Abstract
Studies comparing antibody responses against SARS-CoV-2 between children and adults show conflicting results in blood and are limited in the mucosa. Furthermore, as the results of studies comparing systemic and salivary immune responses are inconsistent, it is unclear whether determination of salivary antibodies against SARS-CoV-2 could be a non-invasive approach to evaluate the humoral immune response.
In this work, by studying and comparing systemic and salivary IgG antibody responses in vaccinated adults, we observed that the salivary immune response against SARS-CoV-2 in vaccinated adults largely reflects that observed at systemic levels. Higher salivary and systemic antibody concentrations were observed in adults who had schedules that included mRNA-based vaccines, a greater number of exposures, a shorter interval time between last exposure and saliva collection and had been exposed to SARS-CoV-2. Detection of salivary antibodies was associated with schedules that included mRNA-based vaccines, time between last exposure and sample collection, and systemic antibody concentrations. This suggests that salivary antibody detection might be compromised in subjects with lower systemic antibody levels.
Comparison of salivary antibody levels between vaccinated children and adults showed a stronger salivary antibody response against SARS-CoV-2 in vaccinated children. This also remained when antibody levels were compared between children and adults who had similar vaccination schedules or the same number of exposures or interval time between last exposure and saliva collection. A multivariable linear regression analysis confirmed these results showing that age, symptomatic exposure, number of vaccine doses and schedules that included mRNA-based vaccines associated with salivary anti-SARS-CoV-2 IgG antibody levels.
Determination of salivary antibodies against SARS-CoV-2 could be a non-invasive approach to evaluate the short-term immune response in children and adults with multiple exposures, especially in settings where blood sampling cannot be fulfilled. The higher levels of salivary anti-SARS-CoV- 2 IgG antibodies observed in children align with studies reporting stronger systemic antibody responses in children, suggesting their adaptive immune responses might be more efficient both locally and systemically. Further studies are required to characterize the salivary microenvironment for a better understanding of the difference observed in the local humoral immune response between children and adults.
Introduction
While several advances have been made in the understanding of SARS-CoV-2 infection and its associated pathology, certain aspects related to the immune response remain uncertain. The lower availability of blood samples from children coupled with the late administration of COVID-19 vaccines in this population led to less knowledge about their immune response following infection and vaccination in comparison to adults.
It is still debatable whether children and adults show similar systemic humoral immune responses against SARS-CoV-2 after infection and vaccination. Studies have reported similar [1, 2], lower [3, 4] or stronger and longer-lasting [5, 6] systemic antibody responses to SARS-CoV-2 infection in children than adults. There are also conflicting results regarding antibody responses following vaccination, with studies showing higher [6, 7], similar [8–10] or lower [4] SARS-CoV-2-specific antibody levels in children compared to adults.
Information about the mucosal immune response against SARS-CoV-2 in children is limited. Using saliva-based tests, the prevalence of salivary anti-SARS-CoV-2 IgG antibodies in unvaccinated and vaccinated children has been studied [11–13], further showing that salivary antibody levels increase following infection or exposure [12, 14] and vaccination [13]. However, studies comparing salivary antibody responses against SARS-CoV-2 between children and adults have yielded contradictory results, showing higher [15] or lower [4] salivary antibody levels in children than in adults.
On the other hand, results from studies comparing systemic and salivary humoral immune responses to SARS-CoV-2 are inconsistent. While several studies have shown a positive correlation between systemic and salivary anti-SARS-CoV-2 IgG antibody levels in both adults [16–18] and children [11, 15], other studies have shown a weak correlation [19, 20]. We have also previously shown that SARS-CoV-2 breakthrough infections in vaccinated adults boost antibody levels not only in the blood but also in the salivary compartment [21]. Therefore, although several studies have demonstrate the validity of the determination of specific salivary antibodies as indicators of seroconversion after natural infection or vaccination [16–18, 22], is still unclear whether saliva could be used as an alternative to blood for the determination of antibodies against SARS-CoV-2.
