SARS-CoV-2 Diagnostics, From Serology to Spit

Paul G. Auwaerter, MD


November 16, 2020

Find the latest COVID-19 news and guidance in Medscape's Coronavirus Resource Center.

This transcript has been edited for clarity.

Hello. I'm Paul Auwaerter with Medscape Infectious Diseases, speaking virtually from IDWeek with another segment on Chasing the Sun, the 24-hour COVID-19 overview that was a tremendous effort by so many.

One of the more interesting segments — and one with many questions — had to do with SARS-CoV-2 diagnostics.

SARS-CoV-2 Serology

It was an interesting choice for them to start this section with serologic diagnostics. Florian Krammer, who's at Mount Sinai in New York, discussed this segment and commented that the initial impetus was to develop an accurate serologic test for the identification of donors of plasma, so that convalescent plasma could be used as a therapeutic modality. Although the S protein is known not only for this virus but for other coronaviruses, such as SARS-CoV-1, it is a target for generating many of the neutralizing antibodies.

His goal, when he was establishing this, was to set up serologic tests that would have a sensitivity and specificity of > 90%, which would be quite admirable. By background, many of the tests developed at this point against a highly identified set of positive and negative controls may do reasonably well, but in wider use they might have high rates of false positives. Therefore, generally, the admonishment has been that currently available serologic tests should not be used to make the diagnosis that someone has had the novel coronavirus or to offer a type of immunity passport that people would find so desirable.

Florian developed a two-ELISA model, so not just a single one against a spike protein but enhancing that by also taking a receptor-binding domain of the coronavirus and looking for that antibody production. You would actually require two steps.

He was finding sensitivities of about 92%, with much higher specificities toward 100%, and this seemed to correlate with abilities to neutralize the virus.

They screened over 65,000 donors to date, mostly who had PCR-confirmed mild coronavirus disease. Interestingly, he found that 99.5% seroconverted, which was quite reassuring for mild disease, as others have reported that people don't make antibodies or that they wane very quickly.

Among this group, 7% had low titer, 20% had intermediate titer, and 70% had high titer. We still don't really know what that means in terms of a correlate of protection, but the hopeful biologic plausibility is that intermediate to high titers would offer [protection]. Indeed, [that may be] some of the impetus for why the US Food and Drug Administration (FDA) gave an emergency use authorization for convalescent plasma with the requirement to at least have them labeled as high or low titer, so that clinicians would know they're giving high titers.

As opposed to other groups, they saw persistence of antibodies even beyond 4 months, which was another important aspect. Clearly, there are some mixed data on this and we'll wait to hear more. He did comment that titers understandably do drop a bit but then seem to stabilize. Perhaps this gives some better hope about the durability of immunity to the coronavirus.

Antigen Tests and Novel Approaches

The next talk on diagnostics was by Esther Babady, and she discussed some of the other tests that we generally do use for diagnostics.

She commented on the antigen test, which offers ease and speed, along with lower cost. However, the antigen tests are only FDA-approved for early illness, when viral loads are quite high, and they do lack the sensitivity of molecular testing. She commented that the FDA has set up a panel to help try to standardize PCRs for sensitivity. I think once [this happens], we'll have a better sense of whether we can actually compare testing and clinical results, as platforms may offer differences.

She also discussed saliva. Is this where we're going? I think many of us are hopeful, but there are generally mixed data. She pointed to the Wyllie paper in The New England Journal of Medicine from September 2020, which said that it might actually outperform the nasopharyngeal swab and molecular testing. I think we still need more experience.

Returning to antigen tests, she commented on the four tests in the United States. Interestingly, although there are some analytical data from these tests presented to the FDA, nothing has been published. She referred to published data from antigen tests available in Europe and China, to give the following ranges: The sensitivities of rapid antigen tests for the coronavirus ranged from 30% to 97%, so clearly there are many opportunities for false negatives. Specificities, though, as expected, were better, in the 90%-100% range.

Last, Audrey Odom John had a very interesting presentation about using dogs and their ability to sniff out specimens or people who might be infected. Of course, this is not terribly practical, but it's a highly interesting and entertaining aspect.

Thanks very much for listening.

Paul G. Auwaerter, MD, is a professor of medicine at the Johns Hopkins University School of Medicine and clinical director of the Division of Infectious Diseases. His areas of clinical expertise include Lyme disease, Epstein-Barr virus, and fever of unknown origin. He has been a Medscape contributor since 2008.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.