Adsorption of deamidated antibody variants on macroporous and dextran-grafted cation exchangers: II. Adsorption...

Single and multicomponent batch adsorption kinetics were obtained for deamidated mAb variants on two commercial cation exchangers, one with an open macroporous structure--UNOsphere S--and the other with charged dextran grafts--Capto S. The adsorption kinetics for the macroporous matrix was found to be controlled largely by pore diffusion. The effective diffusivity estimated from single component data was a fraction of the mAb free solution diffusivity, and its value could be used to accurately predict the adsorption kinetics for two- and three-component systems. In this case, when two or more variants were adsorbed simultaneously, both experimental and predicted results showed a temporary overshoot of the amount adsorbed above the equilibrium value for the more deamidated variant followed by a gradual approach to equilibrium. Adsorption rates on the dextran grafted material were much faster than those observed for the macroporous matrix for both single component and simultaneous adsorption cases. In this case, no significant overshoot was observed for the more deamidated forms. The Capto S adsorption kinetics could be described well by a diffusion model with an adsorbed phase driving force for single component adsorption and for the simultaneous adsorption of multiple variants. However, this model failed to predict the adsorption kinetics when more deamidated forms pre-adsorbed on the resin were displaced by less deamidated ones. In this case, the kinetics of the displacement process was much slower indicating that the pre-adsorbed components severely hindered transport of the more strongly bound variants. Overall, the results indicate that despite the lower capacity, the macroporous resin may be more efficient in process applications where displacement of one variant by another takes place as a result of the faster and more predictable kinetics.