Ok, I know promised a thread on the fascinating and important new research explaining what happened with the rare, serious side effects of AstraZeneca's and Johnson&Johnson's Covid-19 vaccines.

First a caveat: When I was studying molecular biomedicine two of the more complex processes I learnt about were the immune system and the clotting system. What we’re talking about here is right at the intersection between the two, so uhmm bear with me.

So what I will do for now is just describe what the research suggests happened in the patients. I’ll try and do a thread on how they actually did the science, which I just find very satisfying, in the coming days. No promises this time.

Adenoviruses are common cold viruses that circulate all year round and come in dozens of different types. The two vaccines we're talking about here used adenoviruses to shuttle the gene for the Spike protein on SARS-CoV-2 into human cells so the body would produce it and mount an immune response.

(Johnson and Johnson used a human adenovirus type 26 and AstraZeneca used a chimpanzee adenovirus. There were other Covid-19 vaccines too that used adenoviruses, for instance Russia’s Sputnik V. We’ll ignore all that for now.)

The first thing that happened, happened before any immunizations: The people who later developed VITT (the side effect we are talking about here with thromboses in unusual places AND low platelets leading to bleeding) likely had an adenovirus infection at some point.

One response of our immune system to infection is for our B cells to start producing antibodies, Y-shaped molecules that can recognize invading microbes and (amongst other things) bind to them, keeping them from infecting other cells and signaling to the immune system to attack.

Now this is where it gets a little tricky.
Different people produce different antibodies for a few reasons. One is that the instructions for building antibodies in our genome are not straight forward. Instead, we have a set of genetic segments that are mixed and matched when a B cell is created.

Imagine a deck of cards from which you pick a few cards and shuffle them. That is essentially how the part of a B cell's genome that carries the instructions for antibodies is created.
It's a fascinating mechanism and one reason we can produce such a huge array of different antibodies.

But different people also have different versions of that starting deck of cards that B cells mix and match from.
And only people who have one of two particular cards in their decks seem to develop VITT. (The scientific name of these "cards" is IGLV3-21*02 and IGLV3-21*03...Gotta love scientists).

Now, when people were vaccinated, the B cells that had previously been activated against adenovirus were reactivated. And now on top of all the existing variation, the body produces lots of copies of those B cells and introduces little mutations to maybe find an even better antibody.

(Sorry, I'm realizing this is getting reaaaallly long, but you gotta love the amazing ways that evolution has honed our body's ability to produce so many slightly different molecules to ensure that no matter what microbe nature throws at us we'll be able to produce a matching antibody...)

Amongst the B cells being reactivated were B cells that produced antibodies recognizing a particular protein of the adenovirus called pVII.
And in some people, who had one of those two cards I mentioned, so IGLV3-21*02 or IGLV3-21*3, one tiny change flipped what these antibodies recognize.

These people had a particular antibody that recognized pVII and in the process of generating mutations in B cells a single amino acid at position 31 switched from a lysine (which is positively charged) to either an aspartic acid or a glutamic acid (both of which are negatively charged).

This tiny shift changed what the antibody binds to and suddenly it was not binding to pVII but do PF4, an important protein in our blood clotting system. That led to complexes forming of antibodies with PF4 and those complexes activate platelets (thrombocytes) that then release more PF4. And so on.

The result is both blood clotting and the depletion of platelets that are needed to stop bleeding elsewhere leading to the really striking symptoms that patients with this rare side effect showed.
In our story we simplified the NEJM graphic a little to show what happens.

The good news (apart from this thread almost being done):
There is an easy treatment. Doctors can give IVIG (intravenous immunoglobulin). All that means is injecting the patients with other antibodies, essentially flooding their system with a mix of antibodies.

The reason that works? Platelets are activated when the antibodies that are in complexes with PF4 bind to the platelets. But when there are lots of other antibodies that already occupy the binding sites on the platelets then the complexes cannot bind and the platelets are not activated.

It's a little like an asshole billionaire buying up all the houses in his neighborhood so no-one can move in who might aggravate him. Only when doctors do it for our immune system it saves lives...

All of this explains why VITT was so rare and why it could not be picked up in the trials. A lot of things had to come together. The right kind of genetic background with the right kind of antibody and at the end one particular mutation.
There may be other factors too that we don't understand yet.