Biofluid based biochemical and clinicopathological profiling for the detection and stratification of oral potentially malignant disorders.

OBJECTIVES

To assess the clinical applicability of Raman spectroscopy using serum and saliva for the identification and stratification of oral potentially malignant disorders (OPMDs) and to differentiate Oral Leukoplakia and Oral Submucous Fibrosis from tobacco habit controls.

MATERIALS AND

METHODS

This case control study was conducted on 60 patients divided into three groups: leukoplakia (L), Oral submucous Fibrosis (OSMF), and tobacco controls (TC). Serum and saliva samples were collected and analyzed using a confocal Raman spectroscope. Multivariate analysis using principal component analysis (PCA) and linear discriminant analysis (LDA) was performed to generate classification models.

RESULTS

Spectral variations were observed among the groups in both serum and saliva, particularly in protein, lipid, nucleic acid related bands. Serum-based models demonstrated higher classification accuracy (~ 87.5%) compared to saliva (~ 80%) in 2-Class analyses. Combined 3-Class models showed accuracies of 70% and 60% in serum and saliva respectively. Class-specific sensitivity, specificity and predictive values for the three-class serum and saliva models are reported.

CONCLUSION

In this exploratory case-control study, Raman spectroscopy of serum and saliva distinguished tobacco controls from patients with Oral Leukoplakia or OSMF with moderate-to-good classification accuracy, and serum-based models outperformed saliva-based models. These preliminary findings suggest that RS warrants further investigation as a minimally invasive adjunct for classification of OPMDs; validation in larger, prospective, multicentre cohorts with longitudinal follow-up is required before any early-detection or screening role can be established.

CLINICAL RELEVANCE: Raman spectroscopy-based classification might eventually support risk-based triage for biopsy in high-burden settings. The present cross-sectional case-control design demonstrates group discrimination only, and does not by itself establish a screening, early-detection, or longitudinal-monitoring role.