High-Throughput Multiplex SARS-CoV-2 IgG Microsphere Immunoassay for Dried Blood Spots: A Public Health Strategy for Enhanced Serosurvey Capacity

Abstract

Early in the pandemic when diagnostic testing was not widely available, serosurveys played an important role in estimating the prevalence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in different populations. Dried blood spots (DBS), which can be collected in nonclinical settings, provide a minimally invasive alternative to serum for serosurveys. We developed a Luminex-based SARS-CoV-2 microsphere immunoassay (MIA) for DBS that detects IgG antibodies to nucleocapsid (N) and spike subunit 1 (S1) antigens. The assay uses a 384-well plate format and automated liquid handlers for high-throughput capacity. Specificity was assessed using a large collection of prepandemic DBS and well-characterized sera. Sensitivity was analyzed using serology data from New York State SARS-CoV-2 serosurvey testing and matched diagnostic test results. For DBS, the specificity was 99.5% for the individual N and S1 antigens. Median fluorescence intensity (MFI) values for DBS and paired sera showed a strong positive correlation for N (R² = 0.91) and S1 (R² = 0.93). Sensitivity, assessed from 1,134 DBS with prior laboratory-confirmed SARS-CoV-2 infection, ranged from 83% at 0 to 20 days to 95% at 61 to 90 days after a positive test. When stratified using coronavirus disease 2019 (COVID-19) symptom data, sensitivity ranged from 90 to 96% for symptomatic and 77 to 91% for asymptomatic individuals. For 8,367 health care workers reporting detailed symptom data, MFI values were significantly higher for all symptom categories. Our results indicate that the SARS-CoV-2 IgG DBS MIA is sensitive, specific, and well-suited for large population-based serosurveys. The ability to readily modify and multiplex antigens is important for ongoing assessment of SARS-CoV-2 antibody responses to emerging variants and vaccines. IMPORTANCE Testing for antibodies to SARS-CoV-2 has been used to estimate the prevalence of COVID-19 in different populations. Seroprevalence studies, or serosurveys, were especially useful during the early phase of the pandemic when diagnostic testing was not widely available, and the resulting seroprevalence estimates played an important role in public health decision making. To achieve meaningful results, antibody tests used for serosurveys should be accurate and accessible to diverse populations. We developed a test that detects antibodies to two different SARS-CoV-2 proteins in dried blood spots (DBS). DBS require only a simple fingerstick and can be collected in nonclinical settings. We conducted a robust validation study and have demonstrated that our test is both sensitive and specific. Furthermore, we demonstrated that our test is suitable for large-scale serosurveys by testing over 56,000 DBS collected in a variety of community-based venues in New York State during the spring of 2020.

Keywords: COVID-19; SARS-CoV-2; dried blood spot; immunoassays; serology; serosurvey.

FIG 1

Assay specificity determination. (A) DBS and serum collected prior to December 2019 were tested using three lots of beads for DBS (A, B, C) and one bead lot for serum. *, P < 0.005 (B) Serum specimens containing antibodies to respiratory viruses, including other human coronaviruses, were tested. Lower line is cutoff for indeterminate results; upper line is cutoff for reactive results. MFI, median fluorescence intensity; N, nucleocapsid; S1, spike subunit 1.

FIG 2

Analysis of paired serum and laboratory-prepared DBS. MFI, median fluorescence intensity; N, nucleocapsid; S1, spike subunit 1.

FIG 3

Concordance with commercially available SARS-CoV-2 immunoassays. Specimen panels from Access Biologicals were tested, and MFI index values for nucleocapsid (N) and spike subunit 1 (S1) were compared to results from other commercial assays as provided by the panel provider. (A) Thirty sera with comparator results from GSD SARS-CoV-2 IgG (Gold Standard Diagnostics). (B) Fifty sera with comparator results from Liaison SARS-CoV-2 S1/S2 IgG (Diasorin). (C) Twenty sera with comparator results from Advia Centaur SARS-CoV-2 total (Siemens). (D) Twenty-eight plasma samples collected from a single individual between days 1 and 98 after COVID-19 symptom onset. The reactive cutoff values for each test are listed next to the name.

FIG 4

Sensitivity analysis on specimens with prior laboratory-confirmed SARS-CoV-2 infection. (A) Reactivity for individual nucleocapsid (N) and spike (S1) bead sets and reactivity based on “OR” and “AND” result criteria by days after positive diagnostic test. (B) Reactivity for N and S1 bead sets by days after positive diagnostic test and presence/absence of COVID-19 symptoms. (C) Overall assay reactivity (reactive = N or S1 reactive) by days after positive diagnostic test and presence/absence of symptoms. Error bars = 95% CI; *, 0.01 < P < 0.05.

FIG 5

Percent reactive and mean index values for spike S1 and nucleocapsid by symptom category for health care workers. The numbers who reported “yes” and “no” for each symptom are listed. Error bars indicate 95% confidence intervals.