What are the histologic criteria for the diagnosis of incipient invasion in Barrett's mucosa with high-grade dysplasia?

H.D. Appelman (Ann Arbor)

From a pathologist's perspective, there are three recurrent annoying diagnostic problems that are inherent in Barrett's. First, the diagnosis of low-grade dysplasia in Barrett's mucosa is often difficult, since regenerative or reparative epithelium shares features with dysplastic epithelium, including primitive cytoplasm, large nuclei and mitoses. Because of this the category of "indefinite for dysplasia" was invented. In contrast, the diagnosis of high-grade dysplasia (HGD) is somewhat easier, because it has a much greater degree of cellular and architectural disorganization and more profound nuclear changes of hyperchromatism, crowding, pleomorphism and mitoses. However, the second diagnostic problem, although less common than the other two, relates to those dysplastic epithelia that are mainly lowgrade, but they have some features that suggest that the transition to high-grade is occurring. There are no tightly defined histologic guidelines for determining at exactly what point HGD is present. The third major problem is determining if there incipient invasion of carcinoma cells into the lamina propria within a field of HGD. The issue of whether there is incipient invasion usually arises when a biopsy with HGD contains dysplastic tubules that are crowded and/or have strange shapes or in which single cells or strands of cells seem to be disengaged from the adjacent dysplastic tubules. Pathologists try to make the diagnosis of incipient invasion, but what are the criteria they use to do so? Considering the importance of the diagnosis of invasive carcinoma in Barrett's mucosa, there is amazingly little in the way of published criteria, not just for incipient invasion, but for supposedly obvious or clear-cut invasion.

In the new Vienna classification of epithelial neoplasia for the entire gastroenteric tract, the term noninvasive carcinoma "stands for absence of evident invasion" while the term intramucosal carcinoma "stands for invasive into the lamina propria or muscularis mucosae" [1-3]. These definitions were agreed upon by an international group of gastrointestinal pathologists, some of the current giants in the field, but, unfortunately, they neglected to offer us specific criteria for applying those definitions. So, what about the rest of the world's literature; does it offer diagnostic help?

In some studies, the definition of incipient invasion is given as "invasion beyond (or through) the epithelial basement membrane", but this has little meaning unless the term is explained in great detail [4, 5]. Does this mean that the basement membrane is incomplete and any dysplastic cells adjacent to the defect in the basement membrane are invasive? Basement membranes beneath dysplastic epithelium are likely to be attenuated or even absent in small patches, yet we do not usually conclude that such defects indicate that invasion has occurred. Does invasion through the basement membrane mean that cells actually stick out from the dysplastic tubules through defects in the basement membrane into the lamina propria? We don't know, because details are rarely given. Furthermore, the basement membrane is not visible in routine H & E sections. It can only be visualized in special stains for basement membrane components, such as type IV collagen or laminen, and pathologists do not routinely stain biopsies for basement membrane components in order to prove that such defects occurs. In addition, some carcinomas actually synthesize basement membrane material, so that invasive structures may actually be encased in basement membrane.

A more detailed set of criteria for intramucosal carcinoma in gastric mucosa that should be applicable to Barrett's mucosa uses "irregularity of pit outlines with irregular infiltration into the lamina propria" as something that is used operationally, but the strict definition of "irregular infiltration into the lamina propria" is not given [2]. In the paper in which this is discussed the authors mention that separating small glands that are attached to larger tubules can be difficult to distinguish from lamina propria invasion, but no specific criteria for lamina propria invasion were given.

One criterion that seems to be used commonly for invasion is the presence of single dysplastic cells or small clumps of them, within the lamina propria, separated from the tubules of the background dysplasia [6, 7]. The problem with this criterion is that it requires serial sections to ensure that the cells are truly separated from the adjacent tubules.

