August 17, 2022 | A Yale-based diagnostics initiative is building a nationwide network of clinical laboratories embracing an open-source testing protocol for COVID-19 using saliva. Ultimately, the labs could become important players in public health responses to pandemics of the future and, perhaps more immediately, the ongoing monkeypox outbreak, according to Anne Wyllie, Ph.D., a research scientist in epidemiology in the Yale School of Public Health as well as principal investigator for the SalivaDirect Initiative.
To date, 186 labs certified per the Clinical Laboratory Improvement Amendments of 1988 (CLIA) have been designated to use the SalivaDirect COVID-19 PCR test, she reports. More than seven million tests have now been performed.
Although saliva is widely and successfully used as a sample type around the world, saliva-based tests have gotten a bad rap in the U.S. due to a combination of inexperience, misinformation, and the fact that collection and processing methods are largely unstandardized, says Wyllie. In a recent survey, labs doing SalivaDirect and nasopharyngeal swab tests for COVID-19 side by side reported that nasal swabbing had far more invalid or rejected test samples.
A review of 58 studies conducted in 2020 found that the majority concluded saliva was as good, if not better, than nasopharyngeal swabs for SARS-CoV-2 detection and detected an additional 10% of positive cases missed by swab testing (The Lancet Respiratory Medicine, DOI: 10.1016/S2213-2600(21)00178-8). And that was despite methodological discrepancies that skewed the sensitivities reported across studies.
The key point is that “differences in collection and testing protocols matter,” Wyllie says. “Saliva is not a traditional diagnostic sample type [different matrix, consistency, and variability] so you can’t take a method that works for swabs and simply apply to saliva and expect it to work. You wouldn’t do that for blood, and you wouldn’t do that for urine.”
SalivaDirect provides a robust method that is also less costly and time-consuming and more patient-friendly than tests requiring a nasal swab, she continues. Based on the protocol’s proven sensitivity and reliability, the U.S. Food and Drug Administration (FDA) granted SalivaDirect an Emergency Use Authorization (EUA) in August 2020.
“We are eventually going to go for full approval,” says Wyllie. “We have a wealth of real-world evidence and data from different bridging studies we have done... [but] it is not clear how much more work is needed rather than collating the data [in hand].”
Eventually, a nonprofit organization will form outside of Yale and make a 510(k) premarket submission to the FDA, she adds. The assay has already been submitted for an EU mark to regulatory authorities in Europe where approvals are easier to come by.
The SalivaDirect testing protocol is unique and, for the moment anyway, specific to COVID-19. To be clear, it is not a kit, and no company is behind it, says Wyllie. Materials from multiple manufacturers can be used to allow for substitutions and guard against delays related to supply chain shortages.
Importantly, SalivaDirect eliminates the RNA extraction step from the PCR process and accommodates more than a dozen reagents and 26 PCR machines, she continues. One lab technician can manually run 93 samples in three to four hours. Unlike most COVID tests, SalivaDirect is also validated to detect asymptomatic as well as symptomatic infections.
The designation process for using the SalivaDirect protocol is simple and can take as little as two business days, says Wyllie, noting that not all 186 labs in the network have started using the laboratory developed test (LDT) yet. She and her team have handled the upfront validation work, including providing data to the FDA that SalivaDirect performs well for SARS-CoV-2 and can distinguish between individuals who are and are not infected.
Labs designated to use the test have submitted some positive samples, controls, and results to the Yale team to ensure they know how to properly run the test and “to guarantee the best results for the local community,” she says. The labs have additionally signed an agreement that they will abide by EUA regulations, including reporting requirements.
Various schools, universities, workplaces, and sports leagues not under the EUA have adopted the SalivaDirect protocol and are using one of the in-network CLIA labs to handle the processing. “Anyone anywhere can collect their own sample,” she says, no supervision required.
SalivaDirect is essentially a crowdsourced protocol with more than a dozen partners and collaborators. The first major supporter was the National Basketball Association (NBA), which in May 2020 funded the initiative with a $500,000 grant. At the time, Wyllie says, “we couldn’t even get money from NIH [National Institutes of Health]. We couldn’t get money [for COVID-related projects] from basically anywhere [else].”
Early on, efforts to grow the SalivaDirect Initiative received a financial boost from Fast Grants then being offered by charitable individuals and causes to further COVID-19 science. Tempus Labs provided a grant to develop an at-home collection kit, which required multiple usability studies, says Wyllie.
Flambeau, whose RapidX mobile diagnostic testing system was developed with NIH Rapid Acceleration of Diagnostics (RADx) funds, has helped support SalivaDirect’s staffing and ongoing validation work, she says. Lindus Health put out a pro bono recruitment ad for usability studies.
