Development of generic raman models for a GS-KO^TM CHO platform process

Area of Science:

• Biotechnology
• Process Analytical Technology (PAT)
• Spectroscopic Analysis

Background:

• Bioprocess monitoring traditionally uses offline sampling, leading to delays and process variability.
• Implementing Process Analytical Technology (PAT) is crucial for consistent, high-quality biopharmaceutical production.

Purpose of the Study:

• To evaluate the creation of generic calibration models using Raman probes for bioprocess monitoring.
• To develop models independent of specific cell lines for monitoring key process parameters.

Main Methods:

• Utilized Raman probes for in-situ monitoring of bioprocesses.
• Developed calibration models using two different CHO K1SV GS-KO™ cell lines producing monoclonal antibodies (mAbs).
• Monitored glucose, lactate, glutamate, ammonium, viable cell concentration (VCC), total cell concentration (TCC), and product concentration.

Main Results:

• Achieved average prediction errors for glucose, lactate, and ammonium at 0.44, 0.23, and 0.03 g/L, respectively.
• Demonstrated low prediction errors for viable cell concentration (VCC) and total cell concentration (TCC), averaging 1.90 × 10^6 and 1.85 × 10^6 cells/mL.
• Models showed minimal dependence on the specific cell line used.

Conclusions:

• Generic Raman spectroscopic models enable robust, real-time monitoring of critical bioprocess parameters.
• These models facilitate the application of PAT in manufacturing environments, improving process control and product quality.
• The cell line-independent nature of the models supports their use in diverse biomanufacturing settings.