2.2. Investigated SARS-CoV-2 Antibody Tests
Seven SARS-CoV-2 IgG assays (Abbott SARS-CoV-2 IgG, Wiesbaden, Germany; DiaSorin Liaison® SARS-CoV-2 S1/2 IgG, DiaSorin, Dietzenbach, Germany; Epitope EDI™ Novel Coronavirus COVID-19 IgG ELISA Kit, Epitope Diagnostics, San Diego, CA, USA; Euroimmun Anti-SARS-CoV-2 ELISA (IgG), EUROIMMUN AG, Lübeck, Germany; Mikrogen recomWell SARS-CoV-2 IgG, Mikrogen GmbH, Neuried, Germany; SARS-CoV-2 ViraChip® IgG Test, Viramed Biotech AG, Planegg, Germany; and SERION ELISA agile SARS-COV-2 IgG, Institute Virion-Serion GmbH, Würzburg, Germany), as well as one total SARS-CoV-2 antibody test (Roche Elecsys Anti-SARS-CoV-2, Roche Diagnostics, Mannheim, Germany), were included in this study. The recombinant antigens of these tests cover the N protein (Abbott, Epitope, Mikrogen, Roche) and the entire S protein (Virion-Serion), as well as its S1 domain alone (Euroimmun) or together with the S2 domain (DiaSorin). The SARS-CoV-2 ViraChip® IgG test kit uses the purified viral surface antigens S1 and S2 and the N antigen, which are all presented separately at a defined position on a nitrocellulose film (Table S1).
All tests were conducted strictly following the recommendations of the manufacturers on an Architect or Alinity machine (Abbott), a Liaison XL (DiaSorin), or a Cobas e 411 (Roche) or, for the assays of Epitope, Euroimmun, Mikrogen Viramed, and Virion-Serion, on the BEP 2000 system (Siemens Healthcare GmbH, Erlangen, Germany), respectively. The ViraChip® test was then evaluated automatically by help of a ViraChip® reader and the ViraChip® Software.
All borderline/gray zone results were counted as positive. Furthermore, raw data from seven assays were converted into relative indices according to the decision limits set by the manufacturer. Thereby, a signal (S)/cut-off (CO) value of <1 was valued as negative and ≥1 as positive, which corresponds to a previous study [22]. Results of the SARS-CoV-2 ViraChip® IgG could not be converted into relative indices and were only given qualitatively as negative or positive. The proportion of correctly identified samples per test relative to the chosen reference method was calculated and given as assay accuracy.
The convalescent sera were also tested with and without avidity reagent (contains urea in denaturating concentrations) in two versions of an IgG line assay (blot v1: Mikrogen recomLine Coronavirus IgG (Avidität), prototype and blot v2: Mikrogen recomLine SARS-CoV-2 IgG (Avidität)/RUO). The first version is based on the N protein of the human coronaviruses (HCoVs) 229E, HKU1, NL63, and OC43, which are used separately as antigens, and on the N protein of the “classic” SARS-CoV from 2002/2003 and SARS-CoV-2. The second improved version still uses the N proteins of the aforementioned HCoVs and of SARS-CoV-2 but has been expanded to include the S1 antigen and its RBD. A Dynablot Plus system was used to process all samples. Blots were then evaluated automatically with a BLOTrix reader and the recomScan software (all from Mikrogen).