A comparative study in patients with type 2 von Willebrand disease using 4 different platelet-dependent von Willebrand factor assays

Background: Several assays are now available to evaluate platelet-dependent von Willebrand factor (VWF) activity. Objective: To report the results obtained using 4 different assays in patients with von Willebrand disease (VWD) carrying variants mainly in the A1 domain, which is critical for VWF binding to glycoprotein Ib (GPIb) and ristocetin. Methods: We evaluated 4 different assays, 2 gain-of-function mutant GPIb binding (VWF:GPIbM) and 2 ristocetin cofactor (VWF:RCo) assays, in 76 patients with type 2 VWD. Patients and healthy controls were tested using VWF:GPIbM enzyme-linked immunosorbent assay (ELISA), VWF:GPIbM automated, VWF:RCo aggregometric, and VWF:RCo automated assays. Results: There was a good correlation (Pearson's r>0.82) and agreement (Bland-Altman plots assessment) between the 4 assays, although several outliers existed among the type 2B without high-molecular-weight multimers (HMWM). The VWF activity/VWF:antigen ratios, calculated for each assay, were used to establish the percentage of a correct diagnosis of type 2 (ratio<0.60) in these patients: VWF:RCo aggregometric, 2A(100%), 2M(78%), 2M/2A(100%), 2B(68%); VWF:RCo automated, 2A(88%), 2M(89%), 2M/2A(100%), 2B(63%); VWF:GPIbM ELISA, 2A(96%), 2M(67%), 2M/2A(67%), 2B(0%); VWF:GPIbM automated, 2A(73%), 2M(44%), 2M/2A(75%), 2B(84%). In type 2B patients with HMWM, all assays gave a ratio ≥0.60. Conclusion: The VWF:GPIbM-automated assay is the most effective to diagnose as type 2 the 2B variants, whereas the VWF:RCo assays are the most effective in detecting 2M and 2M/2A variants. The VWF:GPIbM ELISA greatly overestimates the activity of the type 2B patients lacking HMWM. In this study, the use of a VWF activity/VWF:antigen ratio cut-off of 0.70 halved the number of misdiagnosed patients.