Colorectal Polyps With Extensive Absorptive Enterocyte Differentiation.

To the Editor.--We were very interested in the recent article by Yokoo et al,[1] entitled "Colorectal Polyps With Extensive Enterocyte Differentiation," and the subsequent letters by Drs Ban[2] and Yokoo[3] on the same subject.

We agree with Yokoo et al[1] that the great majority of colorectal polyps can easily be classified as tubular or villous adenomas and hyperplastic polyps. However, we encounter a not inconsiderable number of polyps that we cannot classify so easily. Some of these polyps have been called mixed hyperplastic adenomatous polyps / serrated adenomas by Longacre and Fenoglio-Preiser.[4]

Recently, we encountered a rectal polyp that closely resembled those reported by Yokoo et al.[1] The patient was a 71-year-old woman. The lesion measured approximately 4.5 cm in diameter and was composed predominantly of villous and branching papillary patterns. Tubular areas were also present. Columnar epithelial cells were generally eosinophilic. Uniform well-polarized, elongated nuclei were usually present in these cells. Goblet cells containing predominantly intestinal-type mucin were scattered throughout the lesion, and many of the columnar surface epithelial cells had a distinct brush border. A single focus of high-grade dysplasia, without invasion and measuring approximately 0.3 cm in diameter, was present in the center of this lesion.

Ki-67-positive nuclei were scattered throughout the lesion, as described by Ban. However, except in the focus of high-grade dysplasia, the majority of these nuclei were localized at the base and lower half of the crypts, villi, and papillary structures (Figure). Predominant staining of the luminal surface nuclei characteristic of tubular and villous adenomas was seen only in the focus of high-grade dysplasia. Intense nuclear staining for p53 was also seen only in the focus of high-grade dysplasia.

[Figure ILLUSTRATION OMITTED]

In agreement with Yokoo et al, we believe that the 6 polyps they studied as well as the polyp we describe here indeed belong to a histologically distinct type. The question of whether they are variants of mixed hyperplastic adenomatous polyp/ serrated adenomas[4] requires further study, as pointed out by Yokoo.[3] We feel that these lesions differ from classic tubular and villous adenomas in 2 principal respects: first, the proliferative activity, as demonstrated by Ki-67 staining, is seen throughout the lesions but predominantly at the base of crypts rather than the surface. Second, the lesions have a focally distinct serrated pattern. Thus, the morphology and proliferative pattern both resemble more closely the family of mixed hyperplastic/ adenomatous polyps or serrated adenomas.

In agreement with Ban,[2] we therefore believe that these polyps are most closely related to serrated adenomas, an entity somewhat ambiguous in its definition, which probably includes various lesions. Provisionally, the polyps discussed here might be termed villous serrated adenomas. The finding of focal high-grade dysplasia in our polyp supports the suggestion of Yokoo et al that these polyps may possess malignant potential.

Our studies were supported, in part, by University of California Davis Medical Center and Kaiser Medical Foundation.

RAMEZ M. G. SAROUFEEM,
FRCS
BYUNG HEE MIN, MD
HENRY TESLUK, MD
MICHAEL J. LAWSON, MD
BORIS H. RUEBNER, MD
Department of Pathology
University of California
Davis Medical Center
Sacramento, CA 95817

[1.] Yokoo H, Usman MI, Wheaton S, Kampmeier PA. Colorectal polyps with extensive absorptive enterocyte differentiation: histologically distinct variant of hyperplastic polyps. Arch Pathol Lab Med. 1999; 123:404-410.

[2.] Ban SC. Colorectal polyps with extensive absorptive enterocyte differentiation [letter]. Arch Pathol Lab Med. 2000;124:493-494.

[3.] Yokoo H. Colorectal polyps with extensive absorptive enterocyte differentiation [reply]. Arch Pathol Lab Med. 2000;124:494.

[4.] Longacre TA, Fenoglio-Preiser CM. Mixed hyperplastic adenomatous polyps/serrated adenomas: a distinct form of colorectal neoplasia. Am J Surg Pathol. 1990;14:524-537.

In Reply.--The polyp Saroufeem et al describe is similar to those we presented in our article, "Colorectal Polyps With Extensive Absorptive Enterocyte Differentiation."[1] However, their polyp contained an area of tubular and villous adenomatous glands with a single focus of high-grade dysplasia. Among the 6 polyps presented in our article, 2 polyps contained an area of tubular adenomatous glands, and another polyp contained areas of tubular and serrated adenomatous glands.

Longacre and Fenoglio-Preiser, among others, pointed out the existence of polyps composed of hyperplastic and adenomatous glands, which are often called mixed hyperplastic and adenomatous polyps.[2] The polyps Longacre and Fenoglio-Preiser studied were composed of adenomatous glands, but they resembled hyperplastic polyps on low-power examination because of their serrated architecture and mucinous differentiation of the glands. However, on high-power examination these glands appeared cytologically neoplastic (adenomatous). Therefore, applying the term "mixed" hyperplastic adenomatous polyps to these polyps is somewhat misleading, since they may be confused with the type of polyps with neoplastic adenomatous glands and coexisting nonneoplastic hyperplastic glands. Longacre and Fenoglio-Preiser pointed out that there is "abundant evidence for the divergent growth and differentiation of hyperplastic polyps and adenomas," and that this phenomenon appears to occur even within a single polyp. We believe that the polyps we studied belong to the group of hyperplastic polyps, because of our observation that the immature glandular cells in the deeper portions of the polyps appeared to differentiate into the absorptive enterocytes on the surface areas of the polyps. However, as mentioned, some of our polyps had coexisting neoplastic foci of tubular and serrated adenomatous glands. Therefore, these polyps belong to the mixed type as described above, but not to the "mixed" type as applied by Longacre and Fenoglio-Preiser in their publication.[2]

Saroufeem et al mention that the columnar epithelial cells of their polyp were generally eosinophilic with well-polarized, elongated nuclei with a distinct brush border. This description suggests that these cells show absorptive enterocyte differentiation, thus we may still consider their polyp to be a hyperplastic polyp with absorptive enterocyte differentiation having mixed components of tubular and villous adenomatous glands and, in addition, a focus of high-grade dysplasia.

Currently, we are studying the patterns of immunohistochemical staining on these various colorectal polyps utilizing antibodies against cytokeratin 20, Ki-67, p53, and Bc1-2. Our preliminary results suggest that the polyps we described as "hyperplastic polyps with extensive absorptive enterocyte differentiation" show a pattern of expression similar to the hyperplastic polyps rather than the serrated or tubular adenomas.

HIDEJIRO YOKOO, MD
Department of Pathology
Northwestern Memorial
Hospital
Chicago, IL 60611

[1.] Yokoo H, Usman MI, Wheaton S, Kampmeier PA. Colorectal polyps with extensive absorptive enterocyte differentiation: histologically distinct variant of hyperplastic polyps. Arch Pathol Lab Med. 1999; 123:404-410.

[2.] Longacre TA, Fenoglio-Preiser CM. Mixed hyperplastic adenomatous polyps/serrated adenomas: a distinct form of colorectal neoplasia. Am J Surg Pathol. 1990;14:524-537.