An Integrated PAT Approach for a Lab Scale Bioreactor IgG3 Cell Culture Process Real Time Monitoring: A Feasibility Study

A model mAb IgG3 cell culture process real time monitoring scheme was reported in this work , to explore the possibility of combining real time spectroscopic techniques and off line characterization tools for a bioreactor cell culture process understanding. Both inline NIR and FTIR Spectroscopy were used to monitor various stages of a model mAb IgG3 cell culture process in real time. In-process nutrient composition , Protein A elution profile , and cell size/morphology were determined offline by Nova 100 Bioanalyzer , Protein A Chromatography , and a digitalized light microscopy , respectively. Without inoculating the bioreactor , a two factor full factorial design (agitation rate with 9 levels , sparge rate with 5 levels) were executed in a 7.5L bioreactor to examine if they impact the NIR real time monitoring. The time series of process NIR spectra was subjected to Principal Component Analysis (PCA). A process trajectory was then constructed based on the PCA results. Three-dimension spectral maps of time-wave number-(NIR or FTIR) spectral intensity were constructed. The process trajectory , 3-D spectral maps , and other process and product characterization results were compared to identify significant process events and elucidate process progression dynamics. It was found that within the scope of DOE , the NIR absorbance of the media was insensitive to either agitation rate or sparge rate. However , the pH adjustments during the bioreactor media equilibration process were characterized by a signal singular point and can be accurately tracked by process trajectory. Applying PCA to the cell culture process NIR clearly demonstrated two distinguishable clusters in the sample map , which correlates well with the Protein A elution profile. The 3-D NIR map demonstrated 3 distinguishable process periods: initial decline , stationary , and growth period. The time series of FTIR spectra along with in-process sample microscopy characterization provided further details regarding some characteristics of those periods. Thus , it demonstrated that combining real time process monitoring and bioreactor in-process sample characterization can help to illustrate process progress and map significant process events.