Hacker News Reader

Comment by usaar333 | Hacker News Reader

> However, making such changes that are not supported by adequate scientific evidence may ultimately be counterproductive to public health.

I remain sympathetic to the argument that we're maintaining too high of a standard of evidence during a rapidly moving pandemic. The question isn't whether this "may be" a harmful decision (of course it might); it is whether it is likely to be. The FDA's memo fails to convince me the expectation is worse taking the single dose strategy.

This reads like a letter by someone who keeps all their money in CDs rather than stocks, because they might lose money.

View on HN

> The question isn't whether this "may be" a harmful decision (of course it might); it is whether it is likely to be.

The statement said there was insufficient data to make that calculation definitively. If you want to take a course of action not borne out by the study, the FDA doesn’t go after you or your doctor for off label use. The FDA is clarifying that the unproven route is, well, unproven.

The FDA is all about medical risk vs provable results. They can’t intelligently comment on the uncontrolled unknown beyond saying that that you’re entering that territory. If you want odds on rolling the dice outside that space, they’re irrelevant to you. So stop looking there.

edit: what you want is a politician with the balls to say whether we should do it anyway and convince other politicians that they should own making that decision even if they turn out to be wrong. But Washington is full of cowards too cowardly to even go to bat for sticking to the studied regime beyond milktoast deference to science so you’re here expecting the FDA to start exceeding their role.

It is not about who has "balls." You seem to think this is just a balancing of risks. It is not. It is a balancing of "what do we actually have data to support" versus "what do we hope to be the case but have no actual evidence to suggest it is so."

Sure, but we do have prior probabilities based on other vaccination campaigns and studies. We also have some understanding of the causal mechanism of this vaccine, so there is other evidence. It’s just not a direct randomized control. People fixate on statistical testing under randomized control, but there’s a lot more information out there that can be used to assess probabilities.

Given that we've never before studied an mRNA vaccine, how do any of the numerous other studies apply to either the Moderna or Pfizer vaccine?

> You seem to think this is just a balancing of risks.

No I don’t. I was saying that the kind of balancing you mention is out of scope for FDA to answer. The fact you can’t find anyone to answer the question you want answered is not the FDA’s problem.

I said go off-label if you want to (although off-label might be slightly the wrong word in this case). Good luck finding a politician that wants to tie themselves to recommending that, in case it turns out wrong.

> politician with the balls

I wish that these decisions wouldn't need to be deferred to politicians. So many of them are selfish idiots.

I'd more trust the vote of relevant medical specialists. I'd rather that these specialist provide the public the best/concise information they have and let people make informed decisions for themselves, or even have a referendum on one dose versus two, and prioritizing.

Politicians have already turned the world upside down for something that looks like it will kill two out of a thousand people when all is said and done, mostly older. Half the population is dumbfounded by their lack of perspective, the other half is scared shitless.

This is why I am frustrated by the FDA. It seems like all they are willing to do is read statistics produced by others.

The suggested dosing is obviously unproven because it has not had a nice large scale study done. Shouldn't this be where they conduct a few rapid small scale experiments?

That data, the existing trials, along with a deep bench of knowledge should be enough to make a well justified determination.

FDA is doing exactly what’s in their statutory responsibility and authority. ie: what congress told them to do. (See my edited addendum to the parent comment.)

When someone observes that a government agency which seems to be doing something nonsensical is actually just following the law perfectly, it always brings to mind H. L. Mencken: "Democracy is the theory that the common people know what they want, and deserve to get it good and hard."

1) The FDA does not run clinical trials.

2) Trials must be large to see results because Covid is rare. The Moderna trial took 30,000 patients 3 months to get to 90 covid cases.

3) This could be a an interesting opportunity for challenge trials, but unfortunately, the key metric for vaccines is reduction in severe cases. Subjecting people likely to have a severe case is unethical when there chance of normally catching it is very low.

Why not look for key indicators of immunity in the authorized dosage group, then compare against indicators in the reduced dosage group?

This is part of my complaint. Why is a large trial, or even a challenge trial (which I also support), the only way to get results here?

There is more to science than reading statistics from trials. They should be able to ask and answer very specific questions based on the working model proven far. Then scale those results to the broader problem.

