4 ms·

The case for a one-dose first strategy gets stronger and stronger, but due to slavish proceduralism, the FDA wouldn’t consider it.

by ummonk 6y ago

thehappypm 6y ago
There is 0 data that any of the vaccines approved by the FDA are efficacious with a single dose. There is 0 data to support that a delayed dose is efficacious. The FDA is making a very, very valid choice so far.
- endgame 6y ago
  There is also minimal data that parachutes are better than placebo for people who jump out of planes. This situation needs wartime thinking, not peacetime proceduralism. What you're saying is that we must assume the worst case, that the FDA-approved vaccines are like 2-pack epoxies: that first doses give 0% immunity, and all the efficacy comes from applying the second. That seems completely absurd.
- gojomo 6y ago
  Your claim of "0 data" is completely wrong. The original phase 3 trials already showed limited support for partial one-dose efficacy. (They even broke it down in their writeups! That's non-zero data!) Observational reports so far on the tens of millions additional doses given (compared to a mere ~15-20K in phase 3 trials) is also suggestive of 1-dose efficiency. (Most recently, a retrospective study in Israel – the nation with the highest proportion of its population immunized – found 51% efficacy against "confirmed COVID" looking at just days 13-24 after a single dose – and that's in a world with far more variants than the original mRNA studies. It's reasonable to conjecture from other results that the efficacy would be even better later, and better against "severe disease".) Plus, there's everything we can reason about from similar diseases & vaccines. The reason to wait for 2nd-dose-plus-7d in study "primary endpoints" is to have a singular, legible, stark readout for rather-simpleminded regulatory processes. But smart people who live in the real world know the actual mechanisms are far more fluid/incremental, with immune processes starting immediately from one dose, then accelerating over weeks (even without a booster).
- mchusma 6y ago
  This is very innacurate. We have the following data: -Moderna data between the 1st and 2nd dose at 28 days -Pfizer data between the 1st and 2nd dose at 21 days -Aztrazeneca data between 1st & second dose at various dosing schedules -J&J data -Novavax data -Data from Isreal real world deployments of the vaccine -Data from UK First Dose First effort -Our Bayseian priors - <30 day boosters are almost never tried in the real world because larger time gaps typically illicit better protective responses. You can read the FDA briefings for Moderna & Pfizer which review the 1 dose data, its very readable. The empirical data is overwhelming from many sources and the data matches our current disease models. All signs suggest overwhelmingly that: (1) 1 dose appears to be 100% effective at preventing death in all candidates >14 days after the 1st dose. (2) The marginal benefit of the second dose is real but comparatively much smaller. (3) The delaying the second dose appears to confer additional immunity. Focusing on 1st doses is by far the best decision.
gojomo 6y ago
Indeed, even the practice of calling certain vaccines "2-dose" or "1-dose" is somewhat arbitrary & misleading, reflecting rushed early decisions about which courses to test. The Pfizer/Moderna mRNA vaccines might be just as effective, 30d-onward after one dose, as the "1-dose" J&J. The tested 21d/28d spacing for the mRNA vaccines might be far shorter than optimal, so the ad-hoc "emergency" delays of booster shots by many jurisdictions might actually be benefitting long-term immunity. The "1-dose" J&J might benefit greatly from a booster 30d, 90d, 180d later. Even the dosings could be far from optimal, with some hints that half or less of the mRNA vaccines may be just as effective, especially in younger patients. And of course there was the AZ/Oxford glitch that mistakenly gave half the intended dose in the 1st shot - & that subgroup seemed to have even less disease after the 2nd full-dose shot. (While this could be a statistical fluke, maybe also a "booster" that's stronger is read by the immune system to mean that the disease is continuing to become even more of a threat, so it reacts even more strongly than to a "same as we beat before" dose. Until lots more study is done – who knows?) So: people who treat the officially/rigorously evaluated doses/schedules as "optimal" or "the only safe course" are way overclaiming what the data shows. They data we have are a series of singular draws from the possibility-space, showing some points of reasonably-safe and -effective approach. With plenty of other data constantly arriving, hinting at other equally- or more-effective approaches, regions & practitioners should be free to use their best-judgement to navigate risk/reward tradeoffs.

