Classification and Quantification of Phenotypic Characteristics of Normal and Dysplastic Oral Cells

Introduction Oral cancer accounts for 2.5% of all cancer cases in the United States. The 5 year survival rate or stage 1 or localized oral cancer is 82%, and the survival rate of unstaged oral cancer is only 50%; therefore, early diagnosis is key. Current clinical practices for determining malignancy of tissue include physical examination and surgical biopsy, which can be plagued by high variability between observers and are invasive for the patient. Proflavine intercalates between base pairs of nucleic acids and has excitation and emission peaks at 455 (+/- 20) nm and 515 nm respectively; therefore, it can be used as a fluorophore to highlight cell structures in vitro. Methods Four cell lines, SCC-25, CAL-27, FaDu, and NCI cancer cells, were cultured in media, stained, and then imaged with a fluorescent microscope. Images were then analyzed to determine the phenotypic differences that existed between the cell types. Normal oral epithelial cells were also recovered and given the same treatment and analysis to provide a control group for comparison. Results The average brightness and size of the nucleus and cytoplasm was determined for each of the four cell types. Average entropy of each cell line was also determined. Discussion These results showed a significant difference in phenotypic characteristics between normal and cancerous cell lines. Using this information, we can begin to construct an algorithm that can be used to classify cells as normal or abnormal in automated and semi-automated screening tests