AstraZeneca COVID-19 vaccine induces robust broadly cross-reactive antibody responses in Malawian adults previously infected with SARS-CoV-2

Abstract

Background: Binding and neutralising anti-Spike antibodies play a key role in immune defence against SARS-CoV-2 infection. Since it is known that antibodies wane with time and new immune-evasive variants are emerging, we aimed to assess the dynamics of anti-Spike antibodies in an African adult population with prior SARS-CoV-2 infection and to determine the effect of subsequent COVID-19 vaccination.

Methods: Using a prospective cohort design, we recruited adults with prior laboratory-confirmed mild/moderate COVID-19 in Blantyre, Malawi, and followed them up for 270 days (n = 52). A subset of whom subsequently received a single dose of the AstraZeneca COVID-19 vaccine (ChAdOx nCov-19) (n = 12). We measured the serum concentrations of anti-Spike and receptor-binding domain (RBD) IgG antibodies using a Luminex-based assay. Anti-RBD antibody cross-reactivity across SARS-CoV-2 variants of concern (VOC) was measured using a haemagglutination test. A pseudovirus neutralisation assay was used to measure neutralisation titres across VOCs. Ordinary or repeated measures one-way ANOVA was used to compare log10 transformed data, with p value adjusted for multiple comparison using Šídák's or Holm-Šídák's test.

Results: We show that neutralising antibodies wane within 6 months post mild/moderate SARS-CoV-2 infection (30-60 days vs. 210-270 days; Log ID₅₀ 6.8 vs. 5.3, p = 0.0093). High levels of binding anti-Spike or anti-RBD antibodies in convalescent serum were associated with potent neutralisation activity against the homologous infecting strain (p < 0.0001). A single dose of the AstraZeneca COVID-19 vaccine following mild/moderate SARS-CoV-2 infection induced a 2 to 3-fold increase in anti-Spike and -RBD IgG levels 30 days post-vaccination (both, p < 0.0001). The anti-RBD IgG antibodies from these vaccinated individuals were broadly cross-reactive against multiple VOCs and had neutralisation potency against original D614G, beta, and delta variants.

Conclusions: These findings show that the AstraZeneca COVID-19 vaccine is an effective booster for waning cross-variant antibody immunity after initial priming with SARS-CoV-2 infection. The potency of hybrid immunity and its potential to maximise the benefits of COVID-19 vaccines needs to be taken into consideration when formulating vaccination policies in sub-Saharan Africa, where there is still limited access to vaccine doses.

Keywords: Antibodies; AstraZeneca COVID-19 vaccine; RBD; SARS-CoV-2; Spike; VOC.

Fig. 1

Antibody responses following mild/moderate SARS-CoV-2 infection. A Presence of anti-Spike IgG in convalescence serum (n = 52). B Pseudovirus neutralisation activity in anti-Spike IgG seropositive sera following mild/moderate COVID-19 (n = 45). Data was log 10 transformed and statistics were calculated using ordinary one-way ANOVA and p value adjusted for multiple comparison using Šídák’s multiple comparisons test. Vertical bars represent geometric mean and horizontal bars represent 95% confidence intervals. A p < 0.05 was regarded as statistically significant

Fig. 2

SARS-CoV-2 specific antibody responses in re-infected individuals. Two participants had a laboratory-confirmed SARS-CoV-2 reinfection during the follow-up period of the study. They were both re-infected during the beta variant wave. Kinetics of anti-Spike, anti-RBD, and neutralisation activity in serum of A 46-year-old male and B 49-year-old female. S, full-length Spike; RBD, receptor-binding domain; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2

Fig. 3

Neutralisation activity against the original variant in recovered mild/moderate COVID-19 patients with various levels of anti-Spike and RBD IgG antibodies. Magnitude of neutralisation activity against the original variant (D614G) in first wave sera with varying concentrations of A anti-Spike IgG and B anti-RBD IgG antibodies. The Spike and RBD protein antigens were from the original D614G variant. Vertical bars represent geometric mean and horizontal bars represent 95% confidence intervals. Data was log 10 transformed and statistics were calculated using ordinary one-way ANOVA and p value adjusted for multiple comparison using Šídák’s multiple comparisons test (n = 30). A p < 0.05 was regarded as statistically significant. S, full-length Spike; RBD, receptor-binding domain; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2

Fig. 4

Antibody responses following adenovirus COVID-19 vaccination in previously infected adults. A Concentration of anti-Spike and anti-RBD IgG antibodies pre- and post-vaccination with a single dose of AstraZeneca COVID-19 vaccine. All antigens were from the original strain. B Representative plate of the haemagglutination test against the wild type, alpha, beta, gamma, and delta variants using pre- and post-vaccination sera. The sera were diluted 2-fold for 6 dilutions. CR033 and EY-6A are used are positive control, and red blood cells from blood group O negative are used as negative control. C Heat map showing collated data of HAT titres against the wild type, alpha, beta, gamma, and delta variants in pre- and post-vaccination sera. D Magnitude of neutralisation activity against the beta, D614G, and delta variants in pre- and post-vaccination sera. E Number of individuals with neutralisation activity against Beta or D614G or Delta variants in pre- and post-vaccination sera. Statistics were calculated using RM one-way ANOVA and p values adjusted for multiple comparison using Holm-Šídák’s multiple comparisons test. A p < 0.05 was regarded as statistically significant. S, full length Spike; RBD, receptor-binding domain; HAT, haemagglutination test; NC, negative control; PC, positive control; SARS-CoV-2, severe acute respiratory syndrome coronavirus