Wow! Britain OKs Astra Zeneca Oxford Vaccine, holds no 2nd shot in reserve

[Snorlax94]

What do people think? It’s so aggressive yet their situation is so dire. It could in a way be helpful for other countries because they are basically running a massive experiment on their population. I guess this is one way to find out if transverse myelitis is a risk with this vaccine.

It seems a inconsistemt though that the Russian and Chinese rushed, unpublished vaccinations were characterized as dangerous and unethical yet this move is “bold.”

I though Oxford wouldn’t even report until late January to early Feb and  would be approved in early Feb at the earliest. Have they even published Phase 3 preliminary data?

https://www.nytimes.com/2020/12/30/world...eneca.html

LONDON — Britain became the first country on Wednesday to give emergency authorization to the coronavirus vaccine developed by AstraZeneca and the University of Oxford, opening a path for a cheap and easy-to-store shot that much of the world will rely on to help end the pandemic.

In a bold decision to accelerate vaccinations, a British government advisory body directed clinicians to give as many people as possible their first dose of a coronavirus vaccine, without reserving supplies for planned second doses. 

Instead of administering the two shots within a month, clinicians will wait as long as 12 weeks to give people their second doses, the government said, a decision that applies to the Oxford-AstraZeneca vaccine as well as the Pfizer-BioNTech shot that Britain authorized early this month

[DocSavage87]

What do people think? It’s so aggressive yet their situation is so dire. It could in a way be helpful for other countries because they are basically running a massive experiment on their population. I guess this is one way to find out if transverse myelitis is a risk with this vaccine.

It seems a inconsistemt though that the Russian and Chinese rushed, unpublished vaccinations were characterized as dangerous and unethical yet this move is “bold.”

I though Oxford wouldn’t even report until late January to early Feb and  would be approved in early Feb at the earliest. Have they even published Phase 3 preliminary data?

https://www.nytimes.com/2020/12/30/world...eneca.html

LONDON — Britain became the first country on Wednesday to give emergency authorization to the coronavirus vaccine developed by AstraZeneca and the University of Oxford, opening a path for a cheap and easy-to-store shot that much of the world will rely on to help end the pandemic.

In a bold decision to accelerate vaccinations, a British government advisory body directed clinicians to give as many people as possible their first dose of a coronavirus vaccine, without reserving supplies for planned second doses. 

Instead of administering the two shots within a month, clinicians will wait as long as 12 weeks to give people their second doses, the government said, a decision that applies to the Oxford-AstraZeneca vaccine as well as the Pfizer-BioNTech shot that Britain authorized early this month

Hard to say what's going on without studying all the details of the available studies.  Probably their reasoning: (1) there's a fair amount of effectiveness even with one dose from prelim data, (2) even if it doesn't prevent infection, the other main thing you want to reduce is the severe cases that could lead to death, and it seems it's very effective in doing that.

[Goose]

This seems a bit reckless to me. The clinical trials performed so far had a specific protocol for the second dose. Why bother with a trial if you aren't going to use that protocol for administration? No way in the world the FDA would go along with such a thing IMHO. Astra Zeneca is still running their trial in the USA and has not even attempted to get an EUA in the US with the results they have from their previous trials because they (correctly I think) believe it would not be granted. Hopefully their USA results are more useful to them.

It may well be that the change in protocol will not affect the results in immunity, but we don't know that. If it "works", that will be great, but if it doesn't work very well, imagine the reaction of the public when people who have received the vaccine get sick in large numbers. Confidence in such a treatment will plummet. The AZ vaccine had not so great efficacy in the trial they actually did complete. The results of this new protocol may not be even as good. That said, if the UK is much more efficient at getting vaccine into arms than we have been here in the USA, they may get better results faster than we do.. Still a gamble I think.

[BostonCard]

I though Oxford wouldn’t even report until late January to early Feb and  would be approved in early Feb at the earliest. Have they even published Phase 3 preliminary data?

Yes

https://www.thelancet.com/journals/lance...1/fulltext

BC

[BostonCard]

The Oxford/AZ vaccine situation is different than the China/Russia situation.

