6 ms·

We could retarget the mRNA batches tonight if it wasn't for the FDA not taking this seriously.

by wbl 5y ago

H8crilA 5y ago
Uhm, it is serious if you know anything about drug development history. It always is. The current mRNA Covid vaccines have a pretty high incidence rate of ongoing rapid swelling and rashes (in the 1/100 to 1/1000 range), and it's still unclear what will happen to the hundreds of thousands of affected people. Source: https://en.m.wikipedia.org/wiki/Pfizer%E2%80%93BioNTech_COVID-19_vaccine#Adverse_effects https://en.m.wikipedia.org/wiki/Pfizer%E2%80%93BioNTech_COVI... Also to be found in the adverse reactions database (VAERS) by the CDC.
- astrange 5y ago
  VAERS reports aren't sufficient to show causation, because they just show whatever people happened to report. It's the same reason drugs show the thing they're treating as a side effect. People are reporting the symptom because they already had it.
gus_massa 5y ago
Retargeting is one of those things that everyone is 99% sure it's safe. But 99% sure in millions of people means thousand of problems and perhaps deaths. So before signing any authorization it's better to wait until the clinical trials are done instead of using gut feelings. (Waiting too much is also bad, so it's a tradeoff.)
- dragonwriter 5y ago
  > Retargeting is one of those things that everyone is 99% sure it's safe. But 99% sure in millions of people means thousand of problems and perhaps deaths. You seem to be assuming “99% sure it is safe” means “sure that its 99(.9)% safe”. They are...not equivalent.
  - gus_massa 5y ago
    I know... I know... You are right. But a simple calculation sometimes is better than a correct long calculation in an online discussion :) The correct calculation is to ask them "How sure are you that is at lest safe for X% of the persons?" The replies are like 100% sure that it's safe for 90% of the persons. Also 100% sure that it's safe for 99% of the persons. 97% sure that it's safe for 99.9% of the persons. 93% sure that it's safe for 99.99% of the persons. 87% sure that it's safe for 99.999% of the persons. ... Now you take the derivative of the function (or the inverse of the function?), multiply by -1 if necessary and get a gut feeling probability distribution to use to calculate the integral and get the expected values of deaths of using the new version without a clinical trial. Anyway, you need a lot of 9s, because a FDA approval means almost automatic approval in a lot of countries. There are like 10 vaccines now and 7.000 million people in the word, so my guess that it will be used in 500-100 million people.
- wbl 5y ago
  What has the FDA done to start gathering this data?
  - maxerickson 5y ago
    Nothing. NIH funded the trial for the Moderna booster.
logicchains 5y ago
A nice presentation by the FDA about why they're reluctant to let new treatments through without full phase 3 trials: https://www.fda.gov/about-fda/reports/22-case-studies-where-phase-2-and-phase-3-trials-had-divergent-results https://www.fda.gov/about-fda/reports/22-case-studies-where-...
- wbl 5y ago
  We don't do phase III trials for the flu vaccine every year. Instead we follow the same production formula for the new virus variant. Why not with COVID? The FDA agrees this is desirable but is just foot dragging.
  - jhayward 5y ago
    Do you think there might be some difference in "changes each year for flu vaccine" and "changes for every other vaccine"? Like perhaps, how much study the practice of flu vaccine changes has undergone, what the limited target constraints are, and prior field experience with the sterotypical changes to those specific targets?

We could retarget the mRNA batches tonight if it wasn't for the FDA not taking this seriously.

by wbl 5y ago

We could retarget the mRNA batches tonight if it wasn't for the FDA not taking this seriously.

Uhm, it is serious if you know anything about drug development history. It always is. The current mRNA Covid vaccines have a pretty high incidence rate of ongoing rapid swelling and rashes (in the 1/100 to 1/1000 range), and it's still unclear what will happen to the hundreds of thousands of affected people. Source: https://en.m.wikipedia.org/wiki/Pfizer%E2%80%93BioNTech_COVID-19_vaccine#Adverse_effects https://en.m.wikipedia.org/wiki/Pfizer%E2%80%93BioNTech_COVI... Also to be found in the adverse reactions database (VAERS) by the CDC.

astrange 5y ago
VAERS reports aren't sufficient to show causation, because they just show whatever people happened to report. It's the same reason drugs show the thing they're treating as a side effect. People are reporting the symptom because they already had it.

gus_massa 5y ago

Retargeting is one of those things that everyone is 99% sure it's safe. But 99% sure in millions of people means thousand of problems and perhaps deaths. So before signing any authorization it's better to wait until the clinical trials are done instead of using gut feelings. (Waiting too much is also bad, so it's a tradeoff.)

dragonwriter 5y ago
> Retargeting is one of those things that everyone is 99% sure it's safe. But 99% sure in millions of people means thousand of problems and perhaps deaths. You seem to be assuming “99% sure it is safe” means “sure that its 99(.9)% safe”. They are...not equivalent.
- gus_massa 5y ago
  I know... I know... You are right. But a simple calculation sometimes is better than a correct long calculation in an online discussion :) The correct calculation is to ask them "How sure are you that is at lest safe for X% of the persons?" The replies are like 100% sure that it's safe for 90% of the persons. Also 100% sure that it's safe for 99% of the persons. 97% sure that it's safe for 99.9% of the persons. 93% sure that it's safe for 99.99% of the persons. 87% sure that it's safe for 99.999% of the persons. ... Now you take the derivative of the function (or the inverse of the function?), multiply by -1 if necessary and get a gut feeling probability distribution to use to calculate the integral and get the expected values of deaths of using the new version without a clinical trial. Anyway, you need a lot of 9s, because a FDA approval means almost automatic approval in a lot of countries. There are like 10 vaccines now and 7.000 million people in the word, so my guess that it will be used in 500-100 million people.
wbl 5y ago
What has the FDA done to start gathering this data?
- maxerickson 5y ago
  Nothing. NIH funded the trial for the Moderna booster.

logicchains 5y ago

A nice presentation by the FDA about why they're reluctant to let new treatments through without full phase 3 trials: https://www.fda.gov/about-fda/reports/22-case-studies-where-phase-2-and-phase-3-trials-had-divergent-results https://www.fda.gov/about-fda/reports/22-case-studies-where-...

wbl 5y ago
We don't do phase III trials for the flu vaccine every year. Instead we follow the same production formula for the new virus variant. Why not with COVID? The FDA agrees this is desirable but is just foot dragging.
- jhayward 5y ago
  Do you think there might be some difference in "changes each year for flu vaccine" and "changes for every other vaccine"? Like perhaps, how much study the practice of flu vaccine changes has undergone, what the limited target constraints are, and prior field experience with the sterotypical changes to those specific targets?