Running trials takes time. It isn't worth it.

This post makes the same point at greater length and considers counterarguments: https://marginalrevolution.com/marginalrevolution/2021/01/fi... What's the expectation in lives/disability of one course vs. the other? That's the first-order question.

Added: https://www.acpjournals.org/doi/10.7326/m20-8137

> We find that under most plausible scenarios, a more balanced approach that withholds fewer doses during early distribution in order to vaccinate more people as soon as possible could substantially increase the benefits of vaccines, while enabling most recipients to receive second doses on schedule.

I've only read the abstract, but this is an example of the kind of analysis you want to see behind a decision, as opposed to "we cannot conclude anything definitive" which is true but a deflection from actual decision-making.

I think this is part of the overall issue with the handling of this pandemic. No one wants to take responsibility.

The FDA's job is not to decide which lives are more valuable than others, or what our society's risk tolerance is. Their job is to keep people safe and to present the data and the options to political leaders.

The FDA clearly has a goal of not tarnishing their reputation. If they make the wrong decision here, then they lose credibility with approvals for future drugs that come out to treat Alzheimer's or diabetes or whatever else.

Ultimately, I think we look to our scientists to make many decisions that aren't really their place to decide. A scientist can tell me that vaccine X is this effective based on their data. Or even that they have a hunch that treatment X will do Y. But ultimately, I think it's on our elected leaders' shoulders to hear the scientists out about what the various risks are and then make a decision. Unfortunately, I don't have faith that our current elected leaders can adequately listen to our scientists counsel and then make wise decisions accordingly.

> The FDA clearly has a goal of not tarnishing their reputation.

I agree with this statement, but I think there is more to it. The FDA is not only concerned about its reputation, but also public trust.

Don't get me wrong, my _gut_ tells me that a single dose of the vaccine would be beneficial to achieving "herd immunity" more quickly.

But if the FDA goes down that route and they are wrong, then they will have demolished the trust associated with many future decisions.

> This reads like a letter by someone who keeps all their money in CDs rather than stocks, because they might lose money

If there were a clear delta in expected value, then we'd do it. But you can't compare a speculative change to requirements with asymmetrical costs/rewards in a policy context during a pandemic to CDs vs stocks.

Don't give in to the temptation to be like the business guys who change requirements the day of launch. We have a path to end the pandemic. Let's take it.

Following the status quo path will likely result in significantly more people dying. It's a gamble either way, and if you want to argue that the risk of a first-doses-first approach is not worth the risk, then quantify your reasoning and explain exactly how many extra deaths you're willing to accept instead of these platitudes about paths.

You can't ask me to quantify my reasoning and then throw out wild unsubstantiated claims out like:

> Following the status quo path will likely result in significantly more people dying

Besides, the clinical trials _are_ quantified. The issue is we don't have data for the potential changes.

Armchair speculation of "maybe half will work" != data

No. But we can ask the FDA to justify its reasoning by sharing their modeling of the expected value of QALY for different courses of actions. It's ridiculous for them to spit out a proclamation like this without providing a framework for the decision other than "has not been tested".

I hear you about the path to end the pandemic, but there's also that scary thing about the virus mutating enough to render the current vaccines useless. That feels like a very compelling reason to at least consider whether a single dose or smaller doses might get more of us immunity more quickly.

Apparently the variant seen in South Africa recently is "concerning" scientists: https://globalnews.ca/news/7553824/coronavirus-variant-south...

Sure, Moderna and Pfizer could whip up a new vaccine to tackle this ... but how much does the new variant spread and further mutate while we're testing new vaccines?

News is generally trustworthy, but it is in their interest to always make everything sound as dramatic as possible. "Potential new strain which may result in mild reduction in vaccine effectiveness" is not a headline you will ever read. It's a very boring headline. It's in the interest of an advertisement supported company to make the most inflammatory headlines possible.

Let the science do it's work, don't try to be clever and cheat the system.

It reads to me like a letter by someone who invests in target date retirement funds and index etfs by dollar cost averaging on an automated schedule, and avoids investing in tulip bulbs, because the available data shows that this is the most prudent approach.

Unlike stocks the risk here is vastly larger than the reward.

