Hacker News Reader

Comment by randcraw | Hacker News Reader

From the Nature article:

"Peak antibody levels were 3.5 times higher in those who waited 12 weeks for their booster shot than were those in people who waited only 3 weeks. Peak T-cell response was lower in those with the extended interval. But this did not cause antibody levels to decline more quickly over the nine weeks after the booster shot."

I care less about short-term peak antibody response than longer-term T-cell response or longest-term immunizing antibody response. On the latter two points, this study was not encouraging or outright silent. Nor was there any indication of how well these folks fared after they returned to the wild.

This info doesn't seem sufficient to draw meaningful conclusions about longer term protection.

View on HN

And how much antibody response is enough? Is more always better? Or is it diminishing returns?

IIRC one of the tricky things in the development of mRNA vaccines was being careful to not overstimulate the immune systems, triggering autoimmune types of reactions.

It seems plausible to me that there is a "safe" level of immune response, and either too much or too little could be less desirable (maybe for different reasons).

I'd also be curious to hear about an expert weigh in on "original antigenic sin" in this context. (My lay understanding: your immune system gets partially locked onto its first response to an antigen, and may fail to respond as well as it could to future variants).

If you power up the response to massive levels, does it also augment the downsides of that "original antigenic sin"?

This is exactly what they are finding.

https://www.medrxiv.org/content/10.1101/2021.05.03.21256520v...

The Pfizer vaccine is shown here to elicit increased cytokine immune response to many types of non-covid agitants like mold and bacteria. The danger there would be cytokine storm in the extreme or cytokine-induced inflammation in mild cases.

Might be chance (it is spring) but my allergies were rocked after the shots. I wonder if revving up your immune system also revs up allergies since they're just an immune over-response?

Fascinating - that was my experience as well. But I can't say whether it's just been an especially bad allergy season in general.

It hadn't occurred to me that the two could be related, but my allergies were pretty quiet until I got the first shot, then it was like I'd just motorboated a vase full of pollen.

Again, could just be timing.

Second shot in a week, so we'll see if it happens again!

I was hammered by allergies before I had any shots. After the second shot my allergies got a lot better. Probably totally cured my allergies. Magic.

Except it didn't because past week or so they've been back again.

The second Pfizer shot made my allergies better for about two days. They're back :(

Could it be because the immune system is busy with something else than fighting harmless agents ?

It’s a known strategy.

One guy infested himself with various worms to give his immune system something productive to do.

I myself didn’t really eat for a month once. That ended a multi month long flare. Idea is it kills off immune system for awhile.

Your body has a lot of possible immune responses. Runny nose, inflammation, fever. My guess is that it just switches strategies to something less annoying.

cmon guys it's mid may, it's peak grass pollen season (in the northern hemisphere)... that has nothing to do with vaccines. smh, anectdotal this, anectdotal that

Anecdotal, yes: I've never had allergies before (at least not ones I acknowledged) and now that I'm fully vaccinated, I've got crazy responses. Interesting nonetheless.

Are you saying you not only had an allergic response afterwards, but that your new allergies persist to this day? If so, very interesting. I wonder if a severely-heightened immune response could cause your body to sensitize to otherwise harmless agents if the conditions are right?

Which shot did you get? The different vaccines also use different adjuvants (or even none in case of BionTech), so maybe that also plays a role.

It is worth noting that cytokine responcse is implicated in many models of dementia, including alzheimer's. It is terrifing to imagine senarios where this leads to a billion early onset cases a few decades from now

Normal Covid infection apparently causes cytokine storms as well, so maybe there was no escaping it assuming every individual would eventually either have Covid or get the vaccine.

I hate to even say this, but maybe it's possible that excessive vaccinations increase mental diseases in the elderly?

and too few vaccinations increases fatal sickness for the elderly.

Or just a general lack of elderly

Why are you asking this? This seems fairly out of left field.

not to worry ! Lots of cytokine inhibiting drugs on the market, like anti-tnf drugs and anti-IL biologic drugs. All for the low price of about $5-10k/month. Ask your doctor about ustekinumab/adalimumab/vedolizumab. May cause cancer, demyelination and death.

As pointed out by the preprint, the same effect is observed in serval other common vaccines (BCG, Measles, etc.)

Trial data for the mRNA vaccines has not yet indicated an elevated risk of experiencing a cytokine storm.

There’s really not much to get excited about here.

If the infection model of Alzheimer's Disease is correct, then it might actually provide a protective effect.

Related to this, vaccinations are correlated with a lower risk of Alzheimer's:

https://www.alz.org/aaic/releases_2020/vaccines-dementia-ris...

