14 Sept 2023 Episode 349 Update on BA.2.86 and other variants in view of the upcoming XBB.1.5 booster

Thu, 09/14/2023 - 18:05

Episode 349: Updating on emerging variants BA.2.86, EG 5.1, Flip variant etc and their  potential impact on the course of the epidemic and on the effect of the booster campaign with XBB.1.5 variant

Dear colleagues,

Sorry for the long title with so many cryptic abbreviations, but things have become more complex than in the “bad old times” with monolithic alpha, beta, gamma delta waves….  No matter how complicated the present situation is, the good news is that, despite all the mutations in those variants, there is no convincing evidence yet that any of those will create a new 2020-2021 disaster.   Although some of these variants are intrinsically very fit, they are confronted now with a rather robust “hybrid herd immunity”, as a consequence of repeated infections and vaccinations in the population.  However, as you know, we must keep a watchful eye on this vicious virus.  The purpose of this Episode  is to provide you with the scientific elements to form your own opinion.

For more background on the new variants see also Ep 304 on XBB.1, Ep 331 (XBB.1.16),  Ep 342 (EG.5) and Ep 345 (BA.2.86). 

For those who are a bit lost, I can recommend to read first  Ep 349-0 “The conversation” on BA.2.86.



Ep 349-1: Sheward bioRxiv 2 Sept 2023   New variants in a complex network


Circles show estimated variant growth advantage:  FL.1.5.1 and HK.3 have a clear advantage, as compared to XBB.1.5.


As a reminder:

XBB is a recombinant between a BA.2 variant (BJ.1) and a BA.2.75 variant (BM.1.1.1)


From this phylogenetic tree, it is evident that BA.2.86 is very different from the BA.2 ancestor (over 30 mutations), but also from the recombinant variants XBB.1.5 (the basis for the next booster) and from other “emerging variants” such as XBB.1.16, EG.5 and FL.1.5.1.   The latter three carry an important immune escape mutation F456L. Moreover, in addition, some carry the “flip” mutation L455F as well (e.g. FL.1.5.1 and XBB.1.16, but also HK.3), providing even more immune escape.     

See Ep 349-2: ECDC report on variants Sept 2023




Ep 349-3: ECDC report of Sept 7

  • In recent weeks, signals of SARS-CoV-2 transmission have increased from previously very low levels in the EU/EEA. Factors unrelated to the genetic evolution of SARS-CoV-2 likely contributed to observed increases in epidemiological indicators, such as large gatherings and increased travel during the seasonal holidays, as well as waning levels of immunological protection against infection – but not severe disease – in the population.
  • Emergence and subsequent dominance of a group of related Omicron sub-lineages, XBB.1.5-like variants carrying the F456L mutation
  • A highly mutated Omicron sub-lineage, BA.2.86: …. low-level community transmission
  • No indication that infection with XBB.1.5-like+F456L variants or BA.2.86 are associated with more severe disease or a reduction in vaccine effectiveness against severe disease
  • Vaccination efforts should focus on protecting those at risk of progression to severe disease, e.g. people aged over 60 years and other vulnerable individuals irrespective of age (such as people with underlying comorbidities or immunocompromised conditions).



See also very similar conclusions of CDC (Ep 349-4 A and B)


Global transition (from Ep 349-9)


Ep 349-5: Morten Rasmussen Eurosurveillance 31 August on first 10 cases of BA.2.86 in Denmark

Given the widespread distribution of BA.2.86 globally (US, UK,  South-Africa, Israel) and within Denmark in clinical samples and in wastewater, and the high number of mutations conferring immune-evasive potential, the possibility that BA.2.86 will outcompete currently dominating variants cannot be ruled out.

 Our very early clinical data suggest a clinical picture in line with the typical COVID-19 illness caused by previous variants (no increased severity)



Ep 349-6: UKSHA 9 Sept: Data from England on BA.2.86

As of 4 September 2023, there are 34 BA.2.86 sequenced cases in England, with cases located in East of England, London, and North West England. Of 34 total cases in England, 5 were hospitalised (2 have unknown hospitalisation status), and no deaths due to COVID-19 have been reported.

