key: cord-0971066-65aljtks
authors: Nörz, Dominik; Olearo, Flaminia; Perisic, Stojan; Bauer, Matthias F.; Riester, Elena; Schneider, Tanja; Schönfeld, Kathrin; Laengin, Tina; Lütgehetmann, Marc
title: Multicenter Evaluation of a Fully Automated High-Throughput SARS-CoV-2 Antigen Immunoassay
date: 2021-08-09
journal: Infect Dis Ther
DOI: 10.1007/s40121-021-00510-x
sha: b0704811d5aff5320ac6265e268d1bc824b25c26
doc_id: 971066
cord_uid: 65aljtks

INTRODUCTION: Molecular testing for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) continues to suffer from delays and shortages. Antigen tests have recently emerged as a viable alternative to detect patients with high viral loads, associated with elevated risk of transmission. While rapid lateral flow tests greatly improved accessibility of SARS-CoV-2 detection in critical areas, their manual nature limits scalability and suitability for large-scale testing schemes. The Elecsys(®) SARS-CoV-2 Antigen assay allows antigen immunoassays to be carried out on fully automated high-throughput serology platforms. METHODS: A total of 3139 nasopharyngeal and oropharyngeal swabs were collected at 3 different testing sites in Germany. Swab samples were pre-characterized by reverse transcription real-time polymerase chain reaction (RT-qPCR) and consecutively subjected to the antigen immunoassay on either the cobas e 411 or cobas e 801 analyzer. RESULTS: Of the tested respiratory samples, 392 were PCR positive for SARS-CoV-2 RNA. Median concentration was 2.95 × 10(4) (interquartile range [IQR] 5.1 × 10(2)–3.5 × 10(6)) copies/ml. Overall sensitivity and specificity of the antigen immunoassay were 60.2% (95% confidence interval [CI] 55.2–65.1) and 99.9% (95% CI 99.6–100.0), respectively. A 93.7% (95% CI 89.7–96.5) sensitivity was achieved at a viral RNA concentration ≥ 10(4) copies/ml (~ cycle threshold [C(t)] value < 29.9). CONCLUSION: The Elecsys SARS-CoV-2 Antigen assay reliably detected patient samples with viral loads ≥ 10,000 copies/ml. It thus represents a viable high-throughput alternative for screening of patients or in situations where PCR testing is not readily available. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s40121-021-00510-x.

Reverse transcription real-time polymerase chain reaction (RT-qPCR) remains the gold standard for detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in clinical specimens due to its unparalleled analytic accuracy [1, 2] . However, the sudden surge in demand for molecular testing continues to overstretch the capacity of both diagnostic laboratories and reagent suppliers, leading to reporting delays and sometimes inadequate availability of testing where it is urgently needed. SARS-CoV-2 antigen immunoassays have recently emerged as an alternative to nucleic acid amplification tests, most prolifically in the form of rapid, point-of-care, lateral flow tests (LFTs) [3] [4] [5] . Despite their inherent disadvantage in sensitivity compared with PCR, mathematical modeling suggests that tests could be effective for infection control if appropriate testing frequency were adopted [6] . Using currently available rapid antigen tests for mass testing does however bring about additional challenges, as this assay format is highly manual in nature and largely unsuitable for automation. Moving the SARS-CoV-2 antigen assay from LFT to a high-throughput immunoanalyzer using electrochemiluminescence detection technology could improve those issues, while further increasing overall testing capacity. The aim of this study was to evaluate the sensitivity and specificity of the new fully automated Elecsys Ò SARS-CoV-2 Antigen assay, using samples characterized by the gold standard: RT-qPCR.

In this multicenter study, nasopharyngeal and oropharyngeal samples were tested following routine diagnostics at the University Medical Center Hamburg-Eppendorf, the Hospital of Stuttgart, and Hospital Ludwigshafen (Supplementary Table 1 

Antigen detection was performed at two testing sites (University Medical Center Hamburg-Eppendorf [UKE] and testing site of Augsburg) using the Elecsys SARS-CoV-2 Antigen assay on the cobas e 411 and cobas e 801 immunoanalyzers. The Elecsys SARS-CoV-2 Antigen assay (Roche Diagnostics International Ltd, Rotkreuz, Switzerland) is an automated electrochemiluminescence immunoassay (ECLIA), developed for in vitro qualitative detection of SARS-CoV-2 antigen. It utilizes monoclonal antibodies directed against the SARS-CoV-2 nucleocapsid protein in an antibody sandwich assay format for the detection of SARS-CoV-2 in upper respiratory tract specimens [7] . The time to result for the assay can be as little as 18 min, and the analyzer automatically calculates a cutoff based on the measurement of two calibrators: one negative (COV2AG Cal1) and one positive (COV2AG Cal2). The results obtained are interpreted as recommended by the manufacturer as either 'reactive' or 'non-reactive' in the form of a cutoff index (COI), i.e., 'non-reactive' if COI \ 1.0 or 'reactive' if COI C 1.0.

