Immunogenicity of Ad26.COV2.S vaccine against SARS-CoV-2 variants in humans

The Ad26.COV2.S vaccine1–3 has demonstrated clinical efficacy against symptomatic COVID-19, including against the B.1.351 variant that is partially resistant to neutralizing antibodies1. However, the immunogenicity of this vaccine in humans against SARS-CoV-2 variants of concern remains unclear. Here we report humoral and cellular immune responses from 20 Ad26.COV2.S vaccinated individuals from the COV1001 phase I–IIa clinical trial2 against the original SARS-CoV-2 strain WA1/2020 as well as against the B.1.1.7, CAL.20C, P.1 and B.1.351 variants of concern. Ad26.COV2.S induced median pseudovirus neutralizing antibody titres that were 5.0-fold and 3.3-fold lower against the B.1.351 and P.1 variants, respectively, as compared with WA1/2020 on day 71 after vaccination. Median binding antibody titres were 2.9-fold and 2.7-fold lower against the B.1.351 and P.1 variants, respectively, as compared with WA1/2020. Antibody-dependent cellular phagocytosis, complement deposition and natural killer cell activation responses were largely preserved against the B.1.351 variant. CD8 and CD4 T cell responses, including central and effector memory responses, were comparable among the WA1/2020, B.1.1.7, B.1.351, P.1 and CAL.20C variants. These data show that neutralizing antibody responses induced by Ad26.COV2.S were reduced against the B.1.351 and P.1 variants, but functional non-neutralizing antibody responses and T cell responses were largely preserved against SARS-CoV-2 variants. These findings have implications for vaccine protection against SARS-CoV-2 variants of concern.


Article
Comparable IgG, IgM and IgA subclasses and Fc-receptor binding were observed across the variants, with only a slight loss in FcγR2b binding compared to the WA1/2020 strain (Fig. 2b). RBD-specific ADCP, ADNP and ADCD responses were comparable against the WA1/2020, B.1.1.7 and B.1.351 variants (Extended Data Fig. 2). These data show robust spike-and RBD-specific Fc-effector functions against these SARS-CoV-2 variants.

Cellular immune responses to variants
Spike-specific cellular immune responses were assessed by pooled peptide ELISPOT assays in peripheral blood mononuclear cells on days 57 and 85. IFNγ ELISPOT responses were comparable to WA1/2020, B.1.351, B.1.1.7, P.1 and CAL.20C at both time points, with no evidence of decreased responses against the variants (Fig. 3a). No spike-specific ELISPOT responses were observed in vaccine recipients who received placebo. Spike-specific CD8 + and CD4 + T cell responses were evaluated by multiparameter intracellular cytokine staining (ICS) assays on days 57 and 85 (Extended Data Fig. 3 Fig. 4). Central memory CD27 + CD45RA − and effector memory CD27 − CD45RA − CD4 + and CD8 + T cell responses were also comparable across these variants (Extended Data Figs. 5, 6). These data show that spike-specific cellular immune responses were not detectably affected by SARS-CoV-2 variants. Polyfunctional analyses showed that CD8 + T cells were primarily IFNγ, TNF and both IFNγ and TNF responses, whereas CD4 + T cells were primarily TNF; IFNγ and TNF; IL-2 and TNF; and IFNγ, IL-2 and TNF responses (Extended Data Fig. 7).
To evaluate the specificity and breadth of individual T cell receptors (TCRs) after vaccination, TCRβ sequencing 20 was performed to define the repertoires of 8 convalescent individuals and 19 participants receiving the vaccine and 5 receiving placebo on day 63 (Extended Data  Table 1). To identify SARS-CoV-2 specific TCRs, the observed TCRs were compared to a TCR dataset that had previously been determined to be SARS-CoV-2-specific and enriched in subjects with natural infection relative to placebos 21 . The breadth (unique rearrangements) and depth (frequency of TCRs) of TCRs specific to either spike or non-spike SARS-CoV-2 proteins were determined, although these analyses may have underestimated total T cell responses. Higher breadth of spike-specific TCRs was observed in vaccine recipients compared with placebos (P = 0.0014, Wilcoxon rank-sum test) (Fig. 4a, Extended Data Figs. 8,9). By contrast, the breadth of non-spike TCRs was comparable in vaccine recipients and controls, as expected because the vaccine did not contain any non-spike immunogens. Substantial breadth of CD8 + and CD4 + T cell responses was also observed (Fig. 4b).
Discussion SARS-CoV-2 variants have emerged with several mutations in targets of neutralizing antibodies, such as the E484K mutation. Median pseudovirus neutralizing antibody titres induced by Ad26.COV2.S were 5.0-fold lower against the B.1.351 variant and 3.3-fold lower against the P.1 variant as compared with the original WA1/2020 strain, which is a comparable reduction of psVNA titres that has been reported for other vaccines 4,6,7 . By contrast, functional non-neutralizing antibody responses and CD8 + and CD4 + T cell responses were largely preserved against SARS-CoV-2 variants of concern.
In the phase III ENSEMBLE trial 1 , Ad26.COV2.S was evaluated in the USA, Latin America including Brazil, and South Africa. In South Africa, 95% of sequenced viruses from COVID-19 cases were of the B.1.351 variant, and in Brazil, 69% of sequenced viruses from COVID-19 cases were of the P.2 lineage. Protective efficacy of Ad26.COV2.S against severe or critical disease was similar in all geographic locations by day 28, and protective efficacy against moderate to severe disease was only slightly reduced in South Africa compared with the USA. Although the mechanistic correlates of protection for COVID- 19 are not yet known, the robust protective efficacy in these regions despite reduced neutralizing antibodies raises the possibility that functional non-neutralizing antibodies and/or CD8 + T cell responses may also contribute to protection. Indeed, TCRβ sequencing revealed substantial breadth of T cell responses in individuals vaccinated with Ad26. COV2.S. Alternatively, it is possible that low levels of neutralizing antibodies are sufficient for protection. In a non-human primate model, adoptive transfer of purified IgG was sufficient for protection against SARS-CoV-2 if titres of psVNA exceeded a threshold of approximately 50, but CD8 + T cells also contributed to protection if antibody titres were subprotective 22,23 .
In conclusion, neutralizing antibody responses elicited by Ad26. COV2.S were reduced against the B.1.351 and P.1 variants, but other functional antibody responses and T cell responses were largely preserved against these variants. The relevance of these immune parameters to mechanistic correlates of vaccine efficacy remains to be determined.

Online content
Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at https://doi.org/10.1038/s41586-021-03681-2.