(367d) Process Control Strategy Development for an Integrated Continuous Platform for Monoclonal Antibody Manufacturing

Process and product understanding is at the root of manufacturers’ efforts to produce safe and effective medicines. The 2009 ICH and FDA Guidance for Industry Q8(R2) Pharmaceutical Development describes a quality by design (QbD) approach that “emphasizes product and process understanding and process control, based on sound science and quality risk management” (1). During the product development stage, the critical quality attributes required to ensure product quality are defined and their relationship to the critical process parameters which impact them are determined. Manufacturers can use a risk-based approach to define a design space within which the combination of process inputs and process parameters have been demonstrated to provide an assurance of product quality (2,3). Consistent product quality is ensured by development of a control strategy that maintains the process within the bounds of the design space.

Process models are an efficient way to define a design space, from the level of individual unit operations or of connected unit operations up to the level of the entire manufacturing process. Process models can be used by manufacturers to ensure product quality, to make predictions, to modify procedures, and for control. Insights gained from process models can lead to improved troubleshooting capabilities during manufacturing and improved control of the critical quality attributes through long-lasting changes in process operating protocols.

Here we present a platform for the cultivation of this understanding via a fully instrumented, integrated continuous manufacturing testbed for monoclonal antibodies (mAbs). The testbed consists of 4 parallel upstream bioreactor setups including 4 perfusion devices, with one reactor assembly integrated with a fully continuous downstream purification system including Protein A chromatography, in-house designed viral inactivation, and ion exchange chromatography. This presentation describes the design and implementation of integrated control strategies for the system. The control strategy discussion begins with tuning of lower level controls (i.e. pH, DO, VCD, media addition, etc.) during process development and for extended cultivations.Following experimental demonstration of the control strategies for integrated operation, some first-principles and data-driven models (4) for the prediction of product-related quantities (i.e. N-linked glycoform distribution) are constructed and validated experimentally. These models are then used to inform a control strategy for the stable, continuous manufacturing of a monoclonal antibody that explicitly controls critical product quality attributes rather than only process quality attributes and process operations.