Re: Guest Post - Covid questions

1

There also appeared to be an industrial-labor thing going on in some places where hospital industry leaned on "not airborne" as justification not to go harder on PPE, and unions were, justifiably I think, mad and pointing at airborne evidence.


Posted by: Schuyler Colfax | Link to this comment | 05-10-21 6:35 AM
horizontal rule
2

Question 6 is interesting, because AFAIK developing the mRNA vaccines took about a weekend after the virus was sequenced, and that doesn't take long either. The long delay was all about safety and efficacy testing to meet regulatory standards.

I'm all for regulatory standards, but there's clearly got to be some kind of room for speeding things up procedurally. I have wondered about leapfrogging efficacy testing -- once a vaccine is tested and established to be safe, and there's good theoretical/lab based reason to think it should be effective, would it be so bad to roll it out broadly on an emergency basis while the efficacy testing is happening? Expensive and possibly pointless, but six months earlier on the vaccine rollout would have saved a lot of lives.


Posted by: LizardBreath | Link to this comment | 05-10-21 6:54 AM
horizontal rule
3

3 is badly phrased. It should be more like, "Can we shrink the global infection reservoir quickly enough that there won't be time for a vaccine-resistant variant to break out?"


Posted by: MC | Link to this comment | 05-10-21 7:03 AM
horizontal rule
4

I think diseases are usually less deadly when you have been exposed to a similar one. Even if Covid evolves to get around vaccination, it may be less deadly for the simple reason that we still have some immune response. Flu mutates even faster, but people have lifelong better protection against the first strain of flu they were exposed to.


Posted by: Unfoggetarian: “Pause endlessly, then go in.” (9) | Link to this comment | 05-10-21 7:44 AM
horizontal rule
5

2: I don't think the annual flu vaccine goes through clinical trials each time. I can imagine that covid updates, if necessary, would more likely follow that path than having to run the full testing gauntlet each time.


Posted by: Doug | Link to this comment | 05-10-21 7:50 AM
horizontal rule
6

2: Safety is established from the same clinical trials as efficacy, and for obvious reasons can't be speeded up all that much. If everyone in your clinical trial drops dead after 03 days, that is useful data point to have before too many people have been vaccinated. If one tenth of one percent drop dead after 30 days, while the disease has a somewhat lower mortality rate, that's useful information to have before you vaccinate 100 million people.

While it's apparently true that the first vaccines approve were developed over a weekend, tht doesn't mean that every vaccine developed that weekend was approved. Merck, among others, started out with promising vaccine candidates (maybe also developed in the same weekend) but they never got past early safety and efficacy tests. If those had been distributed on a wider scalre before adequate testing thst would have been a Bad Thing.


Posted by: unimaginative | Link to this comment | 05-10-21 8:30 AM
horizontal rule
7

6: You very plausibly know more about this than I do -- I know almost nothing. But my understanding is that safety trials are Phase I, and are completed before the efficacy trials in later phases, like so: https://www.fda.gov/patients/drug-development-process/step-3-clinical-research

I'm not saying that this process should be shortened for vaccines generally, but it seems as if in an emergency there might be some use for provisionally approving a vaccine that's safe but only probably effective, as opposed to waiting to be certain it's effective.


Posted by: LizardBreath | Link to this comment | 05-10-21 8:36 AM
horizontal rule
8

7: Isn't the vaccine-escape concern a reason to not rush to give out maybe-effective vaccines?


Posted by: Minivet | Link to this comment | 05-10-21 8:39 AM
horizontal rule
9

Spell that out?


Posted by: LizardBreath | Link to this comment | 05-10-21 8:42 AM
horizontal rule
10

I'm dubious about the premise (that immunized human human lungs really matter in presenting new challenges to the virus).

A Bat's immune system is an insanely hostile place for invaders compared to those of other mammals, I think roughly because flight demands a very high metabolism compared to anything else.
So thinking about reservoirs needs to somehow weight by how tough the reservoir environment is from the virus' perspective, and my claim is that bat sinuses and lungs punch orders of magnitude above their volume from this perspective.
This one's paywalled but good and current:
https://pubmed.ncbi.nlm.nih.gov/33473223/
Maybe I can find something more apposite later.

Here's a fantastic essay for thinking about small particles afloat:
https://science.curie.fr/wp-content/uploads/2016/04/Purcell_life_at_low_reynolds_number_1977.pdf



Posted by: lw | Link to this comment | 05-10-21 9:19 AM
horizontal rule
11

I don't think I understand 10.
Are 10.1 and 10.2 separate points or connected?
Is 10.2 saying non-bat hosts are benign environments where coronoviruses go soft and decadent?


