[De novo cancer and adenoma-carcinoma sequence of the colorectum--clinicopathological differences between de novo carcinoma and carcinoma with the sequence]

Abstract

The adenoma-carcinoma sequence ("The great majority of colorectal carcinomas arises from adenomas") proposed by Morson et al (1972) has been generally accepted world wide. Considering the sequence from the viewpoint of human carcinogenesis in general, however, a few contradictions and mysterious phenomena are found in the development and growth process of the colorectal carcinomas. It is evident that the histogenesis of the adenoma-carcinoma sequence including such contradictions is logically false. Consequently, we should fundamentally reconsider the histogenesis of colorectal carcinoma. The premise for the induction of the histogenesis of colorectal carcinoma is that carcinomas are much more objectively differentiated from adenomas with severe atypia, and that hyperplastic glands with slight atypia are differentiated from adenomas with slight atypia. In order to objectify the histological interpretation of atypical grades, conversion of complicated histological designs into real numbers is required. Two discriminant functions with two variables (N/C ratio and density of tubuli in a unit area) for differentiating carcinoma from adenoma have been determined using computed image processing. Based on the discriminant functions for differentiating objectively between carcinoma, adenoma, and hyperplasia, the histogenesis of colorectal carcinomas has been deduced as follows: 70-80% of colorectal carcinomas arise from the colorectal mucosa (de novo carcinoma) and 20-30% arise from adenoma. Observing the various clinicopathological phenomena in colorectal adenomas and carcinomas from the viewpoint of histogenesis, the contradictions do not enter the logically constructed model. Furthermore, the biological behavior of de novo carcinoma is clinicopathologically different from that of carcinoma with the sequence.