In another study the authors mentioned different criteria for invasion in discussing three specific cases. The diagnosis of intramucosal carcinoma in biopsies in these cases was based on "glandular crowding or intraluminal necrosis, but actual invasion was not demonstrated". Furthermore, more traditional evidence of invasion, namely "single-cell invasion, proliferation of irregular shaped glands, and sheets of neoplastic cells or stromal desmoplasia were not seen". As it turned out, when these three esophagi were resected, no invasive carcinoma was found, so the criteria used to identify invasion, namely glandular crowing or necrosis, may not have been the right ones [8].

Desmoplasia, that is, fibroblastic stromal proliferation usually including collagen and mucopolysaccharides, has been mentioned as a helpful hint in identifying invasion. There is one basic problem with desmoplasia. It virtually never occurs when the carcinoma is limited to the mucosa, but it develops when the carcinoma reaches the submucosa. Thus it is of no use in identifying incipient invasion of the lamina propria.

There is one indirect but very powerful histologic hint that an invasive carcinoma is present when the biopsy only contains HGD. That is an ulcer appearing either as granulation tissue from the ulcer bed or as a separate fragment of exudate in the biopsy [9, 10]. Dysplasias virtually never ulcerate, while carcinomas commonly do so.

If the criteria for invasion are so unsettled, how then do we pathologists make that diagnosis. The answer given at the OESO Congress four years ago by Riddell has not changed [11]. He said that "the only criterion for the confident or unequivocal diagnosis of invasive carcinoma in the lamina propria is the pathologists own subjective and personal interpretation that invasion may or may not be present. If in many instances minimal invasion is not based on objective criteria but on hunches, experience, and the fear of missing an underlying carcinoma, then there are no minimal criteria for" such invasion. What this says is that experienced pathologists have developed subjective criteria for detecting incipient invasion in biopsies, and such criteria appear to correlate with the finding of invasion once the esophagus has been resected. They try to create objective descriptions of these subjective criteria, but they often fail in these attempts. I have been using such a subjective approach for years, and I suspect that so have virtually every other pathologist who deals with intraepithelial neoplasia in the columnar gastrointestinal tract. It is likely that this unsettled bit of reality explains why the literature on this subject is also so unsettled.

References

1. Schlemper RJ, Riddell RH, Kato Y, et al. The Vienna classification of gastrointestinal epithelial neoplasia. Gut 2000;47:251-255.

2. Lauwers GY, Riddell RH. Gastric epithelial dysplasia. Gut 1999;45:784-790.

3. Geboes K, Van Eyken P. The diagnosis of dysplasia and malignancy in Barrett's oesophagus. Histopathology 2000;37:99-107.

4. Peters JH, Clark GWB, Ireland AP, et al. Outcome of adenocarcinoma arising in Barrett's esophagus in endoscopically surveyed and nonsurveyed patients. J Thorac Cardiovasc Surg 1994;108:813-822.

5. Van Sandick JW, van Lanschot JJB, Kuiken BW, et al. Impact of endoscopic biopsy surveillance of Barrett's oesophagus on pathological stage and clinical outcome of Barrett's carcinoma. Gut 1998;43:216-222.

6. Haggitt, RC. Barrett's esophagus, dysplasia, and adenocarcinoma. Hum Pathol 1994;25:982-993.

7. Lewin KJ, Appelman HD. Tumors of the esophagus and stomach. Fascicle 18, 3rd series. Atlas of tumor pathology. Washington: Armed Forces Institute of Pathology, 1996:135.

8. Cameron AJ, Carpenter HA. Barrett's esophagus, high-grade dysplasia, and early adenocarcinoma: a pathological study. Am J Gastroenterol 1992;92:586-591.

9. Burke AP, Sobin LH, Shekitka KM, Helwig EB. Dysplasia of the stomach and Barrett's esophagus: a follow-up study. Mod Pathol 1991;4:336-341.

10. Montgomery EA. Unpublished data, September, 2000.

11. Riddell RH. Are there minimal criteria for the confident or unequivocal diagnosis of invasive carcinoma in the lamina propria both in Barrett's mucosa and in squamous mucosa? In: Giuli R, Galmiche JP, Jamieson GC, Scarpignato C, eds. The esophagogastric junction. Paris: John Libby Eurotext, 1998:1171-1172.