Antigen-specific T cell responses in SARS-CoV-2 mRNA-vaccinated children

Abstract

SARS-CoV-2 mRNA vaccines elicit humoral responses in children that are comparable to those in adults. However, early-life T cell responses are distinct from adult ones, and questions remain about the nature and kinetics of mRNA vaccine-induced T cell responses in children. We report that Pfizer BNT162b2 mRNA vaccination elicits a significant antigen-specific CD4⁺ T cell response in the ≥12-year-old cohort. This response is weaker in magnitude in the 5- to 11-year-old cohort and is not improved by a higher vaccine dose (Moderna mRNA1273, 100 μg), suggesting distinct developmental programming that may underscore early-life T cell immunity. Increased effector phenotypes of antigen-specific T cells in younger children correspond with elevated anti-receptor binding domain antibody levels, albeit at the cost of memory generation. These studies highlight aspects of age-specific adaptive immune responses and the need for careful consideration of priming conditions including vaccine dose and adjuvant in the pediatric population.

Keywords: T cell immunity; early life immunity; mRNA vaccine; pediatric COVID-19.

Graphical abstract

Figure 1

Pediatric mRNA vaccination elicits antigen-specific T cell responses PBMCs from Pfizer COVID-19 mRNA (BNT162b2)-vaccinated 5- to 11-year-old (n = 8, 5–11y) and ≥12-year-old (n = 8, 12+y) children were analyzed for the presence of antigen-specific CD4⁺ and CD8⁺ T cells at indicated times over the vaccination series. All paired longitudinal samples were analyzed within a single batch. (A) Representative flow cytometry dot plots show identification of CD4⁺ AIM⁺ T cells by co-expression of activation-induced markers (AIMs) 4-1BB and CD200 on non-naive CD4⁺ T cells. Values in black show frequency of 4-1BB⁺CD200⁺ cells in non-naive CD4⁺ T cells and in red show the number of 4-1BB⁺CD200⁺ events and the number of events in the parent non-naive CD4⁺ T cell gate. (B) Frequency of AIM⁺ non-naive T cells was calculated by subtracting the frequency of 4-1BB⁺CD200⁺ cells in stimulated samples from paired unstimulated samples. Summary plots of background-subtracted non-naive CD4⁺ AIM⁺ T cell frequencies from individual 5–11y and 12+y participants over the study period. Solid lines connect individual donors sampled at indicated time points. Statistics show two-tailed p value summary of paired t tests comparing post-vaccination time points V1, V2, V6, and VB to baseline V0. (C) Summary plot of the (left) frequency and (right) numbers per million CD45⁺ PBMCs of non-naive CD4⁺ AIM⁺ T cells from grouped 5–11y and 12+y participants over the study period. Statistics show two-tailed p value summary of unpaired t tests comparing 5–11y and 12+y groups at each time point. Error bars are SEM. (D) Summary plots of background subtracted non-naive CD8⁺ AIM⁺ T cell frequencies, identified by expression of at least three of four activation-induced markers (4-1BB, CD200, LAMP-1, and IFNγ), from individual 5–11y and 12+y participants over the study period. Solid lines connect individual donors sampled at indicated time points. Statistics show two-tailed p value summary of paired t tests comparing post vaccination time points V1, V2, V6, and VB to baseline V0. (E) Summary plot of the (left) frequency and (right) numbers per million CD45⁺ PBMCs of non-naive CD8⁺ AIM⁺ T cells from grouped 5–11y and 12+y participants over the study period. Statistics show two-tailed p value summary of unpaired t tests comparing 5–11y and 12+y groups at each time point. Error bars are SEM. ns: non-significant, ∗p < 0.05, ∗∗p < 0.01.

