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

Broadsheet number 52: Molecular genetics of colorectal cancer

The molecular genetics of colorectal cancer is presented in an order that ascends from the basic to the applied: molecular mechanisms, morphogenesis, classification and diagnosis. Major consideration is given to the nature of genetic instability and the role of this mechanism in driving neoplastic progression. It is shown how the fundamental principle of genetic instability cuts across applied research, tissue diagnosis and clinical management with respect to both sporadic and inherited forms of colorectal cancer.

Clinicopathological features of superficial depressed-type colorectal neoplastic lesions

OBJECTIVE

We performed this study to analyze the endoscopic findings, dissecting microscopic features, and p53 immunostaining in superficial depressed-type (depressed) colorectal neoplastic lesions.

METHODS

Dissecting stereomicroscopy was used to ascertain the size and pit pattern of lesions removed by endoscopic snare polypectomy. Immunohistochemical staining of p53 was performed with an antigen retrieval system using a monoclonal antibody to p53.

RESULTS

All depressed neoplastic lesions (submucosal carcinoma, n = 6; high-grade dysplasia, n = 14; and adenoma, n = 30) were small (< 1 cm in diameter) and were detected as a depression with or without a marginal elevation on colonoscopic examination. In the dissecting microscopic study, submucosal carcinomas and lesions of high-grade dysplasia almost exclusively showed irregular small pits, with the exception of four malignant lesions with moderate submucosal invasion in which the pit structure was absent. In contrast, adenomas had either regular small (29/30 lesions) or oval pits (1/30 lesions). Rates of p53 positivity were 100%, 64%, and 7% in depressed submucosal carcinomas, lesions of high-grade dysplasia, and adenomas, respectively, thus the prevalence of p53 positivity was significantly higher in the former two groups than in the adenoma group.

CONCLUSIONS

The high frequency of invasive carcinoma and high-grade dysplasia found in depressed colorectal neoplastic tumors, despite their small size, indicates that these lesions may be a subtype of colorectal tumor with more aggressive malignant potential at an earlier stage.

Loss of heterozygosity and microsatellite instability in de novo versus ex-adenoma carcinomas of the colorectum

Small adenocarcinomas of the colorectum showing no evidence of origin from an adenoma have been called de novo carcinomas, a name that implies an origin via a different molecular genetic mechanism than the usual colorectal carcinoma which develops from an adenoma. Using microsatellite analysis, 35 early (pT1) de novo and 36 pT1 ex-adenoma carcinomas were compared using 8 microsatellite loci at 6 different chromosomal loci (1p, 2p, 8p, 5q, 17p, and 18q) known or hypothesized to be important for colorectal carcinogenesis. The rate of loss of heterozygosity (LOH) at the 17p locus (near the p53 gene) was significantly higher in the de novo than in the ex-adenoma group (73 vs. 37%, P = 0.004). The rates of LOH at the other loci (including the APC and DCC genes) and the rate of MSI were not significantly different in the two groups. These results indicate that de novo carcinomas of the colorectum develop via a similar carcinogenetic pathway as conventional ex-adenoma carcinomas; however, their higher rate of LOH at 17p is evidence for a biologically more advanced lesion with more frequent p53 mutations, consistent with clinicopathological data indicating that de novo carcinomas are more aggressive than ex-adenoma carcinomas.

Growth patterns in various macroscopic types of noninvasive intramucosal colorectal carcinoma with special reference to apoptosis and cell proliferation

PURPOSE

Apoptotic cell death and cell proliferation play important roles in the histogenesis and development of colorectal carcinoma. The aim of this study was to examine the relationship between apoptosis and cell proliferation in various macroscopic types of intramucosal colorectal carcinoma in relation to the expression of p53 and bcl-2.

METHODS

One hundred forty cases with endoscopically or surgically resected intramucosal colorectal carcinoma were studied. There were 57 cases of polypoid-type carcinomas, 55 cases of superficial-type carcinomas, and 28 cases of granular-type, laterally spreading tumors. Polypoid-type carcinomas were pedunculated, subpedunculated, or sessile polyps. Superficial-type carcinomas were flat lesions with a smooth, even surface. Granular-type, laterally spreading tumors were superficially spreading lesions with aggregates of nodules and a granular surface. Apoptotic cells were identified by the in situ DNA nick end labeling method. Ki-67, p53, and bcl-2 expression were examined immunohistochemically.

