PROCESS DEVELOPMENT AND CLINICAL PRODUCT MANUFACTURING - Risk Management Applications in Pharmaceutical and Biopharmaceutical Manufacturing - page 107

ensure procedures are being carefully followed; oversight of personnel gowning
to ensure procedures are carefully adhered to; gowning operations performed in
the correct order; etc.
Once the initial Ishikawa analysis has been performed, the hazards are arranged
according to the group's assessment of their potential impact based on prior
knowledge and experience. As this is a subjective assessment, a democratic
approach can be appropriate where majority rules. The facilitator should be care-
ful to avoid letting a particular person override the majority because they have
a powerful personality—this will introduce bias and detract from the usefulness
of the exercise. For example, QC may have experience in analyzing a similar
formulation and explain that they had ongoing problems with the specificity of
the method because of one particular excipient. Without that piece of the puzzle,
the “analytical method” might have been marked as low risk. Now it can remain
low risk, if the excipient in question is eliminated. However, if the excipient
is selected because other considerations take priority, the analytical team will
immediately have a high risk item for failure of the method and will need to
develop an appropriate mitigation strategy.
Table 6.3 shows how the risks identified in the Ishikawa diagram might have
been prioritized by the group and the rationale for the same. Table 6.3 could serve
as a preliminary risk assessment with proposed controls for the development of
the control strategy. This should be formalized as a controlled document, assigned
a version number and date, and signed off by key players involved in performing
the assessment.
6.4 USE OF DESIGN OF EXPERIMENT (DOE) TO ELIMINATE AND
STUDY (C)PPS AT LABORATORY SCALE
Where a large number of factors, which could potentially impact the success
of the process, are identified, design of experiment allows a systematic method
to reduce the number of tests needed to study these factors as well as to study
potential interactions between parameters. When combined with prior knowledge,
it provides a superb tool for exploring and learning how process parameters need
to be controlled in order to achieve the (C)QAs. The outcome of small-scale
experiments will always require close scrutiny and documented verification in
a formal protocol when initial full-scale batches are produced. For elucidation
of these concepts and the requirement to capture the information subsequently
generated in a formal development report, refer to FDA's Process Validation
Guidance [9].
For example, consider the synthesis of an API where, during the synthesis
step, there are numerous parameters that might influence the process such as the
following:
• concentration of reactants;
• pressure;