Model-based analysis of a hydrophobic interaction chromatography for antibody-drug conjugate purification

Antibody-drug conjugates (ADCs) are a class of biopharmaceutical drugs designed as a targeted therapy for treating cancer while sparing healthy cells. ADCs are complex molecules composed of an antibody linked to a biologically active cytotoxic drug. The drug-to-antibody ratio (DAR), which represents the number of drugs conjugated to an antibody, is an important quality attribute of ADCs. The conjugation process typically yields a complex DAR profile, which requires further purification to remove undesired DAR species. Separation of DAR species post conjugation reaction can be achieved using hydrophobic interaction chromatography (HIC). HIC utilizes the hydrophobicity differences of different drug-loaded ADCs to separate them by reversible interaction between the proteins and the hydrophobic stationary phase. The buffer greatly influences the binding interaction between hydrophobic proteins and a HIC resin, but the process is also sensitive to variations in temperature, resin attributes, and the solid-liquid ratio of the column. A mechanistic model that captures these critical process parameters and material and column properties was established from calibration studies using multiple well-characterized HIC columns and resin lots. This model enabled in-silico characterization of the HIC process and facilitated understanding of unexpected process performance observations. It revealed the importance of controlling resin-specific hydrophobic capacity to minimize the DAR separation variations. It led to identifying an effective process control strategy to ensure a consistent DAR profile.