Thromb Haemost 2021; 121(12): 1558-1561
DOI: 10.1055/a-1585-6264
Invited Editorial Focus

Heparin-Induced Thrombocytopenia and Vaccine-Induced Immune Thrombotic Thrombocytopenia Antibodies: Fraternal—Not Identical—Twins

Theodore E. Warkentin
1  Department of Pathology and Molecular Medicine, Michael G. DeGroote School of Medicine, McMaster University, Hamilton, Ontario, Canada
2  Department of Medicine, Michael G. DeGroote School of Medicine, McMaster University, Hamilton, Ontario, Canada
3  Transfusion Medicine, Hamilton Regional Laboratory Medicine Program, Hamilton, Ontario, Canada
4  Service of Benign Hematology, Hamilton General Hospital (Hamilton Health Sciences), Hamilton, Ontario, Canada
5  McMaster Centre for Transfusion Research, McMaster University, Hamilton, Ontario, Canada
› Author Affiliations

PF4-Dependent Immunoassays in Patients with Vaccine-Induced Immune Thrombotic Thrombocytopenia: Results of an Interlaboratory Compariso

A major development during the fight against the global pandemic caused by SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) has been the extraordinarily rapid emergence of effective vaccines that elicit an immune response against the “spike” protein of this novel coronavirus.[1] [2] Indeed, vaccines have been central to management of this pandemic.[3] Remarkably, a small proportion of otherwise (mostly) healthy vaccine recipients develop a prothrombotic thrombocytopenic syndrome—vaccine-induced immune thrombotic thrombocytopenia (VITT)—that strongly mimics immune heparin-induced thrombocytopenia (HIT) on clinical, laboratory, and serological grounds.[4] [5] [6] [7]

As shown in a multi-laboratory study headed by Professor Ulrich Sachs, and reported in this issue of Thrombosis and Haemostasis,[8] while there are many serological similarities between VITT and HIT antibodies, they are not identical. In essence, VITT and HIT antibodies (particularly, “autoimmune” HIT antibodies[9]) are fraternal, not identical, twins that despite sharing the key unifying property of anti-platelet factor 4 (PF4) reactivity nonetheless bind to different regions on this target protein.

What insights gained from 50 years of research into HIT can be applied to 5 months of experience evaluating VITT? [Table 1] lists three fundamental tenets of HIT serology.[10] [11] [12] [13] [14] [15]

Table 1

Fundamental tenets of HIT serology

Law #1

Patient serum/plasma always contain antibodies that recognize PF4 in standard solid-phase PF4/polyanion immunoassays, i.e., ELISAs (sensitivity >99%[a]).

Law #2

Pathogenic anti-PF4 antibodies that cause HIT have platelet-activating properties (i.e., are detectable in platelet activation assays); PF4 supplementation may be required for optimal sensitivity (overall sensitivity >99%).

Law #3

Rapid PF4-dependent immunoassays have high sensitivity (at least 85–90%) but generally not >99% sensitivity as seen with ELISAs.

Abbreviations: ELISA, enzyme-linked immunosorbent assay; PF4, platelet factor 4; HIT, heparin-induced thrombocytopenia.

a Although occasional HIT sera will test negative in a particular ELISA, they will generally test positive in another PF4-dependent ELISA.

Many clinicians do not recognize how special these first two fundamental tenets of HIT serology are. After all, other platelet immunology disorders—(auto)immune thrombocytopenia (ITP), drug-induced ITP, neonatal alloimmune thrombocytopenia—often do not have readily detectable antibodies in patient serum/plasma (indeed, poor sensitivity of platelet antibody assays in ITP makes this disorder a diagnosis of exclusion). In contrast, HIT is defined as a “clinical–pathological” disorder because the key pathological criterion—detectability of serum/plasma HIT antibodies—is fundamental to diagnosis.[16] The fact that patient serum/plasma contains detectable HIT antibodies is probably crucial to explaining HIT pathogenesis, namely that free plasma antibodies are required to assemble over time the in situ multimolecular complexes (multivalent PF4/polyanion/immunoglobulin G [IgG]) on platelet surfaces required to engage the platelet FcγIIa receptors, and thus to trigger the platelet activation response that underlies HIT.[10]

A striking feature of VITT is its triggering by one of two adenoviral vector vaccines (ChAdOx1 CoV-19 [AstraZeneca] and Ad26.COV2.S [Johnson & Johnson/Janssen]),[4] [5] [6] [7] [17] neither of which contains heparin. In as many as 1 or 2 individuals among 100,000 so vaccinated,[18] approximately 1 week later there can emerge this disorder characterized by one or more unusual and oftentimes severe thrombotic events with marked thrombocytopenia, hypercoagulability (elevated D-dimers, reduced fibrinogen), and—bizarrely—positive testing in standard PF4-dependent enzyme-linked immunosorbent assays (ELISAs) and in functional (platelet activation) assays.[4] [5] [8] Thus, VITT strongly resembles HIT—particularly the severe variant known as “autoimmune HIT” (where antibodies strongly activate platelets even in the absence of heparin).[9] Indeed, it was this close clinical resemblance to HIT that led to the speedy identification of pathogenic anti-PF4 antibodies playing a fundamental pathogenic role in VITT. However, as Sachs et al have shown, VITT antibodies do not precisely mimic HIT antibodies in their reactivity profile.

