Microscopic adenoma in nonpolyposis coli: incidence and relation to basal cells and lymphoid follicles

Gut-associated lymphoid tissue or so-called “dome” carcinoma of the colon: Review

AIM

To present a comprehensive review of the etiology, clinical features, macroscopic and pathological findings, and clinical significance of Gut-associated lymphoid tissue or “dome” carcinoma of the colon.

METHODS

The English language medical literature on gut- or gastrointestinal-associated lymphoid tissue (GALT) or “dome” carcinoma of the colon was searched and appraised.

RESULTS

GALT/dome-type carcinomas of the colon are thought to arise from the M-cells of the lymphoglandular complex of the intestine. They are typically asymptomatic and have a characteristic endoscopic plaque- or “dome”-like appearance. Although the histology of GALT/dome-type carcinomas displays some variability, they are characterized by submucosal localization, a prominent lymphoid infiltrate with germinal center formation, tumor-infiltrating lymphocytes, absence of desmoplasia, and dilated glands lined by columnar epithelial cells with bland nuclear features and cytoplasmic eosinophilia. None of the patients reported in the literature with follow-up have developed metastatic disease or local recurrence.

CONCLUSION

Increased awareness amongst histopathologists of this variant of colorectal adenocarcinoma is likely to lead to the recognition of more cases.

Prostaglandin-endoperoxide synthase 2 (cyclooxygenase-2), a complex target for colorectal cancer prevention and therapy

A plentiful literature has linked colorectal cancer (CRC) to inflammation and prostaglandin-endoperoxide synthase (PTGS)2 expression. Accordingly, several nonsteroidal antiinflammatory drugs (NSAIDs) have been tested often successfully in CRC chemoprevention despite their different ability to specifically target PTGS2 and the low or null expression of PTGS2 in early colon adenomas. Some observational studies showed an increased survival for patients with CRC assuming NSAIDs after diagnosis, but no clinical trial has yet demonstrated the efficacy of NSAIDs against established CRC, where PTGS2 is expressed at high levels. The major limits for the application of NSAIDs, or specific PTGS2 inhibitors, as adjuvant drugs in CRC are (1) a frequent confusion about the physiological role of PTGS1 and PTGS2, reflecting in CRC pathology and therapy; (2) the presence of unavoidable side effects linked to the intrinsic function of these enzymes; (3) the need of established criteria and markers for patient selection; and (4) the evaluation of the immunomodulatory potential of PTGS2 inhibitors as possible adjuvants for immunotherapy. This review has been written to rediscover the multifaceted potential of PTGS2 targeting, hoping it could act as a starting point for a new and more aware application of NSAIDs against CRC.

The mucosa beyond the polyps in juvenile polyposis

Little is known of early histologic changes in the mucosa of the colon aside from the polyps in Juvenile Polyposis. Provided with a surgical specimen of a total colectomy of a 6-year-old boy with this condition, this report describes those changes. The mucosa depicted a peculiar serrated profile of the uppermost part of the crypts due to elongation of them, dilated openings, and scant stroma. Also present were frequent aberrant crypts. Early juvenile polyps presented associated with lympho-glandular sites as distorted and microcystically dilated crypts containing granular and filamentous mucoid material. The findings possibly represent the abnormal cytologic potential of this genetic condition.

Colorectal carcinoma with dome-like phenotype: an under-recognised subset of colorectal carcinoma?

BACKGROUND

The term dome carcinoma has been applied to a variant of colorectal carcinoma, thought to derive from M-cells of the gut-associated lymphoid tissue. Its distinguishing morphological features include a non-polypoid plaque-like lesion composed of closely apposed cystically dilated glands lined by a single layer of non-mucinous cells, intensely PAS-positive intraluminal material, and a close spatial relation to lymphoid stroma.

AIMS AND METHODS

A search in the literature for such cases and the authors' experience with carcinomas sharing morphological details with dome carcinoma are presented to direct focus on this unique phenotype of colorectal carcinoma and to expand on its morphology.

RESULTS

Four such examples, all stage pT1, pN0 have been previously reported. Here two additional cases, with several features of dome carcinoma, stage pT1 and pT2, respectively, are added. An extensive intramucosal component, unassociated with adenomatous growth, a pink quality of the lesional cells, low grade budding, absence of cytoplasmic pseudofragments, and absence of necrosis characterised the present cases as well as intact MMR-proteins and loss of APC. As opposed to two of the previously reported cases, intraepithelial lymphocytes were unapparent in these cases and the lymphoid stroma was effaced along with tumour progression in one of the present cases. Hence, a range of appearances is encompassed by the dome carcinoma. The uncommon reporting of dome carcinoma may be due to lack of awareness of this particular subset of colorectal carcinoma. Indeed, one of the current cases was signed out as a conventional carcinoma, despite the comment in the pathology report of an unusual morphology.

