key: cord-1028462-3pts3s3x
authors: Montalvo Villalba, María C.; Sosa Glaria, Elena; Rodriguez Lay, Licel de los A.; Valdés Ramirez, Odalys; Vallina García, Dayana; Arencibia Garcia, Amely; Martinez Alfonso, Javier; Menes Llerena, Dunia M.; Torres Pérez, Loida; Resik Aguirre, Sonia R.; Guzman Tirado, Maria G.
title: Performance evaluation of Elecsys SARS‐CoV‐2 Antigen immunoassay for diagnostic of COVID‐19
date: 2021-10-29
journal: J Med Virol
DOI: 10.1002/jmv.27412
sha: e6d70c8cb1425979940c4a6909290909d6ce4ff3
doc_id: 1028462
cord_uid: 3pts3s3x

One of the challenges for control and prevention of severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) infection is the early diagnostic at the point of care. Several tests based on qualitative antigen detection have been developed; one of these is Elecsys SARS‐CoV‐2 Antigen immunoassay (Roche Diagnostics). In total, 523 nasopharyngeal swabs were randomly selected with the aims to evaluate sensitivity, specificity, cross‐reactivity, positive and negative predictive value (PPV, NPV), and agreement of Elecsys SARS‐CoV‐2 Antigen immunoassay using reverse transcription‐polymerase chain reaction (RT‐PCR) STAT‐NAT® coronavirus disease‐2019 as reference test. Cross‐reactivity was estimated using samples positive by RT‐PCR to other respiratory viruses (influenza virus, parainfluenza virus, rhinovirus, coronavirus OC43, and HKU1). The overall sensitivity of Elecsys SARS‐CoV‐2 Antigen was 89.72% (288/321); specificity was 90.59% (183/202); and cross‐reactivity to other respiratory viruses were not detected. Elecsys SARS‐CoV‐2 Antigen immunoassay showed a high sensitivity in samples with cycle threshold value <30, which ranged from 92.81% to 95.40%, independently of symptoms. PPV and NPV were 93.81% and 84.72%, respectively. The κ coefficient was 0.79 (95% confidence interval: 0.73–0.84), showing substantial agreement between both tests. The results suggest Elecsys SARS‐CoV‐2 Antigen immunoassay could be used as an alternative to RT‐PCR testing, or in complement with it, to identify infectious individuals and reduce SARS‐CoV‐2 transmission.

sensitivity and specificity, for SARS-CoV-2 diagnosis is the real-time reverse transcription-polymerase chain reaction (RT-PCR). However, to get accurate results with RT-PCR, it is necessary to establish a good workflow, with trained personal, equipment specifics, and laboratories (preanalytical, analytical). Furthermore, the results could take time, although not more than 24 h.

To reduce infection and make opportune decisions at the point of care (POC), a rapid and reliable diagnostic test for detection of SARS-CoV-2 infection is needed. For this purpose, several rapid tests for qualitative antigen detection (PanbioTM COVID-19 Ag Rapid Test, COVID-19 Ag Respi-Strip) have been developed. The principle of the lateral-flow immunochromatographic is the basis of the majority of rapid tests; obtaining results in minutes without any instrumentation. LUMIPULSE is another assay, which uses a LUMIPULSE G600II automated immunoassay analyzer, to detect SARS-CoV-2 antigen quantitatively; furthermore, results are ready in 30 min. 2 SOFIA SARS Antigen FIA, STANDARD F COVID-19 Ag FIA, and FIC assay are other antigen tests that also need an analyzer. 3, 4 Nucleocapsid is the most used SARS-CoV-2 protein for antigen detection, since it is the most expressed virus derived-protein, and its nucleotide sequence is conserved in time. 5 Nonetheless, there are other lateral flow assays (LFA) designed to identify viral spike proteins using glycans, nanoparticles, and antibodies. 6 Peto 7 developed an LFA to detect all SARS-CoV-2 structural proteins, which showed good sensitivity and specificity. Clinical samples collected from the upper respiratory tracts, like nasopharyngeal swabs (NPS) and saliva, have been used for antigen detection; being sensitivity the main problem of these assays. 8, 9 Elecsys SARS-CoV-2 Antigen test is an immunoassay that runs on the cobas ® e systems, produced by Roche Diagnostic for the qualitative detection of SARS-CoV-2 antigen in NPS. The manufacturer reports Elecsys SARS-CoV-2 Antigen tests to have a throughput of 300 tests per hour, depending on the analyzer used.

