Abstract
Simple SummaryRaman spectroscopy, a light scattering technique that provides the biochemical fingerprint of a sample, was used on samples taken from patients with cancer and precancerous lesions. This information was then used to build a classifier to identify cancer and the precancerous phases. The ability to distinguish cancerous tissue from normal and precancerous tissue is diagnostically crucial as it can alter the patients’ prognosis and management. Moreover, as cellular changes are often present at the tumour margin, the ability to distinguish these changes from cancer can help in preserving more of the tissue and maintaining aesthetics and functionality for the patient.Early diagnosis, treatment and/or surveillance of oral premalignant lesions are important in preventing progression to oral squamous cell carcinoma (OSCC). The current gold standard is through histopathological diagnosis, which is limited by inter- and intra-observer errors and sampling errors. The objective of this work was to use Raman spectroscopy to discriminate between benign, mild, moderate and severe dysplasia and OSCC in formalin fixed paraffin preserved (FFPP) tissues. The study included 72 different pathologies from which 17 were benign lesions, 20 mildly dysplastic, 20 moderately dysplastic, 10 severely dysplastic and 5 invasive OSCC. The glass substrate and paraffin wax background were digitally removed and PLSDA with LOPO cross-validation was used to differentiate the pathologies. OSCC could be differentiated from the other pathologies with an accuracy of 70%, while the accuracy of the classifier for benign, moderate and severe dysplasia was ~60%. The accuracy of the classifier was lowest for mild dysplasia (~46%). The main discriminating features were increased nucleic acid contributions and decreased protein and lipid contributions in the epithelium and decreased collagen contributions in the connective tissue. Smoking and the presence of inflammation were found to significantly influence the Raman classification with respective accuracies of 76% and 94%.
Highlights
Oral cancer (OC) is the 16th most common cancer worldwide, 354,864 new cases and 177,384 deaths having been reported in 2018 [1]
The results show a very high accuracy in connective tissue (AUC = 0.94) and, to a lesser extent, in epithelium (AUC = 0.69) (Figure S5)
According to the Receiver operating characteristic (ROC) curves, the accuracy of the classifier was highest for the SCC class (AUC = 0.71), intermediate (AUC~0.6) for the benign, moderate and severe classes, and lowest (AUC = 0.46) for the mild, resulting in misclassification with benign and moderate
Summary
Oral cancer (OC) is the 16th most common cancer worldwide, 354,864 new cases and 177,384 deaths having been reported in 2018 [1]. The major risk factors for developing oral cancer are smoking and alcohol consumption, which can work synergistically [2,3,4] Premalignant lesions such as leukoplakia (white patch) and erythroplakia (red patch) carry an increased risk of malignant transformation [5]. The gold standard for diagnosing OC and dysplasia is through a conventional clinical oral examination, followed by a biopsy of any suspicious lesions and their histopathological examination [8]. The issue with this method is that it is subjective and prone to inter- and intra- observer errors [9]. A biopsy may not be representative of the whole lesion, as studies looking at the histology of tumours post operatively and comparing them to the preoperative biopsies have found that, in a significant number of cases, a neoplasia or carcinoma in-situ was misdiagnosed [10]
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call