Page 77 - Assurance of Sterility for Sensitive Combination Products and Materials

Aseptic processing 63

may also uncover unaddressed process variables and weaknesses missed in
earlier evaluations. They are a check on the accuracy and effectiveness of
assumptions. Process simulations are not a substitute for good process design
and should not be relied upon to be the primary judge of the effectiveness
of the aseptic process. Confidence in the aseptic process is primarily gained
through proper design of the aseptic process with aseptic process simula-
tions being used as for a tool to confirming the acceptability of the process
or uncovering missed process variables and weaknesses.
As stated in PDA TR 22, the aseptic process simulations are designed to:
• evaluate capabilities of aseptic processing operation,
• simulate the aseptic process from the point of sterilization to closure of
the container, substituting a microbiological growth medium for the
sterile product,
• assess changes made to an aseptic processing operation which might
impact the sterility of the final product,
• identify weaknesses in aseptic processing which might contribute to the
microbiological contamination of the product,
• valuate proficiency of aseptic processing personnel,
• comply with current Good Manufacturing Practice requirements, and
• demonstrate appropriateness of operating practices used in support of
aseptic processing.
Solely running aseptic process simulations will not:
• Validate the aseptic process: The validation of the aseptic process involves a
holistic approach, incorporating validation and qualification of numer-
ous aspects of process design and contamination control.
• Assure process control: Process control is assured through careful evaluation
of the critical quality attributes, process steps, process variables, risks, and
control strategies to mitigate those risks.
• Determine sterility assurance level: SAL (sterility assurance level) or PNSU
(probability of non-sterile unit), such as those referenced in sterilization
processes, cannot be calculated or predicted for aseptic processing by the
outcome of aseptic process simulations, due to lack of statistical correla-
tions between observed parameters and desired outcome. In aseptic pro-
cess simulations, each fill is considered as a separate study with a separate
set of conditions. Combining those results to fit a statistical model would
not be accurate.
• Qualify clean room personnel: Participation in successful aseptic process
simulations will not be enough to provide assurance that clean room
personnel are properly trained and capable of performing assigned tasks.