Rapid and noninvasive diagnosis of oral and oropharyngeal cancer based on micro-Raman and FT-IR spectra of saliva

Fast, sensitive, and noninvasive techniques are needed for better health care management, particularly when traditional biopsies could be replaced with appropriate analyses of body fluids, such as saliva. Here is presented a proof-of-concept study, which aims to test a recently developed saliva samples preparation method, for oral and oropharyngeal cancer diagnosis, using micro-Raman and Fourier transform infrared (FT-IR) spectroscopic techniques. The detected biomarker bands and the cancer classification rates are compared and discussed. Saliva samples were collected from healthy donors and pathologically confirmed oral and oropharyngeal cancer patients. Principal components analysis (PCA) and principal components analysis-linear discriminant analysis (PCA-LDA) chemometric methods were applied to build discrimination models for the test and control groups. Based on the differences between salivary spectra of healthy and cancer patients, several biomarker bands were identified. Noteworthy, a significant vibrational biomarker band at 2064 cm(-1), assigned to thiocyanate, was observed in both the FT-IR and Raman data-set. Other cancer characteristic Raman bands were 754 cm(-1) (tryptophan), 530 and 927 cm(-1) (lysozyme), 1001 cm(-1) (phenylalanine), while the FT-IR biomarker band was located at 1075 cm(-1) (phosphodiester bonds stretching in DNA, RNA). The oral and oropharyngeal cancer was classified with an accuracy of 90% based on the micro-Raman data and 82% based on the FT-IR data set, respectively. The study showed that oral and oropharyngeal cancer can be differentiated from control saliva samples based on their respective micro-Raman and FT-IR spectral signatures, due to the biomolecular modifications induced by the disease. (C) 2021 Elsevier B.V. All rights reserved.

Automated summary

This study tested a saliva samples preparation method for oral and oropharyngeal cancer diagnosis using micro-Raman and FT-IR spectroscopic techniques. Biomarker bands for cancer classification were identified, with a significant vibrational band at 2064 cm(-1) observed in both datasets. The oral and oropharyngeal cancer could be differentiated with 90% accuracy using micro-Raman data and 82% accuracy using FT-IR data.