Immunohistochemical expression of fatty acid synthase, apoptotic-regulating genes, proliferating factors, and ras protein product in colorectal adenomas, carcinomas, and adjacent nonneoplastic mucosa.

The normal mucosa-adenoma-carcinoma sequence in colon pathology provides an attractive model of tumor progression. The role of tumor suppressor genes, oncogenes, and proliferative markers in tumorogenesis has evolved considerably in the last decade. By immunohistochemistry means, we have studied p53, bcl-2, c-myc, p21-ras, ki67, and fatty acid synthase (a fatty-acid-synthesizing enzyme) in normal, dysplastic, and neoplastic mucosa. The results were correlated with clinicopathological features and overall survival (OS). Formalin-fixed, paraffin-embedded archival material from 100 nonconsecutive adenomas and 100 adenocarcinomas (ADCs), including adjacent-to-tumor nonneoplastic mucosa (ANNM), from patients with a 5-year follow-up period were studied. Negative controls were obtained from colon resections for nonneoplastic disease. Fatty acid synthase was associated with ADC (P = 0.0001). p53 protein was associated with high-grade dysplasia adenoma (AHGD), ADC (P = 0.0001), and pT stage (P = 0.003). bcl-2 was associated with adenomas with low-grade dysplasia (P = 0.009); c-myc was associated with ANNM (P = 0.005) and pT stage (P = 0.006). p21-ras was associated with AHGD (P = 0.0001) and ANNM (P = 0.01). Ki67 was associated with AHGD (P = 0.02) and ADC (P = 0.0001). Univariate analysis on neoplastic tissue revealed histological grade, pT stage, pN stage, p21-ras, and p53 to be significant markers of OS; p21-ras, p53, and c-myc were reliable markers when evaluated on ANNM. Multivariate analysis revealed pT stage, pN stage, and p21-ras to be independent prognosticators of OS on ADC; p21-ras and c-myc staining in the ANNM were correlated with worse survival (OS). We suggest that the evaluation in concert of clinicopathological data and immunohistochemical markers on both normal and abnormal colon tissue provides an attractive model of tumor progression; moreover, it may give important messages about the prediction of survival.