Hybrid modeling for quality by design and PAT‐benefits and challenges of applications in biopharmaceutical industry

Abstract

This report highlights the drivers, challenges, and enablers of the hybrid modeling applications in biopharmaceutical industry. It is a summary of an expert panel discussion of European academics and industrialists with relevant scientific and engineering backgrounds. Hybrid modeling is viewed in its broader sense, namely as the integration of different knowledge sources in form of parametric and nonparametric models into a hybrid semi-parametric model, for instance the integration of fundamental and data-driven models. A brief description of the current state-of-the-art and industrial uptake of the methodology is provided. The report concludes with a number of recommendations to facilitate further developments and a wider industrial application of this modeling approach. These recommendations are limited to further exploiting the benefits of this methodology within process analytical technology (PAT) applications in biopharmaceutical industry.

Multivariate analysis of the effect of operating conditions on hybridoma cell metabolism and glycosylation of produced antibody

“…First principles models can be used to elucidate the metabolic processes and their impact upon the final glycoprotein production. When combined with multivariate data analysis models, such as PLS demonstrated in this study, the resulting hybrid models have the potential to offer better prediction and control of mammalian cell cultivations …”

“…The research reported here investigates the link between process operating conditions, cell metabolism and the production of different glycosylated forms of a monoclonal antibody (mAb) produced by a murine hybridoma cell line as a first step towards formulating such combined hybrid models . A set of 12 cultivations was selected for model development.…”

Current state and perspectives in modeling and control of human pluripotent stem cell expansion processes in stirred‐tank bioreactors

“…A critical step in such model development is the identification of empirical, data‐driven correlations of the growth rate of differentiated/undifferentiated cells and the cell aggregation characteristics as a function of cultivation conditions. Depending on the amount of experimental data available, these correlations could be based on simple regressions, artificial neural networks (ANNs), or other techniques applied within a hybrid modeling approach . ANN is a powerful approximation technique that may be applied for modeling of the process submodels when a clear phenomenological explanation and knowledge based on fundamental physical and chemical laws are not available.…”

Liquid‐to‐solid ratio control as an advanced process control solution for continuous twin‐screw wet granulation

“…14,15 The present work is based on the former approach, in which the model for control purposes is based solely on measurement data coming from the actual plant. In this respect, purely data-driven models as well as the combination of mechanistic and data-driven knowledge, so-called hybrid models, can be applied for process control applications.…”