Quantification of Polysorbates Using the Waters Charged Aerosol Detector

Significance of the Topic

Non-ionic surfactants polysorbate 20 (PS20) and polysorbate 80 (PS80) are ubiquitous excipients in monoclonal antibody (mAb) formulations where they preserve protein stability, reduce aggregation, and limit nonspecific adsorption during manufacture, storage and delivery. Their accurate measurement at low concentrations in protein-rich matrices is critical for product quality, stability studies and regulatory control, but remains analytically challenging because polysorbates are highly heterogeneous, lack UV chromophores, and occur at concentrations orders of magnitude lower than protein levels.

Objectives and Study Overview

This application study aimed to demonstrate a robust trap-and-elute reversed-phase liquid chromatography (RPLC) workflow coupled to a charged aerosol detector (CAD) for quantifying PS20 and PS80 in mAb formulations. Specific goals were to: establish a trap-and-elute protocol that removes protein matrix, evaluate CAD operating parameters (notably ion trap voltage and power function values) to optimize sensitivity and linearity, define method calibration range, LOD/LOQ, and verify accuracy and precision via spike-recovery experiments using NISTmAb.

Methodology and Key Experimental Conditions

The analytical approach used a trap-and-elute RPLC sequence designed to retain polysorbates while diverting proteins to waste under acidic aqueous loading conditions, followed by strong organic elution to produce a single, integrable polysorbate peak. Important method parameters:

• Sample and standards: PS20 and PS80 stock at 20 mg/mL; calibration standards prepared in water across 0.005–0.75 mg/mL.
• Mobile phases: A = 2% formic acid in water; B = 2% formic acid in isopropanol (100% IPA used for elution).
• Injection volume: 30 µL; column temperature: 30 °C; sample temperature: 10 °C.
• Trap chemistry: Oasis MAX mixed-mode (used in hydrophobic trapping mode via 2% formic acid) enabling protein exclusion and polysorbate retention.

Used Instrumentation

Specific instrumentation reported in the study:

• LC system: ACQUITY Premier with Binary Solvent Manager.
• Trap column: Oasis MAX Column, 30 µm, 2.1 × 20 mm.
• Vials: QuanRecovery with MaxPeak HPS vial and pre-slit PTFE silicone cap.
• Detector: Waters Charged Aerosol Detector (CAD) integrated with Empower CDS.
• Key CAD settings: sampling rate 5 Hz, time constant Normal, evaporation temperature 40 °C, ion trap voltage variable (investigated up to 600 V); PFV adjustable (optimized during study).
• Fluidic configuration: double-valve column manager with an additional downstream CAD divert valve for waste/detector routing.

Main Results and Discussion

Trap-and-elute performance and CAD behavior were characterized as follows:

• Matrix removal: Acidic aqueous loading protonates proteins, minimizing hydrophobic interactions with the polymeric trap and causing proteins to elute in the void to waste, while PS20/PS80 remain retained by hydrophobic interactions and are eluted later with 100% isopropanol as a single broad peak suitable for integration.
• Valve configuration: A double-valve arrangement with an intermediate fluid-sweep step provided superior reproducibility and lower %RSD by flushing unswept volumes and equilibrating fluidic paths; the downstream CAD divert valve further protected the detector from residual matrix during non-detection intervals.
• Ion trap voltage optimization: Increasing ion trap voltage (evaluated 20–600 V) reduced baseline noise and improved calibration linearity but decreased absolute signal intensity. The highest tested voltage (600 V) yielded the best linearity (lowest residual sum of squares), demonstrating a trade-off between sensitivity and linear response range. PFV and ion trap voltage were shown to be interdependent and should be optimized jointly.
• Calibration and sensitivity: Calibration curves for both PS20 and PS80 across 0.005–0.75 mg/mL gave excellent linearity (reported R² up to 0.9999) with low %RSD across levels. PS80 exhibited a higher calibration slope than PS20, attributed to its higher average molecular weight and greater hydrophobicity, which enhance aerosol formation efficiency.
• LOD/LOQ: Because the solvent composition change produces an unavoidable gradient artifact that co-elutes with the analyte peak, classical S/N approaches were not applicable. Following ICH Q2(R2) recommendations, LOD and LOQ were calculated from the standard deviation of blank injections and calibration slope: LOD = 0.002 mg/mL and LOQ = 0.006 mg/mL (reported for PS80).
• Spike-recovery: NISTmAb was spiked at ~0.1, 0.3 and 0.5 mg/mL levels for PS20 and PS80; measured recoveries closely matched spikes with minimal percent deviation and low %RSD across six replicates, confirming accuracy and repeatability in a protein matrix.

Benefits and Practical Applications

The developed trap-and-elute CAD workflow offers multiple practical advantages for biopharmaceutical analysis:

• Universal response for non-volatile surfactants: CAD does not rely on chromophores, enabling consistent detection of polysorbates and similar excipients.
• Robust matrix handling: The trap-and-elute strategy allows direct analysis of protein-containing formulations without extensive sample preparation.
• Adjustable detector parameters: PFV and ion trap voltage tuning provide analysts control over sensitivity, noise suppression, and linear dynamic range to match application needs.
• Regulatory-ready data management: Seamless integration with Empower CDS supports automated processing, standardized calibration, audit trails and report generation suitable for QC environments.

Future Trends and Opportunities

Potential developments and broader applications emerging from this work include:

• Advanced CAD method models: Systematic studies to map PFV and ion-trap settings for wider analyte classes could enable predictive optimization and standard operating procedures.
• Higher-throughput implementations: Shorter trap columns, faster gradients and optimized fluidic switching may increase sample throughput for QC labs.
• Degradation-product profiling: Combining CAD quantitation with orthogonal techniques (MS or ELSD) could improve detection and identification of polysorbate degradation species relevant to stability studies.
• Extended excipient panels: Applying the trap-and-elute CAD approach to other non-volatile excipients (e.g., poloxamers) or complex formulation additives.
• Automation and transferability: Standardized, software-driven method templates in Empower will facilitate method transfer between labs and regulatory submission packages.

Conclusion

The trap-and-elute RPLC method coupled with a modern CAD provides a robust, reproducible and regulatorily compatible platform for quantifying PS20 and PS80 in mAb formulations. Key advantages include effective removal of protein matrices, a single consolidated polysorbate peak for straightforward integration, tunable CAD parameters that enable improved linearity and noise control, and strong spike-recovery performance in a representative mAb. The approach is well suited to both method development and routine QC settings where reliable surfactant quantification is required.

References

The original application note cites literature on polysorbate characterization, trap-and-elute/ELSD methods, CAD linearity modeling and regulatory guidance, including ICH Q2(R2) for analytical procedure validation. Representative citations provided in the source document include peer-reviewed reviews and method papers on polysorbate analysis, Waters application notes on CAD optimization and Empower integration, and ICH guidelines for analytical method development and validation.