Hacker News Reader

Comment by dcolkitt | Hacker News Reader

Let's be very clear. Covid is currently killing more Americans every three days than the total number of global birth defects attributed to Thalidomide over its entire life.

The FDA refuses to approve AstraZeneca, despite the fact that it's approved in the UK, and doing so would prevent hundreds of thousands of American Covid deaths. The only answer the naysayers can give is to sputter and point at Thalidomide.

So, I want all of us to be crystal clear. If they're wrong about AstraZeneca, then the type II errors of the FDA will cancel out the Thalidomide type I errors by three orders of magnitude. And hereby, no intellectual honest person should ever bring up Thalidomide again, unless explicitly qualifying the much large type II failure of not approving AstraZeneca.

If you believe that AstraZeneca is the next Thalidomide, I want you to bet money on it. If you're willing to bet other people's lives, then you should be willing to bet your own money. I will accept any bet, up to $10,000, that the number of deaths attributable to AstraZeneca over the next two years will not exceed ten thousand deaths globally. (Far less than the number of Americans immediate approval would save).

If you're not willing to bet money than AstraZeneca is the next Thalidomide, then stop insisting that the FDA let hundreds of thousands die by implicitly treating it that way.

View on HN

> The FDA refuses to approve AstraZeneca.

AstraZeneca has not applied to the FDA for an emergency use authorization. They are expected to apply soon.

That's because the FDA basically told them it would be rejected unless they spent months collecting more data. Similar situation with the Novavax vaccine, which seems to have even better results than the AstraZeneca one.

Is there somebody specific this comment is directed at? I feel like I'm missing a lot of context.

No. It's just that there's no penalty for being anonymous and wrong on the internet, even here, and people get a thrill and a small sense of boost in their self-esteem putting down experts in a field. They don't know more than the experts, but they suffer no consequences from saying this. That's all. The FDA is one of the crown jewels of the United States and maybe the world. It breaks hearts in the development of pharmaceuticals and medical devices that barely miss efficacy thresholds and that earns it enemies. It also prevents dishonest scam artists and psychopaths from selling fake or harmful cures as medicine, and those people, as we've clearly seen recently, have no sense of decency and attack anything that prevents them from personal benefit.

Thank god for the FDA. Thank god for Frances Kelsey.

If you think I'm wrong, why aren't you taking my bet?

Consider that the other commenter likely does not know who you are or whether you are trustworthy, even on the off chance that they would be interested in taking such a bet with a trustworthy person.

Your bet is nonsense, and nobody is making the argument you're framing the bet against. There's not the slightest similarity in the massive teratogenicity of thalidomide vs waiting a few months for safety data on a vaccine. Not getting an EUA is not at all equivalent to a black label. You completely ignore we have two vaccines that work better already, that have research behind them from post-SARS1 and MERS research with the mRNA approaches. Astra will get approved after it proves its safety.

Where did I miss anyone mentioning sars-covid-2 besides you? You are the one that brought it up, and I still have no idea why you did. The FDA is right to make sure vaccines are safe. We know AZD1222 is less effective already, so it better be at least equivalently safe and that requires data.

Are you confused and think you're replying to https://news.ycombinator.com/item?id=26048470 ?

> Not getting an EUA is not at all equivalent to a black label.

It is equivalent though, because without an EUA it's illegal to give that vaccine. Right now, we are supply constrained on vaccines - the choice for the people on the margin is not "AstraZeneca now or Moderna now", it's "AstraZeneca now or Moderna in a few months, during which time there's an uncontrolled pandemic raging". Or rather, that would be the choice, except that the AstraZeneca vaccine is illegal to distribute so the only choice is "take your chances with no protection at all".

> There's not the slightest similarity in the massive teratogenicity of thalidomide vs waiting a few months for safety data on a vaccine.

You're right. They're not comparable. Spuriously waiting for Covid data is way worse than thalidomide. The latter resulted in two thousand deaths, whereas the former will result in two million deaths.

