Booster Shots Already: Why?

There are five fascinating things about our choice to use booster shots. I will try to explain each, but unfortunately, some introduction is necessary first.

COVID Vaccines: How Good are They Really?

Global governments have bet the farm on vaccination eradicating (remember when we used to talk about "herd immunity"?) or greatly prolonging the emergence of SARS-CoV-2 as a widespread pathogen to such a great extent that they're deploying vaccine passports for routine, daily activities, demonstrably not a great idea almost regardless what you want to achieve.

It was immediately apparent that SARS-CoV-2 was very contagious. Later studies found it to have had a base initial R0 of around 5.7, compared to around 1.3 for the average influenza strain. That meant that achieving an effective vaccine was always going to be extremely hard: a 40% effective flu vaccine could bring Reff close to or below 1, but a 40% effective SARS-CoV-2 vaccine would come nowhere close.

The vaccines that we created achieved fairly spectacular protection at first, but these numbers were a deceptive snapshot: they were compiled very early in the pandemic, before SARS-CoV-2 had a chance to show us its moves. mRNA vaccines generally hold great promise; the problem here is not the platform, but the target virus.

Because of the high R0, slow vaccine production and distribution, and rapid viral evolution, vaccines should have been anticipated to fail, and that prediction wasn't hard. Treatments (like monoclonal antibodies) and non-pharmaceutical interventions were always the best path.

Evolution: The Virus Fights Back

At first, most experts thought that SARS-CoV-2 evolved fairly slowly. It has since become apparent that it evolves very quickly. This evolutionary speed could only have become visible after the pandemic had run for awhile, but we had enough hints from early studies that we should have anticipated it.

This happened because SARS-CoV-2 relies on recombination, and recombination wasn't directly visible when the pool of haplotypes was very limited, because there were few different mutations to combine. The pace of evolution thus initially appeared to be mostly guided by just SNP's, which is relatively slow in SARS-CoV-2's case. But this proved enormously deceptive.

As soon as mutations began to pile up, recombination could begin to happen in earnest. The more mutations there are, the more recombination can happen for two reasons: there are more alleles to choose from overall, and the genotypes become less likely to overlap, allowing for easier co-infection. Thus, evolution and emergence of new strains will accelerate from here on out, modulo global viral loads, which are highly unlikely to drop far enough with multiple viable serotypes already circulating widely and weakening vaccination.

We've long known that coronaviruses can recombine and use that ability to evolve very quickly, and we should have recognized that SARS-CoV-2 utilizes recombination heavily once we knew how error-prone its RdRp is. We never developed vaccines against coronaviruses that cause colds for a reason: we couldn't. We also didn't really need to. Common human coronaviruses are rarely dangerous.

Sarbecoviruses are Different

Unlike the four common alpha and beta coronaviruses endemic in humans, sarbecoviruses like SARS and MERS are extremely pathogenic in humans. SARS-CoV-2 is actually on the light side of pathogenicity compared to them. What does this imply for the future pathogenicity of SARS-CoV-2? It's anyone's guess at this point.

When 229E, an alphacoronavirus, emerged in the pandemic of 1889-1890, it caused many fatalities, but the CFR was already lower than SARS-CoV-2's was at the outset and it has become less virulent since then.

Viruses find an optimal level of pathogenicity for transmission: it's virtually always bad to kill your host, but disabling them -- and particularly their immune systems -- is generally beneficial. This level will depend on many other factors of the virus, and as SARS-CoV-2 is a sarbecovirus, other sarbecoviruses are closer genetically and a better guide than the other coronaviruses. However, no other sarbecovirus has seen prolonged transmission in humans, so comparison with MERS and SARS must come with a big asterisk.

One troubling indicator is that each successive globally dominant strain of SARS-CoV-2 has been more dangerous than its predecessor. We do not and cannot know yet whether this trend will continue, stabilize, or reverse. Thus, there is enough room for concern that SARS-CoV-2 will remain dangerous or become even more dangerous, which put governments in a panic.

A Lack of Caution

This understandable panic led to the immediate development of vaccines and their widespread deployment despite known unknown long-term risks. Specifically, ADE has been demonstrated in SARS-CoV-2, but we don't know the extent to which it contributes to pathogenesis. ADE could evolve to be more severe.

Furthermore, we still do not know how much OAS will be a factor. It's often a critical issue with pathogens that rapidly antigenically drift, like SARS-CoV-2, so as soon as we learned of its evolutionary pace, we should have been much more cautious.

Whether deliberate or not, there is a fortunate safety valve in current vaccines: they are not as immunogenic as they could have been. Determining proper dosage of medication is always a challenge as trade-offs are made between intended effects and side effects, and in this case, we appear to have room to increase the immunogenicity of the current vaccines. That is the best underlying support for booster shots, as increased immunogenicity can compensate for antigenic drift. There will be limits as side effects become more severe, however.

Fascinating Things about our Booster Shots

So, the overall goal here is really just to boost the vaccination course's overall immunogenicity. Will that work? Probably for awhile. But there are key nuances:

1. We are using the original vaccines rather than modified versions with a different S protein, even though we could design, manufacture, and approve a vaccine targeted at a different S, say, Delta's. Why is that? The most likely explanation by far is that the existing vaccines lie at the base of the phylogenetic tree and thus have a higher probability of cross-reactivity with more strains, and the authorities are aware of this, but it could also involve OAS.
2. Booster shots will probably only buy time. It is unlikely that the durability of immunity from B cells will be increased enough to compensate for the rapid antigenic drift. Additional T cells should be more durable, so booster shots should help more against severe disease than against infection and mutation, but as evolution continues unabated, this too will fade(e.g. through mutations like S:P272L).
3. We are doubling down on OAS and ADE not being problems. Not only would they be more severe, but we would also lose our room to maneuver around them conferred by the lower immunogenicity.
4. We have chosen to use our limited supply of vaccines for booster shots rather than as a primary course for underserved countries, much to the consternation of the WHO. Besides the human toll on these countries, they are also sources of dangerous variants.
5. We feel that the protection that will remain by winter will be so insufficient domestically that it is worth this miserable trade-off. This is proven by our timing of booster administration. If we wanted to immediately boost protection levels, we could have at any time. We knew durability ebbed within several months and things looked bad in Israel. We had doses on hand that other people didn't want. Instead, we strategically chose awhile ago to delay announcement of a booster until it will have peak efficacy for the colder months.

But will booster shots work indefinitely? Almost certainly not: only if global viral load is reduced fast enough through other means or the number of viable and dangerous haplotypes/serotypes is sufficiently limited. I believe it likely that we are facing a virus that evolves and replicates well enough that vaccination will ultimately fail. It will have bought us time to work on treatments, but it will also have bought the virus time to evolve, created the conditions for ADE and OAS, and induced immunoevasive selective pressures.