By now we can be quite sure that BA.2.86, the new highly mutated version of SARS2, will cause another big global wave of infections. It has been spotted in at least eight countries by now and looks basically identical in all of them, which means it has already rapidly started spreading. If it was bad at spreading itself and nobody had simply noticed it, you would expect to see substantial differences between the genomes.
Of the nine sequences uploaded so far, one was a hospitalized patient. And I expect many more to follow, for a number of reasons. To start with, the antibody mismatch simply reduces the protection that people already have built up. The number of Spike mutations is comparable to the Delta -> Omicron shift. In other words, the “wall of antibodies” should now once again be pretty useless.
Second, in contrast to Omicron that probably jumped from rodents back into our species, this version would seem to have evolved in a chronic infection. The chronic infections tend to favor the evolution of increased virulence, because the easiest way to avoid someone’s antibodies for this virus is to spread from cell to cell, prohibiting these antibodies from binding the virus:
Both fusogenicity and tunneling nanotube formation are a form of immune evasion, as they shield viral particles from the impact of nAbs during cell-to-cell spread (152). However, this trait was subsequently lost by the Omicron BA.1 variant (154,155) and its descendants. As properties such as syncytia formation are likely to be under drift, it cannot be ruled out that they will emerge again by chance. Other changes to viral pathogenesis (for example changes in tropism) can be expected to be gained and lost from variant to variant, under conditions of drift. It is important to note that such changes in viral behavior may have a strong impact on virulence,
despite being difficult to anticipate before the fact or predict from in vitro assays.
In BA.2.86 you see such mutations linked to increased fusogenicity. We already know that increased fusogenicity results in higher virulence. It also has the mutation in the Furin cleavage site that seems responsible for Delta’s higher virulence, (681R). Finally, it also has L452W in Spike. L452R is what’s seen in Delta and is part of what helps Delta increase its fusogenicity (and thus virulence) over Omicron. BA.2.86 turns it into Tryptophan, that hasn’t been seen before. But considering it’s a double nucleotide mutation, it’s likely to offer a significant survival advantage somehow.
Most people seem to think Omicron emerged from a chronic infection, but I don’t. If Omicron came from a chronic infection, you would have to wonder why fusogenicity decreased from Delta to Omicron. Fusogenicity allows spreading from cell to cell by merging cells, unaffected by antibodies. Why would a chronic infection lose this ability? This seems illogical to me. BA.2.86 looks far more like a chronic infection: It shows signs of improved fusogenicity compared to its ancestor.
And interestingly enough, we also begin to see some evidence of what van den Bossche has been warning about: A mutation to Threonine right in the Receptor Binding Domain, which enables a sugar molecule to bind that then effectively prohibits antibodies from binding.
What seems to have happened here is that someone became infected, who lacked the capacity to destroy infected cells (no cytotoxic lymphocytes). As a result, the antibodies had to do the job of keeping this virus under control by themselves. So the virus would have hidden in the cells, learning to spread from cell to cell. And in the process it would have changed its Spike so much, that eventually it stumbled upon a version of itself against which the antibodies didn’t even work anymore. And then it could spread from its host to other hosts again. This would explain why only Spike has significantly mutated.
There is however another aspect that needs to be considered. It’s very unusual, for a new variant to emerge that is so different from other variants, but only in its Spike protein. Because the IgG4 antibodies induced by mRNA vaccination are monovalent and don’t induce cross-links on the Spike protein, they are relatively simple to escape through mutation.
It’s really weird to see such a long list of double-nucleotide mutations that enable antibody evasion. It suggests the virus would have been under immense pressure to escape the IgG4 antibodies. The failure of cytotoxic lymphocytes to destroy infected cells may have been caused by the chronic infection itself, as it’s known to reduce CD8+ T cell counts and chronic infections cause CD8+ T cell exhaustion.
I want to return to van den Bossche’s prediction for a moment.