CONCLUSION

Dome carcinoma may be more under-recognised than rare. The reporting of variants of colorectal carcinoma, displaying histological features suggestive of dome carcinoma, is encouraged in order to obtain more exact knowledge on its putative clinical significance.

Morphology of colorectal lymphoid aggregates in cancer, diverticular and inflammatory bowel diseases

The present study compares the characteristics of colorectal lymphoid aggregates in patients with carcinoma, diverticular disease, Crohn's disease, or ulcerative colitis of the large bowel. A total of 77 patients (41 colorectal cancer, 27 diverticular disease, six ulcerative colitis, three Crohn's disease) undergoing colorectal resection were included. Acetic acid staining, hematoxylin and eosin staining, CD3, CD20, and MIB1 immunostaining were employed in order to assess density, diameter, subepithelial or basal location, cellular profile, and proliferation of lymphoid aggregates in normal-appearing and actively inflamed large bowel. In normal-appearing tissue, mean density of lymphoid aggregates was lower in patients with ulcerative colitis and Crohn's disease than in those with colorectal cancer or diverticular disease. A larger mean diameter of aggregates was observed in patients with Crohn's disease. In inflammatory bowel diseases, a marked increase of the mean density of lymphoid aggregates was observed in actively affected specimens. In Crohn's disease more than in ulcerative colitis, the aggregates had a predominant basal or transmural distribution. In diverticular disease, active inflammation determined a less significant increase of subepithelial aggregates harboring a lower proportion of germinal centers. No significant variations of CD3, CD20, and MIB1 were recorded among the four disease groups. The lymphoid aggregate derangements observed not only in the actively affected mucosa but also in the unaffected colorectal lining of patients with Crohn's disease and ulcerative colitis support a relevant involvement of lymphoid aggregate system in the pathogenesis of inflammatory bowel diseases.

Adenocarcinoma of colon differentiating as dome epithelium of gut-associated lymphoid tissue

AIMS

An early adenocarcinoma of the ascending colon was confined to a mass of gut-associated lymphoid tissue (GALT). The first description of an adenocarcinoma of colon differentiating as dome epithelium is presented.

METHODS AND RESULTS

A plaque-like carcinoma was identified opposite the ileocaecal valve in an asymptomatic 56-year-old man with a family history of colorectal cancer. Malignant epithelium was confined to a mass of GALT filling but limited to the submucosa. Characterization of the neoplasm was undertaken by means of mucin histochemistry, immunohistochemistry, electron microscopy and assessment of DNA microsatellite instability status. The malignant epithelium comprised well differentiated columnar cells with a microvillous brush border and expressing MUC1, but no goblet cells or expression of MUC2. The demonstration of focal clusters of intraepithelial B-lymphocytes supported the presence of functioning M-cells within the malignant neoplasm. The cancer was DNA microsatellite stable despite the finding of tumour infiltrating lymphocytes.

CONCLUSIONS

There is evidence for the origin of colorectal neoplasia from dome epithelium in both experimental models and microreconstruction studies of early adenomas in nonpolypotic human colorectal mucosa. It is suggested that the lymphocyte-rich subset of colorectal cancer that expresses MUC1 but not MUC2 may be differentiating as dome epithelium of gut-associated lymphoid tissue.

Superficial-type adenomas and adenocarcinomas of the colon and rectum: a comparative morphological study

BACKGROUND/AIMS

It has been uncertain whether colorectal carcinomas preferentially arise on preexisting adenomas or de novo. However, from a morphological viewpoint, it seems unlikely that pedunculated or exophytic malignant polyps progress to the deeply ulcerated advanced carcinomas usually found clinically.

METHODS

The morphological features of 26 nonpolypoid, superficial-type colorectal tumors (17 adenomas and 9 adenocarcinomas) were compared to clarify the developmental route of colorectal carcinomas.

RESULTS

The adenomas and adenocarcinomas were very similar in size and gross appearance; however, examination of the surface appearances of unsectioned tumors by dissecting microscopy was helpful for distinguishing the two. Histologically, no adenomatous tissue was found in any case of superficial-type adenocarcinoma. Five of the nine adenocarcinomas, even including those of small size, invaded the submucosal layer, and two showed lymph node metastasis.

CONCLUSIONS

These findings suggest that superficial-type adenocarcinomas show rapid growth and aggressive behavior. We suggest that this type of carcinoma may not progress by the adenoma-to-carcinoma pathway but that it may arise from a very small superficial-type adenoma.

