Association of infection-induced antibody levels with risk of subsequent SARS-COV-2 reinfection among healthcare professionals, Rhode Island, 1 March 2020-17 February 2021.

Numerous studies have investigated vaccine-induced correlates of protection (CoP) against severe acute respiratory syndrome-coronavirus 2 (SARS-CoV-2) infection, but data on infection-induced CoP are limited. Given differences between vaccine- and infection-induced immune responses, in conjunction with low vaccination in many US populations, a better understanding of infection-induced CoP is needed. We used residual sera from a mid-2020 Rhode Island serosurvey of healthcare professionals (HCP) and corresponding state-collected SARS-CoV-2 testing data through February 2021 to generate an analytic cohort of HCP with a first SARS-CoV-2 infection prior to serosurvey blood collection and multiple viral tests after blood collection to assess for reinfection (defined as a positive viral test ≥90 days after their first positive). We tested sera for levels of IgG and IgA targeting ancestral spike (S), receptor-binding domain (RBD), or nucleocapsid (N). We used adjusted Cox proportional hazard ratios to assess the association between categorical antibody level and the risk of subsequent reinfection. Among 170 HCP included in this analysis (median age = 47 years; interquartile range: 35-55 years), 30 were reinfected during the analytic period. Adjusted Cox proportional hazard ratios indicated that higher levels of anti-S or anti-RBD IgG were significantly associated with a lower risk of reinfection. These findings support the use of anti-S or anti-RBD IgG levels as markers of immunologic protection, such as in population serosurveys, or immune-bridging studies in settings of high prevalence of prior infection.  IMPORTANCEThe measurement of antibodies in blood is a relatively simple process and commonly used to estimate overall levels of past infection in populations. But, if someone has antibodies, does this mean that they are protected from being infected again? And are people with higher levels of antibody better protected? There are good data in the literature exploring how antibodies from the coronavirus disease 2019 (COVID-19) vaccination are associated with protection. But, there is still a lot to learn about protection conferred by antibodies that develop after a severe acute respiratory syndrome-coronavirus 2 (SARS-CoV-2) infection. In our study, we measure the levels of six different antibody types developed after infection and compare levels to the risk of subsequent infection to better understand which antibody types are best associated with protection. Our data are important for improving studies that use antibodies as proxies for protection, such as population immunity estimates, or those assessing new prevention products.