Hybrid B- and T-Cell Immunity Associates With Protection Against Breakthrough Infection After Severe Acute Respiratory Syndrome Coronavirus 2 Vaccination in Avon Longitudinal Study of Parents and Children (ALSPAC) Participants

Abstract

Background: Immunological memory to vaccination and viral infection involves the coordinated action of B and T cells; thus, integrated analysis of these 2 components is critical for understanding their respective contributions to protection against breakthrough infections (BIs) after vaccination.

Methods: We investigated cellular and humoral immune responses to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and/or vaccination in 300 adult participants from the Avon Longitudinal Study of Parents and Children (ALSPAC). Participants were grouped by those with (cases) and without (controls) a history of SARS-CoV-2 infection. To provide a quantitative correlate for protection against BI in the 8-month period after the study, Youden index thresholds were calculated for all immune measures analyzed.

Results: The magnitude of antibody and T-cell responses following the second vaccine dose was associated with protection against BI in participants with a history of SARS-CoV-2 infection (cases), but not in infection-naive controls. Over 8 months of follow-up, 2 threshold combinations provided the best performance for protection against BI in cases: (i) anti-spike immunoglobulin G (IgG) (≥666.4 binding antibody units [BAU]/mL) combined with anti-nucleocapsid pan-immunoglobulin (pan-Ig) (≥0.1332 BAU/mL) and (ii) spike 1-specific T cells (≥195.6 spot-forming units/106 peripheral blood mononuclear cells) combined with anti-N pan-Ig (≥0.1332 BAU/mL). Both combinations offered 100% specificity for detecting cases without BI, with sensitivities of 83.3% and 72.2%, respectively.

Conclusions: Collectively, these results suggest that hybrid B- and T-cell immunity offers superior protection from BI after coronavirus disease 2019 (COVID-19) vaccination, and this finding has implications for designing next-generation COVID-19 vaccines that are capable of eliciting immunity to a broader repertoire of SARS-CoV-2 proteins.

Keywords: ALSPAC; SARS-CoV-2; breakthrough infection; hybrid immunity; vaccination.

Figure 1.

Study design. A, Participants from the Avon Longitudinal Study of Parents and Children (ALSPAC) were screened for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)–specific anti-spike (S) immunoglobulin G (IgG) in October 2020, prior to the coronavirus disease 2019 (COVID-19) vaccine rollout. Both the G0 (age 48–70 y) and G1 (age 29–30 y) generations were eligible for screening. Further sampling was conducted in June 2021 and June 2022 for both anti-S and anti-nucleocapsid (N) IgG. Numbers indicate the total valid tests, with the percentages indicating the positivity rates. B, Of the 4819 participants with valid lateral flow test (LFT) results in October 2020, 377 were recruited to this study. Participants attended 1 or more clinics from December 2020 to June 2021, providing biological samples and completing health questionnaires. During this period, participants became eligible to receive COVID-19 vaccines via the UK national vaccination program (text above clinic timepoints indicates percentage vaccinated at each clinic). Following the sampling period, participants continued to complete online questionnaires detailing their LFT and polymerase chain reaction–confirmed SARS-CoV-2 infections. Created in BioRender (A. Halliday, 2025; https://BioRender.com/i7wpnfr).

Figure 2.

Antibody responses to coronavirus disease 2019 vaccination in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)–naive and previously infected individuals. A, Participants were classified as cases based on a previous polymerase chain reaction–confirmed SARS-CoV-2 infection and/or positivity on serum anti-spike and/or anti-nucleocapsid pan-immunoglobulin enzyme-linked immunosorbent assays (ELISAs). B, Serum pseudoneutralizing antibody titers. C and D, Anti-spike immunoglobulin G and immunoglobulin A in serum and saliva, measured by ELISA. Correlation coefficients were calculated using Spearman rank test (r_s). Central bars indicate median responses. Unpaired comparisons were performed using Kruskal-Wallis test with Dunn correction for multiple comparisons. Within each of the case and control groups, responses were compared between 0 and 1, and 1 and 2 vaccine doses. Responses between groups were compared after each dose. Pairwise correlations were assessed with Spearman rank-order correlation (r_s). Correlation coefficients were interpreted as weak (r_s = 0.20–0.39), moderate (r_s = 0.40–0.59), strong (r_s = 0.60–0.79), or very strong (r_s = 0.80–1.00). Statistics are only displayed for comparisons where *P ≤ .05, **P ≤ .01, ***P ≤ .001, and ****P ≤ .0001. Abbreviations: BAU, binding antibody units; IgA, immunoglobulin A; IgG, immunoglobulin G; N, nucleocapsid; ND₅₀, 50% neutralising dilution; OD, optical density; pan-Ig, pan-immunoglobulin.

