Saliva SARS-CoV-2 Antibody Prevalence in Children

Abstract

COVID-19 patients produce circulating and mucosal antibodies. In adults, specific saliva antibodies have been detected. Nonetheless, seroprevalence is routinely investigated, while little attention has been paid to mucosal antibodies. We therefore assessed SARS-CoV-2-specific antibody prevalence in serum and saliva in children in the Netherlands. We assessed SARS-CoV-2 antibody prevalence in serum and saliva of 517 children attending medical services in the Netherlands (irrespective of COVID-19 exposure) from April to October 2020. The prevalence of SARS-CoV-2 spike (S), receptor binding domain (RBD), and nucleocapsid (N)-specific IgG and IgA were evaluated with an exploratory Luminex assay in serum and saliva and with the Wantai SARS-CoV-2 RBD total antibody enzyme-linked immunosorbent assay in serum. Using the Wantai assay, the RBD-specific antibody prevalence in serum was 3.3% (95% confidence interval [CI]. 1.9 to 5.3%). With the Luminex assay, we detected heterogeneity between antibodies for S, RBD, and N antigens, as IgG and IgA prevalence ranged between 3.6 and 4.6% in serum and between 0 and 4.4% in saliva. The Luminex assay also revealed differences between serum and saliva, with SARS-CoV-2-specific IgG present in saliva but not in serum for 1.5 to 2.7% of all children. Using multiple antigen assays, the IgG prevalence for at least two out of three antigens (S, RBD, or N) in serum or saliva can be calculated as 3.8% (95% CI, 2.3 to 5.6%). Our study displays the heterogeneity of the SARS-CoV-2 antibody response in children and emphasizes the additional value of saliva antibody detection and the combined use of different antigens. IMPORTANCE Comprehending humoral immunity to SARS-CoV-2, including in children, is crucial for future public health and vaccine strategies. Others have suggested that mucosal antibody measurement could be an important and more convenient tool to evaluate humoral immunity compared to circulating antibodies. Nonetheless, seroprevalence is routinely investigated, while little attention has been paid to mucosal antibodies. We show the heterogeneity of SARS-CoV-2 antibodies, in terms of both antigen specificity and differences between circulating and mucosal antibodies, emphasizing the additional value of saliva antibody detection next to detection of antibodies in serum.

Keywords: SARS-CoV-2; antibodies; children; humoral immunity; prevalence; saliva.

FIG 1

Age distribution, comorbidity, and sex. (A) The distribution of age across the inclusion period of 24 weeks is depicted for the total number of cases (n = 517), with each bar representing 4 weeks. (B) Comorbidity and male/female ratio of the study sample is calculated for the nonmissing values (n = 514, excluding 3 missing values for comorbidity).

FIG 2

SARS-CoV-2 antibody prevalence estimates in serum and saliva. (A) Prevalence estimate of SARS-CoV-2 RBD total antibodies in the Wantai assay for serum. (B and C) Prevalence estimates of SARS-CoV-2 S- (gray bars), RBD- (green bars), and N-specific (blue bars) IgG (B) and IgA (C) in the Luminex assays for serum (solid bars) and saliva (hatched bars). Prevalence was the calculated proportion with a value above the determined cutoff out of nonmissing values. Estimates are shown with 95% confidence intervals. S, trimeric SARS-CoV-2 spike protein; RBD, the monomeric receptor binding domain of the SARS-CoV-2 spike protein; N, SARS-CoV-2 nucleocapsid protein.

FIG 3

Correspondence of Luminex assays for serum and saliva. (A and B) SARS-CoV-2 S-, RBD-, and N-specific IgG (A) and IgA (B), measured in paired serum and saliva samples (n = 413) by Luminex assay, expressed as MFI. Only samples also measured in the Wantai assay are shown, with the positive Wantai results indicated in blue. Serum and saliva are plotted against each other to reveal the differences between each compartment. The red dotted lines are the cutoff values to discriminate positive and negative measurements in the Luminex assays. Percentages represent data points which are positive for both compartments or for a single compartment as the percentage of total positives in each graph. S, trimeric SARS-CoV-2 spike protein; RBD, protein of only the monomeric receptor binding domain of the SARS-CoV-2 spike protein; N, SARS-CoV-2 nucleocapsid protein; MFI, median fluorescence intensity.

FIG 4

Correspondence of Luminex assays for different antigens and isotypes. (A) S-, RBD-, and N protein-specific IgG, measured in serum (n = 509, left) and saliva (n = 430, right) in the Luminex assay, expressed as MFI. S and N are plotted against each other, and RBD-positive samples are shown in orange. The red dotted lines are the cutoff values to discriminate positive and negative measurements for S and N. (B) SARS-CoV-2 S-specific IgG and IgA in serum (n = 487, left) and saliva (n = 413, right) measured in the Luminex assay, expressed as MFI. IgG and IgA are plotted against each other to reveal the correspondence between the two isotypes. Samples that were also positive in the Wantai RBD total antibody assay are shown in blue. S, trimeric SARS-CoV-2 spike protein; RBD, monomeric receptor binding domain of the SARS-CoV-2 spike; N, SARS-CoV-2 nucleocapsid protein; MFI, median fluorescence intensity.

FIG 5

Combined SARS-CoV-2 IgG antibody prevalence. Combined prevalence of the explorative Luminex assay of IgG in serum (pink bars), in saliva (orange bars), or in serum and saliva combined (yellow bars). The combined prevalence was calculated for children positive for 3/3 SARS-CoV-2 antigens (crosshatched bars) at least 2/3 antigens (hatched bars), and for children positive for at least 1/3 antigens (solid bars). S, trimeric SARS-CoV-2 spike protein; RBD, monomeric receptor binding domain of the SARS-CoV-2 spike; N, SARS-CoV-2 nucleocapsid protein.