Thread by @rishirajgoel

Antibodies come and go, but memory cells are forever (maybe not forever, but at least 6 months). Our latest in @ScienceMagazine: science.org/doi/10.1126/sc…

How long does immune memory last after #mRNA vax? Is it effective vs. variants? What about “boosted” responses?

Full 🧵⬇️...

Antibodies are important for protection, but our immune system can also remember viruses through memory B and T cells

We measured all components of immune memory for 6 months after #mRNA vax. Vaccination in people w/ prior immunity also let us study "boosted" responses

#1 - Antibodies:

2-dose mRNA induces high levels of antibodies (blue). Even higher in "boosted" responses (red)

Antibodies decline over time (THIS IS NORMAL AND EXPECTED). But neutralization declines more slowly than binding antibody, suggesting higher quality antibody persists

What about Delta (B.1.617.2)?

We observe that almost everyone in our study still had neutralizing antibodies to Delta at 6 months. Maybe some decrease relative to wild-type D614G spike, but Delta is clearly not as immune evasive as the Beta (B.1.351) variant

#2 - Memory B Cells:

These are the backup plan when circulating antibodies go down. They are also harder to measure...

Inspired by @profshanecrotty @PepperMarion @NussenzweigL & others, we developed a method to quantify SARS-CoV-2-specific memory B cells in the blood

Unlike antibodies, memory B cells targeting Spike and the Spike RBD actually INCREASE over time

And 2 doses of mRNA vaccine get you to similar levels of B cell memory as the much-hyped "hybrid" immunity. We think this is pretty remarkable...

These memory B cells were able to rapidly produce functional anti-Spike antibodies after re-activation that inhibited RBD binding and neutralized Beta/Delta pseudovirus

We then expanded this approach to study memory B cell responses to different variant RBDs and also non-RBD parts of Spike. We included mild COVID-19 samples here to compare vaccine w/ infection

2-dose #mRNA induced memory to all components of Spike, w/ S2 clearly immunodominant

Here is the kicker - we found #mRNA vax generates memory B cells that are better at cross-binding variants than infection at 6 months

Memory from infection evolves more over time compared to vax but doesn't seem to generate as good of an endpoint response vs. Beta (B.1.351)

For a subset of these memory B cells, we could observe their evolution from binding only wild-type RBD to also binding variant RBD. This evolution was associated w/ higher somatic hypermutation, a process that happens in immunological "boot camps" called germinal centers

Data from @TheBcellArtist lab has shown that germinal centers are still active in lymph nodes up to 15 weeks post-mRNA vax: nature.com/articles/s4158…

So it is possible that ongoing germinal center activity after vax may continue to improve the quality of memory against variants

#3 - Memory T Cells:

What about T cells? CD4+ T cells are important for helping antibody and B cell responses. CD8+ T cells are important for killing virus-infected cells

Working with @SetteLab, @markmpainter @divijmathew developed an assay to detect SARS-CoV-2 specific T cells

Vaccination generated durable CD4+ T cell memory at 6 months. CD8+ T cells were also detectable but a bit more variable at memory timepoints in this assay

These data are consistent w/ findings from @Dani6020 summarized here by @profshanecrotty:

https://twitter.com/profshanecrotty/status/1437795168195948547?s=20

With all of this data, we constructed an "immunological map" of vaccine responses

Notice how samples at 6 months cluster far away from baseline pre-immune samples! Even though antibodies decline from peak levels, mRNA vaccines still generate durable multi-component immune memory

For the aficionados - we also investigated relationships between different components of the immune system

CD4+ T cell responses ~2 weeks after the first dose correlated w/ antibody responses out to 6 months, suggesting these cells are important for coordinating long-term memory

Finally, we analyzed what "boosted" responses might look like

Boosting prior immunity w/ vax increased antibody levels via memory B cells but didn't change their decay rate relative to 2-dose #mRNA. And there wasn't much change in the long-term frequency of memory cells

Boosters are a complicated topic. Based on our data, we think antibody recall will extend protection vs. "breakthrough" infx for a while, but not forever

OTOH memory cells seem durable and may explain continued protection vs. severe disease w/o a boost: theatlantic.com/science/archiv…

To summarize:
- Antibodies decrease but memory B/T cells are stable for ≥6 mo
- Immune memory is still effective vs. variants
- "Boosting" from memory = significant (but temporary) increase in antibodies w/ less impact on already durable memory cells

science.org/doi/10.1126/sc…

We think this is mostly good news. Immunological memory after #mRNA vax meets our expectations for long-term immunity. Our data may also provide some information on what to expect from booster vaccines. Lots more discussion in the full paper that doesn't fit a 280 character limit

This was team science from the start @markmpainter @divijmathew @s_apostolidis co-led this work w/ @Bad2theBeaker @EJohnWherry. Collabs w/ @Penn_IFI @SCOTTeHENSLEY @PaulFBates @SetteLab @KirbyInstitute also critical. @amylizbaxter @Facts_Matter @OldridgeDerek helped build methods

Additional thanks to folks like @KatherineJWu @ewencallaway @jwgale for highlighting this work. The immune system is complicated and their science communication is more important now than ever. Stay tuned for more vaccine science from the @WherryLab

A few more thoughts on boosters -

Vax is not a one size fits all concept. 3rd dose clearly has benefit for >65, immunocompromised, etc. who did not have an optimal response to the initial vax series: nature.com/articles/s4159…

We looked at mostly young and otherwise healthy folks