“I’ve never felt the weight of my profession more than I have in the past 10 days,” says University of Washington Medical Center (UWMC) virologist Alex Greninger, MD, PhD.
Over that span, he estimates, the lab where he is assistant director has tested more than 4,000 patient specimens for the new coronavirus — the cause of a disease that has stricken more than 127,000 people worldwide and killed more than 4,700, according to the live coronavirus tracker developed and maintained by Johns Hopkins University.
“If we report a negative test, and they release somebody, and we’re wrong — you’ve put them, family, friends, co-workers at risk,” Greninger says. “That’s why it’s so critical to make sure the tests are accurate.”
As the coronavirus spreads, researchers at academic medical institutions have jumped into the containment battle by developing and deploying tests to detect the virus. They hope to help fill a nationwide shortage of such tests — but they’ve been impeded by bureaucratic roadblocks that might have delayed thousands of tests in recent weeks.
Labs at UWMC, Johns Hopkins, and Stanford University are collectively running nearly 1,500 tests a day on processes that they’ve developed in recent months, while other academic institutions await federal approval to implement their own tests, including Columbia University’s Mailman School of Public Health, the University of North Carolina (UNC) School of Medicine, the Mayo Clinic, and the University of Texas Medical Branch (UTMB) at Galveston.
The researchers note that academic medical institutions are particularly well-suited to develop the tests because of their lab capacity, staff expertise, culture of collaboration across disciplines, and dedication to their research missions.
“We develop tests all the time,” says Melissa Miller, PhD, director of the Clinical Microbiology Laboratory at the UNC School of Medicine. “We know how to validate them. We have colleagues and collaborators on campus or on other campuses to get specimens, to get ideas from. We’re not doing this in a silo.”
“Our laboratories have worked with these kinds of viruses for 35 years,” says Gregory A. Poland, MD, director of the Vaccine Research Group at Mayo Clinic.
Yet until recently, researchers at those labs could do little more than watch the virus grow into a pandemic while the tests they created to track it sat unused.
A slow start
The virus emerged in China in late December 2019, and by January, researchers at university labs began working to develop procedures, called assays, to measure for the presence of the virus in specimens. Typically, coronavirus tests use polymerase chain reactions to detect genes within a sample (such as from a throat swab) that are specific to the virus. Most of the researchers worked not only with samples of the virus but with “contrived samples,” which substitute for unmodified body fluid or tissue. The Centers for Disease Control and Prevention (CDC) explains the tests.
“I saw what was going on” with the spread of the virus overseas, says Benjamin Pinsky, MD, PhD, a clinical virologist at Stanford. “I was concerned that we need to have testing capacity, so we moved ahead with the thought that we would be able to perform testing at some point.”
By mid- to late-February, researchers at Stanford, UWMC, and UNC had developed tests for the virus, yet they saw no quick path to get those tests validated for clinical use on patient samples. For that, they needed an Emergency Use Authorization (EUA) from the Food and Drug Administration (FDA), which required an approval process that many researchers found significantly more onerous than the process for other tests developed by labs and far too slow to respond to the rapidly expanding coronavirus outbreak.
Stanford had validated its test “to the best of our ability,” Pinsky says, but its process got stalled by the federal requirements for an EUA.
In early February, the FDA issued one EUA: for the coronavirus test that the CDC distributed to public health labs. Soon thereafter, the CDC announced that one part of the test didn’t always perform properly, rendering many results inconclusive. The CDC sent replacement test kits — a process that took up to two weeks.
“How could this not have been tested and ready to go? I don’t understand that,” Poland says.
The snafu frustrated university lab researchers who held tests they couldn’t deploy.
“It’s in the tertiary care centers and academic medical centers where you have people who are involved in research and who naturally would develop an assay against something like this,” Poland says. “But they’re prohibited from using it” because the test isn’t approved by the FDA.
One experience particularly irked Greninger. While preparing to submit documents to the FDA to get feedback on a test approval application, the agency informed him that an electronic submission would not suffice; he had to send hard copies to an office in Maryland by FedEx.
