Although SARS-CoV-2 infects the upper respiratory tract, we know little about the amount, type, and kinetics of antibodies (Ab) generated in the oral cavity in response to COVID-19 vaccination. We collected serum and saliva samples from participants receiving two doses of mRNA COVID-19 vaccines and measured the level of anti-SARS-CoV-2 Ab. We detected anti-Spike and anti-Receptor Binding Domain (RBD) IgG and IgA, as well as anti-Spike/RBD associated secretory component in the saliva of most participants after dose 1. Administration of a second dose of mRNA boosted the IgG but not the IgA response, with only 30% of participants remaining positive for IgA at this timepoint. At 6 months post-dose 2, these participants exhibited diminished anti-Spike/RBD IgG levels, although secretory component-associated anti-Spike Ab were more stable. Examining two prospective cohorts we found that participants who experienced breakthrough infections with SARS-CoV-2 variants had lower levels of vaccine-induced serum anti-Spike/RBD IgA at 2–4 weeks post-dose 2 compared to participants who did not experience an infection, whereas IgG levels were comparable between groups. These data suggest that COVID-19 vaccines that elicit a durable IgA response may have utility in preventing infection.

Our study finds that a local secretory component-associated IgA response is induced by COVID-19 mRNA vaccination that persists in some, but not all participants. The serum and saliva IgA response modestly correlate at 2–4 weeks post-dose 2. Of note, levels of anti-Spike serum IgA (but not IgG) at this timepoint are lower in participants who subsequently become infected with SARS-CoV-2. As new surges of SARS-CoV-2 variants arise, developing COVID-19 booster shots that provoke high levels of IgA has the potential to reduce person-to-person transmission.