Digital models can be used to optimize drug production processes more effectively than traditional trial and error methods. But digital models could do much more if industry uses them to guide process development and plans for them at the earliest possible stage, says Seyed Mansouri, PhD, professor, Technical University of Denmark.
“The techniques used to model current processes should inform the creation of new processes. Drug companies that build modeling into the earliest stages of process development are going to realize the full benefits,” adds Mansouri. “Also, the ability to combine these models with data to develop robust economic forecasts, quantify process sustainability, and identify areas where new technologies can be used are other advantages.”
Mansouri suggests models built with data from live processes could serve as digital sandboxes where engineers can try out ideas in a more economically viable way than traditional pilot plant-based methods.
“Such models can be used as testbed for evaluating different optimization and operational strategies. With digitalization becoming more vital to enterprises, such mechanistic models can be coupled with plant data to develop digital twins,” he continues. “These tools can help in saving cost, driving sustainability, and making informed decisions for biopharma manufacturing.”
Generic model
Mansouri and colleagues made the case for the greater use of digital modeling in early phase process development in a new publication. The team proposed a plant-wide digital model for the production of a “generic” biopharmaceutical compound that is designed to capture the dynamics from every unit operation within the manufacturing process.
The idea was to use the model as a tool to evaluate different processing scenarios in continuous and semi-continuous fermentation processes.
“The case study presented considers both an upstream and downstream process model for a semi-batch production. This model and strategy behind building it can pave the way for manufacturers to develop their own in-house tools,” says Mansouri. “This will be valuable to address the ever-growing challenges in the biopharma sector for new process development, process optimization, and scaleup.”
Biopharmaceutical industry use of model-based tools is likely to increase, according to Mansouri, who predicts that those firms willing to re-use modeling techniques will benefit most.
“As we advance in exploiting the potentials of data and move on with digitalization, the adoption of model-based tools will also increase significantly. Our ability to re-use and recycle models as well as extending our understanding from a fundamental point of view will be the next steps,” he tells GEN.
