Study population and setting
There is currently limited data on whether development of IgG antibodies following infection from SARS-CoV-2 leads to protection against reinfection. The authors sought to determine whether being antibody positive for SARS-CoV-2 predicted risk of future infection compared to being antibody negative. The authors conducted a retrospective cohort study using a national sample in the US with commercial laboratory results linked to longitudinal claims data, electronic health records and billing data and medical records. Patients with at least one antibody test on or after January 2020 were classified by their antibody status on first index test, and then compared on their risk of having evidence of viral shedding using a nucleic acid amplification test (NAAT) through August 26, 2020. Subsequent antibody tests were also included in the study to examine loss of detectable SARS-CoV-2 antibodies over time.
Summary of Main Findings
The authors identified 3,257,478 patients with at least one index antibody test of whom 88.3% were negative and 11.6% were positive at baseline. Median follow-up was 47 days for the seronegative group and 54 days for the seropositive group. In total 18.4% of those who were seropositive at baseline converted to seronegative by the end of study. Among those positive at baseline, 11.0% (n=41,587) had a NAAT test during follow-up, while among those negative at baseline, 9.5% (n=273,735) did. The ratio of positive NAAT tests at follow-up over the first 0-30 days comparing those who had a positive antibody test at baseline to those who were antibody negative at baseline was 2.85 (95% CI: 2.73 – 2.97), likely reflecting shedding of viral particles from the initial infection. After 30 days, the ratio of NAAT positive tests was lower for those who were initially antibody positive compared to those initially antibody negative (ratio 0.67; 95% CI: 0.60-0.74 from 31-60 days, 0.29; 95% CI: 0.24-0.35 from 61-90 days and 0.10; 95% CI: 0.05-0.19 at >90 days) likely reflecting antibody protection against reinfection in the short term.
The study sample was very large giving reasonably precise estimates of the relationship between antibody positivity and infection and allowing for breaking the results up into time periods to distinguish between first infections and repeat infections. In addition, the cohort included participants from around the US allowing for broad generalizability of the result in terms of region.
One limitation of this analysis is the potential for misclassification of the antibody and NAAT results and the use of any type of antibody to define antibody positivity, even though IgM antibodies, for example, would not be expected to provide protection over time. Both of these limitations could lead to an underestimate the protective effect of longer lasting antibodies. There is likely some measurement error in the NAAT test itself but there also limited ability to differentiate between new and persisting viral shedding, particularly early after the first antibody test. Finally, selection bias may have impacted the results given that it isn’t clear who would have had more than one antibody test or the reasons for testing – including, curiosity, symptoms, or work requirements – and how that relates to possible risk of exposure.
There is currently limited data on the duration of protection from antibodies following SARS-CoV-2 infection, and as such, this adds important information on short term risk of reinfection.
This review was posted on: 29 January 2021