In this work we investigated and compared the systemic and salivary IgG antibody responses against SARS-CoV-2 in vaccinated adults, to analyze whether the determination of salivary antibodies against SARS-CoV-2 could be a non-invasive approach to evaluate the humoral immune response following infection or vaccination. Salivary antibody levels were also compared between vaccinated children and adults to analyze whether the specific humoral immune response differed within the salivary compartment.
Materials and methods
Study Design, Participants and Samples
The results of adults presented in this study are part of an ongoing observational prospective cohort study started at the beginning of the pandemic among healthcare workers from the Academia Nacional de Medicina, to study the humoral immune response against SARS-CoV-2 after vaccination and/or infection. The results of the children included in this work are part of an observational cohort study started on February 2022 to analyze the prevalence of salivary anti- SARS-CoV-2 IgG antibodies in children.
Blood and saliva sample pairs (n = 101) from adults who had received the primary vaccination schedule and a booster dose were collected between December 2021-November 2022. Saliva samples (n = 80) from children up to 18 years old who had received two or three vaccine doses were collected between February-August 2022. The median time between last exposure to SARS- CoV-2 antigens (through vaccination, infection or exposure) and sample collection for all samples was 58 (21-270), for adult samples was 50 (21-234) and for children’s samples was 77 (21−270) days (Table 1).
In both adults and children, samples were obtained from unexposed individuals and from those who were exposed to SARS-CoV-2 at any stage of vaccination. Exposed subjects included those with confirmed past SARS-CoV-2 infection and those who were household contacts. All subjects with confirmed past SARS-CoV-2 infection had mild disease based on the World Health Organization classification [23]. Among adults, 49/101 had a confirmed past SARS-CoV-2 infection, 3/101 were household contacts and 49/101 were unexposed subjects. 11/49 of the infected subjects and 3/3 household contacts were exposed before the Omicron variant circulated in Argentina. The remaining 38/49 subjects were infected when the only circulating variant in Argentina was Omicron. Among children, 11/80 had a confirmed past SARS-CoV-2 infection, 35/80 were household contacts and 34/80 were unexposed subjects. 13/46 of the children who have been infected/household contacts were exposed before the Omicron variant circulated in Argentina, while the remaining 33/46 children were exposed when Omicron was the only circulating variant in Argentina. Of the adults exposed to SARS-CoV-2, 48/52 had mild symptoms, while 4/52 reported no symptoms. Of the SARS-CoV-2 exposed children, 26/46 reported mild symptoms, while 20/46 had no symptoms.
Information regarding demographic characteristic, SARS-CoV-2 exposure and vaccination schedules are detailed in Table 1. Vaccines from the same platform were administered on similar dates and with similar interval time between doses. This study was approved by the Academia Nacional de Medicina Ethics Committee.
Written informed consent was obtained from adult participants and from parents for children participants.
Sample collection and processing
Plasma or serum samples were obtained immediately after centrifugation of the peripheral blood and stored in aliquots at −20°C until used. For saliva sampling, individuals spat their first saliva of the day into a tube, without drinking, brushing teeth or eating, before collection. Saliva samples were centrifuged at 17,000 x g for 10 min (4°C) and the supernatant was stored at −20°C until used.
SARS-CoV-2 Antibody ELISA
Detection of anti-spike SARS-CoV-2 IgG antibodies both in blood and saliva was performed by ELISA (COVIDAR-IgG, Laboratorios Lemos S.R.L, Buenos Aires, Argentina). The plates of the assay are coated with a purified mixture of the spike protein and the receptor binding domain (RBD) of the SARS-CoV-2 from the original viral variant from Wuhan (GenBank: MN908947). Determination of specific antibodies in blood was performed following the manufacturer’s instructions [21]. Determination of salivary antibodies was performed with the conditions we had previously established for the salivary measurements [21]. Basically, salivary IgG anti-spike antibodies were determined following the manufacturer’s instructions, without performing the first sample dilution. Antibody concentrations in binding antibody units (BAU) per mL (BAU/mL) were obtained by interpolating the OD 450 nm values of the samples into a calibration curve performed with the provided standard (400 BAU/mL).