I dare you to make the case for this causing 2M incremental deaths. Please proceed.

The UK, where the main difference is the approval of the AZ vaccine, is forecast to reach herd immunity four months before the United States.[1] America is currently averaging 3500 deaths per day.[2] That's an unecessary 127 deaths per 100,000 persons.

Since we're comparing total worldwide deaths from Thalidomide, that would result in 6 million unnecessary deaths if FDA policies were adopted globally.

[1]https://www.bloomberg.com/graphics/covid-vaccine-tracker-glo... [2]https://covidtracking.com/data/charts/us-daily-deaths

So if the AZ vaccine was the only vaccine available to save Americans’ lives (which it’s not) and it was 100% effective (which it’s not) and the FDA delayed its rollout by a month, by your argument the FDA’s delay would have caused 105,000 (3500 x 30) unnecessary deaths.

1. Those numbers don’t work. There are other vaccines in play which can have a disproportionate impact on the death rate if they’re properly targeted at people at highest risk of severe infection. You don’t need herd immunity to prevent most deaths.

2. I’m still not seeing the ‘two million’ deaths you mentioned?

The context is that there's a pandemic going on and the FDA is being a bit slow at approving vaccines. Historical failures of the drug industry like thalidomide explain why the FDA is currently extremely cautious about approving drugs. So you can imagine why some people might get a little worked up when they see a story about thalidomide at this moment in history.

The FDA approved emergency use authorizations for both Pfizer and Moderna within 3 weeks of receiving the applications.

Johnson & Johnson just submitted their application, with emergency use approval expected by the end of the month if there is nothing wrong with J&J's data.

I don't see how this is "a bit slow". It's rather fast, considering that the FDA does their own analysis of the submitted raw data and has a committee of outside experts and representatives of consumers, industry, and sometimes patients also weigh in.

It’s not actually being slow once submissions are made . As said, other vaccines were approved more quickly. And if you want a comparison, it’s acting much more quickly than the EMA seems to be.

Now, the interesting part is why they (apparently) delayed the submission of the AS vaccine by requesting more data...?

Just the system. 10 years from now people look back at covid and say:

"They had save and effective vaccines in June 2020. Why didn't they immediately distribute and scale the production of those vaccines?"

We also had ineffective ones (GSK, Sanofi, Merck—-twice!).

Some of that time was needed to sort the wheat from the chaff.

Because producing vaccines at scale is nothing like webshit at scale?

This is Pascal’s wager applied to medicine. I don’t find it terribly convincing for the same reasons.[1] https://en.wikipedia.org/wiki/Pascal%27s_wager

[1]my eternal life elixir might look like snake oil, but you don’t know that

Vaccine manufacturers certainly aren’t willing to take that bet. They are exempt from any liability of harm. [1] At least before COVID you could litigate in secret vaccine courts. Now even that is off the table.

[1] https://www.cnbc.com/2020/12/16/covid-vaccine-side-effects-c...

Allowing a problematic vaccine at this point will likely have further reaching consequences than the people who experience side effects. Public doubt over vaccines is high and a bad one getting through will give people legitimate cause for concern going forward. The current anti-vaxxer movement is nothing compared to what will happen if a bad Covid vaccine gets approved.

Make sure you count the number of people who will die from preventable disease due to fear of vaccines over the coming decades. Plus add in the likelihood of another pandemic and the deaths caused by both the additional scrutiny over vaccines for it and the public's reluctance to trust them.

Astra Zeneca could not ensure that it dosed people correctly in their trials. This is clown school stuff. The head of the Oxford team described textbook p-hacking at a press conference and admitted they were doing it, saying essentially that it didn't matter if some datasets showed that the vaccine didn't work so long as some subsets of data did. Given this, ANY papers from this group are highly suspect. This isn't quite as extreme as what Didier Raoult stuff, but it's basically the same thing, just less extreme and slightly more sophisticated.

