Assessment of salivary antibody response to BNT162b2 mRNA COVID-19 vaccination.

To the Editor,

Universal vaccination is promoted as the single most effective strategy in the fight against coronavirus disease 2019 (COVID‐19). The first available vaccine was BNT162b2 (Pfizer/BioNTech, COMIRNATY), which is based on a messenger ribonucleic acid (mRNA) technology and specifically induces the production of antibodies against the spike (S) glycoprotein of the severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2).1, 2 Clinical trials and recent studies point at the development of an efficient and robust systemic humoral response following vaccination, that can be monitored using serological immunoassays registered for quantitative measurement of anti‐SARS‐CoV‐2 S antibodies in serum.1, 2, 3, 4 The humoral immune response might be expected not only in blood but also in the mucosa and salivary glands. Since SARS‐CoV‐2 is mainly transmitted through direct or indirect contact with mucosal membranes, the presence of mucosal antibodies might directly prevent or limit virus transmission. Saliva is an easily accessible and noninvasive sample, and the presence of anti‐SARS‐CoV‐2 S antibodies in saliva has been already confirmed in previously affected COVID‐19 patients.5, 6, 7 However, there is a scarcity of data on the possible presence of anti‐SARS‐CoV‐2 S antibodies in salivary samples obtained from mRNA vaccine recipients. Therefore, the present study aimed to determine the antibody response against SARS‐CoV‐2 S glycoprotein in paired serum and saliva samples from BNT162b2 vaccine recipients and to assess the possible correlation between serum and saliva antibody titers.

The study was performed at the Department of Laboratory Diagnostics, University Hospital Center Zagreb, Croatia, and included 43 adult healthcare workers (median age 52 years, from 27 to 63; 6/43 males) who received both doses of the BNT162b2 vaccine and were not previously affected by COVID‐19. The median time from administration of the second vaccine dose was 71 days (from 71 to 89 days). Paired blood and saliva samples were obtained from each study participant. Blood was drawn in 5 ml serum tubes with a gel separator (Becton Dickinson) while stimulated saliva was collected shortly after the blood draw by a standardized protocol using a Salivette collection device (Sarstedt AG & Co.). All analyses were performed in fresh samples, within 4 h from the collection. Antibody titers in both serum and saliva samples were determined using the quantitative automated electrochemiluminescence immunoassay Elecsys Anti‐SARS‐CoV‐2 S (Roche Diagnostics) applied to Cobas e801 analyzer (Roche Diagnostics). This assay measures total antibodies against SARS‐CoV‐2 S glycoprotein, by using SARS‐CoV‐2 S receptor binding domain recombinant antigens that predominantly capture anti‐SARS‐CoV‐2 S immunoglobulin G (IgG), but also IgA and IgM. The declared repeatability for serum samples ranges from 0.9% to 2.9%, depending on the concentration level, while similar repeatability was obtained in our laboratory for saliva samples, being 1.2% at 1.3 U/ml. The results are expressed in U/ml and values below 0.8 U/ml in serum are considered negative. The interference study performed by the manufacturer proved no impact of endogenous IgA on assay results up to 16 g/L, which is above the expected levels of IgA in saliva, making its interfering effect highly improbable. Antibody titers in saliva were normalized to total protein concentrations, which were measured on Alinity c biochemistry analyzer (Abbott Laboratories) using the automated turbidimetric urinary and cerebrospinal fluid assay whose measuring range (70–2000 mg/L) covers the expected total protein concentrations in saliva.

Data normality was assessed using the Shapiro–Wilk test. Spearman's rank correlation coefficient (ρ) was used to determine the correlation between anti‐SARS‐CoV‐2 S antibody titers in paired serum and saliva samples, p < 0.05 was considered statistically significant. Statistical analysis was performed in MedCalc, version 19.5.2 (MedCalc).

Participation in the study was voluntary and all participants gave their informed consent before enrollment. The study was approved by the University Hospital Center Zagreb Ethics Committee (Protocol no. 8.1‐21/14‐2; 02/21 AG).

From 43 study participants, 4 had antibody titers in saliva below the detection limit (<0.4 U/ml) and were assigned the value of 0.4 U/ml to be included in the statistical analysis. Spearman's ρ was 0.606 (p < 0.001), showing a moderate correlation between antibody titers in serum and saliva (Table 1). The graphical presentation of results provided in Figure 1 reveals the existence of one outlier with significantly higher antibody titers both in serum and saliva, as compared to other participants. The residuals were found to be normally distributed.

Table 1

Correlation between anti‐SARS‐CoV‐2 S antibody titers in paired serum and saliva samples of BNT162b2 mRNA vaccine recipients (N = 43)

	Anti‐SARS‐CoV‐2 S antibody titer in serum (U/mL)	Anti‐SARS‐CoV‐2 S antibody titer in saliva (U/mg proteins)	ρ
Median (IQR)	1274.0	2.5	0.606
	(872.3–1784.0)	(1.7–3.9)	(p < 0.001)

Abbreviations: IQR, interquartile range; ρ, Spearman's rank correlation coefficient

Figure 1

Graphical presentation of the correlation between anti‐SARS‐CoV‐2 S antibody titers in paired serum and saliva samples of BNT162b2 mRNA vaccine recipients (N = 43)

The present study supports recently published data that antibodies against SARS‐CoV‐2 S glycoprotein can be detected in the saliva of BNT162b2 mRNA vaccine recipients,10, 11 and further reveals that the systemic and mucosal antibody response is maintained for more than 2 months from the boost dose, however, exhibiting large interindividual variability. Moreover, it was demonstrated that antibody titers in serum are positively, but only moderately, correlated with paired saliva antibody titers. This finding suggests that the presence of antibodies in saliva might only partly derive from blood and that vaccination induces direct production of antibodies by mucosal‐associated immune cells. Further studies are required to understand the duration and protective value of salivary immune response as well as to identify the specific immunoglobulin classes present in saliva.

CONFLICT OF INTERESTS

The authors declare no conflict of interest.

AUTHOR CONTRIBUTIONS

Ivana Lapić conceived and designed the study, collected samples, performed laboratory analyses, analyzed the data, and wrote the manuscript. Dragana Šegulja collected samples, performed laboratory analyses, and analyzed the data. Dunja Rogić designed the study, analyzed the data, and co‐wrote the manuscript. All authors drafted the article and approved its final version for submission.