Saliva may offer a surprisingly useful window into the stomach, according to a new systematic review and meta-analysis published in Cellular Oncology. Researchers led by Maryam Koopaie, Reihaneh Khalilianfard, Mahnaz Fatahzadeh and Sajad Kolahdooz pooled studies of salivary nucleic acid biomarkers and found that these molecules can distinguish patients with histologically confirmed gastric cancer from controls with modest to promising accuracy, depending on how the biomarkers are combined.
The team followed PRISMA-DTA guidelines and registered the review in PROSPERO (CRD420261395418). They searched MEDLINE, Embase, Web of Science, Scopus, the Cochrane Library and Google Scholar for studies that evaluated salivary nucleic acid biomarkers against histologically confirmed gastric cancer. Methodological quality was assessed with QUADAS-2, and sensitivity, specificity, diagnostic odds ratio and the area under the summary ROC curve were pooled using bivariate random-effects models.
Eleven records were included in the systematic review, but only six were eligible for the meta-analysis, a small evidence base that the authors acknowledge directly. The pooled results were stratified by diagnostic model: single biomarkers, multiple-biomarker panels, and integrated biomarker models that also incorporated demographic or other biological variables.
Single salivary biomarkers showed modest diagnostic performance, with pooled sensitivity of 0.70, specificity of 0.75, a diagnostic odds ratio of 7.33 and an AUC of 0.79. In other words, a lone salivary nucleic acid marker missed roughly three in ten cancers and misclassified about a quarter of people without the disease, which is far from sufficient for clinical use on its own.
Combining several biomarkers into a panel improved accuracy. Multiple-biomarker panels reached a pooled sensitivity of 0.74, specificity of 0.81, diagnostic odds ratio of 8.83 and AUC of 0.81. The diagnostic odds ratio, which summarizes overall discrimination, rose from 7.33 to 8.83, indicating that panels outperform single markers, though the improvement is incremental rather than transformative.
The strongest results came from integrated models evaluated in validation phases, which combined salivary biomarkers with demographic or other biological variables. These models achieved a sensitivity of 0.81, specificity of 0.81 and a diagnostic odds ratio of 18.20, more than double that of the biomarker panels. However, the authors note that between-study heterogeneity was high for sensitivity in these analyses, meaning the estimates varied considerably across the included studies.
The motivation for the work is clinical: gastric cancer remains a leading cause of cancer-related mortality worldwide, and endoscopy, while accurate, is invasive and impractical for population-based screening. A non-invasive test based on a saliva sample could, in principle, make large-scale screening more feasible, which is why salivary nucleic acid biomarkers have attracted growing research interest.
Despite the encouraging pooled estimates, the authors are explicit about the limitations of the current evidence. The included studies relied on case-control designs, showed methodological heterogeneity, were subject to spectrum bias, and carried a high risk of bias in patient selection.
The review concludes that salivary nucleic acid biomarkers, particularly within validation-phase integrated models, represent a promising non-invasive approach for gastric cancer detection, but that standardized collection protocols and large-scale multicenter prospective validation are essential before any clinical implementation. Until such validation is completed, the pooled figures should be read as an early signal of potential rather than a ready-to-deploy diagnostic test.
It is also worth noting that the article is being shared early as an accepted, peer-reviewed manuscript that is citable and carries a permanent DOI, but is subject to further edits before the final Version of Record appears. Readers should therefore treat the reported numbers as current best estimates from a small, heterogeneous evidence base, and expect the final published version to be the definitive record of the analysis.
Subject of Research: Diagnostic accuracy of salivary nucleic acid biomarkers for gastric cancer detection
Article Title: Diagnostic accuracy of salivary nucleic acid biomarkers in gastric cancer detection: a systematic review and meta-analysis
Article References: Koopaie, M., Khalilianfard, R., Fatahzadeh, M., & Kolahdooz, S. (2026). Diagnostic accuracy of salivary nucleic acid biomarkers in gastric cancer detection: a systematic review and meta-analysis. Cellular Oncology. https://doi.org/10.1007/s13402-026-01311-5
Image Credits: AI Generated
DOI: 10.1007/s13402-026-01311-5
Keywords: gastric cancer, salivary biomarkers, nucleic acids, systematic review, meta-analysis, diagnostic accuracy, non-invasive screening, PRISMA-DTA, QUADAS-2, Cellular Oncology, Diagnostic, accuracy
News Source: Nathaniel Bowman. (October 28, 2026). Saliva biomarkers show promise for detecting gastric cancer, review finds. Scienmag.
