Model-Based Design Space for Protein A Chromatography Resin Selection

Abstract

As demand for biopharmaceuticals rises, manufacturers are required to meet multiple competing key performance indicators (KPIs) such as process sustainability, efficiency and product efficacy and quality. Advanced process optimisation and control in biopharmaceutical manufacturing is challenged by the lack of online Process Analytical Technologies (PAT). This results in processes relying heavily on wet-lab experimentation, which may be costly and inefficient. In this work, a novel methodology for evaluating process robustness and alternative operating strategies using design space identification is proposed to accelerate process design and optimisation. The focus in this work is on the initial separation step for the purification of monoclonal antibodies (mAbs) separating the majority of process impurities generated upstream using affinity (protein A) chromatography. A high fidelity process model is used to computationally explore the multidimensional design space. The performance and robustness of the process under three different resin properties and a variety of input conditions are evaluated using the framework. Three scenarios for each of the resins are considered resulting in a total of nine design spaces. The results indicate that using a higher column protein A density resin can increase operational flexibility.