Aaron Siri is misrepresenting a perspective piece in NEJM authored by Daniel A. Salmon, Walter A. Orenstein, Stanley A. Plotkin, and Robert T. Chen concerning vaccine safety studies. You can read the perspective via these screenshots (it's paywalled). Some comments from me 🧵:





The gist of the perspective is that vaccine safety studies take a lot of time. The US currently has a means to compensate those harmed by vaccines via the NVICP, and the budget for the NVICP comes from an excise tax on vaccines, and it has a large surplus at the moment.

One thing on my mind more in recent times is the way that we've managed to make health, and in particular public health, feel normative- and in some ways that's a double-edged sword.

Undeniably, it's far more positive than it is negative. If someone drinks milk and gets paralyzed from Guillain-Barre syndrome today, that is a shocking event.

If someone takes a medication and develops lead poisoning from it being tainted, that is a shocking event.

We now have the beginning of the COVID-19 vaccines session. First a review of the epidemiology of COVID-19 hospitalizations:


Age continues to be a key risk factor for hospitalization, though rates have declined markedly over time. cdc.gov/vaccines/acip/…
Importantly, children <6 months old have the second highest hospitalization rate by age second to those over 75, underscoring the importance of immunization in pregnancy. There are also racial disparities in hospitalization rates.

I'm seeing some people read this correspondence in NEJM and walk away with the takeaway that pasteurized milk carries some meaningful risk of having replication-competent avian influenza (H5N1) and... that really isn't my takeaway:
nejm.org/doi/full/10.10…
Here's the part that has people concerned: basically, the correspondence suggests that if levels of virus in the milk are *very* high, some tiny bit of it might remain functional after pasteurization by the high temperature, short time (HTST) method, suggesting an infection risk.

The VRBPAC meeting to discuss the strains to be included in the upcoming season's COVID-19 vaccines is starting:
fda.gov/advisory-commi… It wouldn’t be a VRBPAC meeting unless committee members asked questions at a time that it didn’t make sense to ask a particular question and then spend 10 minutes discussing the discussion of a question to be answered later on

Very cool new preprint provides insight into how we might improve mRNA therapeutics (and maybe vaccines?) by explaining how lipid nanoparticles cause inflammation:


Thread 🧵biorxiv.org/content/10.110… Context: Most people think about mRNA-LNPs (lipid nanoparticles) today is probably with vaccines, but originally mRNA was intended primarily as a therapeutic. The problem here, it became apparent, was that mRNA is a very potent inducer of inflammation.

ICYMI (I did) Moderna did share some data for mRNA-1283, their next-generation COVID-19 vaccine, in their most recent Vaccines Day presentation:


The results are interesting. Thread🧵s29.q4cdn.com/435878511/file… mRNA-1283 was studied in a Phase 3 trial of 11,000 people randomized 1:1 to receive either mRNA-1283.222 or mRNA-1273.222 (the BA.4/5 bivalent). 1283 contains just the RBD of the spike protein instead of the entire thing. The idea here is this will help the immune response to...

People are asking questions about this paper:


Here are my 2 cents on the matter 🧵nature.com/articles/s4146… Firstly, we need to get our terminology straight. When I use the term imprinting, I am referring to the tendency of the immune system to rely on its memory cells instead of recruiting naive cells in response to a new antigen.

Berenson also whined about imprinting, which is an issue that actually deserves some attention, if only because it has been explained very poorly to the general public and there’s actually a lot of nuance to it. Here’s a thread to explain it 🧵 Imprinting (aka original antigenic sin, antigenic seniority, primary addiction) is probably most readily explained through an example.

Suppose I encounter strain X of some virus. I make an antibody response with memory B cells and plasma cells and I survive my infection. Some time passes and now a new strain of this virus, strain Y, has appeared. My immune system still has memory from its encounter with strain X of this virus, and some of the antibodies and memory B cells can recognize strain Y. The question now is do I bother making a strain Y-specific response from scratch, or do I just tweak the responses I have saved from strain X? By virtue of the fact that with infections time is of the essence and memory responses can go much faster than new responses from naive cells, the immune system will typically choose to tweak the response from X than to start from scratch against Y.

That choice- to use the memory the immune system already has rather than to make a brand new response from naive cells- is known as imprinting.

There's a new study about why some people might get long COVID that recently came out and it actually seems very promising:

News and views for it here:


A thread explaining the work 🧵nature.com/articles/s4159…
nature.com/articles/s4159… Before I get into this study though, I need to explain a bit about human biology. Iron is a really important micronutrient, essential for many processes including oxygen transport by hemoglobin. However, iron is also very important for bacterial infections in particular and bacteria really try to make use of our iron stores whenever they can so that they can reproduce and cause disease.

Because of this, when we experience acute inflammation, like from an infection, our body has a tendency to sequester iron so that bacteria cannot access it. If this occurs persistently, it can cause a drop in hemoglobin levels and red blood cell levels, at which point it is known as anemia of (chronic) inflammation or anemia of chronic disease.

A few players are really important in this process. One of these is a hormone made by the liver called hepcidin, which inhibits a protein called ferroportin, responsible for importing iron into the circulation or into cells. The other is interleukin 6 (IL-6), which causes the liver to make hepcidin. The effect of this is that we have a much harder time absorbing iron from our diet and the iron that we do have ends up trapped inside specialized macrophages that recycle iron by taking up aged red blood cells. This inflammation also reduces levels of transferrin, the iron transporting protein. With persisting inflammation however, the response eventually also makes it harder to make new red blood cells by inhibiting the effects of the hormone erythropoietin, which normally stimulates red blood cell production. Depending on the specific reason for the inflammation, red blood cell lifespan may also be shortened. Taken together, these conditions create a state of iron starvation in our body.

