Article Text

Download PDFPDF

Gastrointestinal epithelial neoplasia
  1. C A Rubio
  1. Gastrointestinal and Liver Pathology, Research Laboratory, Department of Pathology, Karolinska Institute and Hospital, 171 76 Stockholm, Sweden; Carlos.Rubio{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

We read with interest the viewpoint “Gastrointestinal epithelial neoplasia: Vienna revisited” by Dixon (


For many years Western gastrointestinal pathologists have followed the recommendations of British gastrointestinal pathologists. We learned that terms such as carcinoma in situ should be banned from the diagnostic terminology as it could lead to misinterpretation by surgeons and to unnecessary surgical intervention.

The Vienna classification1 has introduced new avenues to the understanding of the process of carcinogenesis in the gastrointestinal tract. For some Western pathologists in the Vienna group who also received histopathological training in Japan, the concept of intraepithelial carcinoma (that is, carcinoma in situ) and of intramucosal carcinoma appeared natural. Although during the first day of discussions other Western pathologists appeared reluctant to accept such controversial notions, the discussion became less intense during the second day, and at the end a consensus was reached, gaining finally the pages of this journal.1

The Vienna classification1 dismembered the concept of dysplasia from that of carcinoma in its earlier forms. After many years of studying adenomas we now know that low grade dysplasia may progress to high grade dysplasia. On the other hand it remains elusive whether carcinoma in situ is preceded by high grade dysplasia or develops without a prodromic phase. By the same token we do not know whether carcinoma in situ antedates intramucosal carcinoma. If those microscopic realities of colorectal carcinogenesis are being ignored, how are we going to learn in a correct sequential fashion the intricate molecular footsteps that telescope from dysplasia to submucosal carcinoma? As that Pandora box is being presented to pathologists we should treasure it by opening it little by little.

One criticism of the Vienna classification1 may be that although various categories of neoplasia were listed, the histopathological criteria for each one of the lesions were not verbalised, thus postponing the opportunity for its worldwide acceptance. Notwithstanding, some Western pathologists have started to herald the new “doctrine” by providing histopathological descriptions (criteria) for each one of the various categories proposed in Vienna.2

“To see or not to see” is not the question, as all lesions are there. As an example, dysplasia can be differentiated from carcinoma in situ.2 Dysplasia in the glandular gastrointestinal mucosa is characterised by spindle or cigar shaped, elongated, pleomorphic, hyperchromatic nuclei, and regular nuclear membrane whereas carcinoma in situ displays large vesicular nuclei, irregular conspicuous nucleoli, and scalloped nuclear membranes. Bridges of nucleolus associated chromatin reaching irregular chromatin deposits are seen along the nuclear membrane. Bridges of chromatin are also seen connecting angular chromatin clumps. The nuclear polarity is disrupted, and marked cell pleomorphism and aberrant mitosis are present. Structural alterations may occur such as budding or branching crypts or tubules, with epithelial septa and back to back glands, and cribriform growth of epithelial cells in clusters and sheets. Those structures are confined to the basement membrane of the epithelial layer.2 But surprisingly, despite those differences, high grade dysplasia and carcinoma in situ are still being regarded as synonyms in the Western literature.

The present discussion is beyond the usefulness of the Vienna classification as a tool for proper treatment; the discussion aims to point out our present lack of knowledge regarding the histogenesis of lesions represented by categories 4.2, 4.3, and 5.1 of Vienna1 and their correct identification for future molecular research.

The viewpoint of Dixon appears to be in concert with the desire of many Western pathologists who are willing to embrace this new “doctrine” in order to acquire accurate information on the histological steps followed by early neoplastic lesions of the gastrointestinal tract. Only then will we be able to translate such events into molecular terms.

Perhaps the spirit of Johan Strauss has succeeded in orchestrating not only fiddlers but also workers engaged in the microscopic diagnosis of gastrointestinal epithelial neoplasias.