We have assayed two different monoclonal-antibody-purified concentrates (A and B) and one conventional concentrate (C), against the 3rd International Standard for factor VIII concentrate, using one-stage, two-stage and chromogenic methods. One-stage assays performed with immunodepleted plasmas gave lower potencies than with haemophilic plasma for all concentrates, though the discrepancies were most marked for the two highly purified products. The absence of von Willebrand factor in one of the immunodepleted plasmas appeared to contribute towards the low potencies observed. In addition, potencies of product A were 50% higher by one-stage assays (haemophilic plasma) than by two-stage or chromogenic methods. These results indicate the need for careful evaluation of assay methodologies for assessment of factor VIII:C activity in highly purified concentrates.

A chromogenic factor Xa generation method has been developed for comparing the co-factor activity of factor VIII concentrates at physiological factor VIII concentrations (1 iu/ml). In the presence of thrombin all concentrates gave similar rapid rates of factor Xa generation, but in the absence of thrombin there were major differences between the rates of Xa generation between different products. High purity products, particularly those prepared by monoclonal antibody purification from plasma and recombinant sources, gave more rapid Xa generation than most intermediate-purity products. There was a very strong correlation between the rate of Xa generation and the difference in factor VIII potency by one-stage and two-stage assays. These results suggest the possible presence of small amounts of activated factor VIII in some concentrates, but differences in von Willebrand factor content could also contribute towards the different rates of factor Xa generation observed.

Summary The inhibitory effect of factor VIII concentrate products on IL‐2 secretion by human T‐cells was investigated. The six products used widely in the lJ.K. showed very different activities varying from almost total inhibition to no significant effect. There appeared to be no obvious relationship between inhibitory activity and protein composition but factor VIII itself was not responsible for the effect as affinity purified products were entirely non‐inhibitory. The two wetheated products were most inhibitory whereas dry‐heated products were less inhibitory or non‐inhibitory. However, a wet‐heated version of a non‐inhibitory dry‐heated product was also non‐inhibitory, suggesting that the composition of the concentrate rather than anti‐viral treatment is important for immunosuppressive activity. A product treated by the solvent/detergent procedure showed considerable inhibitory activity. Immunoglobulin and albumin products did not inhibit IL‐2 secretion to any significant extent, but factor IX concentrates were inhibitory. We suggest that inhibition of IL‐2 secretion by factor VIII concentrates may be related to the immunosuppression observed in haemophiliacs treated with high dose factor VIII products and that our results should be considered by clinicians and manufacturers of factor VIII products.

The molecular-weight distribution of proteins in factor VIII concentrates has been analysed by fast-protein liquid chromatography. The proportion of high-molecular-weight (HMW) aggregates in one product increased on freeze-drying and heating, with fibrinogen and fibronectin being the main protein components of the HMW peak. In all other concentrates, the HMW peak was less than or equal to 5% of the total protein content and there were no differences in HMW content according to purity or method of viral inactivation.