Thread by @ewanbirney

It is hard month in January in UK, in particular the NHS critical care, but also now in "schools out" lockdown. Here are some thoughts from grey January London on COVID.

Context. I am an expert in human genetics/genomics and computational biology; I know experts in viral genomics, infectious epidemiology, clinical trials and other fields. I have COIs: I am long established consultant to @nanopore that makes a COVID test + I am on the OX/Az trial

Reminder: SARS_CoV_2 is an infectious virus which causes a horrible disease (COVID) in a subset of people (more likely older, male and overweight). A substantial proportion of the people who get the disease die.

If we let the virus spread across populations, in particular ones with large at risk groups, many people would die, many more suffer a horrible disease and most importantly the healthcare systems in countries would not cope with the influx of patients

[just to complete what would happen; if this was allowed then the healthcare service would be forced to make choices about insisting on some people dieing when they shouldn't and far more deaths would occur, potentially catastrophically].

Since April many countries have suppressed (either partially or fully) widespread transmission of the SARS_CoV_2 virus by a combination of Test-Track-Isolate and broad societal measures, which range from everyday actions (mask wearing) to regional/national restrictions.

(this has been variably successful; broadly far more successful in Asia than in Europe and North America, and substantial variation across Europe).

Due to excellent clinical research across the year, the care for people with COVID has dramatically improve - 50% of COVID patients die compared to April, and improvements continue to come in. Large, straightforward, robust clinical trials have been key to make these gains.

We also now have extensive testing schemes across most developed and many developing countries, with unprecedented volumes of tests being processed each day.

In early December two lineages of the SARS_CoV_2 virus were recognised to have different biological properties, one first discovered in England (B.1.1.7) and one first discovered in South Africa (B.1.351).

Analysis of the progression of these lineages, in particular B.1.1.7 in England shows that they transmit faster - quantifying that "faster" rate is complex (it hinges on how one models aspects of it) but it is substantial.

My own rule-of-thumb at the moment is that B.1.1.7 is 1.5x more transmissible (I favour a multiplicative model) but note also aspects like the infection cycle timing might have changed. There are many fine details to nail down.

Importantly by back-tracing data, scientists could show that B.1.1.7 transmission was not held (R <1) by the November lockdown in the UK; B.1.1.7 was relatively low incidence then but due to the depth of UK genomic surveillance its growth at these low levels could be seen

This feature of growth coupled with high and growing case load tightened restrictions progressively in England over late December, resulting in a new "national lockdown" importantly with schools shut (though many fine details changing from April's lockdown).

Giving the timings it is unclear whether this new lockdown will hold against B.1.1.7 (the data is simply not in yet); one can make arguments either way on the current partial data, though I am hopeful that there are regimes that hold B.1.1.7 (pointing to Kent LTAs in particular)

Here, speculation on twitter is not so productive - data which will get from both case reports and importantly the random surveys from ONS and REACT will be key.

B.1.1.7 is a big change in the pandemic course in my view, but it came at a similar big pivot point - the availability of vaccines that had passed rigorous clinical trials. We have at least 3 vaccines that work (certainly vs severe disease) and are safe.

A number of countries have expedited the roll out of these vaccines, even before it was clear about the presence of B.1.1.7 - notably Israel, US and also UK (which was wise in the circumstances) as well as UAE and Bahrain.

With the approval of the BioNTech/Pfzier vaccine many EU countries are rapidly vaccinating as well, notably Denmark, Italy and Germany and with the approval of Oxford/AstraZeneca in India and Bangledesh one of the largest vaccination schemes ever in scale and timing is starting

A number of Chinese firms have produced vaccines, some used in the middle east and many inside China, leading to a good starting vaccination rate in China, but with a similar awesome scale needed.

Thankfully both previous data about the immune response by vaccines (strong and with T-cell immunity) and more recent designed experiments suggest that the vaccines will work against B.1.1.7 and B.1.351

2021 has hope - vaccines - but huge challenges - and B.1.1.7 and B.1.351 has dramatically increased those challenges. It is unclear to me what levels of TTI+NPI in each country can hold transmission of B.1.1.7 below 1; but it must be more than what was needed in 2020.

Remember the scary behaviour of exponential growth - if B.1.1.7 reproduction is substantially above 1 it will double at some rate. Ireland's recently doubling of case numbers in under a week seems likely to me to be due to a substantial amount of B.1.1.7.

Although we know more about this virus and more forewarned, the fact that we were not able to prevent previous lineages moving around Europe (eg, the Spanish lineage in the summer) suggests to me it will be very hard to contain B.1.1.7 and B.1.351

This leads to the rather straightforward conclusion - 2021 is a race between vaccine logistics and B.1.1.7 / B.1.351 spread across the world. And, as ever, this is a humans vs virus fight where we have to pull together to win.