Months into the pandemic, still no easy answers on coronavirus testing

An abrupt government recommendation this week that fewer people get tested - while at the same time the Trump administration moved to purchase 150 million rapid tests - has made a disorganized testing regime even murkier.

The controversy sparked by the revised standards and, separately, the fraught decisions about testing that confront local school officials and university administrators open a window onto the persistent problems dogging testing in the United States.

"I started to talk to colleagues nationally, and I was taken aback that most [public school] districts and most charter networks really had thrown in the towel quickly and said testing was too expensive and too complicated," said Laura Clancy, chief talent officer for a nonprofit charter school network in Philadelphia and Camden, N.J., that plans to open its 24 schools on a limited basis by Oct. 1 pending community transmission rates and local guidance.

Some universities are already closing their campuses because they can't keep up with outbreaks. A growing chorus of scientists say occasional screening, or surveillance, is not enough; they say students and others need be able to screen themselves at least twice a week.

While President Donald Trump has declared that we have enough or even too much testing, Democratic presidential nominee Joe Biden promised in his convention speech that "if I'm president . . . we'll develop and deploy rapid tests with results available immediately."

Political rhetoric aside, testing companies and policy advocates face regulatory, scientific, logistical and ethical hurdles more than half a year after the virus appeared in the United States.

Turnaround times for most tests have improved in recent weeks, in part because of a substantial drop in the number of tests administered. Still, the Trump administration estimates that over the past month, 1 in 5 tests took more than three days to come back - rendering results largely useless. The equipment needed at the nation's biggest private lab companies is expensive and on back order.

Many companies have developed faster, cheaper tests, but there is a trade-off: Such tests are not nearly as sensitive as the standard coronavirus tests that can take days or weeks to process. The Food and Drug Administration is opposed to letting people take these tests at home unless they meet higher senstivity standards, fearing they would create a false sense of security. And public health experts remain deeply divided.

Testing serves two purposes. One is diagnostic, in which it is used to confirm the presence of the virus in people who have symptoms or a known exposure to someone who is already sick. The other is to detect people who are infected but don't know it. That second purpose is especially important for screening asymptomatic people out of large groups, at schools and elsewhere.

Two kinds of tests exist to look for active infections. Molecular tests search for the virus's genetic material. Antigen tests look for spiky proteins on the surface of the virus.

The gold-standard test is molecular - a polymerase chain reaction (PCR) test performed with a long nasal swab, known to be highly sensitive because of an expensive and time-consuming amplification process that finds even a small bit of infectious material.

Antigen tests are simpler and cheaper but less sensitive; they need more virus in the body to generate a positive result.

The White House just announced a deal to produce 150 million new rapid tests that yield results in 15 minutes, without added equipment. But these tests still must be analyzed by a medical professional. Technically they are approved for use only on people with symptoms, although the White House is touting them as a potential solution for schools and businesses.

That announcement came days after the introduction of new guidance on testing from the administration. It replaces advice that everyone who has been in close contact with an infected person should get tested to find out whether they had contracted the virus. Instead, the updated guidance says those without symptoms "do not necessarily need a test."

Some experts argue the only way out of our current morass is testing that is even cheaper and simpler, performed entirely at home - even if it is not as sensitive.

"This can work better than a vaccine if it could just be produced at scale," said Michael Mina, a Harvard University epidemiologist who has become a vocal advocate for rapid at-home testing. The tests "are not going to tell you if you definitely do or do not have any virus in you. And that's not their role. They're going to tell you if on a given day when you wake up and use this test inside of your house, you are at risk of transmitting the virus to other people."

- - - 

The novel coronavirus is particularly difficult to contain because it has a long incubation period and often leaves no symptoms behind. A test could be taken too early to catch the virus, or the results could come too late, when a person is already infecting others.

That's why some public health experts argue in favor of rapid tests: If they're cheap and convenient enough, they could be taken repeatedly by the same person, something unlikely to happen with more expensive PCR testing.

While the rapid tests might not catch every infection, experts say they should detect the most contagious cases.

"The notion of what a 'junk test' is has to be adapted for the role it's playing," said Carl Bergstrom, a biologist at the University of Washington and another proponent of rapid at-home testing. "It's a substantial shift in what you're looking for."

Bergstrom suggests the FDA develop a separate track for approving such tests, with lower standards and appropriate warnings. People who are sick and need the right treatment, or who know they were exposed to the coronavirus, could get highly accurate PCR tests. Everyone else would proactively take rapid tests, using them as a potential indicator of infection.

An FDA official speaking on the condition of anonymity to explain the administration's thinking said the agency is willing to approve less-sensitive rapid tests that are performed at point-of-care sites, such as a clinic or pharmacy, where a medical professional can offer advice on how to interpret the results. But the agency is loath to open the floodgates to rapid tests performed completely at home.

"Random testing of healthy people just all over the country is not what we're advocating," Adm. Brett Giroir, who leads the White House testing efforts, told reporters recently.

Some critics view even the few antigen tests that have met the FDA's standards as suspect.

