Coupling high-throughput and modeling approaches to streamline early-stage process development for biologics

Area of Science:

• Biopharmaceutical Process Development
• Chromatographic Separations
• Protein Purification

Background:

• Downstream process development for monoclonal antibodies (mAbs) traditionally uses lab-scale methods, leading to lengthy timelines.
• Increasing complexity of biologics necessitates streamlined and efficient development strategies.
• Automated high-throughput screening and mechanistic modeling are emerging as powerful tools in bioprocessing.

Purpose of the Study:

• To propose and validate a streamlined workflow for early-stage downstream process development.
• To integrate high-throughput screening and mechanistic modeling for accelerated development.
• To reduce timeline and material requirements in bioprocess development.

Main Methods:

• Utilized automated resin plate screens for condition screening and parameter calibration.
• Employed mechanistic modeling to define robust process conditions for manufacturing.
• Confirmed process space using miniaturized RoboColumns, avoiding traditional lab-scale columns.
• Applied the workflow to cation-exchange (CEX) and multimodal cation-exchange (MMCEX) processes.

Main Results:

• The integrated workflow successfully narrowed screening conditions and calibrated binding isotherm parameters.
• Mechanistic models accurately characterized a robust range of conditions for early manufacturing.
• Miniaturized RoboColumns confirmed the predicted process space.
• Demonstrated utility for both CEX and MMCEX processes, predicting mAb product and aggregate impurities.
• Model-predicted parameter sensitivities aligned with traditional design-of-experiment (DOE) models.

Conclusions:

• The proposed workflow offers a practical and efficient roadmap for early downstream process development.
• This approach significantly reduces development timelines and material consumption.
• Enhanced process understanding is achieved through the integration of high-throughput screening and mechanistic modeling.