Posted by: Mossy Character | Link to this comment | 05-10-21 9:29 AM
horizontal rule
12

2: I think that's the plan for variant updates, a similar regulatory pathway to the way the flu gets updated every year.


Posted by: Bostoniangirl | Link to this comment | 05-10-21 9:37 AM
horizontal rule
13

5 beat me to it, but I feel fairly confident that the FDA explicitly said they would do that for updated versions.


Posted by: Bostoniangirl | Link to this comment | 05-10-21 9:39 AM
horizontal rule
14

Yes, hypothesis to reject before proceeding with further reading is that non-bat environments are relevant to thinking about the virus' evolution.

I don't think one can extrapolate from antibiotic-resistant strains of bacteria or fungi like C auris , those are not like viruses. Looking at viruses where immunization has lost effectiveness due to adaptation of the virus would be a starting point, maybe in herd animals? There's a literature about viral adaptation to new hosts, that's a subject I'd like to know more about-- basically to think about this, to get a suitable framing, I would suggest to look at relevant solved or understood problems rather than the pile of maybes that is this virus.


Posted by: lw | Link to this comment | 05-10-21 9:39 AM
horizontal rule
15

And I don't think I held or need to hold 10.1 as a premise.* Specifically, the S. African variant AIUI developed AZ resistance by accident, not as a response to AZ: when AZ was rolled out the resistance was already there. Similarly the double-mutant (or some such terminology) variant in India.**
*Pretending here that I'm arguing we should be worried. I want reasons not to be worried.
**Which AIUI has experts worried, even if not resistant.


Posted by: Mossy Character | Link to this comment | 05-10-21 9:39 AM
horizontal rule
16

whoops 1st sentence of 14 is a mess. s/relevant/irrelevant/ . Maybe I should wait before I can express myself more clearly to chime in


Posted by: lw | Link to this comment | 05-10-21 9:41 AM
horizontal rule
17

15. Oh sure, thinking about efficacy of particular vaccine against particular strain makes sense, presents challenges.

I was addressing the thought that human partial response will create environments that grow terrifying new strains, I don't beleive that's a direction worth thinking about.


Posted by: lw | Link to this comment | 05-10-21 9:44 AM
horizontal rule
18

15 without seeing 14.
17: What do you mean by partial human response? I'm not necessarily thinking about the human-virus arms race, just the sheer size of the petri dish the virus has to play in. I'm wondering if terrifying new strains could emerge independent of human response, just by the volume of reproduction the virus is getting in human hosts. OP.3 is asking how big a petri dish is needed to make such strains probable, and whether the dish can be shrunk fast enough.
I would suggest to look at relevant solved or understood problems rather than the pile of maybes that is this virus.
This is a great suggestion. I'd love for the biologically literate reprobates to take a shot at that and report back to us schlubs.


Posted by: MC | Link to this comment | 05-10-21 10:09 AM
horizontal rule
19

9: Isn't there a concern that insufficiently-effective vaccines could be more conducive to the virus learning how to get around the mRNA-directed antibodies?


Posted by: Minivet | Link to this comment | 05-10-21 10:18 AM
horizontal rule
20

Response to 7:
Start with a disclaimer: I know a fair amount about new drug trials, although my knowledge is a decade out of date, and not as much about vaccine trials.

All clinical trials concern safety as well as efficacy. Phase I trials are exclusively about safety, since the patients are healthy going in, it can't measure whether the drug cures anything. If the drug passes, it is considered safe enough to be tested further, not necessarily safe enough to be released generally.

The website you linked describes the purpose of Phase II trials as "efficacy and side effects," and Phase III as "efficacy and monitoring of adverse reactions." I.e., both efficacy and safety. Many drugs that show efficacy in Phase II and Phase III are rejected because the side effects or adverse reactions are too severe.

Also Phase I trials usually have less than 100 people, so the 1% side effects won't show up.

I agree that there may be short cuts that speed up approval in emergency situations without compromising the safety of patients very much. The FDA has an emergency approval process for this situation, and it was used for the Covid vaccines to great success. The government also speeded up distribution by committing to purchase large amounts of the vaccines before they were approved, figuring it was definitely worth risking a waste of money on a product, even if it was not yet worth risking a loss of life or health. Good idea! I'm not convinced that the process could have been speeded up much more than it already was.



