Novel diagnostic assays for heparin-induced thrombocytopenia

Abstract

Laboratory testing for heparin-induced thrombocytopenia (HIT) has important shortcomings. Immunoassays fail to discriminate platelet-activating from nonpathogenic antibodies. Specific functional assays are impracticable due to the need for platelets and radioisotope. We describe 2 assays that may overcome these limitations. The KKO-inhibition test (KKO-I) measures the effect of plasma on binding of the HIT-like monoclonal antibody KKO to platelet factor 4 (PF4)/heparin. DT40-luciferase (DT40-luc) is a functional test comprised of a B-cell line expressing FcγRIIa coupled to a luciferase reporter. We compared these assays to polyspecific and immunoglobulin (Ig)G-specific PF4/heparin enzyme-linked immunosorbent assays (ELISAs) in samples from 58 patients with suspected HIT and circulating anti-PF4/heparin antibodies. HIT was defined as a 4Ts score ≥ 4 and positive (14)C-serotonin release assay. HIT-positive plasma demonstrated greater mean inhibition of KKO binding than HIT-negative plasma (78.9% vs 26.0%; P < .0001) and induced greater luciferase activity (3.14-fold basal vs 0.96-fold basal; P < .0001). The area under the receiver-operating characteristic curve was greater for KKO-I (0.93) than for the polyspecific (0.82; P = .020) and IgG-specific ELISA (0.76; P = .0044) and for DT40-luc (0.89) than for the IgG-specific ELISA (P = .046). KKO-I and DT40-luc showed better discrimination than 2 commercially available immunoassays, are simple to perform, and hold promise for improving the specificity and feasibility of HIT laboratory testing.

Figure 1

Assay results in HIT-negative and HIT-positive subjects. Results from HIT-negative and -positive subjects are shown for the polyspecific ELISA (A), IgG-specific ELISA (B), KKO-I (C), and DT40-luc (D). Solid horizontal lines represent mean values. Dashed horizontal lines represent the cutoff associated with the most northwest point on the receiver operating characteristic curve (ie, the cutoff at which sensitivity and specificity are optimized) for each assay.

Figure 2

ROC curves. ROC curves are shown for the polyspecific ELISA, IgG-specific ELISA, KKO-I, and DT40-luc. The AUCs for these assays were 0.82, 0.76, 0.93, and 0.89, respectively. The AUC for KKO-I was significantly greater than the AUC for the polyspecific (P = .020) and IgG-specific ELISAs (P = .0044). The AUC for DT40-luc was significantly greater than the AUC for the IgG-specific (P = .046) but not the polyspecific ELISA (P = .28).