key: cord-0925548-tl1zok1j authors: Toptan, T.; Eckermann, L.; Pfeiffer, A. E.; Hoehl, S.; Ciesek, S.; Drosten, C.; Corman, V. M. title: Evaluation of a SARS-CoV-2 rapid antigen test: potential to help reduce community spread? date: 2020-12-07 journal: nan DOI: 10.1101/2020.12.04.20240283 sha: ca768aebe9893c6a1431ab4c8a0d4626af41c8da doc_id: 925548 cord_uid: tl1zok1j Background: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) can spread from symptomatic patients with COVID-19, but also from asymptomatic individuals. Therefore, robust surveillance and timely interventions are essential for the control of virus spread within the community. In this regard the frequency of testing and speed of reporting, but not the test sensitivity alone, play a crucial role. Objectives: In order to reduce the costs and meet the expanding demands in real-time RT-PCR (rRT-PCR) testing for SARS-CoV-2, complementary assays, such as rapid antigen tests, have been developed. Rigorous analysis under varying conditions is required to assess the clinical performance of these tests and to ensure reproducible results. Results: We evaluated the sensitivity and specificity of a recently licensed rapid antigen test using 137 clinical samples in two institutions. Test sensitivity was between 88.2-89.6% when applied to samples with viral loads typically seen in infectious patients. Of 32 rRT-PCR positive samples, 19 demonstrated infectivity in cell culture, and 84% of these samples were reactive with the antigen test. Seven full-genome sequenced SARS-CoV-2 isolates and SARS-CoV-1 were detected with this antigen test, with no cross-reactivity against other common respiratory viruses. Conclusions: Numerous antigen tests are available for SARS-CoV-2 testing and their performance to detect infectious individuals may vary. Head-to-head comparison along with cell culture testing for infectivity may prove useful to identify better performing antigen tests. The antigen test analyzed in this study is easy-to-use, inexpensive, and scalable. It can be helpful in monitoring infection trends and thus has potential to reduce transmission. Since the beginning of COVID-19 outbreak in December 2020, the global demand for the 32 Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) testing has been steadily 33 increasing. Already back in March 2020, hospitals and laboratories around the world announced 34 their concerns about reagent, consumable material shortages, and limited personal protective 35 equipment. Yet, timely detection and isolation of SARS-CoV-2 infected cases and identification of 36 their contacts are pivotal to slowing down the pandemic. 37 The main public health strategy during a pandemic relies on robust and easy to perform 38 diagnostic tools that can be used to test large number of samples in a short time. To date the gold 39 standard diagnostic method for SARS-CoV-2 detection [1] is based on real time reverse 40 transcription-PCR (rRT-PCR) technology which has been promptly implemented by the World 41 and a number of commercial assays [4] . The SARS-CoV-2 rRT-PCR has high specificity and 43 sensitivity [5, 6] . However, the type and quality of the patient specimen [7, 8] , stage of the disease, 44 and the degree of viral replication and/or clearance have an impact on the test outcome [9] . These 45 factors are critical not only for PCR-based but also for other diagnostic test systems aiming to 46 detect the presence of the virus. Hence interpreting a test result for SARS-CoV-2 depends on the 47 accuracy of the test, but the prevalence and the estimated risk of disease before testing should 48 also be taken into consideration. 49 In many countries SARS-CoV-2 testing is extended to asymptomatic population, e.g. in 50 schools, airports, nursing-homes, and workplaces. This leads to a growing gap between the large 51 number of demand and the laboratory capacities to preform rRT-PCR tests, especially in 52 developing countries. Despite high specificity and sensitivity, rRT-PCR has a disadvantage in point 53 of care testing, because it usually requires professional expertise, expensive reagents and 54 specialized equipment. Therefore, alternative assays, such as rapid antigen detection tests, which 55 At the Institute of Virology, Charité Berlin, stored samples (swab resuspended in 1.5 mL of 106 phosphate-buffered saline) were anonymized before testing. After thawing at RT all samples were 107 analyzed by antigen test and rRT-PCR in parallel. RNA extraction for rRT-PCR was done by using 108 the MagNA Pure 96 system, using 100 μl of sample, eluted in 100 μl. rRT-PCR was done as 109 published previously [1] . 110 At the Institute of Virology in Frankfurt the SARS-CoV-2 test (Cobas, Roche, Basel, Switzerland) 111 was performed on the rRT-PCR automated Cobas 6800 system. Of the swab-dilution, 1000 µl 112 aliquots were mixed with lysis buffer (1:1 ratio) and 500 μL aliquots were transferred to barcoded 113 secondary tubes, loaded on the Cobas 6800 system, and tested with Cobas SARS-CoV-2 master 114 mix containing an internal RNA control and primer-probe sets towards ORF1 and E-gene 115 between the antigen test and rRT-PCR techniques was evaluated using the Cohen's weighted 128 kappa index (K value) [16] . K value interpretations were categorized as follows: <0.20 is poor, 129 0.21-0.40 is fair, 0.41-0.60 is moderate agreement, 0.61-0.80 is substantial agreement and 0.81-130 1.00 is almost perfect agreement [17] . 