A two-arm analysis of the immune response to heterologous boosting of inactivated SARS-CoV-2 vaccines

Abstract

Several vaccine programs were introduced during the COVID-19 pandemic, which included inactivated virus, DNA viral vectors and mRNA vaccines. Booster programs are recommended, especially for those in high-risk groups. However, many of these booster programs involve heterologous vaccines. This study enrolled volunteers who first received two full-dose CoronaVac vaccinations before receiving heterologous boosters with DNA- and/or mRNA-vaccines for an additional 2 doses (n = 40) or an additional 3 doses (n = 16). Our results showed no difference in side effects, neutralizing antibodies, or T-cell responses for any of the heterologous vaccination programs. However, the neutralizing capacity and IFN-γ responses against the Omicron variant in volunteers who received 4 or 5 doses were improved. Polarization of peripheral memory T cells after stimulation in all booster groups with Omicron peptide showed an increased trend of naïve and central memory phenotypes of both CD4+ and CD8+ T cells, suggesting that exposure to Omicron antigens will drive T cells into a lymphoid resident T cell phenotype. Our data support a continuous vaccination program to maximize the effectiveness of immunity, especially in people at high risk. Furthermore, the number of boosting doses is important for maintaining immunity.

Figure 1

Comparison of plasma inhibitory efficacy against B1.1.529 (Omicron) and Wuhan Hu-1 following different vaccination strategies. Serum samples of participants with 4 doses (n = 40) or 5 doses (n = 16) were collected to evaluate % inhibition against B1.1.529 (Omicron) and Wuhan Hu-1 SARS-CoV-2. (A) Shows a flow of participants through the study. In (B–D) data represent individual participant measurements (open purple circle for 4 doses, open green circle for 5 doses) with the horizontal lines showing median and 95% confidence interval (95% CI) and dash line represents a positive cut-off. (B) Shows differences in % inhibition of different viral strains and different vaccinations regimens with responses against Omicron (open purple circle, open green circle) and against Wuhan Hu-1 (open purple triangle, open green triangle). (C,D) Show % inhibition of B1.1.529 (Omicron) with different vaccination strategies after 4 doses or 5 doses, respectively. Data were analysed using Kruskal–Wallis test with Dunn’s multiple comparisons test. Statistical significance was determined at a p-value < 0.05.

Figure 2

Specific T cell responses measured as IFN-γ release upon in vitro stimulation with SARS-CoV2 receptor-binding domain (RBD) or S1S2 peptides after four or five doses of vaccination. Peripheral blood mononuclear cells (PBMCs) from participants after 4 doses or 5 doses were stimulated with RBD or S1S2 peptides. IFN-γ release was quantified using QuantiFERON. Data are shown as individual dot plots after medium (control) subtracted IFN-γ levels (open purple circle for 4 doses, open green circle for 5 doses) with horizontal lines showing median and interquartile ranges. IFN-γ levels release by samples after 4 doses and 5 doses in response to RBD (A) or S1S2 peptides (B) were compared using Mann Whitney test. Levels of IFN-γ release by samples after different vaccine combinations of 4 doses (C) or 5 doses (D) were analysed using Kruskal–Wallis test with Dunn’s multiple comparisons test. Statistical significance was determined at a p-value < 0.05.

Figure 3

CD4+ T helper lymphocyte subset changes upon stimulation of PBMC from participants with B.1.1.529 (Omicron) peptide after 4 or 5 doses of heterologous vaccination. PBMCs from participants were cultured with medium control (Resting) or Omicron peptide prior to surface marker staining and flow cytometry analysis for CD4+ T helper lymphocytes. Average percentages of T helper lymphocytes with naïve, central memory (CM), effector memory (EM), and effector memory RA (EMRA) phenotypes from participants receiving 4 doses (A) or 5 doses (C) are shown. Absolute cell numbers of each phenotype were calculated and individually plotted, with lines connecting the same participant before and after stimulation (purple open circle; cell count in Resting samples from the 4th dose group, purple filled circle; cell count after stimulation of samples from the 4th dose group) (B). The same analysis was performed on samples from the 5th dose group (green open circle; cell count in Resting, green filled circle; cell count after stimulation of samples from the 5th dose group) (D). Statistically-significant changes in the paired samples were calculated with a paired t test, with significance determined at a p-value < 0.05.

Figure 4

CD8+ cytotoxic T lymphocyte subset changes upon stimulation of PBMC from participants with B.1.1.529 (Omicron) peptide after 4 or 5 doses of heterologous vaccination. PBMCs from participants were cultured with medium control (Resting) or Omicron peptide prior to surface marker staining and flow cytometry analysis for CD8+ T cytotoxic lymphocytes. Average percentages of cytotoxic T lymphocytes with naïve, central memory (CM), effector memory (EM), and effector memory RA (EMRA) phenotypes from participants receiving 4 doses (A) or 5 doses (C) are shown. Absolute cell numbers of each phenotype were calculated and individually plotted, with lines connecting the same participant before and after stimulation (purple open circle; cell count in Resting samples from the 4th dose group, purple filled circle; cell count after stimulation of samples from the 4th dose group) (B). The same analysis was performed on samples from the 5th dose group (green open circle; cell count in Resting, green open circle; cell count after stimulation of samples from the 5th dose group) (D). Statistically-significant changes in the paired samples were calculated with a paired t test, with significance determined at a p-value < 0.05.