Stronger and durable SARS-CoV-2 immune response to mRNA vaccines in 5 – 11 years old children with prior COVID-19

Background and objectives: mRNA vaccines elicit a durable humoral response to SARS-CoV-2 in adults, whereas evidence in children is scarce. This study aimed to assess the early and long-term immune response to the mRNA vaccine in children with or without previous SARS-CoV-2 infection. Methods: In a multicentre prospective observational study, we profiled the immune response to the Pfizer Bio-NTech (BNT162b2) vaccine in 5 – 11-year-old children attending the University Pediatric Hospital of Padua and Bambino-Gesù Hospital in Rome (Italy) from December-2021 to February-2023. Blood samples were collected pre-, 1, and 6-months after vaccination. Neutralizing antibodies (NAbs) and anti-spike-receptor-binding-domain (anti-S-RBD) IgG titers were analyzed through Plaque Reduction Neutralization Test (PRNT) and chemilumi-nescent immune-enzymatic assay (CLIA), respectively. Immune cell phenotypes were analyzed by flow cytometry. Results: Sixty children (26 [43 %] female, median age = 8 years [IQR = 7 – 10.7]) were enrolled in the study, including 46 children with a laboratory-confirmed previous COVID-19 (SARS-CoV-2-recovered) and 14 SARS-Abbreviations

To date, research has predominantly focused on adolescents and adults, and there is a knowledge gap on the long-term immune response to COVID-19 vaccination in children.
Studies on individuals who recovered from COVID-19 have shown an inverse correlation of antigen-specific antibody titers with age, with more intense humoral responses to SARS-CoV-2 infection in young children than adults [10,11].Additionally, children who recovered from SARS-CoV-2 infection had a higher expansion of regulatory T (Tregs) and B (Bregs) cells than adults, likely by mitigating the viral-induced inflammation/chronic immune activation, leading to a higher specific production of anti-SARS-CoV-2 antibodies [12].
The emergence of multiple SARS-CoV-2 variants of concern (VOCs) has raised alarms about the potential increased transmission and evasion from vaccine-conferred immunity [13].Therefore, a better understanding of the antigen-specific immune response and durability after mRNA vaccination in healthy and immunocompromised children with or without previous SARS-CoV-2 infection is necessary to inform public health policies and optimize vaccination strategies in the pediatric population.

Study design and data collection
A prospective, observational study was conducted from December/ 2021 to February/2023 to profile the longitudinal immunogenicity of mRNA COVID-19 vaccine on 5-11-year-old children attending the Pediatric University Hospital of Padua and the Bambino Gesù Children's Hospital in Rome (Italy).
Children met the following criteria for enrollment: a) aged between 5 and 11 years at the time of vaccination and b) had received a 2-doses primary series of the Pfizer BioNTech mRNA COVID-19 vaccine (BNT162b2) (10-μg doses with an interval between doses of 21 days) or c) had received at least one dose of the BNT162b2 within 12 months after previous laboratory-confirmed SARS-CoV-2 infection [14].
Whole blood samples were collected at three time points: prevaccination, ~1 month after vaccination, and ~ 6 months after vaccination.A detailed description of the blood sample collection procedure is reported in the eMethods in the supplement.
Patient follow-up was stopped in the event of a new probable or confirmed close contact with a COVID-19 case and/or a laboratoryconfirmed SARS-CoV-2 infection.
Participants' anonymized demographic and clinical characteristics as well as laboratory findings from enrollment and follow-up were entered into a web-based database using the REDCap® platform (Vanderbilt University, Tennessee).The study protocol is approved by the local Ethics Committee (Prot.N • 0,070,714 of November 24th, 2020; last amendment Prot.N • 0,024,018 del 5/4/2022).Parents/legally authorized representatives were informed of the research proposal and provided written consent to use the routinely collected patient data.