Would this manifest as increased side effects from the vaccine?

Yes, this is why I’m curious whether the MRNA interacts badly with pregnancy, since pregnancy involves a poorly understood modulation of the immune system (so it doesn’t attack the fetus)

And GI function, which affects us all, every day.

The data on pregnancy shows no impact. In fact if I remember correctly there were fewer incidents than the placebo group. So maybe it helps reduce pregnancy related issues??

For a statement written with such apparent certainty, a link to a study (or any similar substantiation) would be pretty useful

There are multiple studies:

https://www.nytimes.com/2021/05/13/health/vaccine-pregnancy....

And from Pfizer and Moderna's research: https://www.latimes.com/science/story/2021-01-27/covid-19-va...

There is now a large trial of pregnant women getting the vaccinces. So far no evidence to suggest it is problematic.

Hmm, the first trial is just 103 women, the second one is 84. Wow wait, the Pfizer one was _12_ women and Moderna was 13. I'd be more convinced when a more comprehensive trial with thousands of participants is done. Still, thanks for the links!

Some more info from the uni that performed the study[1]:

> In relation to the cellular (or T cell) immune response, which plays an important role in supporting and maintaining antibody production, the team found that within the three-week interval group, 60% had a confirmed cellular response at two to three weeks following the second vaccine - although this fell to only 15% eight to nine weeks later.

> The proportion of participants showing a cellular response in the 12-week-interval group was only eight per cent at five to six weeks after the first vaccine, but this rose to 31% two to three weeks after the second vaccine. Research is required to further explore these variations in responses.

You're correct that the evidence isn't dispositive yet, but it's an encouraging signal. Antibody response in the short-term is positive and in the medium-term cellular response possibly converges.

More importantly, the implications for the first doses first strategy are positive. Those who know how to reason under uncertainty saw the UK's strategy as a calculated gamble based on a model that extended the point information provided by the vaccine efficacy RCT; critics were generally incoherent but their claim was that it was likely to cause large _negative_ effects on efficacy, worth paying the cost of vaccinating far fewer people.

On top of the priors that led to the decision, we now have signal from the way the winter surge played out in the UK, and the increased antibody response alone provides further weak signal that the delay was worth it - and conversely, that the US's approach cost lives).

The evidence still isn't dispositive, but crucially, there's no such strong basis for believing the alternative either: that extending doses _reduces_ efficacy and that vaccine availability should be limited in the name of adhering rigidly to the RCT-studied timeline.

The status quo bias of medical culture killed a lot of people during this pandemic by continuing their habit of pretending that certainty exists where none does and refusing to engage in critical thinking about uncertainty. Refusing to explicitly extrapolate a model from few high-quality samples really just means "implicitly extrapolating a crappy model through bias and guesswork". This may be adaptive in the normal medical context, where the status quo is generally fine, but is an incredibly poor fit for a fast-moving pandemic that was killing thousands a day in the US alone.

[1] https://www.birmingham.ac.uk/news/latest/2021/05/covid-pfize...

I completely agree with everything you said.

That said, the most effective defense I've seen for the seemingly incoherently cautious status quo approach is that the medical community doesn't operate on the span of one pandemic, but is rather in a perpetual struggle. So even if the short-term expected value of an excessively cautious approach is negative, resulting in more deaths and a longer pandemic, it might be outweighed by reducing the chance of damaging public faith in medicine long-term by an erroneous more aggressive move.

I still don't agree, but that very long-term view is the only defense I see for the approach given what we knew at the time.

Thank you for bringing that up. That makes a lot of sense, and public health absolutely does need to consider mass psychology and their credibility.

The theory that they're overindexing on protecting that credibility doesn't really comport with their behavior the rest of pandemic, however. Leaving aside all of the blood on the hands of mindless bureaucracy (like effectively banning testing for Covid in the early pandemic when it was most crucial), there has been decision after decision that prioritized status quo bias and RCT-worship even when the long game suggests it would damage their credibility (the fervent anti-mask recommendations in the early pandemic is a great example, as is the bizarre insistence on recommending precautions against fomites over aerosols until extremely late).

More importantly, "public health is misleading you at the cost of your health/life because they're playing the long game" doesn't actually preserve much credibility. It certainly casts public health figures in a more positive light, but I'm not really thinking in terms of blaming individuals over institutions/cultures anyway. It doesn't really matter _why_ public health advice is wrong in obvious and significant ways. What matters is the ultimate conclusion: if you have basic scientific literacy and an IQ over 105, form your own broad understanding of the science. This isn't a call to jettison epistemic humility: you're not going to be able to understand things anywhere close to as well as an individual epidemiologist can. But the point is that listening to the science as filtered through public health advice _is even worse_, in the same way that getting nutrition advice from the USDA will.

