On ultra-wealthy Fisher Island, billed as “Florida’s Most Exclusive,” every resident and worker is being tested for antibodies to COVID-19. The price tag? A relatively modest $31,000 in all.
In Colorado’s San Miguel County, home to the fancy resort town of Telluride, a plan to test roughly 5,500 residents teetered before stabilizing in recent days. It was spearheaded by a biotech company whose executives are part-time residents.
And in the rich Marin County, California, enclave of Bolinas, residents are benefiting from free antibody tests as part of a $400,000 study financed by donations raised by a local venture capitalist.
While diagnostic tests—to determine who does and does not have currently have COVID-19—remain more difficult to access than many experts believe is necessary to end lockdowns, antibody tests remain the ultimate prize. Rich and well-connected people across America have jumped at the chance to receive early blood-based antibody testing for COVID-19 to determine if they have previously been infected and might be carrying protective armor against a terrifying pandemic.
But a flood of antibody tests, many of which have yet to be independently checked for quality and some of which carry disturbingly high rates of false positive and false negative results, are threatening to do more harm than good.
In other words, antibody tests are gobbling up elite cash without necessarily offering what 1 percenters desperate for proof of immunity have been hoping for. In comments to Reuters on Wednesday reflecting the absurdity of the moment, the CEO of biotech giant Roche called the glut of dubious antibody tests a “disaster,” even as the company hopes to produce its own.
“I think that the desire, the hope is that these tests will reveal who is immune from re-infection from SARS-CoV-2,” said Mary Hayden, chief of the division of infectious diseases and director of the clinical microbiology lab at Rush University Medical Center in Chicago. “And so far, we have no information about that. We don’t know whether any antibodies that are detected are protective.”
Misinterpreted results, she said, could lead people to put themselves in potentially unsafe situations. “We’re really trying to emphasize to people that even if you do have these antibody tests done, regardless of the result, you should continue to employ all of the safety measures that are recommended,” Hayden said.
Not everyone has gotten the message. Good Morning America anchor George Stephanopoulos was accused of flouting social distancing and mask-wearing guidance while sheltering in the Hamptons after his wife tested positive for the virus. As if in response, Stephanopoulos announced on April 21 that he had tested positive for COVID-19 antibodies, “confirming that I cleared the virus after weeks without symptoms.”
On a community level, antibody (or serology) testing may indeed help researchers with surveillance efforts to estimate the true prevalence of the viral disease in different parts of the country. Multiple university-backed studies have been published in recent days or are currently underway. Even Major League Baseball is getting in on the action with a study of up to 10,000 of its employees. The tests can also help assess vaccine candidates and determine which antibody-producing volunteers may be best suited for donating their plasma to treat desperately sick patients.
“We absolutely have an understanding now that these tests will have a place in helping us move forward,” said Nathan Ledeboer, medical director of the Clinical Microbiology and Molecular Diagnostics Laboratories at Froedtert and the Medical College of Wisconsin in Milwaukee.
The issue, he and other experts stressed, is that even high-quality antibody tests still have little to say about whether a person with antibodies against the coronavirus is truly immune from being re-infected or infecting others. But many test developers are muting that message or implying otherwise, and those claims are being further undercut in some cases by poor performances that have rendered the tests essentially useless.
So far the FDA has granted emergency use authorization to only four of the more than 110 tests on the market, but has permitted the others to remain available. On March 16, the agency asserted that the other antibody tests “should be validated prior to use.” The guideline, however, did not require such validation to be independently checked—that is, by anyone other than the company making them—prior to marketing.
Antibody testing, meanwhile, has become a critical component of many cities’ and states’ phased-in plans to reopen. The tests, while not a replacement for nasal swab-based diagnostics to identify active infections, can indicate whether someone in any given area was exposed to the SARS-CoV-2 virus in the past. If further tests can establish that individuals who no longer have active infections have acquired some level of immunity, they could be cleared to rejoin the workforce. Based on past experience with other coronaviruses, scientists believe that some level of protective immunity is likely, but they haven’t yet performed the experiments needed to prove it.
Angela Rasmussen, a virologist at Columbia University in New York, worries about public officials relying on the antibody test data to make “huge policy decisions” that affect both public health and the economy. “I think it’s incredibly dangerous to just be like, ‘Oh, antibody-positive, cool. Problem solved. Let’s start making policy about who can come out and go back to work,’” she told The Daily Beast.
False positive test results could be especially problematic if the testing on its own is used as the basis for issuing “immunity certificates” or a virtual clean bill of health, Rasmussen added. “If you are sending a significant number of people out into the world with the assumption that they are positive for antibodies and therefore protected, and a significant number of those people are not, that could obviously put those people at a higher exposure risk,” she said.
People who mistakenly believe they’re protected and contract a viral infection, in turn, may initially dismiss their symptoms and put others at risk; a considerable body of research now suggests that even asymptomatic and pre-symptomatic people—or “silent spreaders”—can transmit the coronavirus.
