Post‐adenoviral‐based COVID‐19 vaccines thrombosis: A proposed mechanism

There have been several reports of thrombosis and thrombocytopenia after vaccination with the ChAdOx1 nCoV‐19 vaccine (Oxford–Astra‐Zeneca) and Ad26.COV2.S vaccine (Johnson & Johnson/Janssen), a recombinant adenovirus serotype 26 vector encoding the SARS‐CoV‐2 spike glycoprotein.1., 2. These cases clinically manifest as disseminated intravascular coagulation‐like phenomena. Recently anti‐platelet factor 4 (PF4) autoantibodies detected by ELISAs were deemed as the number one culprit, but this finding has one main issue. The main issue is: Because these cases manifest as early as 5 days post‐vaccination, there is typically not enough time for immunological tolerance to break and to generate high titer and class‐switched, high‐affinity IgG antibodies to trigger the proposed mechanism. However, it is possible that anti‐PF4 is a byproduct of an initial mechanism that in turn can eventually lead to thrombocytopenia and amplify a viscous cycle. Interestingly, a proposed mechanism was FcγRIIa (CD32a)‐dependent.¹ What is the initial mechanism? Bye et al.³ recently demonstrated the role of aberrant glycosylation of anti‐SARS‐CoV‐2 IgG as a pro‐thrombotic stimulus for platelets. They showed that in COVID‐19 patients, anti‐SARS‐CoV‐2 IgG‐spike glycoprotein immune complexes can, through FcγRIIa, which is the only FcγR present on the surface of platelets, activate platelets and lead to their adhesion to endothelial cells. The latter further triggered endothelial cells to produce von Willebrand factor (VWF). Additionally, VWF has been reported to be as high as five‐fold in severe cases of COVID‐19 compared to other cases.4., 5., 6. More in‐depth research is needed to substantiate such a ready‐to‐go mechanism; that being said, here I propose the following: a single vaccine dose contains 5 × 10¹⁰ adenoviral particles. If all is accidentally injected into the blood, for an approximate blood volume of 5000 ml, there will be an adenoviral load of 10⁷/ml. Even lower levels or leaks from the injection site (over time) would culminate in still high‐level adenoviremia. Although these adenoviruses are claimed to be replication‐deficient, they are still able to travel to distant sites in the body and infect a range of permissive cells. Once infected, cells such as epithelial, endothelial, and fibroblasts, etc., that are not primarily antigen‐presenting cells, may also secrete copious amounts of soluble spike glycoproteins leading to a relatively high level of SARS‐CoV‐2 spike “antigenemia.” It is important to highlight that although chimpanzee adenovirus and human adenovirus 26 use different cellular receptors, that is, sialic acid‐bearing glycans and Coxsackie‐adenovirus receptor, respectively, both receptors are expressed on a large range of tissues.⁷ Although a vaccine that uses a modified spike that may not be shed from cells, because adenoviremia can reach very high levels and infect a large number of cells, even focal expression of the spike is enough to trigger this mechanism. Nonetheless, this may not be possible in the case of mRNA‐based vaccines as lipid nanoparticles cannot survive in the enzymatically hostile environment of plasma and are rapidly cleared by the reticuloendothelial system. In a person with a prior SARS‐CoV‐2 infection and/or with cross‐reactive antibodies to common coronaviruses (CoVs), a large volume of the aforementioned immune complexes may form shortly after vaccination with adenovirus‐based vaccines. Now using the Bye et al. study, in which IgG antibodies within these immune complexes are aberrantly glycosylated (for instance afucosylated), the abovementioned mechanism can be triggered. The platelet adhesion to endothelial cells may also be one of the causes for severe thrombocytopenia observed in these cases. It was also previously shown that afucosylated antibodies were much more common among severe and critical COVID‐19 cases.⁸ Therefore, all three conditions—the amount of adenoviral vector leakage to the circulation, presence of specific and/or cross‐reactive antibodies, and high enough titer of aberrantly glycosylated antibodies—need to be present to trigger such a mechanism. This may explain the rarity of the clinical observation. It is worth mentioning that the spike glycoprotein expressed in these vaccines is in fact the full spike antigen in its trimeric form. This means it contains highly cross‐reactive domains (such as S2) that can be bound by antibodies against common CoVs. Very shortly after vaccination, anamnestic immune responses to common CoVs are activated and antibody titers can be found in very high titers.⁹ Last, the apparent clinical response to intravenous immunoglobulin (IVIg) in these cases could very well be due to the competition between high‐titer non‐specific IgGs in the IVIg with the previously mentioned immune complexes through their Fc ends for CD32a on the surface of platelets. The latter would preclude their further adhesion to endothelial cells, which can lead to the reversal of thrombocytopenia.

The proposed mechanism here needs to be substantiated. Remedial actions would be to observe best practices in administering vaccines, possible reduction of the vaccine dose, and to avoid vaccinating those with underlying coagulopathies or thrombocytopenia.

CONFLICT OF INTEREST

None declared.