Evaluation of a Broad Panel of SARS-CoV-2 Serological Tests for Diagnostic Use

Abstract

Serological testing is crucial in detection of previous infection and in monitoring convalescent and vaccine-induced immunity. During the Severe Acute Respiratory Syndrome Corona Virus 2 (SARS-CoV-2) pandemic, numerous assay platforms have been developed and marketed for clinical use. Several studies recently compared clinical performance of a limited number of serological tests, but broad comparative evaluation is currently missing. Within this study, a panel of 161 sera from SARS-CoV-2 infected, seasonal CoV-infected and SARS-CoV-2 naïve subjects was enrolled to evaluate 16 ELISA/ECLIA-based and 16 LFA-based tests. Specificities of all ELISA/ECLIA-based assays were acceptable and generally in agreement with the providers’ specifications, but sensitivities were lower as specified. Results of the LFAs were less accurate as compared to the ELISAs, albeit with some exceptions. We found a sporadic unequal immune response for different antigens and thus recommend the use of a nucleocapsid protein (N)- and spike protein (S)-based test combination when maximal sensitivity is necessary. Finally, the quality of the immune response in terms of neutralization should be tested using S-based IgG tests.

Spectrum Bias and Individual Strengths of SARS-CoV-2 Serological Tests—A Population-Based Evaluation

“…Comparing different tests to each other, we found similar concordance of our ELISA_G with the COBAS_N, whereas those previous findings also suggested high comparability of our ELISA_G with the commercially available Euroimmun test [5]. In contrast to others reporting lower agreement for the COBAS_S with neutralization [33], we found a high concordance of the COBAS_S with neutralization, in line with the manufacturers statement in the validation [34] and also in agreement with others comparing COBAS_S to ACE-2 Inhibition as a surrogate for neutralization [35].…”

“…A similar strategy, though using the SARS-CoV-2 nucleoprotein (N) instead of the spike protein for antibody detection, was applied like for many other now commercially available SARS-CoV-2 antibody tests-for validation and approval of the Roche COBAS/ELECSYS SARS-CoV-2-N (nucleoprotein) platform, where 204 samples of 69 PCR-positive patients were tested at various time points, and 5272 pre-December 2019 samples were used to test for specificity [2]. Though various studies and meta-analyses have been carried out comparing different serological tests during the pandemic [3][4][5], most of these studies used predefined clinical samples relying on pre-COVID-19 and symptomatic/PCR-verified cohorts or other, non-population based sera or convenience samples to derive performance metrics for test accuracy.…”

Ultrasensitive Detection of SARS-CoV-2 Spike Proteins Using the Thio-NAD Cycling Reaction: A Preliminary Study before Clinical Trials

“…In other words, simultaneous detection of spike and nucleocapsid protein detection enabled us to detect SARS-CoV-2 at the level of 10 4 viruses/assay. This simultaneous detection offers not only good detection sensitivity, but also avoids false positive testing results, as suggested by Werner et al [19].…”

“…It is difficult to detect the spike proteins of SARS-CoV-2 [13], but the simultaneous detection of nucleocapsid proteins and spike proteins is suitable for COVID-19 antigen tests [19]. As a follow-up to our previous study of the detection of SARS-CoV-2 nucleocapsid proteins, our present study developed an ultrasensitive thio-NAD cycling ELISA to detect SARS-CoV-2 spike proteins, which were prepared from UVB-irradiated and inactivated viruses.…”

“…The nucleocapsid proteins of SARS-CoV-2, however, are hardly distinguishable from those of conventional coronaviruses and SARS-CoV [16][17][18]. Thus, to avoid false-positive test results, a combination of tests against both the nucleocapsid and spike antigen proteins is strongly recommended [19], because the spike protein, which is the reason for the name coronavirus, is coronavirus-specific [20,21]. The binding affinity of the spike protein in SARS-CoV-2 with ACE2 is higher than that of SARS-CoV [22].…”