I just discovered your (great) newsletter while searching for information on a vaccination-related matter on which you'd commented previously. I was wondering if you'd be able to provide an update of sorts, as well as comment on a related matter.
In your 11/18/20 issue, in response to a reader whose husband voiced concerns about mRNA technology, you mentioned that a somewhat less far-fetched concern would be that vaccination could generate an autoimmune response. I've seen these issue raised elsewhere here and there.
Do you view this problem as more, less, or about equally likely to occur as you did in November? Do you believe that we have seen signs of such reactions already, if they were going to occur? On a more theoretical level, are there reasons why we *wouldn't* expect such a reaction to occur, e.g., something distinctive about the vaccine-generated spike vs. the virus-generate spike? How worried should we be?
Similarly, an article appeared recently in the MDPI journal Vaccines (not to be confused the with Elsevier journal Vaccine) suggesting that signaling from the spike protein, via either infection or the vaccine, could lead to pulmonary arterial hypertension (see: https://www.mdpi.com/2076-393X/9/1/36/htm), a frightening prospect. How seriously would you take these concerns? (The article has been gaining some traction on antivax-ish corners of the internet. The authors don't seem like cranks, though I know MDPI's peer review practices have been subject to some criticism.)
Virtually everyone I know is at least partially vaccinated, as am I. Reading these things after the fact has honestly made me start to panic a bit.
I just want to say that I really love this question. I will write out an extensive response to it a bit later today, because it's a deep scientific question that is going to require that, but I wanted to say how much I like it.
But, for now, I am going to say that I do not think there is any reason to be concerned. More on this later.
Thanks. There are many scientists and science communicators making laudable efforts to communicate the basics, but it's incredibly difficult to find someone who can answer these more complex questions.
Thank you for your kind words about the newsletter, and your insightful questions.
I’m glad you’ve revisited the concern about autoimmunity that I raised before we had approved vaccines. We’ve learned a lot since then, and there’s a lot more to say. At the time, I was basing this possibility on two items. One, that “long COVID” appeared to be an autoimmune condition mediated by exposure to natural infection, and two, that in the past there have been some other vaccines that have induced apparent autoimmune reactions in some patients. I thought it possible that some of the observed long COVID symptoms might be the result of autoimmune reactions induced by antibodies to the spike protein, and if that were the case, there could be some patients who experienced autoimmune reactions to the vaccines as well. My theory here was that the spike protein could contain some sequence that acts as a molecular mimic of something in the human body, and so antibodies against it might cross-react with that human target.
In the time since then, it has become clear that autoimmune reactions are not really happening in response to these vaccines to the degree one would expect in the above case. Another thing that we have learned is that, apparently, vaccination actually seems to alleviate long COVID symptoms in patients who experience them. This latter situation could arise if the vaccines somehow provide a strong and specific immune response that induces the immune system to “prune” back inappropriate responses that developed during natural infection (one of the headlines in tomorrow’s issue with discuss the evidence that natural infection produces worse antibody responses compared with vaccination), but it could also arise if the mechanism behind long COVID turns out to have nothing to do with autoimmunity. There are now immunologists who think that long COVID may actually be the result of persistent virus somewhere in the patient. This would be a complicated explanation, but the vaccine remedying the condition makes it something that has to be considered.
The only apparent autoimmune reaction that I’ve heard about with any COVID-19 vaccine is the vaccine-induced thrombocytopenic thrombosis syndrome that has made negative headlines for the AstraZeneca and J&J vaccines. It appears that in the extremely small population of patients who have this reaction, antibodies are raised against natural clotting factors. This does not appear to have anything to do with the SARS-CoV-2 components of these vaccines, since the mRNA vaccines do not have this problem. It may have to do with the adenovirus vectors that these vaccines use. However, it is exceedingly rare and likely there is something about the few patients in whom this has occurred that made the reaction possible.
