Every so often there is a study that goes so viral that people start sending it to me and asking me if there’s anything to it. Sometimes I act and write about such studies; sometimes I do not. Sometimes I roll my eyes at the study and think it’s not worth bothering with, only to see how widely it is being disseminated to spread fear, uncertainty, and doubt about vaccines, cancer treatment, conventional medicine, and the like. When that happens, I generally break down and look at the study. So it has been this week with a study out of Sweden that is being spread far and wide that claims that spike protein gets into the nucleus and interferes with the repair of DNA damage (specifically double-stranded breaks in DNA) by blocking the action of BRCA1, a very important DNA damage repair protein, and 53BP1. For reference, BRCA1 mutations can predispose to a very high lifetime risk of certain cancers, in particular breast and ovarian cancer.
[Orac note: Sorry this is late. Somehow the scheduled posting feature of WordPress didn’t work, and I didn’t notice it until now.]
This is the study. And this is the way antivaxxers are spinning it:
One example of what I’m talking about is this video, which has been viewed 1.6 million times already in which Dr. Mikolaj Raszek wildly speculates that the results of the study urgently need to be confirmed and could demand the reconfiguration of vaccine design because of the risk of cancer from spike protein from vaccines:
Spoiler alert: Even if this study is correct (and its methodological shortcomings produce a great deal of doubt about that, as I’ll discuss), it’s just an in vitro study, and, more importantly, from a strictly biological and logical standpoint it doesn’t mean that COVID-19 vaccines cause cancer through interference with DNA damage repair. Dr. Raszek describes this study as a “game changer.” Another spoiler alert: It’s not.
So what is the key “finding” of this study. I’ll cut to the chase (unusual for me) and list the abstract:
Severe acute respiratory syndrome coronavirus 2 (SARS–CoV–2) has led to the coronavirus disease 2019 (COVID–19) pandemic, severely affecting public health and the global economy. Adaptive immunity plays a crucial role in fighting against SARS–CoV–2 infection and directly influences the clinical outcomes of patients. Clinical studies have indicated that patients with severe COVID–19 exhibit delayed and weak adaptive immune responses; however, the mechanism by which SARS–CoV–2 impedes adaptive immunity remains unclear. Here, by using an in vitro cell line, we report that the SARS–CoV–2 spike protein significantly inhibits DNA damage repair, which is required for effective V(D)J recombination in adaptive immunity. Mechanistically, we found that the spike protein localizes in the nucleus and inhibits DNA damage repair by impeding key DNA repair protein BRCA1 and 53BP1 recruitment to the damage site. Our findings reveal a potential molecular mechanism by which the spike protein might impede adaptive immunity and underscore the potential side effects of full-length spike-based vaccines.
First of all, note that this study really has nothing to do with vaccines, but the authors can’t help but throw in some verbiage that allowed antivaxxers to easily make it about vaccines. The scientists here are testing the hypothesis that COVID-19 mRNA vaccines that use full-length spike protein from SARS-CoV-2, the coronavirus that causes COVID-19. Yet, they can’t help but speculate in the abstract that their work could “underscore the potential side effects of full-length spike-based vaccines.” Later, in the discussion, they write:
In contrast, our data provide valuable details on the involvement of spike protein subunits in DNA damage repair, indicating that full–length spike–based vaccines may inhibit the recombination of V(D)J in B cells, which is also consistent with a recent study that a full–length spike–based vaccine induced lower antibody titers compared to the RBD–based vaccine . This suggests that the use of antigenic epitopes of the spike as a SARS–CoV–2 vaccine might be safer and more efficacious than the full–length spike. Taken together, we identified one of the potentially important mechanisms of SARS–CoV–2 suppression of the host adaptive immune machinery. Furthermore, our findings also imply a potential side effect of the full–length spike–based vaccine. This work will improve the understanding of COVID–19 pathogenesis and provide new strategies for designing more efficient and safer vaccines.
Interestingly, the reference cited doesn’t seem to show anything of the sort. In fact, it doesn’t even mention vaccines, much less receptor binding domain (RBD) COVID-19 vaccines or even the RBD itself. (RBD is just the part of the spike protein that binds ACE-receptor to enter the cell.) Rather, it is a paper about how SARS-CoV-2 mRNA can be found in exosomes, which might help spread the infection. I rather suspect that the authors meant to cite this paper (although I could be mistaken), which reviews data and marshals arguments to suggest that RBD-targeted vaccines are at least not inferior to the current vaccines that use the entire spike protein and very well might be superior. Either way, though, the paper under discussion doesn’t really support the idea that spike protein causes cancer.
Let me boil it down in a way that I’ve boiled it down to people who have written to me. Let’s, for the sake of argument, assume that this paper, even though it’s an in vitro paper using nothing but cell culture methods, is completely valid and the spike protein can enter the nucleus of cells, either from infection or from mRNA vaccination, and bind to important proteins involved in DNA damage repair, thus inhibiting their ability to fix DNA damage, which in the long run can lead to cancer. You know what produces a crapton of spike protein in the cells of many different organs throughout the body? You guessed it! SARS-CoV-2 infection!