RESULTS

The superficial-type carcinoma apoptotic index (30.9 percent) was significantly lower than that of polypoid-type carcinoma (54.4 percent) and granular-type, laterally spreading tumor (60.7 percent). The superficial-type carcinoma proliferative index (67.3 percent) was significantly higher than that of polypoid-type carcinoma (42.1 percent) and granular-type, laterally spreading tumor (28.6 percent). In superficial-type carcinomas the proliferative index in p53-positive carcinomas was significantly higher, and the apoptotic index was higher in carcinomas with a lower proliferative index. There was no significant difference in apoptotic index, proliferative index, or p53 protein overexpression between de novo carcinomas and those that had arisen in precursor adenomas.

CONCLUSIONS

The pattern of cell death and proliferation may vary with different macroscopic types of intramucosal colorectal carcinoma. Superficial-type colorectal carcinomas especially demonstrate diminished apoptosis and increased cell proliferation. This may be useful in understanding their biologic behavior.

Invasion depth diagnosis of depressed type early colorectal cancers by combined use of videoendoscopy and chromoendoscopy

BACKGROUND

Depressed type early colorectal cancers are found less frequently than other polypoid cancers although they have a higher submucosal invasion rate. Recently videocolonoscopy and chromoendoscopy have become available and precise descriptions of these lesions are now routine. Because endoscopic mucosal resection is designated for intramucosal and focally extended submucosal (m-sm1) cancers, an evaluation of the characteristic findings indicating invasion depth with these modalities is important.

METHODS

Between January 1991 and March 1996, 64 depressed type early colorectal cancers were detected and treated. When a faint abnormality of the mucosa was suspected by routine videocolonoscopy, 0.1% of indigo carmine solution was sprayed on the mucosal surface (chromoendoscopy). Colonoscopic findings of m-sm1 cancers and moderately and massively extended submucosal (sm2-3) cancers were retrospectively reviewed and compared with confirmed histologic findings.

RESULTS

Characteristic colonoscopic findings needed for surgical operation were as follows: (1) expansion appearance, (2) deep depression surface, (3) irregular bottom of depression surface, and (4) folds converging toward the tumor. By using these findings, the invasion depth of depressed type early colorectal cancers could be correctly determined in 58 of 64 lesions (91%).

CONCLUSIONS

Characteristic colonoscopic findings obtained by a combination of videocolonoscopy and chromoendoscopy are useful for determination of the invasion depth of depressed type colorectal cancers, an essential factor in choosing a treatment modality.

Somatically acquired genetic alterations in flat colorectal neoplasias

Somatically acquired mutations in several genes have been reported as playing an important role during colorectal tumorigenesis. Two alternative groups of carcinomas, termed LOH+ and RER+, have been defined on the basis of their genetic anomalies, a biallelic inactivation of the APC or the TGF-betaRII genes, occurring as an alternative, in LOH+ or RER+ tumors. It is a generally accepted hypothesis that most of colorectal cancers (CRC) develop from a pre-existing adenomatous polyp. Such benign lesions are usually exophytic polyps, a small proportion of adenomas having been described as flat lesions. The latter histological category has thus been proposed to bear specific genetic alterations. In order to examine this hypothesis, we have characterized a series of 44 flat colorectal neoplasias for their RER status and for somatic APC, KRAS and TGF-betaRII genes mutations. Flat colorectal neoplasias were found to be of the RER+ subtype in 22% of cases, all of them exhibiting a TGF-betaRII mutation. A mutation of the APC and KRAS genes has been found in 42% and 4% of tumors, respectively, none of these tumors being of the RER+ subtype. With the exception of a low KRAS mutation rate, flat adenomas appear to follow tumorigenesis pathways very similar to those identified in exophytic adenomas and carcinomas.

Differences in the diagnostic criteria used by Japanese and Western pathologists to diagnose colorectal carcinoma

BACKGROUND

In view of the many studies of early stage colorectal carcinoma from Japan, it is essential to know whether the criteria for the histologic diagnosis of colorectal carcinoma are similar in Japan and Western countries.

METHODS

Eight expert pathologists from Japan (4), North America (2), and Europe (2) individually reviewed microscope slides of 20 colorectal lesions from Japanese patients who had undergone endoscopic mucosal resection or surgery because early stage carcinoma and/or adenoma was suspected. The pathologists indicated the pathologic findings on which they based each diagnosis.

