Abstract
Small-scale cell culture is used for development and optimization of biologics drugs in the biopharmaceutical field. The creation and maintenance of a scaled-down production bioreactor is an arduous and complex task. Monitoring the cell culture throughout the production stages takes daily hands-on manipulation(s), and consistent sample removal from the bioreactor for testing. There is an ongoing need to minimize the cell culture removal from the production bioreactor, and to keep the closed system of the bioreactor protected from exterior pressures. Online and inline Raman spectroscopy probes coupled with predictive analytics have the potential to be developed as a process analytical technology (PAT) that reduces the need to remove cell culture from the bioreactor for daily sampling. This will ensure that there is less risk for external pressures or contaminations as the system will be continuously closed for the duration of the batch fed production.
This thesis delves into the Raman spectroscopic technique and the ability to develop predictive analysis for several different parameters based on the produced spectra. This technique will be used for thirteen cell culture production experiments executed on three different cell lines to produce three different monoclonal antibody products in the small-scale Manufacturing Sciences and Technologies (MS&T) laboratory at AstraZeneca Pharmaceuticals.
Advantages to inline and online Raman spectroscopy in cell culture are that they require very little hands-on manipulation after the predictive models have been created. The Raman instruments are user friendly and take little expertise to use, and if successful, can easily be transferred to the larger scale of production. This could give hourly insights into bioreactor and cell culture where current monitoring methods are maximum of two readouts per day. This could lead to faster response times if a critical process parameter has exceeded the set specification range. Additionally, alert limits could be set to minimize the critical process parameter excursions altogether.