Scientists have discovered distinct bacterial patterns in the saliva of patients with a deadly form of esophageal cancer.
Bacteria living in the mouth may hold clues to a deadly form of esophageal cancer that often goes undetected until treatment options are limited. Researchers in South Africa have identified distinctive bacterial patterns in patients’ saliva, raising the possibility that a simple, inexpensive test could eventually help identify people who need further screening.
The study, published in Communications Medicine, was conducted by scientists at the Sydney Brenner Institute for Molecular Bioscience (SBIMB) at Wits University and Columbia University in New York. Combining bacterial DNA sequencing with machine learning, the team developed a model that distinguished cancer patients from unaffected individuals more effectively than one based on clinical and demographic information alone.
A Deadly Cancer With Mysterious Geographic Patterns
Esophageal squamous cell carcinoma (ESCC) is especially common in parts of eastern and southern Africa, including South Africa’s Eastern Cape and KwaZulu-Natal provinces. High rates also occur in regions of China and Iran, forming a geographic belt that scientists have struggled to explain.
Unlike esophageal adenocarcinoma, which is strongly associated with acid reflux, obesity, and lifestyle factors, ESCC has less clearly understood causes. It also frequently affects relatively young adults.
“A feature of the cancer is the young age of many patients, with a mean age of 50, and 18% of patients presenting before the age of 40,” says Distinguished Professor Christopher Mathew of the SBIMB.
“This geographical occurrence is also puzzling, and may reflect genetic susceptibility, a specific environmental exposure, or a combination of factors that researchers have still not been able to pin down,” explains Dr. Wenlong Carl Chen, a researcher at SBIMB.
Known Risk Factors Cannot Explain Every Case
Analysis of data from the Johannesburg Cancer Study identified several ESCC risk factors, including smoking, heavy alcohol consumption, rural residence, lower educational attainment, and the use of biomass or other household fuels.
“Still, they do not fully explain the number of cases, the geographic clustering, or why some patients develop the disease without these obvious exposures.”
To investigate another possible indicator, researchers examined the oral microbiome, the community of microorganisms living in the mouth. DNA sequencing revealed distinct differences between saliva samples from cancer patients and unaffected controls.
“We found clear differences between the saliva bacteria of people with oesophageal squamous cell carcinoma and unaffected controls. Using genetic sequencing of bacteria and machine-learning methods, the team identified a distinctive microbial pattern associated with the cancer,” says Chen.
Saliva Bacteria May Signal Changes in the Esophagus
Several bacterial species were more abundant in ESCC patients, including Fusobacterium nucleatum, a microorganism previously implicated in other cancers. The microbiome-based model also performed better than one relying solely on clinical and demographic data.
However, the findings do not establish that these bacteria cause cancer. The microbial changes could instead result from the disease itself, particularly as a growing tumor narrows the esophagus and disrupts normal swallowing.
“But even that could be useful. A blockage in the esophagus develops gradually, often before patients arrive at the hospital, unable to swallow properly. If microbial changes in saliva can help flag that something is already changing, they could one day help identify people who need earlier referral for endoscopy.”
A saliva or cheek swab test could eventually help doctors prioritize patients for endoscopy, particularly in communities with limited access to specialized care. Such a test would not replace endoscopy, however, and the current approach has not been established as an early cancer detection method.
Searching for Genetic and Environmental Clues
Researchers are also investigating whether inherited genetic differences increase susceptibility to ESCC and whether tumors contain evidence of earlier environmental exposures.
One approach involves studying mutational signatures, distinctive patterns of DNA damage preserved in cancer cells. Certain chemicals, pollutants, smoke exposures, or water contaminants may leave recognizable genetic traces, potentially helping scientists identify contributors to the disease.
“This is critical research which has generated important foundational knowledge that will take us to the next step, and hopefully to the clinic in the long run,” says Professor Michèle Ramsay, Director of the SBIMB.
What Researchers Need to Confirm
The saliva model has undergone internal validation but still requires testing in independent populations. Oral bacterial communities vary with geography, diet, and environmental conditions, meaning results from one population may not translate directly to another.
The next research phase will include patients with confirmed ESCC, healthy individuals, and people with benign esophageal conditions that can also cause swallowing difficulties.
These comparisons will help establish whether the bacterial patterns are specific to cancer or reflect esophageal obstruction more generally. The answer will determine whether saliva testing could eventually become a practical tool for identifying patients who need further investigation.
Reference: “A generalizable cross-continent prediction of esophageal squamous cell carcinoma using the oral microbiome” by Shahd ElNaggar, Wenlong Carl Chen, Leanne M. Prodehl, Thomas K. Marumo, Muhammed U. Khan, Christopher G. Mathew, Paul Ruff, Zhezhen Jin, Alfred I. Neugut, Anil K. Rustgi, Anne-Catrin Uhlemann, Tal Korem and Julian A. Abrams, 28 February 2026, Communications Medicine.
DOI: 10.1038/s43856-026-01468-y
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.