RESULTS

For 11 slides that showed adenoma according to the Western pathologists with low grade dysplasia according to at least half of them, the Japanese diagnosed definite carcinoma with or without adenoma in 4 cases and adenoma in 5, and in 2 cases they were equally divided between a diagnosis of adenoma and carcinoma. For five slides showing adenoma with high grade dysplasia according to the Western pathologists, the Japanese diagnosed definite carcinoma with adenoma in three cases and adenoma in one, and in one case they were equally divided between a diagnosis of adenoma and carcinoma. For one case in which the Western pathologists were equally divided between a diagnosis of carcinoma and adenoma with high grade dysplasia, all the Japanese pathologists diagnosed definite carcinoma with or without adenoma. Three slides showed definite carcinoma with or without adenoma, according to both the Western and the Japanese pathologists. The presence of invasion was the most important diagnostic criterion of colorectal carcinoma for the Western pathologists, whereas for the Japanese the nuclear features and glandular structures were more important.

CONCLUSIONS

In Japan, colorectal carcinoma is diagnosed on the basis of nuclear and structural criteria, even in cases considered by Western pathologists to be noninvasive lesions with low grade dysplasia. This diagnostic practice may contribute to the relatively high incidence of early stage colorectal carcinoma reported in Japan as compared with Western countries.

Induction of aberrant crypt foci and flat-type adenocarcinoma in the colons of dogs by N-ethyl-N'-nitro-nitrosoguanidine and their sequential changes

Sequential endoscopic observation of dog colons was performed during colon carcinogenesis. Two beagle dogs were given suppositories containing N-ethyl-N'-nitro-N-nitrosoguanidine (ENNG) every day for five months. In month 3, aberrant crypt foci (ACF), a putative preneoplastic lesion, were found in the colons of both dogs, but not in an untreated dog. The frequency of ACF increased until month 10, and then decreased. In month 9, very small lesions, less than 1 mm in diameter, which were similar to human early flat tumors, were first noticed. One of these lesions grew to about 7 mm in size without a change in its shape for 10 months. There were more than ten flat-type tumors in the two dogs, but such lesions were not found in the untreated dog. By biopsy, two of the lesions were proved to be well-differentiated adenocarcinomas histologically. Four polypoid lesions were found in one of the carcinogen-treated dogs. Thus, flat-type adenocarcinomas were induced in the dog colon by ENNG, and their development was followed by magnifying endoscopy.

Significance of histologic criteria for the management of patients with malignant colorectal polyps and polypectomy

BACKGROUND

The management of invasive cancer in colorectal polyps (malignant polyps) is controversial, particularly with regard to the different sets of histologic criteria for deciding whether malignant colorectal polyps should be treated by polypectomy alone or be followed by surgical resection. We report on the outcome of patients in accordance with the histologic assessment of their malignant polyps.

METHODS

Malignant polyps were defined as having favourable histology (free margin, grade I or II, and no angiolymphatic invasion) or unfavorable histology (no free margin, grade III, or angiolymphatic invasion). Malignant polyps with favourable histology were treated by endoscopic polypectomy alone, whereas further therapy was recommended for malignant polyps with unfavourable histology. Residual cancer in a resection specimen and local or metastatic recurrence during the follow-up period (mean, 60 months; range, 12 - 120) were defined as adverse outcome.

RESULTS

Thirty-seven malignant polyps were detected in 35 (0.5%) of 6605 patients. Five of these 35 patients were treated by primary bowel resection and analysed separately. In the other 30 patients the following unfavourable histologic signs were detected in 20 (62.5%) of 32 malignant polyps; no free margin in 16, grade III in 1, and angiolymphatic invasion in 3 polyps. Twelve polyps with favourable histology had no adverse outcome; in contrast, 5 of 20 polyps with unfavourable histology had an average outcome (P < 0.05).

CONCLUSIONS

Locally excised malignant polyps without unfavourable histologic signs may not need further surgical treatment; for all other malignant polyps an ensuing bowel resection is recommended.

MUC1 and MUC2 mucins in flat and polypoid colorectal adenomas

AIMS

To examine the expression of MUC1 and MUC2 apomucins and distribution of MUC phenotypes (MUC2+/ MUC-, MUC2+/MUC1+, MUC2-/ MUC1+, MUC2-/MUC1-) in colorectal tubular adenomas in order to compare the distribution of phenotypes in flat and polypoid adenomas.

