(660f) Gas Transfer Characterisation Methodology to Improve Bioreactor Scale-Up/Scale-Down

During cell culture process development, bench scale bioreactors are routinely used to optimize and characterize the operating conditions.  These vessels can be 1:100 to 1:10000 the scale of the manufacturing vessels.  Regularly we observe that the dissolved carbon dioxide concentration is lower in the scale down reactors compared to that in the manufacturing scale.  This difference in carbon dioxide levels has resulted in shifts in productivity or metabolism and may impact the representativeness of, and hence the conclusions drawn from the scale-down runs.  To address this gap in the scale down model, we developed a systematic and mathematical method to conserve the dissolved carbon dioxide concentration across the scales.  The method uses a mass balance-based approach to characterizing the gas transfer coefficients (kLa) in bench- and intermediate- scale bioreactors that incorporates gas transfer from both the head space and sparged gas bubbles.  We found that the headspace contribution to gas transfer which is significant at small scale can be predicted by using parameters such as temperature, working volume, dissolved oxygen, agitation rate, and gas flow rates.  After the bioreactors are fully characterized with respect to gas/liquid mass transfer, we can make rational decisions on how to operate the bioreactors to conserve the dissolved carbon dioxide levels.  Example data from a CHO-based process comparing bench-scale to production scale bioreactors will be shown.