3.2.2. Inline spiking viral clearance test for integrated
flow-through two column and virus filter
To separately examine the virus removability of the virus filter, an
inline spiking viral clearance test for the Planova BioEX filter
implemented in the flow-through two column chromatography was carried
out. Due to clogging from impurities in the virus spike, the flow rate
was reduced to produce a flux of 40 LMH to the filter (compared to 76
LMH for the end-to-end continuous process) to avoid exceeding the
operating pressure limit of Planova BioEX (0.35 MPa) before reaching the
target throughput. It should be noted that viral clearance determined
for the lower flux process (40 LMH) can be taken as confirming viral
clearance at 76 LMH, as lower flux is a worse case condition for virus
filtration.
As shown in Figure 5, a feed solution of 10 mg/mL mAb without virus
spike was supplied to the two columns and the same mAb spiked at 10%
MVM or X-MuLV was supplied through inline spiking. The pressure
increases on the filter were observed until the peak after the process
pause for the loading of mAb solution spiked with MVM or X-MuLV compared
to the minimal pressure increase observed for loading of mAb solution
without spike. The pressure profile of the virus-spiked mAb filtration
shows fluctuations that are considered to be due to pulsation of the
pump due to air bubbles. The pressure increases with increasing load
volume are greater for the filtration of the X-MuLV spiked solution and
approach the upper operating pressure limit of the filter at the target
throughput. The more stable filtration pressure of the un-spiked mAb is
similar to the pressure profile for the filter incorporated in the
end-to-end continuous process shown in Figure 3.
Due to the concern that virus breakthrough to permeate may occur due to
the pressure drop during a process pause as previously described (Ajayi
et al., 2022; Fan et al., 2021; Johnson et al., 2021), a 60 min process
pause was included after loading the virus-spiked mAb, and virus
removability for inline spiking of MVM or X-MuLV was evaluated for both
the flow-through fraction of the mAb and the flow-through fraction with
the buffer wash. As shown in Table 2 for mAb spiked with MVM or X-MuLV,
complete clearance was achieved for both the flow-through fraction and
the flow-through plus wash with MVM LRV of ≥ 5 and X-MuLV LRV of ≥ 3.
The relatively lower X-MuLV LRV for the filter than the two column
process can be attributed to the lower X-MuLV spike titer. These results
confirm that the Planova BioEX filter has high virus removability for
both MVM and X-MuLV and that the process pause did not affect virus
removal.