Immunogenicity and reactogenicity of heterologous ChAdOx1 nCoV-19/mRNA vaccination

Abstract

Heterologous priming with the ChAdOx1 nCoV-19 vector vaccine followed by boosting with a messenger RNA vaccine (BNT162b2 or mRNA-1273) is currently recommended in Germany, although data on immunogenicity and reactogenicity are not available. In this observational study we show that, in healthy adult individuals (n = 96), the heterologous vaccine regimen induced spike-specific IgG, neutralizing antibodies and spike-specific CD4 T cells, the levels of which which were significantly higher than after homologous vector vaccine boost (n = 55) and higher or comparable in magnitude to homologous mRNA vaccine regimens (n = 62). Moreover, spike-specific CD8 T cell levels after heterologous vaccination were significantly higher than after both homologous regimens. Spike-specific T cells were predominantly polyfunctional with largely overlapping cytokine-producing phenotypes in all three regimens. Recipients of both the homologous vector regimen and the heterologous vector/mRNA combination reported greater reactogenicity following the priming vector vaccination, whereas heterologous boosting was well tolerated and comparable to homologous mRNA boosting. Taken together, heterologous vector/mRNA boosting induces strong humoral and cellular immune responses with acceptable reactogenicity profiles.

Fig. 1. Immune responses against the SARS-CoV-2 spike protein after vaccination with homologous and heterologous prime–booster regimens.

Immune responses were compared between individuals who received either homologous ChAdOx1 nCoV-19 vector/vector vaccination (n = 55), heterologous ChAdOx1 nCoV-19 vector/mRNA vaccination (n = 96) or homologous mRNA/mRNA vaccination (n = 62). a,b, Spike-specific IgG levels (a) and neutralizing antibodies (ab) (b) were quantified by ELISA and a surrogate neutralization assay. c,d, Percentages of SARS-CoV-2 spike-specific (c) and SEB-reactive (d) CD4 and CD8 T cells were determined after antigen-specific stimulation of whole-blood samples, followed by intracellular cytokine analysis using flow cytometry. Reactive cells were identified by coexpression of CD69 and IFN-γ among CD4 or CD8 T cells and subtraction of background reactivity of respective negative controls. e, Correlations between spike-specific T cell levels, antibody responses and numbers of plasmablasts. f, Cytokine expression profiles of spike-specific CD4 and CD8 T cells in all individuals showing single or combined expression of IFN-γ, IL-2 and TNF-α (gating strategy shown in Extended Data Fig. 5). a–d, Bars represent medians with IQR. Individuals who received the mRNA-1273 vaccine are indicated in gray (1x vector/mRNA, 9x mRNA/mRNA). Differences between groups were calculated using a two-sided Kruskal–Wallis test with Dunn´s multiple comparisons post test. e, Correlation coefficients (r) were analyzed according to two-tailed Spearman (Supplementary Table 1). f, Bars represent means and standard deviation; ordinary one-way ANOVA tests were performed. a,b, Dotted lines indicate detection limits (DL) for antibodies, indicating negative, intermediate and positive levels or levels of inhibition, respectively, per the manufacturer’s instructions. c,d, Dotted lines indicate detection limits for SARS-CoV-2-specific CD4 T cells. f, Analysis was restricted to samples with ≥30 cytokine-positive T cells (n = 31 (CD4) and n = 15 (CD8) for vector/vector, n = 89 and n = 73 for vector/mRNA and n = 58 and n = 24 for mRNA/mRNA).

Fig. 2. Reactogenicity after primary and secondary vaccination with homologous and heterologous prime–booster regimens.

a–c, Reactogenicity within the first week after priming and after the booster dose was self-reported based on a standardized questionnaire, and was analyzed after the first vaccination with either vector (V, n = 150) or mRNA vaccine (mRNA, n = 48), and after the second vaccination with homologous (V/V, n = 54; mRNA/mRNA, n = 48) and heterologous (V/mRNA, n = 95) vaccine regimens with respect to local/systemic reactions in general (a), stratified for local (b) and various systemic adverse events (c). d, Individual perception of which of the two vaccinations had the greater affect. Comparisons between groups were performed using the X² test. GI, gastrointestinal.

