Ex vivo photodynamic diagnosis to detect malignant cells in oral brush biopsies.

In this study we proved the efficiency of the fluorimetric detection of a minimum number of malignant cells ex vivo. The goal of this work was to investigate whether the combination of photodynamic diagnosis (PDD) with oral brush biopsy might become a suitable chair-side tool to detect early oral carcinoma. Small numbers (100-500) of established human tumour cells-small cell lung carcinoma (OAT 75), transitional cell carcinoma of the bladder (SW1710) and human embryonic kidney cells (HEK293)-were incubated with 2 mM 5-aminolaevulinic acid (5-ALA). In addition, 50 brush biopsies from volunteers were prepared. After 2 h and 3 h of incubation, all samples were investigated by spectrofluorometry. Measurements were performed in capillaries. For excitation (405 nm) and detection of fluorescence spectra, a fibre microprobe-spectrofluorometer system (fibre 400 microm) was used. A minimum of 100 malignant cells and 3 h of incubation with ALA were needed to detect a typical spectrum for protoporphyrin IX (PPIX). Some epithelial samples from brush biopsy showed strong (bacteria related) PPIX autofluorescence, which increased after the addition of 5-ALA. From the testing of various antibiotics and antiseptics it emerged that 0.4 mM chlorhexidine strongly reduced fluorescence in brush biopsies from healthy volunteers, whereas the fluorescence signal of established cancer cell lines decreased only a little. The experiments revealed that, by means of an optical microprobe, very few cancer cells (100) can be detected. The addition of chlorhexidine before the incubation of brush biopsies with 5-ALA increases the reliability of the test by largely reducing the autofluorescence signal due to the presence of bacteria. Chair-side diagnostics of epithelial carcinoma seem feasible.