COVID Transmissions for 9-1-2021

Does it matter what your antibody level is?

Sep 01, 2021

Greetings from an undisclosed location in my apartment. Welcome to COVID Transmissions.

It has been 652 days since the first documented human case of COVID-19. In 652, an Arab fleet handed the Byzantine Navy a crushing defeat, turning back 500 Byzantine ships. The Byzantine Empire had to this point been the unquestioned superpower of the Mediterranean, so this represented a major setback. To combat this threat, the Empire developed a weapon that has taken on an almost legendary status—Greek Fire.

Speaking of superweapons, today we discuss the mRNA vaccines, true superweapons against COVID-19. I want to contextualize a paper I saw that talked about antibody levels, because I think it’s important to understand the subtleties here.

Also, I address a reader comment against the unproven, highly speculative ivermectin therapy that some are unwisely pushing as an option for protection against COVID-19.

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:

Share COVID Transmissions

Now, let’s talk COVID.

Moderna’s vaccine supposedly yields antibody levels 2x as high as Pfizer’s

This study made headlines yesterday: https://jamanetwork.com/journals/jama/fullarticle/2783797

In the study, 2500 healthcare workers who got different vaccines (specifically the Pfizer and Moderna vaccines) had their antibody levels measured and compared with one another. The authors of this brief letter examined antibody levels in vaccinated people who had previously been infected compared with those who had not, and in those who received different vaccines regardless of their antibody status.

Previous infection followed by vaccination caused much higher antibody titers to be observed—by nearly a factor of 10 compared to people without that previous infection experience. Wow! I guess. We’ll come to the interpretation of this in a minute.

Now to turn to the differences between vaccines. Here, the Moderna vaccine yielded approximately 3-8 times the antibody titer as the Pfizer vaccine. What does this tell me?

Pretty much nothing. Firstly, these aren’t huge effects. The research letter itself presents the data in a way that makes this clear:

Image shows a series of 6 graphs showing the difference in antibody levels between various conditions. It’s not important to understand each individual graph; what’s important is that the graphs all use a log scale, and on that log scale, none of the conditions look wildly different. This choice of a log scale is deliberate, and a standard choice—it’s large differences, in multiple orders of magnitude, that are worth writing home about in antibody studies.

As you can see, using standard visualization approaches, these differences all look pretty small. The nearly 10x increase described a moment ago seems to me to be most likely to have a real impact on immune protection, but it’s still pretty small. That brings me to the second point. You see, as I’ve mentioned a few times in the past few weeks, we still don’t know what immune responses actually correspond to protection against COVID-19. These correlates of protection may take some time to sort out, as we compare the real-world performance of vaccines against our molecular understanding of them. Right now there is no test a person can get that tells them if they are certainly protected against COVID-19. Such tests do exist for diseases like measles, because we do understand the correlates of protection in that context.1 People often ask me what their COVID-19 antibody test results mean, and I am forced to answer “they mean you got an antibody test.” I can’t say much else! We do not know right now, for COVID-19, whether antibodies are actually the beginning and end of the story of protective immunity. We don’t know of a level of antibodies that is reasonably protective. So if I see a particular number for antibody levels in a particular person, all I can say is, “That sure is a number. Real numbery.”

For this reason, I can’t say whether these explorations of antibody levels actually make a difference. The 10x increase in those with previous infection speaks to me as potentially more meaningful than the 3-8x difference between the two vaccine brands, but both findings could be totally meaningless. Why? Well, even if antibodies are a correlate of protection, maybe you only need 100 units2 per milliliter to be protected—everything in this study brought about mean responses above 1500 units per milliliter. Again to pick on measles, it’s thought that 0.12 units per milliliter is the protective threshold for that disease, but the measurement system used to establish that for measles isn’t quite the same so we can’t actually cross-compare these thresholds. The reason I mention it is to point out that protective thresholds can be quite low, actually. If the protective threshold for COVID-19 is low, then it really doesn’t matter if you have 100x the amount of antibodies that you need or 1,000,000x the amount. Both situations would be protective.

