Scientists Seek Ways To Speed Up COVID Testing

As the coronavirus pandemic grows, so do the number of people griping about battles with insurance companies for approvals to get a COVID-19 test and week-long waits for results that feel useless by the time they arrive.

Public health experts agree: California needs to do far more coronavirus testing, especially among people without any symptoms, before offices, schools and more businesses can safely reopen. But that could make the waits for results even longer.

So how can the Golden State quickly fix its broken testing system?

The answer could be as simple as "pooling" a dozen samples into one test, like labs already do with blood, some experts say, or as innovative as using genetic sequencing to perform tens of thousands of tests in a single day with next-day results.

"I recognize our responsibility to do more testing," Gov. Gavin Newsom said Friday, pointing to pooled testing as a solution. "We'll be pushing that very, very aggressively."

Universities and health systems across the state, from UCLA and Kaiser to Stanford and Sutter, are working on the effort, he said.

Instead of taking nasal swabs from, say, eight people and analyzing them all separately, pooling allows labs to combine the samples and run a single analysis. If the grouped sample comes back positive, all eight people have to be tested again individually. But if it comes back negative, it's more than likely no one in the group has the virus.

"The goal is to identify outbreaks before they occur," said James Zehnder, director of clinical pathology at Stanford.

The process could stretch supplies and let fire departments, colleges, sports teams, businesses and others test large groups and "really help open the economy and schools and so forth," he said.

Stanford Medicine recently got the green light from the U.S. Food and Drug Administration to start performing pooled testing and plans to screen asymptomatic nursing home residents, first responders, health care workers and others in the coming weeks.

There are some trade-offs with pooling. Testing in batches reduces the sensitivity, or likelihood that the results are accurate, and it works best in places where there isn't a lot of virus spreading already and people can be tested on a regular basis. And because individual samples still need to be collected, the supply chain issues - a lack of nasal swabs, for instance - that have hampered existing testing efforts in places still matter.

It's difficult to say exactly how much pooling will increase testing numbers, because it's not clear yet exactly how many samples labs can run at once. But, Zehnder said, the number could range from four or five to 20. Regardless, he said, "it's a huge force multiplier. That's where the real power is."

And Stanford is working on automating parts of testing, such as unscrewing tubes and pipetting samples, which could speed up the process.

The Santa Clara Valley Medical Center in San Jose is also performing pooled testing in a bid to save both time and resources. The county health system says that it's increased testing to an average of about 3,000 a day.

The pandemic "is revealing a lot of shortcomings in our health care systems," said Zehnder, at Stanford, "but also opportunities."

Still, health experts say labs performing a few thousand tests here and there are just a tiny fraction of the overall testing needed to reopen schools and businesses.

"We're not doing enough testing in California or anywhere in the country," Bob Kocher, a member of the state's testing task force recently said during a panel at the Commonwealth Club.

The state could jump from just north of 100,000 tests a day to 400,000 if it could get enough sample-collection materials, Kocher said. But while demand is rising exponentially, he said, manufacturing is growing at a linear rate.

San Mateo-based Helix is looking at another way to massively increase testing: through a process called next-generation sequencing. Essentially, scientists can look for the virus with machines typically used to detect cancer risk and genetic diseases.

The upside of next-generation sequencing, said co-founder and chief scientific officer James Lu, is that the tests are fast, scientists can run a lot at one time and they rely on some different supplies than more traditional coronavirus testing. Each instrument, he said, could potentially run 25,000 tests a day.

The process can also be used to look for mutations in the virus, which can give infectious disease and public health experts an idea of where certain case clusters originated and how COVID-19 is spreading. And it can help scientists understand whether genetic differences lead to variations in how the coronavirus affects different people.

But launching testing on such a major scale is challenging, and while many of the supplies are different, patients will still need to be swabbed, so those supply constraints are still limiting.

Helix hopes to be running next-generation sequencing to test for the virus with next-day results later this summer or fall, and, Lu said, "we're considering pretty dramatic scaling."

In April, UCSF and Mammoth Biosciences announced they were using a different tool - gene-editing CRISPR technology - to develop a coronavirus test that could also boost testing and provide results quickly.

Kocher thinks driving down the cost of testing - currently the tests cost $100 to $200 apiece - is particularly important, so more people can be tested and more frequently, something he hopes happens before the flu season.

"If we don't get with that plan quickly," he said, "I think there'll be real problems in the fall."