The study protocol was approved by the Research Ethics Committee of the Faculty of Medicine, Chulalongkorn University (IRB number 572/63), and that committee waived the requirement for consent because the samples used were obtained during the course of routine preventive measures and were de-identified and anonymous.
Specimen collection
A cross-sectional set of serum samples derived from 245 patients in whom SARS-CoV-2 infection had been confirmed via real-time RT-PCR of nasal swab specimens or who had a previous SARS-CoV-2-positive diagnostic test result in their medical record from the hospital and public health center were used in the study. Samples from all 245 patients (138 from the National Blood Center, 107 from Bangkok Metropolitan Administration Hospital) were obtained during the period between COVID-19 symptom onset and serum sample acquisition for serology testing to monitor the presence of antibodies against SARS-CoV-2. For samples from asymptomatic patients, the period was calculated from the first date of SARS‑CoV-2 detection via real-time RT-PCR to the date of serum sample acquisition. The time enrolling in this study ranged from 2 days to 60 days. A total of 130 blood donor specimens collected in 2018 were used as negative control samples.
IgM/IgG rapid diagnostic test
The Standard Q COVID-19 IgM/IgG Combo test (SD biosensor, Chungcheongbuk-do, Republic of Korea) is a rapid chromatographic immunoassay for the qualitative detection of SARS-CoV-2-specific antibodies. The test utilizes recombinant COVID-19 nucleocapsid protein conjugated to colloidal gold particles as detectors, and SARS-CoV-2-specific IgM and IgG antibodies are detected simultaneously. A violet test line is visible if SARS-CoV-2 antibodies are present in the specimen. The visual intensity of the violet test line varies depending on the amount of SARS-CoV-2 antibodies present in the specimen. A visual intensity ratio ranging from 0 to 3 compared to a control line is scored, where 0 = no intensity, 1 = weak intensity, 2 = medium intensity, and 3 = strong intensity. In the current study all results of this assay were interpreted by three different people independently, to measure positivity.
Spike protein IgG and IgA ELISAs
Anti-SARS-CoV-2 IgG and IgA ELISAs (EUROIMMUN, Lubeck, Germany) are enzymatic immunoassays that provide semi-quantitative in vitro determination of human IgG and IgA antibodies against the S1 domain of the SARS-CoV-2 spike protein. Optical density was measured at 450 nm. The results can be evaluated semi-quantitatively by calculating the ratio of the control or patient sample over the extinction of the calibrator. Samples with a cutoff ratio higher than 1.1 were considered positive. All ELISAs were tested and interpreted automatically using the EUROIMMUN Analyzer I-2P machine (Lubeck, Germany).
IgG nucleoprotein chemiluminescent microparticle immunoassay
The SARS-CoV-2 IgG chemiluminescent microparticle immunoassay (CMIA) (Abbott Ireland Diagnostics Division, Sligo, Ireland) is an automated two-step immunoassay for the quantitative detection of IgG antibodies against SARS-CoV-2. Samples were analyzed using an Abbott ARCHITECT I 1000SR instrument. The ARCHITECT I system calculates the mean calibrator chemiluminescent signal. Results derived from test samples are measured as relative light units (RLU) and determined via comparison with the calibrator, and the cutoff is 1.4. There is a direct relationship between the amount of SARS-CoV-2 antibodies in the sample and the RLU detected by the system.
Statistical analysis
Sensitivity and specificity for detection of the presence of SARS-CoV-2 IgG antibodies were evaluated. Statistical analyses were performed using IBM SPSS statistics for Windows, version 21 (IBM Corp., Armonk, NY, USA). Agreement rates and kappa coefficients between immunoassays and RDTs were analyzed. The association between the amount of SARS-CoV-2 antibodies and the time period after COVID-19 symptom onset was analyzed using the Chi-square test. Correlations between IgG RDT visual intensity scores and IgG immunoassay relative ratios were analyzed using one-way ANOVA. P < 0.05 was considered statistically significant.