Now compare that to vaccination, in which a small amount of mRNA is injected into skeletal muscle (usually the deltoid muscle), where it gets taken up by some muscle cells, there to make SARS-CoV-2 spike protein, which is then displayed on the cell surface and stimulates an immune response. (It is true that vaccine-spike protein does get into the blood—for instance, from the Moderna vaccine—but it’s at an incredibly low concentration, so much so that it took the development of special assays even to detect it, as I discussed a while back.) Basically, if there is a cancer risk or risk of other harm from mRNA-based COVID-19 vaccines, the very worst we could expect is that tumors might develop at the injection site. Of course, even that is very unlikely, again taking the findings of this study at face value, because muscle cells do not replicate and DNA damage repair mechanisms tend to impact cancer only in replicating cells. (Again, compare that to the crapton of spike protein that a SARS-CoV-2 infection produces in the cells of the lungs and other organs, which if you believe the conclusions of this study as spun by antivaxxers, would imply that it is COVID-19 survivors who need to worry about an increased risk of cancer, not those who are vaccinated.) To go beyond that requires postulating that the vaccine mRNA somehow gets into other cells (such as the various white blood cells that circulate in the area, fibroblast in the connective tissue of the muscle fibers, or into the cells of the vasculature in the area) and causes cancer would produce the same result: Local cancers.
Let’s look at it this way. The authors speculate:
Although it is debatable whether SARS–CoV–2 directly infects lymphocyte precursors [26,27], some reports have shown that infected cells secrete exosomes that can deliver SARS–CoV–2 RNA or protein to target cells [28,29].
Remember, the entire hypothesis behind the study is that the spike protein can interfere with DNA damage repair mechanisms in immune cells—a process known as V(D)J recombination—which are important in the the mechanisms of DNA recombination occurring in specific immune precursor cells that lead to the production of specific antibodies. It is this mechanism by which, the authors propose, SARS-CoV-2 might inhibit the immune response to its infection. Yet right here the authors admit that it’s not clear whether SARS-CoV-2 directly infects these cells and have to resort to a handwaving argument about how exosomes (small lipid particles) might carry SARS-CoV-2 mRNA, a process whose significance, even if it occurs, is not yet clear in terms of its ability to contribute to the progression of COVID-19 infection. I’ve also discussed exosomes before in the context of germ theory denial, in which certain deniers have claimed that SARS-CoV-2 doesn’t exist and what is being seen is actually exosomes.
You can stop reading right here if you want, as these are the main reasons why this study does not cast doubt on the safety of mRNA-based COVID-19 vaccines through the supposed ability of the spike protein to inhibit DNA damage repair. Or you can join me as I continue the discussion, specifically describing how the study itself has a number of problems that call into doubt whether its data support its conclusions. At best, this paper is difficult to interpret and not at all clear evidence for the conclusions that its authors make, which is probably why it was published in an MDPI journal, rather than a good virology or molecular biology journal, whose reviewers would certainly have demanded more controls and grilled the authors more about their methods. (MDPI journals tend to be bottom-feeding crappy journals, in my experience and estimation.)
Before I discuss some of the problems with this paper, I have to noe that I’m not an immunologist or an expert in DNA damage repair mechanisms. I am, however, a PhD molecular biologist in addition to being a surgeon, and I’m not alone in noticing the problems. So let’s look at what the authors did. In one experiment, they transfected (introduced plasmid DNA into the cells) HEK293T and HEK293 cells with DNA plasmids expressing the cDNAs encoding the spike protein, as well as a number of other SARS-CoV-2 proteins (NSP1, NSP9, NSP13, NSP14, NSP16, and nucleocapsid proteins), after which they did subcellular fractionation and found that the nuclear fraction was enriched for spike protein, but not just the spike protein. They found Nsp1, Nsp5, Nsp9, Nsp13, Nsp14, and Nsp16 proteins were localized in the nucleus too.
This result brings up another issue. This study is not the first study to look at the subcellular localization of SARS-CoV-2 proteins. (Indeed, figuring out where in the cell these proteins go would be one of the obvious early experiments to do, and, of course, it was done.) Normally, when you get a result like this, you’re supposed to compare your results to past results. If your results agree with past results, great! If they don’t, you really do need to explain why you think you got different results and possibly even do experiments to figure out why there was a discrepancy.
So here’s the older study, which found:
As can be seen in the Supplementary Fig. S2, the viral proteins are either cytoplasmic (NSP2, NSP3C, NSP4, NSP8, Spike, M, N, ORF3a, ORF3b, ORF6, ORF7a, ORF7b, ORF8, ORF9b and ORF10) or both nuclear and cytoplasmic (NSP1, NSP3N, NSP5, NSP6, NSP7, NSP9, NSP10, NSP12, NSP13, NSP14, NSP15, NSP16, E and ORF9a).
