As biopharma companies fine-tune their strategies for continuous bioprocessing, bioburden control has become a top priority. The approaches that worked for batch processing are inadequate for continuous processing streams, so new strategies are needed.

The issue is that a drug substance may move through a continuous bioprocessing configuration within 24 hours, thus completing it before tests can be returned. Integrating bioburden testing into continuous bioprocessing expedites detection and has the potential to eliminate human sample-handling errors.

The 2004 FDA Guidance on Process Analytical Technology (PAT) described a two-tier testing strategy. It hasn’t been routinely applied, however, because, as Gene Schaefer, PhD, senior fellow at the National Institute for Innovation in Manufacturing Biopharmaceuticals (NIIMBL), tells GEN, “In-process micro-testing has long been considered a special activity, removed from the test plans for other quality attributes.”

Irina Ramos, PhD, director, bioprocess technologies & engineering, AstraZeneca, and Michelle Najera, PhD, principal scientist, purification process design, Just-Evotech Biologics, along with Schaefer, recently identified user requirements for rapid bioburden testing, focusing on fault detection. Their paper outlines mandatory, as well as “nice-to-have,” features and evaluates rapid microbial testing technologies for near-real-time detection.

“The greatest challenge,” they point out, “is timely decision-making for where and when to divert and segregate material,” when bioburden is detected. Exploring various scenarios beforehand with an eye to “test duration and frequency alongside resident times,” are essential to effective decision-making.

The other significant challenge is managing large amounts of data (which can be accomplished via artificial intelligence), in addition to sample volumes, equipment integration, and equipment sterility.

Control system

The two-tiered control system they outline uses established PAT control methods in which the first test determines whether the system is closed, aseptic, and generating an appropriate in-process signal. The second test confirms (or refutes) the initial findings and identifies specific contaminants.

“Regardless of which rapid technology is chosen to test bioburden, it is important to build a control strategy that relies on results in the context of residence time, sampling point, and simpler diversion/segregation points of the process, to preserve product quality,” Ramos tells GEN.

Real-time data is critical in making near-real-time decisions during processing. To be most useful, she stresses, it must be linked to a well-defined decision framework that standardizes actions based on test results.

Implementing in-process bioburden testing for continuous manufacturing is technically challenging today, but will ease as new as new equipment, connections, and automation, are developed.

“The availability of devices that…are sufficiently accurate to trigger in-process decisions makes a two-tiered control feasible,” Schaefer says, “especially when the alternative is running a process with a rapid cadence at risk.”

Previous articleMaking Bioprocessing More Sustainable
Next articleGetting the Economical Edge on Cheaper Cell Therapy Manufacturing