And that’s all under the assumption that antibodies are the real key to protection here, an assumption that nobody should really be making at this time. That’s not to say that this work is without scientific merit—there are actually a few interesting hypotheses that I’d like to see tested based on this paper—but in this instance I’m talking about practical, everyday utility. That’s just too limited at this time.

So when you see headlines that say “X generated more antibodies than Y,” this isn’t necessarily evidence that X generates better protection than Y. I wouldn’t make any choices based on this finding—at least for now. Perhaps down the line, when correlates of protection are well understood, but we’re just not there yet.

What am I doing to cope with the pandemic? This:

Noticing Strollers

The impending arrival of our first child has made me hyperaware of strollers and stroller brands in the world around me. It’s something that once you become attuned to seeing, you really can’t stop noticing. Which brands are popular? How to people modify them?

My wife and I already made our decisions on this score, but attunement to this aspect of the world doesn’t go away quickly, it seems—now that I’ve bitten this particular apple, I’m stuck in a world where I can tell the difference between an UPPABaby Vista V2 and a Chicco Corso LE, from 50 feet away.

A reader who self-identifies as “Just Another Bozo on the Bus” wrote in to say:

Hi John, I love your great public service! I have a question about another study. One of my friends is one of those people who has drunk the Kook-Aid hook, line, and sinker. He forwarded this study to me from the American Journal of Therapeutics which seems to show that Ivermectin has some therapeutic use. What is your opinion on this study? I see that it has cited the retracted Elgazzar paper. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8088823/. I am very skeptical, but then again there is a history of groupthink in medicine, as in other fields.

As ever, thank you for the kind words about the newsletter. I’m going to presume “Kook-Aid” was a deliberate play on words for “Kool-Aid”—love it. Even if it was a typo, let’s pretend it was deliberate, because it’s funny.

This reader’s senses with regard to the reliability of the linked study is on-target. The thing about any kind of “review”—aka “research synthesis”—is that if you synthesize together things that are no good, your synthesis is also no good. This is similar to how if you use fresh vegetables in a recipe vs ones you found rotting in the trash,3 you’re going to get a different result. In other words, garbage in means garbage out.

The Elgazzar study was a complete fabrication that distorted public perception of ivermectin because of its falsehoods. Including it means the entire review has been distorted by those falsehoods. It’s a shame, really. Scientific fraud hurts everyone.

That said, it’s not beyond the realm of possibility that ivermectin actually does something against COVID-19. That’s why it’s currently being studied in clinical trials. But it’s equally possible that it does nothing, and since it can have harmful adverse effects, that’s why it should only be currently used in clinical trials and not by people self-medicating. Ivermectin boosters are fond of saying that the drug is “safe,” but nothing is categorically safe. Everything—water, clothing, air—has a safe dose and an unsafe dose, and a safe context for use vs unsafe contexts for use. Ivermectin is no exception, and the context for use as an antiparasitic medication is very different than the proposed uses in COVID-19.

On the other hand, we know the safe dosing of vaccines, monoclonal antibodies, steroids, and supplemental oxygen in the COVID-19 context. Stick to these proven approaches until meaningful, genuine clinical trial results show us what place, if any, ivermectin may have in COVID-19 therapy.

I said much of this, with less belabored explanation, in my reply:

The inclusion of the discredited and fraudulent Elgazzar paper basically disqualifies this review from being worthwhile to consider.
That said, I think ivermectin deserves clinical trials to assess its efficacy and safety in COVID-19 applications. I understand that several are underway and await their results, but in the meantime I do not know of any blueprint available for safe and efficacious use of the drug. We have several lines of other COVID-19 defenses, thankfully, beginning with vaccination and continuing through to monoclonal antibody cocktails and eventually steroids and oxygen. Nearly every COVID-19 death is preventable without the use of experimental drug regimens.

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.

Viral Transmissions

Discussion about this post