Virus clearance was calculated by comparing the virus titers prior to and after the virus removal or inactivation step. Virus validation studies were conducted in the Pathogen Safety Department of Biotest, Dreieich, Germany. Virus validation studies addressed the worst case conditions (Fig. 1) assumed Z-VAD-FMK price to be low concentrations of alcohol in precipitation studies, reduced S/D concentrations in inactivation studies and elevated temperatures or high volumes per filter area in filtration studies. Studies per step and virus were performed at least in duplicate to demonstrate the reproducibility of results. Resuspended fraction II+III was spiked with test virus, fraction III was precipitated at assumed worst case

conditions, using 15% ethanol at −4 °C. Tests were performed at pH 5.15, pH 5.25 and at pH 5.45. The precipitated fraction III was removed by centrifugation and the supernatant was filtered through a depth filter and was tested for virus removal. Same was done without depth filtration. Precipitation of fraction III, centrifugation, filtration

and collection of filtrate was done at −4 °C. Fraction III supernatant (17% ethanol) was spiked with LY294002 mouse test virus and filtered at −4 °C through a Cuno SP depth filter (Cuno, Meriden CT, USA) and tested under worst case conditions, i.e. elevated volume per filter area of ∼500 kg/m2 and pH 5.10, beyond the lower limit of production scale and at pH 5.45, beyond upper limit. For S/D treatment, test material was spiked with test virus and treated at combined worst case conditions with a reduced concentration of 0.2% (w/w) TNBP and 0.8% (w/w) Triton X-100 and

at a reduced temperature of 25.5 °C. The concentrations of S/D are at the lower end of the production range. To prevent potential inactivation of viruses by low pH during S/D treatment, runs were performed at neutral pH 6.20. Eluate from Q Sepharose (GE Healthcare Bio-Sciences; Uppsala, Sweden) was spiked with test virus and passed through Galeterone a 75 nm prefilter followed by a 35 nm filter (Asahi Kasei Bioprocess; Brussels, Belgium) at worst case conditions, such as high temperature of ≤29.5 °C, high volume per filter area (150 kg/m2) and constant high pressure of ≥0.8 to ≤1.0 bar. Filtration was performed in dead end mode, where the feed is pumped directly against the filter membrane without any tangential flow. The conditions for validation of fraction III precipitation included lowering the ethanol concentration, shortening the precipitation time and increasing the temperature. The virus reductions observed at pH 5.10 are shown in Table 1 as a worst case condition. Validation of virus removal was tested at low pH of 5.10, reduced amount of filter aid and at a ratio of sample volume to filter area that was higher than the ratio used in manufacturing. Virus reductions were less than 1 log10 at every condition tested and none of the data were used to estimate total virus clearance.