Posted by: unimaginative | Link to this comment | 05-10-21 10:26 AM
horizontal rule
21

I don't know, but I'm pretty sure that while that may actually be a problem, it's not necessarily a problem. That is, if a vaccine researcher or someone else with relevant knowledge said that that is a problem with this virus in these circumstances, that'd be a really good reason not to shortcut efficacy testing. But I haven't seen anyone with knowledge saying that is a problem.


Posted by: LizardBreath | Link to this comment | 05-10-21 10:29 AM
horizontal rule
22

21 to 19.


Posted by: LizardBreath | Link to this comment | 05-10-21 10:39 AM
horizontal rule
23

I'm not convinced that the process could have been speeded up much more than it already was.

No kidding, this was the drug development equivalent of being dealt a straight flush.


Posted by: apostropher | Link to this comment | 05-10-21 2:02 PM
horizontal rule
24

The main reason I have to disagree with you, LB, is that your line is the libertarian line -- let everyone opt into vaccine trials and get rid of regulation, and everything will be great -- and they're usually wrong about everything.


Posted by: Cala | Link to this comment | 05-10-21 2:11 PM
horizontal rule
25

Contrarian take: by accelerating mRNA vaccine tech by years, COVID may end up saving lives on net if the tech works for high mortality diseases like TB or malaria where vaccines have had many failures.


Posted by: SP | Link to this comment | 05-10-21 3:07 PM
horizontal rule
26

Counterpoint: India.


Posted by: Moby Hick | Link to this comment | 05-10-21 3:11 PM
horizontal rule
27

25: Except Oxford had the malaria vaccine before the COVID one, or at least that is what I had been assuming based on their trial.

Re: shortcuts, I think in vitro efficacy would likely be sufficient for modified mRNA vaccines, but I thought the FDA hadn't produced a final guidance yet.

The biggest timesaver here was a clever trial design that overlapped phases. They also had a ton of money thrown at the issue to be able to afford to run virtually everything concurrently. Usually (because most things fail), it's a huge waste of resources to run things concurrently rather than sequentially.


Posted by: ydnew | Link to this comment | 05-10-21 4:09 PM
horizontal rule
28

A lot of this discussion is premised on the idea that the virus can successfully evolve its way past the antibodies spun up by the vaccines. There's relatively little evidence of this - the SA AZ situation is probably the strongest - and the virus doesn't have unlimited space to maneuver. A Spike protein that is sufficiently different to evade the antibodies has a good chance of being too different to actually infect cells.


Posted by: Nathan Williams | Link to this comment | 05-10-21 4:11 PM
horizontal rule
29

With pounds of salt:

2. SARS-CoV-2 mutates rapidly, and at least some variants reduce the efficacy of at least some vaccines.
The mutation rate isn't more rapid than other viruses per se, it's that there are lots more people infected, and it's pretty transmissible. Originally, scientists were relieved that it wasn't particularly rapid.

3. Can we get to global herd immunity before a vaccine-resistant outbreak?
We won't ever get to global herd immunity. We are still working on fucking polio erradication. India eradicated polio in 2014. We can get pretty good global coverage in a few years. It's not clear where exactly we need to be, but 80% seems conservative to me.

3.1 How much is global immunity compromised by subpar Chinese vaccines?
It's basically not. They are garbage compared to the good ones and I hope they don't kill people with bad manufacturing lots, but even a 50% efficacy (which it seems like is roughly where they are) is better than nothing. They are going to where there is little to no distribution of better options. Most importantly, they do not preclude being vaccinated with something that works better at a later date. There is literally no way the Moderna vax is going to be rolled out in resource poor areas with bad infrastructure. The better vaccines (and more coming, still) will act as boosters. Lots of vaccines require multiple rounds of immunization. Think of this as round 1.

4. What countermeasures are possible? Inoculation with multiple differently acting vaccines?
Yep. We need to do this. There's some data accumulating suggesting that mix-and-match will actually give better coverage. US companies are already designing and working on how approval will work for variant vaccines, which they are describing as "boosters." I suspect an eventual vaccine might have multiple mRNA sequences in one shot to improve coverage.

5. Can variants be distinguished by tests alone, without full sequencing?
Yes and no. You could make rapid tests to look for a particular strain, but they won't be as "smart" as sequencing since they will find only what you're looking for. If you are looking for new, emergent strains, sequencing gets you that. If you're tracking a particular strain at levels where you might want to deploy a rapid test, you're probably already a little behind.