The main difference is that Oxford/AZ has run a phase 3 clinical trial, and as of the data cut-off of the Lancet article, they had over 5000 volunteers vaccinated with a median follow-up of 3.4 months.  Given that their cut-off was Nov. 4, they have undoubtedly recruited more volunteers and followed the existing ones for at least another 1.5 months.  From the MHRA's instructions for healthcare providers:

https://assets.publishing.service.gov.uk...Zeneca.pdf

The overall safety of COVID-19 Vaccine AstraZeneca is based on an interim analysis of pooled data from four clinical trials conducted in the United Kingdom, Brazil, and South Africa. At the time of analysis, 23,745 participants ≥18 years old had been randomised and received either COVID-19 Vaccine AstraZeneca or control. Out of these, 12,021 received at least one dose of COVID-19 Vaccine AstraZeneca. The median duration of follow-up in the COVID-19 Vaccine AstraZeneca group was 105 days post-dose 1, and 62 days post-dose 2.

While ideally you would like a safety database with 15,000 volunteers followed for at least 2 months, which I believe is the FDA's guidance, 12,000 volunteers total, half of whom have been followed for at east two months after the second dose is not bad.

The safety is a little harder to assess than what the FDA provided for the Pfizer/BioNTech and Moderna vaccines. The Information for Healthcare Practitioners (somewhat similar to the Package Insert or USPI) states, "Very rare events of neuroinflammatory disorders have been reported following vaccination with COVID-19 Vaccine AstraZeneca. A causal relationship has not been established."

The term "very rare" has a specific meaning, and means side effects with a frequency of <1 in 10,000, which, given that 12,000 volunteers have been vaccinated, means a single volunteer suffered from such an event. However, the language is not clear as to whether "very rare events of neuroinflammatory disorders" means multiple different neuroinflammatory disorders, each of which happened to one volunteer, which would be a bit more concerning. Also, the confidence interval around a single event is between 2 in 1 million and 1 in 2000; again, the latter would be a bit more concerning.

The other thing that the Oxford/AZ vaccine has that the Chinese and Russian ones didn't is randomized, placebo controlled efficacy data. The problem is that the data is confusing. Because of the quirk in the study (basically, if you read the Lancet article, you will see that they thought they were giving a larger dose than they were actually giving initially, so a couple thousand volunteers received a half dose followed by a full dose. Perhaps a bit paradoxically, the interim data suggested that this dosing regimen might work better (up to 90% efficacy). Now, there are all sorts of reasons not to take that data as gospel, and it is worth waiting for a follow-up study that definitively establishes a precise efficacy measure.

That being said, both regimens work against placebo, and both clear the 50% efficacy bar. Moreover, none of the cases of COVID-19 requiring hospitalization in the Lancet article (there were 10), occurred in the vaccinated group, so the protection against moderate or severe disease may be better.

The data presented by the MHRA is a bit different; they found that in the primary analysis (≥15 days after the second dose, 5 hospitalizations and case of severe disease, again, all in the placebo.

The interesting data lies in the exploratory analysis they looked at.

The level of protection gained from a single dose of COVID-19 Vaccine AstraZeneca was assessed in an exploratory analysis that included participants who had received one dose. Participants were censored from the analysis at the earliest time point of when they received a second dose or at 12 weeks post-dose 1. In this population, vaccine efficacy from 22 days post-dose 1 was 73.00% (95% CI: 48.79; 85.76 [COVID-19 Vaccine AstraZeneca 12/7,998 vs control 44/7,982])

Moreover, only 2 hospitalizations (none severe) occurred after the first vaccine dose (on days 1 and 10 post-vaccination). And lastly, the data presented by the MHRA contains this intriguing nugget:

Exploratory analyses showed that increased immunogenicity was associated with a longer dose interval (see Immunogenicity Table 3). Efficacy is currently demonstrated with more certainty for dose intervals from 8 to 12 weeks. Data for intervals longer than 12 weeks are limited.

