Multivariate data analysis was performed on data from a mammalian cell culture process producing antibodies.  Data from the small scale and large scale bioreactor runs were used for the analysis.   To develop a scale-down model of the production bioreactor process, data from the large scale were used as a reference dataset to set acceptance criteria of the scale-down model.   The results show that the satellite runs conducted at small scale runs using the same inoculum and medium lots as large scale runs were unable to meet the acceptance criteria.  Titer, viable cell density and viability trended lower, and lactate, Na+ concentration and osmolality trended higher in the satellite runs compared to large scale runs.  It was hypothesized that due to higher lactate accumulation in the satellite runs more base was added to control pH at the set-point, which resulted in higher Na+ levels and increased Osmolality in the bioreactor.  Based on the data analysis (Multivariate and univariate), operating conditions in the 3L bioreactor runs were adjusted to meet the acceptance criteria for a scale-down model.  In this presentation, the approach used to develop a scale-down model using multivariate and univariate data analysis tools for mammalian cell culture bioreactor process will be presented.

Using multivariate data analysis to develop a scale-down model for a mammalian cell culture process

243rd National Meeting (2012)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.