The bivalent booster probably won’t stop you from getting covid this winter
Covid might be mutating too fast for vaccines to keep up.
Covid vaccines worked great in 2021, with effectiveness of around 90% against infection.
Covid vaccines didn’t work as well in 2022. Vaccinated people still died at a lower rate than those who didn’t get the shot. But people still caught covid, whether they were vaccinated or not. Many people I know caught it twice this year: once with BA.1 and once with BA.5.
In 2021, my life felt almost normal. In 2022, I went back to wearing a mask.
There’s a new booster shot that’s been approved, a bivalent vaccine that combines mRNA for the original strain of covid and for the BA.5 strain.
Is this shot going to make life like 2021, again, where the shot protects you fully? Or is it going to be more like this year, where you still get sick, just not as severely?
When the new shot was approved, we didn’t really know. As many critics complained, the vaccine was approved with no new trial, just some animal testing — data from 8 mice.
That criticism wasn’t entirely fair, there had been human tests done on a very similar bivalent booster. It takes months to run a phase 3 trial, if we wait for those kinds of results, covid will always be ahead of us. But, we still don’t know much about how well this one will work against different variants.
There are two main reasons why it could fail:
The new shot could be less effective against BA.5 than hoped.
Or, whichever variant comes after BA.5 could already have mutated enough that it’s resistant to BA.5 antibodies.
New trial data
A recent experiment measured the response from different combinations of vaccines.
They compared volunteers who’d had 3 shots of the old vaccine, 4 shots of the old vaccine, 3 shots of the old plus one of the bivalent, or 3 shots of the old plus a breakthrough infection with BA.5.
A month after people got the shot, they took blood from each participant and tested their antibodies against various versions of the covid virus.
I found it a bit difficult to parse the graphic they presented:
So I turned this into a slightly more readable table:
These antibody titer numbers are a measure of immunity. A higher number is better, that’s a higher level of immunity.
I shaded the numbers for groups that have low immunity and are at more risk of catching covid.
Breakthrough infection with BA.5 brings the antibody levels up to around 4,500. One article says that you need levels about 1/3 that high to have immunity. So, somewhere around 1,500.
All the vaccinated people are immune to the old strains of covid.
People who caught BA.5 this summer won’t catch BA.5 again.
People who got the bivalent booster might be immune from BA.5, but they might not. They have pretty close to 1/3 of the antibodies of people that had a BA.5 infection, so it’s not clear. Maybe some would be protected, some would not.
The next strain that hits this winter won’t be BA.5.
BA.2.75.2 is one of the candidates for the next strain of covid.
It looks like no one has enough immunity to avoid infection with that.
The experiment suggests that:
- Everyone that got a bivalent booster can still catch the next variant.
- Everyone that caught BA.5 this summer can still catch the next variant.
- People that caught BA.1 and BA.5 could get covid a 3rd time, this year.
This is kind of depressing.
I think there are 2 main reasons why things went wrong.
Maybe it should have been a monovalent booster
One surprising finding from the experiment is that getting a bivalent booster shot gives almost the exact same response as getting 4 doses of the original vaccine.
Getting 3 shots and then a breakthrough infection of BA.5 gave a different reaction, it gave better protection against BA.5 than the bivalent booster did.
Bivalent means that the booster is designed to protect against two strains of covid, both the original strain and the new BA.5 strain.
Pfizer and Moderna started developing it in late 2021, when both Omicron and Delta were circulating, and people needed protection against both.
At some point in 2022, Omicron took over and Delta mostly disappeared.
It might have made sense, at that point, to switch to a shot that was purely for Omicron. But, I don’t know, maybe testing and production were already in motion, so they kept making it bivalent.
So, why did a BA.5 infection work differently than the bivalent shot?
Part of the problem might come from something called “Original Antigenic Sin”.
Your immune system has a memory of the pathogens it’s seen before. When it sees the same virus again, it can quickly create antibodies to fight it off. But, if it sees a virus that is similar to the old one, it might also create antibodies that are tailored to the old virus, not the new one. So the immune response will be quick but might not be very effective.
While the basic concept is clear, we’re still learning how exactly Original Antigenic Sin will work, with respect to covid infection and vaccination.
Some anti-vaxxers have used the Original Antigenic Sin concept to claim that people who got vaccinated for covid in 2021 will never be able to protect against future strains of covid. That has not proven to be true — the data above shows that you can still develop strong immunity to Omicron infection, even after 3 of the old vaccines.
For a more reasonable take, Shin Jie Yong wrote a piece speculating that you will be able to adapt to new covid strains, but maybe future vaccine updates won’t work as well, because of that immune memory.
We’ve moved past theorizing about this and started seeing that it is a real problem. With the new boosters, the problem might be that using both the old and the new strains triggers the immune memory of the old strain. Your body recognizes the spike proteins from the old virus, mounts an immune response against that, and mostly ignores the new one.