Statistical Analysis
Data analyses were performed using the GraphPad 10.2.3 Prism software (GraphPad Software, San Diego, CA, USA). Geometric mean concentrations (GMC) of specific antibody levels with 95% confidence intervals (95% CI) were calculated. Antibody levels between groups were compared with Mann-Whitney test. The Spearman coefficient of rank correlation was used to assess the correlation between specific salivary and blood antibodies. A multivariable linear regression model was performed with age, sex, symptomatic exposure, number of vaccine doses, number of exposures, vaccination schedules that included mRNA-based vaccines, time between most recent antigen exposure and sample collection as independent variables and salivary antibody levels as dependent variable. The linear regression coefficients (β) with 95% CI were calculated. In all cases, a value of p< 0.05 was considered indicative of a significant difference.
Results
IgG specific salivary and blood antibody responses in vaccinated adults
Specific antibodies were detected in 100% of blood samples (GMC BAU/mL, 95% CI: 3154, 2461−4043) and in 82% of saliva samples (GMC BAU/mL, 95% CI: 67.0, 42.8−104.7). Subjects with higher systemic antibody levels (≥GMC: 3154 BAU/mL) also showed higher salivary antibody concentrations than those with lower systemic antibody levels (GMC BAU/mL, 95% CI, Saliva: 156.8, 92.7−265.2 vs 24.2, 11.1−52.7; p< 0.001). Furthermore, higher systemic antibody levels were found in subjects with detectable salivary antibodies compared to those without detectable salivary antibodies (GMC BAU/mL, 95% CI, Blood: 3841, 2979−4952 vs 1316, 701.1−2469; p< 0.01) (Figure 1A-D). Specific antibody levels in blood and saliva were positively correlated (r= 0.5, p<0.0001).
Blood and salivary anti-SARS-CoV-2 antibody levels were analyzed in vaccinated adults according to: vaccination schedules (A), number of exposures (B), time between last exposure and sample collection (C), exposure to SARS-CoV-2 (D). IgG anti-spike antibody concentrations (BAU/mL) with geometric means are shown. Dotted line indicates the assay detection limit (4.03 BAU/mL). Subjects with detectable salivary antibodies (white circles or squares) and those without detectable salivary antibodies (black circles or squares) are shown. p values were determined by Mann-Whitney test.
Therefore, we analyzed how variables previously reported to be associated with systemic anti- SARS-CoV-2 IgG antibody levels, relate to antibody levels in both the blood and the salivary compartment. As we observed similar results among subjects who received one or two doses of mRNA-based vaccines, they were combined into a single group (non-mRNA + mRNA (3d)) for the presentation of the results. Levels of anti-spike SARS-CoV-2 IgG antibodies were higher in subjects who received three doses of schedules that included mRNA-based vaccines (non-mRNA + mRNA (3d)) compared to those observed in individuals who received three doses of schemes not including mRNA-based vaccines (non-mRNA (3d)) (GMC BAU/mL, 95% CI, Blood: 6335, 3990−10058 vs 2494, 1875−3318; p< 0.01; Saliva: 183.3, 106.5−315.6 vs 47.2, 27.1−82.2; p< 0.01)
(Figure 1A). Subjects with a greater number of exposures to SARS-CoV-2 antigens (through vaccination, infection or exposure) showed higher antibody concentrations than those with a lower number of exposures (GMC BAU/mL, 95% CI, Blood: 5586, 4237−7365 vs 1615, 1162−2245; p< 0.0001; Saliva: 99.0, 52.6−186.4 vs 43.2, 23.5−80.0; p< 0.01) (Figure 1B). Antibody concentrations decreased as the interval time between last exposure and sample collection increased (GMC BAU/mL, 95% CI, Blood: 5448, 4087−7263 vs 1847, 1305−2614; p< 0.0001; Saliva: 179.3, 113.3−283.7 vs 22.0, 11.1−43.5; p< 0.0001) (Figure 1C). Subjects exposed to SARS-CoV-2 showed higher antibody levels than those unexposed (GMC BAU/mL, 95% CI, Blood: 5183, 3871−6940 vs 1807, 1275−2561; p< 0.0001; Saliva: 104.5, 54.8−200.0 vs 42.9, 23.6−78.0; p< 0.01) (Figure 1D).
