What coronavirus antibody tests tell us — and what they don’t

Widespread testing could reveal who has had COVID-19, but not whether they’ll get it again

Antibody test

To get a better handle on how far the coronavirus has spread, researchers are developing blood tests to detect antibodies in people who were previously infected (one from Mirimus, Inc., shown). But such tests don’t indicate whether those antibodies can protect against future COVID-19 infections.

Misha Friedman/Stringer/Getty Images News

As some countries begin to reopen in the midst of the ongoing coronavirus pandemic, experts are racing to ramp up the development and use of blood tests that pinpoint people who have been exposed to the virus that causes COVID-19 and are no longer infected.

The tests detect antibodies, proteins made by the immune system to fight infection (SN: 3/27/20). People who carry antibodies specific to the novel coronavirus, called SARS-CoV-2, have been infected previously, even if they didn’t know it. For those people, discovering that they have these virus-fighting antibodies could raise hopes of immunity and a return to normal life.

But scientists are also working to uncover what these blood tests really tell us. At this point, there isn’t enough evidence to confirm that recovered people are protected from the disease and, if so, for how long, the World Health Organization said in a statement on April 24. So people hoping for that assurance may be disappointed.

For researchers and public health officials, though, the tests can reveal the true extent of the pandemic. The U.S. National Institutes of Health announced April 10 that researchers had begun recruiting people for a nationwide study that aims to test as many as 10,000 volunteers without an official COVID-19 diagnosis, which could help clarify how many people across the country have actually been infected. A number of similar, more local studies are also under way.

The goal is to fill in the gaps created by trouble rolling out diagnostic tests, which detect the virus’s genetic material and can catch an active infection. Those tests have faced roadblocks such as flawed tests and supply shortages, leaving some sick people wondering whether their symptoms were from COVID-19 or a different respiratory infection.

Such tests can’t detect the virus once the infection has cleared. But antibodies typically stick around in the body after the virus has disappeared, giving scientists a glimpse into the past. So for people who weren’t able to get a diagnostic test, the antibody test “will give us the ability to let them know yes, you did have COVID-19,” says Aneesh Mehta, an infectious disease physician at Emory University in Atlanta.

Knowing how many people have already been exposed to the virus is also a step toward understanding when the pandemic might end (SN: 3/24/20). High numbers of immune people can protect the population as a whole from outbreaks, creating what’s called herd immunity. Researchers estimate that around one-third to two-thirds of a population would need to be infected with SARS-CoV-2 to reach herd immunity.  

Positive or negative

For an individual, an antibody test result isn’t black and white: exposed or not exposed, immune or not immune.  

This is in part because antibody tests are not 100 percent accurate, says Angela Rasmussen, a virologist at Columbia University. “They don’t accurately detect every single antibody, and they may have both false positives and false negatives.”

In the tests, a small sample of a patient’s blood is taken and exposed to proteins that match parts of the virus. If any antibodies specific to the coronavirus are present, they should recognize and bind to the virus components. Such antibodies might attach to any of a variety of places on the virus — including spots that are similar among closely-related viruses. As a result, researchers have to carefully choose which part of the new coronavirus to use.

The tests usually detect two types of antibodies. One, called IgM, is typically produced about a week after infection and could identify patients who may still be infected.Levels of IgM begin to wane as the body makes another type of antibody called IgG, which can persist for longer periods of time.

The best antibody tests are both highly sensitive — detecting a wide range of IgM or IgG antibodies that recognize different parts of a viral protein — and highly specific, meaning the detected antibodies are for only that virus. Coronaviruses that cause colds, for example, also circulate around the globe. Antibody tests with low specificity and high sensitivity might detect antibodies against cold viruses and give a false positive. But a test with high specificity and low sensitivity could miss antibodies, resulting in a false negative. Timing is also crucial, as patients who have not been infected long enough to develop antibodies would test negative.   