SalivaDirect has frequently cross-collaborated with Mirimus Labs, best known for its pioneering work doing saliva pooling at schools in the northeast, to obtain needed patient samples, Wyllie adds. In terms of monetary partners, the Rockefeller Foundation has supported salaries for some team members and the development of key public health communication materials.
Other collaborators stepped in to provide something for the EUA for the COVID-19 test that the SalivaDirect Initiative hadn’t the resources or means to independently accomplish. “It was very much a team effort to get this through,” she says.
Several labs join the SalivaDirect network each month and usage of the protocol is being driven largely by its ease of implementation, Wyllie says. Labs that pivoted to COVID and are now returning to their normal clinical diagnostic routine view SalivaDirect as “a very easy, streamlined test to have in the background” to handle ongoing outbreaks (currently from omicron BA.5 variant), Wyllie says.
They can also charge for the test under the licensing agreement with Yale. “We don’t take any fees,” she notes.
A peer-to-peer support system emerged on its own during the onboarding process, which originally involved calls twice a week but has since scaled back to once a month. “Some of our labs are telling us they’ve never had anything like this before,” says Wyllie, noting how they have been sharing their experiences with proficiency testing as well as cross-validating samples.
The test itself uses readily available materials and very few reagents and takes much less time than one requiring full RNA extraction, so staff and instrument time are preserved for other testing, continues Wyllie. Among the reasons labs might choose to not use SalivaDirect, beyond unsubstantiated beliefs that saliva is difficult to work with, is that they may not have the capacity to validate new methods.
Wyllie, who has been working with saliva as a sample type for nearly a decade now, cites clear collection instructions as one of the keys to success. When the SalivaDirect protocol was initially being developed in partnership with the NBA, one person at Drug Free Sport International was responsible for distributing collection instructions to players and staff.
“We didn’t give any formal training in collection to anyone,” she says. “We received 3,800 samples and had only 12 invalid results, showing we could release the protocol... into the wild.”
The plan is to expand the SalivaDirect protocol to other viruses, says Wyllie, starting with monkeypox. The virus can be transmitted, and is thus detectable, through saliva. Flow Health in California has already deployed a saliva-based LDT for the viral zoonotic infection.
Currently, the FDA is advising that swab samples be taken directly from a lesion when testing for the monkeypox virus. But by then people may have been infectious for two to four weeks, thwarting containment efforts, she says.
Given the possibility of oral lesions with the virus, many labs are now exploring saliva-based testing, continues Wyllie. “We would love to be able to [do this]... if we could get the samples that we need to validate the SalivaDirect protocol for monkeypox.”
Labs in the SalivaDirect network could potentially join forces to quickly scale much-needed monkeypox testing to communities across the country, she says. Collaboration would also enable demonstration of how early the virus is detectable in exposed individuals who are “feeling a bit off” (e.g., fevers and muscle aches) post-contact but do not yet have the telltale lesions.
Now that monkeypox has been declared a Public Health Emergency by the Biden Administration, the floodgates have theoretically opened for EUA of monkeypox diagnostic tests as well as vaccines. The agency will hopefully be prioritizing its response such that collaborative, higher-impact tests such as SalivaDirect get reviewed ahead of proprietary tests of one or two big labs, says Wyllie.
Meanwhile, the SalivaDirect Initiative will continue monitoring the emergence of new COVID-19 variants and the ability of the single-target assay to detect their presence, she adds. The team is also working to expand the test to other respiratory pathogens, including influenza and respiratory syncytial virus, and health conditions such as concussions and diabetes.
“We are incredibly small and under-resourced so... we are doing what we can,” Wyllie adds. “I do not have anyone in my lab who is full-time SalivaDirect.” Influenza samples have also been nearly impossible to come by.
SalivaDirect just held its first annual conference, which attracted about 115 laboratories, diagnostic developers, and academics to talk about successes, partnerships, and advances made through their collective pandemic response, Wyllie says. “The idea was not to be a SalivaDirect horse and pony show.”
The attendees included representatives from University of Illinois Urbana-Champaign, which has a similar saliva-based, RNA-extraction-free test (covidSHIELD), although it is not open source. Clemson University was represented by Delphine Dean and some of her laboratory staff who have made low-cost, saliva-based testing broadly available across campus.
Wyllie says she was particularly encouraged by the camaraderie and banter that organically arose between people at the conference, which lasted under two days. “Everyone wished it could have been longer.”
One of the major goals of the conference was strengthening connections between labs composing the fledgling SalivaDirect network and to set a common path forward, says Wyllie. Being a network member signals interest in expanding access to saliva-based testing—and not only through use of the SalivaDirect protocol.