This same team has claimed to have vaccines for multiple viral diseases in the past including for example Chikungunya. None of their studies could be replicated. It's not in the popular psyche in the US because these are not diseases that affect people in the West, but finding supposed cures for them are great for getting grants. Nothing they have done has actually been found to work in the fullness of time. It's not like it sort of worked but not well enough to be statistically significant or that it had some low incidence side-effect. The stuff they came up with was mostly indistinguishable from placebo in trials run by third parties.

They have made crazy claims about their Covid research since March 2020 while serious scientists who have produced stuff that actually works were more focused on stuff in the lab. They continue to make outlandish claims, for example claiming they had developed the first ever monoclonal antibody drug for Covid. This was in the weeks after Trump and ex NJ Gov Christie were successfully treated with Regeneron's and Eli Lilly's monoclonal antibody drugs respectively, and of course the British press and Oxford boosters spread it around in the media.

The "studies" that were published that show "efficacy" had vanishingly few over 65s in the treated population but had over 65s in the control. If you look at it carefully, it's not clear the claimed efficacy is much more than a difference in population characteristics. Their apologists claim they performed worse than Pfizer/Moderna because it's possible the treated took more risks because they had symptoms from the vaccine and so knew they were immune. The only problem with this claim is that the Moderna and Pfizer vaccines caused more flu-like symptoms.

This is pure conjecture, but I suspect the UK and EU approvals have more to do with Brexit politics than actual science. It also has to do with the fact that the head of the Oxford team is from the "right" family background. Unlike the US, this sort of thing matters a lot in the EU, even the UK. The UK government is looking good but the reduction in UK deaths is only partially real. More than half the decrease in Covid deaths in the UK is fake because excess deaths went up simultaneously with a decrease Covid deaths. Of course the EU doesn't want to look bad post Brexit so they had to approve and show they were vaccinating their population just as quickly as the UK was. Despite the EU approval, Poland, Germany and France are taking steps to limit the usage of this vaccine in their countries. I wouldn't be surprised if more European countries do so as well.

I for one sincerely hope the Astra Zeneca vaccine is NOT approved by the FDA. Would you rather have a real vaccine or pretend to have been vaccinated and get infected later? Even if the Astra Zeneca vaccine has the claimed 70% efficacy (it doesn't against the new strains), and if R0 is >5 as it is for the new South African and UK variants, it will only slow, not stop the epidemic.

Meanwhile, the US is doubling the number of people vaccinated per day every two weeks, and this is with vaccines that incontrovertibly work about as well a any vaccine in history. Exponential growth is bad when it's an epidemic but it's very, very good when it's a startup or the scaling up of a treatment. Resources are always limited. I'd much rather spend them on getting more Pfizer and Moderna vaccines into people's arms than waste time, energy and resources on something that isn't a solution even in the best possible scenario.

The JNJ and Novavax vaccines are looking just as good and are much easier to manufacture and distribute. The top line numbers are a bit worse only because their treated population was different and more challenging to get an immune response out of and infected with the strain the Moderna and Pfizer vaccines were tested on plus many people who had the new strains. The Astra Zeneca vaccine looks like a candidate for countries where a cold logistics chain is hard but it really isn't. It adds the spike protein to an adenovirus. Adenovirus infections generally don't cause many symptoms but after one or two infections you are immune. If you are immune to adenoviruses, your immune system is likely to take out the vaccine before it has the chance to train your immune system to go after the spike protein. The problem here is that adenovirus infections are far more common in areas where cold logistics chains are difficult.

I can't say this will cause Thalidomide like side effects but I would happily take a bet at 10-1 odds that within 5 years, data will come to light that shows serious problems with either the safety of efficacy of the Astra Zeneca vaccine. That implies that I think the chances there are problems are >10%, which is much too high for approval given the IFR of Covid.

Efficacy is irrelevant. Rolling out AZ vaccines will not slow down Pfizer/Moderna vaccination rates, because the bottleneck is supply. Even assuming J&J + Novavax there's still a drastic supply bottleneck.

