Thread by @mark_toshner

This paper on thromboinflammation is everywhere this week. Unfortunately nobody commenting on it seems to have read it!

I have, and it definitely doesn't say what commentators are suggesting. 🧵nature.com/articles/s4158…

Quick summary. Spike interacts with fibrinogen and drives a thromboinflammatory cascade. In the paper, this is tied most closely to lung & brain pathology. I am not going to dissect the figures- take it on face value as true; this is not my problem with the paper/messaging.

My problem is the explicit link to long covid present in the paper (from the abstract on) & prominent/signal boosted in most of the media.

This is how most big SM commentators have described the paper. Note they are all explicit about long covid.

and more, the list goes on. Publically then this has become a story about long covid.

So what is the paper? It is a purely mouse paper in ... exclusively acute models. I couldn't find a model over 7 days. There is no long covid data here. The long covid hypothetical rests on the concept that there is very common viral (and intact spike) persistence in long covid.

That hypothesis rests on the shoulders of this paper in 31 patients with PASC/long covid
ncbi.nlm.nih.gov/pmc/articles/P…

Here is the relevant data where there is no significant S1 subunit but reactivity to full length spike. I can’t see this work validated anywhere else but I might be wrong. BTW it looks like spike binding in the Nature paper is in the S1 subunit.

I don't know what their antibody is reacting to and certainly might be non-S1 spike but it seems not controversial to say this is exploratory and needs validation and much more work to confirm it is real.

So what we have is an acute mechanism in mice. I don’t doubt the acute clot link (I've published on it, so skin in that game). However what nobody is seeing in clinical practice is unusual persistence of significant non-resolving thrombus-see this 🧵).

https://x.com/mark_toshner/status/1818948926948426178

I also don't doubt a significant story in altered immune mechanisms and a link to complement and thrombotic pathways (different however from "thrombosis").

Serious scrutiny needs to be applied here as clinical trials are being proposed with serious side effects on a hypothesis where we are over-weighting preclinical biomarker & animal data in questionably relevant models and ignoring the actual data from humans.