Case identification and definitions
Children who had evidence of SARS-CoV-2 infection (i.e., a molecular or antigenic COVID-19 nasopharyngeal swab (NPS) and/or positive SARS-CoV-2 serology test at the pre-vaccination assessment) preceding vaccination were identified as SARS-CoV-2-recovered.For each SARS-COV-2-recovered participant, the index date of infection was established based on the date of the first positive NPS.SARS-CoV-2 infections before December 11, 2021, or after January 7, 2022, were classified as pre-Omicron or Omicron infections, respectively, based on Italy's predominant circulating SARS-CoV-2 VOC [15].Children were identified as SARS-CoV-2-naïve if they had no analytical evidence of SARS-CoV-2 infection at the pre-vaccination assessment.
Children were further classified into four groups based on the presence, or not, of comorbidities: 1) healthy children (i.e., no comorbidities) (HC), 2) immunocompromised children (i.e., an immuneinflammatory chronic renal and rheumatological disease requiring an immunomodulant therapy) (IC), 3) solid organ transplant recipients (i.e., liver or renal organ transplant recipients on anti-rejection regimen) (SOT), and 4) having a previous diagnosis of multisystem inflammatory syndrome in children (MIS-C).

Flow cytometry
Peripheral blood mononuclear cells (PBMCs) were thawed, washed, and stained for 20 min in the dark with the Live/Dead Fixable Near-IR Dead Cell Stain Kit (Life Technologies, Carlsbad, CA, USA) and with monoclonal antibodies, as previously described (eMethods and eFigure1 in the supplemental) [12].

CXCL10 mRNA expression
T-cell activation was assessed following an incubation period of 16-18 h by measuring the CXCL10 mRNA expression by direct real time PCR (dRT-PCR) (bKIT™ Immunofinder dqTACT [Ref.HK032X300]), a method already validated with respect to an IGRA assay, quantitatively measuring IFNγ.The assay used allow to amplify the mRNA of the gene CXCL10, which is a proxy of the activation of CD8 + cells and the detail of the procedure is reported in eMethods in the supplemental [18].Every sample (whole blood from every patient) was stimulated in a single biological replicate and tested in a single technical replicate.
The production and long-term persistence of antibody and immune cells titers following mRNA vaccination were assessed by comparing: 1) the median and interquartile range (IQR) of anti-S-RBD IgG antibody titers and immune cells, and 2) the geometric mean titer (GMT) and the 95 % confidence interval (95 % CI), of NAbs against WT virus and BA.2 VOC.Median and IQR and GMT and 95 % CI of antibodies and immune cells titers were evaluated overall and stratified by SARS-CoV-2recovered and SARS-CoV-2-naïve participants at 1-and 6-months after vaccination.The Wilcoxon Rank Sum test and the sample t-test were performed, respectively.
To describe the antibodies and immune cells' kinetics over time, we conducted a longitudinal analysis on a intra-participant-paired sample from a sub-cohort of 39 SARS-CoV-2-recovered and 5 SARS-CoV-2-naïve children tested at 1-and 6-months after vaccination.The decay of the Abs was evaluated as Absvalueat6mo.−Absvalueat1mo. Absvalueat6mo.
*100.The Signed Rank test and the paired t-test were performed, where appropriate, to compare values at 1-and 6-months after vaccination among SARS-CoV-2recovered and SARS-CoV-2-naïve children, respectively.
The association between anti-S-RBD IgG and NAbs titers at 6-months follow-up was assessed with a linear regression model.The strength of associations between variables was evaluated by the Pearson correlation coefficient, using the logarithm (base 2) of the antibody titers given data skew.
Moreover, to better investigate the effect of mRNA vaccination in subjects with pre-existing immunity from a prior SARS-CoV-2 infection, we conducted a longitudinal analysis on an intra-subject-paired sample from a sub-cohort of 18 SARS-CoV-2-recovered children tested at prevaccination, 1-, and 6-months after vaccination to evaluate the trend over time of antibodies and immune cells.The increase from pre-to 6months assessment was evaluated as Absvalueat6mo.−preAbsvalue Absvalueat6mo.
*100.Finally, Tregs and Bregs frequencies were correlated with levels of NAbs to WT and BA.2, and anti-S-RBD IgG antibody titers evaluating the Spearman correlation due to data skew.
All analyses were also performed, stratifying by HC, IC, SOT, and MIS-C.However, since the multiple stratifications and comparison, we performed a descriptive analysis only.
Analyses were performed using the Statistical Analysis System software (version 9.4; SAS Institute, Cary, North Carolina).Statistical significance was set at the level of 0.05.All P values were 2-sided.Graphs were made using GraphPad Prism version 9 (GraphPad Software, La Jolla, CA).
Fig. 1B (eTable2 in the supplemental) shows the antibody kinetics over time.All individuals showed decreased anti-S-RBD IgG and WT and BA.2 NAbs titers, regardless of a previous COVID-19.Further crosssectional analysis of 6-month antibody titers demonstrated that anti-S-RBD IgG remained highly correlated with NAbs, indicating that anti-S-RBD IgG antibodies retain their functional characteristics and neutralization capacity over time (eFigure3 in the supplemental).Fig. 1C (eTable3 in the supplemental) shows the antibodies kinetics from pre-vaccination, 1-, and 6-months post-vaccination in SARS-CoV-2recovered participants.Both anti-S-RBD IgG antibodies and NAbs titers showed a robust increase at a 1-month time point post-vaccination compared to the pre-vaccination titer.In addition, despite a decay in antibody titers over time, both anti-S-RBD IgG antibodies and NAbs titers at 6-months remained higher than the pre-vaccination levels.
In an analysis assessing the humoral response to SARS-CoV-2 vaccination stratifying participants by comorbidity (i.e., HC, IC, SOT, and MIS-C), both the SARS-CoV-2-recovered SOT and the SARS-CoV-2naïve SOT recipient showed a 3.5-fold and 5.5-fold lower anti-S-RBD IgG titers at 1-month after vaccination compared to SARS-CoV-2-recovered and SARS-CoV-2-naïve HC, respectively.Similarly, SOT recipients achieved a lower neutralizing response than HC against both WT and BA.2 strains at 1-month after vaccination, regardless of a previous SARS-CoV-2 infection.Moreover, SOT recipients showed lower antibody responses  compared to HC at 6-months after vaccination, irrespective of prior SARS-CoV-2 infection (eFigure4, eTable4 in the supplemental).Conversely, MIS-C participants recorded similar anti-S-RBD IgG antibody titers to HC, at both 1-and 6-months post-vaccination.WT and BA.2 NAbs titers were also similar between MIS-C and HC at both 1-and 6-months after vaccination (eFigure4, eTable4 in the supplemental).

