(438b) Using Mechanistic Modeling of Chromatography to Increasing Process Understanding

Although the past decade has seen a rise in the use of mechanistic modeling of chromatography unit operations to increase process understanding, the biopharmaceutical industry’s preferred approach to characterization of these processes is still the resource-intensive and experimentally-based design of experiments (DoE). Although this traditional approach does increase process understanding to a limited degree, resource pressures invariably limit the size (number of experiments) and scope (number of parameters investigated) of these studies and therefore result in a diminished level of process understanding. Due to inherent limitations, DoE based process characterization studies often do not adequately address important sources of process variability, including the impact of: feed stream variability, difficult to control parameters (e.g. column bed height, resin-lot), and variability in buffer preparation. Further, DoE studies are not well suited to identify the edges-of-failure for a process. As mechanistic modeling of chromatography unit operations is uniquely suited to address all of these limitations, we have modeled an intermediate cation exchange purification step using a targeted set of calibration experiments. Using this model, we demonstrate the increased process understanding that can be achieved through mechanistic modeling of chromatography unit operations.