Much attention (>1000 primary tweets) has been going to a paper titled "Recent SARS-CoV-2 infection abrogates antibody and B-cell responses to booster vaccination"

The study is complex and hard to summarize. Indeed the title is not an accurate summary IMO
medrxiv.org/content/10.110…
The study was done at a transitional time in the epidemic, so its current relevance is not too clear. It studied effects of infection on responses to a Wuhan booster after a Wuhan primary series. So that's a Wuhan vaccine sandwich with Delta or BA.1 infection in the middle.
And now we have BA5 in the boosters, and the study is not designed to study what introducing a new antigen will do to the old response, or what an old response will do to a new antigen.
Now we know the who and when, we can discuss the main results:
1, Shot 3 induced more antibodies 30 days later for all subjects
2, In 6 subjects who had COVID within 6mo of shot 3, day 0 and day 60 Ab levels were similar to day 60 for everyone else (so no change from 0 to 60)
That's it. Where's the "abrogation", you might wonder. It's in the lack of difference between day 60 and day 0 for the recently infected, forgetting that something happened in the immune system since everyone saw more Abs at day 30.
The authors attributed the higher Ab levels at day 30 to new plasmablasts, whereas a lack of difference at day 60 for the recently infected with the highest baseline abs suggest already activated memory B cells couldn’t be activated further.
So a more accurate title might be "Boosting during high antibody production induces new plasmablast formation without further memory B cell restimulation"
You might wonder why the authors would choose a title that says "Recent SARS-CoV-2 infection abrogates *antibody* and B-cell responses to booster vaccination" when the booster did induce more antibodies (transiently), but you'd have to ask the authors that
After Fig 1B above, the authors focus on the 60-v-0 day difference for the rest of the paper. In Figs 1C-E, they rescale everything as fold diff from day 0 to 60. Because prior-infected have higher baseline abs, as expected they have lower fold induction.

Cyan = prior infected
The rescaling shows again some prior-infected return to preboost Ab levels 60 days post-boost. Now however we can't tell that they had the higher baselines, whereas we saw that in Fig 1B. We also can't tell they induced Abs at day 30. So Figs 1C-E throw away info from Fig 1B.
Fig. 2 examines # of B cells reacting to spike N-terminal domain (NTD) or receptor-binding domain (RBD) (both parts of S1). Panel B shows the % of all B cells (CD19+) that react to these antigens. You can see prior-infected (cyan) have the highest baselines and stays pretty flat
Panel C replots B cell #s at day 60 relative to 0. The prior-infected again shows no change. But they had made more antibodies at day 30, so hmmm what does it mean?

The easiest explanation is the B cell population has hit a ceiling; new B cells are few in # or replacing old ones
So are 2x vaxxed then infected then boosted really any worse off than 2x vaxxed then boosted? Actually they are, in a very subtle way: the % of B cells that can recognize BA.1 appears lower for 2x vaxxed then infected then boosted (cyan) than 2x vaxxed then boosted (lavender).
Note these two groups didn't see BA.1 antigen ever, so this is a measure of somatic hypermutation. To me this is the only aspect where suboptimal results are seen in the infected-between-shots-2-and-3 group: there is less SHM. The final diff is 15% (60 vs 70% dual-reactive)
The rest of the paper gets into cell signaling and cell cycle diffs, looking for molecular correlates of this refractory period in memory B cell SHM, picking the 6 most recent infection in the prior-infected group. Don't think it's necessary to discuss here.
What is important to know is the precise dates of infection in the prior-infected group. 6 of them had distinctly higher baselines and less boosting as a result. These are presumaby those with infection within 6mo of boosting. But were they closer to 2mo (the minimum) or 6mo?
In any case what do the results say about when to get bivalent boosters after infection? It's unclear IMO. Not only are we missing important data on timing, but previously Omicron-infected getting a bivalent booster would be getting only their 2nd exposure to Omicron spike.
This is very different from the study subjects who were getting a 3rd or 4th exposure after infection. To know what Omicron-infected can expect from the bivalent booster, we are better off looking at studies of Wuhan-infected patients getting Wuhan-vaxxxed for the first time.
Thx eagle-eyed @Nicolewithno_h who spotted different times on the X axis of Fig 5D, implying 5 of the 6 were 120-170 days = 4-6 months out from their infection. That means at most 1 patient, the 2-month one, could possibly have been Omicron (the study ending in 3/2022).
And NIH guidance is to suggest waiting 3mo after infection, based on a study finding ≥3mo between infection and vaccination was slightly better than <3mo:
cdc.gov/vaccines/covid…

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Michael Lin, MD PhD 🧬

Michael Lin, MD PhD 🧬 Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @michaelzlin

Sep 7
I had that long thread with detailed data, but I know it's not easy to follow. So in the spirit of Twitter here's the short form:

2022 saw a new virus family: BA2-3-4-5. This family is distinct from the vaccines and viruses we got before.