At worse, we'll just give a bunch of people a placebo. I don't believe there's lack of efficacy, but with millions of subjects, the answer will be apparent in a matter of weeks. If so we'll simply re-innoculate those people with Pfizer/Moderna. There's literally zero cost to this scenario, besides monetary spending that's a rounding error in the federal budget.

The only consideration should be safety. I'm so confident in safety that I'm willing to give you 10-1 odds, up to $500/$5000 that there will be no more than 35,000 deaths attributed to the AZ vaccine globally within five years.

What about increased deaths due to people taking more risks because they think they are safe? Most deaths are in over 65s. AZ has released absolutely no data on over 65s, nada. There is independent data from Germany that shows the AZ vaccine is essentially indistinguishable from placebo in over 65s.

What you are suggesting may be the right thing to do somewhere other than the US, so long as its accompanied by an effective warning that the vaccination may or may not work so you still have to be careful. The important operative word regarding the warnings being effective. My comments are entirely in response to an earlier commenter stating that the FDA should approve AZ, which means it's US specific. The single biggest bottleneck right now is vials, not vaccine production. AZ vaccine production will compete for this.

Besides, do you really think AZ is going to be able to magically scale up from 0 in the US with the UK and EU already fighting over production and do this faster than an absolute lazer focus on increasing Pfizer/Moderna production and distribution? The US is currently doubling the number of people vaccinated per day every 2 weeks. We did this despite a massive snowstorm on the East Coast. It might seem bad but the rate of vaccinations is growing exponentially. It's mostly a solved problem. You just don't see it yet. Exponential growth! It's the same thing that makes high R0 epidemics so bad, only in reverse.

There is data from germany? There was a media blitz about low covid infection prevention in ages over 65, but the confidence interval was between something like -130% to 90%, so it's more of a case of missing data

Most deaths are in over 65s. If the confidence interval on the vaccine working on over 65s is -130% to 90%, how can you possibly claim this vaccine will do anything. Their conclusions were based on data submitted by Astra Zeneca and data Astra Zeneca did not initially submit.

Breaking news this AM: South Africa just halted the use of the Astra Zeneca vaccine. It's being spun as being done because of evidence it doesn't work on the new strain. I strongly suspect once all the data is in, this vaccine will be found to be indistinguishable from placebo.

There is no data from Germany. The German commission on vaccination concluded, based on the data provided by AZ, that a) there is no reason to believe that AZ is unsafe and b) there is currently insufficient data on efficacy in ages over 65 to recommend vaccination.

Would you mind sharing the references underlying your points?

(This isn’t ‘citation required’ snark - I’m genuinely interested.)

> What about increased deaths due to people taking more risks because they think they are safe? Most deaths are in over 65s. AZ has released absolutely no data on over 65s, nada. There is independent data from Germany that shows the AZ vaccine is essentially indistinguishable from placebo in over 65s.

Then why not allocate AZ to the population it is most effective? The whole argument with efficacy across different brackets appears like a (simple) maxflow problem.

A friend of mine was part of the AZ trials, the control wasn't a placebo because lack of side effects would let the person know which group they were in. Instead it was some other common vaccine that would produce similar side effects as the real one.

This was the case with all the trials, not just the AZ one. However, there was a difference in side effects between the vaccinated and controls in all three. The difference was larger for the Moderna/Pfizer vaccines.

Also, there wasn't one AZ trial. There were 8 or 9 in different countries. But for reasons only known to AZ/Oxford, their studies are based on data only from 2 and part of the data from a third.

You make some interesting claims, do you have any sources? Would be interested to read more.

> They have made crazy claims about their Covid research since March 2020 while serious scientists who have produced stuff that actually works were more focused on stuff in the lab. They continue to make outlandish claims, for example claiming they had developed the first ever monoclonal antibody drug for Covid. This was in the weeks after Trump and ex NJ Gov Christie were successfully treated with Regeneron's and Eli Lilly's monoclonal antibody drugs respectively, and of course the British press and Oxford boosters spread it around in the media.