What van den Bossche predicts is that variants will eventually emerge, that have glycosylation patterns in the NTD that make the infection enhancing antibodies obsolete. These antibodies pull open the Spike protein and thereby make it easier to infect cells, but they prohibit the virus from becoming attached to dendritic cells. The dendritic cells tend to transport such viral particles to the lower lungs and thus these antibodies have the simultaneous effect of increasing vulnerability to infection, while decreasing the severity of infection. This is argued here.
Does BA.2.86 fit the definition of such a variant, against which these antibodies no longer work? Well returning to the presentation, the NTD supersite, where the antibodies bind, has a whole bunch of antibody escaping mutations. There is even a weird insertion of four amino acids, which tends to be very rare and could have the same effect by obscuring immunogenic regions through structural changes.
It seems likely that these mutations prohibit the polyreactive non-neutralizing antibodies from binding. The immune system would need to develop new antibodies against these regions, but this runs into problems as the new NTD will probably be less immunogenic.
The general pattern I have illustrated for XBB will just continue: Spike will evolve to become less immunogenic. The IgG antibodies our B cells stumble upon through somatic hypermutation that do react with the new BA.2.86 Spike, will be cross-reactive: They will increasingly also bind to self.
The problem you run into with these IgG antibodies that also bind to our own proteins is that they interfere in your body’s own functioning. IgG1 and IgG3 antibodies would cause autoimmune problems, by causing your immune system to attack healthy cells. If you deploy IgG4 antibodies cross-reactive with self, then that is problematic in situations where you do need your immune system to destroy your own cells.
Finally, as a descendant of a persistent infection, you would expect BA.2.86 to be better at establishing persistent infections. This variant appears to be the descendant of an infection that would seem to have continued for over a year.
In a chronic infection, a virus can develop mutations that help it to survive in your body, at the cost of its ability to spread through the air, pass through the mucous membranes and infect another person.
BA.2.86 doesn’t have a lot of new threonine and serine mutations that would enable complete antibody evasion. However, it may be very good at establishing persistent infections, that eventually lead to such mutations that cover the Spike protein in sugar molecules and make the antibodies powerless.
However, as I have explained before, there are many different routes we can think of for a virus to survive our neutralizing antibody response. Glycosylation is just one route to worry about. With BA.2.86 it’s still very early, but we already see evidence of one such route: Improved fusogenicity. This is likely to result in higher intrinsic virulence than other Omicron variants.
The bigger problem I think we’re faced with is as following: There’s ultimately nothing that requires a new variant to force the dominant variant into extinction. Variants that are sufficiently different can just co-exist. And then over time, they can adjust themselves, to make use of the immunological conditions produced by the other variant, as observed for Dengue. This is the serotype scenario I have mentioned before. Have a look at Australia:
This is XBC.1.6* In Australia, a Delta-BA.2 hybrid. You can see it’s just not being forced into extinction by the variants that are dominant elsewhere.
Here you can see the Spike mutations. XBC.1.6 has 14 mutations EG.5.1 doesn’t have, EG.5.1 has 17 (the XBB family as a whole has 15) that XBC.1.6 doesn’t have. They’re different enough that neither is able to wipe the other out in Australia. Instead they exist side by side, which suggests they’ll now evolve to stop interfering with each other.
This is different from when Omicron first emerged. Omicron was so good at spreading itself, that it practically wiped out Delta. Delta got no real chance to adopt, except in the failed zero COVID utopia Australia, where it gave birth to XBC.1.6.
But that seems much less likely to happen now. Instead, it looks like you’re going to have multiple very divergent families circulating together: The XBB family (of which EG.5.1 is part), the BA.2.86 family and the XBC (Delta-Omicron hybrid) family.
None of this really had to happen. It happened because of a number of mistakes. To start with, there are the failed zero COVID utopias, Australia, New Zealand and China. These managed to keep SARS2 out for over a year, until they faced more infectious variants. Then they lost and gave those variants the chance to adapt in a population with very little immunity.
That’s how Australia bred the very fit XBC.1.6 variants. XBC* went extinct in the Philippines, where it first emerged. It’s only in the failed zero COVID utopia Australia, which never had a big Delta wave, where it can train to become a globally co-circulating variant, a separate serotype.
