Tissue autofluorescence as an intermediate endpoint in cancer chemoprevention trials.

Intermediate endpoints which reflect responsiveness to chemopreventive agents are needed in clinical trials. A potential problem in the assessment of these endpoints is their requirement for invasive biopsies. Secondly, their expression within the aerodigestive tract is doubtfully uniform. Without clinically definable disease, the potential for biopsy sampling error is significant. Our approach to this problem is the analysis of intrinsic tissue fluorescence. Various cellular components exist within tissues with the innate capacity to emit a characteristic spectral signature when excited at a particular wavelength of light. It can be postulated that cells in varying stages of progression towards malignancy will differ in both the qualitative and quantitative nature of these intrinsic cellular fluorophores which include folate derivatives and cytokeratins, as well as various vitamins and coenzymes. Using current bioptical technology, we have tested the applicability of tissue autofluorescence to chemoprevention studies through various model systems. In an N-nitroso-N-methylbenzylamine (NMBA)-induced rat esophageal cancer model, alteration of the 380 nm emission pattern corresponded to disease progression from normal mucosa through dysplasia to invasive cancer. In a multicellular tumor spheroid model, trans-retinoic acid (RA) altered autofluorescent profiles at multiple wavelengths including intensities at 340 nm, 450 nm, and 520 nm. Such RA-induced alterations corresponded to changes in the state of spheroid differentiation. In vivo human studies are in progress which suggest that aerodigestive mucosal neoplasias can be discriminated from normal mucosa by their autofluorescent profile. Because aerodigestive mucosa can be scanned without the need for invasive biopsies, the assessment of tissue autofluorescence as an intermediate endpoint may prove valuable.