COVID vaccines aren’t believed to be indefinitely effective, so it’s very possible that less effective large scale vaccination effort would cycle through and fail to stop the spread long term. However, a sufficient vaccination effort could completely eliminate the disease globally.

On what basis do you say that? Stocks could potentially involve you losing everything, and the "reward" of cutting the number of necessary doses in half is that we could potentially reach herd immunity much faster.

Bear in mind that the expectation is only about 1.9 billion doses total will be produced in 2021 - so we are talking about the difference between maybe 1/3 or 1/2 of the first world being immunized, and hitting herd immunity this year.

I'm not saying we should, the FDA is probably correct that we should follow the dosing schedule we know works rather than throwing away a whole year of vaccination efforts, but it's obviously desirable.

(it's unfair but I'm sure that most of the doses are going to end up in the US, Europe, and other western-aligned countries first, then China and Southeast Asia, and the developing world is going to get the shit end of the stick, the developed world is going to hit herd immunity long before let's say India or Africa.)

You're not going to die from bad stock advice.

You may die from COVID.

Why don't we start playing options? Everyone gets a tenth of a dose so we can get it done next week. Just start a whole new study, made up of everyone.

Vaccines generally do not eliminate a pathogen and they are not expected to eliminate this Coronavirus. Smallpox is the only human disease to ever be successfully eliminated.

Relatively few vaccines exist, and even fewer old ones, but on the whole they have been quite effective. https://www.vaccines.gov/diseases In total both Smallpox (human) and Rinderpest(cattle) have been eliminated in the wild. Seasonal flu is kind of an interesting case as each year is arguably a different disease with the old one having been eradicated. That’s hopefully a model which will work for covid-19.

Polio is very close to being eliminated, with only 125 known cases in 2019, which was a spike. Rubella dropped from 670 thousand in 2000 to under 15 thousand in 2018.

Ovine rinderpest elimination is considered a reasonable goal, but still 15+ years off.

> COVID vaccines aren’t indefinitely effective

Citation?

We don’t have specific data on covid, but it’s closely related to other diseases where immunity from contacting the disease tapered over time.

The best estimate we can make on the vaccine is. "From what we know of the duration thus far of immunity, I would be surprised if it turns out to be a 20-year duration, but I would also be surprised if it was less than a year,"

We don’t have the capacity to make enough vaccines for global herd immunity in under a year. Which is why reduced long term efficiency is a significant concern over a faster rollout schedule.

> We don’t have specific data on covid

So you're making a super strong statement that's not very justifiable.

> but it’s closely related to other diseases where immunity from contacting the disease tapered over time.

I mean, maybe? Let me know when there's more then extremely minimal evidence of reinfection. We are at the point now where we are coming up on a year of significant spread and it's still missing. Additionally, have we ever seen an mRNA vaccine for a coronavirus? Nope. Evidence here is at best uncertain.

> We don’t have the capacity to make enough vaccines for global herd immunity in under a year. Which is why reduced long term efficiency is a significant concern over a faster rollout schedule.

Would we have enough if we split doses? Why does the US need global herd immunity?

> So you’re making a super strong statement that’s not very justifiable

We don’t think of each annual flu as it’s own disease, but they are caused by closely related strains not a single virus. Similarly, COVID-19 was originally named SARS-Cov-2 due to it’s close relationship with SARS-CoV also known as SARS. https://www.cdc.gov/coronavirus/types.html

Which is why the vaccine could both be developed so quickly in the first place and we can make reasonable predictions about the vaccine. They didn’t create a successful vaccine in under a week from scratch without a lot of knowledge of closely related diseases. It’s that same knowledge which is behind the dosing schedule, and why it was tested like that.

> Why does the US need global herd immunity?

Ending the vast social and economic harm from social distancing efforts. With local herd immunity things go back to normal even if the rest of the world is dealing with the disease.

PS: The demographics of the developing world even support a staggered approach. First a focus on vaccinations for the elderly and medical workers globally then local herd immunity based on population demographics. Local herd immunity protects those who the vaccine can’t, but evenly spreading out vaccinations to the younger population accomplishes little.