Anyway, to turn back to the possibility that long COVID has something to do with autoimmunity. If this is the case, then I suspect that it happens because the full, competent virus disrupts the quality of the antibody response. The vaccines don’t do this, and this is probably because they are lacking in whatever component of the full virus is responsible for this antibody response disruption. I’ll reiterate that it’s not clear whether long COVID actually has an autoimmune mechanism underlying it, but if it does, then I think my hypothesis here reconciles why we don’t see autoimmunity yielding the same effects in response to the vaccines.
Now, to your second question. Wow, that’s a bad paper. I’m actually not sure where to start. I looked into the references (since that is a review), and it is mostly the review authors citing themselves. Those authors are not virologists or vaccinologists and it is apparent that they do not understand the vaccines they are writing about. Their work used the full-length SARS-CoV-2 spike protein without modifications, in a tissue culture setting that models a process that, if it goes on for a long time, might cause disease. Most of the available vaccines use a different construct, and even if the effect they observed in their tissue culture experiments is real, they haven’t demonstrated that it actually happens with the construct used in most vaccines.
That said, I don’t think it’s realistic at all to expect what they have seen in tissue culture to happen in any real way in vaccinated patients. Their experiments were done in cultured lung cells to model pulmonary hypertension. Pulmonary artery hypertension is a condition that can take a long time to develop into symptomatic disease, and what’s more, it takes place in the lungs. None of the vaccines are delivered to vascular cells in the lungs. All of the currently-approved vaccines are delivered to muscle in the arm. They express the spike protein there, and there is no mechanism that can be reasonably expected to deliver the spike protein from muscle cells, far from major arteries and veins, to the lung vasculature. The spike protein does not have a means to exit cells and transport itself such a great distance.
Imagine I am wrong about that, though. The vaccines currently on the market contain a small amount of material that expresses the spike glycoprotein for a short period of time. Imagine that some small fraction of that expressed spike glycoprotein travels, by some mechanism currently unknown to science, to the lungs from the muscles of the arm. There, it causes some cellular protein activation that, if it persists, could eventually lead to pulmonary artery hypertension. Except, it won’t persist—because your body is developing an immune response specifically engineered to destroy the spike glycoprotein. Within a couple of weeks of getting vaccinated, an army of cells and antibodies are patrolling the vaccinated person’s body trying to destroy every single molecule of spike protein that they encounter. Meanwhile, the actual vaccine components degrade eventually. They are not persistent in the body. They are used to make spike protein for a week or two, maybe even less time, and are then digested and destroyed.
So, to summarize: the effect shown in the paper hasn’t been shown in a living system, it hasn’t been shown to happen with any vaccine construct, it hasn’t been shown to be competent to cause disease, and there are several reasons to be skeptical that there is any mechanism by which the observed experimental effect could be seen in an actual living system, particularly not for long enough to cause disease. With all of those points separating these results from the likely reality, I don’t take these concerns seriously at all.
What’s more, if the proposed spike protein-mediated risk for PAH is even possible, it’s clear that the risk posed by virus infection, which expresses large quantities of the spike protein right inside the lungs, would without a doubt be higher than any risk that the vaccine could possibly pose. Since the vaccine doesn’t express spike protein in the lungs and actively helps you prevent the expression of spike protein in the lungs, it seems that even if this effect is real, the vaccine is actually your best method to prevent this only-possibly-real effect.
I’m honestly stunned that a journal targeted to vaccinologists would even run such a paper. But, we should know by now—peer review is not even close to a perfect system. It has its uses, but it also has its share of failures.
I'm jealous. They didn't give me a sticker at the Aqueduct. I'm only 4 days from my 2-week anniversary! Which is good, because business travel is a thing again, and Sunday I leave for Massachusetts.
Hi John,
I just discovered your (great) newsletter while searching for information on a vaccination-related matter on which you'd commented previously. I was wondering if you'd be able to provide an update of sorts, as well as comment on a related matter.
