Prediction of malignant transformation in oral epithelial dysplasia using infrared absorbance spectra

Abstract

Oral epithelial dysplasia (OED) is a histopathologically-defined, potentially premalignant condition of the oral cavity. The rate of transformation to frank carcinoma is relatively low (12% within 2 years) and prediction based on histopathological grade is unreliable, leading to both over- and under-treatment. Alternative approaches include infrared (IR) spectroscopy, which is able to classify cancerous and non-cancerous tissue in a number of cancers, including oral. The aim of this study was to explore the capability of FTIR (Fourier-transform IR) microscopy and machine learning as a means of predicting malignant transformation of OED. Supervised, retrospective analysis of longitudinally-collected OED biopsy samples from 17 patients with high risk OED lesions: 10 lesions transformed and 7 did not over a follow-up period of more than 3 years. FTIR spectra were collected from routine, unstained histopathological sections and machine learning used to predict malignant transformation, irrespective of OED classification. PCA-LDA (principal component analysis followed by linear discriminant analysis) provided evidence that the subsequent transforming status of these 17 lesions could be predicted from FTIR data with a sensitivity of 79 ± 5% and a specificity of 76 ± 5%. Six key wavenumbers were identified as most important in this classification. Although this pilot study used a small cohort, the strict inclusion criteria and classification based on known outcome, rather than OED grade, make this a novel study in the field of FTIR in oral cancer and support the clinical potential of this technology in the surveillance of OED.

Prediction of malignant transformation in oral epithelial dysplasia using machine learning

“…The performance of the MLA and the PCA-LDA study [ 5 ] are summarised in table 1 . The two approaches yield very similar results with the MLA showing a slight improvement over the PCA-LDA in terms of sensitivity, specificity and precision.…”

“…In four of these comparisons the differences are comparable to the spectral resolution of 4 cm −1 . The apparent discrepancy between 1614 cm −1 and 1628 cm −1 arises because the feature identified in the PCA-LDA analysis is distinctly asymmetric (see figure 3(c) of [ 5 ]) and this skews the value to a higher wavenumber. There is no corresponding entry against the PCA-LDA value of 1020 cm −1 because this wavenumber is not found in the top 150 metrics (equivalent to 0.3% of all metrics calculated).…”

“…This study utilised samples from seventeen patients with biopsy-proven OED. The collection, preparation and classification of samples has been described previously [ 5 ]. All patients have given written informed consent to a UK NHS Research Ethics Committee approved study that was carried out in compliance with the Helsinki Declaration (Liverpool Central REC ref: EC 47.01).…”

“…The collection and processing of FTIR spectral images has also been described previously [ 5 ] and these results are re-analysed in this work using a supervised machine learning algorithm (MLA) that was developed for the characterisation of oesophageal cancer cells [ 6 ] and recently applied to the characterisation of metastatic oral cancer tissue [ 2 ].…”

“…It has been shown that it is possible to classify oral squamous cell carcinoma (OSCC) by analysing FTIR spectral images [ 2 ], and other studies have successfully associated vibrational spectroscopy data with histopathological classification of potentially malignant oral lesions [ 3 , 4 ]. A recent study [ 5 ] used a supervised, retrospective analysis of tissue samples from patients with high-risk oral epithelial dysplasia (OED) lesions. As these archival samples came with prolonged, clinical follow-up and known outcomes (transformation or no transformation to cancer), it was possible to train and test a principal component analysis (PCA) linear discriminant analysis (LDA) model for the prediction of malignant transformation for OED.…”