Thread by @michaelzlin

🎉🎇🙂 Wonderful news! @US_FDA will allow mix & match boosters!

As the linked article below says, booster freedom helps local health officials. And it's also the right thing scientifically, especially for #JnJers

This is the happiest I've ever felt from a FDA decision!

https://twitter.com/V2019N/status/1450223481178578945

I had some hope FDA might do this, after Dr. Marks' said in Friday's VRBPAC meeting that they'd like to know what info was needed to approve mixed boosters, and some VRBPAC members replied that the "Mix and Match Booster Trial" results were clear enough

nytimes.com/2021/10/16/us/…

I'd learned from repeated lack of action not to hope too much. Gathering info on the benefits of heterologous boosts for #JnJers and seeing it tossed aside for various nonreasons made me feel like Sisyphus rolling a stone

So this news is like Sisyphus finally cresting the hill

I got into the J&J issue because I couldn't help correcting what I saw as a widespread unbalanced treatment of vaccine findings in March. I'm Pfizer myself so it didn't matter directly to me, but as a scientist I didn't like seeing data misused

https://twitter.com/michaelzlin/status/1372278100805836801

We ask people to spend time gathering data for robust statistics, so to then interpret the findings in clearly biased ways (picking the end of confidence interval that suits your argument) makes a mockery of statistics and disrespects the effort put in to gather the data.

It took until Friday's VRBAC — 6 mo — for the pretense that J&J was as good as the others to be refuted by authorities on the record (by VRBAC chair Dr. Monto, then FDA's Marks and CDC's Cohn). Until then all you'd hear was "no evidence of differences"

https://twitter.com/michaelzlin/status/1449039346116358144

But the RCT always indicated J&J provided less case protection and was unclear on hospitalization protection; the gap in case protection correlated well with antibody levels. I keep citing this paper because it was both early and correct.
nature.com/articles/s4159…

When Delta came, it was clear J&J VE wouldn't be better than against the original strain (70%), and when Pfizer VE waned to that level, people went bananas with anxiety.

But each time someone asked whether #JnJers were okay you'd get PRs with reassuring conclusions and concerning numbers, and news orgs would repeat the conclusions and not interpet the numbers, such as below
nytimes.com/2021/07/01/hea…

or below. It was like getting trapped in some horror movie where only the protagonist can see the monster and everybody else thinks he/she is crazy
nytimes.com/2021/08/06/sci…

Meanwhile Europe had put out data showing AZ+RNA was better than AZ+AZ, and safe, and AZ+RNA (similar to JJ+RNA) became widely implemented elsewhere. It was the perfect zero-risk high-yield solution for J&J's issues.
nature.com/articles/d4158…

That's why @PaulSaxMD and I wrote this opinion piece i August to recommend heterologous boosting, first for immunocompromised JnJers (who as of today still don't have an officially recommended boosting option!) and later hopefully for all #JnJers
nytimes.com/2021/08/29/opi…

And that's why I started gathering all data I could find on J&J efficacy against Delta, estimating ~50% overall VE in late August, which AFAIK was the first publicly available estimate of J&J VE based on real world cases

https://twitter.com/michaelzlin/status/1432121982397599745

I also estimated 2x lower efficacy vs hospitalizations in late August relative to Moderna

https://twitter.com/michaelzlin/status/1430794839415660544

These were confirmed by Amanda Cohn at VRBPAC on Friday, and by papers such as medrxiv.org/content/10.110… (graphic below, even bigger hospitalization diff than I'd seen in the OK data) and cdc.gov/mmwr/volumes/7… (CDC's own study)

Finally there was one more PR announcing 94% VE by J&J boosting itself at 2mo. But the 94% was based on 15 US cases, when J&J also saw 75% based on 66 WW cases. There were better Ab levels after a 6mo boost, but even then I doubt you'd get 94% VE vs Delta

https://twitter.com/michaelzlin/status/1440390832633253888

The press repeated the number and the implication that the only available booster would be J&J again.
nytimes.com/2021/09/21/hea…

But then NIH burst on the scene with their eagerly awaited mix & match booster trial, which produced spectacular results... both in terms of data quality and in the benefits of RNA boosters for JnJers, finally bringing them up to (and exceeding) other vax

https://twitter.com/michaelzlin/status/1448335699866836993

Then yet another plot twist: The VRBPAC was convened to discuss the J&J booster application and the agenda didn't include the NIH study. The J&J recitation of the dubious 94% number went unchallenged.

But then the very final plot twist: Marks put the NIH study on the agenda and gets favorable comments on the heterologous boost. And now it seems we might actually get an approval for booster freedom sooner rather than later.

So that's the whole story, it seems, which went on far longer than I hoped, with more twists and turns than I'd have thought possible.

In a sense everyone acts out the part assigned to them: J&J scientists propose their own products because that's what they will get rewarded for, FDA has to wait for the data, NIH comes up with an objective trial to get fair data etc. Who says science can't be exciting?

And as to whether my tweets and our opinion piece swayed any important minds, I don't know, but it doesn't matter as long as the right decision is made this week. In the meantime I hope it was useful for some people.

But I will be very glad when this saga is finally truly over

And this is why it mattered (linking threads)

https://twitter.com/michaelzlin/status/1450319601275793409