A COVID and vaccination info-dump
Apr. 19th, 2021 01:20 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
garoteEwan Birney is an expert in genetics and computational biology. He also talks with experts in infectious epidemiology, viral genomics, clinical trials, and testing. He's an established consultant to Oxford Nanopore (which makes sequencing machines) and he was on the Ox/Az vaccine trial.
A COVID update, distilled form a series of his tweets on April 18th:
Background:
SARS-CoV-2 is an infectious virus which causes a horrible disease (COVID) in a subset of people, often leading to death. If we let infection progress at the virus' natural rate many people would die, and no healthcare system can cope with this rate of disease. Because of this, restrictions on social contacts and ways to suppress transmission have been implemented over most of the world in the last year. These have been variably successful, and shifts in behavior over the year have created "wave" like patterns of infection.
Two things we've learned from these patterns:
1. Backtracing works (e.g. Japan)
2. Ventilation is key to lower risk: Be outdoors. Exchange air.
Mutations:
As expected, the virus mutates all the time, and some of these mutations change the biology of the virus. Those biological shifts were small (but measureable in retrospect) over early 2020, but more substantial changes in biology occurred near the end of 2020. The B.1.1.7 lineage (first seen in England), the B.1.351 lineage (first seen in South Africa) and the P.1 lineage (first seen in Brazil) are three well-known examples, with high transmissbility (B.1.1.7) and some immune escape (P.1 / B.1.351).
(As an aside, it's hard to disentangle transmission from partial immune escape from aggregate numbers, and pinning down the biology of these viruses "in the wild" is far more complex than one might imagine.)
Vaccines:
Thankfully in 2020 a number of vaccines have been produced, currently in two groups - mRNA vaccines (BioNtech/Pfzier + Moderna) and Adenovirus vaccines (Ox/Az and J&J). Chinese and Russian vaccines also exist but less data is available for them (at least in developed world publishing.) All these vaccines work against the old strain and have low number of adverse events. We know the BioNtech/Pfzier and Ox/Az vaccines work due to the high vaccination levels in the UK and Israel with this strain.
The two adenovirus vaccines (Ox/Az and J&J) do have a very rare adverse reaction, also an immune reaction, but at the 1-in-million to 1-in-100,000 level. Different countries are adapting their roll-out to handle this in different ways. A consistent view is that the risk of this adverse reaction is well below many other healthcare and life risks, and the benefit - in particular for old and middle aged individuals - of avoiding COVID is far far higher. Still, it is complex to navigate.
Both in-vitro and small scale trials say that the Ox/Az vaccine is less effective with the B.1.351 (South Africa) lineage, and the BioNtech/Pfzier to a smaller degree. A similar mixed tale is true for the P.1 strain, with my read being that B.1.351 is the most clearly "escaped" from antibodies.
However, we have virtually no data on severe disease. This is unsurprising because the number of people who get severe disease is much lower, and countries with high vaccination levels and good national data collection (UK and Israel mainly currently) do not have much of the B.1.351 lineage.
There are reasons to think severe disease will have more "cross-strain" protection: All the vaccines have shown strong protection against severe disease (though not complete; nothing in biology is complete), and in vitro studies and studies from vaccinated individuals show strong a T-cell response, which is far harder (but not impossible) for a strain to mutate around and escape. But nothing here beats real world data. Hopefully the French studies in the Moselle area (higher B.1.351) will give us a readout on the BioNtech/Pfzier performance, and there should be an accumulation of data in the UK over time.
It's going to be thin pickings for a while because in both these settings the people organizing "test, trace, and isolate" strategies want to eliminate the B.1.351 strain, and other potential immune escape strains -- quite rightly.
Tactics going forward:
Israel has exited into what feels from my London standpoint something like the summer of 2020: There are some controls, face masks mandates, "green passports" to go inside, and lots of outdoor stuff, but it looks like a pretty normal life. Bravo Israel!
