Is Omicron A “Vaccine-Escape” Variant?

Derek Thompson at The Atlantic has a nice roundup of the latest information on Omicron’s spread. Among the links that he’s gathered are this report which mentions the fact that neutralizing antibodies struggle to restrain the variant, but also this preprint, which analyzes the mutations characterizing the variant and concludes that “SARS-CoV-2 has not evolved extensive T-cell escape mutations at this time.” Moreover, in this Twitter thread, Mia Malan summarizes data from South Africa showing that while two doses of Pfizer vaccine (hardly anyone in South Africa has had a booster shot at this point) are only 33% effective at preventing infection, they are 70% effective at preventing hospital admission. So all of that is consistent with the Type 3 scenario: antibodies struggling to keep up with viral growth, but killer T-cells mopping up the infection as soon as the immune system can mobilize them.

The Immunosuppressed Patient Theory, And Its Implications

But there’s more. As discussed in this preprint (a summary is here), Omicron has a huge number of mutations compared to all other known variants, and is not part of the SARS-CoV-2 lineage that includes Delta, the presently-dominant strain (I am indebted to Jim Olds (GMU) for pointing this out and educating me on what it means). This is extremely significant, because it lends support to the theory that the variant developed in an immunosuppressed patient. Normally a new variant emerges with mutations that produce a much smaller divergence from its parent lineage, because each mutation actually bears a greater risk of killing the virus than it does a chance of improving the virus’ fitness. It’s as if the virus particles are crowded into a “mutation casino” full of one-armed bandits, and occasionally pulling the levers: most of the time the player loses, and he usually goes bankrupt if he plays long enough, but once in a while there is a big payoff and you get something like Delta: a small set of mutations that create a big advantage in reproduction rate. Then Delta takes over the casino, and the genome drifts slightly. To get the high level of mutations observed in Omicron, you need a long-term reservoir where the virus can slowly and endemically evolve to its radically-changed new state, adapting gradually to the local conditions in its host. The “immunosuppressed patient” hypothesis is attractive because (a) such patients make ideal reservoirs, as their immune systems pose no threat to the virus, which therefore has plenty of time to evolve to suit itself; (b) Omicron probably emerged in South Africa, and (c) South Africa, sadly, has plenty of such hosts, due to the AIDS epidemic.

There are some interesting implications of the “immunosuppressed patient” theory that have not really received enough attention:

1. A SARS-CoV-2 virus that adapted to survive in such a patient long enough to develop the vast constellation of mutations that festoon Omicron would necessarily have had to figure out how not to kill its host. If it had not done so, it would never had had time to develop all those mutations. That is to say, some of those mutations must necessarily code for lower virulence (clinical-speak for “producing less severe disease”);
2. A SARS-CoV-2 virus that came of age in such a patient has never experienced a fully-militarized human immune system. There is no evolutionary pressure whatever for any of those mutations to code for “hide from killer T-cells”, since killer T-cells adapted to SARS-CoV-2 are a non-issue in an immunosuppressed organism. Emerging from the host to infect new individuals with functional, vaccinated immune systems that know all about SARS-CoV-2, the virus would be in for a nasty shock.

So the “immunosuppressed patient” theory, which is suggested by the highly-mutated form of Omicron, in fact also predicts both lower virulence and lack of a vaccine-escape capability. Lower virulence is basically established for vaccinated patients; also, this Twitter thread summarizes research showing that the variant has diminished tendency to grow in the lower lung tissue, making it is less dangerous, basically because it “appears to prefer to infect bronchial epithelial cells instead of alveolar epithelial cells” (Gordon Pusch pulled this out of the paper on an Argonne Slack channel, thanks Gordon). And lack of vaccine escape capability dovetails nicely with the evidence for “Type 3-ness” of Omicron’s breakthrough infections laid out above.

In other words, a lot of this picture really hangs together well, for something that’s been patched together in crisis mode during the course of a little over a month. That doesn’t make the picture rock-solid, but in my opinion it does make it worth placing some bets on.

An Exit, Or At Least A Reprieve?

What’s the bottom line here then? I have actually been optimistic about Omicron representing the exit ramp from the COVID-19 pandemic since the first reports from South Africa became intelligible, to an extent that I think may have shocked some of my colleagues. If I’m correct about the above (yes, it’s very speculative and much of it could be wrong) I think that it is entirely possible that Omicron has the perfect set of characteristics to tame this epidemic:

1. It causes much less severe disease than Delta, the dominant-prevalence variant until Omicron showed up;
2. It knows nothing of human immunity, and killer T-cells primed by COVID-19 vaccines can wipe it out easily;
3. It transmits 5-6 times faster than Delta, and is in the process of kicking Delta off the planet.

That last point is significant. The dynamics of newer, fast-transmitting SARS-CoV-2 variants are apparently such that they entirely displace previous variants. Check out the per-country plots at the excellent covariants.org site: when Alpha, or Delta showed up, the only variants that survived were other slight variations of the same strain (think “blue-eyed instead of green-eyed Delta”) which transmitted at about the same rate. The previous strains did not remain at some lower prevalence: they just disappeared from the genome so far as sequencing of specimens could ascertain. Some certainly survived in reservoirs (as Omicron’s parent lineage did), but the community transmission phase was over for those variants. This is a bit weird, and needs more of an explanation than “the new variant just reproduces faster”, for reasons that I’ll write about in a few days. The main point, however, is that given these dynamics it seems very likely that Delta—a much more virulent variant than Omicron—is doomed, and soon. At current rates, it should be gone before the end of January. And good riddance.

And so perhaps we’ll be left with a version of SARS-CoV-2 that causes a nuisance cold in people who had the sense to get a vaccine (where available), and which current vaccines can handle well.This is not to say that it couldn’t evolve again, and that it won’t be necessary to tune new vaccines, or re-re-boost, or some such thing. And we still have a big job rolling out enough vaccines for the rest of the world, both as a moral duty and to try to shut down the mutation casino before the next big fitness payoff. But we may catch a bit of a break on this pandemic, at least for a while. There’s been so little good news lately, I’d gladly take even a temporary reprieve.

And, who knows, maybe it is an exit. Wouldn’t that be nice?