A modern reassessment of glycoprotein-specific direct platelet autoantibody testing in immune thrombocytopenia

Abstract

Platelet autoantibody (PA) testing has previously shown poor sensitivity for immune thrombocytopenia (ITP) diagnosis, but no previous study used both 2011 American Society of Hematology (ASH) guidelines for ITP diagnosis and 2012 International Society on Thrombosis and Haemostasis (ISTH) PA testing recommendations. We therefore performed a comprehensive retrospective study of PA testing in adult patients with ITP strictly applying these criteria. Of 986 PA assays performed, 485 assays in 368 patients met criteria and were included. Sensitivity and specificity of a positive test result for diagnosis of active ITP (n = 228 patients) were 90% and 78%, respectively. Sensitivity and specificity of a negative test result for clinical remission (n = 61 assays) were 87% and 91%. Antibodies against both glycoprotein IIb (GPIIb)/IIIa and GPIb/IX were required for the presence of antibodies against GPIa/IIa in patients with ITP. Logistic regression analysis revealed that more positive autoantibodies predicted more severe disease (relative to nonsevere ITP, relative risk ratio for severe ITP and refractory ITP was 2.27 [P < .001] and 3.09 [P < .001], respectively, per additional autoantibody); however, serologic testing did not meaningfully predict treatment response to glucocorticoids, intravenous immunoglobulin, or thrombopoietin receptor agonists. Sixty-four patients with ITP had multiple PA assays performed longitudinally: all 10 patients achieving remission converted from positive to negative serologic results, and evidence for epitope spreading was observed in 35% of patients with ongoing active disease. In conclusion, glycoprotein-specific direct PA testing performed using ISTH recommendations in patients meeting ASH diagnostic criteria is sensitive and specific for ITP diagnosis and reliably confirms clinical remission. More glycoproteins targeted by autoantibodies predicts for more severe disease.

Graphical abstract

Figure 1.

Flow diagram describing application of inclusion and exclusion criteria for the study. ^aWithin 4 days before PA assay. ^bWithin 30 days before PA assay. ^cPer ASH 2011 clinical practice guidelines. ^dOne of these 28 patients had 2 assays performed, and thus 29 assays were excluded according to this criterion. ACD-A, acid citrate dextrose solution A.

Figure 2.

Serologic patterns of platelet autoantibody assays in ITP patients. Results of PA assays in patients with ITP (360 assays total in 260 patients).

Figure 3.

Relation of glycoprotein-specific autoantibody number with disease severity. Multinomial logistic regression model for probability of disease severity based on number of glycoprotein-specific autoantibodies (error bars represent 95% confidence intervals). Model only included patients with active disease at the time of PA testing.

Figure 4.

Pattern of possible epitope spreading observed in patients with increasing numbers of glycoprotein-specific autoantibodies on interval testing. Median durations between assays given. White text delineates new autoantibodies present in subsequent assay not present in previous assay. Platelet count at the time of testing was similar at first test compared with repeat test (88 × 10⁹/L at first test and 71 × 10⁹/L at repeat test, P = .17 by paired Student t test).