Thread by @dgurdasani1

Myth busting thread on naturally immunity from infection in children. One of the arguments used some, including members of JCVI have been high exposure in kids, which means 1 dose may be adequate. The evidence actually suggest poor neutralisation in kids & rapid waning of Abs 🧵

First, many people don't seroconvert after infection, and with young people this proportion can be even higher. This means that they may not have detectable Abs after infection. Ab levels generally do correlate with protection, so absent antibodies are problematic.

Of course, we cannot rule out T cell responses in these children, but in the absence of clear evidence showing these are present, we cannot assume this, as some appear to have done.

Recent study from the CDC showing non-seroconversion (did not develop Abs 3 weeks after confirmed infection) in 36% of the cohort studied. Those who didn't develop antibodies were on average *10 years younger* than those who did (included adults 20+).
wwwnc.cdc.gov/eid/article/27…

Even when antibodies are present, neutralising antibody responses measured against pseudovirus were lower in hospitalised children compared to adults who donated plasma, or developed severe COVID-19. Here is a Nature Immunology study that showed this.
nature.com/articles/s4159…

And when Abs are present, they seem to wane fairly quickly in children- possibly because children are more likely to have mild or asymptomatic acute infections. Here's the school serology study from England showing 50% of children losing Abs in 4 months:

https://twitter.com/dgurdasani1/status/1437705606492827649?s=20

Another study looking at antibodies among infected children compared to their parents. It's very clear that children had rapid waning of immunity by 180 days, with much lower immunity at 6 months compared to their parents.

Even worse is emerging evidence suggesting long COVID in children is associated with lower levels of Abs during acute infection. V. problematic for studies that have used serology to define those with exposure to study long COVID! h/t @ahandvanish

medrxiv.org/content/10.110…

Similar in hospitalised patients. Ab levels predict long COVID, so antibody positivity *should not* be used to assign people 'exposure' or 'caseness' when examining long COVID, because this will just lead to misclassification & underestimating prevalence.

https://twitter.com/ahandvanish/status/1377041277653504000?s=20

In summary:
- young people tend to have lower seroconversion at acute stage
-Abs in children less likely to be neutralising than adults
-Abs in children wane faster than in adults
- we don't know about T cell immunity, but we cannot assume children protected without evidence

And even if kids had Abs, let's be very clear- this doesn't always mean protection against delta, which is able to escape even from immunity to previous lineages to an extent. Vaccines provide immunity through a safe mechanism, and boost immunity for those with prior infection.

So everyone talking about 'naturally immunity' in the 'majority of children', or natural immunity in children being sufficient and them needing either no or 1 dose of vaccination - please cite evidence for these claims, or stop pushing unevidenced misinformation.

Please feel free to link this thread to anyone who uses these unevidenced narratives. The problem is that these will likely lead to vaccine hesitancy, despite being completely without evidence. It's disappointing to have seen these repeated again & again my members of JCVI & PHE.

And also all those talking about rigour and case-control studies in long COVID- please take a moment to consider how you would minimise misclassification among controls, given the sheer number of children infected and the low pick up from serology *and* PCRs in kids.

Unless you're able to discuss the limitations of these case-control analyses in a sensible way, please don't pretend these are rigorous just by virtue of having a control, without any consideration of what that control is.