In your 11/18/20 issue, in response to a reader whose husband voiced concerns about mRNA technology, you mentioned that a somewhat less far-fetched concern would be that vaccination could generate an autoimmune response. I've seen these issue raised elsewhere here and there.
Do you view this problem as more, less, or about equally likely to occur as you did in November? Do you believe that we have seen signs of such reactions already, if they were going to occur? On a more theoretical level, are there reasons why we *wouldn't* expect such a reaction to occur, e.g., something distinctive about the vaccine-generated spike vs. the virus-generate spike? How worried should we be?
Similarly, an article appeared recently in the MDPI journal Vaccines (not to be confused the with Elsevier journal Vaccine) suggesting that signaling from the spike protein, via either infection or the vaccine, could lead to pulmonary arterial hypertension (see: https://www.mdpi.com/2076-393X/9/1/36/htm), a frightening prospect. How seriously would you take these concerns? (The article has been gaining some traction on antivax-ish corners of the internet. The authors don't seem like cranks, though I know MDPI's peer review practices have been subject to some criticism.)
Virtually everyone I know is at least partially vaccinated, as am I. Reading these things after the fact has honestly made me start to panic a bit.
I just want to say that I really love this question. I will write out an extensive response to it a bit later today, because it's a deep scientific question that is going to require that, but I wanted to say how much I like it.
But, for now, I am going to say that I do not think there is any reason to be concerned. More on this later.
Thanks. There are many scientists and science communicators making laudable efforts to communicate the basics, but it's incredibly difficult to find someone who can answer these more complex questions.
Thank you for your kind words about the newsletter, and your insightful questions.
I’m glad you’ve revisited the concern about autoimmunity that I raised before we had approved vaccines. We’ve learned a lot since then, and there’s a lot more to say. At the time, I was basing this possibility on two items. One, that “long COVID” appeared to be an autoimmune condition mediated by exposure to natural infection, and two, that in the past there have been some other vaccines that have induced apparent autoimmune reactions in some patients. I thought it possible that some of the observed long COVID symptoms might be the result of autoimmune reactions induced by antibodies to the spike protein, and if that were the case, there could be some patients who experienced autoimmune reactions to the vaccines as well. My theory here was that the spike protein could contain some sequence that acts as a molecular mimic of something in the human body, and so antibodies against it might cross-react with that human target.
In the time since then, it has become clear that autoimmune reactions are not really happening in response to these vaccines to the degree one would expect in the above case. Another thing that we have learned is that, apparently, vaccination actually seems to alleviate long COVID symptoms in patients who experience them. This latter situation could arise if the vaccines somehow provide a strong and specific immune response that induces the immune system to “prune” back inappropriate responses that developed during natural infection (one of the headlines in tomorrow’s issue with discuss the evidence that natural infection produces worse antibody responses compared with vaccination), but it could also arise if the mechanism behind long COVID turns out to have nothing to do with autoimmunity. There are now immunologists who think that long COVID may actually be the result of persistent virus somewhere in the patient. This would be a complicated explanation, but the vaccine remedying the condition makes it something that has to be considered.
The only apparent autoimmune reaction that I’ve heard about with any COVID-19 vaccine is the vaccine-induced thrombocytopenic thrombosis syndrome that has made negative headlines for the AstraZeneca and J&J vaccines. It appears that in the extremely small population of patients who have this reaction, antibodies are raised against natural clotting factors. This does not appear to have anything to do with the SARS-CoV-2 components of these vaccines, since the mRNA vaccines do not have this problem. It may have to do with the adenovirus vectors that these vaccines use. However, it is exceedingly rare and likely there is something about the few patients in whom this has occurred that made the reaction possible.
Anyway, to turn back to the possibility that long COVID has something to do with autoimmunity. If this is the case, then I suspect that it happens because the full, competent virus disrupts the quality of the antibody response. The vaccines don’t do this, and this is probably because they are lacking in whatever component of the full virus is responsible for this antibody response disruption. I’ll reiterate that it’s not clear whether long COVID actually has an autoimmune mechanism underlying it, but if it does, then I think my hypothesis here reconciles why we don’t see autoimmunity yielding the same effects in response to the vaccines.
