Validation of a new automated chemiluminescent anti-SARS-CoV-2 IgM and IgG antibody assay system detecting both N and S proteins in Japan

Abstract

PCR methods are presently the standard for the diagnosis of Coronavirus disease 2019 (COVID-19), but additional methodologies are needed to complement PCR methods, which have some limitations. Here, we validated and investigated the usefulness of measuring serum antibodies against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) using the iFlash3000 CLIA analyzer. We measured IgM and IgG titers against SARS-CoV-2 in sera collected from 26 PCR-positive COVID-19 patients, 53 COVID-19-suspected but PCR-negative patients, and 20 and 100 randomly selected non-COVID-19 patients who visited our hospital in 2020 and 2017, respectively. The repeatability and within-laboratory precision were obviously good in validations, following to the CLSI document EP15-A3. Linearity was also considered good between 0.6 AU/mL and 112.7 AU/mL for SARS-CoV-2 IgM and between 3.2 AU/mL and 55.3 AU/mL for SARS-CoV-2 IgG, while the linearity curves plateaued above the upper measurement range. We also confirmed that the seroconversion and no-antibody titers were over the cutoff values in all 100 serum samples collected in 2017. These results indicate that this measurement system successfully detects SARS-CoV-2 IgM/IgG. We observed four false-positive cases in the IgM assay and no false-positive cases in the IgG assay when 111 serum samples known to contain autoantibodies were evaluated. The concordance rates of the antibody test with the PCR test were 98.1% for SARS-CoV-2 IgM and 100% for IgG among PCR-negative cases and 30.8% for SARS-CoV-2 IgM and 73.1% for SARS-CoV-2 IgG among PCR-positive cases. In conclusion, the performance of this new automated method for detecting antibody against both N and S proteins of SARS-CoV-2 is sufficient for use in laboratory testing.

Fig 1. Linearity analyses of SARS-CoV-2 antibody titer.

The dilution linearities of SARS-CoV-2 IgM (A, B) and SARS-CoV-2 IgG (C, D) were investigated. A sample was diluted with non-reactive serum in 5 to 8 steps; each sample was then analyzed with two replicates. The linearity studies for SARS-CoV-2 IgM (E) and SARS-CoV-2 IgG (F) were performed according to CLSI EP06-A. The sample of high antibody titer was diluted with the sample of low antibody titer in 6 to 9 steps; each sample was the analyzed with two steps.

Fig 2. Prozone phenomena and a hook effect of SARS-CoV-2 antibody titer.

The prozone phenomena of SARS-CoV-2 IgM (A) and SARS-CoV-2 IgG (B) were investigated. We diluted two serum samples from infected patients with non-reactive serum in 10 steps; each sample was then analyzed with two replicates. We diluted a serum sample with a saline in 10 steps and measured SARS-CoV-2 IgG titer (C); each sample was then analyzed with two replicates.

Fig 3. Time course of serum antibody titers in COVID-19 subjects.

The time courses of the SARS-CoV-2 IgM and SARS-CoV-2 IgG titers in sera collected before and after the onset of COVID-19 were examined in seven patients.

Fig 4. Serum antibody titers in sera from 2017.

The SARS-CoV-2 IgM/IgG titers of sera collected from subjects (n = 100) in 2017 were measured.

Fig 5. Interference from autoantibodies in SARS-CoV-2 IgM/IgG assay.

We collected sera from patients with autoimmune diseases and measured the SARS-CoV-2 IgM (A) and IgG (B) titers. *p < 0.05, **p < 0.01. Control, randomly selected outpatients who visited the hospital in 2020 (n = 20); RF, rheumatoid factor-positive group (n = 25); dsDNA, anti-double-strand DNA antibody-positive group (n = 26); M2, anti-mitochondrial M2 antibody-positive group (n = 20); P-ANCA, myeloperoxidase antineutrophil cytoplasmic antibody-positive group (n = 20); SS-A, anti-Sjögren’s syndrome A antibody-positive group (n = 20); Suspected, suspected COVID-19 group (n = 53); Positive, COVID-19-positive group (n = 26).