131 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted December 7, 2020. ; https://doi.org/10.1101/2020.12.04.20240283 doi: medRxiv preprint The use of stored clinical samples for validation of diagnostic methods without person related data 133 is covered by section 25 of the Berlin hospital law and does not require ethical or legal clearance. 134 The use of anonymized clinical samples for validation of diagnostic methods does not require 135 ethical clearance by the Goethe University, Frankfurt. 136 137 Rapid antigen test sensitivity and specificity were evaluated by two independent institutions 139 using various number of clinical samples. rRT-PCR was used as a reference test system. We were tested in cell culture for infectivity. All Ag-test positive (n:16, red circles) and three Ag-test 149 negative (red-filled blue circles) samples displayed CPEs after inoculating in Caco-2 cells (Table 150 S2). Intensities of the test bands were compared to control band and designated as follows: +++ 151 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted December 7, 2020. ; https://doi.org/10.1101/2020.12.04.20240283 doi: medRxiv preprint (test band intensity stronger than the control), ++ (test and control bans intensity are similar), + 152 (test band intensity is weaker than the control). 153 In the Institute of Virology, Charité, Berlin, a total of 67 stored patient samples were 154 available for the study. Of these, 58 were rRT-PCR positive with cycle threshold (cT) range 155 between 18.77-40 corresponding to 2.5x10 9 -1380 RNA genome copies/ml (Table S1) , 156 representing 86.6% (58/67) of the clinical samples analyzed ( Figure 1A) . When the rRT-PCR 157 results were used as a reference, the antigen test diagnosed SARS-CoV-2 infection status with a 158 sensitivity of 77.6% (45/58) and a specificity of 100% (9/9) ( Table 1) . After re-evaluating the data 159 based on the acceptable analytic sensitivity and limit of detection suggested by WHO [18], we 160 identified 48 samples with ≥10 6 RNA genome copies/ml. Rapid antigen test performed with 89.6% 161 sensitivity for this sample set (Table S1 ). Of these, 40 samples had ~2.23x10 6 or more RNA 162 genome copies/ml and reacted positive with the antigen test ( Table 1 ). In contrast samples with 163 less than 7.63x10 5 RNA copies/ml were negative ( Figure 1A , Table S1 ). Cohen's weighted kappa 164 value of 0.482 indicated moderate agreement between the rRT-PCR and the rapid antigen test 165 ( CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted December 7, 2020. Certain rapid tests may be used at the point-of-care and thus offer benefits for the detection 174 and management of infectious diseases. In order to assess the potential of the rapid antigen test 175 in this context, 70 nasopharyngeal samples freshly collected from individuals living in a shared 176 housing were analyzed head to head by rRT-PCR using Cobas 6800 system, rapid antigen test, Table S2) . 180 The antigen test diagnosed the infection status with a sensitivity of 50% (16/32) and a specificity 181 of 100% (Table 3) . Re-evaluating the data based on the limit of detection, sensitivity was 182 determined to be 88.2% for samples with cT values <28, and it was reduced in the group of 183 samples with cT values ≥ 28 (6.7%) ( Table 3) . Cohen's weighted kappa value of 0.521 indicated 184 moderate agreement between rRT-PCR and the rapid antigen test ( Table 4 ). The overall 185 concordance between the rRT-PCR and the antigen test was 77.1% (54/70) ( Table 4) . 186 rRT-PCR is a highly sensitive method to detect viral RNA molecules from clinical samples. 187 However, viral RNA can persist in different body parts and can be detected in specimens for much 188 longer than the presence of viable virus [19] . Thus demonstration of infectivity on permissive cell 189 lines in vitro is a more reliable surrogate for infectivity and virus transmission. Therefore, we 190 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted December 7, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted December 7, 2020. ; Table 4 . Cohen's weighted kappa coefficient between rapid antigen test and rRT-PCR. 206 In order to investigate potential cross reactivity among common coronaviruses and other 207 respiratory viruses, infectious and heat inactivated (4 h at 60°C) cell culture supernatants were 208 tested ( Table 5) . SARS-CoV-1 and SARS-CoV-2 tested positive with the antigen test, as 209 expected. The antigen test did not display any cross-reactivity with the other respiratory and 210 endemic corona viruses listed in Table 5 We further evaluated the detection sensitivity among different SARS-CoV-2 isolates. Here 216 we used cell culture supernatant collected from Caco-2 cells infected with seven different isolates 217 [15] and SARS-CoV-1 (Figure 3) . The virus stocks were thawed at room temperature and a total 218 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted December 7, 2020. ; https://doi.org/10.1101/2020.12.04.20240283 doi: medRxiv preprint of six 10-fold dilutions were prepared in PBS. The antigen test was performed and evaluated 219 immediately ( Figure 3A) . In parallel, aliquots of the dilutions were mixed with lysis buffer used for 220 ORF1 and E-gene that resulted in similar cT values ( Figure 3B, Table S3 ). 