The combination of magnifying colonoscopy and dye spraying is helpful in determining the nature of colonic lesions as non-neoplastic, adenomas, or invasive carcinomas. Therefore it may be possible to determine, at the time of colonoscopy, which lesions require no treatment, which can be removed endoscopically, and which should be removed by surgery.

Background-There is a diVerence in the location of colorectal mucosal lesions and invasive cancers. Aims-To ascertain whether the location of colorectal neoplasms reflects the carcinogenesis pathway. Methods-The subject material consisted of 4147 neoplastic lesions that had been resected endoscopically or surgically from 5025 patients. Mucosal lesions and submucosal cancers were classified into depressed and non-depressed types endoscopically or histologically. The relations between macroscopic type, size, histology, and location were investigated. Conclusion-There may be some mechanisms that promote the progression of mucosal lesions to invasive cancers in the left colon and rectum, whereas a de novo pathway from depressed type lesions may be implicated in some cancers of the right colon. (Gut 1999;45:818-821) Results-(a)

Although, there is a substantial difference in the overall diagnostic accuracies, it is not statistically significant. Therefore, we conclude that magnifying colonoscopy is at least as accurate as EUS for preoperative staging of early CRCs.

Background: Recent Japanese studies have shown that histogenesis of small colorectal carcinomas can be divided into two groups: polypoid growth arising from polypoid neoplasia, and non polypoid growth arising from flat or depressed neoplasia. This classification should be verified with genetic as well as morphologic characteristics. Subjects and Methods: In order to classify our subject into polypoid growth and non polypoid growth types both histologically and endoscopically, we selected 42 colorectal carcinomas <2 em in size (35 submucosal and seven more advanced). C,linicppathological findings, presence or absence of Ki-ras gene mutation and overexpression of p53 protein were compared between the two types. Results: Histologically, the cases were divided into 27 of the polypoid growth type and 15 of the nonpolypoid growth type. None of the non polypoid growth cases contained adenomatous remnant, whereas this was found in 75% of the polypoid growth cases. No Ki-ras mutation was observed in any of the nonpolypoid growth cases, although it appeared in 440/0 of the polypoid growth cases. Regarding the overexpression of p53 protein, no significant difference was observed between the two types. The histological and the colonoscopic polypoid growth-nonpolypoid growth classifications correlated well with each other(agreement rate 98%), except for one lesion, which was classified as polypoid growth type endoscopically but as nonpolypoid growth type histologically. Conclusions: The histologically defined polypoid growth-nonpolypoid growth classificationmayindicate a difference inpathway of colorectal carcinogenesis. Also, colonoscopic polypoid growth-nonpolypoid growth classification is available for preoperative estimation of the genetic characteristics of small carcinomas.Key words: carcinogenesis -nonpolypoid growm -Ki-ras mutation -overexpression ofp53protein -endoscopy INTRODUCTIONRecently, Vogelstein and his co-workers (l,2) reported an interesting scenario of genetic alterations involving colorectal carcinogenesis, suggesting that Ki-ras mutation occurs early in tumorigenesis, while p53 mutation occurs later. However, this work was based on the adenoma-carcinoma sequence theory Received July 9, 1997; accepted November 18, 1997 For reprints and all correspondence: Takahiro Fujii, Department of Medicine, National Cancer Center Hospital East, 6-5-1, Kashiwanoha, Kashiwa, Chiba 277, Japan Abbreviations: PG, polypoid growth; NPG, nonpolypoid growth; EMR, endoscopic mucosal resection; FAP, familial adenomatous polyposis; HNPCC, hereditary nonpolyposis colorectal cancer; SM-C, submucosal carcinoma; Adv-C, advanced carcinoma; PCR-SSCP, polymerase chain reaction single-strand conformation polymorphism obtained from polypoid adenomas to invasive carcinoma, and may not be applicable to flat or depressed tumors (3,4). Yamagata et al. (5) reported that flat adenomas manifested a lower incidence of the Ki-ras codon 12 mutation than did polypoid adenomas, and suggested that the above difference may indicate that the colorect...

We macroscopically classified 25 gastric and 23 colorectal advanced cancers into “contracted” and “uncontracted” types, and found immunohistochemically that integrin subunit α3 was more frequently expressed in the extracellular matrix (ECM) in the former than in the latter (75%:9/12 vs. 38%: 5/13 in gastric and 86%:6/7 vs. 25%:4/16 in colorectal cancers, respectively). Integrin subunit α3 was also expressed more frequently in cancers producing transforming growth factor‐beta (TGF‐β), which is related to ECM deposition, integrin expression and cell mobility, than in those which did not produce TGF‐β (67%:10/15 vs. 40%:4/10 in gastric and 57%:4/7 vs. 38%:6/16 in colorectal cancers, respectively). In addition, integrin subunit α3 was not expressed in 2 benign gastric ulcers combined with gastric cancer elsewhere in the stomach. On the other hand, a retrospective analysis of 107 cases of rectal cancer which recurred after a curative operation revealed that local recurrence was more frequent in “contracted” than “uncontracted” types (44%:ll/25 vs. 26%:21/82). These results may suggest that the abundant interstitial fibrosis which leads to remarkable gastric or colorectal wall contraction is a result of the interaction between cancer cells and ECM, along with the expression of integrin and/or the production of TGF‐β, This fibrosis may also be closely related to the mode of gastric and colorectal cancer invasion.