Naked Science Forum

Life Sciences => Physiology & Medicine => COVID-19 => Topic started by: evan_au on 26/03/2020 20:51:21

Title: Can an antibody test distinguish infected vs cured patients?
Post by: evan_au on 26/03/2020 20:51:21
Today on Australian TV, a doctor advised that they have produced a COVID-19 test which is styled after a pregnancy test. They hope to start production next week.

He claimed 99% specificity (ie it is unlikely to be fooled by other viruses) and something like 85% sensitivity (ie if your infection is new, it may not pick it up).

The test is based on detecting immune molecules in a drop of blood.

It is one thing to detect that 1."You have an active infection, and must stay in isolation for 2 weeks"
vs 2. "you had an infection, and are now cured, and can start doing normal work activities despite travel bans on other people"...

Can an antibody test distinguish infected vs cured patients?

And would we give people a certificate (preferably on an app) that describes their status, and forms a passport to say "I am now immune, and can travel around"?
- Immune people would still need to follow social distancing rules; they may not be generating virus themselves, but they may transfer virus from infected to non-infected individuals by close contact.

See: https://en.wikipedia.org/wiki/Immunoassay
Title: Re: Can an antibody test distinguish infected vs cured patients?
Post by: set fair on 26/03/2020 20:57:57
An antibody test can tell whether there are abtibodies to  the coronavirus in your blood - there will be a % of false positives and a % of false negatives. That's all it tells you.
Title: Re: Can an antibody test distinguish infected vs cured patients?
Post by: scientizscht on 26/03/2020 21:14:01
Is it specific to COVID-19 or to coronaviruses in general?
Title: Re: Can an antibody test distinguish infected vs cured patients?
Post by: chris on 27/03/2020 17:48:49
Yes, an antibody test can discriminate between people who have never been infected, those with acute infection and those who have been infected in the past.

The basis of this is that a non-immune (unexposed) person will have no antibodies in their bloodstream to the infectious entity in question; someone who is acutely infected makes an initial, short-lived antibody response dominated by antibodies of the IgM class. As the immune response matures, over about 2-3 weeks post-exposure, the dominant circulating antibodies "class switch" to IgG antibodies. These provide the long-lasting, protective neutralising response that gives us future immunity.

Antibody tests can discriminate between each of the three above situations; there are antibody tests against IgM and IgG responses; used together these can tell if a person either has had, has got or has never had the infection.
Title: Re: Can an antibody test distinguish infected vs cured patients?
Post by: evan_au on 27/03/2020 22:14:09
Thanks, chris. Now you mention it, the doctor may have mentioned IgG and IgM in the interview...

Apparently this test has just received expedited approval from the Therapeutic Goods Administration in Australia (the equivalent of the US FDA) to begin deployment with medical staff (ie not for home use).

The test takes 15 minutes to process a sample.

It was developed by a US company.
See: https://www.theguardian.com/australia-news/2020/mar/24/rapid-covid-19-testing-kits-receive-urgent-approval-from-australian-regulator