Yes, I'm aware of the relationship to SARS. Did you know for the original SARS, natural immunity lasts ~3yrs? https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2851497/

Makes all you're saying have even less sense.

> They didn’t create a successful vaccine in under a week from scratch without a lot of knowledge of closely related diseases. It’s that same knowledge which is behind the dosing schedule, and why it was tested like that.

Was there a SARS vaccine? What same knowledge would give the current dosing schedule then? Why does it trump the new knowledge gained from an actual RCT of the vaccines?

> With local herd immunity things go back to normal even if the rest of the world is dealing with the disease.

Huh? So then the US doesn't need the global herd immunity like I was suggesting?

> PS: The demographics of the developing world even support a staggered approach. First a focus on vaccinations for the elderly and medical workers globally then local herd immunity based on population demographics. Local herd immunity protects those who the vaccine can’t, but evenly spreading out vaccinations to the younger population accomplishes little.

Okay, how is this relevant to whether or not there should be a longer time between initial dose and booster? It's totally irrelevant.

> Similarly, COVID-19 was originally named SARS-Cov-2

No, it wasn't. The disease and the virus that causes it were originally referred to as “the 2019 novel coronavirus” (or, 2019-nCoV for short.)

SARS-CoV-2 was announced as the name of the virus by the International Committee on Taxonomy of Viruses on 11 February 2020; COVID-19 was announced as the name of the disease by WHO also on 11 February 2020.

I am specifically referring to the virus, which I hoped was obvious. To avoid confusion with the disease SARS, the WHO sometimes refers to SARS-CoV-2 as "the COVID-19 virus" in public health communications. And from there it’s entered the popular press.

But yes, the disease was named COVID-19 thus the overlap, and current confusion over the viruses name.

We have a known vaccine that works with 2 doses. We don’t know if 1 dose will work. We have single-dose vaccines with higher temperature tolerance from AstraZeneca and J&J ready to be approved.

Let’s not snatch defeat from the Jaws of victory. Let’s let the 2 dose vaccines goes to those who need it the most and the most at risk. They need more care and monitoring to entire they get both doses.

In a few months when the other vaccines come online then we have unimaginable more flexiblity.

Now is not the time to confuse Americans who apparently are easily confused. Take two doses now. Splitting them up and then some getting Moderna and some getting Pfizer is just a clusterfuck.

How much uncertainty we accept depends a lot on the potential downsides. For example there isn't much of a downside to wearing masks, so that is a measure you could defend on rather flimsy data if you had to.

The potential downside here is very large, it could mean that the vaccine loses a lot of efficacy if the delay to the second dose is too large. It also could create circumstances that favor mutations that escape the vaccine, because you have a lot of people with a weaker vaccination response while the virus is circulating widely.

>This reads like a letter by someone who keeps all their money in CDs rather than stocks, because they might lose money.

That's a good thing in medicine and science- it's called the "precautionary principle". We have what we know (two does in the trials) working, so trying something new like a one dose regimen is risky.

I think the FDA is dodging the obvious since the US doesn't want to alienate the UK:

R-Value with UK mutation: 3?

Effectiveness with 1 dose: 66%?

Probability of breeding a new mutation if you gradually vaccinate everyone in the UK up to only 1 dose:

99.9%?

Once 2/3 of people have 1 dose, mutations that make the vaccines worthless should dominate given the available pool, and then the available pool is back to everyone again.

From what I recall from Dr. John Campbell’s excellent daily videos, the UK strategy was not to give only one dose, but to simply delay the second dose a few weeks so that more people could get their first injection from the initial supply(presumably supply will increase over time). That sounds much lower risk. I think the single dose proposal was a US thing.

All of the mRNA vaccines target the spike protein. Coronavirus spike protein mutations have only ever been observed to occur in bats- never humans. Therefore the number of vaccinated people will almost assuredly not increase the likelihood of a resistant strain.

I would accept a similar kind of reasoning in an argument towards caution, but not in an argument for seeing what happens with prolonged partial but insufficient immunity. Aside from near certainty that a change to the target protein is game over, we have no idea what other mutations could render immune responses triggered by the vaccines worthless, and I don't see the ethics in an experiment to find these mutations.

step 1. do no harm.

Reply on news.ycombinator.com