mRNA Vaccines Enhance Neutralizing Immunity against SARS-CoV-2 Variants in Convalescent and ChAdOx1-Primed Subjects

Abstract

To identify the most efficient methods of immunological protection against SARS-CoV-2, including the currently most widespread variants of concern (VOCs)-B.1.1.7, B.1.351 and P.1-a simultaneous side-by-side-comparison of available vaccination regimes is required. In this observational cohort study, we compared immunological responses in 144 individuals vaccinated with the mRNA vaccines BNT162b2 or mRNA-1273 and the vector vaccine ChAdOx1-nCoV-19, either alone, in combination, or in the context of COVID-19-convalescence. Unvaccinated COVID-19-convalescent subjects served as a reference. We found that cellular and serological immune responses, including neutralizing capacity against VOCs, were significantly stronger with mRNA vaccines as compared with COVID-19-convalescent individuals or vaccinated individuals receiving the vector vaccine ChAdOx1-nCoV-19. Booster immunizations with mRNA vaccines triggered strong and broadly neutralizing antibody and IFN-γ responses in 100% of vaccinated individuals investigated. This effect was particularly strong in COVID-19-convalescent and ChAdOx1-nCoV-19-primed individuals, who were characterized by comparably moderate cellular and neutralizing antibody responses before mRNA vaccine booster. Heterologous vaccination regimes and convalescent booster regimes using mRNA vaccines may allow enhanced protection against SARS-CoV-2, including current VOCs. Furthermore, such regimes may facilitate rapid (re-)qualification of convalescent plasma donors with high titers of broadly neutralizing antibodies.

Keywords: RNA vaccines may facilitate rapid (re-) qualification of convalescent plasma donors with high titers of broadly neutralizing antibodies; heterologous vaccination regimes using mRNA vaccines may allow enhanced protection against SARS-CoV-2, including current VOCs.

Figure 1

Average antibody and neutralization titers after different anti-SARS-CoV-2 vaccination regimes. Serum samples from 21 BNT162b2- (cohort A), 10 mRNA-1273- (cohort D), and 29 ChAdOx1-nCoV-19 (cohort E)-vaccinated individuals were collected at different timepoints after their first vaccination, as indicated. Second vaccination was given 3 weeks after first vaccination in BNT162b2-, and 4 weeks after first vaccination in mRNA-1273-vaccinated individuals. Serum samples from 162 COVID-19-convalescent individuals, collected at different timepoints after positive pharyngeal swab, served as the control cohort (cohort I). Plotted are average results for (A) neutralization of ACE2–RBD interaction, (B) anti-spike IgG titers, and (C) anti-spike IgA titers. (A) Neutralization capacities > 30% were considered positive, neutralization capacities > 70% were considered strong. Error bars indicate SEM; significance levels shown are between vaccinated individuals (colored circles) and convalescent individuals (grey circles) only. Statistical analysis at each timepoint was performed using Dunn’s multiple comparisons test. Significance levels between ChAdOx1-nCoV-19-vaccinated individuals and BNT162b2- or mRNA-1273-vaccinated individuals ranged at the same level as those between convalescent individuals and BNT162b2- or mRNA-1273-vaccinated individuals. Significance levels were * p < 0.05, ** p < 0.005, and *** p < 0.0005. Abbreviations: COVID-19: coronavirus disease 2019; ACE2: angiotensin-converting enzyme 2; RBD: receptor-binding domain.

