Longitudinal Analysis of Antiphospholipid Antibody Dynamics after Infection with SARS-CoV-2 or Vaccination with BNT162b2

Abstract

Antiphospholipid antibodies (aPL) comprise a group of autoantibodies that reflect prothrombotic risk in antiphospholipid syndrome (APS) but may also be present in a small proportion of healthy individuals. They are often transiently elevated in infections, including SARS-CoV-2, and may also be associated with vaccine-induced autoimmunity. Therefore, we aimed to investigate the dynamics of aPL in COVID-19 patients and in individuals (healthcare professionals—HCPs) after receiving BNT162b2 vaccine and to compare aPL levels and positivity with those found in APS patients. We measured solid-phase identifiable aPL, including anticardiolipin (aCL), anti-β2 glycoprotein I (anti-β2GPI), and anti-prothrombin/phosphatidylserine (aPS/PT) antibodies in 58 HCPs before and after vaccination (at 3 weeks, 3, 6, and 9 months after the second dose, and 3 weeks after the third booster dose), in 45 COVID-19 patients hospitalized in the ICU, in 89 COVID-19 patients hospitalized in the non-ICU (at admission, at hospital discharge, and at follow-up), and in 52 patients with APS. The most frequently induced aPL in COVID-19 patients (hospitalized in non-ICU) were aCL (50.6% of patients had positive levels at at least one time point), followed by anti-β2GPI (21.3% of patients had positive levels at at least one time point). In 9/89 COVID-19 patients, positive aPL levels persisted for three months. One HCP developed aCL IgG after vaccination but the persistence could not be confirmed, and two HCPs developed persistent anti-β2GPI IgG after vaccination with no increase during a 1-year follow-up period. Solid-phase aPL were detected in 84.6% of APS patients, in 49.4% of COVID-19 patients hospitalized in the non-ICU, in 33.3% of COVID-19 patients hospitalized in the ICU, and in only 17.2% of vaccinated HCPs. aPL levels and multiple positivity were significantly lower in both infected groups and in vaccinated individuals compared with APS patients. In conclusion, BNT162b2 mRNA vaccine may have induced aPL in a few individuals, whereas SARS-CoV-2 infection itself results in a higher percentage of aPL induction, but the levels, persistence, and multiple positivity of aPL do not follow the pattern observed in APS.