> as is the bizarre insistence on recommending precautions against fomites over aerosols until extremely late

I watched Contagion for the first time last week, and they had the same attitude toward fomites, where it was spread through knobs and cash registers.

I’m convinced that the movie was interpreted as a documentary on what to do rather than science.

Now I question how much hand-washing helps prevent the spread of other respiratory viruses.

Looks like, for flu, in this study, fomites were responsible for 4% of cases:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6121424/

… and

> Mask wearing is much more useful than hand washing for control of influenza A in the tested office setting. Regular cleaning of high-touch surfaces, which can reduce the infection risk by 2.14%, is recommended and is much more efficient than hand-washing.

But I’m just a guy that spent 2 minutes searching studies on Google of respiratory viruses, why make decisions so easily based on that when you can base decisions on a movie on Netflix.

Very interesting, thank you. Call it Gell-Mann Amnesia, but I had not thought to extend my skepticism of the basic competence of public health authorities to flu recommendations. I suppose my model is that they were bad in the fast-moving epistemic environment of the pandemic, and that this wouldn't extend to relatively stable and well-understood cases like influenza.

Is having a lower peak cellular response meaningful in the long run, or not so much?

Great question, that I don't know the answer to. The answer to that question is part of the uncertainty that attenuates the signal I described as "weak". I think the reason I like the approach I'm describing is that you can model uncertainty at every level of knowledge: you don't need to become an epidemiologist to reason about the facts (with correspondingly higher error bars than a domain expert would have), _and_ you don't need to blindly trust that experts in every domain are somehow magically optimal when it comes to reasoning under uncertainty.

To put it another way, that's the kind of information I would spend the time digging into once knowing the answer would push my model across a decision threshold. I look forward to seeing the follow-up research, which is usually the quickest way to fill in gaps in the "primitives".

Isn't there a potential downside to the UK approach in an environment where variants are increasing? I've seen some data suggesting that with only one jab the mRNA vaccines are only about 20% effective against the UK, SA and Brazilian variants (no data yet on the India variant), but that number jumps in to the 70% range with the 2nd dose. A delay of 12 weeks leaves a lot of people in that 20% efficacy range for those variants. Now some parts of the UK are starting to see increasing cases of the Indian variant.

> with only one jab the mRNA vaccines are only about 20% effective against the UK, SA and Brazilian variants

Do you have a source for this I could see? I did some light Googling and couldn't find anything that said as much; it also runs contrary to the impact of Britain's choice of widespread vaccination over incremental effectiveness gains (though this is of course just a correlation). A study I did find[1] looking at effectiveness in the UK has first-dose effectiveness at 70%, roughly as expected (I'm using the UK here as a noisy proxy for the UK variant, but I think that's reasonable).

Leaving aside the veracity of this concern, the counterfactual was facing the massive winter surge many countries saw while intentionally choosing to vaccinate half as many people. It's not clear at all that it would be good policy, especially given the ability to change policy[2] as the reality on the ground changes. Seeing how public health has conducted themselves throughout pandemic, I simply don't have the blind faith that they handled this cost-benefit analysis competently.

To pick just one example I came across[3]: as much as I respect Bob Wachter, the Chair of Medicine at UCSF, he was exemplary of this flaw in critical thinking ability that medical culture indoctrinates into HCWs. He came around to FDF on New Year's, after a month of arguing against it. It's amazing to see him discover the concept of quantitative rigor in the face of uncertainty when speed is crucial. This is especially striking when contrasted against his prior argument in the opposite direction, which consists of handwavy platitudes around "unnecessary curveballs".

My sister and her husband are both (Ivy League-educated, very successful) physicians, one a clinician and one a clinician scientist. We've had a decade-long conversation about this tendency, the degree to which it exists, the degree to which it's salutary, how it should affect patient/informed-citizen behavior, etc. Part of my relative confidence in my model of this epistemic culture is that these pitfalls were entirely predictable, and indeed, I pre-registered concerns[4] about the pandemic's characteristics and how modern medical culture was likely to interact with it.

