Thread by @dgurdasani1

Important paper out in Science TLDR;
-possible immune imprinting (where being exposed to a previous variant could means impaired protection against a new variant)
-hybrid immunity (infection + 3-dose vax) not always better than vax
-T cell response v. impaired with omicron🧵

This was a study of healthcare workers. It looked at HCWs who had had mild/asymptomatic infection during different variant waves or had remained infection-free and been vaccinated (3 doses) - to look at their immune response to different variants based on their history.

These were the histories looked at:
-no infection + 3-dose vax
-original (Wuhan strain) inf + 3-dose vax
-alpha + 3-dose vax
-delta + 3-dose vax
-3-dose vax + omicron
-Wuhan + 3-dose vax + omicron (re-infection)
colours match those in plots below

First, let's talk about overall immunity to omicron- irrespective of past exposure. Antibody response irrespective of history *much much* lower against omicron than for Wuhan. The fig below shows the antibody response to Wuhan (top panel) & to the omicron variant (bottom panel).

The blocks left-to-right are response before vaccination, after 1 dose, 2-3 wks after 2 doses, 20-21 wks after 2nd dose (waning), 2-3 wks after 3rd dose, 14 wks after booster in HCWs. The colours are as above- blue is no prior infection, red is prior Wuhan, green is prior alpha.

So what do we see?
Look at the first block in the top and lower panel- this is prior to vaccination responses to Wuhan (top panel) and omicron (bottom) among those unexposed (blue), and previous Wuhan (red) or alpha (green).

Prior infection with Wuhan or alpha is associated with higher antibody response to Wuhan prior to vax (red and green boxes, top panel, leftmost box)- compared to those with no exposure or vax. We don't see this with omicron. No response seen even in those with prior infection.

Then we look at 1st dose (2nd box left-to-right). After the 1st dose, antibodies to Wuhan & omicron are higher in those with prior Wuhan/alpha exposure (red & green boxes higher than blue in both panels in 2nd box). So we see hybrid immunity but Abs much lower against omicron.

Then 2-3 wks after the 3rd dose, antibody response becomes at par (but not higher) among those with no previous infection compared with those with any prior variant infection (3rd box- three boxes at same level).

But immunity wanes very differently against Wuhan and omicron based on prior exposure (4th box left-to-right). Almost all (19/21) HCWs infected during the alpha wave no longer showing detectable antibody against omicron (2nd panel), but HCWs infected during the Wuhan wave did.

So what happens with boosting? 2-3 wks after boosting, again brings everyone on par regardless of past infection (5th box left-to-right). But again no increase in antibody among those with prior infection of any time compared to those who were unexposed and boosted.

But 14 wks down the line, those with previous infection with Wuhan, or previous infection with Wuhan followed by omicron have *much lower* response than those infected with *omicron alone* without any prior infection.

So Wuhan + 3-dose vax + omicron does not boost your response to omicron in the way 3-dose-vax + omicron does. In fact the response remains the same as those who had prior infection with the Wuhan variant without re-infection with omicron (pink and red, last box)

As the authors say, this indicates "that prior Wuhan Hu-1 infected individuals were immune imprinted to not boost antibody binding responses against B.1.1.529 (Omicron) despite having been infected by B.1.1.529 (Omicron) itself"

What does this mean?
It means that it's possible that a persons early exposure may partly determine future response to other variants. So an immune system may still respond to the original variant because it's been 'primed' to do so but not the one it needs to be responding to.

This is important to understand further, because it means that infection + vaccination doesn't necessarily improve your immune response. If an early infection was with a specific variant different from the current one, it could even dampen the response to the new variant.

This fig summarises it. It shows that overall response to omicron is much lower than to other variants *even if you had prior infection with omicron + vax*. And if you had prior Wuhan infection + vax, then omicron doesn't boost like it does if you didn't have previous Wuhan.

But what about T cells? We've been told that they're a lot more robust to virus adaptation and escape. Unfortunately, this study shows that T cell responses are also massively affected with omicron.

The study shows that 14 wks after the 3rd dose (9/10, 90%) of triple-vaccinated, previously infection-naïve HCW showed no cross-reactive T cell immunity against B.1.1.529 (Omicron) S1 protein (rightmost box here - most responses to omicron undetectable)

None (0/6) of HCW with a previous history of Wuhan Hu-1 variant responded to omicron S1 protein - *even if they were then infected with omicron after*. So omicron infection was unable to boost T cell immunity against itself among those with a previous history of Wuhan infection.

And even in those who had only 3-dose vax + omicron breakthrough (so no prior Wuhan type infection) - the T cell response to omicron although higher than those with other histories of infection- was *much* lower than responses to other variants.

So even omicron breakthrough didn't boost protection much against omicron. Cross-reactivity to other variants was much better than to omicron even among those with omicron infection. Omicron seems to evade T-cell response quite well.

More and more evidence suggesting that SARS-CoV-2 seems adept at evading T cell restriction, and has multiple mechanisms to do so.

https://twitter.com/VirusesImmunity/status/1522939444856193024?s=20&t=wcRXSAJe6bm5XVqxuZyTfw

This is an extremely important study- suggesting that 'imprinting' by previous variants may alter and even impair protective immunity against subsequent variants- a process described by the authors as - 'hybrid-immune-damping"

We do need to understand this better. These are laboratory studies, so we need to see how this affects protection from infection & severe disease among those with different past infection profiles. Neutralising antibodies have shown strong correlation with clinical protection.

Vitally important that going forward clinical studies try to stratify outcomes by histories of past infection + vaccination, to see if we see this reflected in vaccine protection outcomes as well.

Study here:
science.org/doi/10.1126/sc…

And also important to reflect on the narrative that 'hybrid immunity' is always better, and infection is a 'good booster' to vaccination. Not only does infection come with serious risk of long COVID & severe illness (despite boosting), it may impair response to future variants.

And also time to reflect on overreliance on T-cell protections that seem at least in the laboratory very significantly impacted against omicron, with very little T-cell response seen against omicron even in those boosted, and previously infected.

I want to really congratulate the authors on this amazing and vitally important piece of work. I don't think we'll see this widely covered in the media, but it's one of the most important pieces of research to emerge during the pandemic, and so carefully and elegantly done.

Sadly, anything that has nuance is lost in scientific discussions & communication. But I hope this will at least make some in the scientific community who've been pushing infections as 'boosters' & suggesting T cell protections are enough rethink.

I'm sure I'll be attacked for this- I'm already mentally prepared. I would say if you're going to attack me please try to challenge me on what I've actually said and what the paper shows rather than launch ad-hominems because you don't like the message here.

Also, this's a complex piece of work. I had to read it a few times to ensure I understood it correctly. While I've studied immunology, I'm not a specialist. I've tried to keep msging close to what the authors say, but if I've got something wrong happy as always to be corrected.