Overview

Could your spit save your life? Researchers think it might. Saliva is full of clues about what’s happening inside the body, and new tests are being developed to see whether it can reveal many types of cancer long before symptoms appear. A large review of 29 studies, covering more than 11,000 people, found that saliva tests were accurate 85% of the time. They correctly picked up cancer in 76% of cases and gave the right “all clear” to those without cancer.¹ 

Together, these findings suggest that saliva tests could one day provide a simple, needle-free method for detecting cancer earlier. So how do these tests work, what can they actually detect, and how close are we to seeing them in everyday healthcare? This article takes a closer look. 

What are saliva tests?

Saliva might seem like an unlikely place to look for signs of disease, but it contains hundreds of biological molecules that mirror what’s happening throughout the body. These include fragments of DNA and RNA, proteins, hormones and even chemical by-products from tumours. By analysing these biomarkers, scientists can look for patterns linked with specific diseases – much like a blood test does, but without needles or lab visits.^2,3

In practical terms, saliva tests are simple: a person either spits into a clean tube or uses a soft swab inside the mouth, then seals the sample for analysis in a lab or with portable devices. Saliva collection is painless, low-risk, and easy to repeat. Because saliva collection is quick, painless and non-invasive, it can be done almost anywhere – at a GP surgery, dental practice or even at home. Samples can then be sent to a laboratory or analysed with point-of-care devices under development. This ease of collection makes saliva an attractive option for screening large numbers of people or for repeat testing over time.^2,3

Increasingly, researchers are also using artificial intelligence and machine-learning tools to sort through the thousands of molecules found in saliva and identify the most reliable patterns. This ease of collection – combined with powerful analysis methods – makes saliva an attractive option for screening large numbers of people or for repeat testing over time.⁴

Can saliva tests detect cancer?

Researchers are exploring saliva testing for a wide range of cancers, not just those in the mouth. Studies have looked at biomarkers linked to cancers of the pancreas, stomach, colon, breast, lung and even the brain.^1,2 

The idea is similar to a “liquid biopsy” – instead of drawing blood, saliva is collected to see if tumour-related molecules have entered the mouth through the bloodstream or local tissues. Most of these tests are still in the early stages, but the research so far suggests that saliva could become a gentle, non-invasive way to help detect cancer and work alongside existing blood tests or biopsies in the future.^1,2,3  Because saliva contains thousands of different molecules, researchers are increasingly using machine-learning tools to sift through the data and identify the most reliable patterns, which may help turn these early-stage discoveries into simple, reliable tests for everyday care.⁴

One example already being tested is lung cancer. Researchers have investigated whether patterns of tiny chemical compounds in saliva could help identify the disease without blood tests or scans. They found clear differences between the saliva of people with lung cancer and those without. In testing, their approach achieved a performance score known as “AUC” of 0.79 – a level scientists consider promising, but still at an early research stage.⁵

Why saliva testing looks promising – and what’s holding it back

 Potential benefits

• Quick and needle-free: just spit or swab – no needles, no special equipment
• Comfort and convenience: it can be done in GP or dental settings, or even at home
• Cost-effective and scalable: easier to collect and transport samples from large numbers of people
• Inclusive screening: it could reach people who avoid or struggle with blood tests, reducing health inequalities
• Repeat testing: simple enough for regular monitoring or follow-up without added stress^1,2,3 

Current limitations

• Still experimental: most saliva tests for cancer are in research or early trials, not routine care
• Variable accuracy: Sensitivity and specificity differ between studies and cancer types
• Standardisation needed: how saliva is collected, stored and analysed can affect results
• Influenced by everyday factors: food, drink, oral hygiene, medications and smoking can all change saliva composition and potentially affect test accuracy
• Requires confirmation: positive results would still need follow-up with established diagnostic tests^1,2,3

What’s next for saliva testing?

