Su1547 Clinicopathological Features of Laterally Spreading Tumors in the Colorectum-Large Scale Study

Flat and depressed neoplasms typified as laterally spreading tumors (LSTs) are considered one of the risk factors for the development of advanced colorectal cancer, and knowledge of their clinicopathological features is of great importance. LSTs have been morphologically described two distinct subtypes-granular type(LST-G) and non-granular type (LST-NG). Some reports showed these pathological characteristics based on the differences of these subtypes; however, only a small number of the studies have been conducted showing clinical features of LSTs. The aim of this study was to investigate the clinicopathological features including body mass index (BMI), past history, frequency of synchronous and metachronous tumor of large number of LSTs. In a retrospective review of the colonoscopy prospectively completed database (February 1998 - October 2009) at the National Cancer Center Hospital, we selected cases of colorectal neoplasms based on the following criteria: endoscopic diagnosis of LSTs with subsequent endoscopic or surgical resection and observation by chromoendoscopy using indigocarmine dye spraying to clarify LST subtypes. Familial adenomatous polyposis (FAP), inflammatory bowel disease (IBD), and lesions diagnosed as residual or recurrent tumors were excluded from this analysis. We assessed patients' backgrounds, and endoscopic and pathological findings in their medical records and evaluated the following factors: gender, age, BMI, past history of cancer, serum level of HbA1C. We also compared the following characteristics of the tumor: location, depth of invasion, size, and frequency of synchronous and metachronous tumors defined as LSTs and/or advanved colorectal cancers. A total of 393 LSTs were included in the analysis, of which 187 (48%) were LST-NG and 206 (52%) LST-NG. The rate of male patients was higher with LST-NG [64% (120/187) vs 51%(106/206); P=0.011]. Patients with LST-NG were older than those with LST-G (67.2 vs 64.8, P=0.016). The rate of LST-NG in the rectum is significantly lower than that of LST-G [6% (12/187) vs 32% (65/206); P<0.001]. The LST-NG had a significantly higher frequency of submucosal invasion compared to LST-G [35% (122/187) vs 18% (168/206); P<0.001] but the mean tumor size of LST-NG (30mm) was smaller than that of LST-G (48mm, P<0.001). The detection rate of synchronous and/or metachronous tumor was 9.1% (17/187) vs 8.7% (18/206) (P=0.902)), between LST-G and LST-NG but there was no significant difference. There were also no significant differences of other factors. The most distinct clinical feature observed was the localization of LST-G and LST-NG. Despite the smaller lesion size, the rate of submucosal invasion of LST-NG was significantly higher than that of LST-G. LSTs might have a higher incidence of synchronous and/or metachronous tumors.