COVID-19 Seroprevalence and Active Infection in an Asymptomatic Population

Abstract

In response to the COVID-19 pandemic, immediate and scalable testing solutions are needed to direct return to full capacity planning in the general public and across the Department of Defense (DoD). To fully understand the extent to which a population has been affected by COVID-19, active monitoring approaches require an estimation of overall seroprevalence in addition to accurate, affordable, and rapid tests to detect current SARS-CoV-2 infection. In this study, researchers in the Air Force Research Laboratory's 711th Human Performance Wing, Airman Systems Directorate evaluated the performance of various testing methods for the detection of SARS-CoV-2 antibodies and viral RNA in asymptomatic adults working at Wright-Patterson Air Force Base and the surrounding area during the period of 23 July 2020-23 Oct 2020. Altogether, there was a seroprevalance of 3.09% and an active infection rate of 0.5% (determined via the testing of saliva samples) amongst individuals tested, both of which were comparable to local and national averages at the time. This work also presents technical and non-technical assessments of various testing strategies as compared to the gold standard approaches (e.g., lateral flow assays vs. ELISA and RT-LAMP vs. RT-PCR) in order to explore orthogonal supply chains and fieldability. Exploration and validation of multiple testing strategies will allow the DoD and other workforces to make informed responses to COVID-19 and future pandemics.

Keywords: COVID-19; SARS-CoV-2; antibodies; asymptomatic; saliva; seroprevalance; testing.

Figure 1

Overview of the study. Participants enrolled in the serosurvey (Aim 1a or Aims 1a and 1b) and/or the molecular survey (Aim 2).

Figure 2

Breakdown of seropositive participants based on W-ELISA. The two participants represented by the IgG+/IgM+ bar had a sample test positive for both antibodies. All other positive participants tested positive to the same single antibody in one to four blood samples.

Figure 3

Comparison of serological tests for both IgG and IgM antibodies across either 4 assays (A) or 5 assays (B). In all comparisons, 20 assays were conducted on control samples and the remainder was on participant samples. Test performance was compared to the W-ELISA data; p < 0.001 (^***). Alfa, Alfa LFA; AXON, AXON LFA; BC, Beckman Coulter immunoassay; CH, CareHealth LFA; W-ELISA, Weighted ELISA.

Figure 4

Characterization of Thermo Fisher TaqPath™ RT-PCR COVID-19 assay using contrived positive saliva samples with a range of SARS-CoV-2 genomic copy equivalents (GCE). An equal amount of MS2 phage was added to each sample as an extraction control. Values represent Ct values ± standard deviation of three reactions. Dotted line represents the detection threshold (Ct = 37).

Figure 5

RT-PCR and RT-LAMP comparison using synthetic standards. Dilutions of synthetic control RNA in water were extracted and assayed (n = 4 or 9 tests per dilution). For RT-PCR assays, the viral genes were considered detected if 2 of the 3 viral genes were detected.