In vivo photo-detection of chemically induced premalignant lesions and squamous cell carcinoma of the rat palatal mucosa.

Photo-detection using in vivo fluorescence was studied for different stages of chemically induced premalignant lesions and squamous cell carcinoma (SCC) of the Wistar rat palatal mucosa. It was found that the epithelial dysplasia (numerically expressed in the epithelial atypia index (EAI) of the rat palate, induced by repeated application of the carcinogen 4-nitroquinoline 1-oxide (4NQO), showed an increase approximately proportional to the duration of the application period. Photo-detection of the lesions using Photofrin-induced fluorescence was studied with dual-wavelength excitation and the subtraction of images, in an attempt to reduce the autofluorescence. The Photofrin dose was 2.5 mg kg-1. This was based on a dose-response study for normal tissue damage by photodynamic therapy (PDT) in this animal model, because the underlying rationale was to study photo-detection as a method of locating additional (early) malignancies in patients treated by PDT. Fluorescence intensities 24 and 48 h after injection of Photofrin were shown to increase with the duration of 4NQO application and with increasing EAI. For an EAI greater than 15, there was a statistically significant difference (p < 0.01) between the fluorescence signals obtained with and without the injection of Photofrin. Fluorescence signals of these lesions without the use of Photofrin (autofluorescence) also showed an increase with increasing stages of epithelial dysplasia of the rat palate. However, the fluorescence signals obtained with Photofrin were always higher than those of the autofluorescence. From this study, we conclude that photo-detection with Photofrin has potential in distinguishing chemically induced premalignant lesions and squamous cell carcinomas from the normal rat palatal mucosa. Photofrin (2.5 mg per kg of body weight) certainly adds to the sensitivity of photo-detection, but autofluorescence alone also has promising features for detecting premalignant and malignant lesions of the oral mucosa.