Now, to your second question. Wow, that’s a bad paper. I’m actually not sure where to start. I looked into the references (since that is a review), and it is mostly the review authors citing themselves. Those authors are not virologists or vaccinologists and it is apparent that they do not understand the vaccines they are writing about. Their work used the full-length SARS-CoV-2 spike protein without modifications, in a tissue culture setting that models a process that, if it goes on for a long time, might cause disease. Most of the available vaccines use a different construct, and even if the effect they observed in their tissue culture experiments is real, they haven’t demonstrated that it actually happens with the construct used in most vaccines.
That said, I don’t think it’s realistic at all to expect what they have seen in tissue culture to happen in any real way in vaccinated patients. Their experiments were done in cultured lung cells to model pulmonary hypertension. Pulmonary artery hypertension is a condition that can take a long time to develop into symptomatic disease, and what’s more, it takes place in the lungs. None of the vaccines are delivered to vascular cells in the lungs. All of the currently-approved vaccines are delivered to muscle in the arm. They express the spike protein there, and there is no mechanism that can be reasonably expected to deliver the spike protein from muscle cells, far from major arteries and veins, to the lung vasculature. The spike protein does not have a means to exit cells and transport itself such a great distance.
Imagine I am wrong about that, though. The vaccines currently on the market contain a small amount of material that expresses the spike glycoprotein for a short period of time. Imagine that some small fraction of that expressed spike glycoprotein travels, by some mechanism currently unknown to science, to the lungs from the muscles of the arm. There, it causes some cellular protein activation that, if it persists, could eventually lead to pulmonary artery hypertension. Except, it won’t persist—because your body is developing an immune response specifically engineered to destroy the spike glycoprotein. Within a couple of weeks of getting vaccinated, an army of cells and antibodies are patrolling the vaccinated person’s body trying to destroy every single molecule of spike protein that they encounter. Meanwhile, the actual vaccine components degrade eventually. They are not persistent in the body. They are used to make spike protein for a week or two, maybe even less time, and are then digested and destroyed.
So, to summarize: the effect shown in the paper hasn’t been shown in a living system, it hasn’t been shown to happen with any vaccine construct, it hasn’t been shown to be competent to cause disease, and there are several reasons to be skeptical that there is any mechanism by which the observed experimental effect could be seen in an actual living system, particularly not for long enough to cause disease. With all of those points separating these results from the likely reality, I don’t take these concerns seriously at all.
What’s more, if the proposed spike protein-mediated risk for PAH is even possible, it’s clear that the risk posed by virus infection, which expresses large quantities of the spike protein right inside the lungs, would without a doubt be higher than any risk that the vaccine could possibly pose. Since the vaccine doesn’t express spike protein in the lungs and actively helps you prevent the expression of spike protein in the lungs, it seems that even if this effect is real, the vaccine is actually your best method to prevent this only-possibly-real effect.
I’m honestly stunned that a journal targeted to vaccinologists would even run such a paper. But, we should know by now—peer review is not even close to a perfect system. It has its uses, but it also has its share of failures.
I'm jealous. They didn't give me a sticker at the Aqueduct. I'm only 4 days from my 2-week anniversary! Which is good, because business travel is a thing again, and Sunday I leave for Massachusetts.
Aw man, there should be a service you can write to and get a sticker, that's unfortunate.
If stickers are enough motivation for people to get vaccinated, every site should have them in spades!
I'm glad you're close to fully IgG'ed, Carl. That's great news :)
"The thing about the J&J vaccination at this point, though, is that it’s not really central to the US strategy ..."
Apparently many Americans like it for the pure convenience of a single visit, and only one round of adverse effects (for those who have them).
https://www.washingtonpost.com/health/johnson-and-johnson-vaccine-preference/2021/04/28/75ee6662-a770-11eb-bca5-048b2759a489_story.html