[FRIAM] Papers on asymptomatic transmission and serology

[Barry MacKichan]

In the distributed FRIAM meeting yesterday I mentioned these results, 
but I thought you might also want to see the commentary with them. This 
is an email from last week from one of my daughters (Joanna) who is a 
microbiologist at Victoria University in Wellington, NZ.

—Barry

————————————

First, the tests. I don't know what happened in the US with testing, 
maybe some of that has come out or maybe will come out. What I do know 
is that testing for RNA viruses is pretty straightforward - you extract 
RNA from the sample and try to quantify it using a PCR-based method. 
This is done around the world for routine surveillance of various RNA 
viruses (norovirus, influenza). There may be a few viruses that use 
antibody-based testing, but that wouldn't be used for the coronavirus. 
The RNA genome sequence of this virus was published early on by Chinese 
scientists, which countries were meant to use to design tests. The WHO 
normally only provides tests to low income countries, relying on high 
income countries (like the US or NZ) to develop their own tests. The 
primers and other reagents that work, all that information is out there. 
There was no need to reinvent the wheel with the tests, but from what I 
understand, the problem with some of the kits sent out by the CDC early 
on had a problem with one of the reagents (not the science behind the 
test). Validation is of course important, but again, tests have been 
clearly validated overseas. The usual regulatory safeguards will have to 
be relaxed around the world to keep up with the need for mass testing. 
The WHO says, test, test, test! That is going to be the key to beating 
this thing. I should add that I have full confidence that, despite 
initial missteps, the US is going to eventually get the testing right. 
It's absolutely essential if you (1) don't want to be under nationwide 
lockdown for 18 months, or (2) don't want 2 million Americans to die.

The craziest thing about this virus is the degree to which transmission 
is being driven by asymptomatic people. A Science paper just came out 
claiming about 85% of cases were transmitted by an asymptomatic person. 
Asymptomatic people may be less contagious, but they likely make up for 
it by going to family dinners, work, getting on airplanes, shaking 
hands, giving hugs, etc. Here's the Science paper:

https://science.sciencemag.org/content/early/2020/03/13/science.abb3221

When I first heard about asymptomatic transmission of the virus being 
key, I was very skeptical. Asymptomatic people don't cough or sneeze - 
so how do they transmit the virus? It turns out that they shed large 
amounts of virus anyway - such that breathing or talking is enough to 
infect someone nearby. I became less skeptical when I saw how insanely 
rapid the spread of this thing has been around the world. When 
asymptomatic people are your vectors, tests are absolutely critical. 
It's the only way of knowing if someone could be transmitting the 
disease. And if you don't have testing, you have to assume everyone is 
infected, which is why lockdown is the only alternative response.

The other reason I was skeptical about asymptomatic transmission, or the 
presence of a lot of asymptomatic people, is that this virus kills. How 
can it kill 15-20% of people over 80 but cause an asymptomatic infection 
in so many other people? I don't have an answer for that, but it's the 
essential reason that this virus has shut down the globe like it has. 
All our usual tricks don't work particularly well. I will say that 
microbes with these two extreme outcomes (no apparent illness, vs deadly 
infection) are relatively unusual, and that's why we are in this 
unprecedented situation. On the other hand, that particular combination 
is probably what largely drove the AIDS epidemic, so some of this is not 
new. But for a respiratory infection, it is unusual, and unlike AIDS, it 
is spreading much more rapidly.

And there is still much we don't know about asymptomatic people. Are 
many people never showing symptoms? Or do most of those asymptomatic 
people eventually go on to develop illness? This information is coming - 
for that you need serology - a retrospective look into a population to 
see who was actually exposed to the virus. The great news is that a 
paper that was posted in the past couple of days describes the first 
ELISA test for the virus. That is, they synthesised the (presumed) main 
viral antigen, the spike protein, in the lab, coated plates with it, and 
are now able to tell whether people have antibodies to that spike 
protein. Although this paper hasn't been peer-reviewed yet, it is out of 
a credible lab.

If you want to read the paper yourself:
https://www.medrxiv.org/content/10.1101/2020.03.17.20037713v1

The implications of this paper are important:
They only tested a small sample, but people who'd recovered from the 
virus clearly had antibodies to the spike protein. Non-infected people, 
and one person who had recently recovered from a confirmed infection 
with a common, milder coronavirus (which has a similar spike protein, 
attaches to the same human cell receptor) had ZERO antibodies. To make a 
huge extrapolation, there is likely little or no existing immunity to 
this thing (possible exception of SARS survivors), which is another 
explanation for why it has spread so rapidly.
Scaling up will enable screening of people to see whether they have 
protective immunity to the virus (due to natural infection). This would 
enable you to deploy healthcare workers with immunity to the frontlines 
- i.e., hospitals, caretakers in nursing homes.

This will also enable people to go back and study the wider population 
of places like Wuhan or Seattle. The current data suggest that about 20% 
of Wuhan residents got the illness. But it's possible that many more 
people were infected entirely asymptomatically (i.e., never became ill 
but carry antibodies to the virus). If only 20% of your population 
infected crashes the healthcare system, there is no clear strategy for 
relying on herd immunity. If it turns out it was actually closer to 60 
or 80% who were infected (enough for herd immunity) that changes things. 
Specifically, it would suggest that Wuhan is less likely to get a 
resurgence of disease if restrictions are eased. At this point we have 
no idea.
Adoptive antibody transfer - giving antibodies from someone who has 
recovered to someone fighting off the illness - can be explored.

To extrapolate even further, it may turn out that the differences in 
mortality or degree of sickness are not due to preexisting immunity; 
more likely the answer will be in variations in our underlying 
physiology (for example, maybe the virus mainly infects a cell type that 
hasn't matured in most children, rather than that children are largely 
immune).
In the meantime, we all want to avoid crashing the healthcare system, as 
has now happened in Wuhan, Italy, and Iran. And avoid getting ill, 
especially if in a more vulnerable category. I don't think there's any 
reason to assume that you will necessarily eventually get it. Even in 
the worst case scenarios being played out, it is not 100% of the 
population that is infected. Being very careful, until there is a 
vaccine, can ensure you can be in that part of the population that can 
avoid it altogether. Life is long, and we'll get through this challenge 
together!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://redfish.com/pipermail/friam_redfish.com/attachments/20200321/2f3f3bdb/attachment-0001.html>