Comparable neutralisation evasion of SARS-CoV-2 omicron subvariants BA.1, BA.2, and BA.3.

The SARS-CoV-2 omicron (B.1.1.529) variant has rapidly become globally dominant, displacing the previously dominant delta (B1.617.2) variant. The viral spike (S) protein is the key target of the neutralising antibody response, and the omicron variant harbours more than 35 mutations in the S protein, which allow highly efficient evasion from neutralising antibodies. In keeping with these findings, the omicron variant efficiently spreads in populations with a high percentage of convalescent or vaccinated individuals.2, 3

The three main subvariants of the omicron variant are BA.1, BA.2, and BA.3. Initial data suggest that BA.2 might have a growth advantage over BA.1, posing a rapidly increasing threat to health systems. The omicron subvariants display remarkable differences regarding S protein mutations, particularly with respect to the N-terminal domain and the receptor-binding domain (appendix pp 2–3), which are known to harbour key epitopes of neutralising antibodies.5, 6 Here, we compared BA.1, BA.2, and BA.3 for sensitivity to neutralisation by antibodies induced by infection and vaccination, using pseudoviruses as a model system, which adequately mirrors SARS-CoV-2 neutralisation by antibodies.

We analysed particles harbouring the S protein of B.1—which is identical to the wildtype strain apart from the D614G mutation—and S proteins of BA.1, BA.2, and BA.3. We first examined neutralisation by antibodies from convalescent patients, who were infected during the first (February to May, 2020) and second (December, 2020, to February, 2021) waves of COVID-19 in Germany (appendix pp 2–3, 4–6). Neutralisation of particles bearing the B.1 S protein (B.1pp) was robust, whereas neutralisation of BA.1pp and BA.3pp was at least 32-times less than B.1pp (BA.1 p=0·0020; BA.3 p=0·0020). Neutralisation of BA.2pp was also diminished, but the reduction was less pronounced than that measured for the other omicron subvariants (9·2-times less than B.1pp; p=0·0020).

Analysis of neutralisation by antibodies induced by double vaccination with BNT162b2 (BNT) yielded similar results as neutralisation with antibodies from convalescent patients (appendix pp 2–3). Particles harbouring the S proteins of BA.1 and BA.3 showed 17-times lower neutralisation than B.1pp (BA.1 p=0·0020; BA.3 p=0·0020), whereas neutralisation of BA.2pp was 9-times reduced (p=0·0020). Triple BNT vaccination induced a more potent antibody response, and only modest evasion of neutralisation was seen for particles bearing omicron S proteins (BA.1 2·5-times, p=0·0039; BA.2 1·9-times, p=0·012; BA.3 2·4-times, p=0·0039; appendix pp 2–3). Finally, neutralisation by antibodies induced in fully vaccinated (three vaccine doses) individuals with breakthrough infection during the fourth wave in Germany (October, 2021, to January, 2022, dominated by the delta variant) was most potent and neutralisation of particles bearing omicron S protein was 9–12-times less efficient than B.1pp (BA.1 p=0·0020; BA.2 p=0·0039; BA.3 p=0·0039; appendix pp 2–3). However, no significant differences were observed between BA.1pp, BA.2pp, and BA.3pp (appendix pp 2–3).

Our results show that all presently circulating omicron subvariants evade neutralisation by vaccine-induced antibodies with comparably high efficiency, suggesting that increased antibody evasion is not the reason for the current expansion of BA.2 in several countries.4, 8 Since currently available vaccines provided robust protection against early omicron isolates circulating in South Africa from Nov 15 to Dec 7, 2021, which was likely to be BA.1, our results suggest that this protection should extend to all omicron subvariants.

SP acknowledges funding from Bundesministerium für Bildung und Forschung (BMBF; grant numbers 01KI2006D, 01KI20328A, 01KX2021), the Ministry for Science and Culture of Lower Saxony (grant numbers 14-76103-184, MWK HZI COVID-19), and the German Research Foundation (DFG; grant numbers PO 716/11-1, PO 716/14-1). MSW received unrestricted funding from Sartorius, Lung research. H-MJ received funding from BMBF (grant numbers 01KI2043, NaFoUniMedCovid19-COVIM 01KX2021), Bavarian State Ministry for Science and the Arts, and DFG through the research training groups RTG1660 and TRR130, the Bayerische Forschungsstiftung (Project CORAd), and the Kastner Foundation. GMNB acknowledges funding from the German Center for Infection Research (grant number 80018019238) and a European Regional Development Fund (Defeat Corona, grant number ZW7-8515131, together with AD-J). All other authors declare no competing interests.