Biomedical Engineering Reference
In-Depth Information
between components upon mixing. An increasing number of pharmaceutical and
biopharmaceutical companies are using in-line or off-line spectroscopic instruments
to perform raw material testing in their supply chain. The library approach using NIR is
themost common in industry. The identity of a class of rawmaterials is usually confirmed
with a spectral library of know “good” rawmaterials that have been confirmed through a
use test.
As outlined by Reicht [46], the selection of samples is critical to the success of the
application. Two sets of samples are required: one for the construction of the library and
an independent one for external validation purposes to verify performance. The number
of batches required to train the system depends on the required specificity of the method.
The training set must collectively describe the typical variation of the substance being
analyzed. Identification involves a smaller number of different batches (3-5) while
qualification requires at least 20. With an appropriate calibration setup, NIR provides
simultaneous quantitative measurements, such as moisture content and particle size
determinations of raw materials. In-line probes such as OD, NIR, and fluorescence are
placed into rawmaterial or fed-batchmedia storage tanks where an assessment is made of
the material prior to opening a valve to introduce the component or complex media into
“fed-batch” fermentation or cell culture processes.
Other off-line techniques are also being used to gain a deeper process understanding
of how individual components in simple and complex raw materials impact process
variability. Techniques such as NMR, HPLC (with DAD or MS), and induction coupled
plasma mass spectrometry (ICP-MS) are commonly employed for these purposes. The
data are commonly correlated with a raw material “use test” to assess its impact on cell
culture or fermentation critical process parameters (such as cell viability or product titer).
These data are then coupled with chemometric visualization tools to more readily
identify potential root cause of the raw material discrepancies. These types of in-depth
insights into complex rawmaterials are beginning to create a paradigm shift onwhat level
of analysis is needed to appropriately control process variability and product safety and
know what additional testing must be performed prior to introduction into the
manufacturing process. Many of the off-line analyses are amenable for conversion
in-line analysis, if the process requires this level of control.
An example of an analysis of a complex raw material component is monitoring a
vitamin breakdown product in minimum essential media (MEM) by UVanalysis. Lanan
and Kiistala [53] showed correlation of photostability of vitamins in aminimumessential
media with cell culture productivity using two UV-based approaches. The first was an
HPLC-based and the second used a 96-well plate-based approach. They demonstrated
quantification of five vitamins in the MEM solution in addition to a vitamin photo-
stability breakdown product. They were able to show that an increase in a breakdown
product of one of the vitamins in the solution correlated with a loss of cell viability.
Further research revealed that the breakdown product was, in fact, toxic to cells. With this
proof-of-concept work, they determined that an optical density probe might be used to
monitor vitamin solutions in the tanks prior to mixing with the other fed-batch media
components. Such a system would allow engineers to test the material, just prior to use.
Scientists and engineers are also learning the importance of key trace elements
contained in their more complex rawmaterials. Kiistala, Lanan, Houde, andDonegan [54]
