Enhancing Bioprocess Development Through Automated Spent Media and Titer Analysis by LC-MS with Data Connectivity Via Statistical Software

Importance of the Topic

Automating sample preparation and LC-MS analysis in bioprocess development addresses critical challenges such as manual errors, long turnaround times, and variability in data quality. Integrating these automated workflows with advanced statistical software enables rapid and reliable monitoring of both product quality attributes and nutrient utilization. This streamlined approach enhances decision-making during upstream process optimization, accelerating development timelines and ensuring consistent biomanufacturing performance.

Study Objectives and Overview

This study demonstrates a fully integrated platform for end-to-end automation of bioreactor sample handling, LC-MS assays, and data connectivity to multivariate analytics. Key aims include:

• Implementing automated pipetting and sample prep for clarified bioreactor harvests.
• Executing three targeted LC-MS and LC-UV assays for intact glycoform, spent media metabolites, and aggregation profiling.
• Connecting assay outputs directly to data visualization and statistical discovery software for real-time insights.

Methodology

Daily samples were withdrawn from Ambr® micro-bioreactor campaigns and processed on an Andrew+ Pipetting Robot using specifications for clarification, dilution, and protein A cleanup. OneLab software monitored reactor aspiration files and automatically generated assay sequences. Prepared plates were sealed and transferred directly to the BioAccord LC-MS System, executing predefined methods for each assay. Following acquisition, raw data files were exported as CSV and funneled into specialized software tools for processing, visualization, and statistical analysis.

Instrumentation

• Andrew+ Pipetting Robot, BenchHub, and Plate Sealer+
• BioAccord Compact LC-MS System with Automation Portal
• OneLab Workflow Software with integrated file watcher
• Bioprocess Monitor Application and waters_connect connectivity
• Intact Mass Software for glycoform and aggregation calculations
• JMP Statistical Discovery for multivariate data analytics

Main Results and Discussion

Assay 1 (Intact Glycoform Monitoring) applied a dilute-and-shoot approach to day-10 harvests, comparing direct analysis with protein A–purified controls. UV chromatograms revealed consistent detection of light chain and main IgG peaks, demonstrating the robustness of the simple prep method.

Assay 2 (Spent Media Metabolite Profiling) used a 1:400 dilution to quantify amino acids and key energy metabolites. Statistical overlays across eight bioreactors highlighted significant variations in arginine, isoleucine, leucine, and lysine, informing nutrient feed strategies.

Assay 3 (Aggregation Monitoring) processed clarified samples through SEC-UV to quantify high- and low-molecular-weight species. Automated integration via Intact Mass software produced reproducible HMWS/LMWS percentages, enabling timely aggregation risk assessments.

All datasets were integrated within JMP to perform multivariate analytics. Waters_connect and Bioprocess Monitor applications facilitated structured data flow, ensuring seamless entry into visualization dashboards for trend identification and outlier detection.

Benefits and Practical Applications

• Reduced manual intervention and standardized protocols improve reproducibility.
• Automated data collection accelerates process development and comparability studies.
• Real-time connection to statistical tools supports proactive process control.
• Comprehensive profiling of titer, glycosylation, aggregation, and metabolites informs both upstream optimization and downstream purification strategies.

Future Trends and Applications

Advancements in PAT-enabled automation and machine-learning algorithms will further enhance closed-loop control in bioproduction. Expanded assay libraries can integrate glycan, lipid, and peptide mapping alongside metabolite panels. Additionally, cloud-based analytics and digital twins may enable predictive modeling across scales, from micro-reactors to commercial bioreactors.

Conclusion

The described automated workflow unifies sample preparation, LC-MS analysis, and statistical data connectivity, delivering rapid, high-quality insights into critical quality attributes and nutrient dynamics. This platform supports accelerated decision-making in bioprocess development, ultimately reducing time to clinic and improving product consistency.

References

1. LC-MS Analytics in Bioprocessing: Automation-Driven Analysis of Product Quality Attributes and Nutrient Monitoring. Application Note, 2024.
2. New Software for Targeted Analysis of Titer and Media Components Using LC-MS in Bioproduction. White Paper, 2025.
3. JMP® Marketplace Add-In – Bioprocess Monitor. JMP Statistical Discovery LLC.