Using Fisher’s exact test, detection of specific antibodies in saliva was associated with schedules that included mRNA-based vaccines (p< 0.05; OR=7.8) (Figure 1A), the time between last exposure and sample collection (p< 0.001; OR=7.8) (Figure 1C) and systemic antibody concentrations (p< 0.01; OR=7.1) (Figure 1A-D).
Comparison of IgG specific salivary antibody responses between vaccinated children and adults
Considering all samples, higher salivary antibody concentrations were observed in children compared to adults (GMC: 104.0 BAU/mL, 95% CI: 58.7−184.1 vs 66.0 BAU/mL, 95% CI: 42.8−104.7; p< 0.05). No significant differences were found in antibody levels between adults receiving three doses of non-mRNA-based vaccines (non-mRNA (3d)) and children receiving two doses of non-mRNA-based vaccines (GMC: 47.2 BAU/mL, 95% CI: 27.1−82.2 vs 27.3 BAU/mL, 95% CI: 13.7−54.3; p=0.3). However, antibody levels in adults who received three doses of schedules that included mRNA-based vaccines (non-mRNA + mRNA (3d)) (GMC: 183.3 BAU/mL; 95% CI: 106.5−315.6) were higher than in children who received two doses of non- mRNA vaccines (p< 0.001). Children who received three doses of mRNA vaccines (mRNA (3d)) (GMC: 486.0 BAU/mL; 95% CI: 220.7–1069) or two doses of a non-mRNA-based vaccine plus one dose of an mRNA-based vaccine (non-mRNA + mRNA (3d)) (GMC: 1997 BAU/mL; 95% CI: 878.6–4538) showed higher antibody concentrations than adults who received three doses of schedules that did or did not include mRNA-based vaccines (p< 0.01) (Figure 2A). Salivary antibody concentrations were higher in children who had three exposures compared to adults who had three exposures (GMC: 138.4 BAU/mL, 95% CI: 64.3−298.0 vs 43.2 BAU/mL, 95% CI: 23.5−80.0; p< 0.001) (Figure 2B). Children who had four exposures (GMC: 495.3 BAU/mL; 95% CI: 178.2–1377) showed higher antibody levels than adults who had three or four exposures (GMC:
Salivary anti-SARS-CoV-2 antibody levels were compared between vaccinated adults and children according to: vaccination schedules (A), number of exposures (B), time between last exposure and sample collection (C), exposure to SARS-CoV-2 (D). IgG anti-spike antibody concentrations (BAU/mL) with geometric means are shown. Dotted line indicates the assay detection limit (4.03 BAU/mL). p values were determined by Mann- Whitney test.
99.0 BAU/mL; 95% CI: 52.6–186.4) (p< 0.001) (Figure 2B). Antibody concentrations were higher in children compared to adults who had the same interval time (≤58 days) between last exposure and saliva collection (GMC: 396.0 BAU/mL, 95% CI: 185.8−844.1 vs GMC: 156.3 BAU/mL, 95% CI: 98.0−249.6; p< 0.01) (Figure 2C). Unexposed adults showed lower antibody levels than exposed children (GMC BAU/mL, 95% CI: 42.9, 23.6−78.0 vs 150.8, 74.5−305.5; p< 0.001) (Figure 2D). Age (β1: -24.6, 95% CI: -39.2−-9.9; p< 0.001), symptomatic exposure (β2: 871.7, 95% CI: 249.5−1494; p< 0.01), number of vaccine doses (β3: 881.4, 95% CI: 90.5−1672; p< 0.05) and schedules that included mRNA-based vaccines (β4: 726.0, 95% CI: 189.5−1263; p< 0.01) associated with salivary antibody levels in a multivariable linear regression analysis (p< 0.0001).
