A new serological assay for SARS-CoV-2, the virus which causes COVID-19, was developed by a microbiology lab at the Icahn School of Medicine at Mount Sinai Health System.
Florian Krammer, PhD, professor of microbiology at Mount Sinai, talked to Contagion
® about his laboratory’s new work.
Interview transcript (modified slightly for readability):
Could you explain what this test is and what different role a serological assay plays compared to the SARS-CoV-2 acute infection diagnostics we've heard a lot about the past few weeks?
The PCR [polymerase chain reaction] or nucleic-acid based test is to figure out if somebody is acutely infected with the virus. The serological test is to figure out if somebody had been infected with the virus, it's not for finding acute infections.
You might, during an acute infection, already see some antibody, but typically that comes later.
It has a lot of uses in research. We can use it to better understand the antibody response and the dynamics. We can now perform serosurveys to figure out how widespread spread the virus actually is. People who might not have symptoms might still produce an immune response. Those are the research reasons for doing this. Now, there's 2 more practical reasons.
There is a large initiative to look for people who seroconverted, and then ask them to donate plasma and maybe use that as a therapeutic. This has been done in China to a certain degree and in Italy.
There are no clinical trials that show that this is effective, but there's anecdotal evidence that it might work. China is also doing a clinical trial. If you can screen people and look who has a strong immune response, you can identify them and ask them to donate. You could also do that with a neutralization assay, but that takes a few days and you need to do that at biosafety level 3. While the ELISA assay that we developed is very easy to do—there's no infectious virus involved at all, and the output is relatively high. With 1 operator, the current setup that he has, you can run a few thousand samples per week.
The other more hypothetical use is that we don't know what the real infection rate in the population is. We know from human coronaviruses that—for at least some time, as long as the immune response is there—once you seroconvert you are immune to reinfection. It doesn't last forever. It lasts maybe for 1, 2, maybe 3 years. But you get a protective immune response for some time.
Everybody's at home, self-quarantining. If you would know who was already exposed and infected and who has an immune response, we might be able to let these people go back to work. Because if they're immune, and they can't get infected anymore, they're no danger for others because they're not able to spread the virus on anymore. That is more hypothetical. There's no approved use of that assay for that purpose right now. But that's something that we may want to think about when we think about restarting the economy.
A lot of health care workers also probably got exposed early in the game, and they might now be immune. Knowing that you're immune might be some comfort if you have to deal with COVID-19 patients.
Could you elaborate a bit about how the test was developed and how it works, because you mentioned that you're not actually working with a live infectious agent when you're doing this, is that right?
That is correct. We made recombinant versions of the surface protein of the virus, the spike, and then you run relatively simple ELISA with that.
A very simple assay, very old school, and that's it. It takes about 3 hours and you're done.
You mentioned 2 to 3 years for other human coronaviruses. Is that how long we can predict this immunity the last or is there more research that needs to be done there?
There needs to be research done of course, it's a different virus, but that's what we know for the human coronaviruses. That's was observed for people who got infected with SARS-CoV-1 in 2003.
We published a preprint
. Last Friday, there was another paper from a group in the Netherlands and they did something similar, and they were already able to show that their results correlate with neutralizing antibodies. But again, we need to understand it better. The problem in general is that nobody cared about human coronaviruses before this happened. There were only few cases of SARS. Studies at some point stopped.
What should clinicians who want this test available to them do?
What we started to do is distribute the plasmids for making the antigens—the recombinant spike protein—widely. We're shipping these out—now for days—to many different universities and health care centers.
We also provided the protocol of how this can be set up; people can do that locally. In the last few days commercial kits also popped up. We don't know how good they are yet. But that's certainly also an option, right?
In the end it's up to the clinical labs to set this up and validate the assays and make sure that they get approval to use it as diagnostic. How far, who is getting tested then, and with what consequence is a different story that's above my pay line.
Thanks very much. That was all really useful information.
For more information, visit the Krammer Lab homepage.
To stay informed on the latest in infectious disease news and developments, please sign up for our weekly newsletter.