METHODS

Endoscopically resected specimens of 35 flat and 15 polypoid tubular adenomas measuring less than 10 mm were examined and compared for the expression of MUC1 (MUSE11) and MUC2 (CCP58) and combined MUC phenotype distribution using conventional immunohistochemistry.

RESULTS

There was no significant difference between flat and polypoid adenomas in their expression of MUC1 and MUC2 and the MUC phenotype distribution when stratified by grade of histological atypia. Adenomas with low grade atypia showed more extensive MUC2 expression than MUC1 (MUC2+/MUC1-phenotype). Expression of MUC1 was more extensive in adenomas with high grade atypia and the majority displayed either MUC2+/ MUC1+ or MUC2-/MUC1+ phenotypes.

CONCLUSIONS

MUC2/MUC1 phenotypes were similar in flat and polypoid adenomas when stratified by grade of atypia. High grade atypia was characterised by reduced MUC2 and increased MUC1 expression in both types of adenoma. The phenotype MUC2-/MUC1+ occurs in tubular adenomas and cannot be a specific marker for de novo colorectal cancer.

Mixed epithelial polyps in association with hereditary non-polyposis colorectal cancer providing an alternative pathway of cancer histogenesis

A member of a hereditary non-polyposis colorectal cancer (HNPCC) family developed two colorectal cancers and multiple polyps within four years of a negative colonoscopic examination. One of the cancers was only 4 mm in diameter and showed the gross and endoscopic appearances of a de novo carcinoma. Microscopic examination of multiple levels revealed a mixed hyperplastic polyp/adenoma (mixed polyp) in contiguity with the cancer. The colon harboured additional polyps of which five were tubular adenomas, seven were hyperplastic polyps and seven were mixed polyps (architecturally compatible with hyperplastic polyps but with atypical cytology). Atypical features of the mixed polyps included tripolar mitoses, bizarre chromatin aggregations and multinucleation. One mixed polyp showed DNA microsatellite instability. Under the influence of the mutator defect, hyperplastic polyps may develop atypical or adenomatous features and show progression to carcinoma. Such an alternative morphogenetic pathway could explain the differing molecular and pathological profiles of cancers showing DNA microsatellite instability.

Effect of K-ras mutation on morphogenesis of colorectal adenomas and early cancers: relationship to distribution of proliferating cells

The authors evaluated the association of the K-ras mutation with the distribution of proliferating cells and the macroscopic appearance of colorectal tumors. A total of 122 colorectal adenomas and 96 early cancers were classified macroscopically as follows: (1) polypoid, exceeding 3 mm in height; (2) flat or hemispherically elevated 3 mm or less; and (3) depressed. The intramucosal areas of these tumors were examined by Ki-67 immunostaining and nested polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) to detect proliferating cells and the K-ras codon 12 mutation. The distribution pattern of the Ki-67-positive cells was of two types: diffuse (D), being positive throughout the crypts, and superficial (S), being positive mainly in the superficial areas of the crypts. The K-ras mutation and the D type distribution of the Ki-67-positive cells were significantly less common (P < .005) in the nonpolypoid, depressed, or elevated tumors versus the polypoid tumors. The incidence of the K-ras mutation was significantly (P < .0001) associated with the D type distribution. The K-ras mutation may involve a disorder of cell proliferation that leads to the polypoid growth of colorectal tumors.

Diagnosis of colorectal tumorous lesions by magnifying endoscopy

BACKGROUND

The magnifying colonoscope allows 100-fold magnified viewing of the colonic surface.

METHODS

We examined 2050 colorectal tumorous lesions by magnifying endoscopy, stereomicroscopy, and histopathology and classified these lesions according to pit pattern. Based on stereomicroscopy, lesions with a type 1 or 2 pit pattern were nontumors, whereas lesions with types 3s, 3L, 4, and/or 5 pit patterns were neoplastic tumors.

RESULTS

The pit patterns observed by magnifying endoscopy were fundamentally similar to those demonstrated in stereomicroscopic images. When the diagnosis by magnifying endoscopy was compared with the stereomicroscopic diagnosis, there was agreement in 1130 of 1387 lesions (81.5%). True neoplasms could be differentiated from non-neoplastic lesions. Of lesions with a type 5 pit pattern with a bounded surface, 11 of 22 (50%) were found to be invasive cancers with involvement of the submucosal layer. If this pit pattern is found to involve a relatively broad area of the mucosal surface, extensive malignant invasion (sm-massive) should be strongly suspected.