In 2020, Cuba achieved good control of the COVID-19 pandemic; but since January 2021 with the third wave, the number of confirmed cases increased. to improve SARS-CoV-2 diagnosis, the Cuban Health System inaugurated 27 laboratories for molecular diagnosis around the country. Currently, the National Reference Samples were selected randomly based on the number of samples necessary to perform the assay. In total, 523 NPS were collected, grouped according to the epidemiological definitions of NPS samples used in the NRLRV ( Table 1) 

This assay is an in vitro electrochemiluminescence immunoassay Sensitivity and specificity evaluated with RT-PCR Cobas SARS-CoV-2 by Roche manufacturers were 94.5% (symptomatic patients) and

99.9% (asymptomatic and symptomatic subjects), respectively. 10

Qualitative detection of SARS-CoV-2 RNA from NPS was assessed using STAT-NAT ® COVID-19 MULTI (SENTINEL Diagnostic) multiplex assay; based on the simultaneous detection of RdRP and ORF1b

genes. Its limit of detection is 10 copies/reaction and the assay was performed in Rotor-Gene Q 3000 (QIAGEN). Samples with a threshold cycle (C t ) of fewer than 40 were considered positives.

The relationship between the amount of viral RNA, measured as C t value of the RT-PCR, and SARS-CoV-2 antigen detection was estimated. Therefore, C t values were stratified as high (C t ≤ 26), moderate (26 < C t ≤ 35), and low (35 < C t ≤ 40) viral concentration.

Sensitivity was calculated as the proportion of samples reactive to Elecsys SARS-CoV-2 Antigen immunoassay relative to the total number of samples (n = 321), that were positive by SARS-CoV-2

RT-PCR testing.

This percentage was considered as the number of samples nonreactive with Elecsys SARS-CoV-2 Antigen immunoassay, relative to the total amount of samples that were negative (n = 202) by SARS-CoV-2 RT-PCR assay. In this group, cross-reactivity and viral interference were assessed using 17 NPS samples negative to SARS-CoV-2 RT-PCR. These NPS were collected as part of the routine diagnosis for respiratory virus infections during 2019. First, the samples were tested with the Elecsys SARS-CoV-2 Antigen assay; the test included the following samples: influenza virus A (n = 9) and B (n = 1); parainfluenza virus type 3 (n = 3); coronavirus OC43 (n = 2) and HKU1 (n = 1); and rhinovirus (n = 1 3.2 | Sensitivity of Elecsys SARS-CoV-2 Antigen assay relative to C t and comparing C t with COI 

The overall mean of the CV of 42 evaluated samples was 4.47% (95% 

Early diagnostic and the application of a reasonable test algorithm to detect SARS-CoV-2 infection are critical to reduce its transmission.

Elecsys SARS-CoV-2 Antigen assay decreases the time of results to minutes, with a high throughput of samples per hour.

The sensitivity of Elecsys SARS-CoV-2 Antigen assay was lower (89.72% vs. 94.5%) in comparison with that reported by the manufacturer, who used NPS from symptomatic individuals with C t < 30.

However, overall sensitivity independently of the presence of symptoms is in agreement with the minimum performance requirements set by WHO for antigen tests at ≥80%. 12 In agreement with other studies, the sensitivity of Elecsys SARS-CoV-2 Antigen assay decreased with the increase of C t values; probably due to declining of virus concentration under the detection limit of the assay. For this reason, some authors recommend these assays in the early phases of acute infection. 13 Good performance was reported by Porte et al. 3 in samples with C t ≤25 using rapid SARS-CoV-2 antigen detection tests. It should also be noted that C t values and COI were obtained with different viral targets. SARS-CoV-2 RT-PCR was based on the simultaneous detection of RdRP and ORF1b genes, indispensable for viral replication and usually detected at low viral load (≥10 1 copy/µl). 14 Conversely, Elecsys SARS-CoV-2 Antigen assay detects nucleocapsid protein expressed in the infectious viral progenies, which usually do not exceed the amount of viral RNA production, reducing the sensitivity of the antigen assays. 15 Correlation between C t values and antigen concentrations was not detected. Amendola et al., 8 PPV is a measure of the performance of an assay and depends on infection prevalence. Overall, PPV (93.81%) was higher than NPV (84.72%). As a result, the likelihood that a "reactive" result is correct is highest in samples tested for reference of confirmatory cases, virus tracing at 5 days of diagnosis, and contact cases.

