Neutralization escape by SARS-CoV-2 Omicron subvariant BA.2.86

Abstract

The Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) variant BA.2.86 has over 30 mutations in spike compared with BA.2 and XBB.1.5, which raised the possibility that BA.2.86 might evade neutralizing antibodies (NAbs) induced by vaccination or infection. In this study, we show that NAb titers are substantially lower to BA.2.86 compared with BA.2 but are similar or slightly higher than to other current circulating variants, including XBB.1.5, EG.5.1, and FL.1.5.1. Moreover, NAb titers against all these variants were higher in vaccinated individuals with a history of XBB.1.5 infection compared with vaccinated individuals with no history of XBB.1.5 infection, suggesting the potential utility of the monovalent XBB.1.5 mRNA boosters.

Keywords: BA.2.86; Neutralizing antibody; SARS-CoV-2; Vaccine; mRNA.

Figure 1.. Transitions in SARS-CoV-2 variant lineages in 2023.

(A) Global frequencies based on GISAID data. Note that recent sampling is biased due to sparse sampling given the inevitable time delays between sampling and sequence submissions to GISAID. EG.5.1 sampling is currently increasing the fastest globally. (B) shows the countries were the emergent lineages are most sampled. EG.5.1 has become the dominant lineage in China, and is increasing everywhere it is sampled. XBC.1 and XBC.1.6 are common in Australia (magenta), but no longer increasing there. The recently detected highly divergent variant BA.2.86 has 66 sequences available in GISAID as of this writing, 2023-09-30; its sublineage BA.2.86.1 and has spread globally more quickly, and has been sampled 238 times. As BA.2.86.1 is distinguished by only two mutations (ORF1a:K1973R, C12815T), both outside of spike, both are grouped here and called BA.2.86. BA.2.86 can be visualized on the global plot in (A) as the thin blue line at the top. The county where BA.2.86 has the highest frequency of sampling is South Africa, while the country where BA.2.86.1 has the highest frequency of sampling is Sweden. The FL.1.5.1 sublineage is increasing in the USA, and New York is showing the most rapid increase in FL.1.5.1 sampling frequency at the state level which is highlighted here. FL.1.5.1 can be differentiated from FL.1 by 2 additional mutations in Spike, F456L and T478R. Of note, these are common mutations in currently expanding lineages.

Figure 2.. Spike mutations in current circulating SARS-CoV-2 variants.

Substitutions in the ancestral BA.2 Omicron lineage relative to the Wuhan/WIV04/ reference strain (https://gisaid.org/WIV04/) are shown in grey. Additional substitutions found in XBB.1.5 are highlighted in teal. Additional substitutions in all other study variants relative to these two patterns are indicated for each variants studied. The mutational pattern used matched the consensus form of the expanding lineage, although in BA.2.86 the I670V mutation was present in the two earliest sampled intact sequences. The highlighted region in XBC.1.6 is a recombinant fragment from a Delta variant.

Figure 3.. Structural models of BA.2.86 mutations.

(A) Structural mapping of key neutralizing antibody epitopes. Spike trimer structure in the one-up configuration from Benton et al. (PDB: 7A94) is used. RBD class and epitope definitions are from Barnes et al., and NTD supersite epitope definitions are from Cerutti et al.. Epitopes are color-coded per legend. (B-D) Mutations in BA.2.86 versus BA.2, Wuhan-1, and XBB.1.5. Sites with either new or different mutations in BA.2.86 with respect to the comparator variant are shown in red, and sites that have mutations in the comparator strain but not in BA.2.86 are shown in blue. An N-linked glycosylation site introduced in BA.2.86 at Spike N354 (resulting from a K356T change) and insertions and deletions are in bold as they may be particularly impactful. The pie charts on the right show the fraction of sites with amino acid differences between BA.2.86 and the comparator variant that fall in each of the epitopes from (A); grey indicates fraction of sites with sequence differences that did not fall in any of the epitopes from (A). The total number of sites with sequence differences is indicated in the center, and sites that occur in multiple epitopes are counted in each epitope.

Figure 4.. Neutralization escape from SARS-CoV-2 variants.

Neutralizing antibody (NAb) against the WA1/2020, BA.1, BA.2, BA.5, XBB.1.5, XBB.1.16, EG.5, EG.5.1, FL.1.5.1, XBC.1.6, and BA.2.86 variants by luciferase-based pseudovirus neutralization assays at baseline and at 6 months in individuals who did or did not receive the bivalent mRNA booster in fall 2022. Also shown are participants at 6 months who were infected during the XBB.1.5 predominant period. The horizontal red bar reflects median values. The bivalent mRNA booster cohort with no XBB infection had n=32 participants, and the bivalent mRNA booster cohort with XBB infections had n=12 participants. The no bivalent booster cohort with no XBB infection had n=9–19 participants, and the no bivalent booster cohort with XBB infection had n=5 participants.

Figure 5.. NAb titers induced by bivalent mRNA vaccines against SARS-CoV-2 variants during peak immunogenicity.

NAb against the WA1/2020, BA.1, BA.2, BA.5, XBB.1.5, XBB.1.16, EG.5, EG.5.1, FL.1.5.1, XBC.1.6, and BA.2.86 variants by luciferase-based pseudovirus neutralization assays at week 3 following the bivalent mRNA booster in fall 2022. The horizontal red bar reflects median values. This cohort had n=27 participants.