Discussion
Data from studies comparing antibody responses against SARS-CoV-2 after infection or vaccination between children and adults are conflicting in blood [1–10] and limited in mucosa [4, 15]. On the other hand, as studies comparing systemic and salivary humoral immune responses to SARS-CoV-2 have yielded contradictory results [11, 15, 16–22], it remains yet to be determined whether saliva could be used as an alternative to blood for the determination of antibodies against SARS-CoV-2.
In this work, we investigated and compared the systemic and salivary IgG antibody responses against SARS-CoV-2 in vaccinated adults. Analysis of different variables previously reported to be related to systemic anti-SARS-CoV-2 IgG antibody levels showed similar results for blood and saliva. Higher antibody concentrations were observed in both the blood and saliva of adults who had schedules that included mRNA-based vaccines, a greater number of exposures, a shorter interval time between last antigens exposure and saliva collection and had been exposed to SARS- CoV-2. Furthermore, a positive correlation was found between systemic and salivary anti-SARS- CoV-2 IgG antibody levels, with higher salivary antibody concentrations observed in subjects with higher systemic antibody levels. However, specific antibodies were detected in 100% of blood samples and in 82% of saliva samples. Detection of specific salivary antibodies was associated with schedules that included mRNA-based vaccines, time between last exposure and sample collection, and systemic antibody concentrations. These results, together with the fact that most subjects without detectable salivary antibodies showed lower systemic antibody levels, suggest that salivary antibody detection would appear to depend on variables that influence systemic antibody levels.
Therefore, although our results showed that the salivary immune response against SARS-CoV-2 in vaccinated adults largely reflects that observed at systemic levels, salivary antibody detection might be compromised in subjects presenting lower systemic antibody concentrations. This is in agreement with previously reported results showing that salivary antibody responses were mainly detected in subjects showing higher serum antibody levels [24]. However, we observed that the prevalence of salivary antibodies was high in both adults (82%) and children (81%) [12], unlike to that previously reported in unvaccinated infected subjects [11, 13, 24] but in agreement with those reported in vaccinated subjects [13, 18, 22], suggesting that multiple exposures are required to enhance systemic antibody levels and allow salivary antibody detection.
Salivary antibody concentrations were also compared between vaccinated children and adults to analyze whether specific antibody responses differed within the salivary compartment. We observed a higher salivary antibody response against SARS-CoV-2 in vaccinated children compared to adults. This also remained when antibody levels were compared between children and adults who had similar vaccination schedules (three doses combining mRNA and non-mRNA based vaccines) or the same number of exposures. Furthermore, when considering the same interval time between last exposure and saliva collection, children also showed higher antibody concentrations than adults. Higher antibody levels were also observed in children exposed to SARS-CoV-2 compared to unexposed adults. These results were confirmed in a multivariable linear regression analysis, showing that age, symptomatic exposure, number of vaccine doses and schedules that included mRNA-based vaccines were associated with salivary anti-SARS-CoV-2 IgG antibody levels.
The biological and molecular basis for a favorable outcome after SARS-CoV-2 infection in children compared to adults is not fully understood. Several hypotheses have tried to explain the milder presentation of the disease in children, including a protective role of pre-existing cross-reactive antibodies against seasonal coronaviruses, a lower expression of the angiotensin-converting enzyme 2 (ACE2) and a less propensity to develop an exacerbated pro-inflammatory response, among others [25]. Furthermore, it is not entirely clear the relative contribution of the antibody response to SARS-CoV-2 to the favorable outcome in the pediatric population. In the current work, we showed that children had higher salivary anti-SARS-CoV-2 IgG antibody concentrations than adults. Given that mucosal immunity plays a key role in prevention and early defense against infection, being the entry route for the virus and the site of first encounter with the immune response, it is tempting to speculate that the stronger salivary antibody response observed in children could help prevent or limit infection, thus contributing to the favorable outcome observed in the pediatric population.