CONCLUSIONS

The magnifying colonoscope provides an accurate instantaneous assessment of the histology of colorectal tumorous lesions. This may help in decision making during colonoscopy.

Colorectal adenomas in surgical specimens from subjects with hereditary non-polyposis colorectal cancer

An understanding of the early morphogenesis of hereditary non-polyposis colorectal cancer is relevant to screening strategies. If most of these cancers were to evolve through the classical adenoma-carcinoma sequence, screening and removal of adenomas at relatively long intervals might be a safe and cost-effective approach. We have reviewed 131 cancers from 117 affected members of 34 such families. One hundred and four cancers were initial symptomatic lesions, eight were cancers detected in asymptomatic screened individuals, one was a synchronous cancer and 18 were metachronous cancers. None of the 131 cancers was a small, superficial type. Residual adenoma (contiguous with cancer) was present in three out of three (100%) in situ cancers, eight out of nine (89%) cancers involving only submucosa, four out of 14 (29%) cancers limited to the muscle coat and 13 of 105 (12%) cancers extending beyond the muscle coat. Twenty-one out of 28 (75%) residual adenomas had a villous component. Only one was flat. Of the eight asymptomatic cancers, seven arose within tubular (two) or tubulovillous adenomas (five). The eighth was not associated with an adenoma but was 35 mm in diameter and extended through the bowel wall. Discrete adenomas (contiguous excluded) were present in 22% of surgical specimens and 31% of specimens from subjects older than 50 years. A relatively high proportion (30%) had a villous component and 43% were at least 10 mm in diameter. Patients with one or more discrete adenomas in their first surgical specimen were more likely to develop multiple cancers.(ABSTRACT TRUNCATED AT 250 WORDS)

Colorectal adenoma progression and genetic change: is there a link?

Focal neoplastic change occurs frequently within the colorectum. Yet, of the several hundreds of microadenomas that are likely to be present within an individual colorectum, only one or two will develop into a clinically diagnosable adenoma. In turn, only a fraction of adenomas will progress to malignancy. The risk that a particular microadenoma will end its natural history as a carcinoma varies according to clinical context. The risk is very low in familial adenomatous polyposis (FAP), but relatively high in hereditary non-polyposis colorectal cancer (HNPCC). This variation is governed by the timing and ordering of the underlying mutational events. In FAP, inactivation of the wild-type APC gene occurs early, whereas K-ras mutations are late events. The converse appears to apply in the case of sporadic adenomas. In flat adenomas, which are known to be relatively aggressive, K-ras mutations may not occur at all. In HNPCC, mutational events are accelerated as a result of defective DNA mismatch repair. The evolution of colorectal adenoma occurs through a variety of quite distinct genetic pathways.

Frequent and characteristic K-ras activation and absence of p53 protein accumulation in aberrant crypt foci of the colon

BACKGROUND/AIMS

The relationship of aberrant crypt foci (ACF) to colorectal carcinogenesis is still controversial. Histological examination and analyses of K-ras mutations and p53 gene expression were performed to characterize ACF.

METHODS

ACF were identified microscopically in grossly normal colorectal mucosa. The ACF were separated into two pieces, one for histological and immunohistochemical examinations and the other for molecular analysis. K-ras mutations in codons 12 and 13 were analyzed by polymerase chain reaction amplification, followed by restriction fragment length polymorphism and sequencing analyses. Intranuclear p53 protein was immunostained by the avidin-biotin complex method.

RESULTS

Histologically, elongation and apical branching of the crypts in ACF were striking. K-ras mutations were detected in 58% of ACF (33 of 57; 46% [26 of 57] in codon 12, and 12% [7 of 57] in codon 13) and in 44% of adenocarcinomas (11 of 25; all in codon 12). In ACF, GAT mutations (12 of 26) were as frequent as GTT mutations (11 of 26) in codon 12, although GTT mutations in codon 12 were predominant in adenocarcinomas (10 of 11). No accumulation of p53 protein was detected in any ACF, although it was detected in 52% (13 of 25) of the colorectal carcinomas.

CONCLUSIONS

ACF do not seem histologically to be neoplasms, although genetically they are monoclonal lesions. K-ras mutation is critical in the formation of ACF, but p53 alteration could play a causal role in tumor progression.