https://orcid.org/0000-0001-6445-582X
This editorial refers to ‘Predictors of mortality in thrombotic thrombocytopenia after adenoviral COVID-19 vaccination: the FAPIC score’, by J. Hwang et al., https://doi.org/10.1093/eurheartj/ehab592 and ‘Immune complexes, innate immunity, and NETosis in ChAdOx1 vaccine-induced thrombocytopenia’, by S. Holm et al., https://doi.org/10.1093/eurheartj/ehab506.
Clinical and laboratory characteristics of VITT are variable. Onset of symptom is usually 4 days or more after vaccination. The latest date post-vaccine that cases have been reported was initially 25–30 days; however; as this syndrome evolves, cases associated with longer elapsed time may be identified. Patients who present early after vaccination may have not yet developed thrombosis and may have less severe thrombocytopenia or other laboratory abnormalities. The FAPIC score is newly developed; risks for mortality based on total score may change as data from larger numbers of VITT cases are incorporated in the model. At the vascular level, multiple thromboinflammatory pathways are activated in VITT, resulting in the characteristic findings of this clinical syndrome. CVST, cerebral venous sinus thrombosis; ICH, intracranial haemorrhage; FAPIC Score—fibrinogen, age, platelet count, ICH, and CVT; PF4, platelet factor 4; ELISA, enzyme-linked immunosorbent assay.
Severe acute respiratory syndrome coronavirus 2 (SARS CoV-2) has surprised us from the beginning. Initially thought to be confined to an isolated location as with previous SARS outbreaks, it instead relentlessly marched across the world, infecting >183.9 million people.1 In a profound testament to scientific progress, achievement, and collaboration, vaccines were rapidly developed to combat this pandemic. New vaccine strategies used mRNA sequences or adenoviral vectors containing DNA sequences to deliver the coding sequences of the spike protein and harnessed the recipient’s own cells to translate this code into spike protein to stimulate an immune response. Randomized blinded placebo-controlled vaccine trials enrolled >97 805 participants and demonstrated efficacy.2–5 Emergency use authorization enabled these vaccines to be used in early December 2020—just less than 1 year after the first reported cases of COVID-19. These vaccines have demonstrated ability to protect from infection and severity of COVID-19.6 , 7 Yet, as with other unexpected turns associated with COVID-19, rare but severe complications have developed in some vaccine recipients.
Reports from Europe following the early rollout of the ChAdOx1 nCoV-19 adenoviral vector vaccine noted patients with thrombosis in unusual locations, especially the cerebral venous sinuses [cerebral venous sinus thrombosis (CVST)], accompanied by a set of specific clinical and lab findings including short duration of time from vaccination, thrombocytopenia, and elevated D-dimer levels (Graphical Abstract). Astute clinicians recognized the similarity to heparin-induced thrombocytopenia (HIT) when patients treated with heparin developed worsening clinical status. Assays used to diagnose HIT—platelet factor 4 (PF4) enzyme-linked immunosorbent assay (ELISA) and platelet activation assays—yielded surprisingly positive results with higher values than typically seen with HIT, in patients without heparin exposure.8–10 Vaccine-induced immune thrombotic thrombocytopenia (VITT), the name most frequently used for this syndrome, appears to be similar to autoimmune HIT.11 Immunoglobulin G antibodies directed against PF4 complexed with an unknown polyanion have been demonstrated to induce platelet activation in patients with VITT by binding to the IgG receptor on platelets, with subsequent triggering of thrombosis and the resultant clinical features of VITT.8–10 Although direct causality has not been demonstrated, many regulatory agencies have concluded that the link to the vaccine is stronger than initially believed. A recent review of cases of CVST prior to the emergence of COVID-19 found a low rate of associated thrombocytopenia in 8.4% of 865 patients, with 0.5% having a platelet count <50 000/μL, with no PF4 positivity in those that were tested, implying that the VITT findings of thrombosis and thrombocytopenia are related to vaccine use.12
In this issue of the European Heart Journal, work by Hwang13 and by Holm14 and their respective colleagues examine important clinical and laboratory findings in patients with confirmed VITT to better understand factors contributing to this disorder.
Hwang and colleagues performed a rapid systematic review with pooled analysis of reported cases of VITT following vaccination with the ChAdOx1 nCoV-19 vaccine and developed a scoring system to predict mortality. Forty-nine published cases through to 28 April 2021 with sufficient clinical information were selected according to standard guidelines. From these cases, five variables were found to be associated with mortality—age <60, platelet count <25 000/μL, fibrinogen <150 mg/dL, intracranial haemorrhage, and CVST—with a scoring system constructed assigning 1 point for each variable, predicting risk of death based on the total number of points. Variables in the FAPIC score (fibrinogen, age, platelet count, ICH, and CVST) are additive, with the presence of all five associated with 100% mortality in this 49 patient analysis (Graphical Abstract). Evaluation of the effect of treatment on mortality was hindered as treatments were reported for only half of the patients.