Figure 3.

T-cell responses to coronavirus disease 2019 vaccination in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)–naive and previously infected individuals. Magnitude of SARS-CoV-2–specific T-cell responses against spike 1, spike 2, nucleocapsid, membrane, and NSP3B peptide pools measured by enzyme-linked immunosorbent spot assay. Participants were classified as cases based on a previous polymerase chain reaction–confirmed SARS-CoV-2 infection and/or positivity on serum anti-spike and/or anti-nucleocapsid pan-immunoglobulin enzyme-linked immunosorbent assays (see Figure 2). Central bars indicate median responses. Unpaired comparisons were performed using Kruskal-Wallis test with Dunn correction for multiple comparisons. Within each of the case and control groups, responses were compared between 0 and 1, and 1 and 2 vaccine doses. Responses between groups were compared after each dose. Statistics are only displayed for comparisons where *P ≤ .05, **P ≤ .01, and ****P ≤ .0001. To facilitate the presentation of data that included zero counts on log scales, zero counts were plotted as 1 for visualization purposes only; all statistical analyses were performed on the raw data values. Abbreviations: ELISpot, enzyme-linked immunosorbent spot assay; N, nucleocapsid; PBMC, peripheral blood mononuclear cell; SFU, spot-forming unit.

Figure 4.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)–specific CD4⁺ and CD8⁺ T-cell responses in previously infected individuals. Graphs show the percentage of CD4⁺ (A) and CD8⁺ (B) T cells producing the indicated cytokine or combination of cytokines after a brief stimulation with spike 1, spike 2, membrane, nucleocapsid, or NSP3B peptide pools, assessed by intracellular cytokine staining and flow cytometry. Pie charts indicate the proportion of T cells, within total cytokine⁺ T cells, producing each cytokine, and that display 1 or more functions. Statistics from monofunctional SARS-CoV-2–specific CD4⁺ and CD8⁺ T cells in previously infected individuals were calculated using a Kruskal-Wallis test with false discovery rate method of Benjamini and Hochberg correction for multiple comparisons. P values are reported in Supplementary Table 2. To facilitate the presentation of data that included zero counts on log scales, zero counts were plotted as 1 for visualization purposes only; all statistical analyses were performed on the raw data values. Abbreviations: IFN-γ, interferon gamma; IL-2, interleukin 2; M, membrane; MIP-1β, macrophage inflammatory protein 1β; N, nucleocapsid; S, spike; TNF-α, tumor necrosis factor alpha.

Figure 5.

Association between immune responses to prior severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and coronavirus disease 2019 vaccination, and susceptibility to SARS-CoV-2 breakthrough infection (BI). Correlation of post–second vaccination antibody and T-cell responses in participants who self-reported a SARS-CoV-2 infection in the subsequent 8 months (BI) and those who did not (no BI). A, SARS-CoV-2-naive individuals (with BI, n = 14; no BI, n = 19). B, Previously SARS-CoV-2–infected individuals (BI, n = 8; no BI, n = 18). See Supplementary Figure 10 for derivation of Youden index thresholds. Abbreviations: BAU, binding antibody units; ELISpot, enzyme-linked immunosorbent spot assay; IgA, immunoglobulin A; N, nucleocapsid; ND₅₀, 50% neutralising dilution; pan-Ig, pan-immunoglobulin; PBMC, peripheral blood mononuclear cell; S, spike; SE, sensitivity; SFU, spot-forming unit; SP, specificity.

Figure 6.

Association between prevaccination severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)–specific T-cell intracellular cytokine responses and susceptibility to breakthrough infection. Correlation between SARS-CoV-2–specific T-cell intracellular cytokine responses in participants with a history of SARS-CoV-2 infection prior to coronavirus disease 2019 vaccination (cases) and postvaccination breakthrough infections. A and B, Magnitude of baseline (prevaccination) single cytokine/CD107a-producing (monofunctional) SARS-CoV-2–specific CD4⁺ (A; n = 47) and CD8⁺ (B; n = 51) T cells specific for the indicated SARS-CoV-2 proteins among cases. Central bars represent median responses. Statistics calculated by t test (Mann-Whitney). C and D, Proportions of single cytokine/CD107a-producing CD4⁺ (C; n = 40) and CD8⁺ (D; n = 43) T cells within the respective membrane-specific T-cell population. Statistics are only displayed for comparisons where *P ≤ .05. Abbreviations: BI, breakthrough infection; IFN-γ, interferon gamma; IL-2, interleukin 2; M, membrane; MIP-1β, macrophage inflammatory protein 1β; N, nucleocapsid; TNF-α, tumor necrosis factor alpha.