NEW YORK – In a rigorous evaluation of clinical samples, researchers at the University of Washington's Department of Genome Sciences have shown an extraction-free COVID-19 assay using home-collected dry swab samples has 100 percent sensitivity compared to standard RT-qPCR testing.
The SwabExpress protocol — which uses a simple cotton swab and other widely available reagents — could help overcome ongoing supply chain issues.
The UW team's evaluation was published last week in Clinical Chemistry, and they now hope it will be helpful to other labs, particularly in the face of the more highly transmissible SARS-CoV-2 delta variant.
"We wanted to put the protocol out there — because it is very modular, the workflow will work for other people using a different PCR platform on the back end," said Christina Lockwood, a clinical lab director and senior co-author of the study. "You can use whichever piece seems like it fits best in your specific laboratory workflow for a laboratory-developed test," she added.
The researchers examined thousands of prospectively and retrospectively collected samples. Overall, they showed the extraction-free front-end protocol had a sensitivity of 100 percent and a specificity of 99.4 percent compared to standard extraction-based methods.
The assay detects two SARS-CoV-2 genes — the spike, or S gene, and Orf1b. The study determined that the RT-qPCR analytical sensitivity to be 2 molecules SARS-CoV-2 per microliter of dry swab eluate for the Orf1b assay and 4 molecules per microliter for the S gene. This limit of detection is comparable to the limit of detection of numerous other RT-qPCR-based tests that have been issued Emergency Use Authorization by the US Food and Drug Administration, the authors said.
Notably, this is also more sensitive than other extraction-free SARS-CoV-2 testing protocols, including ones that have been granted EUA.
Lea Starita, a senior co-author on the paper and lab director, said in an interview that there are a few reasons that running extraction-free RT-qPCR on a home-collected dry nasal swab could show high sensitivity.
For example, typical viral transport media, or VTM, and universal transport media, or UTM, contain salts, which are PCR inhibitors, Starita said, and essentially necessitate extraction if only just to remove the inhibitors.
"Because we are shipping dry swabs, we could find a buffer that was very friendly to PCR," she said.
Lockwood also noted that the SwabExpress protocol uses carefully selected primer and probe pairs.
Many extraction-free protocols use saliva and pooled samples. Nasal samples tend to have higher concentration of virus than saliva, however, and are less likely to contain contaminating substances. And while pooling can increase testing throughput, it dilutes the viral RNA which can lower sensitivity.
The team was inspired to develop the protocol after spending untold person-hours trying to procure supplies. Swabs, UTM or VTM, and plastic consumables like pipette tips had been difficult to find, "and, once we found them, we had to pay exorbitant shipping fees, only to have them not show up," she said.
The team also wanted to simplify its protocol, "so we could do more with fewer people," Starita said.
The overall turnaround time is also significantly faster with the SwabExpress protocol, Lockwood said. "The accessioning and barcoding are easier when we don't have the complication of UTM or VTM," she added. "The smaller anterior nares swabs help that, as well."
Indeed, the swab the team sends to patients resembles consumer brand Q-tips, according to the study.
While at-home sample collection was once controversial, the pandemic has shown it to be a reliable method. There are currently at least 17 diagnostics developers whose EUAs include home collection kits, although these all appear to be collected into liquid transport media.
The UW team had also previously validated the swab for flu testing and found that patients were able to perform the collection reliably.
"Even when people make pretty big mistakes, we still don't seem to have a problem getting accurate results for their tests," Starita said. For example, early on, the team was using midturbinate swabs and found that if a patient put the wrong end of the swab in their nose, the test results still weren't impacted.
The buffer used to reconstitute the dry swab is another innovation of the method.
Sanjay Srivatsan, first author on the Clin Chem study and UW graduate student, spearheaded many of the efforts, Starita and Lockwood noted.
This included an exhaustive evaluation of the ways different buffers used to reconstitute the dry swab impact PCR cycle thresholds. He compared Ct values for each of the three probes in the assay for elution buffers containing one of three detergents across tenfold dilutions.