RT-qPCR testing was performed with the qualitative cobas Ò SARS-CoV-2 assay on the cobas Ò 6800 system (Roche Molecular Systems, Inc., Branchburg, NJ, USA), according to the manufacturer's instructions [8] . Quantification was performed using cycle threshold (C t ) values of Target-2 (envelope protein coding gene) [9] using reference material by Qnostics Ltd. (Glasgow, UK) as the quantification standard. The formula for conversion of the C t value to log10 copies/ml is (C t 9 -0.30769) ? 13.2. The lower limit of quantification was set to C t = 33 corresponding to 1000 copies SARS-CoV-2 RNA/ ml, in accordance with previous studies [9] . It has to be noted that the PCR test and quantification standard used for generating quantitative results were not recommended for this purpose by the respective manufacturers, and reliability is expected to be lower than commercial quantitative solutions once they become available, which represents a limitation of this study.

Sensitivity and specificity, including 95% confidence intervals (CIs), were assessed as per Altman [10] . Pearson correlation was used for linear regression. The significance threshold was set at a two-sided alpha value of 0.05. Statistical analysis was performed with STATA (version 15) and GraphPad Prism (version 86 9.0.0).

Qualified researchers may request access to individual patient-level data through the clinical study data request platform (https://vivli. org/). Further details on Roche's criteria for eligible studies are available here: https://vivli.org/ members/ourmembers/. For further details on Roche's Global Policy on the Sharing of Clinical Information and how to request access to related clinical study documents, see: https://www. roche.com/research_and_development/who_ we_are_how_we_work/clinical_trials/our_commit ment_to_data_sharing.htm.

A total of 3139 respiratory samples were analyzed in this study (Supplementary Table 1 Table 1 ). The cumulative sensitivity was 93.7%, 100.0%, and 100.0% for samples with [ 10,000 copies (* C t \ 29.9), 100,000 copies (* C t \ 26.6), and 1,000,000 copies (* C t \ 23.0) ( Table 1) , respectively.

For samples from symptomatic patients, the assay reached an overall relative sensitivity of 68.4% (95% CI 62.9-73.5, N = 313), increasing Table 2 ).

To analyze the overall assay performance, samples were sorted by SARS-CoV-2 RNA copies/ ml and individual sensitivities calculated for sets of 20 samples (shown using a heat map in Fig. 2a) . Of note, the sensitivity of the assay was 95% and 50% at concentrations of around 2 9 10 5 and 2 9 10 4 -5 9 10 3 viral copies/ml, respectively, but \ 10% for samples with \ 10 3 copies/ml. When the analysis was focused on samples containing RNA levels beyond 95% sensitivity of the immunoassay (10,000 copies/ ml; C t \ 29.9), a significant linear correlation was observed between the Elecsys SARS-CoV-2 Antigen assay COI value and SARS-CoV-2 RNA copies/ml (p \ 0.0001; r 2 = 0.786; Fig. 2b ). Also, in this group, a sensitivity of 98.3% (95% CI 94.0-99.8) was achieved in the sub-analysis of patients tested \ 5 days after symptom onset. Fig. 2 a Elecsys SARS-CoV-2 Antigen assay results in relation to cobas SARS-CoV-2 RNA copies/ml. Dark blue fields represent positive and light blue fields represent negative Elecsys SARS-CoV-2 Antigen test results. On left side, sensitivity was calculated every & 20 samples included in the interval of the heatmap (copies/ml); b correlation between Elecsys SARS-CoV-2 Antigen assay (log10 [COI], y-axis) and cobas SARS-CoV-2 PCR assay (log10 [RNA copies/ml], x-axis). Closed blue circles: measurements within linear range considered for the trendline and correlation (SARS-CoV-2 RNA C 10 5 copies/ml; COI C 1). Open blue circles: measurements outside of the linear range of Elecsys SARS-CoV-2 or one cobas SARS-CoV-2 PCR Target negative, not considered for trendline and correlation. p \ 0.0001; r 2 = 0.786. Ag antigen, COI cutoff index, C t cycle threshold

The introduction of rapid SARS-CoV-2 antigen testing into the market was met with a lot of initial enthusiasm because of the promise of broad availability and easy application. While there are many different methods for antigen detection in clinical practice, such as enzymelinked immunosorbent assays and electrochemiluminescence immunoassays, the first wave of commercial tests consisted predominantly of rapid, point-of-care LFTs with optical read-outs. Preliminary performance data demonstrated good sensitivity in high-viralload samples, but low concentrations of virus are frequently missed, thus leading to the sobering conclusion that antigen tests cannot provide the same definitive results as RT-qPCR [5, [11] [12] [13] . Although PCR tests are analytically superior, their effectiveness for infection control has often been hampered by supply shortages and reporting delays.