Ultrastructural study of M cells from colonic lymphoid nodules obtained by colonoscopic biopsy

The present study was undertaken to investigate ultrastructurally the epithelium covering lymphoid nodules obtained from colonoscopic biopsies of the human colon and rectum. Colonoscopy using the dye spraying contrast method was performed in nine patients who showed x-ray evidence of lymphonodular hyperplasia. Fifty-two colonoscopical biopsy specimens of lymphoid nodules were obtained from the ascending, transverse, and descending colon and rectosigmoid region. All specimens were observed by light and electron microscopy. Light microscopy disclosed large lymphoid follicles protruding into the lumen with a "dome-type" configuration. These extended to the lamina propria of the mucosa and were associated with a massive lymphoid aggregation extending as far as the muscularis mucosa from the submucosa. The epithelium covering these nodules contained a few goblet cells and many lymphocytes. Observation of the elevated surface at the apex by scanning electron microscopy revealed M cells with sparse microvilli in the dome epithelium surrounded by crypts. Transmission electron microscopy disclosed M cells enfolding many immature or mature lymphocytes and plasmocytes. The M cells had cytoplasmic microvilli (so-called "microfolds") on their surfaces, well-developed tubulovesicular systems, and vacuoles in the cytoplasm. The basic structure of the M cells as observed by scanning and transmission electron microscopy was the same as that of M cells in the Peyer's patches of humans and mice. The apical surface of the colonic lymphoid follicles in Crohn's disease patients was associated with erosions observed by scanning electron microscopy. The erosions proved to be the naked surface of the dome after removal of the epithelium, and many holes from 2.0 to 6.0 microns in diameter were observed on the naked surface. At high magnification, lymphocytes were seen projecting from holes (18%) on the naked surface of the dome. These ultrastructural findings indicate that human colonic lymphoid follicles are very similar to those seen in other species.

Effect of gut-associated lymphoid tissue on cellular proliferation in proximal and distal colon of the rat

In previous studies, chemically induced colonic carcinomas were found to originate preferentially from crypts adjacent to lymphoid tissue. Proliferative parameters and mucosecretion were analyzed in proximal and distal rat colon in relation to the proximity of lymphoid patches. Animals received an intraperitoneal pulse of bromodeoxyuridine 1-hr before death. In both proximal and distal colon, crypts located at the immediate proximity of the lymphoid formations contained fewer mucous cells (P less than 0.001), but a higher percentage of proliferative epithelial cells (P less than 0.001) than the crypts far from lymphoid formations. The labeling index was higher in crypts adjacent to lymphoid patches compared to crypts distant from lymphoid patches only in the lower third of the crypts. The association of an increased proliferative activity and a decrease in differentiated mucosecreting cells in colonic crypts adjacent to lymphoid patches could be related to the particular sensitivity of these crypts cells to the effects of mutagens and carcinogens.

Relative growth of adenomatous polyps of the colon. Stereology and allometry of multiple polyposis

The volume of the adenomatous mucosa (V), the area of the surface epithelium (Ss), the area of the glandular epithelium (Sg), and the Sg: Ss ratio were calculated in a series of 14 adenomatous polyps (APs) of a case of multiple polyposis of the colon. The equation of simple allometry was used to study the relative growth of the four series of values. Ss grew isometrically with size; Sg overgrew Ss and accounted for most of the increase in V. The Sg:Ss ratio increased with Sg and V.

Gut-associated lymphoid tissue and 1,2-dimethylhydrazine intestinal tumors in the rat: an histological and immunoenzymatic study

The association between chemically-induced intestinal carcinoma and gut lymphoid patches was studied in 20 male Sprague-Dawley rats 7 months after the first of 16 weekly injections of 1,2 dimethylhydrazine (DMH). The lymphoid patches of DMH-treated rats and of 14 untreated control animals were systematically studied histologically on sections of "swiss-rolled" whole intestine. It was found that 78% of the small-intestine carcinomas and 73% of the colorectal carcinomas were associated with intestinal lymphoid patches. Furthermore, misplaced and often atypical glandular crypts were often found in the parafollicular or interfollicular areas of lymphoid patches, in treated as well as in control animals. These glands could be the origin of the lymphoid-patch-associated carcinoma. Immunohistological staining with monoclonal antibodies (MAbs) against T-lymphocyte antigens or anti-IgM serum labelling B lymphocytes clearly localized early carcinoma and atypical glands in the T-dependent, interfollicular and parafollicular area of lymphoid follicles. An MAb directed against la-antigen stained some well-differentiated carcinomas and some atypical glands found in control rats. On the other hand, lymphoid patches, when not invaded by a carcinoma, were not modified in their number, size, morphology or cellular composition in DMH-treated rats as compared to control animals.

Lymphoglandular complexes of the colon: structure and distribution

The structure and distribution of lymphoglandular complexes of the colon are described. The cellular composition of these complexes, as detailed by immunohistochemistry and electron microscopy, suggests that they are sites of antigen processing. Variations in structure and in the number of complexes that are found in certain colonic diseases are documented.

