Demographics of study participants

A total of 4,371 serum samples from participants was collected between May and August 2024, following multiple waves of COVID-19 (Fig. 1b). Participants ranged in age from 6 months to 80 years (mean (SD) age: 17.9 (16.1) years) and comprised 57% females (Table 1). Study participants were stratified into five age groups: 6 months–4 years (19%, n = 830), 5–11 years (26%, n = 1,135), 12–17 years (16%, n = 717), 18–59 years (36%, n = 1,547), and ≥ 60 years (3%, n = 142) (Fig. 1c). Overall, 62.9% of participants had received at least one dose of a COVID-19 vaccine containing the ancestral strain composition (Table 1), while 36.9% remained unvaccinated, including all children aged 6 months–4 years, 59.1% of those aged 5–11 years and 12% of individuals aged 12–17 years (Fig. 1d) and 0.2% of participants did not report their vaccination history. Only 44.2% of participants self-reported prior SARS-CoV-2 infection (Table 1), with the highest frequency observed in adults (56.2%) and the lowest in young children aged 6 months–4 years (28.2%) (Fig. 1e).

Comparison of anti-N Ig and anti-RBD Ig levels and seroprevalence across age groups

The overall seroprevalence of anti-N Ig was 95.1% (95% CI: 94.5–95.8%) (Fig. 2a). The highest seropositivity rate was observed in adults aged 18–59 years (97.5%), followed by adolescents aged 12–17 years (96.9%), children aged 5–11 years (95.1%), children aged 6 months–4 years (90.4%), and older adults aged ≥ 60 years (88.7%). These findings suggest that the majority of the population has been exposed to SARS-CoV-2. Furthermore, children aged 6 months–4 years and those aged 5–11 years had significantly lower median anti-N Ig levels than those aged 12–17 years and 18–59 years, but similar levels to participants aged 60 years and older (Fig. 2b).

Anti-N Ig and anti-RBD Ig seroprevalence and their responses across age groups in general population of Thailand, between May and August 2024. (a) Seroprevalence of anti-N Ig (b) Median levels of anti-N Ig (c) Seroprevalence of anti-RBD Ig and (d) Geometric mean titers (GMT) of anti-RBD Ig. Participants were categorized into five groups: 6 months–4 years (6m–4Y), 5–11 years, 12–17 years, 18–59 years, and ≥ 60 years. The numbers above the graphs in panels a and c indicate the seropositive rates of SARS-CoV-2 antibodies with 95% confidence intervals (95% CI), and the numbers below the graphs indicate the number of samples for each age group. Number above the graphs in panels b and d indicate the median and geometric mean titers for each age group. Group differences were analyzed using the Kruskal–Wallis test. Horizontal dashed lines indicate the limit of detection for anti-N Ig (1.0 COI) and anti-RBD Ig (0.8 U/mL). Statistical significance is indicated as follows: *p < 0.05; **p < 0.01; ***p < 0.001.

The seroprevalence of anti-RBD Ig across all age groups was remarkably high at 98.8% (95% CI: 98.5–99.1%) (Fig. 2c). Notably, all adults aged ≥ 18 years had detectable anti-RBD Ig, while seropositivity rates among children and adolescents ranged from 96.5 to 99.7%. The majority of individuals aged 5–11 years, 12–17 years, 18–59 years, and ≥ 60 years had detectable anti-RBD Ig concentrations in the range of 1,000 to < 10,000 U/mL, whereas most children aged 6 months–4 years had lower concentrations, ranging from 100 to < 1,000 U/mL. Similarly, individuals aged 12–17 years, 18–59 years and ≥ 60 years had significantly higher geometric mean titers of anti-RBD Ig than those aged 6 months–4 years and 5–11 years (Fig. 2d). Based on both anti-N Ig and anti-RBD Ig, the majority of the population was seropositive for both antibodies, indicating prior infection or hybrid immunity.