The case for a one-dose first strategy gets stronger and stronger, but due to slavish proceduralism, the FDA wouldn’t consider it.

by ummonk 6y ago

The case for a one-dose first strategy gets stronger and stronger, but due to slavish proceduralism, the FDA wouldn’t consider it.

thehappypm 6y ago

There is 0 data that any of the vaccines approved by the FDA are efficacious with a single dose. There is 0 data to support that a delayed dose is efficacious. The FDA is making a very, very valid choice so far.

endgame 6y ago
There is also minimal data that parachutes are better than placebo for people who jump out of planes. This situation needs wartime thinking, not peacetime proceduralism. What you're saying is that we must assume the worst case, that the FDA-approved vaccines are like 2-pack epoxies: that first doses give 0% immunity, and all the efficacy comes from applying the second. That seems completely absurd.
gojomo 6y ago
Your claim of "0 data" is completely wrong. The original phase 3 trials already showed limited support for partial one-dose efficacy. (They even broke it down in their writeups! That's non-zero data!) Observational reports so far on the tens of millions additional doses given (compared to a mere ~15-20K in phase 3 trials) is also suggestive of 1-dose efficiency. (Most recently, a retrospective study in Israel – the nation with the highest proportion of its population immunized – found 51% efficacy against "confirmed COVID" looking at just days 13-24 after a single dose – and that's in a world with far more variants than the original mRNA studies. It's reasonable to conjecture from other results that the efficacy would be even better later, and better against "severe disease".) Plus, there's everything we can reason about from similar diseases & vaccines. The reason to wait for 2nd-dose-plus-7d in study "primary endpoints" is to have a singular, legible, stark readout for rather-simpleminded regulatory processes. But smart people who live in the real world know the actual mechanisms are far more fluid/incremental, with immune processes starting immediately from one dose, then accelerating over weeks (even without a booster).
mchusma 6y ago
This is very innacurate. We have the following data: -Moderna data between the 1st and 2nd dose at 28 days -Pfizer data between the 1st and 2nd dose at 21 days -Aztrazeneca data between 1st & second dose at various dosing schedules -J&J data -Novavax data -Data from Isreal real world deployments of the vaccine -Data from UK First Dose First effort -Our Bayseian priors - <30 day boosters are almost never tried in the real world because larger time gaps typically illicit better protective responses. You can read the FDA briefings for Moderna & Pfizer which review the 1 dose data, its very readable. The empirical data is overwhelming from many sources and the data matches our current disease models. All signs suggest overwhelmingly that: (1) 1 dose appears to be 100% effective at preventing death in all candidates >14 days after the 1st dose. (2) The marginal benefit of the second dose is real but comparatively much smaller. (3) The delaying the second dose appears to confer additional immunity. Focusing on 1st doses is by far the best decision.

gojomo 6y ago

Indeed, even the practice of calling certain vaccines "2-dose" or "1-dose" is somewhat arbitrary & misleading, reflecting rushed early decisions about which courses to test. The Pfizer/Moderna mRNA vaccines might be just as effective, 30d-onward after one dose, as the "1-dose" J&J. The tested 21d/28d spacing for the mRNA vaccines might be far shorter than optimal, so the ad-hoc "emergency" delays of booster shots by many jurisdictions might actually be benefitting long-term immunity. The "1-dose" J&J might benefit greatly from a booster 30d, 90d, 180d later. Even the dosings could be far from optimal, with some hints that half or less of the mRNA vaccines may be just as effective, especially in younger patients. And of course there was the AZ/Oxford glitch that mistakenly gave half the intended dose in the 1st shot - & that subgroup seemed to have even less disease after the 2nd full-dose shot. (While this could be a statistical fluke, maybe also a "booster" that's stronger is read by the immune system to mean that the disease is continuing to become even more of a threat, so it reacts even more strongly than to a "same as we beat before" dose. Until lots more study is done – who knows?) So: people who treat the officially/rigorously evaluated doses/schedules as "optimal" or "the only safe course" are way overclaiming what the data shows. They data we have are a series of singular draws from the possibility-space, showing some points of reasonably-safe and -effective approach. With plenty of other data constantly arriving, hinting at other equally- or more-effective approaches, regions & practitioners should be free to use their best-judgement to navigate risk/reward tradeoffs.