Biomedical Engineering Reference
In-Depth Information
3. Lack of rationale for protocol acceptance criteria such that it is not defend-
able to a regulatory agency.
4. Acceptance criteria are too restrictive and likely results in unnecessary
failures or deviations.
5. Lack of process understanding. No development data available
before validation (e.g., no supporting data on equipment capability),
hence no assurance of
successful validation protocol design and
execution.
6. Process is not robust and validation sampling is not representative of
manufacturing conditions, hence increased likelihood of finding problems
during commercial manufacturing.
7. Small-scale studies not representative of commercial-scale manufacturing:
operating ranges were not challenged or verified.
8. Sampling plan is not appropriate and/or assays are not suitable.
9. No statistical consideration for sampling plan.
10. Inadequate training on validation and sampling procedures.
A science- and risk-based approach may alleviate some of these potential
sources of failures, and potential factors that should be considered are described
in the following sections.
8.4.1 Process Steps
Every processing step does not pose the same level of risk; hence, validation
efforts need to be proportional to the level of risk to product quality and patient
safety. For example, leachables from upstream operations would be a lower risk to
product quality when there are subsequent purification steps compared to down-
stream operations. Hence, vendor-generated data for leachables may be sufficient
for an upstream product contract surfaces, but robust leachables/extractables stud-
ies are required for downstream equipment. Risks during physical processing of
drug substance and drug product (e.g., mixing components, and compression) are
different from risks associated with chemical and biological processing, where the
active molecule is formed via chemical or biochemical reactions and subsequently
purified.
8.4.2 Closeness to Patient
As the process moves from upstream to downstream steps in manufacturing,
product quality and patient safety risks increase [14]. For example, bioburden
contamination in a bioreactor could be detected easily and likely result in discard-
ing the batch, and the problem could be resolved before starting the next product
batch. On the other hand, contamination in the final product vial is unlikely
to be detected by limited batch release testing and poses a greater risk to the
patient. Leachables from product vial stoppers may cause an adverse reaction to
the patient [15].
Search WWH ::
Custom Search