We investigated the effect of antibody solution conditions (ionic strength, pH, IgG concentration, buffer composition, and aggregate level (dimer content)) on filter performance for a virus removal filtration process using the Planova 20N, a virus removal filter. Ionic strength and pH affected the filter flux. A consistent high flux was maintained at an ionic strength greater than 10 mM and at pH 4-8 under a typical buffer composition (sodium chloride, citrate, acetate, and phosphate). Optimum IgG concentration was 10-20 mg/mL allowing for high throughput (kg/m(2) of IgG). Dimer content negligibly affected the flux level. Under high throughput conditions, virus spiking did not affect flux whereas a parvovirus logarithmic reduction value greater than 5 was maintained. From the results of zeta potential analyses for IgG and the membrane, we considered that electrostatic interactions between antibodies and the membrane affect filter performance (flux level and throughput). These results indicate that the Planova 20N filter is applicable for a wide range of solution conditions typically used in antibody processing.

Confirmation of virus filter integrity is crucial for ensuring the safety of biological products. Two main types of virus filter defects may produce inconsistent and undesirable performance in virus removal: improper pore-size distribution across the membrane; and specific damage, such as tears, broken fibers, or pinholes. Two integrity tests are performed on each individual filter manufactured by Asahi Kasei Medical to ensure the absence of these defects prior to shipment. In this study, we verified that typical usage of Planova™ BioEX filters would not improperly shift the pore-size distribution. Damage occurring during shipment and use (e.g., broken fibers or pinholes) can be detected by end-users with sufficient sensitivity using air-water diffusion based leakage tests. We prepared and tested filters with model pinhole defects of various sizes to develop standard acceptance criteria for the leakage test relative to porcine parvovirus infectivity logarithmic reduction values (LRVs). Our results demonstrate that pinhole defects at or below a certain size for each effective filter surface area have no significant impact on the virus LRV. In conclusion the leakage test is sufficiently sensitive to serve as the sole end-user integrity test for Planova™ BioEX filters, facilitating their use in biopharmaceuticals manufacturing.