Carcinoma in situ in nonpolypoid mucosa of the large intestine. Report of a case with significance in strategies for early detection

A case of carcinoma in situ of the flat, nonpolypoid mucosa of the large intestine in a 32-year-old man is reported. This case is of unusual importance because it was detected by endoscopy and was effectively managed; it sheds further light into the histogenesis of large intestinal carcinomas; and it points to alternate strategies for early detection and prevention of large intestinal carcinomas.

Background mucosal changes of primary advanced large intestinal cancer in patients without familial polyposis coli

In 337 cases of primary advanced large intestinal cancer in patients without familial polyposis coli, the risk factor for cancer was discussed chiefly from the background mucosa surrounding and apart from the cancers. The following findings were obtained: (1) in the mucosa surrounding cancers, adenoma was seen in 23 cases (6.8 per cent), adenomatous changes of the basal cells in 211 cases (62.6 per cent), and hyperplastic glands in 167 cases (49.6 per cent), (2) in the mucosa at least 10 mm from the cancers, adenoma, adenomatous changes of the basal cells, and hyperplastic glands were seen in 42 cases (12.5 per cent), 129 cases (38.3 per cent), and 40 cases (11.9 per cent), respectively. Therefore, it is suggested that microscopic adenoma and adenomatous changes of the basal cells might be a sign of premalignancy in patients without familial polyposis coli.

Human large intestinal epithelium: light microscopy, histochemistry, and ultrastructure

Despite numerous reports of morphologic characteristics of premalignant and malignant large intestinal epithelium, the literature lacks comprehensive reports of the morphologic features of the epithelium of the normal large intestine, except of the rectum. Large intestinal epithelium from 41 persons was obtained, and samples from the ascending, transverse, descending, and rectosigmoid areas were studied by light microscopy, histochemical techniques, and transmission and scanning electron microscopy. The morphologic features and histochemical reactions of the various segments of the large intestine are different. Neutral mucopolysaccharide is predominant in the ascending colon, whereas the rectum has predominantly or exclusively acidic mucin. Only three basic epithelial cell phenotypes have been identified: undifferentiated cells, mucous cells, and endocrine cells. The columnar cells at the surface between the crypts appear to be a variant of mucous cells. Compared with other segments, the rectum shows an unusually high concentration of endocrine cells, positively correlating with the high incidence of carcinoid tumors in that segment of the large intestine. The mucous cells in all segments contain large mucous vacuoles and small apical vesicles. The apical vesicles show variable electron density, being most dense in the ascending colon and becoming progressively less dense at the transverse and descending colon and most electron-lucent in the sigmoid colon and rectum. Ultrastructurally, the mucin shows a variable degree of heterogeneity in the proximal segments. This study suggests that some of the previously described ultrastructural features of abnormal large-intestinal epithelium may be only the result of failure to compare the so-called abnormal cells with normal cells from the same region. Well-controlled studies of the abnormal epithelium of a particular segment of large intestine must include the normal epithelium from the identical segment as control in order to make interpretations accurate.

Comparative study of familial polyposis coli and nonpolyposis coli on the histogenesis of large-intestinal adenoma

The entire length of the large intestine resected from six patients with familial polyposis coli (FPC) was step-sectioned, and microscopic adenomas composed of one to several glands were prepared into complete serial sections. Histopathologic comparison of these microscopic adenomas was made between FPC and nonpolyposis coli (those not included in the special disease group of FPC). The histogenesis of adenomas in FPC and nonpolyposis coli was considered as follows. (1) In nonpolyposis coli, adenomas arise from basal cells of the deep layer of the mucosa. (2) In FPC, most of the adenomas (81.2 per cent) arise in the same manner as described above, although there are some (18.8 per cent) that appear to have arisen in the middle or superficial layer of the mucosa. (3) Therefore, it is believed that the morphologic difference of the large-intestinal mucosa between FPC and nonpolyposis coli comes not from the difference in the number of adenomas present but from the difference in the position and/or timing of appearance of adenomas.

Histogenesis of microscopic adenoma and hyperplastic (metaplastic) gland in nonpolyposis coli

Histogenesis of microscopic adenoma in nonpolypoid colons (those not included in the special disease group of familial polyposis coli) was investigated using complete serial sections with the following results: Adenoma arises from basal cells in the deep layer of the mucosa; and two types of basal cells found were 1) those which had already undergone changes to adenoma and 2) those which are in their transitional stage to adenomas. Early-phase growth of adenoma is brought about by branching. Individual hyperplastic (metaplastic) glands constituting so-called hyperplastic (metaplastic) polyps are considered to be only an expression of one variant in the growing process of adenomas and the glands showing these changes most characteristically have a serrated pattern. Based on these findings, the histogenesis of adenoma in nonpolypoid colons is shown schematically.