Another big issue is that in the U.S., overall prevalence of the coronavirus still appears to be relatively low. A preliminary study of residents in California’s hard-hit Santa Clara County, which hasn’t been peer reviewed and has attracted criticism over its sampling methods and other shortcomings, suggested that true COVID-19 cases may be 50- to 85-fold higher than the county’s confirmed cases.
Even so, Ledeboer noted that the infection’s estimated prevalence was less than 6 percent, at most, in that disputed study. Given that low prevalence rate, he said, an antibody test’s false positive rate becomes even more important. “Assume a false positive rate of 5 percent, which isn’t terrible,” he said. “But in a population in which you have a prevalence of 6 percent, you have a nearly equal chance of getting a false positive and a true positive.” In other words, the chances that a positive test result is truly accurate would amount to a virtual coin flip.
On Thursday, Gov. Andrew Cuomo released preliminary results from a New York Department of Health study that tested supermarket shoppers throughout the state. Roughly 14 percent of 3,000 participants tested positive for COVID-19 antibodies, though the non-random sample means the high numbers may not accurately reflect New York’s overall prevalence.
Elsewhere, horror stories have begun to accumulate. In Laredo, Texas, in fact, city officials snapped up 20,000 antibody tests to conduct community-wide testing. But as recounted in The New York Times, the manufacturer’s advertised reliability of 93 to 97 percent was actually closer to 20 percent when tested by the health department.
Recent debuts like a University of Washington-developed antibody test that went live on Tuesday have received far more favorable reviews based on initial reliability tests. Ledeboer said the data he’s seen suggest that Colorado’s San Miguel County is likewise using a high-quality antibody test that has been independently evaluated. In an April 21 press release, the county noted that its testing had turned up 26 positive results, 4,659 negative results and 72 “borderline” results.
People who received that borderline result would need to be retested to clarify their status, he said. “For the 26 people who have the antibodies, they likely have immunity, but we need more data on that,” Ledeboer explained. The mere presence of antibodies, he reminded, doesn’t mean they cannot transmit the virus. Depending on the timing of their tests and initial symptoms, if any, they may need to be followed up with a test that directly looks for the presence of the virus to rule out an active infection.
San Miguel County spokesperson Susan Lilly confirmed that residents who have tested positive have been asked to quarantine for 14 days from the date of the test and have been offered nasal swab or saliva-based tests to check for direct evidence of an active case. Once cleared, she said, the positive antibody results do indicate “presumptive immunity,” though she stressed that the duration of that immunity remains unknown. Lilly also said that the county’s antibody testing has provided one among many data points that will be used in officials’ decision-making over reopening. “It’s a very deliberate and measured process,” she said.
Some antibody tests have been likened to the yes-no results of a rapid pregnancy test, while others provide a more in-depth look at the amount of antibodies produced. But even the latter tests are limited in what they can reveal.
“Certainly from a policy perspective in going forward, we don’t really know enough to say, ‘OK, you have antibodies. You’re good,’” Rasmussen said. “We’re going to need to have much more broad rollout of these serology tests and actually start following some of the people who are sero-positive to get a better understanding of what it actually means to be antibody positive, and if you are antibody positive, how many antibodies you need to be protected.”
So-called neutralizing antibodies ultimately yield the protection so many are so hungry for, Ledeboer said, but recent preprints have yielded conflicting evidence about what fraction of patients produce these antibodies and in what quantity.
At Tuesday’s White House briefing, FDA commissioner Stephen Hahn said the agency’s more flexible policy for antibody tests was aimed at allowing developers “the opportunity to quickly get their test to market and address the needs of the nation as posed by the epidemic.” He reiterated the expectation that manufacturers and developers self-validate their tests.
Hahn said 140 test developers were pursuing applications under the FDA’s emergency use authorization, and he asserted that the agency was cracking down on developers who falsely claim they’ve already gained authorization or approval. Critics say the larger problem is the lack of independent quality checks on tests that aren’t explicitly authorized and are nonetheless available for commercial sale under the relaxed regulations.
The FDA, Hahn said, is now working with the CDC and the National Cancer Institute to perform an independent validation on some of the marketed tests. “We expect to hear more information this week about that, and we will provide that information in a transparent manner,” he said.
As the U.S. and individual states grapple over the best way forward, Hayden said antibody testing has a lot of potential for good. But she and other experts agreed that more realistic expectations are needed, as well as smart conversations about how best to protect individuals and how antibody testing might aid in that effort. “We really have a lot of expertise in this country; we need to take advantage of that expertise,” Ledeboer said.
In the meantime, it’s safe to assume that wealthier enclaves and individuals will continue to shower cash upon antibody test manufacturers of varying quality. In that sense, one thing money can still buy is the illusion of security.