According to WHO guidelines, validation of antigen tests should be carried out using stratification by viral load expressed by C t values and days postonset of symptoms, being the last criteria one limitation of this study.

Clinical performance of antigen tests depends on the rate of viral replication, the kinetics of viral protein expression, the VTM used, and the quality of the specimen. Since the VTMs that were available in POC were not described by the manufacturer in their evaluation, it was important to confirm that Elecsys SARS-CoV-2 Antigen assay is suitable for these NPS samples.

Barlev-Gross et al. 6 described the higher specificity of LFA based on spike protein in comparison with nucleocapsid LFA. Nevertheless, in the current context with the new SARS-CoV-2 mutations in the spike protein; FN results could arise due to these concerning variants not being recognized.

In conclusion, RT-PCR is the standard technique for SARS-CoV-2 detection, the capacity to support it is limited in low-and middleincome countries. NRLRV receives a high volume of samples, which could lead to a deficit in reagents and disposable materials. The proposition is to incorporate Elecsys SARS-CoV-2 Antigen assay in the diagnosis algorithm of SARS-CoV-2 as a screening method owing to its sensitivity. However, FP results must be evaluated with RT-PCR. This means that samples tested for reference of confirmatory cases, virus tracing at 5 days from diagnosis, contact cases, and suspect cases with reactive results by Elecsys SARS-CoV-2 Antigen assay should be considered as positive; and only FPs need confirmation by RT-PCR to rule out infection.

COVID-19) dashboard

Evaluation of Lumipulse ® G SARS-CoV-2 antigen assay automated test for detecting SARS-CoV-2 nucleocapsid protein (NP) in nasopharyngeal swabs for community and population screening

Evaluation of a novel antigenbased rapid detection test for the diagnosis of SARS-CoV-2 in respiratory samples

Accuracy of a nucleocapsid protein antigen rapid test in the diagnosis of SARS-CoV-2 infection

Expression, purification and immunological characterization of recombinant nucleocapsid protein fragment from SARS-CoV-2

Spike vs nucleocapsid SARS-CoV-2 antigen detection: application in nasopharyngeal swab specimens

COVID-19: Rapid antigen detection for SARS-CoV-2 by lateral flow assay: a national systematic evaluation of sensitivity and specificity for mass-testing

Saliva is a valid alternative to nasopharyngeal swab in chemiluminescence-based assay for detection of SARS-CoV-2 antigen

Salivary SARS-CoV-2 antigen rapid detection: a prospective cohort study

The measurement of observer agreement for categorical data

Antigen-detection in the diagnosis of SARS-CoV-2 infection using rapid immunoassays Interim guidance 11 September 2020 COVID-19: laboratory and diagnosis

Performance of a novel diagnostic assay for rapid SARS-CoV-2 antigen detection in nasopharynx samples

Development of multiplex real-time RT-PCR assay for the detection of SARS-CoV-2

SARS-coronavirus-2 replication in Vero E6 cells: replication kinetics, rapid adaptation and cytopathology

The SARS-CoV-2 nucleocapsid protein is dynamic, disordered, and phase separates with RNA

Variation in false-negative rate of reverse transcriptase polymerase chain reaction-based SARS-CoV-2 tests by time since exposure

Performance evaluation of Elecsys SARS-CoV-2 Antigen immunoassay for diagnostic of COVID-19

The authors would like to thank Beatriz A. Acebedo and Maylin G.Alvarez for their technical assistance. They are grateful to colleagues of COVIred, Latin-American net for exchanging scientific experiences in COVID-19. Also, they recognize Marlon J. A. Milian, and in a special way Dr. Martin Plummer for the grammatical corrections.

The authors declare that there are no conflict of interests. 

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to ethical restrictions.