The next few years will be important for finding out just how well saliva tests really work in everyday life. Researchers are now running much bigger studies to check which saliva markers are most reliable for spotting different types of cancer, and to make sure everyone is collecting and testing samples in the same way. They’re also comparing saliva results with blood and tissue samples results, to see how closely they match. Before these saliva tests can become part of routine care, they’ll need to prove they’re safe, accurate, and precise. It’s a long process, but growing interest in simple, non-invasive tests is helping to speed things up.^1,2,3

If the science holds up, the next big question will be how to fit saliva testing into everyday care. Because collecting a sample is so easy, it could make early cancer screening much more accessible, especially for people who avoid needles or hospital tests. Looking further ahead, researchers are experimenting with AI tools that can scan thousands of molecules in saliva and spot early warning patterns across several cancers at once. In the future, a quick saliva test could act as an early “health check” that helps flag potential problems long before symptoms appear.⁶

Challenges in translating from lab to clinic

While these studies are exciting, it’s important to acknowledge the bridge that still needs to be crossed from research to routine healthcare.^1,2,3

• Checking the results: Many of these promising saliva tests have only been tried on small groups of people so far. They need to be tested again on much larger, more diverse groups to make sure the findings are real
• Different methods give different results: Studies don’t all collect or handle saliva the same way, so results can vary from one test to another
• Everyday things can interfere: What you eat or drink, your oral hygiene, medicines or smoking can all change what’s in your saliva
• Needs follow-up: Even if a saliva test gives a positive result, you would still need standard scans or biopsies to confirm

Summary 

Looking ahead, the most likely path is that saliva tests will first be used to supplement existing screening, rather than replace it. For example, someone at moderate risk of cancer might use a saliva test as a first-line screen; if it flags a signal, they’d then go on to standard imaging or biopsy. Over time, as biomarker panels get better and costs fall, saliva testing could become part of routine check-ups – especially in dental or primary care settings. But getting there will require much more validation, regulatory approval, and careful design of real-world clinical trials. Machine-learning tools are likely to play a key part in analysing the data collected from saliva samples and turning it into simple, reliable tests that can be used in everyday care.

References

1. Rapado-González Ó, Martínez-Reglero C, Salgado-Barreira Á, Takkouche B, López-López R, Suárez-Cunqueiro MM, et al. Salivary biomarkers for cancer diagnosis: a meta-analysis. Annals of Medicine [Internet]. 2020 [cited 2026 Jan 12]; 52(3–4):131–44. Available from: https://www.tandfonline.com/doi/full/10.1080/07853890.2020.1730431.
2. Ghosh AK, Nath A, Elangovan E, Banerjee A, Ramalingam K, Sethuraman S, et al. Exploring Salivary Biomarkers for Tumor Diagnosis: A Narrative Review. Cureus [Internet]. 2024 [cited 2026 Jan 12]; 16(7). Available from: https://cureus.com/articles/273929-exploring-salivary-biomarkers-for-tumor-diagnosis-a-narrative-review.
3. Liu J, Huang D, Cai Y, Cao Z, Liu Z, Zhang S, et al. Saliva diagnostics: emerging techniques and biomarkers for salivaomics in cancer detection. Expert Review of Molecular Diagnostics [Internet]. 2022 [cited 2026 Jan 12]; 22(12):1077–97. Available from: https://www.tandfonline.com/doi/full/10.1080/14737159.2022.2167556.
4. Kuwabara H, Katsumata K, Iwabuchi A, Udo R, Tago T, Kasahara K, et al. Salivary metabolomics with machine learning for colorectal cancer detection. Cancer Science [Internet]. 2022 [cited 2026 Jan 12]; 113(9):3234–43. Available from: https://onlinelibrary.wiley.com/doi/10.1111/cas.15472.
5. Kajiwara N, Kakihana M, Maeda J, Kaneko M, Ota S, Enomoto A, et al. Salivary metabolomic biomarkers for non‐invasive lung cancer detection. Cancer Science [Internet]. 2024 [cited 2026 Jan 12]; 115(5):1695–705. Available from: https://onlinelibrary.wiley.com/doi/10.1111/cas.16112.
6. Hunter B, Hindocha S, Lee RW. The Role of Artificial Intelligence in Early Cancer Diagnosis. Cancers [Internet]. 2022 [cited 2026 Jan 12]; 14(6):1524. Available from: https://www.mdpi.com/2072-6694/14/6/1524.