Thread by @TRyanGregory

We still don't know what it will mean to have a cloud of immune-escaping variants at the same time. There are several possibilities, but for now it's looking like we're still going to be talking about what individual variants are doing rather than the cloud per se. 🧵

Here's what we mean by a "cloud" of variants. There are multiple lineages, all derived from either BA.2 or BA.5, which have independently evolved to have a similar set mutations that confer immune escape.

(Image by @dfocosi)

As I have noted, this is new.

https://twitter.com/TRyanGregory/status/1577617059525140480?t=Ijt6FT3NwuCVknsCIKmLJA&s=19

Rather than one new, deeply divergent new variant (i.e., not descended from the most recently dominant variant, but instead sharing a more distant ancestor with it), we have many variants, all recently evolved and relatively closely related.

These are all still called "Omicron", but they are very different from each other. Many have argued that @WHO should be assigning new Greek letters to several of these variants.

Here's a summary of how naming works, in case you find it confusing:

https://twitter.com/TRyanGregory/status/1576945959824523264?s=20&t=I7F6ss5Q99svYE0n3w8WWQ

Here are the names that @WHO has assigned, shown in context. It's very clear from this that naming, say, "Epsilon", "Iota", and "Kappa" as different variants but everything within "Omicron" as a subvariant is... a choice.

Incidentally, in phylogenetic terms, BA.5 is nested within "BA.2". In technical terms, this makes "BA.2" paraphyletic (and thus not a clade, which by definition must be monophyletic). Lots more about reading evolutionary trees here:

evolution-outreach.biomedcentral.com/articles/10.10…

Back to the cloud (or swarm or soup or cluster (or "clovar" if you like my portmanteau of "cloud" and "variants").

Here's what we have now:

* Multiple variants (all descended from BA.2 or BA.5) evolving in different parts of the world and then circulating at the same time.

* Independent evolution ("convergence" or "homoplasy") of a similar set of mutations that confer immune escape and thus are advantageous in the current environment (i.e., human host immunity as it is now).

* Variants that differ in their combination of immune escape (host antibodies can't bind as well to altered spike protein) and ACE2 binding affinity (ability of the virus to latch onto and enter cells).

(Image by @RajlabN: public.tableau.com/app/profile/ra…)

That is to say, though the variants in the soup have some important mutations in common, they are not identical. Some are slightly better at evading immunity, others are slightly better at infecting cells.

Consider, for example, XBB.1 (Hippogryph) and BQ.1.1 (Cerberus). XBB.1 has higher immune escape (the highest seen in any variant to date) whereas BQ.1.1 has higher ACE2 binding affinity (not nearly the highest ever, though).

As I see it, there are several possibilities for how this will play out over the next few months:

1) One variant will eventually become dominant everywhere.

2) One variant will become dominant in one places, another variant will become dominant elsewhere (then maybe they swap).

3) The same subset of variants in the cloud will become dominant everywhere (and no single dominant variant will take over by itself).

4) Different subsets of variants will become dominant in different places (then maybe they swap).

5) We have turnover in variants within the soup as new ones evolve and older ones disappear, with high numbers of infections but no one variant or subset of variants ever reaching a very high frequency.

There's also the question of what happens in terms of clouds of variants from here on. Again, there are several possibilities:

1) Clouds come and go in waves. All of the current variants will disappear and a new cloud composed of different variants will arrive.

2) Clouds persist without breaks in between, but the membership of the cloud changes. A bit of a "Ship of Theseus" situation (when does it count as a new cloud if all the component variants are being replaced?).

3) This cloud ends and then we don't get clouds like this anymore. Perhaps a new, deeply divergent lineage (not within "Omicron", but something @WHO would call Pi or Rho) arises instead. Could happen if older variants are out there evolving in someone with a persistent infection.

Making predictions about what will happen this winter is difficult. Some things we don't know:

1) Can someone be infected with more than one of the variants in the cloud at the same time?

2) Can someone be infected with one variant and then another in rapid succession?

3) Do the variants differ in acute symptoms or severity?

4) Do the variants contribute differently to the risk and/or effects of long COVID?

5) Will different variants infect different subsets of the population based on their immunity? For example (strictly hypothetically), might XBB infect people who had a Delta infection, while BQ.1.1 infects people who had a BA.1 infection?

So, what's happening so far?

1) We've seen a surge of infections and hospitalizations in Europe, and one starting in Canada as well. Most of this has been a resurgence of the BA.5 variant that caused the summer wave. The impacts of the variant soup haven't been felt yet.

2) Individual variants (and their immediate descendants) are taking off in different places. BQ.1 (Typhon) and BQ.1.1 (Cerberus) are rising quickly in Europe and North America, whereas XBB (Gryphon) and XBB.1 (Hippogryph) are surging in Singapore.

For an excellent summary, see this by @CorneliusRoemer:

github.com/neherlab/SARS-…

Things could change, though. XBB has showed up in North America now, for example. Will it outcompete the BQs? Vice-versa? Both increase? Both plateau without becoming dominant?

We don't know yet.

As always, I will note that high quality, well fit, consistently worn, N95 or better respirators, air filtration, and ventilation are variant-proof. Bivalent boosters don't contain XBB or BQs (they're either wild type + BQ.1 or WT + BQ.5), but are worth getting.

And, again, the way to slow the evolution of new variants is to limit the number of viruses out there mutating and being sorted by the process of natural selection. That is, mitigate transmission.

Huge gratitude to the folks on this list who make all of this information available everyday, on a voluntary basis, so that we can stay informed.

twitter.com/i/lists/156947…

/end