10-fold serial dilutions 222 led to ~3 cT difference in rRT-PCR for each set as anticipated. According to our results the limit 223 of detection was between 100-560 RNA copies/ml which is in line with the manufacturer's findings. 224 We previously identified RG203KR mutations in FFM3, FFM4 and FFM6 and SL mutation in 225 CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted December 7, 2020. ; In this study we validated the assay performance of a recently approved rapid antigen test 241 in two independent institutions using a total of 137 clinical samples. Although the test specificity 242 was 100% for this particular sample set, overall sensitivity was low (50-77.6%), yet re-analyzing Our results suggest that the rapid antigen test can detect SARS-CoV-2 infected individuals 259 with high viral loads and has potential in determining highly contagious individuals. Despite low 260 analytic sensitivity, rapid antigen tests are inexpensive and therefore can be used frequently for 261 detecting infected individuals who are asymptomatic, pre-symptomatic and without known or 262 suspected exposure to SARS-CoV-2 [24]. They can be beneficial in congregate settings, such as 263 a long-term care facility or a correctional facility, workplace, or a school testing its students, faculty, 264 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted December 7, 2020. ; https://doi.org/10.1101/2020.12.04.20240283 doi: medRxiv preprint and staff. Rapid antigen tests probably perform best during the early stages of SARS-CoV-2 265 infection when the viral load is higher. 266 267 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted December 7, 2020. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted December 7, 2020. ; https://doi.org/10.1101/2020.12.04.20240283 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted December 7, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted December 7, 2020. ; . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted December 7, 2020. ; https://doi.org/10.1101/2020.12.04.20240283 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted December 7, 2020. ; https://doi.org/10.1101/2020.12.04.20240283 doi: medRxiv preprint Detection of 2019 298 novel coronavirus (2019-nCoV) by real-time RT-PCR Laboratory testing for coronavirus disease (COVID-19) in suspected human 300 cases CDC 2019-Novel Coronavirus (2019-nCoV) Real-Time RT-302 PCR Diagnostic Panel Sars-Cov-2 Diagnostic Pipeline International external quality assessment for SARS-CoV-2 molecular detection and survey on 306 clinical laboratory preparedness during the COVID-19 pandemic Comparison of seven commercial RT-PCR diagnostic kits for COVID-19 Detection of SARS-CoV-2 in Different Types 312 of Clinical Specimens SARS-CoV-2 detection in different 314 respiratory sites: A systematic review and meta-analysis Interpreting Diagnostic Tests for SARS-CoV-2 Evaluation of a novel 318 antigen-based rapid detection test for the diagnosis of SARS-CoV-2 in respiratory samples Clinical Evaluation 321 of Self-Collected Saliva by Quantitative Reverse Transcription-PCR (RT-qPCR), Direct RT-qPCR Reverse Transcription-Loop-Mediated Isothermal Amplification, and a Rapid Antigen Test To 323 Diagnose COVID-19 Comparison 325 of automated SARS-CoV-2 antigen test for COVID-19 infection with quantitative RT-PCR using 326 313 nasopharyngeal swabs, including from seven serially followed patients Head-to-head 329 comparison of SARS-CoV-2 antigen-detecting rapid test with self-collected anterior nasal swab 330 versus professional-collected nasopharyngeal swab Evaluation 332 of the accuracy, ease of use and limit of detection of novel, rapid, antigen-detecting point-of-care 333 diagnostics for SARS-CoV-2 Optimized qRT-PCR 335 Approach for the Detection of Intra-and Extra-Cellular SARS-CoV-2 RNAs Weighted kappa: nominal scale agreement with provision for scaled disagreement 337 or partial credit The measurement of observer agreement for categorical data Virological 343 assessment of hospitalized patients with COVID-2019 Geographic and Genomic Distribution of SARS-CoV-2 Mutations Predicting infectious 347 SARS-CoV-2 from diagnostic samples Viral RNA 349 load as determined by cell culture as a management tool for discharge of SARS-CoV-2 patients 350 from infectious disease wards. European journal of clinical microbiology & infectious diseases : 351 official publication of the Viral cultures for COVID-19 infectivity 353 assessment. Systematic review Rethinking Covid-19 Test Sensitivity -A Strategy for 355 Containment