24 People, one test: Boosting test efficiency using pooled serum antibody testing for SARS-CoV-2

BackgroundThe global pandemic of COVID-19 (coronavirus disease 2019) is caused by the novel coronavirus SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2), with different prevalence rates across countries and regions. Dynamic testing strategies are mandatory to establish efficient mitigation strategies against the disease; to be cost effective, they should adapt to regional prevalences. Seroprevalence surveys that detect individuals who have mounted an immune response against COVID-19 will help to determine the total number of infections within a community and improve the epidemiological calculations of attack and case fatality rates of the virus. They will also inform about the percentage of a population that might be immune against re-infections. MethodsWe developed a sensitive and specific cell-based assay to detect conformational SARS-CoV-2 spike (SARS-2-S) S1 antibodies in human serum, and have cross-evaluated this assay against two FDA-approved SARS-CoV-2 antibody assays. We performed pseudovirus neutralization assays to determine whether sera that were rated antibody-positive in our assay were able to specifically neutralize SARS-2-S. We pooled up to 24 sera and assessed the group testing performance of our cell-based assay. Group testing was further optimized by Monte Carlo like simulations and prospectively evaluated. FindingsHighly significant correlations could be established between our cell-based assay and commercial antibody tests for SARS-CoV-2. SARS-2-S S1 antibody-positive sera neutralized SARS-2-S but not SARS-S, and were sensitively and specifically detected in pools of 24 samples. Monte Carlo like simulations demonstrated that a simple two-step pooling scheme with fixed pool sizes performed at least equally as well as Dorfmans optimal testing across a wide range of antibody prevalences. InterpretationWe demonstrate that a cell-based assay for SARS-2-S S1 antibodies qualifies for group testing of neutralizing anti-SARS-2-S antibodies. The assay can be combined with an easily implemented algorithm which greatly expands the screening capacity to detect anti-SARS-2-S antibodies across a wide range of antibody prevalences. It will thus improve population serological testing in many countries. FundingThis work was supported by the Bundesministerium fur Bildung und Forschung within the network project RAPID (risk assessment in pre-pandemic respiratory infectious diseases [grant number 01KI1723D, S.P.]).