New Insights into Cell Culture Monitoring: A Look at Chinese Hamster Ovaries

Biopharmaceutical manufacturing is a multistep process. First, it begins with upstream processing, which involves the cultivation of living cells that are used to develop therapeutic proteins (1). A recent study explored this initial process in cell culture monitoring by studying Chinese hamster ovaries.

Close-up of a cute hamster in blue cage | Image Credit: © kulthiwat - stock.adobe.com

Researchers from Zhejiang University, in collaboration with Hisun Biopharmaceutical Co. Ltd., looked at enhancing the understanding and monitoring of cell culture processes by utilizing in-line Raman spectroscopy. Published in the journal Process Biochemistry, Haibin Qu and his team demonstrated a new method using Raman spectroscopy to monitor 27 critical cellular components, showing its utility in biopharmaceutical manufacturing (2).

Biopharmaceutical manufacturing often relies on cultivating cells in controlled environments to produce therapeutic proteins, including monoclonal antibodies (mAbs) and vaccines. However, one of the primary challenges in the field is the complex nature of these biological systems (1,2). Real-time monitoring of the growth and metabolic activity of these cells has been a persistent issue because of the variability and biological complexity involved in cell culture (2). In particular, the intricacies of monitoring critical cellular components—such as amino acids, sugars, and proteins—pose significant analytical hurdles (2).

Of the 27 crucial cellular components monitored in the study, they included amino acids, organic acids, lipids, sugars, and physicochemical parameters such as pH and ion concentration. The study's method enables real-time monitoring of these factors, giving biopharmaceutical manufacturers the ability to track cell health and culture conditions more precisely than ever before (2).

Qu’s study used Raman spectroscopy to develop predictive models for the 27 components. These models provided insights into cell culture dynamics, enabling accurate, real-time measurements. By using chemometrics, the team was able to build and refine these models for enhanced precision (2).

One of the critical contributions of the research is the development of a method to remove anomalous spectra, or outlier data points that can skew results. This allows for a cleaner, more accurate analysis of the cell culture process, further increasing the reliability of the data. According to Qu’s study, the predictive accuracy of the models, represented by the Q² value, exceeded 0.8, which indicates that the models performed fairly well (2). Additionally, the external validation of these models demonstrated a relative percent difference (RPD) greater than 2.0 for all measured components except glucose, which remained a slight outlier (2). The Q² value in multivariate calibration is a measure of the predictive accuracy of a model, representing the fraction of the variance in the dependent variable (parameter concentrations) that is predictable by the independent variables (spectral data), as determined using cross-validation. A higher Q² value for a model, close to 1, indicates better predictive performance, while a value near 0 suggests a poor predictive model.

In their study, the researchers also introduced control charts based on the Raman data to monitor normal and abnormal cell culture conditions. These charts allow manufacturers to quickly identify any deviations from expected performance, including issues like bacterial contamination or insufficient nutrient feeding (2). This real-time monitoring capability is critical for preventing costly disruptions in the cell culture process, reducing waste, and ensuring the consistency of biopharmaceutical products (2).

In situations where cell cultures exhibit unexplained abnormalities, the Raman-based workflow offers diagnostic insights. The team found that this system was particularly adept at identifying subtle changes in the cell environment that could otherwise go unnoticed, providing valuable feedback for troubleshooting and optimizing biomanufacturing processes (2).

The potential benefits extend beyond real-time monitoring. The research workflow provides biopharmaceutical manufacturers with a more in-depth understanding of the cell culture process. This could lead to improved cell line development, more efficient production processes, and ultimately, higher quality therapeutic proteins (2).

The researchers plan to expand their work by further refining their Raman-based models and integrating other advanced techniques such as nuclear magnetic resonance (NMR) spectroscopy. The combination of these analytical tools may unlock even more insights into the behavior of cells under different culture conditions (2).

By employing Raman spectroscopy for real-time monitoring of Chinese hamster ovary (CHO) cell cultures, Qu and his team demonstrated how their method could not only improve the efficiency of cell culture monitoring, but also how it can lead to significant improvements in the quality and safety of biopharmaceutical products.

References

1. Proventa International, Available at: https://proventainternational.com/biopharmaceutical-manufacturing-techniques/#:~:text=The%20manufacturing%20process%20of%20biopharmaceuticals,conditions%20and%20maximize%20protein%20production. (accessed 2024-10-11).
2. Dong, X.; Shi, Y.; Yan, X.; et al. Comprehensive Cell Culture Monitoring: Leveraging In-line Raman Spectroscopy for Enhanced Biopharmaceutical Manufacturing Insights. Proc. Biochem. 2024, 143, 1–12. DOI: 10.1016/j.procbio.2024.04.022