Abstract:
Thromboembolic complications are a hallmark of COVID-19 and are usually associated with multiple organ failure and high mortality. Activated platelets contribute to COVID-19 coagulopathy. Recent evidence suggests a persistent hypercoagulable and/or hypofibrinolytic state in convalescent COVID-19 patients. However, platelets in the convalescent period have not been studied to date. In addition, a rare thrombotic and thrombocytopenic complication, vaccine-induced immune thrombotic thrombocytopenia (VITT), has been described after vaccination with adenoviral vector based COVID-19 vaccines. In VITT, antibodies reacting to platelet factor 4 (PF4) activate platelets via the Fcγ receptor IIa (FcγRIIa) and lead to thrombus formation as well as platelet destruction. The mechanism of anti-PF4 antibody development is not clear. In this work, we aimed to understand antibody–mediated platelet activation in COVID-19 patients in the convalescent phase and the pathophysiology of VITT.
In a prospective study, we investigated platelet phenotype by flow cytometry and global coagulation and fibrinolysis by thromboelastometry in convalescent COVID-19 patients and control subjects. Phosphatidylserine (PS) externalization, CD62P expression, and Glycoprotein VI (GPVI) shedding were measured by flow cytometry. Platelets from COVID-19 convalescents did not exhibit an increased procoagulant phenotype. In addition, GPVI shedding did not differ from that of the control group. Sera from COVID-19 convalescents did not induce significantly higher PS externalization or GPVI shedding in healthy platelets. In thromboelastometry, only one COVID-19 convalescent showed increased maximum clotting strength and prolonged lysis time. Consequently, we did not observe an increased platelet activation/procoagulant phenotype in the convalescent phase after mild COVID-19 treatment.
We then focused on understanding the development of anti-PF4 antibodies in patients with VITT. One of the proposed mechanisms is cross-reactivity between SARS-CoV-2 and PF4. To investigate the correlation between anti-PF4 and anti-SARS-CoV-2 antibodies, we measured antibody levels in vaccinated controls and in patients with clinical suspicion of VITT after vaccination with ChAdOx1 nCoV-19. VITT patients had higher anti-PF4 antibody levels than healthy controls and non-VITT patients. Levels of anti-PF4 did not correlate anti-SARS-CoV-2 antibodies in any of the study groups. These results refute the assumption that the development of anti-SARS-CoV-2 antibodies contributes to the formation of anti-PF4 antibodies in VITT patients.
Finally, we investigated the effects of intravenous immunoglobulin therapy (IVIG) on platelet count and platelet activation in patients with VITT. All patients received a non-heparin anticoagulant. In addition, IVIG was administered at a dose of 1 g/KG body weight for 2 to 5 days. Four patients showed a complete platelet response (platelet count ≥ 100×109/L) and one patient a partial response (platelet count ≥ 30×109/L and at least 2-fold increase in baseline count) after IVIG therapy. D-dimer levels decreased after IVIG therapy, indicating a reduction in thrombus burden. After IVIG therapy, the ability of sera from VITT patients to induce procoagulant platelets in flow cytometry was reduced. However, anti-PF4 antibody levels were not affected by IVIG therapy. These data suggest that IVIG attenuates the activity of pathogenic anti-PF4 antibodies by competing with them for binding to FcγRIIa receptors, which may be associated with a reduction in platelet activation in vivo.
In conclusion, our results do not confirm sustained platelet activation in the convalescent phase of COVID-19. Moreover, antibodies against SARS-CoV-2 are not responsible for the development of anti-PF4 antibodies in patients with VITT. Finally yet importantly, high-dose IVIG therapy is an effective means to cool down the hypercoagulable state in patients with VITT.