Figure 2

Impact of previous SARS-CoV-2 infection on the neutralizing capacity after vaccination with BNT162b2. Serum samples from 15 BNT162b2-vaccinated individuals without history of COVID-19 (cohort B) and 25 COVID-19-reconvalescent BNT162b2-vaccinated individuals (cohort C) were collected at different timepoints after their first vaccination, as indicated. (A) Line graphs show individual courses for the neutralization capacity of ACE2–RBD interaction. (B) Boxplots show median neutralization capacities at the time of first vaccination (0 weeks), second vaccination (3 weeks), and 3 weeks after second vaccination (6 weeks). (C) Boxplots show median anti-NCP IgG titers for COVID-19-naïve versus COVID-19-convalescent donors. (D) Boxplots show neutralization capacities against RBD variants as indicated in BNT162b2-vaccinated COVID-19-naïve versus COVID-19-convalescent individuals. Box central horizontal lines in (B,C) indicate medians, box borders represent IQR, and whiskers indicate minima and maxima. For statistics, Šídák’s test for multiple comparisons (B,D), and the Mann–Whitney U test (C) were used. Significance levels were ** p < 0.005, *** p < 0.0005, and **** p < 0.00005. Abbreviations: COVID-19: coronavirus disease 2019; ACE2: angiotensin-converting enzyme 2; RBD: receptor-binding domain; IQR: interquartile range; n.s.: not significant.

Figure 2

Impact of previous SARS-CoV-2 infection on the neutralizing capacity after vaccination with BNT162b2. Serum samples from 15 BNT162b2-vaccinated individuals without history of COVID-19 (cohort B) and 25 COVID-19-reconvalescent BNT162b2-vaccinated individuals (cohort C) were collected at different timepoints after their first vaccination, as indicated. (A) Line graphs show individual courses for the neutralization capacity of ACE2–RBD interaction. (B) Boxplots show median neutralization capacities at the time of first vaccination (0 weeks), second vaccination (3 weeks), and 3 weeks after second vaccination (6 weeks). (C) Boxplots show median anti-NCP IgG titers for COVID-19-naïve versus COVID-19-convalescent donors. (D) Boxplots show neutralization capacities against RBD variants as indicated in BNT162b2-vaccinated COVID-19-naïve versus COVID-19-convalescent individuals. Box central horizontal lines in (B,C) indicate medians, box borders represent IQR, and whiskers indicate minima and maxima. For statistics, Šídák’s test for multiple comparisons (B,D), and the Mann–Whitney U test (C) were used. Significance levels were ** p < 0.005, *** p < 0.0005, and **** p < 0.00005. Abbreviations: COVID-19: coronavirus disease 2019; ACE2: angiotensin-converting enzyme 2; RBD: receptor-binding domain; IQR: interquartile range; n.s.: not significant.

Figure 3

Impact of a previous SARS-CoV-2 infection on the neutralizing immune response to vaccination with mRNA-1273. Serum samples from 10 mRNA-1273-vaccinated individuals without history of COVID-19 and 3 COVID-19-reconvalescent mRNA-1273-vaccinated individuals were collected at different timepoints after their first vaccination, as indicated. (A) Line graphs show individual courses for the neutralization capacity of ACE2–RBD interaction. (B) Boxplots show median neutralization capacities at the time of first vaccination (0 weeks), as well as 3 weeks and 6 weeks after first vaccination. (C) Boxplots show median anti-NCP IgG titers for COVID-19-negative versus COVID-19-convalescent donors. Box central horizontal lines indicate medians, box borders represent IQR, and whiskers indicate minima and maxima. For statistics, Šídák’s test for multiple comparisons (B) and the Mann–Whitney U test (C) were used. Significance levels were * p < 0.05, *** p < 0.0005, and **** p < 0.00005. Abbreviations: COVID-19: coronavirus disease 2019; ACE2: angiotensin-converting enzyme 2; RBD: receptor-binding domain; IQR: interquartile range; n.s.: not significant.