Extended Data Fig. 1. Study design.

Time between first and second vaccination and the time between second vaccination and analysis is shown for the three vaccination regimens (homologous ChAdOx1 nCoV-19 vector vaccination (V/V, n = 55), heterologous ChAdOx1 nCoV-19 vector/mRNA vaccination (V/mRNA, n = 97) or homologous mRNA vaccination (mRNA/mRNA, n = 64). ^#3 individuals were excluded from further analysis due to IgG positivity in a SARS-CoV-2 nucleocapsid ELISA.

Extended Data Fig. 2. Demographic and clinical characteristics of the study population.

Demographic characteristics and vaccine-related data are shown for the three vaccination regimens (homologous ChAdOx1 nCoV-19 vector vaccination (V/V, n = 55), heterologous ChAdOx1 nCoV-19 vector/mRNA vaccination (V/mRNA, n = 96) or homologous mRNA vaccination (mRNA/mRNA, n = 62)). In addition, information on differential blood counts and on lymphocyte subpopulations is provided.

Extended Data Fig. 3. Representative analysis of spike- and SEB-reactive CD4 and CD8 T cells.

Lymphocytes were identified among total events by backgating of CD4 and/or CD8 positive cells combined with signals for size (FSC) and granularity (SSC). Hight and area signals of FSC were used to exclude doublets. CD4 T cells were identified among single cells by CD4 positive and CD8 negative signals. Likewise, CD8 T cells were defined as T cells being CD8 positive and CD4 negative. In (b) and (c) representative contour plots of CD4 and CD8 T cells of a 37-year-old female are shown after antigen-specific stimulation with SARS-CoV-2 spike peptides or respective control stimuli for negative (DMSO) or positive control (SEB) stimulation. Numbers indicate percentages of CD4 or CD8 T cells co-expressing the activation marker CD69 and the cytokines IFN-γ, IL-2 and/or TNF-α. FSC, forward scatter; IFN, interferon; IL, interleukin; SEB, Staphylococcus aureus enterotoxin B; SSC, side scatter; TNF, tumor necrosis factor.

Extended Data Fig. 4. SARS-CoV-2 spike-specific T cells producing TNF-α and IL-2 after vaccination with homologous and heterologous prime-booster regimens.

Percentages of SARS-CoV-2 spike-specific CD4 and CD8 T cells were determined after antigen-specific stimulation of whole blood samples followed by intracellular cytokine analysis using flow-cytometry. Reactive cells were identified by co-expression of the activation marker CD69 and the cytokine tumor necrosis factor (TNF) α (left panel), interleukin 2 (IL-2, middle panel), or any of the cytokines analyzed (IFN-γ/TNF-α/IL-2, right panel), respectively, among CD4 or CD8 T cells and subtraction of background reactivity of respective negative control stimulations. T cell responses were compared between individuals who either received SARS-CoV-2 vector/vector (V/V, n = 55), vector/mRNA (V/mRNA, n = 96) or mRNA/mRNA vaccines (mRNA/mRNA, n = 62) using two-sided Kruskal-Wallis test with Dunn’s multiple comparisons post test. Bars represent medians with interquartile ranges, individuals who received the mRNA-1273 vaccine are indicated by grey symbols.

Extended Data Fig. 5. Gating strategy and analysis of spike-specific cytokine-expression profiles.

