COVID Transmissions for 3-2-2022
Big developments in SARS-CoV-2 origins; more evidence of immune escape by Omicron
Greetings from an undisclosed location in my apartment. Welcome to COVID Transmissions.
It has been 807 days since the first documented human case of COVID-19. In 807, Vikings landed on the Cornish coast and formed an alliance with the Cornish against the kingdom of Wessex. We’re entering a period here where wars with the Vikings were a big part of life in England.
In today’s world, I want to talk to you about recent information on immune escape by the Omicron variant, and some recent papers that have provided what I believe is strong evidence on the origins of SARS-CoV-2.
This new evidence leaves me feeling pretty much convinced about where this virus came from.
Bolded terms are linked to the running newsletter glossary.
Results of the subscription offer
I wanted to thank everyone who subscribed during the discount period, which has now ended. This newsletter has flourished with your support, adding dozens of paid subscriptions (more than doubling the number), and has also gained quite a few free subscribers as well. I appreciate you all. By reading and sharing what we discuss here—and engaging with the conversation—you help improve the fight against COVID-19. By subscribing with monetary support, you’ve definitively helped this newsletter rank up in Substack search results, too—which helps us fight misinformation. I’m very grateful for all of you.
Even though the subscription discount has ended for now, I still want your support in getting the word out:
Keep COVID Transmissions growing by sharing it! Share the newsletter, not the virus. I rely on you to help spread good information, which you can do with this button:
Now, let’s talk COVID.
Prior immunity vs Omicron variant: Qatari experience
A recent paper (technically, a “correspondence” piece, or a letter to the journal) in The New England Journal of Medicine (NEJM) offers analysis of the effect of prior infection on immunity against newer variants—everything from Alpha to Omicron. Study here: https://www.nejm.org/doi/full/10.1056/NEJMc2200133
The findings for protection against symptomatic disease are as follows, adjusted for vaccination status (which would of course affect protection):
Vs Alpha: 90.3% protection (95% CI:1 60.4% to 97.6%)
Vs Beta: 85.1% (95% CI: 74.5% to 91.3%)
Vs Delta: 92.0 (95% CI: 87.8% to 94.7%)
Vs Omicron: 55.9% (50.5 % to 60.8%)
It is interesting in these numbers that past infection was very protective against the Alpha, Beta, and Delta variants…but not Omicron.
Omicron is the dominant variant of concern in 18 different countries right now, and it is expected to outcompete the past variants of concern wherever it encounters them. Based on this, protection from future infection with Omicron variant virus is halved, if you are relying on recovery from past infection for protection. This is not totally unexpected; protection against symptomatic disease from the Omicron variant with three doses of an mRNA vaccine fell into the 55-80% range as well. However, it does illustrate that immune escape by this variant is not a matter of vaccine escape; it is escape of multiple forms of prior immunity.
As we saw in a recent story covered here, three exposures to the SARS-CoV-2 spike protein produces superior immunity compared with 2 exposures, valuable information for contextualizing what we learned here.
There is no question that past infection offers some amount of protection, but we also know that it is possible, particularly with the Omicron variant, for recovered persons to get sick again. What these data don’t tell you is the state of those individuals after their recovery, and how a certain percentage of them had already died of COVID-19 before this study even started to happen. What I mean is that everyone who gets COVID-19 without being vaccinated has an uncomfortably high risk of dying from it, and those who die of COVID-19 do not get the opportunity to be protected from COVID-19 in the future. Furthermore, even if you survive, lung damage from COVID-19 is not necessarily going to go away, and if you get sick again, having damaged lungs isn’t going to help you. The same goes for a damaged heart, predisposition to blood clots, and many other permanent injuries we know to result from COVID-19.
The reason I say this is to make clear that going out and getting infected with SARS-CoV-2 is still a bad idea compared with vaccination. Eventually I do think everyone is going to have an encounter of some kind of SARS-CoV-2, and I think in that encounter you can gain all the protection listed above, and more, if you are already vaccinated.
In fact, I can confirm that’s true in this study. When the analysis wasn’t adjusted for vaccination status, the numbers were statistically the same as in that list.
Another thing to mention is that this study was performed in Qatar, so I would caution anyone about saying its results apply to all people everywhere. Different nations have different lifestyles and different communities, and it is hard to say how those differences change our relationships with disease. What’s true in Qatar may not be true in Japan. Just something to keep in mind.