From what I can tell, it looks like Oxford started their trial in September [1] and all other reference I can find for the other drug you are mentioning, "bamlanivimab", seems to come after that date. Not sure why this is such an issue? The trials for example started in October [2]. The second reference there also seems quite critical of the drug you are mentioning:

* Bamlanivimab showed no benefit in the ACTIV-3 trial involving hospitalized patients with COVID-19.

* Bamlanivimab failed to achieve the primary endpoint of viral load reduction in a trial involving outpatients with COVID-19 (BLAZE-1). It may have caused some debatable improvements in secondary endpoints.

* Eli Lilly is now testing a cocktail of bamlanivimab plus another antibody. The company seems to realize that bamlanivimab might have some efficacy, but it’s far from a wonder drug.

* The IDSA and NIH guidelines recommend against generally using bamlanivimab.

* Somehow, bamlanivimab has nonetheless achieved an Emergency Use Authorization via the FDA and is being used at some centers.

> The "studies" that were published that show "efficacy" had vanishingly few over 65s in the treated population but had over 65s in the control. If you look at it carefully, it's not clear the claimed efficacy is much more than a difference in population characteristics. Their apologists claim they performed worse than Pfizer/Moderna because it's possible the treated took more risks because they had symptoms from the vaccine and so knew they were immune. The only problem with this claim is that the Moderna and Pfizer vaccines caused more flu-like symptoms.

A few things here:

1. It doesn't seem any of the Pfizer/BioNTech, Moderna or Oxford trials have enough data for over 65s. You can see MHRA public assessments here saying the same thing. Pfizer vaccine In this document on page 48 [3]. Oxford/AZ vaccine on page 53 [4]. Moderna (not the full public assessment report) page 9 [5].

For example, even the FDA says the same about the Moderna vaccine trial:

> The small number participants and cases in some subgroups, such as participants >75 years of age and participants in certain racial subgroups, limits the interpretability of the individual VE results, but are displayed for completeness. [6]

2. Oxford seem to be the only ones who have published their studies on the vaccines, unless I have missed anything? So it's hard to scrutinise the other vaccine studies.

3. Your last comment seems unsubstantiated.

> This is pure conjecture, but I suspect the UK and EU approvals have more to do with Brexit politics than actual science. It also has to do with the fact that the head of the Oxford team is from the "right" family background. Unlike the US, this sort of thing matters a lot in the EU, even the UK. The UK government is looking good but the reduction in UK deaths is only partially real. More than half the decrease in Covid deaths in the UK is fake because excess deaths went up simultaneously with a decrease Covid deaths. Of course the EU doesn't want to look bad post Brexit so they had to approve and show they were vaccinating their population just as quickly as the UK was. Despite the EU approval, Poland, Germany and France are taking steps to limit the usage of this vaccine in their countries. I wouldn't be surprised if more European countries do so as well.

That is pure conjecture, do you have anything to try and justify this? Especially for your last sentence there, it does seem odd countries are not allowing for over 65s considering there isn't enough data from the other vaccines (unless I have misunderstood this, happy to be corrected). They have more data, but it is still not enough data?

> I for one sincerely hope the Astra Zeneca vaccine is NOT approved by the FDA. Would you rather have a real vaccine or pretend to have been vaccinated and get infected later? Even if the Astra Zeneca vaccine has the claimed 70% efficacy (it doesn't against the new strains), and if R0 is >5 as it is for the new South African and UK variants, it will only slow, not stop the epidemic.

I think calling the vaccine not real is a bit inflammatory. Why is it not real? The new study pre-print seems to support that is a very good "real" vaccine [7]. It also does appear that all the vaccines are less efficacious against other variants such as the one in SA, but seems to be similarly effective to the UK one [8]:

> Oxford researchers say their analysis found similar levels of efficacy against the old variant (84%) and the "Kent" B117 one (74.6%), which was spotted in the South East in September and caused a sharp rise in cases before Christmas. > Dr June Raine, chief executive of the UK's regulator, the MHRA, said the findings were "very reassuring". Dr Peter English, consultant in communicable disease control, said: "This is excellent news, as it indicates that the vaccine works effectively at preventing illness, even with the variant virus."

