Skip to main content
Premium Trial:

Request an Annual Quote

ESCMID Presenters Make Case for Saliva Samples in Molecular Infectious Disease Testing


BARCELONA, Spain – Saliva is a promising sample type for molecular diagnostic testing in infectious diseases beyond SARS-CoV-2, such as monkeypox and influenza, according to research presented at the European Society of Clinical Microbiology and Infectious Diseases global congress (ESCMID Global), held here this week.

In addition, during a meet-the-expert session, researchers argued for the increased exploration of and use of saliva in infectious disease diagnostics, noting that its potential as a sample type for large-scale screening and resource-limited areas of the world outweigh its perceived drawbacks — such as lower sensitivity and sample handling and transport issues — compared to more established methods.

In an ESCMID Global poster presentation, investigators led by Anne Wyllie, a research scientist in epidemiology and affiliated faculty member at the Yale Institute for Global Health, presented the results of testing 30 saliva samples from mpox-positive individuals — work that was also published last year in MedRxiv.

Wyllie is known for her previous contributions to the development of the SalivaDirect COVID-19 test, one of the earliest saliva-based COVID-19 assays to receive Emergency Use Authorization (EUA) from the US Food and Drug Administration.

In their study, Wyllie and colleagues collected samples in Spectrum SDNA-100 collection devices from Flow Health and tested them with five different PCR assays. They also examined the stability of mpox detection from samples undergoing various simulated shipping and storage conditions.

As they explained in their poster, current clinical diagnosis of mpox uses lesion swabs, but many patients experience symptoms prior to lesion onset. As a source of transmission recognized by the US Centers for Disease Control and Prevention, saliva could be useful for early infection detection to enable rapid post-exposure vaccination and isolation.

They found that the Logix Smart Mpox (2-Gene) RUO assay from US company Co-Diagnostics, which targets a viral F8L gene, performed better at overall detection, with consistent mean cycle threshold (Ct) values across all assay types.

In addition, mpox virus detection in saliva was comparable across three different extraction-free workflows, and DNA remained relatively stable in raw saliva for 24 to 48 hours under multiple conditions, with notable degradation at 72 hours.

They concluded that their results support the use of saliva samples to reliably diagnose mpox samples, reduce testing costs, and enable easier and cheaper self-collection and testing in low- and middle-income countries.

Meanwhile, researchers from the University of Nebraska Medical Center (UNMC) presented the results of a study examining the utility of self-collected saliva samples and extraction-free PCR for influenza surveillance, building on a previous study published in 2021 in JAMA Network Open by some of the same investigators that assessed SARS-CoV-2 screening and environmental monitoring in a public school district.

They collected saliva as passive drool from 73 patients being evaluated for respiratory illness in the emergency department to compare influenza detection by extraction-free RT-qPCR with clinical respiratory pathogen testing using nasopharyngeal (NP) swabs tested with commercial multiplex molecular panels. They also performed extraction-free testing on residual, self-collected saliva specimens from SARS-CoV-2 community-based surveillance programs targeting asymptomatic individuals in aggregate settings from November 2022 to March 2023.

The UNMC investigators found that influenza detection over time using extraction-free RT-qPCR in the drool samples showed 100 percent agreement with NP reference tests, with five positive and 68 negative patients. In addition, the researchers found that monthly positivity rates in the community-based surveillance programs ranged from zero to almost 3 percent.

Changing perceptions

Saliva made inroads as a sample type for infectious disease MDx testing during the COVID-19 pandemic but has not been elevated to the gold standard of PCR testing on NP swabs.

In the meet-the-expert session, Wyllie was joined by Rachel Louise Byrne, a postdoctoral researcher at the Liverpool School of Tropical Medicine who has also extensively researched saliva as an alternative sample type to NP swabs for SARS-CoV-2 detection.

Wyllie and Byrne noted the common perceived concerns from various stakeholders over saliva samples. Clinicians often cite inferior performance, limited clinical value, insurance reimbursement issues, and the difficulties that come with changing physician routine. Meanwhile, clinical labs frequently bemoan transport issues, difficult sample handling, low sensitivity and specificity, and limited market value.

Despite this, since 2021, the saliva-based diagnostics market has grown more quickly for infectious disease applications than a number of other medical testing categories such as endocrinology, cardiology, autoimmune testing, and oncology and is projected to maintain an almost 8 percent compound annual growth rate through 2026, Wyllie noted, citing data provided by Maikel Boot, a technical director specializing in microbiology at research consultancy PreScouter.