An in-depth assessment of the pathological immune and thrombotic responses by Holm and colleagues found marked differences between five patients from Denmark diagnosed with severe VITT following ChAdOx1 nCoV-19 vaccination compared with vaccinated and non-vaccinated healthy controls. This comprehensive analysis found activation of multiple components of the inflammatory response, with elevations of innate immune response cytokines and markers of endothelial cell damage. Examination of leucocyte phenotypes demonstrated an increase in circulating immature neutrophils with striking evidence of excessive neutrophil activation, degranulation, and formation of neutrophil extracellular traps (NETs) seen in those with VITT but not in those who were vaccinated without development of VITT. Dissection of the specific components of the formed immune complexes in patient sera identified the expected IgG and PF4, but also ligands for innate immune response pathways including receptors for the Fc portion of IgG, PF4 receptors, complement receptors, CD14, a lipopolysaccharide-binding protein made by macrophages that functions as an endotoxin receptor, and platelet glycoprotein Ib. In addition, proteoglycan 4, containing glycosaminoglycan (GAG) chondroitin sulfate and keratan sulfate polyanions, was also found, which the authors suggest might serve as the PF4 polyanion ligand. Most striking were the findings of the composition of an extracted cerebral vein clot, with platelets, IgG, and an excess presence of neutrophils and NETs; absent was solid evidence of the presence of SARS-CoV-2 spike protein (Graphical Abstract).
These studies in this issue of the journal evaluated patients with severe manifestations of VITT identified when this syndrome was just coming to light. The development of CVST and splanchnic vein thrombosis, rare and unusual locations for thrombosis, probably aided recognition earlier than if deep vein thrombosis (DVT) or pulmonary embolism (PE) were presenting findings. Increased reports of VITT following ChAdOx1 nCoV-19 vaccination emerged as recognition of the syndrome increased, and cases following administration of the Ad26.COV2.S vaccine, which also uses an adenoviral vector strategy, were reported.15 , 16 Descriptions of these findings were quickly circulated among the scientific and clinical community, resulting in earlier diagnosis and use of appropriate treatment, leading to improved outcomes and decreased mortality.17 By the end of June 2021, a total of 395 cases of VITT associated with the ChAdOx1 nCoV-19 vaccine in the UK had been reported after 45.2 million administered doses, while 38 cases following the Ad26.COV2.S vaccine have been reported in the USA after 12.3 million doses.18 , 19 Case fatality rates in the UK are now reported at 18%, lower than with the early cases studied by Holm and colleagues in which three of five patients died, or the 41.7% mortality in the Hwang analyses.
The FAPIC score developed by Hwang and colleagues is derived from early severe cases with high mortality. Age is in the opposite direction from what we usually expect, highlighting the concern that VITT is seen primarily in young patients; the authors note that younger patients tended to have more severe clinical characteristics. Sex as a variable was not found to be associated with mortality in this model, somewhat surprisingly given the perception that females are more likely to develop VITT based on early case reports. Whether the association of age with mortality is due to demographic selection of those vaccinated with adenoviral vector vaccines remains unclear at this time. Further development of this predictive score with a larger number of cases spanning the severity of VITT and treated with therapy that does not exacerbate the underlying pathophysiology of VITT is needed to validate its clinical utility and predictive ability.
The findings of Holm and colleagues suggest that excessive activation of neutrophils in VITT plays as significant a role in thrombus development as activation of platelets. As with COVID-19, thromboinflammation—the cross-talk between innate inflammatory responses and activation of coagulation—appears to play a key role in the development of VITT based on the wide array of proinflammatory triggers and findings of Holm and colleagues. The participation of neutrophils and the formation of NETs has been noted previously in HIT as well as in COVID-19.20 , 21 While SARS-CoV-2 infection triggers the thromboinflammatory responses in patients with COVID-19, the trigger(s) associated with vaccination remain unknown. Hypotheses include any one of the constituents of the vaccine itself, including the adenoviral vector, or unique biological characteristics of those who develop VITT independent of the vaccine components. Individual variation in ability to control pathological inflammatory responses in any one of the humoral or cellular immune pathways or within the thrombotic pathways may result in the development of VITT or allow a normally minor response to vaccination to evolve into full-blown VITT. The adenoviral vector as the possible driver, however, is now in question given two suspected reported cases of VITT following administration of an mRNA vaccine in the USA (one published, one personal communication)22 and a small number of suspected cases in the UK. How to interpret these findings in light of the >350 million mRNA total vaccine doses administered in both countries is unclear.
What is clear is that VITT is a rare occurrence that we can now diagnose with identifiable clinical and laboratory features (Graphical Abstract) and can appropriately treat, resulting in lower mortality. Although use of the adenoviral vector vaccines was put on temporary hold in some countries, use by most has understandably resumed given the almost 4 million deaths that have resulted from SARS-CoV-2 infection. Continued investigation to unravel the aetiological triggers of VITT using clinical and laboratory findings is warranted, not only to understand the pathophysiology of VITT and improve treatment but to identify critical knowledge that can be used in the future—with 3.9 billion doses of vaccine administered around the world to date, there is hope that this pandemic will soon be over.1
Conflict of interest: J.M.C. has served on Scientific Advisory Boards and Consulted for Abbott, Alnylam, Anthos, Bristol Myers Squibb, Five Prime Therapeutics, Pfizer, Portola, and Takeda. Research funding has been received by the institution from PCORI, NHLBI, and CSL Behring.
The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.
References
https://coronavirus.jhu.edu/map.html (5 July
