Evaluation of assays to measure ADAMTS13 activity in patients' plasma

Several assays have been developed to measure ADAMTS-13 activity. Although the performance characteristics of these methods were recently evaluated (Tripodi A, et al. JTH 2004), little information is currently available on concordance of these assays in the measurement of plasma ADAMTS-13. The main goal of this study was to investigate whether widely available assays concordantly measure the same amount of ADAMTS-13 activity in plasma. ADAMTS-13 activity using residual collagen binding activity (CBA) and FRET assays were measured in plasma samples of 72 healthy subjects and 124 patients affected with thrombotic microangiopathies. The Spearman’s rho and the Kappa statistics were used to assess the correlation and agreement between assays after categorisation of plasma samples into 5 subgroups with different levels of ADAMTS-13 activity. The measurement of ADAMTS-13 activity by CBA and FRET assays showed a correlation of 0.85 (P<0.0001). The concordance between the two assays was confirmed in 81% of samples (Kappa=0.68, P<0.0001). Major discrepancies were observed in 10 samples: ADAMTS-13 activity was consistently lower as measured by FRET assay compared to CBA (Table I, grey area). Only 3 of these discordant data could be ascribed to plasmatic hyperbilirubinemia (Meyer SC, et al. JTH 2007). Additionally, in other three patients with severe haemolysis, ADAMTS-13 activity was not measurable using the FRET assay due to a non-parallelism of serially diluted plasma samples, in contrast to the consistent CBA measurement. In conclusion ADAMTS-13 activity determined by CBA and FRET assays confirmed an overall good concordance. The major advantage of FRET assay compared with CBA is that the results are available within 1h. However, the disadvantage is related to the use of fluorescent probes, which may be influenced by plasma factors such as bilirubin interfering to the fluorescence emission and consequently to the results of FRET assay