"It's a good time to get into saliva," Wyllie said.

In addition, Wyllie and Byrne argued, the potential upside of spit testing is worth debunking some of the preconceived notions around it. It's ideal for large-scale and repeat testing, and healthcare workers are not required for collection, thus reducing transmission risk. Furthermore, despite the relatively noninvasive nature of NP swabs, drooling into a tube is obviously a more comfortable collection procedure.

Additionally, a growing body of research has challenged many of the negative perceptions around spit testing. Wyllie was corresponding author on a seminal April 2020 paper in MedRxiv demonstrating that saliva is more sensitive for SARS-CoV-2 detection than NP swabs, work that was summarized in correspondence to the New England Journal of Medicine later that year.

Meanwhile, Byrne was co-corresponding author on a January 2023 PLOS One paper demonstrating how saliva sampling could be successfully leveraged for routine screening of healthcare workers. She also was first author on a research letter published in November 2020 in Emerging Infectious Diseases showing that SARS-CoV-2 viral loads in salvia samples were similar to those found in nasal and throat swab samples in 110 patients.

Many other peer-reviewed publications exist demonstrating saliva's potential for MDx testing of SARS-CoV-2 and even other respiratory viruses, including respiratory syncytial virus, human parainfluenza virus, human metapneumovirus, human rhinoviruses, influenza, and adenovirus.

EU regulators have followed the literature, with CE-IVD marks applied to saliva-based COVID-19 diagnostic tests from companies like Roche, Becton Dickinson/CerTest Biotec, and Canary Molecular.

However, the FDA remains somewhat unconvinced, having awarded EUAs to several saliva-based diagnostics, but so far holding back from granting full clearance to any spit-based tests — although it's unclear how many submitted applications there have been.

Wyllie and Byrne noted at ESCMID that there are difficulties in performing saliva-based molecular tests when the healthcare community is already so used to NP swabs. For instance, Wyllie noted, tests using saliva must be designed specifically for saliva — in other words, using the same protocols used on NP swabs may compromise saliva-based test performance.

Furthermore, it may be more difficult to consistently collect quality saliva samples, which are crucial for test performance. NP swabs are generally consistent if the proper technique is used, but saliva samples can vary greatly depending on how patients are instructed to provide the sample — passive drool is preferred over "hocking a loogie."

"A good saliva sample is critical for a good test," Byrne said. "Samples should be clear to white in color. Sometimes it will contain sputum, for instance, which can happen when clear instructions aren't given. With inpatients, it can be difficult to collect samples because they might be dry and dehydrated. But for outpatients, it can be a very promising sample type."

This is something that Wyllie and colleagues learned the hard way when collecting samples from the National Basketball Association as part of a 2020 partnership with Yale's SalivaDirect team to evaluate the use of the test as a COVID-19 surveillance tool.

"NBA samples were every color of the rainbow," she said. Surprisingly, they found that recently consumed food and drink did not affect results very much. Mucins and white blood cells — i.e., mucus — and, strangely, a very high concentration of Colgate toothpaste were the only things that consistently confounded results.

Saliva testing may have public sentiment in its corner. Consumers are already used to so-called "spit kits" as have been provided for years by consumer genomics companies such as 23andMe, and in December, the SalivaDirect team published the results of a survey in MedRxiv demonstrating that, if they could be assured that all sample types performed equally for SARS-CoV-2 testing, consumers overwhelmingly would prefer spit as a sample type.

"But we do understand that not everyone is going to come around to it — it's less dignified drooling into a tube, and there are medications that cause dry mouth," particularly in elderly populations, Wyllie admitted. "We need to look at scenarios where it can facilitate fast and easy testing. At least we need to validate it as another tool when NP swabs are problematic," or, for instance, when patients don't want to do them, as has been seen with some parents regarding their children for RSV testing, she added.

Given its promise as a sample type in resource-limited settings, "I can't help but think about tuberculosis," Byrne said, citing a September 2023 review in Lancet Microbe on the accuracy of multiple alternative sample types to sputum, including saliva, for culture- and nucleic acid amplification-based testing. Although data on saliva samples is limited, oral swabs have shown promise as an alternative sample type, and "essentially you are collecting saliva if you are doing an oral swab," Byrne said. "Hopefully, over the next few years we will see more data."