Second, there is the failed vaccine experiment. It looks unlikely that Delta could have survived in a population where everyone had caught Delta. But then a new version of COVID emerged, Omicron, that had changed its Spike protein. And because everyone had been vaccinated against Spike, the Omicron versions could spread themselves very rapidly. By now it is clear that the population has such a strong original antigenic sin effect, that XBB infection doesn’t induce a neutralizing antibody response against XBB. The XBB family can now just continually reinfect people.
It recently became clear that infection by Omicron causes a massively increased risk of reinfection in the short term by a different variant. But until now, the circulating Omicron variants were generally pretty similar.
But now have a look at this:
The two main variants now have a massive 42 mutation difference between them in Spike. This is how you end up with different simultaneously circulating serotypes, where one serotype infects you just as you were busy recovering from the other serotype.
And that’s why we now get even the usual suspects to admit that the vaccine strategy isn’t working. Even Eric Topol now declares that we need a “variant independent vaccine”:
Eric argues that “a variant-chasing strategy doesn’t work”. Well congratulations for figuring that out. It would have been really useful if you could have mentioned this BEFORE YOU ENDORSED INJECTING EVERYONE WITH A VARIANT-CHASING VACCINE.
Because what happened as a result, is that the whole population became stuck with a variant-chasing adaptive immune response, churning out antibodies against the specific part of the virus that mutates most rapidly (Spike). And the population continues deploying basically that same antibody response, even after the virus has radically changed itself. With the birth of XBB*, it has learned how to use the Wuhan-fixated vaccine induced immune response to its own advantage.
A variant-proof immune response is what the body would normally deploy, if we had not tried vaccinating people against Spike. What does that look like?
It means you deploy an affinity maturated IgM antibody response, just like you do for influenza (in the absence of vaccination with inactivated vaccines). IgM is a very big antibody, with multiple branches. This means that it is extremely difficult for the virus to mutate around it. But if your body deploys IgG due to the vaccines, that means IgM never really gets a chance to learn to fight this virus (affinity maturation). Despite not binding as strong at any particular location as IgG, IgM’s massive size compared to IgG makes it very useful when it does bind: It can bend the whole Spike protein out of shape and obstruct access to receptors.
For the IgG we deploy due to vaccination this is much easier, especially with the monovalent IgG4 that people now deploy after mRNA vaccination forced a shift from IgG3 to IgG4. In simple English: The two arms of an IgG4 molecule are identical to each other, meaning that it’s much easier to mutate around than regular IgG antibodies, which in turn are easier to mutate around than IgM antibodies.
Second, it means that you body deploys Natural Killer cells, that take up residence in your tissues where this virus is encountered. These Natural Killer cells have receptors that allow them to recognize whether a cell has been infected by viruses of this nature, because its receptors resemble the sort of receptors the Spike protein uses to infect cells. The virus can’t really mutate around this. These Natural Killer cells can then either:
- Kill the infected cells.
- Release Interferon Gamma, a molecule that helps the cell get rid of any viral material infecting it.
There are various other elements of the innate immune response too. Susceptible cells get better over time at learning to recognize viral threats. Even your infected tissues themselves get better at coping with infection, through vascular and lymphatic remodeling.
This is what a “variant-proof” strategy looks like.
And these morons should have known this.
If you’re sitting there in August 2023, watching as three families circulate simultaneously, unaffected by the immunity induced by their distant cousins, utilizing the original fixation against Wuhan to their own advantage and declare that “well guys, I think we need a variant-independent vaccine”, you’re a moron. It means you’re incompetent, or worse, dishonest.
“We need a variant-independent vaccine” is not some sort of clever insight. It’s like looking at the energy crisis in Europe today and saying “we need some sort of mysterious glowing blue box that generates cheap and abundant energy”. A clever insight would have been to warn about the risk of depending on Russian energy, before we signed up for it.
They should have figured this out in 2020. That’s their job. Their job is to use their insight, their knowledge, to warn the politicians and the general public against the risk of easy technofixes that may appear to work at first but make everything much worse in the long run. But that job ended up falling in the lap of anonymous dudes with obscure blogs.