Figure 2

Memory and effector profiles of CD4⁺ AIM⁺ T cells Antigen-specific CD4⁺ memory and effector subsets generated at the end of the vaccination series (VB time point) were identified based on the expression of CD27, CD45RA, and CXCR5. Only samples that contained 10 or more AIM⁺ events (identified in Figure 1) were analyzed. (A) (Left) Representative flow cytometry dot plots show the gating and frequency of central memory (CM) CD27⁺CD45RA^neg and effector memory (EM) CD27^negCD45RA⁺ and CD27^negCD45RA^neg cells in CD4⁺ AIM⁺ T cell pools. (Right) Summary plots of the frequency of each subset in 5–11y (n = 4) and 12+y (n = 8) groups. Statistics show two-tailed p value summary of unpaired t tests. Error bars are SEM. ns: non-significant, ∗p < 0.05, ∗∗∗p < 0.001. (B) (Left) Representative flow cytometry dot plots show the gating of CXCR5⁺ cTfh cells. Events in gray are non-naive CD4⁺ T cells and in red are CD4⁺ AIM⁺ T cells. Values in black show frequency of CXCR5⁺ cells within non-naive CD4⁺ AIM⁺ T cell compartment and in red show number of CXCR5⁺ events and the number of events in the parent non-naive CD4⁺ AIM⁺ T cell gate. (Right) Summary plot of the frequency of CXCR5⁺ subset within CD4⁺ AIM⁺ T cells from 5–11y (n = 4) and 12+y (n = 8) groups. Statistics show two-tailed p value summary of unpaired t tests. Error bars are SEM. ∗∗p < 0.01. (C) (Left) Summary plots of anti-spike receptor binding domain (RBD) antibody responses of individual 5–11y and 12+y participants over the study period. Solid lines connect individual donors sampled at indicated time points. Statistics evaluating participant kinetics were calculated using Wilcoxon matched pairs signed rank test. (Right) Summary plot shows levels of anti-spike RBD antibody from grouped 5–11y and 12+ +y participants over the study period. Statistics show two-tailed p value summary of unpaired t tests comparing 5–11y and 12+y groups at each time point. Error bars are SEM. ns: non-significant ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Figure 3

Pediatric T cell response to different doses of the COVID-19 mRNA vaccine The antigen-specific T cell response was compared in 5- to 11-year-old children that received the 10-μg dose of the Pfizer COVID-19 mRNA vaccine to those that received the 100-μg dose of the Moderna COVID-19 mRNA vaccine (mRNA-1273). (A) Summary plot of background subtracted non-naive CD4⁺ AIM⁺ T cell (top) frequency and (bottom) numbers per million CD45⁺ PBMCs from individual 5–11y participants that received the Moderna 100-μg mRNA vaccine over the study period. Solid lines connect individual donors sampled at indicated time points. Statistics show two-tailed p value summary of paired t tests comparing post-vaccination time points V1, V2, V6, and VB to baseline V0. (B) Summary plot of background-subtracted non-naive CD4⁺ AIM⁺ T cell (top row) frequency and (bottom row) numbers per million CD45⁺ PBMCs from 5–11y participants that received either the Pfizer 10-μg or Moderna 100-μg mRNA vaccine compared to 12+y participants that received the Pfizer 30-μg vaccine at each time point. Statistics show two-tailed p value summary of unpaired t tests comparing 5–11y Pfizer and Moderna groups to the 12+y group at each time point. Error bars are SEM. ∗p < 0.05, ∗∗p < 0.01. (C) Summary plot of background-subtracted non-naive CD8⁺ AIM⁺ T cell (top) frequency and (bottom) numbers per million CD45⁺ PBMCs from individual 5–11y participants that received the Moderna 100-μg mRNA vaccine over the study period. Solid lines connect individual donors sampled at indicated time points. Statistics show two-tailed p value summary of paired t tests comparing post-vaccination time points V1, V2, V6, and VB to baseline V0. (D) Summary plot of background-subtracted non-naive CD8⁺ AIM⁺ T cell (top row) frequency and (bottom row) numbers per million CD45⁺ PBMCs from grouped 5–11y participants that received either the Pfizer 10-μg or Moderna 100-μg mRNA vaccine compared to 12+y participants that received the Pfizer 30-μg vaccine at each time point. Statistics show two-tailed p value summary of unpaired t tests comparing 5–11y Pfizer and Moderna groups to the 12+y group at each time point. Error bars are SEM. ∗p < 0.05.