Background and objectives: There is still uncertainty over how the agent of variant Creutzfeld-Jakob disease (vCJD) would partition during the manufacture of plasma derivatives. In this study, a BSE-derived agent was used as a vCJD model to determine the extent to which infectivity could be removed by selected steps used in the manufacture of intravenous immunoglobulin (IVIG).
Materials and methods: Murine-passaged BSE (strain 301V), in the form of a microsomal fraction prepared from infected brain, was used to "spike" the starting material in three experiments. The partitioning of BSE infectivity was measured over Fraction I+III precipitation, borosilicate microfibre depth filtration and Seitz depth filtration, with these steps being examined individually and in series.
Results: Most 301V infectivity partitioned into Fraction I+III (log reduction 2.1). Infectivity remaining in Supernatant I+III was reduced by AP20 glass-fibre depth filtration (log reduction 0.6) and subsequently removed to below the limit of detection by Seitz KS80 depth filtration, giving an overall log reduction of > or = 2.9 for the three steps in series. By contrast, glass-fibre depth filtration gave a log reduction of 2.4 when challenged directly with "spiked" feedstock. Seitz KS80 depth filtration gave a log reduction of > or = 3.1 when challenged directly with 'spiked' feedstock and also removed residual infectivity to below the limit of detection when applied as the final step in series.
Conclusions: Results using a BSE-derived agent suggest that vCJD infectivity should be substantially removed from immunoglobulin G (IgG) solutions by Fraction I+III precipitation and Seitz KS80 depth filtration. The three different process steps examined acted in a complementary manner to one another when operated in series. However, the data demonstrated that it would be inappropriate to add together the reduction factors that had been derived for each step in isolation.