This isn't as damning an indictment of medicine as it may sound (the FDA's Vogonic bureaucracy fetish has way more blood on its hands than any of the flaws in medical culture). And one thing I've picked up from all my conversations with my brother-in-law is that the average consumer of medical information is almost unbelievably stupid (not how he would phrase it) and is far better off following public health (and nutrition) advice as dogma than trying to understand and pick it apart. But if you are basically scientifically-literate, say to the point that you can read and mostly understand the paper I shared in [1], you're likely far worse off uncritically accepting their epistemic flaws than using them as a baseline and doing your own research (and as always, retain a heaping dose of epistemic humility around your error bars and counterfactuals)

[1] https://www.bmj.com/content/373/bmj.n1088

[2] https://www.nytimes.com/2021/05/14/world/india-covid-second-...

[3] https://twitter.com/Bob_Wachter/status/1344667655324585986

[4] This was in conversation with them. Sadly, I have no online proof I can share without self-doxxing

One study from Qatar: https://www.nejm.org/doi/full/10.1056/NEJMc2104974

Do you have a source for this number? All of the evidence I've seen shows the mRNA vaccines are in the 80% effectiveness range after a single dose, but that data was not specific to any variants.

One study from Qatar: https://www.nejm.org/doi/full/10.1056/NEJMc2104974

    For preventing infection:
    B.1.1.7: 29.5% effective after one dose (Pfizer) 
    B.1.351: 16.9%

See Table 1 there for more details.

Would have been really nice if they had separated the effectiveness data by days after first dose, we know the curve is substantial for the first few weeks so it would not be surprising for people to still catch it in week one, I would be surprised by data showing <50% effectiveness in week three and four.

I believe the author said people were misinterpreting the study. It included people who caught it before the protection of the first dose kicked in.

Canada has some opposing data: https://www.cdc.gov/mmwr/volumes/70/wr/mm7020e2.htm

Lots of people in the medical/scientific community expected this result. The best explanation I have heard is that the vaccine companies planned these sub-optimal trials with 2-3 weeks between shots, because it meant getting the vaccines to market much more quickly. Instead of having vaccines available in December, they might not have been available until February.

So far, the evidence is pretty compelling that the vaccines have continued efficacy 6-7 months post-administration. It's obviously impossible to know much beyond a couple months however.

Mind you, there's a delicate balancing game here though. Slower vaccines, even if more optimal, come with their own costs.

> This info doesn't seem sufficient to draw meaningful conclusions about longer term protection.

Is it because the study period is just that i.e. 12 weeks? Even the studies on natural immunity from COVID suggests at least 6 - 8 months antibody persistence because the study period were just that[1][2]. Note that policies are being made now to defer the jab for those who got infected by COVID, based on these studies.

We'll likely see studies confirming long term protection from vaccines & natural immunity after disease in coming months. Btw, How much is long term?

[1]https://www.nih.gov/news-events/nih-research-matters/lasting...

[2]https://www.nejm.org/doi/full/10.1056/NEJMoa2034545?s=09

[3]https://twitter.com/PTI_News/status/1392715110821494789

Does this suggest anything about side effects? Is it likely to make them better, worse, or none at all?

I'm aware that there's not enough information to say yet, but it's something I'd like to see them look out for. The side effects are absolutely not a good enough reason to avoid getting vaccinated, both for your own sake and as a participant in herd immunity, but I've personally witnessed enough unpleasantness to know that there will be people who avoid it just to avoid the side effects.

Being able to provide good advice to help the rational but fearful may help get us over the threshold, despite the large numbers of irrationally fearful. We'll need every single one to get there.

Anecdotally, in my home, my wife had chills and general ickyness from about 12-36 hours post 2nd dose. Myself and my teenager had no side effects at all. Even the soreness at the injection site was less after shot #2 than it was for the first.

Don't let fear of side effects drive you away from getting the necessary booster. As of today, Fauci says I can ditch the mask. I'll still be keeping one in my pocket just in case though.

I really wish more were being done to openly address the more extreme cases of side effects, so we could try to find other contributing factors and perhaps mitigation strategies.

The number of poor reactions is very small, but pretending like they don’t exist or failing to address ways to overcome them only further fuels the fear in those hesitant to get vaccinated, and furthers the divisiveness in an issue we should all be working together on.

Which makes sense if you’re a politician or Dr. Fauci, but doesn’t make sense if you just care about the health and safety of everyone and want to get this damn pandemic over with already.

Forcing or shaming the hold outs to get vaccinated won’t do this, but giving honest confidence in the vaccine to people will.

in particular, a linear change in antibody levels, which trigger an exponential response, doesn't really mean very much!

And 12 weeks is a long time. I honestly would rather get an immunity boost after 3 weeks than wait the 12 for a second. A lot can happen in twelve weeks.

Reply on news.ycombinator.com