What We Know About COVID-19 Antibody Testing

Photo: dpa/picture alliance via Getty Images

Many are hopeful that antibodies could be the savior of the coronavirus pandemic. The thought process goes like this: If you were at some point exposed to the SARS-CoV-2 virus, you will produce antibodies for it, which will keep you immune for some reasonable chunk of time — at least long enough that there would be a vaccine by the time your immunity faded. In this version of the future, people with antibodies, perhaps holders of so-called immunity passports, could go back to work. They could travel. They could help the more vulnerable population.

But according to the World Health Organization, it is unclear if immunity even exists for the SARS-CoV-2 virus. Still, antibody testing is available, and those who know or think they might have been infected with the virus have the ability to be screened. It’s a confusing situation, so we’ve attempted to untangle it here with the help of Gigi Kwik Gronvall, an immunology expert with the Johns Hopkins Bloomberg School of Public Health, and Arnau Casanovas-Massana, who specializes in the epidemiology of microbial diseases at the Yale School of Public Health.

What are antibodies?

When disease-causing agents called pathogens enter your body, your immune system — which includes white blood cells, antibodies, and stem cells, among other things — mobilizes to destroy them. Antigens are protein elements found on the surface of pathogens; they alert the body to an infection and trigger an immune response. That immune response is the production of antibodies. So the existence of antibodies in a person’s blood means they have been exposed to a pathogen, which, in this case, is a virus.

How are antibody tests administered?

As you may know, testing to diagnose a current SARS-CoV-2 infection involves a nose swab. But testing for a past infection, by looking for antibodies in your blood, involves a blood sample. During the test, a health-care provider will draw a sample of your blood and send it to a lab for testing. Many are paying out of pocket for the tests, as well as incurring unexpected lab bills after the fact. If you have insurance, check with your provider before your appointment to confirm that it will cover the cost.

Are the antibody tests that are available for the SARS-CoV-2 virus accurate?

The answer is fairly complicated and somewhat maddening, so let’s break it down into hopefully easier to understand elements.

Some tests are better than others. 

To get antibody tests on the market quickly, the Food and Drug Administration waived its requirement that they be reviewed and approved by the FDA. Instead, antibody tests entered the market with only the manufacturer’s claim of accuracy. In late April, the House Oversight and Reform Committee released a report about how this led to a glut of inaccurate tests. “FDA has failed to police the coronavirus serological antibody test market,” the report said. “Numerous companies appear to be marketing fraudulent tests.”

Luckily, the FDA updated its standards in May.

Now, companies that manufacture tests have to submit a request to be reviewed under Emergency Use Authorization (EUA). Being “approved” under a EUA is still not as strict of a process as the FDA would otherwise undergo, but it does require an FDA review of testing done internally by the test manufacturer. The National Cancer Institute is also working to evaluate commercially available antibody tests in order to provide the FDA with independent research.

Test accuracy mainly depends on two things: sensitivity and specificity. Sensitivity means the test is able to find the antibodies related to SARS-CoV-2 exposure; if the test isn’t sensitive enough, you might get a false negative. Specificity means the test is specific enough to know the antibodies it found are the antibodies related to SARS-CoV-2 rather than something else; if the test isn’t specific enough, you could get a false positive. Testing is never going to be 100 percent accurate, but antibodies tests are now thought to be 95 percent to 99 percent specific.

But test accuracy also depends on disease prevalence.

Even though tests are now fairly reliable in terms of their specificity, a test’s accuracy is also determined by the rate of infection in the population. A lower prevalence of the infection means a higher risk of getting false positives. Right now, the base rate of infection in the U.S. is, according to the New York Times, between 5 percent and 15 percent; this means we could be getting false positives up to half of the time. This is confusing and counterintuitive, but I found James D. Walsh’s explanation over at Intelligencer to be helpful.

