Abstract
Technologies for large-scale manufacturing of viral vectors for gene therapies, such as tangential flow filtration and membrane chromatography, are under development. In these early stages of process development, techno-economic analyses are useful for identifying membrane properties yielding the greatest impact on process performance. In this study, we adapted a techno-economic framework used for monoclonal antibody capture for adeno-associated viral vector purification. We added mechanistic models to simulate flux decline during harvesting and separating full and empty capsids during polishing. Graphical user interfaces were added to help users explore the design search space. We selected a base process and manipulated selected variables to see their impact on large-scale manufacturing performance. These sensitivity analyses revealed that, under the selected process conditions, increasing module capacity reduces cost of goods more effectively than increasing operational flux in tangential flow membrane filtration modules for virus harvesting. Membrane chromatography columns with relatively low dynamic binding capacity (DBC) and short residence time (RT) offered similar or better economic performance than those with high DBC and long RT. Additionally, the difference in equilibrium solid-phase concentration between full and empty capsids as a function of salt concentration significantly affects purity.