Scale-up/down of a monoclonal antibody manufacturing bioprocess using data analytics

The scale up/down of biopharmaceutical processes is still a very challenging task. Different cell lines, although clones of the same cell, display different performance in terms of drug productivity. In drug development, the effect of process parameters on cell line performance is often not completely understood and there is a lack of sound science-based methodologies to address this issue. However, the Industry 4.0 revolution is changing the biopharmaceutical industry standards. Extensive digitalization is determining that, even during the process development and scale-up, a significant amount of data can be collected and exploited. In this study we consider a monoclonal antibody manufacturing bioprocess and we focus on two main objectives: the possibility of identifying the most promising cell lines in terms of drug productivity and performance stability from the early development stages, and the prediction of cell lines performance across scales during scale-up/down. This is possible by taking advantage of the information available in the data using multivariate, multiway and multiblock statistical techniques.