Identification of salivary metabolomic biomarkers for oral cancer screening

Shigeo Ishikawa, Masahiro Sugimoto, Kenichiro Kitabatake, Ayako Sugano, Marina Nakamura, Miku Kaneko, Sana Ota, Kana Hiwatari, Ayame Enomoto, Tomoyoshi Soga, Masaru Tomita, Mitsuyoshi Iino

Research output: Contribution to journalArticlepeer-review

141 Citations (Scopus)


The objective of this study was to explore salivary metabolite biomarkers by profiling both saliva and tumor tissue samples for oral cancer screening. Paired tumor and control tissues were obtained from oral cancer patients and whole unstimulated saliva samples were collected from patients and healthy controls. The comprehensive metabolomic analysis for profiling hydrophilic metabolites was conducted using capillary electrophoresis time-of-flight mass spectrometry. In total, 85 and 45 metabolites showed significant differences between tumor and matched control samples, and between salivary samples from oral cancer and controls, respectively (P < 0.05 correlated by false discovery rate); 17 metabolites showed consistent differences in both saliva and tissue-based comparisons. Of these, a combination of only two biomarkers yielded a high area under receiver operating characteristic curves (0.827; 95% confidence interval, 0.726-0.928, P < 0.0001) for discriminating oral cancers from controls. Various validation tests confirmed its high generalization ability. The demonstrated approach, integrating both saliva and tumor tissue metabolomics, helps eliminate pseudo-molecules that are coincidentally different between oral cancers and controls. These combined salivary metabolites could be the basis of a clinically feasible method of non-invasive oral cancer screening.

Original languageEnglish
Article number31520
JournalScientific reports
Publication statusPublished - 2016 Aug 19

ASJC Scopus subject areas

  • General


Dive into the research topics of 'Identification of salivary metabolomic biomarkers for oral cancer screening'. Together they form a unique fingerprint.

Cite this