Cytiva_APR21_EarlyProcessDevDecisions - 26

How to Design a Robust Process for Cell Therapies

of product manufactured. Typical manufacturers

detect, with significant resolution, the variations

rely on a measure of performance that compares

in the manufacturing process.

the output against what's considered a theo-

One risk to doing very significant process

retical yield. But with cell and gene therapy, the

development is that if one makes progress in

theoretical yield is very challenging to calculate.

advancing the manufacturing process by revo-

Also, there can be variations in shipping, specifi-

lutionary change, we may see a shift where the

cally cold chain concerns.

variability could be higher in the analytics than


in the process itself. In that case, you would lose
the resolution to be able to monitor the process

This is a mathematical measure of the levels of

itself. And then, the next challenge is to advance

variability a system can handle, yet still provide

the analytics, as without this resolution in the

quality outputs. And this is truly a measure of

analytics, the process variability may very well go

robustness. There is variability not only in the

undetected. When looking at the overall product

processes inside the flasks or bioreactors, but also

development lifecycle, CCRM operates in what's

in all the processes involved in manufacturing.

considered the process development sweet spot.

These processes include the measures of perfor-

Our ultimate goal is to support manufacturing for

mance, which use the analytical methods to

Phase I and Phase II clinical trials.

monitor the quality and process performance in
the process variability description.
To understand robust design, imagine a

In the cell and gene therapy world, technical
strategies are often highly variable, because the
manufacturing processes are still quite manual.

process that is relatively in control but still has

However, manufacturing projects are viewed

natural variations. As manufacturing scientists, we

through a different lens than what is used in a

will then work on refining the process to further

research and development or process development

reduce this natural variability. These efforts are

lab. It is not necessary to look for the best perfor-

considered Six Sigma applications. For every

mance for individual unit operations. Instead, overall

process, several analytical methods are used to

performance and performance robustness are key.

help monitor the progress. And each of these

Tip: It is better to have a process that is 70%

methods will have its own natural variability.

efficient over 99% of the times it was run, rather

However, ideally you want the magnitude of vari-

than a process that is 99% efficient only 70% of the

ability in analytical methods to be much less than

time. So, it is best practice to send a process back

the natural variability in the process in order to

for more process development work in order to do

26 |


Table of Contents for the Digital Edition of Cytiva_APR21_EarlyProcessDevDecisions

Cytiva_APR21_EarlyProcessDevDecisions - Cover1
Cytiva_APR21_EarlyProcessDevDecisions - Cover2
Cytiva_APR21_EarlyProcessDevDecisions - Contents
Cytiva_APR21_EarlyProcessDevDecisions - 4
Cytiva_APR21_EarlyProcessDevDecisions - 5
Cytiva_APR21_EarlyProcessDevDecisions - 6
Cytiva_APR21_EarlyProcessDevDecisions - 7
Cytiva_APR21_EarlyProcessDevDecisions - 8
Cytiva_APR21_EarlyProcessDevDecisions - 9
Cytiva_APR21_EarlyProcessDevDecisions - 10
Cytiva_APR21_EarlyProcessDevDecisions - 11
Cytiva_APR21_EarlyProcessDevDecisions - 12
Cytiva_APR21_EarlyProcessDevDecisions - 13
Cytiva_APR21_EarlyProcessDevDecisions - 14
Cytiva_APR21_EarlyProcessDevDecisions - 15
Cytiva_APR21_EarlyProcessDevDecisions - 16
Cytiva_APR21_EarlyProcessDevDecisions - 17
Cytiva_APR21_EarlyProcessDevDecisions - 18
Cytiva_APR21_EarlyProcessDevDecisions - 19
Cytiva_APR21_EarlyProcessDevDecisions - 20
Cytiva_APR21_EarlyProcessDevDecisions - 21
Cytiva_APR21_EarlyProcessDevDecisions - 22
Cytiva_APR21_EarlyProcessDevDecisions - 23
Cytiva_APR21_EarlyProcessDevDecisions - 24
Cytiva_APR21_EarlyProcessDevDecisions - 25
Cytiva_APR21_EarlyProcessDevDecisions - 26
Cytiva_APR21_EarlyProcessDevDecisions - 27
Cytiva_APR21_EarlyProcessDevDecisions - 28
Cytiva_APR21_EarlyProcessDevDecisions - 29
Cytiva_APR21_EarlyProcessDevDecisions - 30
Cytiva_APR21_EarlyProcessDevDecisions - 31
Cytiva_APR21_EarlyProcessDevDecisions - 32
Cytiva_APR21_EarlyProcessDevDecisions - 33
Cytiva_APR21_EarlyProcessDevDecisions - 34
Cytiva_APR21_EarlyProcessDevDecisions - 35
Cytiva_APR21_EarlyProcessDevDecisions - 36
Cytiva_APR21_EarlyProcessDevDecisions - Cover3
Cytiva_APR21_EarlyProcessDevDecisions - Cover4