MINIMALLY-INVASIVE ORAL EXFOLIATED CELLS STUDY FOR PREMALIGNANT LESIONS USING RAMAN MICROSPECTROSCOPY

Abstract

Listen

Translate article

Squamous cell carcinoma of the oral cavity ranks as the 15th most common cancer in the world and the 10th most frequent cancer in males. The present study was undertaken for the development of new methods for early oral cancer detection based on Raman microspectroscopy of exfoliated cells. Exfoliated oral cells were collected by brush biopsy from patients attending Dublin Dental Hospital Dysplasia Clinic (25) and from healthy volunteers (25). Samples of exfoliated cells from normal mucosa and from pre-malignant lesions were collected using an endocervical cytobrush and placed in ThinPrep vials. Slides were prepared using the Thinprep2000 processor with the aim of forming a monolayer of cells for analysis. Raman spectra were acquired from the nucleus and cytoplasm of each cell using an XploRA confocal Raman instrument (HORIBA JobinYvon). As source, a 532 nm laser was focused by a 100X objective onto the sample and the resultant Raman signals were acquired in the 400 to 1800 cm−1 region. Glass spectral contamination was removed using extended multiplicative signal correction Following pre-processing, spectra were subjected to principal component analysis (PCA) and principal component-linear discriminant analysis (PC-LDA). The results show that Raman spectroscopy coupled with PCA could differentiate the nucleus and cytoplasm of the cell, the PC loadings showing that the cytoplasmic regions are dominated by protein bands while thenuclear regions are dominated by DNA bands. Furthermore, patient samples were discriminated from healthy volunteers based on DNA and lipids bands inthe PC loadings. Sensitivities of 91% and 97% and specificites of 98% and 89% were achieved for the cytoplasm and nucleus respectively, using PC-LDA. Thus, the findings of the study support the potential of Raman microspectroscopy for providing molecular level information from oral exfoliated cells and the future potential for screening of minimally invasive brush biopsy samples for oral pre-cancer and cancer.