Recent studies have shown a correlation between the probability of recovering infectious virus from clinical samples and the time of onset of symptoms as well as RNA concentrations ([ 1 9 10 6 copies/ml) within the specimen [14, 15] . This suggests that the risk of transmission is highest during the early stages of disease, when viral titers are at their peak, which would also facilitate detection by antigen test. A recent study by McKay et al. [16] demonstrated good correlation between positive antigen test and positive viral culture; however, negative viral culture should not be misinterpreted as a reliable indicator for ruling out transmission risk. It has further been suggested that a low sensitivity test can be just as effective at detecting infections as a highly sensitive one, if testing can be made possible at increased frequency [6] . Most currently available SARS-CoV-2 antigen tests utilize a rapid, point-of-care LFT format, thus relying on a high degree of human interaction. They are unsuitable for automation and additional human operators are required at the same proportion as testing capacity is scaled up. While there are clear benefits of a point-of-care test that can be carried out with minimal training and requiring no specialized equipment [17] , testing en masse (e.g., in healthcare facilities) is vastly more efficient when performed automatically using high-throughput immunoassay platforms.

In this study, we stratified samples to assess the relative sensitivity of the new Elecsys SARS-CoV-2 Antigen assay for the cobas immunoassay analyzer systems using the gold standard, RT-qPCR. Although the overall relative sensitivity for the entire sample set was 60.2%, among those patients that were tested in the first 5 days from the onset of symptoms, the sensitivity increased to 83.2%. Furthermore, the Elecsys SARS-CoV-2 Antigen assay was able to detect 93.7% of patients with elevated risk of transmission (C t \ 30 and accordingly RNA levels [ 10,000 copies/ml). Overall relative specificity was determined as 99.9% in a total of 2747 negative samples. These results are in line with existing preliminary data on the clinical performance of other available SARS-CoV-2 antigen tests [12, 18] , including other highthroughput SARS-CoV-2 antigen tests like the Lumipulse Ò [19] , demonstrating a cumulative sensitivity of 55.2% compared with PCR.

The main limitation of the present study is that it is not a head-to-head comparison of the different systems on the market, and future investigation is needed to address this concern. Additionally, as antigen tests show the best performance in people with symptoms and within a certain number of days since symptom onset [3] , the presence of symptoms should be considered when interpreting results from these assays. However, the high-throughput Elecsys SARS-CoV-2 Antigen assay might be a valuable analytic option in the coming months in the struggle to control the pandemic.

The novel Elecsys SARS-CoV-2 Antigen immunoassay showed good performance in a broad set of clinical samples compared to RT-PCR. It demonstrated a high sensitivity (93.7%) in detecting samples containing high concentrations of viral RNA (C 10 4 viral RNA copies/ ml), indicative of high transmission potential. Furthermore, a relative specificity of 99.9% ensures a minimal rate of false positives, consequently saving resources on confirmationtesting and unnecessary quarantine. The ability to run the test on high-throughput immunoassay platforms and a time-to-result of approximately 18 min allow for en masse deployment of SARS-CoV-2 antigen testing, for example, in the context of centralized largescale testing schemes. Data Availability. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Open Access. This article is licensed under a Creative

Commons Attribution-Non-Commercial 4.0 International License, which permits any non-commercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creativecommons.org/licenses/by-nc/4.0/.

Laboratory testing for coronavirus disease 2019 (COVID-19) in suspected human cases

Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR

Centers for Disease Control and Prevention. Interim Guidance for Antigen Testing for SARS-CoV-2. 2021

Clinical evaluation of the Roche SD Biosensor rapid antigen test for SARS-CoV-2 in Municipal Health Service Testing Site, The Netherlands

Handling and accuracy of four rapid antigen tests for the diagnosis of SARS-CoV-2 compared to RT-qPCR

Rethinking covid-19 test sensitivity-a strategy for containment

ElecsysÒ SARS-CoV-2 Antigen method sheet v1

Roche Molecular Systems Inc. cobasÒ SARS-CoV-2 Qualitative assay for

Pushing beyond specifications: evaluation of linearity and clinical performance of the cobas 6800/8800 SARS-CoV-2 RT-PCR assay for reliable quantification in blood and other materials outside recommendations

Diagnostic tests. 1: sensitivity and specificity

Low performance of rapid antigen detection test as frontline testing for COVID-19 diagnosis

Comparison of seven commercial SARS-CoV-2 rapid pointof-care antigen tests

Evaluation of the accuracy and ease-of-use of Abbott PanBio-a WHO emergency use listed, rapid, antigen-detecting point-of-care diagnostic test for SARS

Viral cultures for COVID-19 infectious potential assessment-a systematic review

Virological assessment of hospitalized patients with COVID-2019

Performance evaluation of serial SARS-CoV-2 rapid antigen testing during a nursing home outbreak

At-home selftesting of teachers with a SARS-CoV-2 rapid antigen test to reduce potential transmissions in schools

Head-to-head comparison of four antigen-based rapid detection tests for the diagnosis of SARS-CoV-2 in respiratory samples

Comparison of automated SARS-CoV-2 antigen test for COVID-19 infection with quantitative RT-PCR using 313 nasopharyngeal swabs, including from seven serially followed patients