These cases include 28 sequenced as part of an investigation of an outbreak in a care home which reported a high attack rate:

  • 33 out of 38 residents tested positive for SARS-CoV-2
  • 19 of 33 symptomatic
  • 1 hospitalized, none died


As a consequence of this outbreak in a nursing home, the age distribution of all cases skewed towards older women

In total:

Conclusion: Amongst the many co-circulating variants in the world, some are of particular concern:

  • The XBB.1.5 descendants with the additional F456L and L455F “escape” mutations, including EG.5, EG.5.1, XBB.1.16 and others, which are showing wide epidemiological spread in Europe and US
  • The BA.2.86 variant, which is highly mutated.  It has been shown in several countries (including US, UK, Denmark, Sweden, France, Portugal South-Africa and Israel) at low level, but with clear evidence of community transmission.

Those variants are of particular concern, because the next booster is based on XBB.1.5.  The questions are

  • whether some of these variants could escape XBB.1.5 induced immunity
    • after infection
    • or vaccination
  • and what their sensitivity is to therapeutic monoclonal antibodies (mAbs).  These questions will be addressed in the next paragraph.




PAR 2 SENSITIVITY OF NEW VARIANTS to XBB.1.5 induced immunity and to mAbs (based on Ep 349-7 by Eric Topol)

Returning to Ep 349-1: Sheward also shows that the XBB.1.5 infection wave in Sweden has boosted the neutralizing antibodies against both XBB.1.5, but also against BA.2 and BA.2.86, suggesting that an XBB.1.5 vaccine booster may also significantly increase immunity against BA.2.86


Pseudovirus neutralisation by serum samples taken Sept 26 - Oct 2, 2022, prior to the arrival of XBB in Sweden, as well as by current samples (Aug 21-27, 2023 = after many have been infected with XBB).


Unfortunately, the BA.2.86 variant (unlike BA.2 and XBB.1.5) seems to be highly resistant to Sotrovimab




Ep 349-8: Uriu bioRxiv 7 Sept confirms the data by Sheward

  1. BA.2.86 seems to have a higher reproductive number > EG.5.1> XBB.1.16 > XBB.1.5 (based on data from Denmark)


  1. BA.2.86 is resistant to all therapeutic monoclonals, including Sotrovimab, while BA.2 and EG.5.1 are sensitive to Sotrovimab


  1. Sera from subjects repeatedly vaccinated with monovalent (F-G) or even bivalent vaccines (H-I) have very weak neutralizing capacity against EG.5.1 and BA.2.86, these variants are more sensitive to sera from XBB.1.5 breakthrough infections (K).


These data from Japan are slightly different from the previous “X-Twitter” preprint from Yonlung Cao (Bejing)


According to C, there BA.2.86 still shows some sensitivity to Sotrovimab (S309), but IC50 indeed > 1 µg/ml

According to D, BA.2.86 is less (not more) infectious than EG.5.1 and XBB.1.5

According to B repeated superinfections are needed to raise the plasma neutralizing capacity to significant (> 100) titers.

Ep 349-9: Ninaad Lasrado bioRxiv 5 Sept


Clearly, neutralizing Ab titers against BA.2.86 are overall higher than against XBB.1.5 and it’s variants and a, XBB infection boosts these titers.  


All previous results are based on pseudovirus assays. = an artificial “recombinant” and non-replicating virus (based on for instance Vesiculo Stomatitis Virus or VSV), which has been genetically modified to express the various envelopes of SARS-CoV-2 variants.  


Ep 349-10: Khadija Khan medRxiv 8 Sept

Data from South-Africa shows in a live virus neutralization that sera from (un) vaccinated people with various omicron (breakthrough) infections neutralize BA.2.86 to a similar extent as XBB.1.5 live viruses.


(C) Neutralization of BA.2.86 live virus versus XBB.1.5 by sera from vaccinated individuals with breakthrough infection (BTI) with BA.1/BA.4/BA.5/XBB subvariants. Numbers above columns are group geometric mean titer FRNT50 and foldchange.