Now, under normal circumstances you would say that all of this is interesting but we would want it formally tested before allowing an EUA. But we have a pandemic that is killing people, so if you put it all together you come to the conclusion that:

1) The vaccine is pretty safe, though you would want to follow it closely in more people for longer before you were certain.
2) The vaccine is efficacious, and seems particularly promising in preventing hospitalizations and severe cases, though again the data is uncertain.
3) There is promising data that suggests the vaccine seems to work 10 days after the first dose, AND that waiting longer between doses (up to 12 weeks) is better for mounting a more robust immune response.

The outstanding questions are mostly about identifying the best regimen, not necessarily in determining whether it is safe or if it has efficacy. Couple it with the lower cost, wider availability, and easier storage and distribution (you just need a refrigerator), and it seems like an authorization with the data at hand is a reasonable gamble to take. If it were me, and I were given a choice between being able to take the Oxford/AZ relatively soon or wait several months to take the mRNA vaccines, I would think about it but probably take it. Moreover, the strategy of approving one dose for now and waiting a few months for a second dose makes sense based on the data at hand currently (with the understanding that I'd want to see more data before saying so conclusively).

Overall, I think the MHRA is in fact being bold and not wantonly dangerous, though it is not a risk-free gamble.

BC

[Goose]

Overall, I think the MHRA is in fact being bold and not wantonly dangerous, though it is not a risk-free gamble. 


BC

While I understand the rationale presented, I think the MHRA is being "bold" and taking a gamble. The safety data is pretty good, so probably the vaccine isn't going to harm any significant number of people. There is some indication that it may prevent serious illness and some indication it may provide a non-zero degree of immunity. There are also some indications that a longer period between immunizations is more effective. Unfortunately, as we know from hard experience many drugs that gave "some indication" don't live up to expectations. Given the emergency, taking a gamble may be a reasonable risk.

The problem is that the public doesn't understand that rationale and the MHRA isn't really going out of their way to explain that. BC was kind enough to tease out this analysis from available data, but I don't thing it is published in the Times. The public will believe they are getting an effective vaccine that will "protect them". If it is only 50% effective, it isn't going to really do that. The public reaction to that outcome will probably be quite vociferous, and may set back confidence in the MHRA to the point they will no longer be effective going forward.

[BostonCard]

I completely agree Goose.  Here the FDA has done an exemplary job by making public the VRBPAC meetings and publishing the staff assessments in advance.  While I am confident that the MHRA isn't being reckless, they are not doing a good job of explaining the rationales.

That being said, the efficacy data is quite a bit stronger than "giving some indication" of efficacy.  Using hard endpoints, the drug does convincingly reduce the risk of COVID-19 infection.  The 95% confidence intervals robustly exclude zero efficacy, meaning that the vaccines are significantly efficacious, and the lower bound of the confidence interval is above 30%, which was the FDA's bar.  We might not for sure if it is 70% effective or 90% effective, but it is likely be above 50% effective, which is about what you get with your average flu vaccine.  You are right, however, that the MHRA has not communicated this well, and people do need to know that they can still get COVID even if they have the vaccine (this is true also of the Pfizer/BioNTech and Moderna ones, though they get closer to 95% efficacy), so some degree of social distancing is still important in the midst of the pandemic.  But all the vaccines will put a dent in the pandemic, and, promisingly, all of them seem to especially offer protection against more severe disease (requiring hospitalization, ICU admission, or causing death).

There are enough questions that I would understand a delay (especially with two highly effective vaccines already authorized), but I consider the MHRA to be aggressive but not reckless.

BC

[M T]

Pfizer and Moderna have fairly short intervals between doses.  If AZ studies suggest that a longer interval may be more effective, I'd want to ask if either Pfizer or Moderna has any evidence one way or the other on any interval other than the 4 week/3 weeks?  

If there were, could the CDC be imagined to allow a longer time between vaccinations in order to get more people with a first dose faster?

Maybe Britain has the idea of getting the first dose in people now and then recommending a second dose 6 months out.