In the end, Srivatsan found that a Tris buffer with a smidgeon of EDTA, also called low-TE buffer, and without other detergents, was best suited for direct RT-qPCR.
"Low-TE can be quickly prepared using reagents commonly found in laboratories," the authors noted in the study.
One of the other tricky elements of the dry swab protocol had been the viscosity of some patient nasal samples. Starita said this issue also plagues saliva-based testing.
"The snottier samples were performing poorly, but by adding proteinase K we were able to get almost equivalent performance to our extractor," she said.
Illustrative of the supply crisis that continues to plague labs, in the middle of evaluating their protocol the UW team had to switch from using a Roche Magna Pure 96 to a Thermo Scientific KingFisher Flex as the comparator method due to commercial reprioritization.
"This kind of stuff is happening to everybody, so that is one of the reasons we wanted to get this out there — so that people can use this to simplify their workflows, as well," Starita said.
Lockwood also said that the supply chain disruptions are continuing, can be unexpected, and have led to substitutions of reagents and consumables from manufacturers "that may or may not work with some kinds of automation," for example.
"The modular nature of our method, and use of lower cost reagents, make it appealing to labs like ours using LDTs and looking to have high-throughput, low-cost testing," she said.
The LDT journey
The history of the SwabExpress protocol has been surprisingly rocky, which Starita and Lockwood attribute in part to the flux in EUA and LDT regulations over the past year and a half.
An at-home SARS-CoV-2 sampling study that grew out of a UW-supported influenza home-collection project — called Home Collect — was halted by the FDA last year, in a move that was widely criticized by experts.
The Greater Seattle Coronavirus Assessment Network, or SCAN, program was later relaunched and began mailing sample self-collection kits to local residents. In a New England Journal of Medicine study, preliminary data showed SCAN surveillance could help with early detection of community spread.
And, a study by a different team at UW showed last year that at-home self-collected nasal swabs performed comparably to clinician-collected nasopharyngeal swabs.
The lab's dry swab work was initially described online in April of last year, as previously reported, in a preliminary study of 11 samples.
The team has worked with the FDA as regulations evolved, but "because our kits are couriered to someone's home, that home delivery model is a little more challenging," Lockwood said.
And, in Washington, prescriptions are not required for a patient to order their own lab test, and the team chose not to require one for SARS-CoV-2 testing. "At the time, that did not fit under the rubric of what FDA was authorizing," Lockwood also noted.
The US Department of Health and Human Services determined in August last year that the FDA will not require premarket review of LDTs without notice and comment rule-making. Although this stance seems to be under consideration, for now LDTs are regulated by CLIA.
The state of Washington is a CLIA-exempt state, however, meaning the CLIA program is state-administered, so the lab has worked closely with the Washington State Department of Health for its test.
In the end, the team also had to reinitiate its IRB protocol in order to make the protocol acceptable to FDA, Lockwood said. So, the clinical testing done by the lab now is performed under the auspices of a research study.
The team also performs SARS-CoV-2 variant sequencing in its lab for the state of Washington and has seen a similar rate of increases in COVID-19 cases due to the delta variant that is being seen nationally.
Early on in the pandemic, SwabExpress could have really put a dent in the testing problem, Starita said. "I'm hopeful that if we have a next go-around with delta, we can do things a little bit smarter, and without as much red tape," she said.
Lockwood concurred. "I think what is important is to be able to meet the testing demand needs, and in the event that there is evolution of SARS-CoV-2 or another respiratory pathogen that comes up, we want to be able to do that, and we should have a framework nationally that allows us to do that in a facile way," she said.
The team is also hopeful that the method could help labs in other hard-hit countries as well, particularly ones that require low cost and high throughput, as the workflow reduces the cost of consumables and reagents by approximately 90 percent.
"I'm really hopeful that low- and middle-income countries can pick this up," Starita said. While extraction can double the cost of testing, for the SwabExpress protocol, "the swab costs 20 cents, the extraction buffer is practically free, so it is very cheap," she said.