Cellular immune profile up to 6-months following COVID-19 mRNA vaccination
Fig. 2 (eTable5 in the supplemental) shows the immunological parameters compared between SARS-CoV-2-recovered and SARS-CoV-2naïve cases at 1-and 6-months after vaccination.No difference was observed in the frequencies of activated and senescent T (CD4 and CD8) and B cells at both time points.The frequencies of Tregs were higher in SARS-CoV-2-recovered than in SARS-CoV-2-naïve children both at 1-(p < 0.001) and 6-months after vaccination (p = 0.02).Similarly, the frequencies of Bregs were higher in SARS-CoV-2-recovered than in SARS-CoV-2-naïve children both at 1-(p = 0.02) and 6-months after vaccination (p = 0.004) (Fig. 2).Within the SARS-CoV-2-recovered participants, Tregs and Bregs frequencies were positively correlated with levels of anti-S-RBD IgG antibody, WT NAbs, and BA.2 NAbs titers at both 1and 6-months after vaccination (eFigure5, eFigure6 in the supplemental).No correlation was found between regulatory cells and NAbs in SARS-CoV-2-naïve children.
The longitudinal analysis on a subject-paired sample from a subcohort of 18 SARS-CoV-2-recovered participants tested at prevaccination, 1-, and 6-months after vaccination showed that both Tregs and Bregs were boosted at 1-month postvaccination compared to the pre-vaccination assessment, and decreased at 6-months after vaccination (eTable7 in the supplemental).
Immune profile assessed stratifying participants by comorbidities showed similar percentages of activated and senescent T and B cells among the HC, IC, SOT, and MIS-C groups, regardless of a previous SARS-CoV-2 infection.The SARS-CoV-2-recovered SOT recipient showed lower percentages of Tregs and Bregs compared to HC, IC, and MIS-C at both 1-and 6-months after vaccination (eTable8 in the supplemental).

