Methodological approaches to optimize multiplex oral fluid SARS-CoV-2 IgG assay performance and correlation with serologic and neutralizing antibody responses

Abstract

Background. Oral fluid (hereafter, saliva) is a non-invasive and attractive alternative to blood for SARS-CoV-2 IgG testing; however, the heterogeneity of saliva as a matrix poses challenges for immunoassay performance. Objectives. To optimize performance of a magnetic microparticle-based multiplex immunoassay (MIA) for SARS-CoV-2 IgG measurement in saliva, with consideration of: i) threshold setting and validation across different MIA bead batches; ii) sample qualification based on salivary total IgG concentration; iii) calibration to U.S. SARS-CoV-2 serological standard binding antibody units (BAU); and iv) correlations with blood-based SARS-CoV-2 serological and neutralizing antibody (nAb) assays. Methods. The salivary SARS-CoV-2 IgG MIA included 2 nucleocapsid (N), 3 receptor-binding domain (RBD), and 2 spike protein (S) antigens. Gingival crevicular fluid (GCF) swab saliva samples were collected before December, 2019 (n=555) and after molecular test-confirmed SARS-CoV-2 infection from 113 individuals (providing up to 5 repeated-measures; n=398) and used to optimize and validate MIA performance (total n=953). Combinations of IgG responses to N, RBD and S and total salivary IgG concentration (μg/mL) as a qualifier of nonreactive samples were optimized and validated, calibrated to the U.S. SARS-CoV-2 serological standard, and correlated with blood-based SARS-CoV-2 IgG ELISA and nAb assays. Results. The sum of signal to cutoff (S/Co) to all seven MIA SARS-CoV-2 antigens and disqualification of nonreactive saliva samples with >15 μg/mL total IgG led to correct classification of 62/62 positives (sensitivity [Se]=100.0%; 95% confidence interval [CI]=94.8%, 100.0%) and 108/109 negatives (specificity [Sp]=99.1%; 95% CI=97.3%, 100.0%) at 8-million beads coupling scale and 80/81 positives (Se=98.8%; 95% CI=93.3%, 100.0%] and 127/127 negatives (Sp=100%; 95% CI=97.1%, 100.0%) at 20-million beads coupling scale. Salivary SARS-CoV-2 IgG crossed the MIA cutoff of 0.1 BAU/mL on average 9 days post-COVID-19 symptom onset and peaked around day 30. Among n=30 matched saliva and plasma samples, salivary SARS-CoV-2 MIA IgG levels correlated with corresponding-antigen plasma ELISA IgG (N: ρ=0.67, RBD: ρ=0.76, S: ρ =0.82; all p<0.0001). Correlations of plasma SARS-CoV-2 nAb assay area under the curve (AUC) with salivary MIA IgG (N: ρ=0.68, RBD: ρ=0.78, S: ρ =0.79; all p<0.0001) and with plasma ELISA IgG (N: ρ=0.76, RBD: ρ=0.79, S: ρ =0.76; p<0.0001) were similar. Conclusions. A salivary SARS-CoV-2 IgG MIA produced consistently high Se (>98.8%) and Sp (>99.1%) across two bead coupling scales and correlations with nAb responses that were similar to blood-based SARS-CoV-2 IgG ELISA data. This non-invasive salivary SARS-CoV-2 IgG MIA could increase engagement of vulnerable populations and improve broad understanding of humoral immunity (kinetics and gaps) within the evolving context of booster vaccination, viral variants and waning immunity.

Subclinical SARS-CoV-2 Infections and Endemic Human Coronavirus Immunity Shape SARS-CoV-2 Saliva Antibody Responses

“…51 In this study, we utilized saliva anti-SCV2 N IgG as a marker for possible prior infection. Saliva anti-SCV2 N IgG has been reported to have a high sensitivity for prior SARS-CoV-2 infections, 49 and we found an 86% sensitivity for known post-vaccine infections in our cohort. Further, we obtained saliva antibody levels at baseline and then regularly (approximately 4 times a year), increasing the likelihood that we would not miss a transient infection-induced increase in saliva anti-SCV2 N antibodies.…”

“…47,48 At least 2 studies that found increases in saliva IgA attempted to screen out unrecognized infections by testing for serum anti-SCV2 N IgG. 24,29 While serum anti-SCV2 N is often positive shortly after infection, the sensitivity of anti-SCV2 N IgG for diagnosing prior infection is highly variable (52% -100%) 49,50 and its half-life postinfection is short, with one study estimating it to be only 35 days. 51 In this study, we utilized saliva anti-SCV2 N IgG as a marker for possible prior infection.…”

SARS-CoV-2 Antibody Prevalence among Industrial Livestock Operation Workers and Nearby Community Residents, North Carolina, 2021 to 2022

“…Oral fluid samples were separated from device sponges by centrifugation and tested for SARS-CoV-2 nucleocapsid (N), receptor-binding domain (RBD), and spike (S) IgG, using a multiplex bead-based immunoassay based on Luminex technology, which has been described previously 19 , 20 , 52 . Briefly, the multiplex assay included SARS-CoV-2 N, RBD, and S antigens coupled to magnetic microparticles and a background control bead coated with bovine serum albumin (BSA).…”

Detection of anti-SARS-CoV-2 salivary antibodies in vaccinated adults

“…However, the invasive process needed for blood collection can limit the use of serological tests. Therefore, introducing saliva samples, composed of both gingival fluid and salivary secretion, can represent huge advantages as saliva collection is easy, non-invasive, and can be self-administrated with a lower risk of contamination (11,12). Regarding daily mucosal antibody production, immunoglobulin A (IgA) production far exceeds the combined production of all other Ig isotypes.…”