Background: Oral cancer (OC) is the sixth commonest cancer worldwide with alarming mortality. If identified at an early stage, the survival rate would be improved.
Methods: We appraised the feasibility of using (1)H nuclear magnetic resonance ((1)H NMR) based metabolomics in the identification of signature metabolites in serum from patients suffering with oral leukoplakia (OLK, n=100), oral squamous cell carcinoma (OSCC, n=100), and healthy control (HC, n=75). (1)H NMR derived data were processed by principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA) to reveal discriminating metabolites among these groups. Receiver operating characteristic (ROC) curve evaluation was also executed.
Results: NMR-derived serum metabolomics reveals eight differentially expressed biomarkers. Among them four biomarkers (glutamine, propionate, acetone, and choline) were able to accurately (ROC; 0.97) segregate 93.5% of OC cases equated to HC with substantial sensitivity and specificity. Similarly, four biomarkers (glutamine, acetone, acetate, and choline) were able to precisely (ROC; 0.96) discriminate, 92.4% of OLK cases from OSCC with considerable sensitivity and specificity. (1)H NMR-based metabolic fingerprint obtained for oral cancer is remarkable, even for OLK stage.
Conclusion: There is a systemic metabolic response to initial stage of cancer, which carries immense possibility for early appraisal.
Keywords: NMR-spectroscopy; OPLS-DA; Oral cancer; PCA; Serum-metabolomics.
Copyright © 2014 Elsevier B.V. All rights reserved.