Tissue Phenotype of oral premalignant lesions (OPLs): Association with molecular markers and cancer development
Prediction of outcome for OPLs is challenging. Although histology is a good predictor for high-grade lesions [severe dysplasia and CIS], it is less effective for lesions with minimal or no dysplasia, which represent the bulk of lesions seen in the community. Molecular markers may play a major role in such assessments.
Objective: The aim of this study was to assess the potential of the Tissue Phenotype to identify high-risk OPLs by studying its correlation with conventional histopathology, molecular markers (LOH) and cancer development.
Methods: A total of 131 oral mucosa lesions were analyzed. The distribution of the pathology grades was as follows: normal- 30, hyperplasia -25, mild dysplasia -13, moderate dysplasia-10, severe dysplasia- 7 and invasive squamous cell cancer (SCC) -29 Feulgen stained sections were analyzed by High-resolution Image cytometry. A Morphometric Index measuring the degree of abnormality of the nuclei was calculated for each specimen. Each of the dysplastic cases were analyzed for 19 microsatellite loci on seven chromosome arms (3p, 4p, 8p, 11q, 13q, and 17p). The time course data for these patients has been stratified into those which develop cancer and those which have not developed cancer within 10 years.
Results: The Morphometric Index correctly identifies 94% of the high-grade OPLs while maintaining a specificity of 74%. MI was significantly higher in dysplasia with LOH than in dysplasia with no LOH for 4 chromosomes arms (3p, 9p, 4q,17p. The MI identified patients which progressed to cancer for 77% of the time while correctly identifying the patient which do not progress 78% of the time. Individuals with a high MI (higher than 4.5) showed a 13.1 fold increase in relative cancer risk than individuals with a low MI.
Conclusions: These data support the potential utility of Tissue Phenotype as a marker for molecular damages and progressive potential of Oral Premalignant Lesions. (Supported by grant R01DE13124, NIDCR)