Our preprint on the evolution and antigenic properties of the #SARSCoV2 lineage BA.2.86 is now out on medRxiv. The lab of Alex Sigal has isolated the variant and characterized its properties. @corneliusroemer and myself have contributed a phylogenetic analysis of the variant.
The preprint is here:
https://www.medrxiv.org/content/10.1101/2023.09.08.23295250v1
An interactive version of the phylogeny can be found here:
https://nextstrain.org/groups/neherlab/ncov/BA.2.86/paper
In the thread below, I want to summary some more thoughts on the variant.
[1/10]
@corneliusroemer
This phylogeny shows a global sample of BA.2 sequences from early 2022 and samples of BA.2.86 (earliest collection in late July 2023). BA.2.86 connects to the to BA.2 via a very long branch without sampled intermediates. Along this branch, the spike protein has changed in about 30 position through substitutions, deletions, and one insertion.
[2/10]
This divergence is comparable to the distance between viruses from 2020 and the initial Omicron variants, which raised concern that this variant might be very immune evasive and could spread rapidly. The fact that it had spread globally at time of detection added to that concern.
While data by the Sigal lab and others, as well as the dynamics since the initial discovery have dampened this concern, the impact of BA.2.86 on future circulation is still uncertain.
[3/10]
@corneliusroemer
As expected, BA.2.86 is poorly recognized by sera from vaccinees (pre-Omicron, panel E). It is also poorly recognized by sera from individuals with BA.1 infection. But sera from break-through infections generally neutralized BA.2.86 and XBB at similar (rather low) levels (C&D).
This shows is that BA.2.86 and XBB.1.5 are 'similarly far away' from the neutralization profile of these sera, but the antigenic properties of BA.2.86 and XBB might still differ substantially.
[4/10]
We also investigated possible origin and emergence scenarios. The branch leading the BA.2.86 shares two mutations with BA.2 samples from early 2022, which were more common at the time in Southern Africa than elsewhere.
Furthermore, BA.2.86 lacks a mutation that the majority of BA.2 samples in the rest of the world had (see screen shot below).
The long branches dominated by changes in Spike are suggestive of a chronic infection in a immuno-compromised individual.
[5/10]
@corneliusroemer
BA.2.86 itself is geographically very broadly distributed and has been observed in North America, Australia, Asia, Europe, and Southern Africa. But out of the 10 branches of the basal polytomy of BA.2.86, 8 are dominated by samples from South Africa, while most samples in the Northern Hemisphere descend from a second polytomy two mutations away from the base.
Both observations are suggestive of emergence in Southern Africa, though not direct evidence.
[6/10]
Throughout the pandemic, the rate of evolution of SARS-CoV-2 showed a dichotomous pattern where long branches on the backbone of the tree accumulate mutations rapidly at a rate of ~30/year, while typical acute transmission chains are about two-fold slower. I investigated this rate difference in a paper last year. This same pattern of rapid evolution along the long branch and slow accumulation otherwise seems to hold for BA.2.86.
https://academic.oup.com/ve/article/8/2/veac113/6887176
[7/10]
Currently available BA.2.86 sequences where mostly sampled in mid-August and most differ from the basal polytomy by 3-5 mutations.
With that slower rate of 15 mutations per year, 3-5 mutations translate into 2-4 months which suggests an emergence in early May (+/- 6 weeks). A molecular clock analysis using TreeTime that allows for this rate dichotomy is consistent with this rule-of-thumb estimate.
[8/10]
BA.2.86 is clearly transmitting well and has likely caused 100s of thousands infections since its emergence earlier this year. Otherwise it would not have been picked up by the much scaled down surveillance. But it is spreading more slowly than the initial Omicron variants spread at the time.
[9/10]
How competition between XBB-like variants and BA.2.86 will play out is unclear. BA.2.86 could disappear, it could adapt further and take over, or it could co-circulate with other variants and form a separate lineage that is antigenically distinct enough to co-exist with XBB.
[10/10]