As salivary IgG is derived primarily from circulating IgG through transudation [26] and the salivary IgG response in adults highly reflects the systemic antibody response, it is very likely that the salivary antibody response observed in children also largely reflects that present in blood. However, given that there is also some local IgG production, we cannot rule out that a difference in the local immune response mounted may also account to the observed difference in the salivary antibody response against SARS-CoV-2 between children and adults.
Limitations of this study include the lack of information on the systemic antibody response in children, which precluded direct comparison between the systemic and salivary immune responses in children. We had no information on neutralization titers and specific IgA antibody responses in both blood and saliva. Besides, for reasons inherent to the vaccination policies in our country, adults mainly received non-mRNA based primary vaccination schedules and children only received mRNA vaccines as third doses. Therefore, a comparison could not be made with samples from adults vaccinated with three doses of mRNA vaccines or from children vaccinated with three doses of non-mRNA-based vaccines. Despite this, we observed that among subjects who received three vaccine doses, children who only received one dose of mRNA-based vaccine showed higher salivary antibody concentrations than adults who also received one or even two doses of mRNA- based vaccines, suggesting that other factors inherent to children, in addition to the vaccination schedules received, could be contributing to the higher salivary antibody response observed in children. Information about SARS-CoV-2 variants involved in the infections/exposures and their association with antibody levels was not available. However, most infections/exposures both in adults and children occurred when the only circulating variant in Argentina was Omicron.
Altogether, our results showed that the salivary immune response against SARS-CoV-2 in vaccinated adults largely reflects that observed at systemic levels. However, salivary antibody detection might be compromised in subjects with low systemic antibody concentrations, which could be related to lower immune stimulation resulting from less immunogenic vaccines, fewer exposures, or excessive time since last exposure, potentially leading to false negative results compared to blood. Nevertheless, since saliva sampling is non-invasive and allows self-collection unlike blood sampling, salivary antibody determination could be performed as an initial screening to evaluate the immune response against SARS-CoV-2, and in those subjects who do not show detectable salivary antibodies, determination of antibodies in blood should be performed whenever possible. Besides, as salivary immune response is indicative of the systemic antibody response, salivary antibody determination could guide vaccination strategies through the administration of booster doses in those subjects who lack detectable salivary antibodies, so that they can achieve high systemic antibody levels. In conclusion, the determination of salivary antibodies against SARS-CoV-2 could be a non-invasive approach for the evaluation of the short-term immune response in subjects with multiple exposures, either resulting from vaccination combined with infection or from the administration of immunogenic vaccination schedules, both in the adult and pediatric population, especially in settings where blood sampling cannot be fulfilled.
Furthermore, the higher levels of salivary anti-SARS-CoV-2 IgG antibodies observed in children are in agreement with studies reporting stronger systemic antibody responses in the pediatric population. This suggests that children could have more efficient adaptive immune responses, both at local and systemic levels, compared to adults. Additional studies focusing on the characterization of the salivary microenvironment are required for a better understanding of the difference observed in the local humoral immune response between children and adults.
Data Availability
All data produced in the present study are available upon reasonable request to the authors
Data availability statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Funding/Support
This study was supported by a grant from CONICET (PIP 11220210100378CO).
Conflict of Interest Disclosure
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Ethics Statement
This study was approved by the Academia Nacional de Medicina Ethics Committee.
Consent Statement
Written informed consent was obtained from adult participants and from parents for children participants.
Abbreviations
binding antibody units (BAU) per mL (BAU/mL), geometric mean concentrations (GMC), 95% confidence intervals (95% CI).
Acknowledgments
The authors thank all enrolled children and adults for their participation in this study. Some aspects of this work could not have been fulfilled without the generous contribution of the IIHEMA and the Academia Nacional de Medicina, who provide financial support to our ongoing research. This study was supported by a grant from CONICET (PIP 11220210100378CO).
Footnotes
The formatting of the manuscript has been improved.
References
