A “>” preceding the logarithmic reduction factor (R) combined with a “no” result for the parameter “virus breakthrough” indicates removal below the limit of detection for all of the filtrate fractions.

If minimal virus infectivity was found in a minority of the filtrate fractions (i.e., the titers were at the limit of detection) then the result for virus-breakthrough is “minimal”.

The early 2^nd generation down-stream filters investigated (filters SG I and SG II – see tables 2 and 3), however, were found to have robust virus reduction capacities even under worst case load and duration conditions. Virus reduction was found to be effective for both filters (R ranging from 5.0 log₁₀ to > 6.5 log₁₀) although some minimal virus break-through was found in several fractions. Variation of the TMP had no impact on the virus clearance capacity of either of the filters – although it did impact the process feed loads possible before reaching any of the decision points.

Both filters are deemed to be equally suitable for the high-volume long duration filtration which would be needed for a perfusion reactor – and thus also for any smaller volume set-up, including batch filtrations.

It is noted that biochemical parameters, i.e., glucose concentration and osmolality pre and post filtration, confirmed that virus filtration had no significant impact on the composition of the fermenter feed media. Furthermore, the use test regarding cell culture performance with the filtered media showed cell density, cell viability and cell productivity (rate of expression of the target protein) to be within specifications, and fully equivalent to that of unfiltered media (data not shown).

Table 4. Filtration conditions and results for MMV retention using improved second generation filters SG III to SG VI.