A User-Friendly, Regulatory-Compliant Multi-Attribute Method Workflow for Monitoring Critical Quality Attributes and Enabling Method Transfer

A User-Friendly, Regulatory-Compliant Multi-Attribute Method Workflow for Monitoring Critical Quality Attributes and Enabling Method Transfer — Summary

Importance of the Topic

Biotherapeutic monoclonal antibodies (mAbs) require rigorous monitoring of critical quality attributes (CQAs) to ensure safety and efficacy throughout development and manufacture. Traditional QC relies on multiple orthogonal assays that are resource intensive. The multi-attribute method (MAM) using LC-MS offers consolidated, site-specific CQA identification and quantification, supporting faster decision-making, higher information content, and potential streamlining of QC workflows while meeting regulatory expectations when implemented with compliant software and validated transfer procedures.

Objectives and Study Overview

This work aimed to demonstrate a practical, regulatory-compliant MAM workflow that:

• identifies and verifies CQA sites on a model mAb (Herceptin/trastuzumab) using LC-MS/MS;
• validates the method for robust quantitation;
• transfers the workflow to an MS-only configuration suitable for QC monitoring; and
• implements an easy-to-use, compliant software solution for routine trend analysis and data export.

Experimental Design and Key Methodology

Sample stress and processing

Herceptin was thermally and mechanically stressed (40 °C, agitation) to accelerate modification formation. Aliquots were taken at multiple timepoints (0 to 28 days) to follow formation kinetics of modifications such as deamidation, oxidation and isomerization.

Sample preparation for MAM

Protein samples were denatured, reduced, alkylated, buffer-exchanged, and digested with trypsin. Conditions were optimized to minimize artifactual modifications introduced during sample handling.

Acquisition strategies and data analysis

The characterization phase employed LC-MS/MS (Agilent 1290 Infinity III BioLC coupled to Agilent 6545XT AdvanceBio LC/Q-TOF) in Auto MS/MS mode to obtain peptide fragmentation for site localization. For routine QC monitoring, the validated MAM was transferred to LC-MS in MS-only mode (pre-release Agilent 6230C LC/TOF), relying on extracted ion chromatogram (EIC) peak areas of target peptides for quantitation. Data processing and CQA monitoring used pre-release Agilent software (OpenLab CDS 3.0, BioConfirm 14.1 for identification, and MAM Analysis Workflow 1.0 for streamlined QC monitoring). Results were exported for trend analysis.

Used Instrumentation

• LC: Agilent 1290 Infinity III BioLC (AdvanceBio Peptide Map column, 2.1 x 150 mm, 2.7 µm).
• MS (characterization): Agilent 6545XT AdvanceBio LC/Q-TOF in Auto MS/MS mode.
• MS (QC monitoring): Agilent 6230C LC/TOF operating in MS-only acquisition.
• Software: OpenLab CDS 3.0 (data acquisition), Agilent BioConfirm 14.1 (peptide ID and sequence coverage mapping), Agilent MAM Analysis Workflow 1.0 (EIC-based quantitation and report generation).

Method parameters summarized

Key LC-MS conditions included a 0.25 mL/min flow, column temperature ~50 °C, positive ion mode, MS1 range extending to ~3000 m/z, MS1/MSMS acquisition rates around 3 spectra/s, and collision energy scaled with m/z. Reference masses were used for accurate mass correction. Conditions were chosen to balance peptide separation, sensitivity, and fragment coverage for reliable site assignment.

Main Results and Discussion

CQA identification and localization

LC-MS/MS provided comprehensive sequence coverage (~89% combined heavy and light chains) and enabled confident localization of modifications. MS/MS fragment maps resolved site-specific changes such as deamidation at specific Asn residues (example: light-chain peptide showing localization of deamidation) and distinguished among multiple potential sites within a peptide.

Validated CQAs for Herceptin

A set of peptide-level CQAs was defined and validated by MS/MS, including multiple deamidation hotspots, oxidation of methionine, and isomerization events located across heavy and light chains. Representative monitored sites included N55 [HC], N84 [HC], D102 [HC], M255 [HC], N30 [LC], N158 [LC], among others.

Quantitative monitoring and trend analysis

Upon transfer to MS-only acquisition and processing in MAM Analysis Workflow 1.0, CQAs were quantified using EIC peak areas and expressed as percent modified relative to total peptide (modified + unmodified). The workflow produced reproducible quantitation across several orders of magnitude and across triplicate measurements. Stress time-course data showed that modification levels increased over time; some sites exhibited rapid accumulation (notably certain deamidation and isomerization hotspots), with a subset of sites approaching near-complete modification at late timepoints, illustrating site-dependent susceptibilities within the molecule.

Software and regulatory aspects

MAM Analysis Workflow 1.0 provided a user-focused interface for parameter setting, EIC integration, and report export, supporting regulatory-compliant documentation of quantitation and trends. The approach separates MS/MS-driven identification (for initial CQA definition) from routine MS-only monitoring (for QC), aligning with practical QC laboratory needs.

Benefits and Practical Applications of the Method

• Consolidation of multiple attribute assessments into a single LC-MS workflow reduces assay burden and increases throughput.
• Site-specific information enables targeted monitoring of the most susceptible residues, informing process control and stability strategies.
• The validated transfer from MS/MS characterization to MS-only monitoring supports routine deployment in QC labs with simpler acquisition requirements.
• User-friendly, compliance-focused software simplifies method setup, routine analysis, and data export for trend review and regulatory submission support.

Future Trends and Potential Applications

• Broader adoption of MAM in regulated QC will be driven by robust, validated software and standardized transfer strategies that separate identification from routine monitoring.
• Automation of sample prep and integration with laboratory information management systems (LIMS) will improve throughput and traceability.
• Expansion of MAM to other biologic modalities (e.g., bispecifics, ADCs, fusion proteins) and incorporation of orthogonal checks where needed will broaden applicability.
• Advanced data analytics, including machine learning on longitudinal CQA datasets, can aid early detection of drift and provide predictive stability insights.

Conclusions

This study demonstrates a practical MAM workflow for CQA identification and monitoring of a therapeutic mAb. Characterization by LC-MS/MS established site-specific CQAs with high sequence coverage and localization confidence. After validation, the method was successfully transferred to MS-only acquisition for routine QC monitoring. Pre-release MAM Analysis Workflow 1.0 enabled streamlined, regulatory-conscious analysis and exportable trend reporting, illustrating a feasible route to implement MAM in a QC environment.

References

1. Xiu L, Knierman M, Miladi M, Sato S, Barkovich R, Horner-Buxton J. A User-Friendly, Regulatory-Compliant Multi-Attribute Method Workflow for Monitoring Critical Quality Attributes and Enabling Method Transfer. ASMS 2026 Poster MP 387. Agilent Technologies; 2026.
2. United States Pharmacopeial Convention. USP General Chapter <1060> Analytical Data—Interpretation and Treatment. Rockville, MD: USP; 2024.
3. Agilent Technologies. Product and software information referenced in ASMS 2026 materials. Agilent Technologies; 2026.