PROVING THE EFFECTIVENESS OF SANITIZATION METHODS
If a manufacturer only has experience performing a particular unit operation in a grade C or grade D environment, data may
be required to convince internal stakeholders and external regulators that CNC operation is acceptable.
Functionally-closed systems require a sanitization step after the system has been opened, and the effectiveness of the sanitization
step is key to proving that CNC operation is feasible. Clean steam is a well-established mechanism for sanitizing process
equipment. Steam sanitization is not, however, always feasible or economical. If other sanitization methods are used to establish
functionally-closed systems (e.g., caustic solutions or hot WFI), data must be shown to support the effectiveness of those
methods. Some data are available in published literature, but demonstrating the effectiveness of the sanitization method in
the configuration that is proposed may also be needed. Deliberate contamination of process equipment is not desirable, but
a relatively inexpensive test system could be constructed to simulate conditions in the target process equipment.
Consider the case in which the test system is used to challenge the connection of a single-use element to stainless steel
equipment. The test system would initially be clean. Prior to making the connection, the single-use element would be contaminated
with a known amount of a biological contaminant (e.g., bacterial, fungus, yeast, etc.). After connection, a caustic solution
sanitization regimen would be applied and then flushed from the system. Then, a process fluid would be pumped into the system
and circulated. The fluid would be chosen to represent fluids used in the manufacturing process. After the circulation period,
a sample would be withdrawn from the system and tested for the presence of the target organism. The concentration of the microorganism
in the sample would provide information regarding the effectiveness of the sanitization method. The premise is that if the
sanitization method is sufficient to remove a deliberate contamination, it will be sufficient for operation in a CNC space.
ENDORSEMENT BY INTERNAL QUALITY REPRESENTATIVES AND HEALTH AUTHORITIES
Facility design for bulk biopharmaceutical manufacturing is a direct descendent of regulatory expectations for biologics manufacturing.
Design principles for BDS manufacturing have been extrapolated from guidance documents and regulations developed for aseptic
processing (i.e., final product manufacturing). Historically, this design precedent has been copied, repeated, and considered
"industry standard" based on successful licensure of products sourced from these facilities. Facility design concepts have
remained essentially stagnant while process enhancements continue to provide additional assurance that the drug substances
will consistently meet their quality attributes. The following arguments should be considered when seeking endorsement for
CNC processing from internal quality representatives or from external regulators.
Process design and capability for clearance
Bioburden and viral clearance steps in the process should be emphasized when considering measures that mitigate risk. For
example, the use of bioburden reduction and viral filters within the process can provide assurance that in-process intermediates
and the final BDS will meet its predetermined quality attributes.
Precise and specific knowledge drives contamination control strategy
Supportive controls such as air cascade designs, gowning procedures, and cleaning regimens can reduce variability in the manufacturing
environment. Risk analysis should lead to a comprehensive microbial control strategy. Knowledge and understanding of the variables
that may impact process quality should be the foundation for the microbial control strategy.
Routine in-process microbiological testing
In-process bioburden testing of intermediate process steps should be established to ensure process control. Microbiological
alert levels should be established for each process step based on demonstrated process capability. Action levels should be
established to identify excursions that may impact final BDS quality. Data should be actively trended to identify any shifts
or changes in the variability of the process.