Health officials in Germany, the UK and the US are looking at antibody testing to figure out when and how their countries could open back up. Some authorities are even considering issuing “immunity passports” that would help key workers safely and quickly return to their jobs.
But some early Covid-19 antibody tests, and the research that hinges on them, have been deeply flawed. Infectious disease experts warn that while reliable antibody tests could eventually reveal much about how the disease spreads and how deadly it is, it is too early to draw any definitive conclusions from screening studies.
What is an antibody test?
PCR tests, the diagnostic tests most commonly used to determine whether someone is currently infected with the coronavirus, looks for the virus’s genetic material in samples collected from patients’ throats or noses.
Antibody tests, or serology tests, rather than looking for the virus itself, detect antibodies or proteins in the blood that our bodies develop in order to fight it off. “These tests tell you that you’ve been exposed to the virus,” said Akiko Iwasaki, an immunologist at the Yale school of medicine. “Antibody tests can basically give us a history of the infection.”
What exactly are these antibodies that we’re testing for?
An antibody is a bespoke weapon that our bodies develop to fight off a pathogen. Our bodies may make several types of antibodies uniquely engineered to fight off the coronavirus over the course of one or two weeks.
“Think about those pictures of the coronavirus you’ve mostly commonly seen,” said Diane Griffin, a microbiologist and immunologist at the Johns Hopkins Bloomberg School of Public Health. “You know those big, blobby spikes on the surface of the coronavirus? That’s the part of the virus particle that attaches to healthy cells to invade and infect them.”
Antibodies to these spikes on the coronavirus glom on to them and coat them – leaving them unable to hijack our cells. Other types of antibodies may mark infected cells, so other cells can destroy them before the infection spreads further. Some of these antibodies will stick around in our blood long after the virus is gone, protecting us from future infections.
We know that antibodies against certain pathogens – such as smallpox – can provide protection for a lifetime. Antibodies for other infections tend to fade: Antibody levels in people who had recovered from Sars and Mers declined over the course of one or two years.
Scientists are still trying to figure out how much of which antibodies give people immunity to the new coronavirus. Once they do, really good lab tests could tell patients with a certain “antibody level” that they’re safe from infection. The ideal antibody tests would “give you value – similar to how people are used to getting a cholesterol value, or a sugar level”, explained Michael Mina, an epidemiologist at the Harvard TH Chan school of public health.
How do the tests work?
There are two main types of antibody tests: rapid tests that take a finger prick of blood and reveal whether a person has antibodies, and tests that use a lab technique called Elisa and indicate the quantity of antibodies present.
And are they accurate?
In a preliminary study published this month, a team of researchers sought to verify the 14 coronavirus antibody tests currently available in the US and found that only three delivered consistently reliable results. Four of the tests produced false positive rates from 11% to 16%. Others produced false positives about 5% of the time.
The ideal tests would be highly sensitive – meaning they can pick up on the presence of antibodies in the blood of those who have infections – and specific, meaning they detect the right antibodies. Tests that aren’t specific might light up positive after picking up on antibodies to other infections, giving people the false sense that they’re immune.
Both types of tests have shown issues. The rapid yes-no tests, which Boris Johnson gushed were “as simple as a pregnancy test”, turned out to be much more complicated, said Alexander Marson, an immunologist at the University of California, San Francisco, who has led the effort to test the tests.
The Elisa tests are generally considered to be more reliable, but none of the tests studied by Marson and his team had perfect specificity.
“I understand that there’s an urgency during this unprecedented time to move things along, to understand who’s been infected and who is immune,” said Marson. “But before we race to get information from these tests, we need to actually have evidence that they work.”
Why are there so many faulty tests?
In the US, federal agencies were criticized for initially dragging their feet in allowing private and academic labs to develop their own versions of the PCR tests used to diagnose Covid-19. When it came to antibody testing, the government may have swung too far in the other direction, lawmakers and public health experts have said.
The US isn’t the only country to have been duped by companies selling faulty test kits. This month, British officials realized they may have jumped the gun by ordering 2m at-home antibody test kits that researchers at the University of Oxford found to be unreliable.
The Food and Drug Administration allowed about 90 companies, many based abroad, to sell antibody tests that haven’t been vetted and validated by federal scientists. The FDA has since warned that some of these tests are inaccurate.
So, when can I get an ‘immunity passport’?
Not for a while, it would appear. Last week, the World Health Organization warned that the detection of antibodies alone shouldn’t serve as a basis for an “immunity passport” – because scientists don’t know how much of which antibodies will protect people from reinfection.
To get there, we need to collect lots of data over time from lots of people who have had the infection and developed antibodies, explained Yale’s Iwasaki – and track who gets reinfected, and who remains immune.
Does it still make sense to test everyone for antibodies?
Germany is planning to conduct nationwide antibody testing. In California, researchers have sought to test nearly everyone in the small town of Bolinas, and in the Mission neighborhood of San Francisco, administering both PCR and antibody tests.
By using blood samples to conduct lab tests, scientists will be able to try more than one antibody test on the blood, verifying and validating their results, Griffin said. Eventually, these large-scale testing efforts could provide more information about how many people within a community have had the coronavirus. Tallying the number of infections and the levels of antibodies and tracking these communities over time could answer critical questions about how widely the virus has spread, who has immunity and how long that immunity lasts.
“Once we know the answers to these key questions, maybe then we can issue immunity passports,” Iwasaki said. “Until then, we need to be patient.”