To explore the influence of age on the SARS-CoV-2 antibody response, we analyzed the correlation between antibody levels and age of participants ranging from 6 months to 80 years. Anti-N Ig levels exhibited a weak positive correlation with age (Fig. 3a). A similar pattern was observed for anti-RBD Ig by age (Fig. 3b). To determine when children primarily experienced SARS-CoV-2 infection, we analyzed the seropositive rate of SARS-CoV-2 antibodies by age among children aged 6 months–4 years. The results showed that the average anti-N Ig seropositivity rate was 72.5% in children aged 6 months to ≤ 1 year, indicating that a substantial proportion had been infected during their first year of life (Fig. 3c). A steadily increasing trend in anti-N Ig seropositivity was observed, with rates exceeding 90% from two years of age onward, suggesting the gradual accumulation of SARS-CoV-2 exposure over time. Similarly, the anti-RBD Ig seropositivity was remarkably high, with seropositivity rate exceeding 90% (Fig. 3d).

SARS-CoV-2 Ig response distribution by age. The association between age (6 months–80 years) and (a) anti-N Ig and (b) anti-RBD Ig levels was analyzed using Spearman’s rank correlation. Scatter plots show (c) anti-N Ig and (d) anti-RBD Ig seropositivity in children aged 6 months–4 years. Numbers above the plots in panels c and d indicate the seropositive rates for each age subgroup including 6 month to ≤ 1 year, > 1 year to ≤ 2 years, > 2 year to ≤ 3 years, > 3 year to ≤ 4 years and > 4 year to < 5 years, respectively. A second-order polynomial trendline was fitted to the data in panels c and d, with the corresponding equation displayed on each plot.

Neutralization against WT and JN.1 across different age groups

To assess antibody-mediated protection against currently circulating variants, neutralizing activity against WT and JN.1 was measured using the focus reduction neutralization test. Overall, 89% (95% CI: 83.8–93%) of serum samples exhibited neutralizing activity against WT, while 86.5% (95% CI: 81–91%) neutralized JN.1 (Fig. 4a–b). When stratified by age, the seropositive rate of neutralizing antibodies against WT ranged from 97.8 to 100% in participants aged ≥ 12 years, whereas the seropositive rate was 78% in children aged 6 months–4 years and 84.4% in those aged 5–11 years. Furthermore, children aged 6 months–4 years exhibited significantly lower neutralizing antibody titers against WT (GMT: 72.7) compared to older age groups (GMT: 158.5–1871) (Fig. 4a).

Comparison of neutralizing antibody responses against WT and JN.1 across age groups. Participants were categorized into five groups: 6 months–4 years (6m–4Y), 5–11 years, 12–17 years, 18–59 years, and ≥ 60 years. A subset of serum samples (n = 200) was used to measure neutralizing activity against the authentic (a) WT and (b) JN.1 viruses using the focus reduction neutralization test (FRNT). (c) Differences in neutralizing antibody levels against JN.1 relative to WT for each age group. (d) Comparison of neutralizing antibody responses against WT and JN.1 among children aged 6 months–4 years, grouped by birth year. Numbers above the graphs in panels a and b represent the geometric mean titers (GMT), while the numbers below the graphs indicate the seropositive rates of neutralizing antibodies with 95% confidence intervals (95% CI). Closed purple and yellow circles represent WT and JN.1 variants, respectively. The horizontal dashed line indicates the detection limit (FRNT50, GMT = 20), and values were set to 10 for statistical analysis. Differences between groups in panels a and b were assessed using the Kruskal–Wallis test, while comparisons in panels c and d were evaluated using the Wilcoxon signed-rank test. Statistical significance is indicated as follows: ns, not significant; *p < 0.05; **p < 0.01; ***p < 0.001.

In addition, participants aged 12–17 years (91.7%) had the highest seropositive rate of neutralizing antibodies against JN.1, followed by those aged 18–59 years (88.9%), children aged 6 months–4 years (88%), children aged 5–11 years (85.9%), and the lowest rate was observed in individuals aged ≥ 60 years (79.3%). Children aged 6 months–4 years demonstrated higher neutralizing titers against JN.1 compared to other groups, although the difference was not statistically significant and the titers remained relatively low (Fig. 4b).

