Accuracy of Rapid Antigen Testing across SARS-CoV-2 Variants

Variants of SARS-CoV-2 have mutations in the viral genome that may alter the accuracy of rapid diagnostic tests.¹ Molecular tests can be affected by single point mutations, whereas antigen tests may require multiple mutations to change the confirmation of viral protein epitopes. The Omicron variant has numerous mutations in the spike and nucleocapsid proteins,² which has raised concerns about the analytical and clinical accuracy of rapid antigen testing.³

We conducted an analytical accuracy study of two FDA-approved rapid antigen tests—SCoV-2 Ag Detect™ Rapid Test (InBios International, Seattle) and BinaxNOW™ COVID-19 Ag CARD; (Abbott Laboratories, Chicago)—using three using replication-competent variants or strains, including Omicron (B.1.1.529/BA.1), Delta (B.1.617.2), and a wild-type of SARS-CoV-2 (USA-WA1/2020).

Omicron SARS-CoV-2 virus was isolated from a nasal swab collected in Maryland on November 27, 2021 and obtained through BEI Resources. The virus was cultured on Cercopithecus aethiops kidney cells (VERO E6) expressing transmembrane protease serine 2 gene (TMPRSS2) and angiotensin-converting enzyme 2 (ACE2), VERO E6-ACE2 and TMPRSS2. Variant identification of the cultured virus was confirmed through whole genome sequencing and cross-referenced with GISAID (Global Initiative on Sharing Avian Influenza Data, accession number, EPI_ISL_7160424). The replication competence of the virus stock was determined by measuring 50% tissue culture infectious dose (TCID₅₀). A corresponding quantitative measure of infectious viral particles was determined by assaying the plaque forming units (PFUs), through infection of VERO E6-ACE2 and TMPRSS2, using a standard plaque assay protocol.⁴ Similarly, we used the Delta (B.1.617.2) variant (GISAID: EPI_ISL_2103264) and USA-WA1/2020 strain (GenBank: MN985325.1) of SARS-CoV-2, which have known TCID₅₀ values of 1.1 ×10⁶ and 1.6 ×10⁶, respectively, for comparative analytical testing.

Performance and the analytical limit of detection for Omicron detection were measured by spiking negative clinical nasal swab matrices with replication-competent virus to establish three stock viral concentrations of 5.0 ×10⁴, 1.25 ×10⁴, and 3.12 ×10³ TCID₅₀/mL. From stock concentrations, we transferred 20 µL of sample onto nasal swabs to generate high, medium, and low viral concentrations of 1,000, 250, 62.5 TCID₅₀ per swab, respectively. We tested the viral dilutions across variants in triplicate for both the SCoV-2 Ag Detect™ Rapid Test (Figure S1) and BinaxNOW™ COVID-19 Ag CARD (Figure S2). Based on the visual signal intensity, the limit of detection for these rapid nucleocapsid antigen tests was approximately 62.5 TCID₅₀, which is equivalent to 51 PFUs (Figure S3).

• Download figure
• Open in new tab

Supplemental Figure 1.

Images of the SCoV-2 Ag Detect™ Rapid Test across SARS-CoV-2 variants and stratified by viral load of TCID₅₀ per swab. After establishing stock viral concentrations of 5.0 ×10⁴, 1.25 ×10⁴, and 3.12 ×10³ TCID₅₀/mL, we transferred 20 µL of sample onto nasal swabs to generate high, medium, and low viral concentrations of 1,000, 250, 62.5 TCID₅₀ per swab, respectively. We tested the viral dilutions across variants in triplicate using a SCoV-2 Ag Detect™ Rapid Test to establish the analytical limit of detection.

• Download figure
• Open in new tab

Supplemental Figure 2.

Images of the BinaxNOW COVID-19 Ag CARD across SARS-CoV-2 variants at 62.5 TCID₅₀ viral load per swab.

• Download figure
• Open in new tab

Supplemental Figure 3.

Plaque Assay to determine the replication competence of the SARS-CoV-2 isolate hCoV-19/USA/MD-HP20874/2021 (Lineage B.1.1.529; Omicron Variant).

To complement the analytical accuracy study of the SCoV-2 Ag Detect™ Rapid Test, we further conducted a clinical diagnostic accuracy study among 802 participants at multiple testing locations in King County, Washington from February 2021—January 2022, during three distinct phases of SARS-CoV-2 infections (pre-Delta, Delta, Omicron) (Table S1). Participants were ≥18 years old reporting onset of Covid-19-like symptoms within the prior five days. We collected two anterior nasal swabs—one for on-site testing by the SCoV-2 Ag Detect™ Rapid Test and one for reverse transcriptase polymerase chain reaction (rt-PCR) testing. The study received ethical approval from the University of Washington (STUDY00009981), and participants provided verbal informed consent.

Overall, we found non-significant differences in the analytical limit of detection or clinical diagnostic accuracy of rapid antigen testing across SARS-CoV-2 variants (Table 1). The positive percent agreement ranged from 81-91%, with improved sensitivity for swabs with a lower cycle threshold, which correlates to a higher viral load (Table S2). Visual signals were positively associated with viral concentration and had more variation at low concentrations (Figure S1 and S2). We found no real differences for clinical test performance by vaccination status or days since symptom onset (Table S2). The negative percent agreement was high across study periods.

This study provides analytical and clinical performance data to demonstrate the preserved accuracy of rapid antigen testing across SARS-CoV-2 variants among symptomatic adults. Other studies have similarly demonstrated good analytical sensitivity for rapid antigen tests to detect Omicron,^5,6 and with similar limits of analytical detection between the Omicron variant and USA-WA1/2020 strain.⁷ A field study has shown the BinaxNOW™ COVID-19 Ag CARD to have good clinical accuracy for Omicron.⁸ Our study has demonstrated both the analytical and clinical accuracy of rapid antigen testing across several variants of concern, including Omicron and Delta.

A study strength was detecting circulating variants representative of the community prevalence over a 12-month period, and additional clinical and analytical comparative studies may still be warranted for other rapid antigen tests. Since rapid antigen tests may correlate with recovery of replication-competent SARS-CoV-2⁹ and appear to retain accuracy across variants, ongoing home-based rapid antigen testing programs may be an important intervention to reduce global SARS-CoV-2 transmission.