Economic analysis of immunoadsorption systems

Abstract

Immunoadsorption possesses an attractive combination of selectivity and yield, and in some cases may replace many conventional purification steps with a single step. The economic performance of immunoadsorption is dependent on apparatus configuration (e.g. Sepharose® bed, perfusive bed and membrane), operating conditions (e.g. bed depth, flow rate and concentration of adsorbate in the effluent at breakthrough) and antibody binding kinetics. The solution space for the lowest operating cost is potentially very large and is prohibitive to exhaustively investigate in the laboratory. A mathematical model of a generic immunoadsorption process is useful in identifying the most promising configurations, operating conditions and antibodies, without having to resort to experiments that explore every possibility. The model system was chosen to represent the purification of a valuable biochemical product (1000 mg) from mammalian cell culture media or milk where the protein is present at low concentration (2 mg l−1). The antigen–antibody kinetic rate constants for a bovine serum albumin–monoclonal antibody system are available in the literature and were representative for this study. A modified chromatographic model was developed to predict the dimensionless operating cost of the generic immunoadsorption system subject to different apparatus configurations, operating conditions and antibody kinetics. Optimal operating conditions for each of the three configurations were identified. Under these conditions, the operating costs for a perfusive bed and a membrane were comparable and 28% lower than for a packed bed. The antibody binding kinetics significantly affected the dimensionless operating cost of the system and, therefore, the model can help select the most suitable antibody from a panel of prospective antibodies.