COVID Transmissions for 9-6-2021
Correlates of protection might be causatives of protection--but only maybe
Greetings from an undisclosed location in my apartment. Welcome to COVID Transmissions.
It has been 657 days since the first documented human case of COVID-19. In 657, Tang Chinese forces made the Western Turkic Khaganate into a protectorate. Turkic polities in Central Asia form a major buffer for Tang and later Chinese dynasties against outside forces, and in many key ways, the interaction between Chinese leadership and the peoples of Central Asia ends up having a deep impact on overall global affairs for centuries.
Speaking of protectorates, I want to continue to discuss correlates of protection today, because this is in my opinion one of the key scientific issues in the pandemic as it currently stands.
As a reminder, this week we are running on a holiday schedule, which you will have realized from the fact that I released this issue at 6 PM today rather than on its usual time. This issue was intended to come out tomorrow morning, as a matter of automatic scheduling. I decided instead to send it out early so that readers who are, like me, celebrating Rosh Hashana, get a chance to read it before the holiday.
The next issue will come out either Thursday or Friday, depending on my availability, and there will only be two editions of COVID Transmissions this week.
Bolded terms are linked to the running newsletter glossary.
Keep COVID Transmissions growing by sharing it! Share the newsletter, not the virus. I love talking about science and explaining important concepts in human health, but I rely on all of you to grow the audience for this, which you can do by using this button here:
Now, let’s talk COVID.
Correlates of protection, again
If you’ve been reading the last several issues consistently, you’ll have heard a lot about correlates of protection. The bottom line is that we need a sense of correlates of protection against COVID-19 so that we can reliably test individuals in order to know whether or not they are protected. It is a game changer to be able to assess protection against disease by a blood test.
In fact, Florian Krammer, a friend and former colleague, wrote about what correlates of protection can do—and what they can’t do—in a recent issue of Nature Medicine, which I think everyone should read: https://www.nature.com/articles/s41591-021-01432-4
Florian points out the value, and also the limitations, of certain correlates of protection. He also draws a key distinction around “non-mechanistic” correlates of protection. These are things that may correlate with protection, but that are not actually involved in the mechanism of protection. As an analogy, the number of houses owned by US military personnel in a given town may correlate with how well-defended that town is, but it’s not the houses themselves that are responsible for the defense. It’s probably the nearby military base, with equipment and soldiers. In some cases, biologically speaking, there may be a convenient measurement we can take that is correlated with protection, but isn’t actually the cause of that protection. Correlation is, as always, not causation.
However, Florian also points out that neutralizing antibodies are indeed often both correlated with, and the cause of, meaningful disease protection. We don’t know for certain whether that’s true in COVID-19 yet, but it does look increasingly likely.
In fact, in the same issue of Nature Medicine, we see a study that collected a broad base on previously-published results to explore whether antibody titers in humans could serve as a meaningful correlate of protection: https://www.nature.com/articles/s41591-021-01377-8
This paper is a bit less approachable than Florian’s commentary piece, but it is also very meaningful. The authors, importantly, were very clear about what “protection” meant, and asked questions about specific types of protection. For protection against any infection, they concluded that a person must have neutralizing antibody titers that are at least 20% the level seen in the average sample taken from a recovered patient. They were reasonably confident in this estimate too, with a 95% confidence interval from about 14% to about 28%. And that’s for protection from infection at all. For protection from severe disease, the neutralizing antibody titer needed was estimated at 3% that of the average recovered-patient sample, with a 95% CI from about 0.7% to 13%.
Looking at these numbers, I think that if the results can be confirmed in primary research, it’s looking like a neutralizing antibody titer that is more than 15% the average level seen in a recovered patient would be sufficient to give strong protection to most people. It would be above the confidence interval for severe disease and within the confidence interval for protection against infection at all.
For the purposes of disease control, the critical threshold would be more like 30%, since that would confidently place us into the realm of preventing infection altogether.
This isn’t the end of the story, though. This is a research synthesis, and requires additional investigation. Also, the benchmark here was made as a percentage of the average level seen in a recovered patient, something that was probably done to make it possible to cross-compare studies that were done in very different settings. The problem is, there is a lot of variation in the antibody levels in recovered patients. That’s one of the reasons I think that vaccination is the better choice compared to getting COVID-19; the antibody level is more consistent with vaccination. That and the fact that COVID-19 can kill you, and the point of this whole exercise is to avoid getting it. Seeking protection from COVID-19 by getting sick with COVID-19 makes about as much sense as driving off a cliff to avoid a future car accident.
Anyway, what we really need here is a threshold value for the correlate of protection that is not relative to something else, but is instead a strict antibody level that is based on a reliable, reproducible measurement. This paper is a step in the right direction, however, and it does leave me feeling confidence that we will be able to, one day, do a blood test to determine whether or not someone is protected.
Once we can do that, large studies can be done to optimize the vaccination program without anyone needing to get COVID-19. Instead, people can be vaccinated and their antibody responses can be measured against that threshold. We can try out a lot of different vaccination strategies and eventually determine the best one available. What a great thing that will be!
What am I doing to cope with the pandemic? This:
Biking to Queens
For reasons that will be shared here Real Soon Now, I biked across the East River on Sunday to visit a friend in Queens and work on a project together. This was my first time biking across a major NYC bridge, specifically the “Ed Koch/Queensboro Bridge,” or as most people from NYC call it in my experience, the 59th Street Bridge. It’s pretty groovy. There’s a whole song.
The experience wasn’t wildly different from biking anywhere else, except that the view was interesting, but I share it here because it’s an example of how bikes can really get you most places that public transit could too. For me, this trip would have been about 45 minutes on public transit and it was 55 minutes on a bike. Instead of being crammed into a crowded bus or train, I was getting exercise and not getting exposed to COVID-19.
I really don’t know if I’ll ever feel like going back to being a habitual subway rider.
I owe a number of you email replies to personal messages. Rest assured I am not ignoring you, I am just extremely busy until after Rosh Hashana.
You might have some questions or comments! Send them in. As several folks have figured out, you can also email me if you have a comment that you don’t want to share with the whole group.
Join the conversation, and what you say will impact what I talk about in the next issue.
Also, let me know any other thoughts you might have about the newsletter. I’d like to make sure you’re getting what you want out of this.
Part of science is identifying and correcting errors. If you find a mistake, please tell me about it.
Though I can’t correct the emailed version after it has been sent, I do update the online post of the newsletter every time a mistake is brought to my attention.
No corrections since last issue.
See you all next time. And don’t forget to share the newsletter if you liked it.
Always,
JS