"It is better to not test at all and practice social distancing than to deliver a high number of false-negative results, since negative results often lead individuals to engage in risky behaviors that could increase virus spread," said Bob Terbrueggen, president of a Los Angeles genomics company called DxTerity.

Public health experts express concerns that tests taken at home will go unreported, undermining efforts to measure the scope of the pandemic. Already, some states don't report positive antigen test results, resulting in an incomplete portrait of the disease's spread.

With no clear standards, average Americans may be on their own in understanding how best to use these tests.

"You're potentially making consequential decisions, on the individual level, based on test results that are going to be harder to interpret," Jennifer Nuzzo, an epidemiologist at the Johns Hopkins Center for Health Security, told reporters recently.

 

- - -

 

There are already more than 150 coronavirus tests approved for use by the FDA, with little explanation for consumers about what differentiates them.

One school district in Florida's Lake County, for example, decided to screen all teachers and students before the school year began, partnering with a clinic that ended up choosing a rapid antibody test because other options were unreliable or not feasible.

"If you were to see my office . . . there's boxes of rapid tests that we chose not to use," said Matt Cady, practice administrator at Adult Medicine of Lake County. Some antibody tests were recalled by the FDA before they even arrived, while other options gave negative results in people known to be infected.

To cut down on lab time and boost capacity at the nation's big labs, the FDA has approved efforts to pool samples and test groups of them at once, something the administration says at least half of states are doing. But pooling samples works only if most samples test as negative; a positive result means the whole batch needs to be tested again.

Peter Iwen, director of the Nebraska Public Health Laboratory, was one of the first scientists in the country to start pooling samples. Now, he says it's pointless in his state: "The positivity rate in Nebraska is much too high," he said.

The laboratory testing world is dominated by big companies - including Abbott Laboratories, Quest Diagnostics and Laboratory Corp. of America - that carry out about half of U.S. testing. Each machine is costly and processes as many as 2,000 tests a day. The American Clinical Laboratory Association, a trade group, says its members need Congress to establish a federal fund to pay for machines and tests not covered by insurance.

The rest of the testing world is full of small companies.

One called E25Bio in Cambridge, Mass., while still hoping to see its cheap at-home test hit the market, is now aiming for approval with a more sensitive version that requires a lab step.

"These type of antigen tests, while they will never be as accurate and sensitive as PCR, what they are very good at is detecting high viral loads early in infection," E25Bio spokesman Carlos Henri-Ferré said. "What we're talking about is a contagion tracker, a contagion stopper."

Other companies are trying to bypass the debate with affordable, at-home saliva tests that can meet the FDA's standards.

One is an antigen test being developed by a company in Hawaii named Oceanit, currently in clinical trials at The Queen's Medical Center in Honolulu. Oceanit chief executive Patrick Sullivan said that by designing a molecule specifically for the coronavirus antigens, the test achieves greater sensitivity. The current model costs $20, but broader manufacturing could bring it down to $4 or $5, he said.

The company designed the test to be simple enough for 8-year-olds to take before school, but the first clients would be a different unruly bunch: mainland tourists.

"The restaurants, the hotels, the airlines - everybody wants a tool like this," Sullivan said.

The FDA this month granted emergency use authorization for two saliva-based tests that don't require swabs or the chemical testing agents that have been in short supply. The administration is encouraging schools to look into using both. While saliva tests are less common, researchers have found they can be as sensitive as nasal swabs.

SalivaDirect, developed at the Yale School of Public Health with funding from the National Basketball Association, cuts costs down to the level of a cup of coffee - $1 to $4 a test in part by avoiding costly preservatives and using a cheap container. The school is making its test available to any company or laboratory interested in using it.

"Its novelty is its simplicity," said Anne Wyllie, a Yale epidemiologist. "We saw the need for cheap testing because it needs to be accessible around the country. And colleges and schools need frequent testing."

But materials are not the only reason coronavirus tests are hard to come by. Andrew Brooks, a geneticist who runs a Rutgers University lab that in April got the first FDA emergency clearance for a saliva test, said even cheap and fast tests involve staffing and packaging costs. Tests need to be logged, billed and reported to public health authorities.

"What people don't understand is there are other operational issues that impact the ability for the lab to do this, not just reagents," said Brooks, whose lab has analyzed some Major League Baseball tests. "We can't hire people fast enough."

Back in Cambridge, not far from E25, a company called Sherlock Biosciences is developing a paper strip test that offers "lab quality molecular diagnostics in the palm of your hand," in the words of chief executive Rahul Dhanda. Rather than searching for a protein, the test uses the virus itself to create a signal detectable without amplification or special instruments. By next year, Dhanda said, the company could produce a test with "as good or better sensitivity than exists with PCR today."

But he said Sherlock Biosciences was wrestling with the value of holding out for perfection in the middle of a crisis. At the same time, he worries that sacrificing accuracy, even if allowed by the FDA, would make the test harder for people to use, understand and trust.

"We wonder whether we should be making some trade-offs on performance to get the products out faster," he said. "We're torn as to what the right answer is. And I think we're all collectively as a sort of society trying to get to that."