For those planning to go through and read the full paper, I want to mention that they did include protection from severe and fatal outcomes, but the numbers are so small and the confidence intervals so wide that I cannot trust those findings. Protection against serious outcomes is hard to reliably estimate from this dataset.
I will note that the mountain of data for vaccines has given us reliable estimates of effectiveness against serious outcomes and death, and it remains very high for all variants. So there’s that.
At any rate, I hope this serves to illustrate that Omicron is an excellent immune escape artist—that finding, I think, is the most reliable takeaway from these and other data we have seen here before.
Impactful findings on SARS-CoV-2 origins
Two recent preprint papers have made quite a splash in the work of investigating the origins of COVID-19 and the virus that causes it, SARS-CoV-2.
One establishes that there were at least two introductions of SARS-CoV-2 into Wuhan, and that these two introductions were 2 different lineages of virus, which the authors named “A” and “B”: https://zenodo.org/record/6291628#.Yh7pjBPMInd
What’s really important, fact-wise, here in this, the more technical of the two studies, is:
There were two distinct lineages, A and B, that emerged in Wuhan in late 2019
In samples collected early on in Wuhan from humans, these lineages did not have a “transitional haplotype”; in other words, there is no sample representing an intermediate, evolutionary link between these two lineages in human patients early in the pandemic
Circulating genomes in the early outbreak in Wuhan were actually less diverse than originally thought, and grouped into these two lineages quite well
What I haven’t listed here is how we can put these facts together to understand possible emergence scenarios, but I’ll do that now. Low diversity, the presence of two lineages, and the lack of an intermediate sample connecting the lineages as related to each other suggests that these were two competing spillover events.
SARS-CoV-2 entered humans twice in Wuhan in 2019. By the authors’ estimation, once in late November or early December, and another time maybe some weeks later. We could argue for a while over what this means for the “lab leak” hypothesis, but the fact is it is very hard to consider this compatible with a scenario where this virus was released from a laboratory. Laboratory virus stocks are carefully maintained and, if any kind of genetic modification system is in use, do not generally have enough diversity for distinct lineages to be identified. What this looks much more like is two independent spillover events.
The second paper sheds more light on the situation: https://zenodo.org/record/6291628#.Yh7q_xPMIne
This paper uses analysis of patient histories as well as virus lineage data to establish that the Huanan Seafood Market was, almost certainly, the epicenter of the initial outbreak in Wuhan in late 2019.
One of the more interesting figures from this paper is this one:
This is classic epidemiology, right here. The authors look at where cases are, and establish geographically how they are related to each other to understand where the disease originated. This is the same way that John Snow figured out the source of a cholera outbreak in London in the 19th century, but with 21st century tools.
I applaud the authors for not just stopping here. They went further, and established that this distribution is not the result of there being more people near the Huanan market who were likely to become symptomatic than in other areas of the city. They also established that, based on case densities later in the outbreak in Wuhan, this is not simply the result of this particular area being highly-populated or highly-trafficked.
And then they go further. They provide clear photographic evidence that live animals, in particular animals like raccoon dogs that are known to be susceptible to sarbecoviruses like SARS-CoV-2, were being sold in the Huanan Seafood Market at the time of the outbreak:
These photos are representative images, and the species pictured correlate with a list of live animals for sale at the market in November 2019 that has also been found by investigators. We know that SARS-CoV-2 has a very wide host range in mammals. We know that SARS coronaviruses spread from mammals to humans in live animal markets. We know that live animals were being sold in the Huanan Seafood Market in December 2019. We know that cases of SARS-CoV-2 in Wuhan in December 2019 center on the Huanan Seafood Market.
Many people would consider this definitive evidence that SARS-CoV-2 emerged at the Wuhan Seafood Market. The authors of this paper were not “many people”—as seasoned virologists and emerging disease investigators, they went further. It is not enough to find the victims and the smoking gun. They went looking for the bullets.
The Chinese government collected environmental samples from the Huanan Seafood Market at the start of the outbreak. Many were positive for SARS-CoV-2. This alone would not necessarily be meaningful evidence; if the virus were spreading in Wuhan, then it is reasonable that people could have brought SARS-CoV-2 to the market rather than getting it from animals there. That is contradicted by the clustering evidence, but since we didn’t have that evidence until recently, the environmental samples from the Huanan Seafood Market were considered ambiguous evidence until now.