The UK and the US are heading the same way. The UK has a series of step-down relaxations. The first step (schools opening, groups of 6 in gardens/outside) seems to be going well. We are still getting a readout of the second step now (outdoor pubs and restaurants, more mixing.) However, the rise in infection levels in Northern Ireland is telling that this level of vaccination (~45% with one dose) by itself will not suppress transmission, and there are still plenty of people who could get this horrible disease, many of whom will suffer very badly, and some of whom will die.
It looks like Israel made its first transition at around 55% with one dose (but note with a higher second dose level, and all BioNTech/Pfzier). Unfortunately the complex relationship with Palestine, with less vaccination in that population, makes the "real world" mapping complex. Chile is another country with moderately high vaccination that still has high transmission, but with the added complication of a third vaccine (Sinopharm).
Europe:
European countries have picked up their vaccination rates, which now are within a factor of 2 from UK and US rates. A key thing has been the improvement of supply from BioNTech/Pfzier (go Marburg!) This puts all of Europe on the same basic path as Israel, the UK, and the US: A vaccinated population which both protects the more at-risk individuals from disease, and does much of the heavy lifting for transmission suppression. But more is needed.
My biggest fear for Europe is that the rather straightforward solution of "sit tight, suppress transmission until enough people are vaccinated" feels somehow unacceptable to them, or too complex to take on. I know that "sit tight" does not come at zero cost, and a failing of this pandemic is that we never got economic welfare and healthcare operations/delivery into the same framework in most countries where hard decisions have to be made, however, in a time where there is a good solution - shown by Israel, tracked internally, and clearly working - of more vaccination, it is probably one of the easier decisions to make relative to the economics. Contrast this with the situation in autumn of 2020, when far less data was available.
European countries have at least three countries to look at in terms of when and how to relax restrictions: Israel, the UK and the US. None of these are one-to-one mappings, but they are informative. Denmark is an interesting case, with really high levels of testing (RT-PCR and LFDs) and good isolation support, and with some vaccination they look in a stable situation with B117 as the majority strain. So there are other ways of making this suppression "bridge" work well.
A COVID update, distilled form a series of his tweets on April 18th:
Background:
SARS-CoV-2 is an infectious virus which causes a horrible disease (COVID) in a subset of people, often leading to death. If we let infection progress at the virus' natural rate many people would die, and no healthcare system can cope with this rate of disease. Because of this, restrictions on social contacts and ways to suppress transmission have been implemented over most of the world in the last year. These have been variably successful, and shifts in behavior over the year have created "wave" like patterns of infection.
Two things we've learned from these patterns:
1. Backtracing works (e.g. Japan)
2. Ventilation is key to lower risk: Be outdoors. Exchange air.
Mutations:
As expected, the virus mutates all the time, and some of these mutations change the biology of the virus. Those biological shifts were small (but measureable in retrospect) over early 2020, but more substantial changes in biology occurred near the end of 2020. The B.1.1.7 lineage (first seen in England), the B.1.351 lineage (first seen in South Africa) and the P.1 lineage (first seen in Brazil) are three well-known examples, with high transmissbility (B.1.1.7) and some immune escape (P.1 / B.1.351).
(As an aside, it's hard to disentangle transmission from partial immune escape from aggregate numbers, and pinning down the biology of these viruses "in the wild" is far more complex than one might imagine.)
Vaccines:
Thankfully in 2020 a number of vaccines have been produced, currently in two groups - mRNA vaccines (BioNtech/Pfzier + Moderna) and Adenovirus vaccines (Ox/Az and J&J). Chinese and Russian vaccines also exist but less data is available for them (at least in developed world publishing.) All these vaccines work against the old strain and have low number of adverse events. We know the BioNtech/Pfzier and Ox/Az vaccines work due to the high vaccination levels in the UK and Israel with this strain.
