PROCESS LIFECYCLE VALIDATION - Risk Management Applications in Pharmaceutical and Biopharmaceutical Manufacturing - page 210

8.9.1 Determining the Significance of a Process Change
When defining a need to conduct additional validation for a process change, the
risks associated with that change should be evaluated to determine the impact
of the change (e.g., the degree of significance) and the likelihood of that change
leading to any potential failure. The following are examples of questions that
could be asked relating to the significance of the change and answers would lead
to the ranking of the change based on associated risks:
• degree of sameness or similarity to the existing conditions, formulation,
process, etc.;
• impact on a CPP, CQA, AQL, or release criteria;
• nature of the change or correction: permanent or temporary, severe or mild;
• location of the change in the overall process (e.g., final product—early or
late in the process);
• type of dosage form;
• impact on validation: is the change within validated range?
• impact on compliance with a regulation, GMP guideline, regulatory-filed
condition or parameter;
• impact on the patient or customer; and
• impact on the firm's reputation, supply, or business.
The likelihood of occurrence of a failure or problem could be based on the
following:
• amount of process understanding and knowledge;
• past history of the area, part, or item that is being changed;
• deviations in the past with respect to this change or similar type of change;
• ability to detect failure as quickly as possible and no later than before the
product leaves the site or manufacturing area, if there is a problem with the
change (i.e., detection component of RA);
• comparison of before and after the change, whether it includes any studies
that may demonstrate reduction in further likelihood of occurrence; and
• history of inspectional or regulatory observations, citations, etc.
Where a risk scoring tool is used, risk scores obtained by combining the
two components, degree of severity and frequency, would lead to the final RI
score, which could then be used to determine the extent and timing of any
additional validation studies. Note that the detectability component may be added
on the basis of the RA tool used (e.g., FMEA). Similar to Example 1, scales
and predetermined thresholds need to be established before scoring the risks for
severity and frequency (and detectability).