6. How rapidly can vaccines be adapted? Might they have to be developed from scratch, with the same 12-18 month lag?
Depends which vaccine you mean. The JNJ and AZ would take longer to design and work out manufacturing conditions because they require cell culture. The mRNA vaccines would be quicker. Realistically, for the adenoviral ones, I'd guess 2-4 months until you have a process with full QC ready to roll out. For mRNA vaccines, maybe 2-4 weeks (except sterility testing outcomes because those take 2 weeks to get the results).

7. We have animal reservoirs in the original bats, pangolins/civets, mink, cats, dogs, and what else? In theory, what are the risks of jumps into livestock, poultry, rodents? What surveillance countermeasures are possible?
Depends how much you want to spend. Frankly, COVID doesn't seem especially transmissible in a lot of species in the same way, so it probably isn't worth losing sleep over. Are you worried about the livestock or the farmers? I'll go hunt through literature, but I think if it were going to be widely transmissible, we'd know by now.


Posted by: ydnew | Link to this comment | 05-10-21 4:33 PM
horizontal rule
30

Sorry, more on 6: the mRNA ones would not at all be "from scratch." Just some minor molecular biology adjustments to the mRNA, all else the same. The adenovirus ones would also be a fairly minor adjustment. My biggest worry with those is that if you keep boosting, it's possible-to-likely that recipients' immune systems will start responding to the adenovirus vector, and we'll lose these as a good option. Or we'll need different adenovirus vectors, which would mean back to the drawing board.


Posted by: ydnew | Link to this comment | 05-10-21 4:37 PM
horizontal rule
31

30: My uninformed thought before the VITT was that everybody should start with a dose of an and J and then 8-12 weeks later get an mRNA vaccine as a boost.


Posted by: Bostoniangirl | Link to this comment | 05-10-21 6:14 PM
horizontal rule
32

My dad wants me to get 1 shot of Moderna or Pfizer to supplement my J&J. I get that he feels that way as a dad, but it still feels wrong at this point in time. Plus I don't really feel like feeling so lousy again.


Posted by: heebie | Link to this comment | 05-10-21 7:22 PM
horizontal rule
33

If you get just one shot, don't you show up in the statistics as an incomplete or something?


Posted by: Moby Hick | Link to this comment | 05-10-21 8:16 PM
horizontal rule
34

33: I look at the daily dashboard for our state. There have a calculation for 1st shot Moderna or Pfizer, J and J shot, 2nd dose Moddern and Fizer, and fully vaccinated which is the J and J folks plus the people who got their 2nd shot of an mRNA vaccine.

32: J and J is conducting a 2-dose regimen now (57 days post shot or something like that).


Posted by: Bostoniangirl | Link to this comment | 05-11-21 2:09 AM
horizontal rule
35

The UK is trialling mixed vaccine dosing regimes, too.


Posted by: nattarGcM ttaM | Link to this comment | 05-11-21 4:21 AM
horizontal rule
36

32: Wait, like they're offering a booster to people who got the first?


Posted by: heebie-geebie | Link to this comment | 05-11-21 6:55 AM
horizontal rule
37

36: It was a research protocol to see if they got more durable and stronger immunity. It's called ensemble 2. I think they were a bit concerned about the strength of the response in the over 60's.


Posted by: Bostoniangirl | Link to this comment | 05-11-21 7:02 AM
horizontal rule
38

Thanks to all above.


Posted by: MC | Link to this comment | 05-11-21 2:22 PM
horizontal rule
39

Are you worried about the livestock or the farmers?
I'm worried about livestock becoming a reservoir for retransmission to humans. If Covid death rates in livestock were comparable to those in humans, there wouldn't be a huge impact on food supply (until the culls began).
I think if it were going to be widely transmissible, we'd know by now.
Would we? For livestock in developed countries, probably.* But livestock elsewhere? Rodents and pets everywhere? I'd hazard also cat and dog outbreaks are more likely in poorer countries, assuming bigger free-roaming feral populations.
*Though I'd hazard there are far more opportunities for human>animal transmission in the poor world, even if any given transmission is less likely to result in an outbreak in a battery farm. Which goes also for human>rodent.


Posted by: MC | Link to this comment | 05-11-21 4:48 PM
horizontal rule
40

I kind of believe the theory that the intermediate host for covid was farmed mink in China. The evidence being that it spreads so easily to minks in the rest of the world and China claims they've had no mink outbreaks, so they're obviously lying and the obvious explanation of why they'd lie is that they know that's where the outbreak started.


Posted by: Unfoggetarian: “Pause endlessly, then go in.” (9) | Link to this comment | 05-11-21 5:02 PM
horizontal rule
41

I don't trust mink (basically weasels) or the Chinese government.


