Potency measurement of factor VIII in recombinant concentrates

In preliminary studies on the stability of recombinant factor VIII (rFVIII) concentrates post reconstitution, a rise in potency to 200% of labelled values was observed in concentrates stored at 22 °C over 24 h. This was observed in potency estimates by the one-stage clotting, but not the two-stage clotting or chromogenic assays, and was not observed for intermediate-purity product derived from plasma (IPVIII). Use of human serum albumin (HSA), rather than the usual bovine material (BSA), to dilute product for the stability study abolished the rise in potency. Incorporating purified von Willebrand Factor (VWF) in the diluent buffer abolished the rise observed in BSA. A similar rise was observed upon incubating rFVIII in the presence of 10(-4) u of thrombin per mL in HSA buffer. Potency estimates using the HSA in the dilution buffer resulted in severe underpotency in relation to the label claim when using the two-stage clotting and the chromogenic assays, but not the one-stage clotting assay. Predilution in severe haemophilic plasma restored potency levels to those claimed. We conclude that (i) commercial preparations of BSA may be unsuitable for inclusion in buffers for rFVIII studies; (ii) FVIII in rFVIII concentrates is exquisitely sensitive to activation by thrombin, presumably as a result of the lack of VWF; (iii) accurate potency estimation in the two-stage assay systems requires VWF in the assay system.