Though coronavirus antibody tests have flooded the market, the U.S. Food and Drug Administration has so far authorized just eight for emergency use. Based on data the tests’ developers provided to the FDA, the sensitivity and specificity of the tests vary widely, from 88 to 100 percent for sensitivity and 90 to 100 percent for specificity.  

In addition, “certain people just don’t make as much antibody as other people do and don’t respond the same to an infection,” Mehta says. “There will be some people that had the infection, but we won’t be able to detect them.”

Immune or not

Even if the tests are accurate, immunity is not a given. The tests being rolled out now look only for the presence or absence of antibodies, not how effective those particular antibodies are at knocking out the virus. And like antibody levels, that can vary from person to person.

“We need to look at people who do have antibody — and that does seem to be the majority of patients — and see if that antibody is protective,” Rasmussen says.

In one cluster of COVID-19 patients from China, most people produced high levels of neutralizing antibodies that prevent the virus from infecting new cells, according to a preliminary report posted April 6 at medRxiv.org before peer review. Of 175 patients with mild symptoms, about 70 percent developed antibodies around 10 days after symptoms began. Among those, elderly and middle-aged people had the highest levels.

Younger patients tended to have lower levels overall, including 10 people who didn’t have any detectable antibodies. But it’s possible that those 10 people developed antibodies that recognized a different viral protein than the one used in the test, producing a false negative result. It could also mean that a different arm of the immune system — one that targets infected cells and doesn’t leave behind antibodies — may play a heavy hand in recovery.    

If researchers find particularly effective antibodies, they could be used to develop treatments, such as producing lab-made antibodies or giving plasma from recovered patients to those that are sick (SN: 4/3/20).   

Long-term protection

It’s unclear what antibody levels provide the best defense.

“It’s very hard to protect the nose from being reinfected,” says Mark Slifka, a viral immunologist at Oregon Health and Science University in Portland. People with lower antibody levels could be more at risk for reinfection, he says, “but you might also be able to still ramp up a rapid [immune] response” and prevent the virus from spreading deeper into the lungs.  

But even people with protective antibodies aren’t necessarily “bulletproof,” Slifka says. Some people may be completely protected against infection, a state known as sterilizing immunity, which is hard to achieve. Others may get infected again, but have mild or no symptoms.

Anecdotal reports from South Korea and China of patients who tested positive after recovering have suggested that some people could be reinfected. But nasal and throat swabs from 12 of these “reinfected” COVID-19 patients had no infectious viruses despite testing positive for the virus’s genetic material, officials from the Korea Centers for Disease Control and Prevention said in a news briefing on April 23. It’s possible the diagnostic test is detecting lingering fragments of the coronavirus as it is cleared from the body rather than a reinfection.

When two rhesus macaques were infected, allowed to recover and then exposed to the virus again, they weren’t reinfected, according to preliminary findings posted March 14 at bioRxiv.org. That hints that at least in the short term, people may be protected. To know for sure, “we would have to follow recovered COVID-19 patients who are antibody-positive for a long period of time and see if any of them become infected again,” Rasmussen says.

Researchers also don’t know how long SARS-CoV-2 antibodies stick around. Some viruses, such as measles, can trigger protection that lasts a lifetime. Defenses against other viruses can wane over time. Studies suggest that protection against coronaviruses that cause the cold can last for about a year. Antibodies for the original SARS virus, on the other hand, slowly vanished over a few years. Since SARS no longer infects people, it’s unclear whether a lack of antibodies means lack of protection. And because the new coronavirus has been infecting people for only a few months, it’s still unknown whether it will behave similarly.

Despite the unknowns surrounding antibodies and SARS-CoV-2, some countries including the United States are considering using antibody tests as a stepping stone to provide so-called “immunity certificates” to people who test positive that would allow them to reenter society or return to work. Some experts, however, are skeptical.

“We don’t know that having antibodies necessarily means you’re immune, so it could give people a false sense of security about how safe they are,” Rasmussen says.

And, according to the WHO, that could increase the risks that the virus will continue to spread.

More Stories from Science News on Health & Medicine

From the Nature Index

Paid Content