There’s a hell of a disagreement there. The previous study found that spike protein was not nuclear and was only cytoplasmic, but the current study found spike in the nucleus. If I were a reviewer, I would have insisted on an explanation or at least that the authors acknowledge the difference between their results and previously reported results and at least speculate as to why they found something different. Seriously, one has to wonder about their the methodology used.
The authors also used specific assays to test DNA damage repair in response to radiation and other DNA damaging agents. This is important, because what gets lost in the antivax ranting about this study is that the authors found that several SARS-CoV-2 proteins inhibited DNA damage repair. For instance, in addition to spike protein, Nsp1, Nsp5, Nsp13, and NSP14 inhibited a type of DNA repair measured by non-homologous end joining (NHEJ). (You don’t need to know the details for purposes of this discussion.) Similarly, Nsp1, Nsp13, and Nsp14 inhibited a process known as homologous recombination (HR), which is important in DNA damage repair. True, in the authors’ assays, spike protein was the most potent inhibitor of these processes, but several other SARS-CoV-2 proteins that supposedly locate in the nucleus inhibited them too. Indeed, Nsp1 inhibited NHEJ almost as potently as the spike protein. One notes that none of these proteins are targets of COVID-19 vaccines and could only theoretically cause problems during COVID-19 infection. In fairness, the authors note that these proteins also inhibit cell replication and that their apparent effect on DNA damage repair as shown in the assays could be secondary to that, but you know what? They should have done further experiments to see if that was indeed the case!
Did I mention again that this is just a single in vitro study done in only two related cell lines whose relevance to the in vivo situation is not shown? Also, no experiments here show that spike actually interacts with the DNA damage repair proteins postulated, just that the levels of various DNA damage repair proteins remained unchanged by cranking up the expression of spike protein.
Other problems that I’ve seen pointed out include a distinct lack of proper controls, as noted by an actual expert, who noted that the assay lacked positive and negative controls:
Given the lack of followup on the results for the other SARS-CoV-2 proteins supposedly interfering with the VHEJ and HR assays to clarify the mechanism, plus the questions raised about the relative lack of proper controls, I, too find this paper more or less uninterpretable.
I could go on, but it’s unnecessary for the reasons I listed above. Even if this paper is 100% accurate about the effects of spike proteins on DNA damage repair, even if spike does get into the nucleus and interfere with the DNA damage repair proteins and pathways implicated, it does not demonstrate that the spike protein made by COVID-19 mRNA vaccines gets into the nucleus and interferes with DNA damage repair mechanisms, and even if vaccine-produced spike protein did that the very worst thing that could result might be tumors at the injection site. Think of it this way. If the results of this paper actually did translate to humans, it would be a very strong argument not against vaccination for COVID-19 but for being vaccinated. That’s because, if these results were to hold true in experimental animals and then humans, they would imply that the massive amounts of spike protein made by SARS-CoV-2 infection in cells in organs all over the body, particularly the lungs, would lead to a hugely increased risk in future years of potentially deadly cancers in a number of organs, rather than a very small potential risk of a tumor at the vaccine injection site, something incredibly unlikely to be life-threatening. And that’s leaving aside the methodological weaknesses of the paper that call into question its results, plus the fact that it’s just an in vitro study with one cell line.
Finally, it’s been observed that before the pandemic and social media’s ability to amplify and misuse science like this, a paper like this would likely have been published and read by few people, influencing no one but a subset of scientists interested in the questions raised. A few with a special interest in DNA damage repair might have used it as a starting point to carry the research forward with other experiments, but only a few, and the authors’ wild and unfounded speculation about how their results suggest that current COVID-19 vaccines might be unsafe and need their design rethought correctly ignored as unjustified extrapolation from results that are at best preliminary and at worst just plain wrong.
That was then. This is now
In the age of the pandemic, this study has been accessed over 350,000 times, and the video to which I linked above has been accessed over a million times. Another video discussing the study and making many of the same mistakes has been accessed nearly a million times. Meanwhile on Twitter, the Tweets above from researcher Jeremy Stark are the only Tweets I could find pointing out the flaws in the study, in contrast to the hundreds of Tweets touting this study as “evidence” that COVID-19 vaccines are going to cause cancer because the spike protein gets into the nucleus and interferes with DNA damage repair. (I included six such Tweets just to make that point, but there are more than I have time to read, and the number has been climbing ever since this study went viral a few days ago.) Scientists need to be a lot more careful about their speculations in their papers, and reviewers need to call them out on it when they do engage in speculation that doesn’t flow from the results of a study. Whereas before studies like this would have languished in justified obscurity, now they provide grist for the antivaccine movement to spread fear, uncertainty, and doubt about vaccines. It’s even worse than that, though. These days crappy preprint studies that haven’t even passed peer review can be weaponized. We need to do better.