Screening, testing and contact tracing plays a pivotal role in control of the COVID-19 pandemic. To enable this it is necessary to increase the testing capacity. This study compared a SARS-CoV-2 rapid antigen test (RAT) and RT-PCR in 842 asymptomatic individuals from Tarapacá, Chile. A sensitivity of 69.86%, specificity of 99.61%, PPV of 94.44% and NPP of 97.22% with Ct values (Ct > 27) that were significantly higher among individuals with false-negative RAT were reported. These results support the fact that RAT might have a significant impact on the identification of asymptomatic carriers in areas that lack suitable laboratories to perform SARS-CoV-2 real-time RT-PCR diagnostics, or the results take more than 24–48 h, as well as zones with high traffic of individuals such as border/customs, airports, interregional bus, train stations or in any mass testing campaign requiring rapid results.
Given the increase in cases of SARS-CoV-2 infections worldwide, there is a need for a reliable rapid diagnostic test in addition to existing gold standard real-time RT-PCR. Rapid antigen tests (RAT) for SARS-CoV-2 can be performed onsite in mass testing, are inexpensive compared to real-time RT-PCR, do not require specific and expensive equipment, and the results are available within 15 min (CDC, 2021), which could serve to evaluate chains of infection and their interruption. A recent meta-analysis revealed that the average sensitivity and specificity of RAT for SARS-CoV-2 were 56.2% and 99.5%, respectively (Dinnes et al., 2020).
To date, most of these validations were carried out in symptomatic individuals or using previously collected samples (Cerutti et al., 2020, Kruttgen et al., 2021, Porte et al., 2020, Weitzel et al., 2020, Yamayoshi et al., 2020). In contrast, onsite test validation studies in asymptomatic individuals, to support the use of RAT in mass testing and epidemiological surveillance, are limited (Jakobsen et al., 2021, Mina et al., 2020, Pollock et al., 2021, Schildgen et al., 2021, Toptan et al., 2021). This study performed a mass comparison of RAT and real-time RT-PCR test in asymptomatic individuals from a Chilean region.
Materials and methods
Sample collection was coordinated by a specialized team from SEREMI de Salud Tarapacá. Two nasopharyngeal swabs (NSS) samples from asymptomatic individuals were collected by healthcare workers at Iquique city, Tarapacá Region, Chile between 14–17 January 2021, where the prevalence of SARS-CoV-2 infection was 11% according to data provided by the Chilean Ministry of Health (Minsal, 2021).
Taking into consideration that the expected prevalence of positive cases for COVID-19 in asymptomatic individuals can vary from 8–12% and establishing a sampling error of 1.5% and a type I error of 5%, the minimum sample size required was 864 ± 69.7. All participants completed a questionnaire and provided information on demographic characteristics, current and past (14 days) symptoms, and recent exposure to people with COVID-19 (Table 1). One swab was immediately tested at the facility using the SARS-CoV-2 RAT (SD Biosensor, Inc. Republic of Korea) and the result was interpreted according to the manufacturer’s guidelines. The second swab was preserved in specimen transport medium and transported at 4 °C to Laboratorio Médico Bioclinic and Hospital Regional de Iquique for RNA extraction and RT-PCR testing.
Rapid antigen test
The SD Biosensor, Inc. Antigen Test (Republic of Korea, Catalog number 9901-NCOV-01G) is a rapid lateral flow immunoassay for the qualitative detection of SARS-CoV-2-specific antigens present in the human nasopharynx, with sensitivity of 96.52% and specificity of 99.68% (SARS-CoV-2 Rapid Antigen Test Package Insert 2020-08, V 1.0). According to the manufacturer, the results are available within 30 min and all necessary reagents to perform the assay are provided. The assay kits are stable when stored at 2–30 °C.
Viral RNA was extracted using the Mag-Bind Viral DNA/RNA 96 kit (Omega Bio-Tek, Catalog number M6246) on the Kingfisher Flex Magnetic Particle Processor (Thermo Fisher Scientific). Real-time RT-PCR was performed at Laboratorio Médico Bioclinic and Hospital Regional de Iquique using the GenomeCov19 Detection Kit ABM (Applied Biological Materials Inc, Canada, Catalog number G628.v2), with cycle threshold (Ct) values ≤ 40 considered positive for the N and S viral gene regions.
