Deterministic Acceptance Limits for Statistical Equivalence Testing in Hydrogen/Deuterium Exchange Mass Spectrometry

Area of Science:

• Biochemistry and Biophysics
• Analytical Chemistry
• Pharmaceutical Sciences

Background:

• Hydrogen/Deuterium Exchange Mass Spectrometry (HDX-MS) is crucial for protein higher-order structure characterization.
• Existing statistical tools for HDX-MS focus on detecting differences, not similarity required for biopharmaceutical comparability.
• Previous equivalence testing methods had variability due to randomized resampling.

Purpose of the Study:

• To enhance an existing HDX-MS equivalence testing method for biopharmaceutical comparability.
• To eliminate variability introduced by randomized resampling in acceptance limit calculations.
• To develop more stringent criteria for patient risk reduction in biosimilar assessments.

Main Methods:

• Replaced randomized resampling with systematic enumeration of all reference data combinations.
• Evaluated three strategies for establishing stringent acceptance criteria: robust outlier detection, percentile-based method, and partitioning approach.
• Applied the enhanced methods to HDX-MS data from infliximab biosimilars and a NIST mAb.

Main Results:

• Systematic enumeration eliminated resampling-induced variation in acceptance limits.
• The enhanced methods correctly classified equivalent and nonequivalent samples.
• The improved strategy demonstrated suitability for regulatory comparability assessment.

Conclusions:

• The enhanced HDX-MS evaluation strategy provides a more robust and reliable method for biopharmaceutical comparability.
• Systematic enumeration and stringent criteria reduce patient risk by ensuring accurate similarity assessment.
• This refined approach is well-suited for regulatory decision-making in biopharmaceutical development.