The FDA will release under EUA a good antibody test, but time is of the essence to get decent tracking and tracing in-place
Without adequate testing for SARS-CoV-2, the novel coronavirus that is now sweeping rapidly around the world, there can be no isolation of cases and the chain of infection will not be broken.
Tedros Adhanom Ghebreyesus, WHO Director General, 16 March 2020
The FDA has cleared 10 reverse-transcriptase real-time PCR molecular tests under EUA (Emergency Use Authorization), with its 10th one approved today (23 March 2020) per the FDA website. As it is well-known, the United States’ Center for Disease Control (CDC) failed to adequately ramp up testing capability nicely described here, and without adequate testing the healthcare system in the US is basically ‘flying blind‘. It was rightly deemed a ‘national disgrace‘.
While rRT-PCR tests have shown good sensitivity (I understand the detection is 10 viral particles per sample) the specificity suffers a bit (on the order of 70%, where specificity refers to about a 30% false positive rate). Nonetheless given the time it takes to get results back, even with all the scaling up, on top of the time it takes for symptoms to appear (if they appear at all), detecting virus by this method will be a lagging indicator.
Thus the 43K+ cases in the United States today per the JHU resource reflects individuals that pass screening criteria, and show symptoms, and have had access to the test, and that information reported. This site is an ad-hoc effort to get numbers around the total volume of tests given; as of today, 23 March 2020 a total of 294,056 tests have been run, of which 42,164 are positive. This is nice increase since last week, when the numbers were only in the 50,000 range, and we can expect these numbers to rise significantly in the days and weeks to come.
Yet as mentioned before it is a lagging indicator by a week or more, given the time from initial inoculation to the recording of a positive test result. What that means is the 42,164 positives we have measured today means this was the number of positive individuals from a week ago.
A better method is an antibody test. This measures the presence of antibodies in the bloodstream that are against the virus SARS-CoV-2. A positive result here (that that individual has antibodies against SARS-CoV-2) means that person was exposed to the virus and has raised protective antibodies to the disease.
A test like this one can be done via one old technique called ELISA (enzyme-linked immunosorbent assay) or via something familiar to many of us, the home pregnancy test, otherwise known in the industry as a lateral-flow immunoassay (LFI). Lateral flow devices are cheap to manufacture, and take only 10 to 15 minutes to provide a signal.
An article in The Economist states the FDA is currently looking at two tests, and these serological tests have already been put into use in China, Singapore and Korea.
The organization FIND (Foundation for Innovative New Diagnostics) maintains a long list of SARS-CoV-2 immunoassays here, along with an even longer list of rRT-PCR assays. There are something like seven firms from China, along with two from Korea and one from Europe.
Here’s the list of five companies with a presence in the US, and perhaps it is two of these that the FDA will grant an EUA soon. Hopefully very soon – every day matters that we as a country understand the scope of the problem, and take active mitigation steps.
20/20 BioResponse is in Rockville Maryland (not far from where I live actually) and announced they would offer the CoronaCheck Test for COVID-19 as soon as the FDA grants them an EUA. At their website they are already taking credit card orders expecting ‘to ship the week of April 6’. They claim sensitivity of 97.9% and specificity of 91.7%, and I believe this data was generated in China. A manuscript with further details about the antibodis and its development is available here from the Journal of Medical Virology.
Aytu Bioscience is in Englewood Colorado and licensed a test from a Hong Kong firm LB Resources, who obtained it from a China-based developer Zhejiang Orient Gene Biotech. Per this news article, it appears to be also a LFI-based test, and validated in China in a 113-patient clinical trial there.
Biomedomics is a North Carolina firm with connections to China via its founders. Their website states it is currently being evaluated by the FDA for an EUA.
Epitope Diagnostics is a San Diego California firm offering an ELISA-based kit. While it takes about 2 hours for a test, it can be performed in 96-well plate format, and has separate IgG and IgM products.
Lastly Eagle Bio in New Hampshire also offers an ELISA-based kit for IgG and IgM.
The need for high-quality antibody tests has been mentioned by many clinical leaders now for weeks, in order to perform effective contact tracing to get ahead of the infectious spread. Keep an eye out for the approval of two of these aforementioned tests (or one I haven’t heard about) in the coming days (hopefully not weeks!).
The chart contrasting IgM titers (rise quickly, then fall off) and IgG titers (rise slowly and persist) is important. In the current situation, we’d want to know which is responsible for a positive test.
A serology test measures how much antibody against the spike protein is in the blood of the person being tested. As far as I know, the n-CoV-SARS-2 spike protein is very similar to the spike proteins of these closely-related viruses:
* HKU1, NL63, OC43, and 229E (the coronaviruses that cause 20% of colds)
* the coronaviruses behind SARS and MERs (hypothetical, since they aren’t around)
I think this likely means that serology tests for n-CoV-SARS-2 could give a Positive result… if you’ve gotten over a common cold caused by HKU1, NL63, OC43, or 229E in the past few months.
So, when assessing people who are recovering from Covid-19 like symptoms, it would be very useful to distinguish IgM (recent) from IgG (a while back) anti-spike antibodies.
It’s looking as though people who have cleared n-CoV-SARS-2 won’t be re-infected. See this MedCram video by Roger Seheult MD on YouTube, starting at the 8-minute mark.
The implication is that recovering health-care workers who now test Negative for the virus and Positive for anti-spike IgM can safely return to work, not getting re-infected by patients, and not putting them at risk, either.
Getting these antibody tests developed and deployed, fast, is really important!
That’s a great comment.
I might add the suggestion: there’s an unusually wide variance in individual response to infection with n-CoV-SARS-2. Possibly 5%-20% of cases are asymptomatic. Around 80% of cases are mild, equivalent to a cold or flu, and then 15% are more serious and may require hospitalization, and with an additional ~5% extremely severe requiring ventilation. Is it possible that the 5%-20% of asymptomatic cases predominantly are comprised of individuals who have had HKU1, NL63, OC43 or 229E (the coronaviruses that cause 20% of colds) within a recent-enough window to still have active antibodies that provide immunity?
I’m not a doctor . . . . . Thanks!