Something Gronvall says could help is testing multiple times. She compared it with HIV testing: “If you got a positive HIV test, they would immediately follow that up with another test.” If there was a low prevalence of disease in the surrounding population, you would have a third test on top of that. “That would drive the statistical confidence you have in that test,” says Gronvall. But the difference between a positive HIV-test result and a positive antibody-test result is that the former has significant meaning in terms of the need to seek treatment. Since most people are paying for the antibody test out of pocket, and since the usefulness of the result is unclear, many people are getting tested only once, lowering the statistical confidence in their result.

Let’s say I’m confident in my test result. Does having antibodies for a virus mean I’m immune to the virus?

The answer is not necessarily, both in general and with the SARS-CoV-2 virus specifically. Not all antibodies are the same, and they vary as to whether they actually neutralize their corresponding viruses and for how long. Health experts think that antibodies for SARS and MERS coronaviruses last about a year, and there is hope the SARS-CoV-2 virus will react similarly, but there is, as of right now, no proof that it does. In fact, a new study published in Nature Medicine suggests antibodies may only last for two to three months. (Gronvall points out that this doesn’t necessarily mean you can be reinfected; the immune system is complicated, and immunity might come from another aspect of it. It does mean, though, that it could make testing for antibodies more difficult.)

Casanovas-Massana said this misinterpretation is “perhaps the biggest risk” related to antibody testing — that people will incorrectly equate the presence of antibodies with immunity. “People could stop wearing masks or social distancing — the common things that still need to be in place,” he says.

Beyond that, for antibodies to fight off a second exposure to a virus, you need a certain level of the sort of antibodies that will be effective in doing so, which are called “neutralizing antibodies.” If that antibody level dips below what is necessary to protect you, you’ll be vulnerable to the virus again.

The trouble with the SARS-CoV-2 virus is that it is unknown whether antibodies, rather than another arm of the immune system, are what the body uses to fight the virus; beyond that, it is unknown what level of antibodies are needed and for how long that level might keep a person immune. “We don’t know exactly what amount of neutralizing antibodies is enough to control the disease or how long this protection lasts. Three months, a year, ten years — we don’t know,” says Casanovas-Massana.

“We need time to prove immunity,” he says. “We have to wait and see what happens in the next months and years. Unfortunately, that’s how it is.”

“What a positive antibody test will tell you, if it’s an accurate test, is that you have been exposed,” Gronvall says. “We can’t say that antibodies, or a certain level of antibodies, means that you’re immune or not. We can’t say that immunity is going to last forever.”

What are “immunity passports”?

Some governments, including those of the U.K., France, Chile, Italy, and the U.S., have announced that they would consider “immunity passports” as part of their plan to reopen society. The so-called immunity passports, or immunity certificates, would be granted to those considered immune and would allow holders to return to work, travel, and to whatever else. As you know, it is currently unknown whether having been infected means you are immune, either temporarily or for any significant length of time. Unfortunately, use of the passports relies on a different reality.

Even if immunity were known and qualifiable, Gronvall says the concept of a passport could pose serious risks. The first is that we still do not know how long immunity, if it exists, lasts. “So it’s like: How long are you going to keep this passport, and what are the conditions for it, and how are you going to reevaluate it?” she says.

Another issue is that, although it might not seem this way, particularly in comparison with other countries, the prevalence of the disease in the U.S. is technically low. (Gronvall says she thinks it’s around 10 percent at most.) Because of this, offering an immunity passport would only welcome a very small percentage of the population back to society. “A lot of people would be left out of that,” says Gronvall.

And this would lead to the next issue. “If people think their employment, or their travel, or anything else that’s financially important to them is going to rely on getting or having been exposed, it means you’re going to get fraudulent passports, and you’re going to get people deliberately trying to get infected.” This is bad for the obvious reasons — that it is good to not get sick — and also because we do not know yet what the long-term effects of the virus are.

Because there is so much we don’t know, Gronvall urges patience and caution. “Even if you were exposed, wear a mask,” she says. “Not everybody is going to have an easy time with this virus. You can’t hide out in your house forever, but there is no decision that could come from this antibody test that is going to be a total Get Out of Jail Free card.”

What We Know About COVID-19 Antibody Testing