Can you reassure me about the UK government's vaccine rollout plans?

[andrewducker]

The UK is currently planning on rolling the various vaccines out with three months between the first dose and the second one. The idea being to give the first dose to as many people as possible, as soon as possible, giving them a bunch of resistance to the drugs, and that this is better than giving less people more resistance, and overall will save lives.

The UK government has stated that "Short-term vaccine efficacy from the first dose of the Pfizer-BioNTech vaccine is calculated at around 90%" and therefore "Given the high level of protection afforded by the first dose, models suggest that initially vaccinating a greater number of people with a single dose will prevent more deaths and hospitalisations than vaccinating a smaller number of people with 2 doses".

Which all seems, well, pretty sensible actually. So why have the WHO, Pfizer, Dr Fauci (head of the US department that deals with infectious diseases) and some other vaccine specialists said they don't recommend it?*

If we look at the Pfizer data, they did dosing tests**. 10µg, 30µg, and 100µg. The last of which they did only one dose of (possibly to test if this could be a one-dose vaccine). The second dose is given on day 21 (for the 10µg and 30µg doses), so we can see the results on day 28 of giving a second dose. The numbers are the immunogen concentrations.

Here are the results.

	Day 7	Day 21	Day 28	Day 35
10	0.9	534	4,813	5,880
30	0.8	1,536	27,872	16,166
100	1.2	1,778	1,260	N/A

And if you look at that, you can see that a week after the second dose the first two trials are looking great - they result of the second dose is that immunogenicity skyrockets. The results of not giving a second dose is that it...drops. They didn't bother with testing those people at day 35. Presumably because it was clear that two doses were needed.

So if we're dependent on the second dosage, why do the UK government think that 90% is going to be achieved from dosage 1? Turns out it's because they've looked at the figures in the Stage 3 study***, where it says that during the week *after* the second dose is given the efficacy is 90%, and assumed that because it's so soon after the second dose is given, that the second dose can't be responsible, that those figures must be what you'd get without it. But for the 30µg dose (which that graph is for) the immnogenicity shoots up from 1,500 to 27,800 (almost 20 times!) during that week! I can't see how that couldn't easily be responsible for the increase of immunity from 50% to 90%.

So, frankly, I'm finding myself very sceptical about the government's line there. And I would like someone to reassure me that I'm misunderstanding in some way.

To be clear - if you get offered a vaccine, take it. Any immunity is better than none, and it might still be better to cover the country with 50% immunity than to give half the country 95% immunity. But I am very troubled that the government's advice isn't adding up.

(Many thanks to Kenny, whose thread on this started me digging to reassure myself.)


*Some have also come out in favour of it. It's by no means a wash.
**Pfizer Stage 1/2 study. All numbers are from Figure 4.
***Pfizer stage 3 study. The important bit is Figure 3, which compares immunity on different dates.

Comments

[maia]

Is there any chance that partially vaccinating people will encourage the spread of mutations that allow the virus to evade the vaccine?

[andrewducker]

An excellent question!

[mountainkiss]

Yes.

[orangesquare]

Good information on the mutation question here.

More information and links here as well.

[jack]

This is a bit that I'm worried about (although presumably having some countries mostly vaccinated and others mostly not vaccinated will have similar risks anyway).

I hear there's some hope that the RNA vaccine targets spiky bits which make it more infectious, so would make a mutation without them less infectious, but obviously that's just a guess.

[aldabra]

Surely this is all public enough that Pfizer would phone them up and tell them this wasn't going to work?

[mountainkiss]

They have.

[aldabra]

I know they've said there's no data to support it; that's an obvious consequence of the research design. But they could also say their research is not consistent with the interpretation being put on it, which is a much stronger refutation.

[mountainkiss]

This might not be their interpretation?

[mountainkiss]

The reassurance I can offer is that Chris Whitty really does know what he's talking about. He is an epidemiologist, and an extraordinarily clever man. There could be no better candidate anywhere for CMO, and I mean that worldwide.

Promptly undercutting the reassurance is that the decision he's being asked to make isn't "is it okay to vaccinate more people once?" It's "Is it better to vaccinate more people once or fewer people twice, in the context of NHS capacity to treat cases?" So he is deciding on the better of two options, both of which are lousy.

Destroying any remaining peace of mind, the thing that worries me is that this entire plan is dependent on the Government being able to roll out and administrate vaccines in a way that makes sure that there really is only twelve weeks between the two shots.

[mountainkiss]

Actually, I also stand by my FB comment. The political fallout of this decision is so unabashedly negative that there must be significant perceived medical benefits for this decision even to have been taken. Context still per above, though.

[mountainkiss]

My guess - and this is a 100% guess; I am not a medic and not an epidemiologist and have no inside knowledge - is that the goal is to prevent serious illness (and hence (i) mortality and (ii) strain on NHS), rather than reducing case numbers. I would further guess that there is medical reason to infer that more people receiving smaller vaccine dosages maximises this effect. But guesswork.

[andrewducker]

hey seem like reasonable guesses though!

And I do hope it all works out!

[danieldwilliam]

I think you are right. The aim is to reduce the number of deaths rather than the number of infections.

[hilarita]

That doesn't necessarily help. It certainly doesn't reassure me.

I've got a relatively low likelihood of dying from covid (lucky me). I am however already disabled, and can't really afford to get any more disabled than I already am. I thus want to avoid getting infected, because I need to avoid "long covid". So reducing the number of deaths, but not the number of infections doesn't really help me much.

[danieldwilliam]

I see that it wouldn't.

[toothycat]

I am not an expert in vaccines or viruses, but looking at the data and the papers, two things jumped out at me. The first is the comment in the phase 1/2 paper: "Because the titre at which human neutralizing antibodies are protective remains unknown, these findings are not proof of vaccine efficacy. Efficacy will be determined in a pivotal phase III trial." The second is the graph showing efficacy in the phase 3 paper (Fig 3). It shows a levelling off of the incidence of Covid-19 infections from 12 days after the first dose, and the second dose does not affect the slope after that. Unfortunately we don't have data showing what happens with a single 10ug or a 30ug dose - it may be that without the second dose, the protection would cease after a while, and in that case the slope would start rising again.

In the phase 1/2 paper, the authors said they compared the neutralisation titres to those from convalescent patients, and both the smaller doses induced titres as good or better than those seen in survivors: "After the first dose, the RBD-binding IgG GMCs (10-μg dose) were similar to those observed in a panel of 38 convalescent human serum samples, obtained at least 14 days after a PCR-confirmed diagnosis of SARS-CoV-2 infection and/or COVID-19. After the first dose, GMCs were similar in the 30-μg and 100-μg groups and higher than those in the panel of human convalescent sera."

My uneducated guess is that a single dose is as good or somewhat better at producing immunity than actually surviving the disease, and hopefully protection from a single dose of vaccine would last as long as the immunity of survivors or longer. And we know that while reinfection occurs, it is rare.

(I am not taking into account the 100ug dose because dosing is weird because biology is weird. Adding more doesn't always help.)

[andrewducker]

Thank you, that's really helpful.

[problemsdog]

More Or Less (https://www.bbc.co.uk/programmes/b006qshd) had a good discussion on this subject yesterday.

[palmer1984]

A couple of things:

1. The MHRA have approved this schedule and they are an independent body.

2. Most of us in the Uk will get the Astra vaccine, which has been tested and found to be effective at 12 week intervals.