseptic process as far as reasonably practicable”. This liberty in design extends to the definition of “worst-case” conditions as part of a media fill. For example, there is not an expectation that such “worst-case” conditions include artificially created environmental extremes, such as reconfiguration of HVAC systems to operate at worst-case limits4. Also, it is an acceptable practice when designing “worst-case” situations into a media fill that no single run has to include all identified conditions.

Rather, multiple media fills over time should include site selected “worst-case” conditions. The following are recommendations to consider as part of the design of “worst-case” conditions for a media fill program:

1. Holding Times: A media fill should account for the maximum length of time that the various direct product contact filling equipment, components, and sterile materials may be staged (i.e. exposed) prior to the actual filling operation. The “worst-case” condition would include multiple components and materials in one media fill operation that are at their prescribed maximum allowed holding times (i.e. that is not just one component at its prescribed maximum hold time) at which re-sterilization would be required.

Additionally, holding times are also an important consideration for the compounding process in which manufacturing tanks of sterile bulk drug substances have a specified maximum period from sterile filtration until aseptic filling and sealing, which is in addition to any allowed holding time maximum of product within the sterile bulk tank.

The media fill program design should also account for any “bulk vessel integrity for specified holding times” that would normally occur as part of routine manufacturing operations for product held within the tank for an allowed period       of time prior to commencing any sterile aseptic filling operations. This is supported by current industry guidance that indicates that once the simulated product (i.e. microbiological growth medium) is sterilized and placed within “the holding tank vessel from which filling proceeds, it is held for a period of time at least equal to that for aseptically produced materials”.

A preferred “worst-case” application would be to hold the bulk for a time that just exceeds the maximum time for holding product during routine manufacturing operations. The holding time simulation for holding vessels containing sterilized product “should be covered by a process simulation test on a regular basis unless a validated, pressure hold or vacuum hold test is routinely performed”.

This would allow a physical testing method in lieu of having to perform on a routine basis a specific holding of sterilized media within the bulk tank for the site prescribed maximum holding time period between sterilization until the initiation of aseptic filling. This additional simulation process would not be required if the       pressure hold test is routinely performed, which demonstrates that the sterile bulk holding tank maintains integrity. For example, the holding time simulation could be performed on an annual (e.g. regular) basis and combined within a scheduled aseptic filling media fill process and would be performed to assess the continued capability of the operators to perform critical manipulations that could affect the sterility of the bulk product at this stage of manufacture.

Remember that other compounding activities that involve “any aseptic manipulations performed during and at the end of the holding period” such as       sampling, weighing, multiple filtrations, etc. should also be included in the design of a bulk holding time simulation. Obviously, in the event of a process change, additional holding time simulation(s) would be indicated to support the change in the process that may have an impact on the bulk holding tank operation.

2. Aseptic Process Times: The time it takes to perform all necessary aseptic   processing operations is a major consideration in media fill design. The maximum aseptic process filling time required to fill the largest batch size for any      given container/closure size should also include other activities that are part of a normal aseptic processing operation (e.g. operator breaks, necessary equipment change-outs, etc.).

The most conservative position or “worst-case” for assessing the required time to complete a media fill operation would be the full batch size and duration since it most closely simulates the actual production operation. Other media fill design options can be used and may include performing the media fill immediately following the completion of routine production operations, which is also known as “piggybacking”, or alternating the filling of media fill vials with just operating the filling equipment (without vial filling) to allow for continuous processing time while minimizing the total number of media filled vials.

This provides an acceptable alternative for a “worst-case” scenario to simulate the maximum anticipated filling time. In this situation, the containers should at a minimum be filled at the start of the filling run and at the end of the desired filling time to fully encompass operation.

In any case, a supporting rationale document should be generated that explains the duration of a media fill process simulation in the event that the anticipated duration differs from the actual filling process duration. Any rationale that is developed for aseptic operations that do not employ true personnel separation from the process (e.g. restricted access barriers, isolators, blow-fill-seal) “should be carried out so as to cover all shift operations” since the introduction of new personnel may be a potential source of contamination due to differences in how operations may be performed.

Remember, a very important consideration exists if the intent of performing a media fill is to justify extended continuous production (i.e. aseptic filling campaign). In this case, the media fill should encompass the full duration of the aseptic filling campaign. For instance, a “worst-case” condition relative to total aseptic processing time could include the longest operations and manipulations that could potentially occur in a single routine production run. Additionally, routine processes may occur on different shifts and the “worst-case” would account for all of the operational shifts that the routine process would run so as to account for differences on each operational shift.

3. Number of Units: The number of units required to be filled for a media fill is not always expected to be the same as the number of units that are filled in routine production. This is especially true for very large batch sizes. The degree of latitude in determining the numbers of containers to fill as part of the media fill decreases in direct correlation to batch size (e.g. <5,000 units). There is an expectation that “the size of media fills for small batch size products should at least equal the number of containers filled for the commercial product”. For larger batch sizes, it is becoming a common industry expectation to include at least 5,000 units. As a result of the additional time required to perform           interventions as part of the media fill program, the number of containers filled typically exceeds this number for larger batch sizes.