So in summary: persistent inflammation can cause iron to end up trapped in our macrophages and reduce iron absorption from diet, which results in a state of iron starvation.

See reviews here for more details:


nejm.org/doi/full/10.10…
ashpublications.org/blood/article/…
ashpublications.org/blood/article/…

There's a new preprint out that has some really important findings regarding the immunity elicited by mRNA vaccines:


But... it's more complicated than it seems.medrxiv.org/content/10.110… As the preprint writes, the holy grail of vaccinology is the generation of long-lived plasma cells (LLPCs). These are antibody-secreting cells (ASCs) that persist for decades, if not life, ensuring the constant presence of antibodies for their lifetime. Classically, we think of these as living in the bone marrow (at least the ones that make IgG antibodies) where they take up residence in survival niches that supply them with cytokines and nutrients that they need to focus exclusively on their job making antibodies.

The media is doing an atrocious job reporting on this study and it's fomenting panic completely inappropriately:


A thread on what this study actually findssciencedirect.com/science/articl… Using an international collaboration and multiple linked healthcare databases, the study team was able to investigate the safety of COVID-19 vaccines in 99 million people. A sample of this size lets you see changes in the risk of adverse events even if they are *extremely* rare.

In today's installment of "should I credulously take the claims of repeatedly proven liars at face value?" I present: blood donation after vaccines. A claim is circulating alleging that @RedCross isn't letting you donate blood if you've been vaccinated. Is it true? Well, if you spend 3 whole seconds looking at eligibility requirements and vaccination on their page you will see this:


So, in other words, if you have received a live attenuated vaccine for COVID-19, you should defer blood donation for 2 weeks. Why? redcrossblood.org/faq.html#eligi…

A new preprint suggests that the XBB.1.5 mRNA boosters might be special in terms of the antibody response:


Brief threadbiorxiv.org/content/10.110… This preprint compared the contraction of antibody titer after XBB.1.5 mRNA vaccines and titers to other boosters. There are multiple noteworthy findings. Firstly, in those with known hybrid immunity in this cohort, in addition to the markedly higher titers, the highest titers are no longer against the ancestral spike.

I'm not sure about the extent to which this needs to be discussed but in the interest of caution: we absolutely know beyond any reasonable doubt that HIV is the cause of AIDS. A thread🧵 In 1981, a cluster of Kaposi's sarcoma (KS; a cancer associated with immunodeficiency) and invasive infection by the fungus Pneumocystis carinii (now known as Pneumocystis jirovecii) was observed among gay men:

cdc.gov/mmwr/preview/m…
cdc.gov/mmwr/preview/m…

New MOMIVAX data: Maternal COVID-19 vaccination protects infants for the first few months of life. Infants born to boosted mothers were 56% less likely to contract COVID-19 in the first 6 months of life compared to unboosted.
publications.aap.org/pediatrics/art… The extent to which the antibodies matched the circulating variants was relevant: each 10-fold increase in neutralizing titer against D614G, and omicron BA.1 and BA.5 at delivery were associated with a 30%, 46%, 56%, and 60% risk reduction, respectively.

We have a new preprint looking at the antibody responses to the XBB.1.5 mRNA vaccine boosters:


Overall, it's good. Here's what it says 🧵biorxiv.org/content/10.110… First, neutralization: before the XBB boosters, a substantial proportion of people have undetectable neutralization against multiple current variants, including JN.1. After the booster, nearly everyone can neutralize them. The fold increase is also greatest for novel variants.

Formaldehyde is present in some vaccines as a trace residual in quantities ~100x lower than that which occur as a natural byproduct of human metabolism:


The doses are too low for any biological effect, let alone leukemia. sciencedirect.com/science/articl…
The rise in cancer incidence in children reported by the screenshot is driven by ALL:

In part this reflects better ascertainment by cancer registries. However, comparing the risk by vaccination status shows no relationship:
nature.com/articles/66039…
academic.oup.com/ije/article/46…

So this preprint has some incredibly meticulous work on SARS-CoV-2 serology:


It also shows that vaccination can elicit antibodies (or at least... their precursors) that protect against ALL SARS-like viruses!

A thread🧵biorxiv.org/content/10.110… Hybrid immunity (the state of immunity that occurs when a person has been vaccinated and infected, in either order) consistently demonstrates the most robust protection currently achievable against SARS-CoV-2:
thelancet.com/journals/lanin…

More new data from Novavax vs. COVID-19, this time in nonhuman primates. It looks really good- mostly. Let's take a look:


Thread🧵nature.com/articles/s4154… Novavax's COVID-19 vaccine comprises Spike (5 micrograms) with prefusion stabilization and deletion of the furin cleavage site with truncated glycans formulated in nanoparticle rosettes around the Matrix-M adjuvant (50 micrograms):

science.org/doi/10.1126/sc…
nature.com/articles/s4154…

Novavax has released a new preprint looking at the profile of antibody effector functions following repeated doses compared with mRNA vaccines. It has useful data but it's prompted some IMO hasty generalizations:
medrxiv.org/content/10.110… Context: in addition to direct neutralization of SARS-CoV-2 by blocking the interaction between the spike protein and the cell's receptor (usually ACE2) that enables the virus to enter the cell, antibodies have many additional effector functions which can contribute to...