Isotherm model discrimination for multimodal chromatography using mechanistic models derived from high-throughput...

Area of Science:

• Chromatography
• Bioseparation
• Chemical Engineering

Background:

• Multimodal chromatography is crucial for bioseparations, but accurate predictive modeling remains challenging.
• Isotherm formalisms are essential for describing the equilibrium binding of molecules to chromatography resins.
• Understanding and predicting elution behavior under various conditions is key for process optimization.

Purpose of the Study:

• To evaluate and compare the predictive capabilities of various isotherm models for multimodal chromatography.
• To characterize isotherm models based on their complexity and accuracy in predicting resin behavior.
• To identify the most suitable isotherm formalism for predicting the performance of multimodal cation and anion exchange resins.

Main Methods:

• Examined multiple isotherm models based on the stoichiometric displacement framework.
• Determined isotherm parameters by fitting to mAb - Capto MMC and mAb - Capto Adhere batch data.
• Assessed predictive ability using salt gradient elution and gradient slope behavior.
• Quantified model performance using peak characteristics agreement and fit accuracy scores.
• Utilized Akaike Information Criterion (AIC) to evaluate model complexity.

Main Results:

• The extended steric mass action (SMA) isotherm demonstrated superior predictive performance for both Capto MMC and Capto Adhere resins.
• This model accurately captured pH-dependent elution for Capto MMC and yield losses for Capto Adhere.
• Model performance for Capto MMC improved with exclusion of low protein concentration data, unlike Capto Adhere.
• The extended SMA isotherm achieved high accuracy with a minimal parameter set, outperforming more complex models.
• Protein-salt activity coefficient was critical for predicting Capto Adhere behavior, influencing binding and elution.

Conclusions:

• The extended SMA isotherm is the most effective formalism for accurately predicting the behavior of both multimodal cation and anion exchange resins.
• This model provides a balance of high predictive accuracy and reduced complexity, simplifying chromatographic process development.
• The study identified key isotherm parameters and their influence on binding and elution, enhancing understanding of multimodal chromatography interactions.