| Summary: | Oral cancer is one of only four major cancers whose predicted mortality rate will significantly increase in the next 15 years. Many patients have potential precursor lesions but the methods for predicting the small number at high risk of cancer do not have sufficient predictive value or are labour intensive. The aim of this project is to investigate tests capable of predicting malignant transformation through detection of chromosomal instability. Image-based DNA image-based cytometry (‘ploidy analysis’) has been used routinely in the diagnostic Oral Pathology Unit but has not been tested on lesions clinically suspected as having high risk and its predictive values are not known for targeted use. Follow up data was compiled for 252 patients with risk lesions from local databases and cancer registries and a follow up study of malignant transformation was performed. Over than half of the dysplastic lesions, which transformed into cancer were aneuploid. Real time qPCR and QuantiGene Plex DNA assays were applied to samples of oral none dysplastic and dysplastic lesions to detect alterations in gene copy number using markers identified in previous studies in the laboratory. Discrepancies between methods were found, with inconsistent results caused by normalised to different housekeeping genes; RnaseP for qPCR, TPM1 for QGPlex and TERT in both techniques. Both these techniques were found to be insufficiently reliable for clinical use. Fluorescence in situ hybridization (FISH) was applied against Cen3/TP63, Cen7/EGFR, Cen8/PTK2, Cen11/CCND1 and 20ptel/MMP9. This proved almost as effective as image based ploidy analysis but was too labour intensive and time consuming for routine use. FISH revealed previously unrecognized zones of variation within dysplastic lesions and variation in clonal structure within them. Our data show that DNA image-based ploidy analysis combined with dysplasia assessment is the most predictive technique for use in a diagnostic setting and should be considered the reference standard.
|
|---|
