Automated Data Generation for Raman Spectroscopy Calibrations in Multi-Parallel Mini Bioreactors.

Area of Science:

• Biotechnology
• Analytical Chemistry
• Process Analytical Technology (PAT)

Background:

• Raman spectroscopy is a valuable tool for monitoring mammalian cell cultures.
• Existing Raman studies primarily focus on large-scale bioreactor monitoring.
• There is a need for efficient methods to develop robust Raman models for process control.

Purpose of the Study:

• To develop a novel workflow for generating reliable Raman models.
• To enable real-time monitoring and control in biologics manufacturing.
• To correlate spectral variations with actual analyte concentrations for model building.

Main Methods:

• Utilized a high-throughput 250 mL mini bioreactor system.
• Integrated a prototype flow cell for on-line Raman measurements.
• Employed a bioanalyzer for rapid reference measurements.
• Applied a design of experiments (DoE) approach and spiked samples for model optimization.

Main Results:

• Developed robust Raman models for glucose, lactate, glutamine, glutamate, and titer.
• Successfully correlated spectral variations with analyte concentrations in Chinese hamster ovary (CHO) cell cultures.
• Demonstrated the capability to build reliable models for predicting concentrations.

Conclusions:

• The presented mini bioreactor setup facilitates the generation of reliable Raman models.
• This approach supports real-time monitoring and control of biologics manufacturing processes.
• The optimized workflow enhances the predictive power of Raman spectroscopy in cell culture applications.