There's a new vaccine with BA5. Get it.
The BA5-containing vax happens to be called a bivalent booster. Bivalent because it contains both 2019 strain and BA5. Booster because you need to have had a 2019-strain vax to get this BA5-containing vax. But it's easier just to consider it a new vax for the 2022 strains.
If you weren't infected in 2022, the bivalent booster will be your 1st exposure to the 2022 strains BA2-3-4-5. Get it.

If you were infected in 2022, a bivalent booster >3mo later will be your 2nd exposure, and will enhance your immunity to the 2022 strains BA2-3-4-5. Get it.
Read 7 tweets
Sep 4
Random question: why did some restaurants that had outdoor tables on their sidewalk remove them? The sidewalk wasn't going away, and neither is COVID. Is it really that much harder for the waiters to take the order outside? Wouldn't the restaurants profit from the extra seating?
Maybe someone in the restaurant industry can explain?
Funny there's an article about this today for NYC. Indeed there it seems pandemic exceptions to rules prohibiting tables on public sidewalks are ending.

Too bad our civil leaders have to treat return to normalcy as undoing all changes including healthy popular ones
Read 6 tweets
Sep 2
Okay everyone. Small request: If you want to make a scientific point, don't just make a broad assertion "e.g. you're wrong" and then link a paper. If you actually have read the paper and understand it, then explain what evidence in the paper backs up your point...
Also put it in context of what's already known, and comment on the quality of the data. Then you may make a point and I may reply.
Otherwise this drive-by citing tactic creates many issues:
1. I often have no clue what you are trying to say
2. Even if I have a clue what you are trying to say, the paper may not say what you think it says
Read 10 tweets
Aug 31
If you recently got an Omicron infection, should you get the bivalent Wuhan+BA.5 booster that was approved today?

I would recommend it, for an unexpected reason: It will improve your protection not only against BA.5 but also against upcoming BA.2.x variants.

Short 🧵
First, which Omicron you got depends on when you got it. Winter=BA.2, spring-summer=BA.5.

The case is clearest if you got BA.5. A single infection of SARSCoV2 is like one shot of antibody, and provides about 90% protection against reinfection for a few months.
That's if BA.5 was your first COVID19. If not, protection may be lower than 90% for BA.5 reinfection due to some imprinting to an earlier strain.

Anyway, getting bivalent booster a few months after your BA.5 infection (ideally 3, but can be as early as 1) is good for 5 reasons
Read 41 tweets
Aug 31
Just two comments on the bivalent Wuhan+BA5 bivalent booster approval today.

1) Turns out Pfizer is approved for 12+ and Moderna 18+. Could be due to suspected higher myocarditis in Moderna boosters (25+25mcg vs 15+15mcg for Pfizer), but don't know if there are good data on this
2) In case you're looking for it, the provider sheets have no data on neutralization of BA.5 by the booster in mice. Guess only human data go in the clinical evidence section of those sheets, so then we get old data on original vax and Wuhan+BA.1 booster
fda.gov/emergency-prep…
In the June VRBPAC, Pfizer and Moderna said they'd need until October to have a BA.5 booster ready. I'm guessing they assumed they'd need to show human antibody data (e.g. nAb levels to BA.5 virus would at levels suggesting efficacy based on original vax against original virus)
Read 14 tweets
Aug 26
Nice try CDC to try to make Paxlovid rebounds a subset of the very rare disease flare-ups without Paxlovid

Nobody is believing this semantic sleight of hand

It's common. It's not rare.

cnn.com/2022/08/26/hea…
Even the authors of a study quoted in the article as finding a <2% rebound rate say that's an underestimate because they only know when people voluntarily reported rebound symptoms unprompted

"Study not designed to detect rebound found few rebounds" is one way to frame it
Recall same-day detection and Paxlovid is happening now but never happened in clinical trials. Uses are supposed to reflect tested conditions plus maybe extension to mechanistically similar situations. But infection day 1 vs 5 are completely different situations immunologically
Read 6 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(