Statistical analysis considered sensitivity, specificity, Positive Predictive Value (PPV) and Negative Predictive Value (NPV), accuracy, Kappa coefficient, and Wilson score Confidence Interval at 95% (GraphPad Prism version 9.0.1).
This study evaluated the performance of the SARS-CoV-2 RAT (SD Biosensor, Inc. Republic of Korea) compared with the real-time RT-PCR for SARS-CoV-2 detection among asymptomatic individuals at Iquique city, Tarapacá Region, Chile. The sampling was carried out in seven testing sites corresponding to i) workers (n = 56; 6.7%), ii) sanitary residence (n = 239; 28.4%) and iii) general public (n = 547; 65%). A total of 854 individuals were included (mean age: 36.67 years; SD: 16.48 years; males: 51%; females: 41.6%; N/A: 7.4%). Two NSS samples from each individual were collected by healthcare workers at testing sites.
Among a total of 854 NSS submitted, 12 (1.4%) were excluded for lacking real-time RT-PCR results. Among 842 paired NSS from asymptomatic individuals, 54 (6.17%) were antigen-positive and 73 (8.6%) were real-time RT-PCR-positive. Antigen testing sensitivity was 69.86% (51 of 73), specificity was 99.61% (766 of 769), PPV was 94.44% (51 of 54), and NPV was 97.22% (766 of 789). Three paired (0.35%) NSS were antigen-positive and real-time RT-PCR-negative. Accuracy between the two techniques was 97.04% (Kappa coefficient = 0.78, 95% CI: 0.70–0.86) (Table 2).
Given that the Ct value is inversely related to the viral load (Rao et al., 2020), the PCR Ct value data of the 73 samples that tested positive for NSS SARS-CoV-2 real-time RT-PCR detection were reviewed: Ct values were significantly higher among individuals with false-negative RAT (viral gene N: 28.07 ± 4.343; viral gene S: 28.81 ± 3.873) compared with true positives (viral gene N: 19.99 ± 4.535; viral gene S: 20.93 ± 4.487) (Figure 1A).
The k-means clustering analysis was then reviewed to find clusters in an iterative way using the Ct value of viral gene N (Cerutti et al., 2020). Three clusters were identified: c1 = 16.87 (n = 31, strongly positive), c2 = 24.5 (n = 30, moderately positive) and c3 = 31.67 (n = 12, weakly positive), where 96.77% (30/31) of the samples with low Ct value also tested positive for RAT (Figure 1B, cluster 1), while 66.6% (20/30) of the samples with medium Ct value also tested positive for RAT (Figure 1B, cluster 2). In contrast, 8.33% (1/12) of the samples with high Ct value tested positive for RAT (Figure 1B, cluster 3). Three RATs-positive and SARS CoV-2 real-time RT-PCR-negative samples were identified
1. When the bilirubin concentration is = 250 µmol/L, the hemoglobin content is = 9 g/L, the triglyceride content is = 15 mmol/L, the rheumatoid factor content is = 80 IU/mL, and the anti-nuclear antibody (ANA) titer is = 1: 240, anti-mitochondrial antibody (AMA) = 80 U/mL, and mouse IgG content = 1000 µg/mL, it will not interfere with the detection results of this product.
2. Histamine hydrochloride, alpha-interferon, zanamivir, ribavirin, oseltamivir, peramivir, lopinavir, ritonavir, abidol, levofloxacin, azithromycin, Ceftriaxone, meropenem, and tobramycin have been validated to have no effect on the test results of this product.
Cross Reactivity: This COVID-19 Rapid Test Kit IgG + IgM (Colloidal Gold) does not cross react with positive samples of parainfluenza virus antibody, influenza A virus antibody, influenza B virus antibody, chlamydia pneumoniae antibody, mycoplasma pneumoniae antibody, adenovirus antibody, respiratory syncytial virus antibody, hepatitis B surface antibody, type C Hepatitis virus antibody, treponema pallidum antibody, human immunodeficiency virus antibody, EB virus antibody, measles virus antibody, cytomegalovirus antibody, enterovirus 71 antibody, mumps virus antibody, chicken pox-zoster virus, and HKU1 virus antibodies, OC43 virus antibodies, NL63 virus antibodies, and 229E virus antibodies (the other common coronavirus strains).