Dr Graham
McCreath is Head of Process Design at FUJIFILM Diosynth Biotechnologies. In
this post, Graham describes the company's approach to process characterization in
advance of his presentation at BPI Boston in October 2015.
What is process characterization and how does it
fit with the development and scale-up activities associated with new production
processes?
Process Characterization studies
are a series of risk assessments and experiments undertaken as part of process
validation and are carried out during late stage process development. These
studies generate process knowledge and understanding and are used to
develop robust control strategies for process qualification and commercial manufacture.
The experiments are often carried out at laboratory-scale using qualified
scale down models and statistical designs such as DoE. They are typically carried
out once the final scale-up is complete when full scale data is useful in
qualifying the required scale-down models.
What is special about the way FUJIFILM
Diosynth Biotechnologies perform process characterization relative to other
organizations?
FUJIFILM
Diosynth Biotechnologies have been carrying out process characterization for
over 15 years and have completed more than 24 characterization projects.
In that time we have built up considerable experience and expertise that
benefits our clients with respect to how process information is generated,
evaluated and how this is translated into a control strategy and registration
package. Our scientists and statisticians work collaboratively with
clients to formulate and plan out the most appropriate
robust approach for their needs balancing resources and timeline
against residual risk. We have also developed some in-house QbD
software packages as well to support modern process development; for
example, our RAPTA (Risk Assessment Process Template Application) packages
are FMEA based risk assessment tools that we use with clients as a
predecessor to process characterization. The RAPTA tools enables
us to identify, prioritize and document parameters that may be
important in either a manufacturing process or an analytical method.
What skills or competencies should a manufacturer
have in order to effectively characterize a process?
Having an
understanding of current regulatory expectations,
engineering principles and how they apply to scale down, good
documentation practices are all important competencies. It is also is very
useful to have people in the organisation trained
in experimental design. Modern DoE packages are becoming more user
friendly but a good understanding of the pros and cons of design options and
especially the interpretation of statistical data is very important.
As you developed your approach to process
characterization were there any traps that engineers should avoid when
developing their own approach?
Generally,
our advice would be to have a logical, well thought out plan, good
documentation and rationales and not to overlook having a good scale-down
model. The analytical methods to be used, where they are going to be applied
and their development status should also be considered carefully. There is
little point in executing a large experimental program if you don't have faith
in your measurements.
Graham’s presentation is
scheduled to take place on Thursday, October 29 at 2:30pm as part of the Cell
Culture track.
Join me at #BPIconf
Contact me at nick.hutchinson@parker.com
Dr Nick Hutchinson has a Masters and Doctorate in
Biochemical Engineering from University College London, UK where he focused on
laboratory tools for rapid bioprocess development and characterization. He then
worked at Lonza Biologics in an R&D function investigating novel methods
for large-scale antibody purification before moving to an operational role
scaling-up and transferring manufacturing processes between Lonza sites in the
UK, Spain and USA. Nick now works in Market Development at Parker domnick hunter
where his focus is in bringing Parker's strengths in Motion & Control to
Bioprocessing. This will enable customers to improve the quality and
deliverability of existing and future biopharmaceuticals.
No comments :
Post a Comment