Notably, children aged 6 months–4 years exhibited significantly higher neutralizing antibody titers against JN.1 compared to WT. Conversely, participants aged 12 years and older displayed the opposite pattern (Fig. 4c). As SARS-CoV-2 variants continue to emerge and replace predominant strains each year, we examined whether birth year influences neutralizing antibody responses in young children. Children aged 6 months to 4 years were grouped by birth year including 2019–2020, 2021, and 2022–2023 (Table S1). Neutralizing antibody titers against JN.1 were significantly higher in children born after 2021 compared to WT (Fig. 4d). These finding suggests that children had detectable neutralizing activity against both WT and JN.1, despite the absence of vaccination, indicating that asymptomatic SARS-CoV-2 infection likely played a key role in eliciting an immune response in this age group.

Effect of unvaccinated and vaccinated with different number vaccine doses on neutralization

To investigate whether the absence of vaccination or varying numbers of vaccine doses influenced neutralizing activity against WT and JN.1, we compared neutralizing responses between WT and JN.1 within each age group and vaccinations status. Overall, unvaccinated individuals exhibited significantly higher neutralizing activity against JN.1 compared to WT (Fig. 5a). Conversely, vaccinated individuals demonstrated a significantly higher neutralizing response against WT than JN.1.

Impact of no vaccination and number dose of vaccines on neutralizing antibody responses against WT and JN.1. Comparison neutralizing antibody titers against WT and JN.1 detetcted in serum from unvaccinated and vaccinated individuals with different number of vaccine dose were performed across (a) all participants and age-stratified groups, including (b) 6 months–11 years, (c) 12–17 years, (d) 18–59 years, and (e) ≥ 60 years. Horizontal dashed lines indicate the detection limit (FRNT50 GMT = 20). (f) Heatmaps represents the p-values from inter-groups pairwise comparisons of neutralizing antibody titers for all participants and age groups with different dose of vaccination (0 V, no vaccination; 2 V, 2 doses; 3 V, 3 doses; 4 V, 4 doses) using Kruskal-Wallis test with Dunn’s multiple comparisons. Numbers above the graphs indicate the geometric mean titer (GMT). The Wilcoxon signed-rank test was used to compare the neutralizing antibody response between WT and JN.1 in panels a-e. Statistical significance is indicated as follows: ns, not significant; *p < 0.05; **p < 0.01; ***p < 0.001.

Among children aged 6 months–11 years, unvaccinated individuals exhibited significantly higher neutralizing activity against JN.1 than WT, whereas vaccinated children displayed comparable titers against both (Fig. 5b). Moreover, significantly higher neutralizing activity against WT was observed in adolescents (12–17 years) who received two or three vaccine doses and in adults (18–59 years) who received three doses compared to JN.1, indicating the WT-imprinted immunity in vaccinated individuals (Fig. 5c–d). In contrast, no significant differences in neutralizing antibody titers between WT and JN.1 were observed in adults (18–59 years) who received two or four doses and elderly participants (≥ 60 years), regardless of the number of vaccine doses (Fig. 5d–e). These findings suggest that infections with later SARS-CoV-2 variants enhance cross-reactive neutralization against early strains in unvaccinated individuals and mitigate WT-imprinted immunity in vaccinated individuals.

We further analyzed neutralizing antibody titers by comparing responses between different age groups and vaccination statuses for each strain (WT and JN.1) to assess whether neutralizing responses varied by age or number of vaccine doses received. The results showed that unvaccinated individuals had significantly lower neutralizing activity against the wild-type (WT) virus compared to vaccinated individuals, regardless of the number of vaccine doses received. In contrast, no significant differences in titers against the JN.1 variant were observed across age groups (Fig. 5f).