This might be why the antibodies produced by the bivalent booster look almost the same as just getting a 4th shot of the original vaccine.
There are other possible explanations — the hybrid immunity that comes from both infection and vaccination has sometimes been observed to be better than either vaccination or natural immunity alone.
Another theory is that the bivalent shots do actually work better than the old, but they work more slowly — your body first reacts to the old strain, and it doesn’t start making antibodies for BA.5 until after a month. The antibody studies were done 1 month after getting the booster. It’s still possible that the bivalent shots start to work better after 2 months.
Since immune memory might be the problem, the answer could be easy. We could just make a monovalent BA.5 booster, which doesn’t have any trace of the old virus in it. That might give a higher degree of protection, just like a BA.5 breakthrough infection does.
But even if we do that, there’s another problem: covid keeps mutating and we don’t know which strain is coming next.
Covid is mutating faster than we can update the vaccines.
We don’t know what the next strain will be for this winter.
There are a few contenders that are spreading. There’s BQ.1.1. There’s BA.2.75.2. There’s something called XBB.
For anyone who’s not a virologist, it’s a soup of random letters and numbers:
But there’s actually something interesting going on.
Every one of these strains is converging towards something similar. They’re all finding the same few mutations to escape immunity to previous infections.
Antivaxxers often blame the vaccines for driving covid’s evolution. But in this case, it has to be something else.
The old vaccines don’t really protect you from infection from any Omicron strain, whether that’s BA.1 or BA.5. Why would covid need to keep evolving when it can already infect vaccinated people?
Covid is probably evolving to escape the natural immunity people get after infection with Omicron’s BA.5 variant.
BA.2.75.2 is one of the new strains that escape immunity from BA.5, so the next strain should be something similar to that.
Experiments suggest it can reinfect people who caught BA.5 this summer.
This fast mutation is going to be a harder problem to deal with. Maybe we could run lab experiments to try to predict how covid will mutate and try to design vaccines that stay a few months ahead of it.
Or, if the manufacturing and approval process can be sped up even more, we could create a new vaccine as each new wave of covid hits.
Boosters should still limit disease severity
Vaccinated people were much less likely to die from covid, in the last 2 years:
People that got 4 shots did even better than people who got 2 or 3 shots:
When they did get sick with Omicron, people who got 4 shots were 12 times less likely to die than those who didn’t get vaccinated at all.
People who got 4 shots also did as much as 3 times better than those who got 2 or 3 shots.
But there was almost no difference between 2 shots or 3.
That might be because the vaccine does two things.
The first thing it does is it teaches your body how to react to covid, via B and T cells. That’s a long-lasting response that prevents severe disease. If you’ve ever been vaccinated, you have an advantage over someone who’s never been vaccinated or infected.
If your body’s reacted to a vaccine (or covid) before, you have an advantage over someone who hasn’t.
The second thing it does is it increases your antibody levels. That’s a short-term response that lasts about 6 months. While your antibody levels are high, you get an additional advantage where you’re less likely to catch covid and you’re less likely to die from it, even if you do get infected.
My guess is that 2 shots worked about as well as 3, because the people that got 3 shots got their booster back in 2021, already had waning antibody levels by the time they got sick, and no benefit over people who’d had only 2 shots.
The people that got the 4th shot in 2022 still had high antibody levels this summer. Less of them died during the BA.5 wave.
Getting another shot now would increase your antibody levels through the winter.
Who should get the booster?
If you’re at high risk from covid, and haven’t gotten boosted or infected in the last 6 months, you should still get the bivalent booster to limit the severity of covid.
But these results are a bit disappointing. You might still get sick this winter even if you get the shot.
You may still want to do other things to reduce your risk, like wearing a mask.
If you got sick with BA.5 within the last 6 months, the booster is unlikely to give you more protection than you already have.
Bad news for keeping covid under control
The bivalent boosters are still a good tool for people to reduce the severity of covid. But there will probably be a lot of breakthrough infections. The results are less impressive than what I was hoping for.
That’s bad news for covid this winter, a lot of people are going to get infected with covid. Many of them will get reinfected.
It’s also bad news for public confidence.
Most people in the US are already not getting booster shots. Seeing that vaccinated people are getting sick might further reduce confidence in the shots.
Going forward, we need to consider other options. Monovalent booster shots might work better than bivalent. But we also need to put a lot more funding into research for other options, like a nasal vaccine or a universal coronavirus vaccine.
We have to accept that current vaccines alone aren’t going to stop covid. There needs to be as much effort into getting people drugs to treat covid. It’s worth improving the ventilation of public spaces. And, at a personal level, it might still be worth it to be cautious or to wear a mask.
