Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) seroprevalence: Navigating the absence of a gold standard

Abstract

Background: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) seroprevalence studies bridge the gap left from case detection, to estimate the true burden of the COVID-19 pandemic. While multiple anti-SARS-CoV-2 immunoassays are available, no gold standard exists.

Methods: This serial cross-sectional study was conducted using plasma samples from 8999 healthy blood donors between April-September 2020. Each sample was tested by four assays: Abbott SARS-Cov-2 IgG assay, targeting nucleocapsid (Abbott-NP) and three in-house IgG ELISA assays (targeting spike glycoprotein, receptor binding domain, and nucleocapsid). Seroprevalence rates were compared using multiple composite reference standards and by a series of Bayesian Latent Class Models.

Result: We found 13 unique diagnostic phenotypes; only 32 samples (0.4%) were positive by all assays. None of the individual assays resulted in seroprevalence increasing monotonically over time. In contrast, by using the results from all assays, the Bayesian Latent Class Model with informative priors predicted seroprevalence increased from 0.7% (95% credible interval (95% CrI); 0.4, 1.0%) in April/May to 0.7% (95% CrI 0.5, 1.1%) in June/July to 0.9% (95% CrI 0.5, 1.3) in August/September. Assay characteristics varied over time. Overall Spike had the highest sensitivity (93.5% (95% CrI 88.7, 97.3%), while the sensitivity of the Abbott-NP assay waned from 77.3% (95% CrI 58.7, 92.5%) in April/May to 64.4% (95% CrI 45.6, 83.0) by August/September.

Discussion: Our results confirmed very low seroprevalence after the first wave in Canada. Given the dynamic nature of this pandemic, Bayesian Latent Class Models can be used to correct for imperfect test characteristics and waning IgG antibody signals.

Fig 1

A. Seroprevalence by month over the first COVID-19 wave in Canada by individual assays. Each line represents seroprevalence rates (summarized in table below) monthly between April and September 2020 (during the first COVID-19 Wave) based thresholds for each assay. Abbott Architect SARS-Cov-2 IgG assay (Abbott-NP) and three in-house IgG ELISA assays recognizing distinct recombinant viral antigens: full length spike glycoprotein (Spike), spike glycoprotein receptor binding domain (RBD), and nucleocapsid (NP). B. Seroprevalence by month over the first COVID-19 wave in Canada by various composite reference standards (results from four anti-SARS-CoV-2 immunoassays). Each line represents seroprevalence rates based on predefined definitions. CRS based on a combination of reactive samples using Abbott-NP, Spike, RBD and NP. Positivity based on “any two or more” was determined by a reactive sample from two or more assays. Since we are not comparing CRS, we did not include 95% CI for each data point (all overlapping).

Fig 2. Seroprevalence estimates by different BLCM (informative, weakly informative and non-informative priors).

Each line represents seroprevalence rates (summarized in table below) derived from posterior means of three BLCMs (comparing informative, weakly informative and non-informative priors) bi-monthly between April and September 2020 (during the first COVID-19 Wave). Error bars represent 95% Credible Intervals.

Fig 3. Summary comparison of seroprevalence rates by analytical methods.

Each line represents seroprevalence rates (summarized in table below) derived from four analytical methods bi-monthly between April and September 2020 (during the first COVID-19 Wave). > = 2 proteins (positivity was determined by a reactive sample from two or more assays), BLCA-Bayesian latent class analysis with informative priors, results are posterior means and error bars are 95% CrI and Abbott-NP is a single commercial assay.