Unique ELISA-Based Assays for VWF-GPIb Interactions and the Impact of Racial Differences on VWF Testing

Diagnosis of von Willebrand disease (VWD) relies primarily on assays of von Willebrand factor (VWF) function (VWF:RCo) and concentration of VWF protein (VWF:Ag). VWF:RCo is a surrogate measure of VWF activity for VWF interaction with the GPIb receptor on platelets. For VWD screening purposes, some have advocated that VWF:RCo is the most appropriate single test, but variability and reproducibility of VWF:RCo assays between laboratories has been problematic. Alternative assays have therefore been sought. Type 2 VWD variants, 2A, 2B, and 2M, are characterized by a discrepancy between the VWF:Ag and VWF:RCo, yet numerous factors can affect this ratio. Previously, our laboratory has reported a cell based assay to measure VWF interaction with the GPIb complex by flow cytometry, but such assay instrumentation is not available in hemostasis testing laboratories. Plasma samples were collected from 75 healthy donors enrolled in the TS Zimmerman Program for the Molecular and Clinical Biology of VWD, including 44 African American (AA) and 31 Caucasian subjects. VWF:Ag and VWF:RCo levels were performed in a central laboratory, as were collagen binding (VWF:CB) and propeptide (VWFpp) testing. Two ELISA-based assays were developed to measure VWF interaction with GPIb using recombinant GPIbα – one using normal GPIb with ristocetin (VWF:RCo ELISA) and the other a mutant form of GPIbα containing the platelet-type mutations D235Y and M239V that does not require ristocetin (VWF:IbCo ELISA). A monoclonal antibody is used to capture the rGPIb and to orient the GPIb for subsequent interaction with VWF. Serially diluted plasma samples were incubated for 1 hour with or without added ristocetin and monoclonal antibodies to VWF were used to detect the presence of VWF. Both assays utilized a 10 minute agitation at the end of the plasma incubation step. For all subjects, the mean VWF:Ag was 142, the mean VWF:RCo was 124, and the mean VWF:RCo/VWF:Ag ratio was 0.90. The two new assays yielded similar activity results when all the control subjects were analyzed, with a mean of 104 for the VWF:RCo ELISA and a mean of 108 for the VWF:IbCo ELISA. Both assays correlated well with each other and with the VWF:RCo. R squared values as determined by linear regression were 0.79 for the comparison of the VWF:RCo ELISA with the VWF:IbCo ELISA and 0.80 for comparison of either with the regular VWF:RCo. When the results were analyzed by race, however, a significant difference was seen for the two ristocetin-containing assays. The mean VWF:RCo/VWF:Ag ratio for the AA controls was 0.85, compared to 0.95 for the Caucasian controls (p<0.025). For the ristocetin ELISA, the mean was 0.54 for the AA controls and 0.79 for the Caucasian controls (p<0.001). However, no significant racial difference was seen for the VWF:IbCo ELISA with mean of 0.72 for the AA controls and 0.81 for the Caucasian controls (p=NS). Other VWF ratios have been proposed to be used to classify VWD – VWF:CB/VWF:Ag, FVIII/VWF:Ag, and VWFpp/VWF:Ag, but none were significantly different by race. The use of ELISA-based assays to determine VWF function is therefore feasible and may alleviate some of the problems inherent in the traditional VWF:RCo assay, including reproducibility and the technical demands of the assay. The VWF:IbCo assay may also eliminate racial differences in the VWF activity to antigen ratio, thus preventing the potential for erroneous diagnosis of VWD. Furthermore, the ELISA-based assays can be performed using standard hemostasis laboratory instrumentation.