Method development for improving lipid nanoparticle quantification on charge aerosol detector

Importance of the topic

Lipid nanoparticles (LNPs) are a cornerstone delivery platform for mRNA therapeutics and vaccines. Accurate, robust quantification of individual lipid components in LNP formulations is critical for potency, stability, process control, and regulatory compliance. Universal, non-UV detectors such as the charged aerosol detector (CAD) have become attractive alternatives to UV-based detectors because several LNP components lack chromophores. This study addresses method development and data-processing strategies to improve CAD-based quantification of LNP lipids, with emphasis on linearization, weighting, chromatographic selectivity, and automation for regulated environments.

Objectives and study overview

• Evaluate CAD suitability for quantitative analysis of a panel of LNP lipids (including ionizable lipid SM-102, cholesterol, DSPC, DMG-PEG 2000).
• Characterize detector response behavior (heteroscedasticity, non-linearity) and identify transformations/weighting to achieve reliable calibration across several orders of magnitude.
• Optimize chromatographic selectivity and peak shape to improve detection limits, especially for challenging components such as DMG-PEG.
• Implement a compliant-ready workflow using Empower CDS to automate calibration, processing, and reporting suitable for manufacturing QA/QC.

Methodology

• Standards: Lipids sourced from commercial suppliers; stock solutions prepared at 5 mg/mL in methanol and diluted in 90:10 water:methanol (v/v) for calibration series spanning ~4 orders of magnitude.
• Chromatography: Short reversed-phase phenyl-hexyl columns evaluated (ACQUITY Premier CSH Phenyl-Hexyl 1.7 µm, 130 Å; GTxResolve Lipid Phenyl-Hexyl+ RP 1.6 µm, 230 Å, 2.1 × 50 mm). Typical conditions: 3 µL injection, 0.400 mL/min flow, column temperature 50 °C, mobile phases with 0.1% formic acid (A: water, B: organic—ACN or MeOH/ACN blends) using tailored gradients for lipid separations.
• Detector and data acquisition: CAD detection with settings including 35 °C evaporation temperature, ion trap 20 V, sampling rate 10 Hz, and standard time constant. Response acquisition exploits CAD’s built-in power-law transformation (the p-term) to linearize response during acquisition where possible.
• Data treatment: Examined ELSD-like log–log post-acquisition linearization vs CAD power-function linearization. Addressed heteroscedastic residuals by applying appropriate weighting (notably 1/x2) during calibration fitting in Empower CDS to minimize residual sum of squares (RSS) and improve fit quality over a wide dynamic range.

Used instrumentation

• ACQUITY Premier U(H)PLC system with BSM (Waters) and Empower CDS 3.9.0 for acquisition and automated CAD workflows.
• Columns evaluated: ACQUITY Premier CSH Phenyl-Hexyl (1.7 µm, 130 Å) and GTxResolve Lipid Phenyl-Hexyl+ RP (1.6 µm, 230 Å, 2.1 × 50 mm).
• Charged aerosol detector (CAD) with specified settings: evaporation temperature 35 °C, ion trap 20 V, sampling 10 Hz, normal time constant. Nitrogen nebulization/gas flows as per CAD configuration.

Results and discussion

• Detector response characteristics: Both ELSD and CAD show intrinsically non-linear responses to analyte concentration. ELSD data are typically linearized by log–log (post-acquisition) transformation; CAD applies a power-function transform during acquisition (p-term) to linearize response but residual heteroscedasticity remains for some analytes.
• Heteroscedastic behavior and weighting: Raw CAD responses exhibited increasing variability with concentration (heteroscedastic residuals). Application of weighting during regression (e.g., 1/x2) substantially reduced RSS and produced residual distributions more consistent with homoscedastic assumptions, improving calibration performance and fit reliability (R2 values for lipids typically ≥0.998).
• Chromatographic improvements for DMG-PEG: Use of the GTxResolve Lipid Phenyl-Hexyl+ RP column markedly improved DMG-PEG peak performance—an 86% reduction in peak width and a fivefold increase in peak height relative to the initial CSH Phenyl-Hexyl column. This enhancement materially improved sensitivity and limit of detection for DMG-PEG, historically a challenging analyte for CAD quantification.
• Dynamic range and linearity: Calibrations covering approximately four orders of magnitude were achievable with the combined approach of optimized chromatography, CAD acquisition linearization (PFV/p-term), and appropriate regression weighting. Representative figures of merit showed R2 values ≈ 0.999 for SM-102, cholesterol and DSPC, and 0.998 for DMG-PEG after optimization.
• Automation and compliance: Implementing CAD workflows within Empower CDS allowed streamlined processing (application of power-function value, automated weighting, custom fields for component concentration and molar ratio), reporting, and reduced manual intervention—supporting deployment in regulated manufacturing laboratories.

Benefits and practical applications

• Robust quantification of LNP lipid components across a wide dynamic range supports potency and formulation control in mRNA therapeutic manufacturing.
• Improved chromatographic selectivity and peak shape for low-signal components (e.g., DMG-PEG) lower LOD and enable reliable monitoring of minor components and excipients.
• Use of CAD with appropriate acquisition linearization plus weighted calibration yields reproducible, high-quality quantitative data even in the absence of chromophores.
• Integration with compliance-ready CDS software enables high-throughput, auditable workflows suitable for regulated QC environments.

Future trends and potential applications

• Further refinement of CAD acquisition parameters and optimized particle-selection approaches could enhance sensitivity for very low-abundance lipid species and PEGylated excipients.
• Expansion of tailored stationary phases (e.g., specialized phenyl-hexyl chemistries and superficially porous particle designs) will continue to improve peak capacity and resolution for complex LNP mixtures.
• Standardized approaches for heteroscedastic data handling (automated weighting selection, robust regression) will facilitate method transfer and regulatory acceptance across laboratories and vendors.
• Coupling CAD quantification with orthogonal techniques (mass spectrometry for identity/confirmation, evaporative light scattering for cross-checking) could form a robust multipronged control strategy for advanced lipid therapeutics.

Conclusion

By combining chromatographic selectivity improvements, CAD acquisition linearization (power-function application), and statistically appropriate weighting during calibration, the study demonstrates reliable, wide-dynamic-range quantification of key LNP lipids. Notably, chromatographic optimization with a GTxResolve Lipid Phenyl-Hexyl+ column substantially improved DMG-PEG peak metrics and LOD. Implementing these strategies within Empower CDS produces a compliance-ready, automated workflow suitable for high-throughput regulated environments, advancing analytical control for LNP-based therapeutics.

Reference

1. Fekete S, Doneanu C, Addepalli B, et al. Challenges and emerging trends in liquid chromatography-based analyses of mRNA pharmaceuticals. Journal of Pharmaceutical and Biomedical Analysis. 2023;224:115174.
2. Han D, Birdsall R, Simeone J, Fogwill M, Yu Y. Comparing ELSD and CAD Performance on Polysorbate Quantification in Infliximab Drug Products. Waters Application Note. 2022.
3. Birdsall R, Du X, Bigos P, Han D, Bhiwankar N. Automating Charged Aerosol Detection (CAD) Analysis with Empower CDS Using a Single-Vendor Integrated LC Platform. Waters Application Note. 2026.