The authors of this paper used them to connect the virus to the live animals. You see, the sampling locations that were positive for SARS-CoV-2, vs those that were negative, were recorded by the sampling team. Keep in mind that at the time of sampling, no one anywhere had any clear evidence of where the virus first emerged. So the sampling was pretty uniform at the market, to track down where the virus might have originated.
This team, in their current paper, did a cluster analysis on where the SARS-CoV-2 positive samples were found in the Huanan Seafood Market. They overlaid the results of that work onto a map showing where in the market the live animals were found. Not only did SARS-CoV-2 positive samples cluster around specific spots, they were found near specific cages that were used to sell live animals.
It seems that this team has established where the emergence of SARS-CoV-2 occurred to within a few feet, after narrowing it down to the market vs the entire city of Wuhan.
These two papers have me completely convinced. SARS-CoV-2 first entered human populations from wild, live mammals that were captured and sold at the Huanan Seafood Market. There is no other reasonable explanation that is compatible with the following facts:
Two different lineages were found in early cases in Wuhan, suggesting two spillover events
Despite this, all cases centered on the Huanan Seafood Market early in the outbreak
Environmental samples place SARS-CoV-2 in high association with live animals for sale at the market
Pictures of the animals present indicate they were susceptible hosts for SARS-CoV-2
Any scenario involving the Wuhan Institute of Virology would lead to counterfactual evidence to these facts. WIV is not near the Huanan Market. A virology institute would not release multiple different lineages, and they wouldn’t both be associated with cases at a market nowhere near that institute. Environmental samples of the virus would cluster near humans in the market and not animals in the market, if they had been brought there by humans from elsewhere.
I now believe that the Huanan Market is where this virus emerged. It came out of two different animals for sale there, possibly weeks apart, and circulated among employees and their contacts in Wuhan before it was detected through hospitalizations.
It is worth noting that there are possible alternatives to this even still, despite the fact that I am personally convinced. Some doubt that the 2 lineages emerged in 2 events at the same market, on the strength of the evidence, and suspect that one of the lineages may have come from elsewhere—something that would be reflected, perhaps, if we had environmental samples from elsewhere. Some have suggested that prior samples collected from pneumonia patients in Wuhan in November 2019 could change the narrative further. Both of these doubts rely on there being evidence that has not been examined. The problem with that is that we don’t have such evidence and even the suggestion that it exists is speculation. We can only make conclusions with the evidence that we have.
An additional question that one could ask is—where is the SARS-CoV-2 positive animal or animals? It would of course be better if we had samples from an animal sold at the market, and could show that it was infected with this virus. On the other hand, I do not think even this would be definitive because one could always argue that the animal got it from a human. The spatial clustering would suggest otherwise, but the doubters seem unconvinced by that anyway.
The issue here is that it’s hard to find food animals when they’ve already been eaten two years ago. I certainly don’t have remnants of food that I ate two years ago. I hope.
One way around this would be to find the sources of the animals being sold in the Huanan Seafood Market, and test the local animal and human populations for evidence of SARS-CoV-2 or its ancestors. That would help to further connect this chain of events to a definitive narrative without needing a culinary time machine.
The New York Times has picked up coverage of this story: https://www.nytimes.com/interactive/2022/02/26/science/covid-virus-wuhan-origins.html?smid=tw-nythealth&smtyp=cur
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.
What am I doing to cope with the pandemic? This:
Funeral
At this point it is relatively public knowledge, so I can mention that the family situation I was dealing with was the death of an older relative.
Every step of the way of this process has been made more complicated by the pandemic environment—from navigating healthcare, to end-of-life, to funeral planning.
I was not really directly involved in any of this for this person in this particular moment of sadness, but I saw the difficulties they and those closest to them went through and I will admit even that has weighed on me. I hope that we can get past this thing, so that even in the most difficult of times for them, families can be together more frequently.
Several readers shared thoughts on the CDC mask guidance. Here is Sam’s position:
If the CDC is going to recommend that mask-wearing be based on "your personal preference, informed by your personal level of risk," then it needs to level with the public about what exactly those risks are -- in particular, the significant danger of long COVID. I worry that we're inviting a massive wave of disability that neither our healthcare system nor our social welfare programs are really equipped to handle.
