In vivo temporal evolution of ALA-induced normalized fluorescence at different anatomical locations of oral cavity: application to improve cancer diagnostic contrast and potential.

BACKGROUND: The focal goal of this study is to identify optimal accumulation periods for ALA-induced PpIX in different healthy anatomical sites of human oral cavity and different types of abnormal mucosa to improve the accuracy of the clinical applications such as photodiagnosis and tissue grading. MATERIALS AND METHODS: Laser-induced fluorescence (LIF) emission spectra, with excitation at 404 nm from a diode laser, were recorded with a miniature fiber-optics spectrometer from 13 anatomical sites of oral mucosa in 15 healthy volunteers and 30 suspicious sites in 15 patients after topical application of 0.4% 5-ALA solution for 15 min. The optimal accumulation time in different anatomical sites of healthy subjects and abnormal tissues were determined by studying the temporal variation in normalized fluorescence intensities (NFI) at 635, 685 and 705 nm. RESULTS AND DISCUSSIONS: In masticatory anatomical locations such as (gingival and hard palate) and in lining mucosa (inner lip, soft palate, floor of mouth, transition to floor of mouth, alveolus and ventral tongue) except vermillion border of lip (VBL) of healthy subjects (designated as group I), it was observed that optimum time for maximum accumulation of PpIX is 90 min. In comparison, for lateral side of tongue (LST) and dorsal side of tongue (DST) tissues (designated as group II), maximum accumulation of PpIX was observed in 150 min of ALA application. For diverse grade lesions of group I mucosa in patients, maximum accumulation of PpIX was observed in 90 min, whereas, in group II mucosa the optimum accumulation time was 150 min as in the case of healthy mucosa. Further, between different grades oral mucosa, maximum variation in NFI take place at these optimal time periods. CONCLUSIONS: The determination of the optimum accumulation time of ALA in oral mucosa based on NFI helps to improve the diagnostic contrast and accuracy of oral cancer diagnosis, and to plan appropriate timing for ensuing PDT.

Clinical grading of oral mucosa by curve-fitting of corrected autofluorescence using diffuse reflectance spectra.

BACKGROUND: Laser-induced autofluorescence (LIAF) and diffuse reflectance (DR) were collectively used in this clinical study to improve early oral cancer diagnosis and tissue grading. METHODS: LIAF and DR emission from oral mucosa were recorded on a fiber-optic spectrometer by illumination with a 404-nm diode laser and tungsten halogen lamp in 36 healthy volunteers and 40 lesions of 20 patients. RESULTS: Absorption dips in LIAF spectra at 545 and 575 nm resulting from changes in oxygenated hemoglobin were corrected using DR spectra of the same site. These corrected spectra were curve-fitted using Gaussian spectral functions to determine constituent emission peaks and their relative contribution. The Gaussian peak intensity and area ratios F500/F635 and F500/F685 were found to be useful indicators of tissue transformation. The diagnostic capability of various ratios in differentiating healthy, hyperplastic, dysplastic, and squamous cell carcinomas (SCCs) were examined using discrimination scatterplots. CONCLUSIONS: The LIAF/DR technique, in conjunction with curve-fitting, differentiates different grades of dysplasia and SCC in this clinical trial and proves its potential for early detection of oral cavity cancer and tissue grading.

Diffuse reflection spectroscopy: an alternative to autofluorescence spectroscopy in tongue cancer detection.

Laser-induced autofluorescence (LIAF) and diffuse reflection spectroscopy (DRS) are two emerging noninvasive optical tools that have shown immense potential to detect oral cavity pre-cancer. In a recent study, we have used spectral ratio reference standards (SRRS) of LIAF intensity ratios F500/F635, F500/F685, and F500/F705 for grading of tissues belonging to sites other than dorsal side of tongue (DST), lateral side of tongue (LST), and vermillion border of lip (VBL) that exhibited similar spectral shape for normal and abnormal tissues. This led to dismal diagnostic accuracies, and for the three LIAF-SRRS, normal tissue values were often misclassified as squamous cell carcinoma (SCC), which means that the true negatives were being wrongly identified as true positives. This study examines the applicability of the site-specific diffuse reflection spectral intensity ratio (R545/R575) of the oxygenated hemoglobin bands to classify different DST lesions and compares the results obtained with those obtained using LIAF-SRRS. DRS-SRRS of R545/R575 differentiated benign hyperplastic DST tissues from normal tissue with a sensitivity of 86% and specificity of 80%, which were indistinguishable using LIAF-SRRS. Further, in distinguishing hyperplastic tissues from premalignant dysplastic lesions, DRS-SRRS gave a sensitivity of 90% and a specificity of 86%, as compared to sensitivity of 89% and specificity of 72% shown by the three LIAF-SRRS together. The diagnostic accuracy and statistical adequacy of the two techniques were assessed by receiver operating characteristic curve (ROC-Curve) analysis. Three LIAF ratios gave a low overall ROC area under curve (ROC-AUCs) of 0.521, whereas the DR ratio (R545/R575) has shown an improved accuracy of 0.970 in differentiating different tissue types. While distinguishing hyperplastic from dysplastic tissues, the DR ratio gave a higher discrimination accuracy of 0.9. Based on these findings, it can be concluded that the DRS-SRRS technique by virtue of its low cost and higher diagnostic accuracies could be a viable alternate to LIAF-SRRS for in vivo screening of tongue pre-cancers and grading of different tissue types.