The two adenovirus vaccines (Ox/Az and J&J) do have a very rare adverse reaction, also an immune reaction, but at the 1-in-million to 1-in-100,000 level. Different countries are adapting their roll-out to handle this in different ways. A consistent view is that the risk of this adverse reaction is well below many other healthcare and life risks, and the benefit - in particular for old and middle aged individuals - of avoiding COVID is far far higher. Still, it is complex to navigate.
Both in-vitro and small scale trials say that the Ox/Az vaccine is less effective with the B.1.351 (South Africa) lineage, and the BioNtech/Pfzier to a smaller degree. A similar mixed tale is true for the P.1 strain, with my read being that B.1.351 is the most clearly "escaped" from antibodies.
However, we have virtually no data on severe disease. This is unsurprising because the number of people who get severe disease is much lower, and countries with high vaccination levels and good national data collection (UK and Israel mainly currently) do not have much of the B.1.351 lineage.
There are reasons to think severe disease will have more "cross-strain" protection: All the vaccines have shown strong protection against severe disease (though not complete; nothing in biology is complete), and in vitro studies and studies from vaccinated individuals show strong a T-cell response, which is far harder (but not impossible) for a strain to mutate around and escape. But nothing here beats real world data. Hopefully the French studies in the Moselle area (higher B.1.351) will give us a readout on the BioNtech/Pfzier performance, and there should be an accumulation of data in the UK over time.
It's going to be thin pickings for a while because in both these settings the people organizing "test, trace, and isolate" strategies want to eliminate the B.1.351 strain, and other potential immune escape strains -- quite rightly.
Tactics going forward:
Israel has exited into what feels from my London standpoint something like the summer of 2020: There are some controls, face masks mandates, "green passports" to go inside, and lots of outdoor stuff, but it looks like a pretty normal life. Bravo Israel!
The UK and the US are heading the same way. The UK has a series of step-down relaxations. The first step (schools opening, groups of 6 in gardens/outside) seems to be going well. We are still getting a readout of the second step now (outdoor pubs and restaurants, more mixing.) However, the rise in infection levels in Northern Ireland is telling that this level of vaccination (~45% with one dose) by itself will not suppress transmission, and there are still plenty of people who could get this horrible disease, many of whom will suffer very badly, and some of whom will die.
It looks like Israel made its first transition at around 55% with one dose (but note with a higher second dose level, and all BioNTech/Pfzier). Unfortunately the complex relationship with Palestine, with less vaccination in that population, makes the "real world" mapping complex. Chile is another country with moderately high vaccination that still has high transmission, but with the added complication of a third vaccine (Sinopharm).
Europe:
European countries have picked up their vaccination rates, which now are within a factor of 2 from UK and US rates. A key thing has been the improvement of supply from BioNTech/Pfzier (go Marburg!) This puts all of Europe on the same basic path as Israel, the UK, and the US: A vaccinated population which both protects the more at-risk individuals from disease, and does much of the heavy lifting for transmission suppression. But more is needed.
My biggest fear for Europe is that the rather straightforward solution of "sit tight, suppress transmission until enough people are vaccinated" feels somehow unacceptable to them, or too complex to take on. I know that "sit tight" does not come at zero cost, and a failing of this pandemic is that we never got economic welfare and healthcare operations/delivery into the same framework in most countries where hard decisions have to be made, however, in a time where there is a good solution - shown by Israel, tracked internally, and clearly working - of more vaccination, it is probably one of the easier decisions to make relative to the economics. Contrast this with the situation in autumn of 2020, when far less data was available.
European countries have at least three countries to look at in terms of when and how to relax restrictions: Israel, the UK and the US. None of these are one-to-one mappings, but they are informative. Denmark is an interesting case, with really high levels of testing (RT-PCR and LFDs) and good isolation support, and with some vaccination they look in a stable situation with B117 as the majority strain. So there are other ways of making this suppression "bridge" work well.
