Small "flat adenoma" of the large bowel with special reference to its clinicopathologic features

Phenotypic variation in colorectal adenoma/cancer expression in two families. Hereditary flat adenoma syndrome

Clinical, pathologic, and genetic studies on two colorectal cancer-prone families have disclosed right-sided colonic flat adenomas and colorectal cancer. Adenomatous polyp counts exceeded those found in hereditary nonpolyposis colorectal cancer (HNPCC) but were fewer than in familial adenomatous polyposis (FAP). Colon cancer occurred at a later age than in HNPCC or FAP and showed right-sided predominance. The older age of patients with colonic cancer, the right-sided predominance of colon cancer, and the paucity of rectal adenomas make FAP unlikely. Vertical transmission of polyps and colon cancer fit the pattern of autosomal dominant inheritance. A characteristic feature of this phenotype is the predominance of flat adenomas. Molecular genetic studies, with careful description of phenotype, should help clarify classification.

Colorectal adenomas in the Lynch syndromes. Results of a colonoscopy screening program

Forty-four asymptomatic putative Lynch syndrome patients participated in a colonoscopy screening program. There were 18 men and 26 women; mean age was 44 yr. Thirty percent of Lynch syndrome patients had at least one adenoma; 20% had multiple adenomas. In 18% of the patients, adenomas were discovered proximal to the splenic flexure. In a reference group of 88 age- and sex-matched patients, 11% had adenomas, 4% had multiple adenomas, and 1% had right-sided adenomas. Twenty-one Lynch syndrome patients had follow-up colonoscopies. Of 7 patients with adenomas on initial examinations, 6 had adenomas at follow-up. Of 14 patients with negative initial examination results, 3 had adenomas at follow-up. The prevalence of adenomas in the Lynch syndromes is greater than in an unselected reference group. The adenomas are more proximally located, corresponding to the site of cancer distribution in the Lynch syndromes. A high rate of synchronous and metachronous lesions is found. Our findings support the hypothesis that adenomatous changes are the premalignant lesion in the Lynch syndromes. We also found evidence of heterogeneity among Lynch syndrome families in adenoma incidence.

Endoscopic diagnosis of small flat carcinoma of the colon. Report of three cases

Most small carcinomas arise from polyps. Small lesions with cellular features of malignancy and early invasion, but with no histologic evidence of residual adenoma, are rare. Diagnosed by endoscopy, three such lesions are described. They were recognized as mucosal plaques, measuring between 6 and 8 mm in diameter. In each case, there was either synchronous or metachronous carcinoma elsewhere in the colon, as well as benign adenomatous polyps. Colonoscopic identification of such lesions allows inclusion of that bowel segment in any planned resection.

The diagnostic value of combining flexible sigmoidoscopy and double-contrast barium enema as a one-stage procedure

Results from the consecutive examination of 675 patients using both flexible sigmoidoscopy and a double-contrast x-ray technique were analyzed with special reference to the detection of polyps in the rectum and sigmoid colon. A total of 193 polyps were found. Histological examination of 93 polyps revealed that half of those less than 5 mm in diameter and 93.3% of those more than 6 mm in diameter were adenomas. The x-ray examination failed to detect 44% of the proven adenomas smaller than 5 mm, 35.3% of those 6-10 mm in size, and 16.7% of those larger than 11 mm in diameter. These rates were significantly higher than those of flexible sigmoidoscopy, which had corresponding miss rates of 8, 11.2, and 0%, respectively. The double-contrast barium enema (DCBE) failed to detect every second polyp in the rectosigmoid. Every second polyp in the same region proved to be adenoma. The DCBE combined with flexible sigmoidoscopy gives the most reliable and precise diagnosis of various disorders of the rectum and sigmoid colon.

Fiberoptic sigmoidoscopy as the first screening procedure for colorectal neoplasms in an asymptomatic population

As the first screening for colorectal neoplasms, a total of 2243 examinations by fiberoptic sigmoidoscopy (FS) were performed on 1573 asymptomatic patients. One hundred seventy neoplasms were found in 134 patients (6 percent of total examinations), including nine adenocarcinomas (0.4 percent). A flat, intramucosal cancer and a small, 5-mm cancer detected by FS could not be demonstrated by barium enema examination. On further investigation, additional adenomas were identified in the proximal colon in 22 percent of patients in whom neoplasms had first been detected by FS and in only 4.4 percent of those in whom FS findings were negative but other criteria were positive. It is concluded that FS is the procedure of choice for the first screening of asymptomatic patients, not only because it enables detection of flat and small lesions within its observation range, but also because it could lead to detection of proximal neoplasms.

Do all colorectal carcinomas arise in preexisting adenomas?

