Unsupported claim about COVID-19 vaccines, anti-spike antibodies, and myocarditis misrepresented conclusions from a New England Journal of Medicine article
Both SARS-CoV-2 infection and COVID-19 vaccines induce the body to produce antibodies against the spike protein carried by the virus. In addition, the body may also produce antibodies against other antibodies—a certain group of such antibodies are called anti-idiotypic antibodies. Such antibodies could potentially bind to antibodies against the spike protein. However, it is still unknown whether anti-idiotypic antibodies that bind to antibodies against the spike protein are actually produced in the body, and if so, whether they have any consequences on health.
Misrepresents source: The claim is based on a New England Journal of Medicine (NEJM) article. However, the NEJM article doesn’t support the claim. The article built a hypothesis about how a type of antibodies (anti-idiotype antibodies) that could bind to antibodies against the spike protein produced by vaccination or infection could result in myocarditis, using previous studies about other viruses than SARS-CoV-2. But it doesn’t contain experimental data demonstrating that this does indeed occur in the case of COVID-19 vaccines.
In December 2020, the U.S. Food and Drug Administration authorized the emergency use of three COVID-19 vaccines, deeming their benefits to largely outweigh the risks. 59.7% of the U.S. population have since been fully vaccinated and research indicates that vaccination against COVID-19 saved hundreds of thousands of lives.
Over the course of the vaccination campaign, there have been many allegations that question the vaccines’ safety. Health Feedback explained several times how such claims commonly conflated the notions of correlation and causality, or misunderstood how adverse event databases work.
In November 2021, a post on the website Gateway Pundit claimed that a scientific article published in the New England Journal of Medicine (NEJM) proposed a mechanism by which the “antibodies that people produce against the coronavirus spike protein may lead to myocarditis”. This claim originated from a Substack article written by journalist Alex Berenson, who has repeatedly published inaccurate information about COVID-19 and the vaccines.
But as we will explain below, this claim is baseless and misrepresents the content and scope of the NEJM article. To understand why, we must provide context to explain what the article reported.
In the NEJM article, the authors commented on the effect of anti-idiotype antibodies on health and why it could be interesting to study them in the context of SARS-CoV-2 infection or COVID-19 vaccination.
Broadly speaking, anti-idiotype antibodies are antibodies that recognize and bind to other antibodies. For example, the presence of the spike protein in the body due to SARS-CoV-2 infection or COVID-19 vaccination will trigger the body to produce anti-spike antibodies. But the body can also produce antibodies against other antibodies, which may then bind those anti-spike antibodies themselves.
Therefore, in order for the body to produce anti-idiotype antibodies, there are two conditions that need to be fulfilled in theory: we first need to have the pathogen protein in the body—in this case the spike protein—followed by the production of anti-spike antibodies by the immune system. The body can then produce antibodies against the anti-spike antibodies.
The authors of the NEJM article explained that anti-idiotype antibodies can affect some physiological functions of the body. For example, anti-idiotypes antibodies arising after infection by the enterovirus coxsackievirus B3 in mice can cause myocarditis. This is of interest, given that myocarditis is a possible rare side effect of COVID-19 vaccination, with a low incidence of 21 to 27 cases per million of people who received at least one dose of vaccine.
Thus, the authors of the NEJM article suggested that it would be worth investigating whether this side effect could be related to the presence of anti-idiotype antibodies due to the spike protein and anti-spike antibodies from COVID-19 vaccination.
To determine the credibility of the claim, Health Feedback reached out to the authors of the NEJM article, who rejected the Gateway Pundit’s claim. William Murphy, one of the authors and Professor of Dermatology at the University of California Davis stated:
“We only made the claim that the concept of anti-idiotype responses, due to their unique nature, may account for some of the long-lasting or seemingly off-target effects being reported of both infection or vaccine application and should be considered. That is the only conclusion that should be drawn from the article.”
In an email to Health Feedback, Dan Longo, Murphy’s co-author and professor of medicine at Harvard Medical School and deputy editor of the NEJM, also refuted this claim.
It’s important to keep in mind that the NEJM article is not a research article, as it didn’t present experimental or clinical evidence. The article gathered bibliographical information about the effect of anti-idiotype antibodies in the context of infection by other viruses and hypothesized that it could happen in the context of SARS-CoV-2 infection or COVID-19 vaccination. In other words, this article aimed at suggesting a possible avenue of future research, but didn’t provide data that supported the claim, as this wasn’t within the scope of the article. Therefore, the interpretation of the article given by Gateway Pundit is false.
Second, the Gateway Pundit claimed that vaccines would be more dangerous than infection because “the spike protein antibody levels are MUCH higher following vaccination than infection”. Some data indeed suggest that vaccination leads to higher levels of anti-spike antibodies than infection[3,4]. However, this statement is confusing as it might mislead people into thinking that the high level of anti-spike antibodies produced by vaccines can cause myocarditis. However, these aren’t the type of antibodies discussed in the NEJM paper. The article only proposed that anti-idiotype antibodies would be worth investigating as a possible cause of myocarditis. Whether anti-idiotype antibodies are produced upon COVID-19 vaccination, and whether it so happens at a higher rate than following SARS-CoV-2 infection, and whether this could increase the likelihood of myocarditis is unsupported by available data.
The only way this would make sense is if a higher level of anti-spike antibodies led to a higher level of anti-idiotype antibodies. However, there is no data supporting such a possibility.
In that respect, Murphy declared:
“We made ZERO claims regarding the amount of spike antibody levels after vaccination much less compared to after infection. These studies simply have not been performed to allow such a statement to be made due to the extremely variable kinetics of antibody induction, heterogeneity of the population, variable extent of infection (and therefore antigen load), differences in vaccines, and even when assessed. There is no robust evidence to support this claim and we do not even suggest that in the article.”
In summary, the Gateway Pundit misrepresented the conclusions of an article proposing a new avenue of research and inaccurately framed them as pieces of evidence of a causal association between anti-spike antibodies generated by COVID-19 vaccination and myocarditis. While there is an association between myocarditis and certain COVID-19 vaccines, there’s not enough evidence to conclude that this is due to anti-spike or anti-idiotype antibodies, as the article claimed.
Professor, University of California Davis
I can state without equivocation that:
1. We made ZERO claims regarding the amount of spike antibody levels after vaccination much less compared to after infection. These studies simply have not been performed to allow such a statement to be made due to the extremely variable kinetics of antibody induction, heterogeneity of the population, variable extent of infection (and therefore antigen load), differences in vaccines, and even when assessed. There is no robust evidence to support this claim and we do not even suggest that in the article.
2. We only made the claim that the concept of anti-idiotype responses, due to their unique nature, may account for some of the long-lasting or seemingly off-target effects being reported of both infection or vaccine application and should be considered. That is the only conclusion that should be drawn from the article.
- 1 – Murphy & Longo (2021) A Possible Role for Anti-idiotype Antibodies in SARS-CoV-2 Infection and Vaccination. The New England Journal of Medicine.
- 2 – Paque & Miller (1991) Autoanti-idiotypes exhibit mimicry of myocyte antigens in virus-induced myocarditis. Journal of Virology.
- 3 – Ebinger et al. (2021) Antibody responses to the BNT162b2 mRNA vaccine in individuals previously infected with SARS-CoV-2. Nature.
- 4 – Sariol et al. (2021) Function Is More Reliable than Quantity to Follow Up the Humoral Response to the Receptor-Binding Domain of SARS-CoV-2-Spike Protein after Natural Infection or COVID-19 Vaccination. Viruses.