---

> I can't say this will cause Thalidomide like side effects but I would happily take a bet at 10-1 odds that within 5 years, data will come to light that shows serious problems with either the safety of efficacy of the Astra Zeneca vaccine. That implies that I think the chances there are problems are >10%, which is much too high for approval given the IFR of Covid.

The results from the UK rolling out the Oxford vaccine should be released next week I believe. I dare say you will be proved wrong. Mass issues with the vaccine would surely be out in the wild right now and be reported on.

All-in-all, this seems like quite a different understand of the information out there, and I think much of what you say isn't justified. You definitely seem to have a strong dislike for the Oxford team and the vaccine, nor do you seem to be fond of the UK and/or EU institutions from my understanding of your comment. I'd rather place my trusts in the MHRA, EMA and JCVI and similar agencies who have access to all the data and are happy with the information provided to them.

[1] https://www.bbc.co.uk/news/health-54120753 [2] https://emcrit.org/pulmcrit/bamlanivimab/ [3] https://assets.publishing.service.gov.uk/government/uploads/... [4] https://assets.publishing.service.gov.uk/government/uploads/... [5] https://assets.publishing.service.gov.uk/government/uploads/... [6] https://www.fda.gov/media/144673/download [7] https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3777268 [8] https://www.bbc.co.uk/news/health-55951920

The UK is showing decreasing deaths from Coronavirus while excess deaths (deaths over the baseline level of expected deaths not attributable to Covid, which are usually very steady) have gone up by more than half the supposed Covid decrease. Simultaneous with the rollout of the Astra Zeneca vaccine, the UK has had a new very stringent lockdown. The actual decrease, net of excess deaths does not seem any more than one would have expected from the new lockdown.

25% of the participants in the Pfizer and Moderna trials were over 65, 40% over 55, according your citations. 0 over 65 in the Astra Zeneca. There weren't enough over 65s and certainly not enough over 75s to come up with confidence intervals as tight as one would like in those subgroups for Pfizer/Moderna. Remember it's hard to recruit older participants because they often have co-morbidities that can cause all kinds of problems when you analyze the data. That's very different from Astra Zeneca. In particular, it's bad because the Astra Zeneca did actually recruit over 65s for the treatment groups. They just chose not to release that data while retaining some over 65s in the control.

This AM South Africa suspended use of the Astra Zeneca vaccine because data collected by South African authorities found it wasn't effective. It's being spun as it's not effective enough against the new strain. I suspect given a few months, we will find the vaccine doesn't work against any strain.

25% of the participants in the Pfizer and Moderna trials were over 65, 40% over 55, according your citations. 0 over 65 in the Astra Zeneca. There weren't enough over 65s and certainly not enough over 75s to come up with confidence intervals as tight as one would like in those subgroups for Pfizer/Moderna. Remember it's hard to recruit older participants because they often have co-morbidities that can cause all kinds of problems when you analyze the data. That's very different from Astra Zeneca. In particular, it's bad because the Astra Zeneca did actually recruit over 65s for the treatment groups. They just chose not to release most of that data while retaining some over 65s in the control.

This AM South Africa suspended use of the Astra Zeneca vaccine because data collected by South African authorities found it wasn't effective. It's being spun as it's not effective enough against the new strain, you still get the mild Covid etc by Oxford apologists. I suspect given a few months, we will find the vaccine doesn't work against any strain.

Already upvoted but also: thanks for the in-depth comment. I haven't been following vaccines beyond Moderna and Pfizer over the past few months.. woah.

We already have two other vaccines approved and being manufactured in large quantities. It’s not a zero to one situation.

I had no idea that the posting of this article was in any way related to current vaccine approval efforts, but I can assure you that holding a rhetorical gun to my head immediately after reading the article has done nothing but make me extremely skeptical of whatever product AZ is peddling.

Reply on news.ycombinator.com