COVID‐19 vaccination can occasionally trigger autoimmune phenomena, probably via inducing age‐associated B cells

1. AGE‐ASSOCIATED B CELLS

Age‐associated B cells (ABCs) constitute a CD11c⁺ T‐bet⁺ B‐cell population that expands continuously with age in healthy individuals, ¹ but also displays a premature accumulation in cases of autoimmune and/or infectious diseases. ² , ³ , ⁴ , ⁵ In autoimmune settings, ABCs are implicated in the production of autoreactive immunoglobulin G, ² the enhanced antigen presentation to T cells, and the formation of spontaneous germinal centers. ⁶ , ⁷ T‐bet, which is a transcription factor highly expressed in ABCs, is considered to be the master regulator of all these processes, ⁸ although new data suggest that its expression may not be required for the generation of functional ABCs. ⁹

In humans, the ABC subset is also known as double‐negative (DN) B cells because of the lack of immunoglobulin D and CD27 memory marker expression. ¹⁰ , ¹¹ , ¹² DN B cells have been further divided into two subgroups, based on the expression of the follicular homing marker CXCR5. ¹³ More specifically, the CXCR5⁺ cells (DN1) are expanded in elderly healthy individuals and lack T‐bet expression, whereas the CXCR5^– cells (DN2) express T‐bet and are more marked in autoimmune diseases (mostly systemic lupus erythematosus). ¹⁰ , ¹¹ , ¹³ The DN2 cells are hyper‐responsive to Toll‐like receptor 7 (TLR7) signaling, so are poised to generate autoreactive antibody‐secreting plasmablasts. ¹³ In general though, their role in the development of autoimmunity remains elusive.

2. AUTOIMMUNE PHENOMENA FOLLOWING COVID‐19 VACCINATION

Autoimmune disease flares and new‐onset disease following coronavirus disease 2019 (COVID‐19) vaccination have recently been reported. ¹⁴ , ¹⁵ In general, all these cases appeared rare and most of them were moderate in severity and had an excellent resolution of inflammatory features, with the use of corticosteroids alone ¹⁴ , ¹⁵ (indicating that COVID‐19 vaccines are actually safe). The use of TLR7/8 and TLR9 agonists, as adjuvants of the available mRNA and DNA severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) vaccines, may be the trigger of these post‐vaccination autoimmune/inflammatory phenomena, ¹⁴ as it is well known that both TLR7 and TLR9 are involved in the generation and amplification of autoreactive immune responses. ¹⁶

In more detail, the adjuvanticity of COVID‐19 vaccines depends—to a large extent—on the intrinsic adjuvanticity of mRNA or DNA, which respectively stimulate the innate immune system through endosomal and cytoplasmic RNA/DNA sensors such as TLRs. ¹⁷ The stimulation of TLR7 and TLR9 might be expected to produce elevated levels of type I interferon and so upregulate interferon‐stimulated genes ¹⁶ , ¹⁷ that contribute to the pathogenesis of a number of rheumatic diseases. ¹⁸

3. POTENTIAL INDUCTION OF ABCs BY COVID‐19 VACCINATION

Apart from autoimmune diseases, ABCs/DN B cells also expand in infectious diseases—including COVID‐19, ⁴ , ⁵ , ¹⁹ , ²⁰ usually having either an exhausted or anergic phenotype. ²¹ Moreover, circulating DN B cells increase in numbers after vaccination against influenza virus in healthy individuals. ²² Considering all these facts, we believe that there is a strong possibility for ABCs/DN B cells to be induced by COVID‐19 vaccines and then be involved in the autoimmune phenomena that (may) follow.

Our knowledge regarding the role of TLR7 and TLR9 in determining the fate of ABCs, ²³ in conjunction with the fact that agonists that bind to these two receptors are used as adjuvants in the available mRNA and DNA COVID‐19 vaccines, ¹⁴ , ¹⁷ further strengthens our hypothesis. To be more specific, we note that TLR7 and/or TLR9 stimulation, after antigen internalization via B‐cell receptor, leads pre‐immune B cells to an ABC‐poised status. ²³ Signals from interferon‐γ or interleukin‐21 determine the ABC phenotype. ²⁴ Otherwise, especially when TLR9 is engaged but no further signals exist, the cell is led to apoptosis. ²³ , ²⁴ In general, ABC generation is based on the synergistic triggering of B‐cell receptor, TLR7, and interferon‐γ or interleukin‐21 receptors. ³ , ²³ , ²⁴

4. CONCLUSIONS

In this article, we discuss the probability of ABC‐mediated autoimmunity (flare or new‐onset) following COVID‐19 vaccination. We find it important to mention here that, according to observations from a new study, the frequencies of DN B cells decrease in previously SARS‐CoV‐2‐infected individuals after their vaccination against the aforementioned virus, ²⁵ indicating that vaccine response counters the infection‐induced production of potentially pathogenic B cells. At first glance, the results of that study seem to oppose our hypothesis. However, we want to make it clear that we do not call into question the safety of COVID‐19 vaccines (besides, the data derived from participants in observational studies ¹⁴ , ¹⁵ clearly suggest that rheumatic disease flares and new‐onset disease following COVID‐19 vaccination are uncommon, mild to moderate in severity, and in most cases are treated with oral corticosteroids) and we propose an ABC‐induction only in the rare cases that autoimmune phenomena occur after the vaccination, as ABCs are indissolubly associated with autoimmunity. ² , ³ , ⁷ , ⁸ , ¹¹ , ¹² , ¹³ Furthermore, it is wise to keep in mind that ABC‐induction and as a result the estimation of ABC percentages are affected by various parameters, such as the age of the individual, ¹ , ²⁶ the ethnicity (as these cells are more marked in African‐American people), ¹³ , ²⁷ and of course the interval between vaccination and cell counting.

Taking into account the prognostic and/or diagnostic potential of ABCs in rheumatic diseases, ²⁸ we believe that the enumeration of these cells could enable better management of people to be vaccinated (especially those with autoimmune/rheumatic history, as some of the post‐vaccination flares described were severe). ¹⁴ Such an approach may determine the most proper vaccination time‐points for these people and so bring the least side effects and the most effective therapeutic benefits. ²⁹

AUTHOR CONTRIBUTIONS

Conceptualization and Writing‐Original Draft: Athanasios Sachinidis.

Critical Revision and Supervision: Alexandros Garyfallos.

CONFLICT OF INTEREST

The authors declare no conflict of interest.