(272f) Tracking Lot Genealogy in Continuous Drug Substance Manufacturing Processes through Mechanistic Modeling

Due to the nature of continuous processing, varying lot numbers of each given material can be present in a continuous drug substance manufacturing (CDSM) operation at any given time in various concentrations. Tracking all the lot numbers throughout the process is a challenge due to variations in flow rates, lot changes, dispersion in plug flow reactors, one or more steps not being purely continuous, and process disturbances which result in variations in residence time distributions. Despite these challenges, key drivers for tracking lot “concentrations” throughout a CDSM process include:

· Regulatory quality – to prove genealogy in the system for batch release

· Batch definition in production campaigns (reduce bottlenecks in release process for longer runs)

In this talk, an approach will be presented using mechanistic modeling to track lot composition throughout CDSM processes from the feed tanks to isolation using residence time distribution (RTD) models and flowsheet modeling methodologies. This framework will be demonstrated on a realistic CDSM process, quantifying the presence of each lot in the final product. Scenarios related to feed tank operational decisions and subsequent processing steps will be explored.

Such a framework may also be used to monitor lot composition as part of a process’s automation system, aligning with the industry’s efforts to ensure regulatory quality in continuous manufacturing operations. This topic will also be discussed in the presentation.