Posted by: Moby Hick | Link to this comment | 05-11-21 5:04 PM
horizontal rule
42

I guess my Bat and Pangolin Heavy Petting Zoo should continue to stay closed.


Posted by: Moby Hick | Link to this comment | 05-11-21 5:07 PM
horizontal rule
43

39: Gotcha. Preliminary research shows a lot of labs purposely innoculating a variety of species and looking for infection and transmission. So far, cows, pigs, ducks, and chickens don't look especially susceptible (viral replication in 2 of 6 cows in an experiment meant to infect). Cats, dogs, ferrets, minks, and fruit bats are susceptible. Ferrets and minks aren't that surprising. There's a lot of overlap with respiratory virus susceptibility between humans and ferrets (folks test lots of influenza vaccines and therapies in ferrets). Human-cat or human-dog transmissions are relatively rare, requiring very close household contact with a very sick human, and I'm not sure there have been any pet-to-human cases, although those would be very hard to trace in a household. Also, feral dogs and cats are outdoors, which would help keep transmission low.

I trust the researchers trying to find reservoirs. I also figure the US, with its horrific case rates combined with Big Ag and lots of ag schools would have found something if it were likely to be found. I might be wrong, but this seems like something actively being researched. Not sure about surveillance on livestock, but I can tell you that research labs and zoos are looking for this.


Posted by: ydnew | Link to this comment | 05-11-21 6:13 PM
horizontal rule
44

Covid goes from the Trump family women to the Secret Service to the community.


Posted by: Moby Hick | Link to this comment | 05-11-21 6:43 PM
horizontal rule
45

43: Thanks! Rodents?


Posted by: | Link to this comment | 05-11-21 6:49 PM
horizontal rule
46

No thank you. Already had dinner.


Posted by: Moby Hick | Link to this comment | 05-11-21 6:50 PM
horizontal rule
47

45: Results there are evolving. Early variants were not infectious in lab strain mice or rats. Hamsters can be infected and transmit it. Genetically engineered mice with a human ACE2 can get it.(These were developed in the early 2000s for SARS research.) More recently, it has been found on surveillance in deer mice, a common rodent in the US, and the B.1.351 and P1 variants do appear to be able to infect lab mice, but there isn't any transmission data yet, even rodent-to-rodent.

When this is researched, there are a lot of ways to "infect" animals that don't have a lot of connection to real world conditions. For example, a super common way for respiratory viruses is to pipet a solution of cultured virus into a mouse's nose. A more realistic experiment uses a device that aerozolizes virus and puffs it into their lungs via a nose cone. Then, infected mice are put back wither with cage mates (a "household") or adjacently caged animals with shared air. These are not issues for the types of work being done, but if you want a natural experiment about transmission where the goal is to mimic real-world conditions in terms of viral exposure, it just isn't an easy experiment to run - you need to have low viral load exposure and lots of animals to understand that. Environmental monitoring (which is being done) is probably more realistic.

For rodent-to-human transmission to occur, you'd need to be very close, breathing the same air, as a mouse with high enough viral load to infect, and a strain that would be infectious to both. I think it doesn't seem likely, but I guess it's possible.


Posted by: ydnew | Link to this comment | 05-12-21 2:08 AM
horizontal rule
48

47: Fecal transmission? Agreed rodents don't look likely, but they're the quarantine problem from hell, so. And maybe rodent ←→ cat /[domestic animal] ←→ human transmission more likely?


Posted by: MC | Link to this comment | 05-12-21 2:31 AM
horizontal rule
49

I think there has been an animal reservoir for the plague in America for decades but it hasn't been a problem for humans.


Posted by: Moby Hick | Link to this comment | 05-12-21 5:04 AM
horizontal rule
50

bacteria!=viruses!!! Waaahh!


Posted by: | Link to this comment | 05-12-21 5:40 AM
horizontal rule
51

Oh, right. Also, the flea is a force-projection platform launched from a rodent.


Posted by: Moby Hick | Link to this comment | 05-12-21 5:43 AM
horizontal rule
52

And before the reservoir stopped being a problem it was a very big problem for 1000+ years.


Posted by: mc | Link to this comment | 05-12-21 5:49 AM
horizontal rule
53

That was a different continent.


Posted by: Moby Hick | Link to this comment | 05-12-21 5:51 AM
horizontal rule
54

Three continents! And plenty in N America 19C.


Posted by: | Link to this comment | 05-12-21 6:16 AM
horizontal rule