To perform their studies, Sachs and colleagues identified sera from 12 patients with clinical VITT where two laboratories independently demonstrated PF4-dependent platelet-activating properties in patient serum. Thus, they had high confidence that these sera did indeed contain VITT antibodies. They then had four independent laboratories perform various immunoassays—two commercial ELISAs as well as three different rapid assays—to determine their reactivity profiles. The three rapid assays were the particle gel immunoassay (PaGIA; Bio-Rad, Hercules, California, United States), the lateral flow assay (LFA; Milenia Biotec; Giessen, Germany), and the chemiluminescence immunoassay (CLIA; Instrumentation Laboratory, Bedford, Massachusetts, United States). These three immunoassays are all known to have high sensitivity for detecting pathologic HIT antibodies (at least 90%).[12] [13] [14] [19] [20]

What did they find? Both commercial ELISAs showed high sensitivity for VITT antibodies: the IgG-specific PF4/polyvinylsulfonate ELISA (Immucor, Waukesha, United States) tested positive with all 12 sera, with generally strong-positive results (mean optical density: 2.7 units [reference range: <0.40]). A second IgG-specific ELISA (Zymutest HIA, Hyphen Biomed, Neuville-sur-Oise, France), which uses heparin to bind proteins (including PF4) contained within platelet lysate, thereby forming PF4/heparin (and, theoretically, other heparin-dependent antigens), yielded positive results in 11/12 sera tested. (A third ELISA, the IgG-specific PF4/heparin ELISA from Stago—not reported in their paper, but subsequently studied, with data presented on July 17, 2021 at the Scientific and Standardization Committee meeting of the International Society on Thrombosis and Haemostasis—also was positive in 10/11 patient sera tested.) Thus, as is known to be the case with HIT, VITT antibodies are detected with high sensitivity in solid-phase ELISAs that utilize standard plastic wells.

In contrast, all three rapid assays evaluated showed poor sensitivity for detecting VITT antibodies. The PaGIA yielded positive results in only 3/12 (25%) samples tested, the LFA in only 1/12 (8%) samples studied, and most strikingly, the CLIA was positive in none (0%) of 10 samples tested. Given their aforementioned high sensitivity in HIT, this insensitivity of the rapid HIT assays for VITT antibody detection is striking. Indeed, Sachs and colleagues even proposed that an ELISA-positive/CLIA-negative reaction profile, in the appropriate clinical context, could be viewed as laboratory “confirmation” of VITT! Although the authors did not evaluate a fourth rapid assay, the latex immunoturbidimetric assay, others have noted its lack of sensitivity too for detecting VITT antibodies.[5] [17] [21] [22]

Thus, while HIT and VITT share remarkably similar clinical, laboratory, and serological features, they are not identical ([Fig. 1]). HIT and VITT can be conceptualized as fraternal, not identical, twins; they belong to the same family of platelet-activating anti-PF4 immune responses, but with different target antigen regions on PF4. Indeed, a recent study[23] of PF4 epitope mapping has identified a distinct region recognized by VITT antibodies that corresponds to the heparin-binding site on PF4; in contrast, HIT antibodies recognize separate and distinct sites on PF4, although tellingly autoimmune HIT sera with heparin-independent platelet-activating properties also recognize the VITT epitope site. Presumably, rapid assays for HIT are designed to detect antibodies that recognize heparin-dependent epitopes on PF4, and so they fail to detect VITT antibodies directed at a highly localized epitope on PF4 which supports heparin-independent platelet activation by these highly pathogenic antibodies.

Zoom Image
Fig. 1 Serological similarities and differences between HIT and VITT antibodies. The concept is that whereas (PF4-supplemented) washed platelet activation assays and PF4-dependent ELISAs show high sensitivity for both HIT and VITT antibodies, markedly different reaction profiles are seen with rapid HIT assays. PIPA here refers to the classic HIPA test performed with supplemental PF4. CLIA, chemiluminescence immunoassay; ELISA, enzyme-linked immunosorbent assay; HIPA, heparin-induced platelet activation assay; HIT, heparin-induced thrombocytopenia; LFA, lateral flow assay; LIA, latex-enhanced immunoassay (latex immunoturbidimetric assay); PaGIA, particle gel immunoassay; PF4, platelet factor 4; PF4-SRA, PF4-dependent serotonin-release assay; PIPA, PF4-induced platelet activation assay; SRA, serotonin-release assay; VITT, vaccine-induced immune thrombotic thrombocytopenia.

Publication History

Received: 09 August 2021

Accepted: 10 August 2021

Publication Date:
12 August 2021 (online)

© 2021. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany