Evaluation of a Rapid Diagnostic Assay for Detection of SARS-CoV-2 Antigen in Nasopharyngeal Swabs

LETTER

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the virus causing coronavirus disease 2019 (COVID-19), was reported for the first time in Wuhan (Hubei Province, China) in December 2019 (1, 2) and has become a major public health concern all over the world. Early diagnosis is crucial for patient management and outbreak control. Most tests currently used for the detection of SARS-CoV-2 rely on viral RNA amplification by using real-time PCR (RT-PCR) and require a few hours before result release. Hence, highly sensitive immunological diagnostic methods that directly detect viral antigens in clinical samples would be very helpful for rapid and accurate diagnosis of COVID-19.

Here, we evaluated a rapid diagnostic test, COVID-19 Ag Respi-Strip (Coris BioConcept, Gembloux, Belgium), for detection of the SARS-CoV-2 antigen in nasopharyngeal secretions. The assay is ready to use and based on a nitrocellulose membrane technology with colloidal gold nanoparticles sensitized with monoclonal antibodies directed against highly conserved SARS-CoV-2 nucleoprotein antigens. We compared this test with RT-PCR, the current reference assay in virology laboratories of three university hospital groups from Assistance-Publique-Hôpitaux de Paris (APHP) (Saint-Antoine-Tenon-Trousseau, Saint-Louis-Lariboisière, and Kremlin Bicêtre-Paul Brousse). Different RT-PCR methods were used (RealStar [Altona Diagnostics], Bosphore novel coronavirus (2019-nCoV) detection kit [Anatolia Geneworks], Cobas 6800 [Roche], Allplex 2019 novel CoV assay [Seegene]). All assays amplify the SARS-CoV-2 E gene. Cycle threshold (C_T) values were recorded. Nasopharyngeal samples were tested prospectively within a few hours after collection and without any cooling or freezing step, from 1 April to 15 April 2020. Swabs were collected in various transport media (COPAN’s UTM [3 ml], Virocult [1 ml], ESwab Amies [1 ml], 4MRT [3 ml], 0.9% NaCl buffer, and cobas [Roche]). The first four samples collected in cobas medium tested gave invalid results. We therefore excluded such samples from the study. Our analysis included 138 nasopharyngeal samples, of which 94 (68.8%) were positive for SARS-CoV-2 by RT-PCR. Compared to that of RT-PCR, the specificity of the test was 100% (95% confidence intervals [95% CI], 91.8 to 100). Among the 94 RT-PCR-positive samples, the rapid test detected only 47 specimens, resulting in a sensitivity of 50.0% (95 CI, 39.5 to 60.5). In nine positive and eight negative tests, control lines were barely visible. Medians of E gene C_T values differed significantly between positive (median = 21; interquartile range [IQR], 17.0 to 23.0) and negative (median = 28.3; IQR, 25.6 to 33.0) antigenic test results (P < 0.0001) (Fig. 1). A study conducted by the manufacturer mentioned a sensitivity of 76.7% for samples positive with a C_T value under 25 (3). In our study, the test had a sensitivity of 82.2% for C_T values under 25.

FIG 1.

COVID-19 Ag Respi-Strip (Coris) results according to real-time PCR C_T values. All cycle threshold values of E gene real-time PCR-positive assays are shown for positive and negative COVID-19 Ag Respi-Strip assay results. Results gathering C_T values for all real-time PCR-positive assays are depicted by squares. C_T values between samples positive or negative for the antigenic assay are significantly different (* indicates a P value of <0.0001). C_T values corresponding to the Cobas 6800, Allplex, Anatolia, and RealStar assays are depicted by triangles, diamonds, circles, and upside-down triangles, respectively.

In our study, the COVID-19 Ag Respi-Strip (Coris) had a sensitivity of 50% compare to that of RT-PCR. The test was more sensitive for high viral loads and might perhaps be used for patients within a few days after symptom onset, when the load in the upper respiratory tract is at its peak. Considering COVID-19’s current low prevalence of 0.19% in France, prospective studies should be conducted to determine the best settings for its implementation.