Beneficial Effect of Exogenous PF4 on SRA for Detecting Pathogenic HIT Antibodies

Introduction: The laboratory diagnosis of HIT is difficult and often requires to perform an enzyme immunoassay combined with a functional assay, either serotonin release assay (SRA) or heparin-induced platelet aggregation (HIPA), for detecting IgG antibodies that activate platelets in the presence of heparin (anti-PF4/H antibodies). SRA is today considered as the gold standard for HIT diagnosis, but its performances could be improved as suggested by a recent clinical history and subsequent experiments.

Case report: A 74 year-old man was hospitalized for a myeloma and treated with unfractionated heparin (UFH) for right atrium thrombosis. Thrombocytopenia and pulmonary embolism occurred the first day of heparin treatment and HIT was suspected. UFH was replaced by danaparoid sodium, and correction of platelet count was achieved within 3 days. Laboratory tests revealed the presence of significant levels of IgG to PF4/H (OD 1.2) with clearly positive HIPA. However, SRA was also performed to confirm the diagnosis of early onset HIT, but remained negative after testing platelets from two different donors. Unlike HIPA performed with PRP, SRA is done with washed platelets, and we thus hypothesized that PF4 concentration on cell surfaces might be too low for allowing heparin-dependent platelet activation when the patient's sample was tested. We thus performed a new SRA after adding exogenous PF4 (10 μg/ml) in the buffer and positive platelet activation pattern was obtained with a maximal release of 65% with 0.5 IU/ml UFH. This experience therefore prompted us to further evaluate whether addition of exogenous PF4 was beneficial or not on platelet activation induced by PF4/H antibodies when studied in SRA.

Patients and methods: Plasma samples from 45 patients with suspected HIT and significant levels PF4-specific IgG were studied. SRA was systematically performed without and with exogenous PF4 (10 μg/ml). This "optimal" concentration of exogenous PF4 was defined after evaluating the effects of variable concentrations of human PF4 on platelet activation induced by 5B9 a recently developed HIT monoclonal antibody.

Results: SRA was consistently negative when 5B9 was tested at concentrations lower than 5 μg/ml (1 or 2.5 μg/ml). In contrast, SRA was positive with these low 5B9 concentrations when 10 μg/ml of PF4 were present in the platelet buffer. When samples from suspected HIT were studied, conventional SRA (without PF4) was positive in 14 of 45 cases (Group SRA+), which all contained relatively high levels of PF4-specific IgG (median OD = 2.6). After addition of exogenous PF4, SRA was positive in 8 additional cases (Group SRA_PF4+), with maximal release between 32% and 64% and a complete inhibition in the presence of 10 IU/ml UFH. Interestingly, levels of PF4-specific IgG were similar in this group and the 23 samples with whom SRA remained negative (median OD 1.6 vs. 1.3 respectively, p = 0.8). On the other hand, the pre-test probability of HIT (4Ts score) was always intermediate or high in 'SRA+' and 'SRA_PF4+' groups, although it was low in more than 30% of patients with negative SRA.

Discussion: Addition of PF4 in the platelet buffer might improve the detection of pathogenic HIT antibodies using SRA. Our results are in agreement with a recent study (I Nazi et al, Journal of Thrombosis and Haemostasis, 2015; 13: 1900-7), although they were obtained with a lower concentration of exogenous PF4, but which may be observed in vivo.