Surface-enhanced Raman spectroscopy and multivariate analysis for the diagnosis of oral squamous cell carcinoma

The incidence of oral squamous cell carcinoma (OSCC) is increasing worldwide year by year, and its prognosis largely depends on early detection. However, it is difficult to make an early and definitive diagnosis of oral cancer patients by conventional oral examination. This study used surface-enhanced Raman spectroscopy (SERS) technology to achieve an early and noninvasive diagnosis of oral squamous cell carcinoma. Serum samples were collected from 74 patients with OSCC, 15 patients with malignant neoplasm of the salivary gland, and 94 normal subjects. And saliva samples from 25 patients with OSCC, 8 patients with malignant neoplasm of the salivary gland, and 35 normal subjects. High-quality surface-enhanced Raman spectra were obtained using silver nanoparticles (Ag NPs) as the enhancement matrix. The differences in the measured Raman spectra revealed differences in the relative concentrations of biomolecules in saliva and serum from oral squamous carcinoma, salivary gland malignancies, and normal subjects. In combination with a support vector machine (SVM) diagnostic model, the cross-validation method is used to assess model performance in terms of accuracy, sensitivity, specificity, F1-measure, balanced accuracy, Matthews correlation coefficient, and area under the receiver operating characteristic curve. The accuracy of model identification is above 80% for all serum samples, whereas the detection of saliva samples, although poorer than serum samples, is expected to improve the diagnostic accuracy with increased sample size. The results suggest that body fluid SERS combined with SVM has great potential to diagnose OSCC noninvasively and rapidly.

Automated summary

The incidence of oral squamous cell carcinoma (OSCC) is increasing worldwide, and early detection is crucial for prognosis. This study used surface-enhanced Raman spectroscopy (SERS) to achieve early and noninvasive diagnosis of OSCC. By analyzing serum and saliva samples from patients with OSCC, malignant neoplasms of the salivary gland, and normal subjects, the study found that body fluid SERS combined with a support vector machine (SVM) diagnostic model has great potential for the diagnosis of OSCC.