(D) BA.2.86 versus XBB.1.5 neutralization by sera from unvaccinated individuals infected with Omicron BA.1/BA.4/BA.5/XBB family subvariants.

(E) Neutralization of BA.2.86 versus ancestral D614G and BA.1 viruses by sera from vaccinated individuals collected before Omicron emergence.

(F)Neutralization of BA.2.86 versus Omicron BA.1 by sera from vaccinated and unvaccinated BA.1 infected individuals.

Significant p-values were ****p<0.0001, **p=0.0042 by the Wilcoxon Rank Sum test.


Ep 349-11: Yanping Hu bioRxiv 11 Sept confirm the trend seen in 349-9 and 349-11 in life virus neutralization: BA.2.86 seems less immune evasive than XBB.1.6 subvariants


Clearly, EG.5.1 and FL.1.5.1 (flip mutants L455F + F456L) and CH.1.1are most immune evasive, but XBB.1.5 breakthrough infection boosts the neutralizing titers.

Ep 349-12: Spyros Chalkais medRxiv update on 8 Sept of data from MODERNA monovalent XBB.1.5 vaccinees:

These subjects received the XBB.1.5 booster (mRNA-1273.815) as 5th doses,

  • They previously received a primary series and 3rd dose of an original mRNA Covid-19 vaccine,
  • and a 4th dose of a bivalent (omicron BA.4/BA.5 and original SARS-CoV-2) vaccine.

Interim immunogenicity data 15 days post-vaccination are presented in a VSV pseudovirus assay.



Clearly, titers before and after booster are highest against D614G (= original strain) followed by BA.4/BA.5.

Titers against all XBB.1.5 variants and against BA.2.86 are similar, with a clear increase of around 10 X after the booster.


SARS-CoV-2 is trying all kinds of “tricks” to defy the established herd immunity in the human population.  This “struggle” results in a complex picture of emerging new variants and subvariants, most of which are related to XBB.1.5. The highly mutated BA.2.86 is very divergent, it is presently spreading, but still at low levels.

Although the number of infections and hospitalizations worldwide is on the rise, as expected in this season, there is no evidence that any of the new subvariants (incl BA.2.86) results in more severe disease, but data are scattered and preliminary.

The fear was that some of the variants would be less sensitive to the autumn booster, based on XBB.1.5.    

  • Data on the sensitivity of the XBB.1.5 subvariants and BA.2.86 to neutralizing sera from patients with XBB breakthrough infections are somewhat divergent: most pseudovirus-based assays point to a lower sensitivity of EG.5 and especially BA.2.86, but live virus assays show rather similar sensitivity to serum neutralization.  
  • The first data of the vaccine itself are rather reassuring: the booster increases neutralizing titers against all subvariants  (incl BA.2.86) by a factor 10 at 15 days after injection (in a pseudovirus assay).
  • Nevertheless, neutralizing titers, induced by either breakthrough infection or booster vaccination remain lower against presently circulating variants than against earlier omicron variants.
  • There is no data on T cell immunity yet.  

The expectation remains that the XBB.1.5 booster will temporarily limit the chances on infection in vaccinees, but the degree and duration of this relative protection against infection is uncertain, also in view of the further evolution of the virus.

There is a reasonable hope that the booster will prevent severe disease in the most vulnerable (older and/or co-morbid and/or immune suppressed subjects).  This important beneficial effect is probably not so much based on “neutralizing” activity, but more so on T cell mediated immunity and on antibodies activation natural killer cells (ADCC) and  phagocytes (ADCP), because the epitopes, presented on infected cells to antibodies and T cells, are less prone to mutation.  All this, however remains to be shown.   

Unfortunately, while Sotrovimab (S309) was the only mAb still active across many variants, its neutralizing activity against XBB.1.5 subvariants and especially against BA.2.86 is low.  Here also, it is possible (but not proven), that ADCC and ADCP may rescue mAb activity in vivo (as explained in Ep 348-6).  

Antivirals such as Remdesivir or Paxlovid are presumably fully active against all these variants, but this needs to be experimentally confirmed.   

I hope you enjoyed this complex story with some rather encouraging insights.  

Best wishes,