Antibody testing is coming, but for now, it doesn’t prove immunity
While there are numerous antibody — or serology — tests now on the market, very few of them have been approved by the Food and Drug Administration.
The tests and interpretation of results are more complex than a yes or no answer.
The tests are without doubt an important piece of determining the reach of the virus and reopening the economy, but the data is only good if accurate — and if people understand what it does or doesn’t mean.
If someone believes they have antibodies and, therefore, may have immunity, and can go back to work without worrying about exposing others, but if the test is wrong — then that information is worse than no information at all.
Even if a test is perfect, the science is too young to tell people if they are truly immune and for how long.
In one recent study of 14 antibody tests by a team of 50 scientists who worked around the clock for about a month, only three tests delivered reliable results.
Only one of the tests never delivered a false positive, which means it signals a person has COVID-19 antibodies when they really don’t have any, or they have antibodies reacting to another coronavirus but not COVID-19.
A false positive could be a dangerous piece of information, said Dr. Richard Zane, UCHealth’s chief innovation officer, professor and chair of emergency medicine at University of Colorado School of Medicine.
“I’m extremely worried because it has become a free for all,” Zane said.
People must have a thorough understanding of when and why they should be taking an antibody test, the characteristics of the test they are taking and the implications if they do test positive for the antibodies, he said.
Zane said UCHealth “has been working really hard to build capacity to do antibody testing, and we are in the final phase of validating our test.”
He said they are close to deploying it — but are still making decisions on how and for whom those tests will first be administered. It likely makes sense to first test those on the front lines who are most at risk, Zane said.
Steamboat Emergency Center doctor Jesse Sandhu said they have ordered 1,000 FDA-approved antibody tests, which are likely to be back ordered for another several weeks.
Asked about where the county is currently at in terms of community antibody testing, Routt County Director of Public Health Kari Ladrow replied in an emailed interview on April 29, “We continue to be very cautious about antibody opportunities.”
As to what exactly antibodies are and how they appear, “Antibodies get excreted as a response to an infection,” Sandhu said. “And there has to be enough virus in the body to trigger the antibodies.”
There are two types of serology tests on the market, Sandhu explained. One is point of care, which don’t have to be sent to a laboratory.
And there are two types of antibodies, he explained. The IgM antibodies show up about three days after being exposed to COVID-19. Those aren’t as specific to COVID-19. The IgG antibodies show up about seven days after exposure, he said, and those are more specific to COVID-19.
There are also five known coronaviruses, Sandhu said, and one includes a virus known as the common cold.
The point of care tests are easily deployed with a fast turnaround for results, he said. However, they aren’t as good at detecting low levels of antibodies. Still, in terms of gathering information, “It’s better than nothing.”
And your level of detectable antibodies may depend on when you were exposed to the virus, Sandhu explained. There is a critical window of time for antibody tests, he said, in order to determine just how much of the population has been exposed — and when. It isn’t known at this time how long COVID-19 specific antibodies remain detectable in a person’s blood.
The other test is known as an ELISA (enzyme linked immunosorbent assays) test. Those tests — also of which there are currently only a very small number approved by the FDA for COVID-19 — can detect lower levels, Sandhu said. And it is more accurate in terms of being specific to COVID-19 as opposed to the other coronaviruses, he said.
When Sandhu was looking at the latest FDA approvals last week, there were only about two point of care tests and two ELISA tests that had been approved thus far.
The FDA has stepped back on approving antibody tests, Zane said, because “they are all over the place” in terms of showing both false negatives and false positives.
Most of the tests currently out there, he said, have a relatively high probability of showing a positive result when the person does not actually have the specific COVID-19 antibodies.
There has also been a major scandal in Europe, Sandhu said, with several countries purchasing hundreds of thousands of dollars of test kits from China, which ended up being fake. As a result, he said, a lot of Chinese tests have now been blacklisted.
The FDA is requiring tests to disclose its characteristics, Zane said, which is defined in terms of sensitivity — meaning how often a negative is truly negative, and specificity — meaning how often a positive is truly positive.
But a test that says it is 92% specific, Zane explained, sounds good, but it really isn’t. That means there is still a relatively high probability of getting a false positive. UCHealth will not use a test that is less than 97%.
“A bad test can do no good,” Zane said.
On Sunday, the FDA approved a new antibody test created by Roche for emergency use. Roche announced its new test is 99.8% specific — meaning only one in 500 tests will get a false positive.
But even if you have a good test — a great test — which gives you an accurate result, what does that mean?
It can tell you if you have been exposed, and that “can be valuable from an epidemiological perspective,” Zane said.
There absolutely is value in figuring out how much of the population has been exposed to COVID-19, he said. Assuming the test is accurate, gathering information on the prevalence of the disease in the community is “a great goal.”
That data is especially useful given the large number of asymptomatic carriers — which may be as much as 25% of the infected population.
Sandhu noted that with SARS, it was easy to tell who was sick and isolate them. That is not at all the case with COVID-19, given the high number of asymptotic carriers, the high number of cases with mild symptoms and symptoms that mimic the flu.
But what keeps Zane up at night is not only the lack of accuracy in many of the tests on the market, but the fact it is still too early to say whether those antibodies are sufficient to provide immunity. And if they do provide immunity, it isn’t known for how long.
“It worries me that people will feel they won the lottery — and say ‘I’m safe, I have antibodies and I don’t have to worry,” he said. “In all likelihood, antibodies provide some level of immunity, but we don’t know at what level or how long that lasts. There are still so many unknowns, it’s very dangerous to tell people they are immune to COVID-19.”
With a background in virology, Sandhu said he would predict that contracting COVID-19 does give a person some degree of immunity. He has also been studying coronaviruses for decades.
Historically, Sandhu said, coronaviruses are more stable and less prone to rapid mutations. Typically, if someone contracted a coronavirus, they had some degree of immunity, he said.
However COVID-19 hasn’t been around long enough for anyone to know if and for how long contracting the disease makes a person immune.
“We’ve never encountered one this contagious,” he added.
Science is moving faster than it ever has before, Zane said. And many of these questions will have answers — and soon. But not yet.
Doctors who are administering antibody tests have a huge responsibility to communicate what the tests mean to their patients, Sandhu said. When the Steamboat Emergency Center gets their first batch of tests, Sandhu said the test will take about 10 minutes while the educational component will take at least twice that.
As we march toward the ultimate goals of treatment, vaccine and herd immunity, Sandhu emphasized, “slow and steady wins the race.”
Residents in Routt County will likely soon have access to the FDA-approved antibody tests, but “we have to be incredibly careful and thoughtful on how to test, why to test and what it means to be tested,” Zane said.
Support Local Journalism
Start a dialogue, stay on topic and be civil.
If you don't follow the rules, your comment may be deleted.
User Legend: Moderator Trusted User
Are we seeing more bears because there are more bears on the valley floor, or because we’re all spending more time at home? It could be a bit of both.