Discussion
To the best of our knowledge, our study represents one of the first immunological studies evaluating the long-term immune response to the BNT162b2 vaccine in 5-11-years-old children.We evaluated the dynamic changes of the antigen-specific humoral and cellular responses to mRNA COVID-19 vaccination in a prospective cohort of 60 children aged 5-11 years with or without a previous asymptomatic or mild SARS-CoV-2 infection up to 6 months following vaccination.We observed that mRNA vaccination induced robust circulating antibody responses to SARS-CoV-2 with a distinct response profile in both SARS-CoV-2-recovered and SARS-CoV-2-naïve children.We documented that mRNA vaccines elicited higher humoral responses and higher Tregs and Bregs frequencies in SARS-COV-2-recovered compared to naïve-vaccinated children at both 1-and 6-months after vaccination.Our findings are similar to several studies documenting a stronger and more durable immune response to BNT162b2 vaccine in adults [1,20,21].Additionally, these results are in line with recently published studies, showing higher levels of binding antibody and Omicron-targeted NAbs titers after mRNA vaccination in children with hybrid immunity compared to SARS-CoV-2-naïve subjects [22][23][24].These observations offer valuable insights into enhancing pre-existing immunity and lend support to recent research demonstrating the significance of administering the BNT162b2 vaccine to SARS-CoV-2-recovered children to prevent subsequent reinfection [25].
In this study, we strengthened and extended upon previous findings documenting the persistence of detectable specific NAbs and binding antibody titers up to 6-months after vaccination in both SARS-CoV-2recovered and SARS-CoV-2-naïve children, despite an observed decline in their titers over time.Moreover, a cross-sectional analysis of 6-months humoral response also demonstrated that binding antibodies remained highly correlated with neutralizing titers to WT, indicating that S-RBD-specific IgG responses retain their functional characteristics and neutralizing capacity over time.The persistence of higher titers of specific antibodies at 6-months post-vaccination compared to prevaccination, as well as the correlation of anti-S-RBD IgG with neutralizing titers, were also documented in adults, highlighting the importance of boosting a pre-existing immunity to confer a longer durable protection against SARS-CoV-2 reinfection [2].
The repeated exposure upon vaccination boosts SARS-CoV-2-specific T CD4 cells, responsible for recall and expansion of memory B cells, resulting in a more robust humoral response in SARS-COV-2-recovered compared to naïve-vaccinated subjects.Specifically, hybrid immunity, marked by the presence of both spike and non-spike memory T cells, exhibits heightened and broader production of binding and variantneutralizing antibodies, leading to a cross-reactive humoral response to Omicron VOC [21,26].
In this study, we found higher levels of Tregs and Bregs in SARS-CoV-2-recovered than SARS-CoV-2-naïve participants at both 1-and 6months after vaccination.This suggests that the greater expansion of Tregs and Bregs in SARS-CoV-2-recovered children might have resulted in a higher specific production of anti-SARS-CoV-2 antibodies after vaccination in these children compared to SARS-CoV-2-naïve children.Given the lack of disparities in the frequencies of immune activation between SARS-CoV-2-recovered and SARS-CoV-2-naïve children, it is plausible to hypothesize that vaccination does not trigger cytokine storm or aspecific immune activation, resulting in a reduced occurrence of regulatory cells in SARS-CoV-2-naïve children compared to SARS-CoV-2-recovered children following vaccination.Consequently, the greater levels of regulatory cells observed in hybrid immunity, attributed to prior infection rather than vaccination per se, may play a key role in enhancing a greater specific humoral response against the virus, likely through their suppressive activity of the aspecific immune response [12].Changes in circulating Tregs have been investigated after vaccination against pathogens other than SARS-CoV-2 [27].An increased frequency of Tregs was observed after the live attenuated yellow fever vaccine and the Hepatitis B vaccine [28].In contrast, adjuvanted or nonadjuvanted influenza vaccines did not affect the expansion of regulatory cells [28].The effects of different vaccines on Tregs dynamics may depend on the properties of the vaccine.Furthermore, the simultaneous administration of multiple vaccines may trigger complex immunologic milieux where cytokines, B cells and possibly Tregs could contribute to either the enhancement or the dampening of immunologic responses [28].
Interestingly, IC children, who were treated with immunosuppressor/immunomodulant regimens developed a comparable humoral and cellular response to HC, regardless of previous SARS-CoV-2 infection.On the other hand, both SARS-CoV-2-recovered and SARS-CoV-2-naïve SOT recipients showed significantly lower antibody and cell titers compared to HC after 2-dose mRNA vaccination.Similarly, SOT recipients showed lower Tregs and Bregs frequencies than HC.These results might suggest the need for an additional dose of primary series [29], a higher dosage of vaccines [24], or additional preventive strategies such as monoclonal antibodies to prevent severe diseases in these fragile patients.Our findings are in line with several studies demonstrating a lower magnitude of immune response to mRNA vaccines, especially in SOT recipient adults [30][31][32].
Antibody titers and immune cell profiles were boosted with vaccination in MIS-C cases, with a higher anti-S-RBD IgG response than HC, despite a lower neutralization power of their sera compared to HC.Similarly, Perez et al. [33] recently demonstrated that BNT162b2 primary series boosted broadly cross-reactive binding antibodies in children with a previous MIS-C which were maintained for up to 3 months.However, the NAbs levels were low, especially against the Omicron variant.
Our study has several limitations.First, the overall number of subjects was small, which limited our ability to thoroughly explore the impact of an immunocompromising condition on the immune response to mRNA vaccination.Second, the time points used in this study did not allow us to capture the full kinetics of the immune response.An additional sampling timepoint 3 months post-vaccination could have allowed for a better assessment of antibody and cell decays over time.Lastly, the lack of evaluation of memory B cell responses meant we were not able to describe the prolonged germinal centre reactions nor evaluate the durability of cell protection against reinfection.
In conclusion, we demonstrated the magnitude and kinetics of humoral and cellular responses to mRNA vaccination in 5-11 years old children up to 6 months post-vaccination and showed higher antibody peaks for SARS-CoV-recovered children at every follow-up time point.
Previous studies have demonstrated a correlation between SARS-CoV-2-specific antibody levels and protection from reinfection, thus implying the presence of humoral correlates of protection [34].Therefore, the decline in vaccine-induced immunity migh be reflected in an increased susceptibility to reinfection 5-months post-immunization, underscoring the importance of booster vaccination schedules in the pediatric population [21].
This study provides valuable insights into enhancing preexisting immunity with mRNA COVID-19 vaccines in children with a prior SARS-CoV-2 infection.Moreover, these findings can aid in establishing vaccination schedules for non-previously infected children and determining the optimal timing for booster immunization in pediatric patients who have already had COVID-19.

