ABSTRACT
Background Heterogeneity of the population in relation to infection, COVID-19 vaccination and host characteristics is likely reflected in the underlying SARS-CoV-2 antibody responses.Methods We measured IgM, IgA and IgG levels against SARS-CoV-2 spike and nucleocapsid antigens in 1,076 adults of a cohort study in Catalonia between June-November 2020 and a second time between May-July 2021. Questionnaire data and electronic health records on vaccination and COVID-19 testing were available in both periods.Results Antibody seroreversion occurred in 35.8% of the 64 participants infected almost a year ago and non-vaccinated, and was related to asymptomatic infection, age above 60 years and smoking. Among vaccinated, 2.1% did not present antibodies at the time of testing. In previously infected individuals, vaccination boosted the immune response and there was a slight but statistically significant increase in responses after a 2nd compared to 1st dose. Infected vaccinated participants had superior antibody levels across time compared to naïve vaccinated people. mRNA vaccines and, particularly the Spikevax, induced higher antibodies after 1st and 2nd doses compared to Vaxzevria or Janssen COVID-19 vaccines. In multivariable regression analyses, antibody responses after vaccination were predicted by type of vaccine, infection age, sex, smoking, mental and cardiovascular diseases.Conclusions Our data support that infected people would benefit from vaccination. Results also indicate that hybrid immunity results in superior antibody responses and infection-naïve people would need a booster dose earlier than previously infected people. Mental diseases are associated with less efficient response to vaccination.