Thread by @ewanbirney

Thoughts on COVID19 from a dark London sunday night; TL;DR - we should be hopeful, but we have to stay disciplined for a number months more.

Context: I am an expert in genetics + computational biology; I know experts in infectious epidemiology, clinical trials, immunology and viral genetics; I have COIs due to consulting to Oxford Nanopore that make a SARS-CoV-2 test + that I am trial participant on Oxford/AZ vaccine

A reminder. SARS-CoV-2 is an infectious human virus which causes a nasty disease in which a subset of people die, and another subset have a long term disease (LongCOVID) which looks like a viral trigger auto-immune disease.

If we let the virus transmit in our populations in the developed world, with older populations, many people would die or get this disease; as well as that being awful, our health systems would not be able to cope.

Due to remarkable pace of research we have 3 vaccines that work (one which doesn't look as if it is going to work) and better treatments. One vaccine has be licensed in at least 2 countries. The (working) vaccines in particular give us much hope.

The 3 working vaccines (Pfzier/BioNTech, Moderna, Oxford/AZ) have different trials and endpoints; all "work" and have very low adverse effects beyond an initial vaccination pain/fevers. This is frankly - great.

All the vaccines showed lower symptomatic infection rates, and low hospitalisation rates (lower numbers, but clearly there). It is quite hard to imagine that these lower symptomatic rates wont translate into lower infection rates as well.

The Oxford/AZ vaccine had a more complex multi-trial meta-analysis with an error in dosing in one trial which looks like a better dosing regimen. In some sense this is "bank error in your favour" but in practice it makes the decision messier.

Discard that dosing error completely, and one still has a vaccine that works, and although the Pfzier and Moderna vaccines post high efficacy numbers, one needs to be careful about the details of the endpoints used.

Realistically we'd be happy with another trial on Oxford/AZ, but could this be some Phase IV like thing (ie, whilst being distributed) rather than a Phase III thing? How would one do it? I am not an expert here so just watching from the sidelines.

Both the UK and now the US have approved the BioNtech/Pfzier vaccine and it is almost certain the EMA will approve at the end of this year. The complexity of the logistics and vaccine prioritisation will now be a major focus in every country.

But even the in best case, with excellent logistics, this is multiple months until enough of the population has been vaccinated to have significant impacts on transmission rates, coupled with NPIs and TTI schemes...

...so this means we've got to navigated minimally December and January with Test-Track-Isolate schemes and blanket restrictions (NPIs) - Lockdowns/Regional restrictions.

I can sense the weariness of this - from Saxony in the south east of Germany to Voseges in France to the Thames Estuary in England and the South Wales valleys rates have been rising even in the face of lockdowns and restrictions.

Why? I don't think we really know - a mixture of economic imperatives, desire to meet people "as normal" and a sense perhaps that it will all be fine. I know the English data best, and the one silver lining is that the rise is more in the under 30s... but....

As sure as night follows day infection levels rising in the younger generations follows through to the older. So we need to come up with stronger solutions - combinations of TTI ("to the individual" NPIs in effect) and regional NPIs ("Tier rules" in English terms) that work.

Many people discount the improvements that have happened in these. TTI has improved in the places I know - UK, France for example. Testing is now deep, and trying to work out to use the higher testing capacity to suppress transmission is being explored in many countries

The super-cheap and quick lateral flow tests - less sensitive for presence of infection throughout but look sensitive enough for the key infectious periods is one thing. But noone has fully cracked using them -

Slovakia seems to be as much about their strict lockdown as well as testing; the LFA tests in Liverpool definitely found many positives, but the overall impact assessment is harder in the context of general decline of transmission there; watch the Thames Estuary LFA roll out.

We still have further arrows in the testing bow (even more testing!) but it is probably now as much about what happens after testing - ie, isolation support and compliance as it is testing.

Trace still has plenty of space for improvement; the backtracing protocols of Asia have not been fully used in the same way in Europe - I am hopeful though that they could make an impact.

We've got to keep making many many improvements here - work out what works in the contexts of our countries - because we still have too long to go until vaccines will fundamentally change the landscape.

So - please, please be careful. It is madness for us collectively lose the will now to guide our countries home with less loss of life. 2021 will be a better year; let's get there together.