I also worry that, in practice, things won't be left up to people's personal preferences. Some employers will prohibit employees from wearing masks. I'm not sure there shouldn't be legislation protecting people who continue to wear masks in the workplace from being terminated for that reason.
Dr. Linda Post, MD, also shared a perspective:
I actually think the CDC guidelines are a good compromise. With so many people now testing at home, we can't necessarily trust the case numbers. Hospital admissions may be a lagging indicator but for public health purposes they let us know when we are heading into trouble. As a healthcare provider with an insatiable desire for knowledge, I would often go to NIH and CDC sites to see what viruses were out there so I could have a sense of what to expect. With the breakdown in the sentinel lab system this doesn't work as well but it helps. Managed care insurance plans use similar data to help predict when they need to deploy more resources on-line or into the community.
I disagree with Dr. Post, with respect. I think there are ways to make our case surveillance more reliable, and furthermore I think Dr. Post and I are very far apart on how dangerous it is to rely only on hospitalization numbers. Here is my reply:
If there were no other solution for the breakdown in the value of case numbers, perhaps this would be a reasonable move. But there is a solution--random sampling. This has been proposed by several epidemiologists and is used to surveil other diseases. We don't need to fly blind on this.
I disagree wholeheartedly that hospital admissions will let us know when we are "heading into" trouble. By the time hospital admissions begin to tick up, it is too late to stop the spread. This is a virus that can generate as many at 10 cases for every positive case, and it does that within days of initial infection--often before symptoms have appeared. Meanwhile it can take 5-14 days for the disease to become serious enough to lead to hospitalization. With an infectious window of 3-5 days shortly after infection, by the time a single case gets to the hospital, they could have caused anything from hundreds to tens of thousands of downstream cases. We have not seen this kind of unrestrained spread so far because we have been working early enough to restrain the spread, on at least some level.
In the course of the pandemic, politicians have consistently underestimated the meaning of exponential growth. It happens faster than we can comprehend. Using a lagging indicator that takes a week or more to realize changes in the community means many deaths that could have been avoided--and a great deal of personal risk to individuals.
In general, I think many policymakers have continued to make errors with each successive wave where they do not account for the success of our control measures since they still observed bad things happening.
Since our measures were not perfect, we assume our epidemiological understanding reflects inherent realities about SARS-CoV-2, rather than reflecting how the virus spreads when we are trying to control it. So we remove precautions and the virus begins to spread differently.
We have attributed these patterns to changes in the virus (and to some degree that is true), but changes in our behavior also make a big impact on spread. Just because, say, the Omicron variant appeared more contagious than anything seen before doesn’t mean that we didn’t prevent cases of it using what measures we did take. We need to remember that we don’t get to see how many deaths we have actually prevented. Just the ones that happen.
In other words, don’t assume that it could not have gotten any worse. Making that assumption in March 2020 would have been wrong and it’s wrong to make it in March 2022. To think that we can rely on patterns from a time when masks were being recommended on the basis of case numbers, as a way to predict what will happen when they are being recommended on the basis of a lagging indicator, is an inference that I do not feel is very safe to make.
You might have some questions or comments! Join the conversation, and what you say will impact what I talk about in the next issue. You can also email me if you have a comment that you don’t want to share with the whole group, or if you are unable to comment due to a paywall.
If you liked today’s issue, please consider becoming a paid subscriber and/or sharing this newsletter with everyone you know.
Please know that I deeply appreciate having you as readers, and I’m very glad we’re on this journey together.
Always,
JS
A 95% confidence interval (CI) is a measurement of the uncertainty in a given set of results. Measurements of the real world can be both inaccurate and imprecise, and so the strict numbers are only an estimate of what the “true” value would be if we had perfect observational powers. The 95% CI represents the range in which we believe it is 95% likely the true value is found. The wider it is, the less confidence there is around the estimated value.
Not directly relevant to this article, but you saw that Covovax (Novavax's vaccine) was approved by several more countries, I'm sure. There is (according to WHO) no longer a vaccine shortage, just a systemic inability to administer the vaccines for various reasons. Here in the States and in other nations like Germany, it's reluctance from the public. In the less-developed world, it's infrastructure, as far as I can tell.