Figure 4

Impact of heterologous vaccination with ChAdOx1-nCoV-19 and an mRNA vaccine on neutralizing immune responses. Serum samples from 5 vaccinated individuals having received a homologous ChAdOx1-nCoV-19 vaccination (cohort H) and 36 vaccinated individuals having received a heterologous vaccination scheme (cohorts F and G) were collected at different timepoints after their second vaccinations. Line graphs show average courses for (A) the neutralization capacity of ACE2–RBD interaction, (B) anti-spike IgG titers, and (C) anti-spike IgA titers from 26 ChAdOx1/BNT162b2, 10 ChAdOx1/mRNA-1273, and 5 ChAdOx1/ChAdOx1 vaccinated individuals at different timepoints after their second vaccinations, as indicated. Statistical analysis at each timepoint was performed using Dunn’s multiple comparisons test. Significance levels were * p < 0.05, ** p < 0.005, and *** p < 0.0005. Abbreviations: ACE2: angiotensin-converting enzyme 2; RBD: receptor-binding domain.

Figure 4

Impact of heterologous vaccination with ChAdOx1-nCoV-19 and an mRNA vaccine on neutralizing immune responses. Serum samples from 5 vaccinated individuals having received a homologous ChAdOx1-nCoV-19 vaccination (cohort H) and 36 vaccinated individuals having received a heterologous vaccination scheme (cohorts F and G) were collected at different timepoints after their second vaccinations. Line graphs show average courses for (A) the neutralization capacity of ACE2–RBD interaction, (B) anti-spike IgG titers, and (C) anti-spike IgA titers from 26 ChAdOx1/BNT162b2, 10 ChAdOx1/mRNA-1273, and 5 ChAdOx1/ChAdOx1 vaccinated individuals at different timepoints after their second vaccinations, as indicated. Statistical analysis at each timepoint was performed using Dunn’s multiple comparisons test. Significance levels were * p < 0.05, ** p < 0.005, and *** p < 0.0005. Abbreviations: ACE2: angiotensin-converting enzyme 2; RBD: receptor-binding domain.

Figure 5

Impact of vaccination regimes on the neutralizing capacity against variants of concern. Serum samples from 23 subjects after homologous vaccination with BNT162b2, 23 subjects after heterologous vaccination with ChAdOx1/BNT162b2, 23 subjects after homologous vaccination with mRNA-1273, 23 subjects after heterologous vaccination with ChAdOx1/mRNA-1273, and 11 subjects after homologous vaccination with ChAdOx1 were collected 2 weeks after their second vaccinations. Serum samples from 20 unvaccinated COVID-19-convalescent individuals, collected at a median time of 14 weeks after diagnosis, served as control cohort. Samples were analyzed for neutralization capacity against wildtype RBD and three RBD variants of concern. Boxplots in panel (A) summarize neutralization capacities in the five vaccination cohorts and in an unvaccinated COVID-19-convalescent cohort. (B–D) Boxplots show neutralization capacities against wildtype RBD and (B) the B.1.1.7 RBD variant, (C) the B.1.351 RBD variant, and (D) the P.1 RBD variant. Box central horizontal lines indicate medians, box borders represent IQR, and whiskers represent minima and maxima. Significance levels were tested by Šídák’s test for multiple comparisons, with * p < 0.05, ** p < 0.005, *** p < 0.0005, and **** p < 0.00005. Abbreviations: IQR: interquartile range; ns: not significant; RBD: receptor-binding domain.

Figure 5

Impact of vaccination regimes on the neutralizing capacity against variants of concern. Serum samples from 23 subjects after homologous vaccination with BNT162b2, 23 subjects after heterologous vaccination with ChAdOx1/BNT162b2, 23 subjects after homologous vaccination with mRNA-1273, 23 subjects after heterologous vaccination with ChAdOx1/mRNA-1273, and 11 subjects after homologous vaccination with ChAdOx1 were collected 2 weeks after their second vaccinations. Serum samples from 20 unvaccinated COVID-19-convalescent individuals, collected at a median time of 14 weeks after diagnosis, served as control cohort. Samples were analyzed for neutralization capacity against wildtype RBD and three RBD variants of concern. Boxplots in panel (A) summarize neutralization capacities in the five vaccination cohorts and in an unvaccinated COVID-19-convalescent cohort. (B–D) Boxplots show neutralization capacities against wildtype RBD and (B) the B.1.1.7 RBD variant, (C) the B.1.351 RBD variant, and (D) the P.1 RBD variant. Box central horizontal lines indicate medians, box borders represent IQR, and whiskers represent minima and maxima. Significance levels were tested by Šídák’s test for multiple comparisons, with * p < 0.05, ** p < 0.005, *** p < 0.0005, and **** p < 0.00005. Abbreviations: IQR: interquartile range; ns: not significant; RBD: receptor-binding domain.