(a) To characterize spike- and SEB-reactive CD4 T cells regarding their single or combined expression of the cytokines IFN-γ, IL-2, and TNF-α, CD4 T cells positive for combined expression of CD69 and IFN-γ were divided into four subpopulations according to additional expression of IL-2 and TNF-α. Using NOT Boolean Gating, all CD4 T cells which were not CD69 + IFN-γ + were analyzed for CD69 + IL-2+ and CD69 + TNF-α + CD4 T cells. Using OR Boolean Gating, CD69 + IL-2+ and/or CD69 + TNF-α + CD4 T cells were divided into IL-2 single, TNF-α single or IL-2+TNF-α + cells. After subtraction of background reactivity from negative control stimulations, the sum of these 7 subpopulations was set to 100%. A similar strategy was applied for CD8 T cells. (b) Cytokine-expression profiles of spike-specific CD4 and CD8 T cells in all individuals showing single or combined expression of the cytokines IFN-γ, interleukin (IL) 2 and tumor necrosis factor (TNF) α. To allow for robust statistics, analysis was restricted to samples with at least 30 cytokine-positive T cells (n = 178 for CD4 and n = 112 for CD8 T cells). Numbers refer to the percentage of cells expressing the respective cytokine. IFN, interferon; IL, interleukin; TNF, tumor necrosis factor.

Extended Data Fig. 6. Cytokine-expression profiles of SEB-reactive CD4 and CD8 T cells after vaccination with homologous and heterologous prime-booster regimens.

Cytokine expression of CD4 (a) and CD8 T cells (b) after stimulation with Staphylococcus aureus enterotoxin B (SEB), was compared between individuals who either received SARS-CoV-2 vector/vector (V/V), vector/mRNA (V/mRNA), or mRNA/mRNA vaccine combinations. Cytokine-expressing T-cells were divided into 7 subpopulations according to their expression of IFN-γ, IL-2, and TNF-α (single, double or triple cytokine-expressing cells, for gating strategy, see Extended Data Fig. 5). Only samples of the study participants included in Fig. 1f are displayed (at least 30 cytokine-expressing CD4 or CD8 T-cells, respectively, after spike-specific stimulation, n = 31 and n = 15 for V/V, n = 89 and n = 73 for V/mRNA and n = 58 and n = 24 for mRNA/mRNA vaccine regimes). Differences among subpopulations between the groups were determined using ordinary one-way ANOVA test. Bars represent means and standard deviations of subpopulations among all SEB-reactive CD4 and CD8 T cells, respectively. IFN, interferon, IL, interleukin, TNF, tumor necrosis factor.

Extended Data Fig. 7. Immune responses against the SARS-CoV-2 spike protein after vaccination with homologous and heterologous prime-booster regimens in age-matched subgroups.

Immune responses were compared between age-matched subgroups of 50 individuals each who either received homologous ChAdOx1 nCoV-19 vector vaccination (V/V, mean age 47.0 ± 11.2 years, 33 females), heterologous ChAdOx1 nCoV-19 vector/mRNA vaccination (V/mRNA, 47.1 ± 9.2 years, 38 females) or homologous mRNA vaccination (mRNA/mRNA, 46.4 ± 11.5 years, 35 females, p = 0.940). Spike-specific IgG-levels (a) and neutralizing antibodies (b) were quantified by ELISA and a surrogate neutralization assay and compared between groups. Percentages of SARS-CoV-2 spike-specific (c) and SEB-reactive (d) CD4 and CD8 T cells were determined after antigen-specific stimulation of whole blood samples followed by intracellular cytokine analysis using flow-cytometry. Reactive cells were identified by co-expression of CD69 and the cytokine interferon (IFN) γ among CD4 or CD8 T cells and subtraction of background reactivity of respective negative control stimulations. Bars represent medians with interquartile ranges. Individuals who received the mRNA-1273 vaccine are indicated by grey symbols (9 in the homologous mRNA/mRNA group). Differences between the groups were calculated using two-sided Kruskal-Wallis test with Dunn´s multiple comparisons post test. Dotted lines indicate detection limits for antibodies in (a) and (b), indicating negative, intermediate and positive levels or levels of inhibition, respectively as per manufacturer’s instructions, and detection limits for SARS-CoV-2-specific CD4 T cells in (c) and (d). SEB, Staphylococcus aureus enterotoxin B.