QUANTIFICATION OF POLYSORBATE IN MONOCLONAL ANTIBODY FORMULATIONS USING CHARGED AEROSOL DETECTION

Significance of the topic

Polysorbate 20 (PS20) and Polysorbate 80 (PS80) are widely used non-ionic surfactants in biopharmaceutical formulations to stabilize protein therapeutics by reducing surface adsorption and limiting stress-induced aggregation. Reliable quantification of polysorbates in final drug products is critical for product quality and stability assessment, but their structural heterogeneity, lack of strong UV chromophores, and broad chromatographic behavior complicate conventional UV-based assays. The work summarized here demonstrates a trap-and-elute reversed-phase approach coupled with charged aerosol detection (CAD) to overcome these analytical challenges and deliver robust quantitation suitable for QC environments.

Objectives and study overview

The primary goals were to develop and evaluate a trap-and-elute RPLC–CAD method that:

• selectively removes protein matrix while retaining and eluting polysorbates for detection
• achieves broad dynamic range and linear response for PS20 and PS80
• demonstrates accuracy and precision in spike–recovery experiments using a monoclonal antibody matrix (NISTmAb)
• identify instrument and method parameters (notably CAD ion-trap voltage and valve configuration) that optimize performance

The study assessed calibration linearity, limit of quantitation, spike-recovery accuracy/precision, and the impact of valve configuration and CAD ion-trap voltage on baseline stability and linearity.

Methodology

Overview of the analytical strategy:

• Trap-and-elute workflow: samples are loaded onto an Oasis MAX trap to retain polysorbates while directing proteins to waste; a fluid-sweep step clears unswept volumes to minimize carryover; a final valve position elutes polysorbates to the analytical path and CAD detector.
• Chromatographic conditions: a short trap column and a high organic elution were used to generate a clean elution of polysorbate species, recognizing that polysorbates present as complex, broad distributions rather than sharp single peaks.
• CAD detection and optimization: the ion-trap voltage (20–600 V evaluated) was adjusted to remove excess small aerosol ions while transmitting charged analyte particles; higher ion-trap voltages reduced baseline noise and improved linearity.
• Calibration and validation: calibration curves for PS20 and PS80 were established over 0.005–0.75 mg/mL (approx. two orders of magnitude). Spike–recovery experiments were performed by spiking NISTmAb with known PS20/PS80 concentrations and assessing repeatability (n=6 injections per level).

Instrumentation used

Key hardware and software reported:

• LC system: ACQUITY Premier System
• Trap column: Oasis MAX Column, 80 Å, 30 µm, 2.1 x 20 mm (p/n: 186002052)
• Column temperature: 30 °C; Sample temperature: 10 °C; Injection volume: 30 µL
• Mobile phases: A = 2.0% formic acid in water; B = 2.0% formic acid in isopropanol
• CAD settings: sampling rate 5 Hz; time constant = Normal; evaporation temperature 40 °C; ion-trap voltages evaluated up to 600 V (600 V used for final conditions)
• Software: Empower CDS (version 3.9.0)
• Column manager: double-valve configuration implementing trap, fluid sweep, and elute steps to reduce unswept volume and matrix coelution

Main results and discussion

Analytical performance highlights:

• Linearity and dynamic range: CAD response demonstrated excellent linearity for PS20 and PS80 across ~0.005–0.75 mg/mL. Representative R² values reached 0.9999 for PS80 and similarly high values for PS20 when optimized.
• Ion-trap voltage effect: increasing ion-trap voltage improved calibration linearity and lowered residual sum of squares (RSS). Lower voltages (100–200 V) produced noticeable nonlinearity; higher voltages (400–600 V) yielded the most linear responses and the lowest baseline noise.
• Precision and accuracy: spike–recovery tests using NISTmAb at three levels (~0.1, ~0.3, ~0.5 mg/mL) yielded measured concentrations within ~1–3% deviation from expected values for both PS20 and PS80. Relative standard deviations were consistently low (<1% in reported data), indicating high repeatability.
• Limit of quantitation and low-level behavior: calibration and area data indicate measurable response down to 0.005 mg/mL, although percent error and uncertainty increase at the lowest concentration levels—particularly for PS20—so practical LOQ should consider acceptable accuracy/precision criteria for the intended QC application.
• Valve configuration benefit: the double-valve column manager with a fluid-sweep step effectively minimized residual protein matrix and unswept volumes, reducing coelution and improving method robustness.

Benefits and practical applications of the method

The trap-and-elute RPLC–CAD approach offers several practical advantages for quantifying polysorbates in protein formulations:

• Universal, mass-responsive detection for non-chromophoric surfactants, avoiding reliance on derivatization or specialized chromophores
• Robust matrix removal via trap column and double-valve fluid-sweep reduces protein interference and carryover—critical for biologic matrices
• Wide dynamic range and demonstrated linearity support routine QC assays across relevant formulation concentrations
• Simplified sample handling: direct injection of formulated product with on-line trapping minimizes sample prep and potential loss or alteration of surfactant species

Future trends and potential applications

Opportunities to extend and enhance the method include:

• Formal method validation across broader product types and excipient profiles to support regulatory submissions
• Refinement of low-level quantitation (LOQ) criteria and calibration strategies (e.g., power-function value tuning, weighted regressions) to improve accuracy at trace concentrations
• Automation and high-throughput adaptations for lot-release testing in manufacturing environments
• Complementary analyses to characterize polysorbate degradation products (e.g., LC–MS workflows after appropriate sample preparation) to pair quantitative CAD assays with structural identification when degradation profiling is required
• Standardization of trap-valve configurations and CAD parameter sets for cross-lab reproducibility

Conclusion

The reported trap-and-elute RPLC method with CAD provides a robust, precise, and accurate means to quantify PS20 and PS80 in monoclonal antibody formulations. Key enablers of performance were the double-valve trapping/fluid-sweep configuration that minimized matrix coelution and careful optimization of the CAD ion-trap voltage to suppress background ions and extend linearity. The approach is well suited for QC laboratories requiring direct, universal detection of non-chromophoric surfactants and can be further developed and validated for broader regulatory application.

References

1. Martos A, et al. Trends on Analytical Characterization of Polysorbates and Their Degradation Products in Biopharmaceutical Formulations. Journal of Pharmaceutical Sciences. 2017;106(7):1722–1735. doi:10.1016/j.xphs.2017.03.001.
2. Hewitt D, et al. Quantitation of Polysorbate 20 in Protein Solutions Using Mixed-Mode Chromatography and Evaporative Light Scattering Detection. Journal of Chromatography A. 2008;1215(1–2):156–160. doi:10.1016/j.chroma.2008.11.017.