4. Line Speed: The FDA recognizes that “the media fill program should adequately address the range of line speeds employed during production”. Consideration must be given to defining “appropriate combinations of container size and opening as well as speed of the processing line should be used (preferably at the extremes)”. There is a clear connection between filling line speed and container size combinations when considering a “worst case” condition for an aseptic filling process. For instance, there is a common belief that the combination of slowest filling line speed and largest container opening increases the probability for microbiological contamination.

However, there are instances when faster aseptic filling line speeds present a “worst case” condition such as “small ampoules run at the highest speed as the ampoules may be unstable and cause frequent jams thus necessitating frequent             operator intervention”.

This direct relationship between line speed and container size provides the guidance needed for the appropriate “worst-case” filling line speed. The “worst-case” condition is therefore not necessarily including every possible line speed for a filling process, it is bracketing the operation based on the fastest and slowest operational line speeds. The selected approach should be described within a           supporting rationale document.

5. Container Size: The combination of container size (i.e. container opening diameter) and filling line speed have a direct impact on the probability for     microbiological contamination from the surrounding environment. For instance, “filling the smallest units at the fastest speeds (handling difficulty) and the largest units at the slowest operating speed (maximizing exposure)”. As this example         shows and as previously stated, the “worst-case” design will inherently include container size and line speed since they are directly related.

6. Process Matrix: A process matrix, which considers types of product, batch size, container size, fill volume, line speed, etc. is a primary step in determining the number of distinct media fills required to validate a process. As a matter of efficiency, it is recognized that representative bracketing with at least two process variants that encompass “all others with respect to size, fill, container opening, line speed, manipulations, etc.” can serve as an adequate representation of all variants that may be used during actual product filling processes and therefore will represent the “worst-case” process configurations possible. Since “each filling line is a separate entity, using different personnel and constituting a separate manufacturing process” any justification to eliminate the use of media fills for multiple filling lines used for equivalent aseptic processing operations is not a recommended practice.

7. Personnel: For aseptic filling operations, a “worst-case” design condition in a media fill should include a consideration for increasing the number of filling personnel “to more than the number necessary to fill the batch”. This provides assurance that the maximum number of anticipated operators will be accounted for as part of the media fill. This may not be an important consideration for those aseptic processing operations that employ various degrees of direct personnel separation technologies (e.g. full hard panels with glove ports, half suits, etc.) from the aseptic filling process. This can be accomplished by having an observer from Quality Assurance located in the filling area to ensure that the “worst-case” condition exists since this additional individual would not normally be present during routine production operations.

The media fill should be operated by personnel in the same manner as would occur during production. However, while all filling personnel are expected to participate in a media fill at least once per year this does not include performance in all the defined “worst-case” conditions since these are focused to testing the aseptic manufacturing process. That is, the importance is on the       interventions themselves rather than the specific operator that performs them, especially if the interventions are not attributed to a specific activity that is performed only by a specific individual. Another important consideration addressing “worst-case” is operator fatigue, which can occur over extended filling periods since “staying in the clean room too long may lead to mental fatigue,            which could result in lapses in aseptic technique”. Thus, including in the media fill is recommended the longest anticipated continuous filling operation period between operator breaks.

8. Process Setup: The media fill program “should be designed to detect potential contamination from set-up activities” as the setup of a filling operation inherently is a “worst-case” condition in any aseptic filling operation since it may involve a great deal of direct operator intervention into the immediate filling area.

9. Interventions: The media fill for any aseptic manufacturing process “should represent a “worst case” situation and include all manipulations and interventions likely to be represented during a shift”. Process interventions can be considered to     be either routine (typical) or non-routine (atypical) in nature. Routine interventions are those that occur in most processing operations (e.g. line set-up, fill weight adjustments, addition of stoppers, removal of downed vials, sampling,     environmental monitoring, etc.). Non-routine interventions occur sporadically and do not necessarily occur during every production batch. Non-routine interventions of this type include unexpected processing equipment failures or any occurrence  that hinders routine operations of the process (e.g. line jams, vial guide   adjustments, removal and replacement of damaged components, etc.). Non routine interventions should be included in the intervention plan of the media fill program and performed at least once per year. Media fill documentation “should list all interventions that are permitted during normal batch processing”.

It is important that the “interventions representative of each shift, and shift changeover, should be incorporated into the design” of a media fill. However, remember that “justifying mechanical interventions through environmental monitoring and media fills is not a good practice; the process design focus should be on eliminating interventions”.

The media fill should not be used to justify poor practices. The frequency to perform an intervention during a media fill should also be representative of the number of times the intervention occurs in an aseptic manufacturing operation.