Statistical characterization of therapeutic protein modifications

Area of Science:

• Biochemistry
• Analytical Chemistry
• Proteomics

Background:

• Peptide mapping via LC-MS/MS is crucial for characterizing modifications in therapeutic proteins.
• Current statistical analysis methods for LC-MS/MS data lack consensus.
• Accurate site occupancy estimation is vital for therapeutic protein characterization.

Purpose of the Study:

• To propose a statistical approach for therapeutic protein characterization using LC-MS/MS data.
• To address three key statistical goals: site occupancy estimation, differential site occupancy detection, and combined site occupancy estimation.
• To provide a framework for summarizing quantitative mass spectrometry data and performing model-based analysis.

Main Methods:

• Developed a statistical approach for summarizing quantitative LC-MS/MS data.
• Employed statistical modeling and model-based analysis for site occupancy estimation.
• Illustrated the approach with an antibody-drug conjugate and monoclonal antibody intermediate experiment.
• Compared performance against a 'naïve' approach using simulations and orthogonal measurements.

Main Results:

• The proposed approach demonstrated importance of replicated studies and appropriate statistical modeling.
• Achieved reproducible, accurate, and efficient site occupancy estimation.
• Successfully performed differential site occupancy analysis.
• Validated results through computer simulations and orthogonal experimental measurements.

Conclusions:

• Appropriate statistical modeling is essential for accurate and reproducible LC-MS/MS-based protein characterization.
• The developed approach enhances the reliability of site occupancy estimation and differential analysis.
• This work provides a foundation for standardized statistical analysis in therapeutic protein characterization.