“I thought that was unfortunate, in 2020,” he says. “The virus moves faster than the FDA.” Researchers pressed the federal government to ease the way to approve tests more quickly.
On Feb. 28, more than 100 clinical laboratory directors and clinical colleagues (many of them affiliated with academic medical centers) sent a letter to selected U.S. House and Senate committees seeking congressional action to expedite test approvals. “This regulatory process is significantly more stringent than that required for every other virus we test for,” the letter says.
That same day, the American Society for Microbiology sent a letter to the FDA urging it to enable its members (which include academic medical institutions and private labs) to more quickly get emergency authorizations for tests they’d already developed and could start using “tomorrow.”
The emergency authorization process “is proving a hindrance to rapid identification of potential COVID-19 infections,” the letter says, referring to the disease caused by the virus. “If we have tests that are safe and effective that cannot be used, this can put people at risk.”
The next day, the FDA showed that it agreed.
The FDA’s new process
On Feb. 29, a Saturday, the FDA issued a guidance allowing labs that already meet certain federal requirements to have their experimental coronavirus test results validated by a lab that uses a different test. For university labs, that meant turning to a public lab that had been using the CDC test.
For a new test to move to clinical use, the guidance requires confirmation on five positive results and five negative results. But finding positive specimens was difficult in some areas of the country, and that delayed some university efforts.
“We didn’t have any positive” patients confirmed in North Carolina, says Miller. Researchers tapped their networks to track down samples. Miller’s microbiology lab got a viral RNA specimen from a colleague at another UNC lab. Johns Hopkins got genomic RNA from UTMB, says Heba Mostafa, PhD, an assistant professor of pathology who is part of the coronavirus diagnostic assay implementation effort there. UTMB’s World Reference Center for Emerging Viruses and Arboviruses had received viral RNA material from the CDC and announced that it could provide some to labs that needed it.
While getting their positive and negative results confirmed, Stanford, UWMC, and Johns Hopkins began testing specimens from hospitals and physicians. UWMC recently increased to 1,000 tests a day and will probably reach 1,200, Greninger says. Stanford says it was approaching hundreds per day by midweek. Johns Hopkins ran 50 tests on its first day this week, and Mostafa believes it can reach 200.
The labs are conducting these initial tests under what amounts to temporary permission. The FDA guidance says that within 15 days of getting the independent validations, the labs should apply for an EUA, which will formally allow them to run the tests. Meanwhile, the FDA “does not intend to object to the use of these tests for specimen testing.”
As of March 11, according to the FDA, an EUA had been issued for one other test besides that of the CDC, created by the New York State Department of Health for use at various state and city labs. The FDA says it is assessing other EUA requests “quickly” but has to move carefully as well because false test results “could have a broader public health impact beyond the individual patients being tested.”
No one knows how many tests the United States will need in the coming weeks, other than more. On March 6, Vice President Mike Pence, who leads the U.S. COVID-19 task force, said, “We don't have enough tests today to meet what we anticipate will be the demand going forward.”
In a March 11 statement, AAMC President and CEO David Skorton called on the federal government to take several steps to combat the virus, including “increasing the availability and capacity of testing.”
As of that day, according to daily tracking by the American Enterprise Institute, U.S. facilities of all types were running a little over 16,000 tests a day, with California institutions accounting for half the total and public health labs accounting for more than 3,000 others.
The onboarding of university and private labs should take some of the pressure off public labs and speed up test results because of the increased capacity. “One goal is trying to move away from inundating state labs,” says Saskia Popescu, PhD, MPH, a senior infection preventionist and fellow of the Johns Hopkins Center for Health Security.
It remains to be seen how much university labs can contribute to testing needs. Among those that hope to start soon is UNC, which can “comfortably” test 100 a day, Miller says, and might “ramp up” to 300.
At Johns Hopkins, Mostafa believes the contributions will go beyond test results.
“Academic institutions have a critical role also in trying to educate and spread knowledge,” she said. “Especially in a situation with an extensive outbreak, where everyone is scared. This is the mission of an academic institution.”