Figure 6

Comparison of SARS-CoV-2-specific T-cell response after homologous and heterologous vaccination. A total of 336 heparin blood samples from (A) 24 BNT162b2-vaccinated individuals, (B) 15 mRNA-1273-vaccinated individuals, and (C) 41 ChAdOx1-nCoV-19-primed vaccinated individuals were collected at different time points after their first vaccination, as indicated, and incubated overnight with a SARS-CoV-2-spike peptide mix as described in the Materials and Methods section. Then, supernatants were harvested and IFN-γ concentrations measured via ELISA. (A–C) Dot plots show IFN-γ concentrations at individual timepoints, as indicated. Lines connect different timepoints for corresponding individuals. Panel (A) also shows one case of an inverse heterologous vaccination with BNT162b2 and ChAdOx1-nCoV-19; in panel (B), 3 cases of COVID-19-convalescent mRNA-1273-vaccinated individuals are shown. Panel (C) shows results from a total of 205 timepoints from 41 individual ChAdOx1-nCoV-19-primed vaccinated individuals having continued with either homologous or heterologous vaccination regimes, as indicated by different color codes. Black lines indicate IFN-γ courses after first, and colored lines after second vaccinations. (D) Boxplots show peak IFN-γ concentrations 1–2 weeks after second vaccination in the various vaccination groups, as indicated. Box central horizontal lines indicate medians, box borders indicate IQR, and whiskers represent minima and maxima. Significance levels were determined by Dunn’s multiple comparisons test, with * p < 0.05, ** p < 0.005, and **** p < 0.00005. Abbreviations: IFN-γ: interferon-gamma; IQR: interquartile range; ns: not significant.

Figure 6

Comparison of SARS-CoV-2-specific T-cell response after homologous and heterologous vaccination. A total of 336 heparin blood samples from (A) 24 BNT162b2-vaccinated individuals, (B) 15 mRNA-1273-vaccinated individuals, and (C) 41 ChAdOx1-nCoV-19-primed vaccinated individuals were collected at different time points after their first vaccination, as indicated, and incubated overnight with a SARS-CoV-2-spike peptide mix as described in the Materials and Methods section. Then, supernatants were harvested and IFN-γ concentrations measured via ELISA. (A–C) Dot plots show IFN-γ concentrations at individual timepoints, as indicated. Lines connect different timepoints for corresponding individuals. Panel (A) also shows one case of an inverse heterologous vaccination with BNT162b2 and ChAdOx1-nCoV-19; in panel (B), 3 cases of COVID-19-convalescent mRNA-1273-vaccinated individuals are shown. Panel (C) shows results from a total of 205 timepoints from 41 individual ChAdOx1-nCoV-19-primed vaccinated individuals having continued with either homologous or heterologous vaccination regimes, as indicated by different color codes. Black lines indicate IFN-γ courses after first, and colored lines after second vaccinations. (D) Boxplots show peak IFN-γ concentrations 1–2 weeks after second vaccination in the various vaccination groups, as indicated. Box central horizontal lines indicate medians, box borders indicate IQR, and whiskers represent minima and maxima. Significance levels were determined by Dunn’s multiple comparisons test, with * p < 0.05, ** p < 0.005, and **** p < 0.00005. Abbreviations: IFN-γ: interferon-gamma; IQR: interquartile range; ns: not significant.