[Goose]

That being said, the efficacy data is quite a bit stronger than "giving some indication" of efficacy.  Using hard endpoints, the drug does convincingly reduce the risk of COVID-19 infection.  The 95% confidence intervals robustly exclude zero efficacy, meaning that the vaccines are significantly efficacious, and the lower bound of the confidence interval is above 30%, which was the FDA's bar.  We might not for sure if it is 70% effective or 90% effective, but it is likely be above 50% effective, which is about what you get with your average flu vaccine.  You are right, however, that the MHRA has not communicated this well, and people do need to know that they can still get COVID even if they have the vaccine (this is true also of the Pfizer/BioNTech and Moderna ones, though they get closer to 95% efficacy), so some degree of social distancing is still important in the midst of the pandemic.  But all the vaccines will put a dent in the pandemic, and, promisingly, all of them seem to especially offer protection against more severe disease (requiring hospitalization, ICU admission, or causing death).


There are enough questions that I would understand a delay (especially with two highly effective vaccines already authorized), but I consider the MHRA to be aggressive but not reckless.



BC

BC, when you say "hard endpoints" aren't you talking about the phase three trial as designed, not the exploratory analysis? That would be the original protocol, not the modified one of a single dose. An exploratory analysis doesn't qualify as a hard endpoint, or does it? I am almost certain the FDA really resists to the max retrospective endpoints.

In any case, a 50% reduction in COVID-19 cases would be really welcome and would drive Reff below one, subject to the current lockdown restrictions. The problem will then arise about how much you can "ease up" the current restrictions and still control the disease. It may well be that you can't get back to "normal" or anything close to it. You may have a virus that is effectively less transmissible (due to immunity) but still capable of doing serious social damage without controls on large crowds etc. being retained. If the vaccine does in fact reduce mortality much more than 50% then having a larger Reff may be just fine, or at least acceptable. However, we don't know that yet.

One does have to say the same situation could arise with vaccines that are 90% effective, but this is much less probable. In either case the "correct" approach IMHO would be to leave the current regulations in place until the smaller Reff brings the number of cases down to a very small number. Naturally, this process will be much faster with a 90% effective vaccine than a 50% effective one. Then there would be hope that we can control the disease by more "conventional" methods and can lift the controls entirely. However, I fully expect the public and the politicians will have none of it and demand the restrictions be reduced "as soon as possible" or even before. There is also, as you point out, a parallel with the flu vaccine. I know several people who say "I got the shot, I still got the flu. You shouldn't bother getting it" which may make vaccination less than universal, eroding the 50% even more.

I also expect that taking a "more effective" vaccine later may not be safe and/or effective. A trial would have to be run to determine that. Thus, the MHRA may be wedding the public to the Astra Zeneca vaccine AND this particular protocol. That's fine if it turns out to be "good enough". Otherwise, ouch. This is a consideration that IMHO shouldn't be ignored. OTOH, there may be a protocol under which this vaccine is 90% effective. There are indications of this in the existing trial. However, a half dose (sort of) was the first administration, not a full one. I would suggest that finding that protocol should be number 1 priority and maybe outweigh the benefits of a quick administration of a single full dose. One should consider that "rushing to a failure" is often unwise. You may not require a full trial, since safety isn't the issue.

I do agree with you that the MHRA isn't being "reckless". They are making a measured evaluation that the expected benefits will outweigh the risks. I do wonder how much "political considerations" played into this evaluation. It is a cheaper alternative, it is easier to distribute (I am pretty sure there are no -80 freezers on Skye) and it was developed in the UK. It may be easier to get into more arms much faster, which would be good anywhere. That we will never know, probably.

[M T]

It isn't like you wake up one day and most of your population has a 90% or 50% vaccine.  The effectiveness of the vaccine has to be multiplied by the ratio of the population, at least while their product is low.

A 90% effective vaccine in 1% of the population only helps about 1% of the population (end of December, counting first shots).   We may get to 10% of the population by the end of January (16% is best case).  Maybe 25% by the end of February (30% best case)

Then, if the Governor of California has his way, we stir the soup by having kids bring it home long before herd immunity across the adults & with 0% vaccinations in K-12.