In this article, the underlying nature of the adenoma is defined and evidence in support of the adenoma-carcinoma sequence is presented. Emphasis is placed on recent findings from the field of molecular biology which fortify the concept of the adenoma as a step in the evolution of large bowel cancer. Despite the considerable body of indirect evidence in support of the adenoma-carcinoma sequence, the de novo theory still has some support. Although differences between the de novo and adenoma carcinoma concept are largely semantic, the fact that infiltrating carcinoma can arise within small, flat foci of severely dysplastic epithelium (indistinguishable from carcinoma in situ) may be of considerable clinical importance. Other routes of morphogenesis exist also, but these make a relatively small contribution to the overall incidence of colorectal cancer. Even if it is accepted that the majority of colorectal cancers develop in a preexisting adenoma, the systematic removal of adenomas would pose a number of practical difficulties. First, adenomas are extremely common, particularly in populations at high risk of colorectal cancer. A second factor is the recently recognized existence of flat or depressed adenomas which may be small and difficult to visualize endoscopically. A simple screening program based on our knowledge of the evolution of colorectal cancer is outlined.

The possibility of nonpolypoid carcinogenesis in the large intestine as inferred from frequencies of DNA aneuploidy of polypoid and crater-shaped carcinomas

DNA ploidy patterns were studied by cytofluorometry in 60 cases of crater-shaped invasive carcinoma and 30 cases of polypoid tumor (severe dysplasia or submucosally invasive carcinoma located in the pedunculated, semipedunculated, or sessile polyp) of the large intestine. The data were compared with 20 cases of intramucosal differentiated (or intestinal type) adenocarcinoma of the stomach, with special reference to their macroscopic shape and frequency of DNA aneuploidy. DNA aneuploidy was found in 77% of the crater-shaped carcinomas and in 17% of the polypoid tumors of the large intestine. The frequencies were significantly different and the frequency gap amounted to 60%. However, 18 of 20 (90%) gastric adenocarcinomas were nonpolypoid in shape, whereas two (10%) were polypoid. DNA aneuploidy was found in 50% of the gastric adenocarcinomas and 56% of the nonpolypoid gastric adenocarcinomas. This value did not differ from the values reported previously for the submucosally invasive and advanced crater-shaped intestinal type adenocarcinomas of the stomach. Biologic characteristics of adenocarcinomas of the large intestine can be compared with those of intestinal type adenocarcinomas of the stomach, because intestinal type gastric adenocarcinoma is surrounded mostly by intestinalized mucosa and considered to arise from the epithelium under induction or progression of intestinal differentiation. Therefore, we inferred that the frequency gap in DNA aneuploidy between the crater-shaped and polypoid tumors of the large intestine implies that in the large intestine approximately 60% of the crater-shaped invasive carcinomas develop from the small nonpolypoid carcinomas.

Early malignant lesions of the colorectum at autopsy

Examination of 1014 consecutive autopsies revealed four early malignant lesions, comprising: 1) a carcinoma in situ arising from a large (2.5 cm) pedunculated adenomatous polyp; 2) a carcinoma in situ arising from a small (0.8 cm) flat adenoma; 3) an early invasive carcinoma arising from a flat (2.5 cm) adenoma, and 4) an early invasive polypoid adenocarcinoma (0.7 cm) with no identifiable remnants of adenoma. The early malignant lesions encountered in this study reaffirm the importance of the adenoma-cancer sequence in the pathogenesis of colorectal cancers in man. The malignant potential of flat adenomas is emphasized. The occurrence of small carcinomas without evidence of adenomatous elements raises the possibility of de novo origin as an alternative pathway. In the present study, one of four early colorectal cancers may have a de novo origin.

Flat adenoma and flat mucosal carcinoma (IIb type)--a new precursor of colorectal carcinoma? Report of two cases

Two flat adenomas and a flat mucosal carcinoma of the colon were reported in patients with synchronous and metachronous colonic carcinomas. These lesions were almost flat and were not detected by preoperative endoscopic examinations. Colonoscopists should be aware of the presence of flat adenomas, which can be easily missed, and recognize them as lesions that play an important role in the "adenoma-carcinoma sequence."

Epidemiology of polyps in the rectum and colon. Recovery and evaluation of unresected polyps 2 years after detection

In an endoscopic population screening study for colorectal polyps among 200 men and 200 women, 50-59 years of age, 215 polyps less than 5 mm in diameter were left in situ for the present 2-year follow-up examination. The attendance rate was 102 of 106 (96%) for polyp patients and 77 of 90 (86%) in the control group. Of 194 polyps, 143 (74%) in the 102 polyp-bearing individuals were recovered for histological evaluation and 57 polyps were registered as new. Ninety-nine (50%) of the polyps were hyperplastic, 45 (23%) were adenomas, and 45 (23%) were mucosal tags. Both growth and regression of polyps were registered. Regression was commoner in the distal part of the rectum than in the proximal part or distal sigmoid colon. Growth was similar for recovered adenomas and hyperplastic polyps, whereas mucosal tags more often showed diminution in size. No polyp had reached a size of more than 5 mm in 2 years, and no case of severe dysplasia or carcinoma was registered. The estimated total polyp mass more than doubled both for adenomas and hyperplastic polyps. It is concluded that the time interval between initial examination with removal of polyps 5 mm or larger in diameter and the first follow-up examination may safely be set at 2 years.