[Are there differences between ex adenoma and de novo colorectal carcinomas?].

Colorectal carcinoma is a major cause of death throughout the Western world. It is increasingly recognized that any reduction in mortality must be achieved through the detection and removal of early and precancerous lesions. The primary attention for such a preventive strategy has been the polypoid adenoma and surveillance studies have shown a significant reduction in the incidence of carcinoma through systematic polypectomy of suspicious lesions. A potential problem with such a program, however, is raised by reports from Japan that some carcinomas seem to arise without a precursor polypoid adenoma, that is de novo. Although the histopathologic findings in such reports seem to clearly support this idea, this concept is not widely accepted in the Western world. We undertook a series of immunohistochemical (p53, bcl-2, Mib-1, E-cadherin, CD44, Stromelysin-3), and microsatellite analysis studies (on 17p (p53), 18q (DCC), 5q (APC), 8p, 2p and 1p), on groups of de novo and ex adenoma carcinomas in order to see if differences between the two groups of lesions exist. The results of these studies demonstrate that de novo carcinomas share several phenotypic and genotypic features with ex adenoma carcinoma (similar CD44 in the carcinomas, similar rates of LOH at APC and DCC loci), but have significantly higher rates of LOH at 17p, p53 over-expression and ST-3 expression indicating that tumor progression in de novo carcinoma is accelerated. These findings should help clarify the concept of de novo carcinoma and contribute to wider recognition of this important clinicopathologic entity.