Infrared spectrum signature of salivary extracellular vesicles from oral cancer patients – a new potential diagnostic tool

Introduction: Extracellular vesicles (EVs), like exosomes, are found in all body fluids, including saliva. EVs play paracrine and endocrine key roles in both physiological and pathological conditions, particularly in cancer. As EVs carry molecular and genetic markers that characterize their parental cells, they also have an essential diagnostic value. EVs from squamous cell carcinoma cells of the oral epithelial lining are expected to concentrate in saliva. Aim: to determine if the infrared (IR) spectra of salivary EVs from oral cancer (OC) patients and healthy individuals (HI) may present different patterns that could serve for diagnostic purposes. Material and methods: Whole saliva samples were collected from a total of 21 OC patients: 9 individual samples and 12 samples, which were joined into one pooled sample. Whole saliva samples were also collected from 13 HI: 8 individual and 5 samples, which were joined into one pooled sample. EVs were pelleted from all samples of saliva (17 individual samples and 2 pools) using differential centrifugation (12,000g followed by 120,000g). The mid-IR absorbance spectra were measured in the range of 700cm^-1 to 5,000cm^-1, using MIR8025 Oriel Fourier transformation infrared equipped with a PIKE MIRacle ZnSe attenuated total reflectance attachment. Results: Individual and pooled pellets from OC patients showed a similar pattern of spectroscopy which was different from that of the HI (individual and pooled). The IR spectra of OC versus HI was consistently different in the 1,025 (cancer peak) and 1,080 (healthy peak) cm^-1 region, 1,745 cm^-1, 2,855 cm^-1 and 2,925 cm^-1 (low-to-absent in cancer), and in the region of 1,540 cm^-1 (higher peak in cancer). A linear discriminant analysis model showed a sensitivity of 90% and specificity of 100%. Conclusions: This pioneering study showed the existence of a specific IR spectral signature for OC salivary EVs. Though this is a pilot study, the high sensitivity and specificity of the outcomes warrant further analysis that may yield an efficient, reproducible and non-invasive clinical tool for the diagnosis of oral cancer at its very early stages or in oral lesions with potential for malignant transformation.