Declaration of competing interest
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.

Fig. 1 .
Fig. 1.Magnitude and kinetics of the neutralizing antibody titers to Wild-type (WT) and Omicron BA.2 strains and anti-S-RBD IgG antibody titers following mRNA vaccination in SARS-CoV-2-recovered cases compared to SARS-CoV-2-naïve cases.Panel A shows titers of anti-S-RBD IgG, NAbs to WT, and NAbs to Omicron BA.2 in SARS-CoV-2-recovered and SARS-CoV-2-naïve cases at 1-(blu dots) and 6-months (purple dots) after vaccination.The Signed Rank test and the paired t-test were performed, appropriately.Panel B shows the individual (grey lines) and overall (black lines) kinetics of anti-S-RBD IgG, NAbs to WT, and NAbs to Omicron BA.2 titers between 1-and 6-months after vaccination in 39 SARS-CoV-2-recovered and 5 SARS-CoV-2-naïve cases who were tested at both time points.Panel C shows the individual (grey lines) and overall (black lines) boosting effect of mRNA vaccination on anti-S-RBD IgG, NAbs to WT, and NAbs to Omicron BA.2 titers in 18 subjects with pre-existing immunity from a prior SARS-CoV-2 infection who were tested at 3 time points.(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 1
Demographic and clinical characteristics of the study population by SARS-CoV-2-recovered and SARS-CoV-2-naïve cases.