Clinicopathological Characteristics And Treatment Strategies Of Laterally Spreading Colorectal Tumors

Background and AimsColorectal cancer is one of most common neoplasms in the world, i.e., it is the second most common cancer diagnosed in women and the third most common in men, accounting for one fourth of worldwide cancer deaths in2008. Although precise etiology remains to be defined, an unhealthy life style has been shown to contribute to colorectal carcinogenesis, such as a diet high in fat and red meat, or low in fiber, fruits and vegetables, and is associated with physical inactivity. Due to the rapid adoption of Westernized lifestyles, there is a rapid increase in colorectal cancer incidence in Asia-Pacific regions, including China. To date, endoscopy is frequently used for the early detection of colorectal cancer to effectively prevent late stage cancers. However, in recent years, nonpolypoid colorectal cancer has drawn much attention. Laterally spreading tumors (LSTs) of the colorectum is a large and relatively flat neoplastic lesion, which typically extends laterally rather than vertically along the colonic wall and belongs to the class nonpolypoid colorectal neoplasia. LST is defined as a lesion greater than10mm in diameter. Recent studies have indicated that LSTs represents17.2%of advanced colorectal neoplasia. LSTs can also develop into a deeper submucosal invasive cancer. Thus, the use of colonoscopy for LST diagnosis is crucial to effectively prevent colorectal cancer. LSTs can be treated with endoscopic resection (such as endoscopic mucosal resection, piecemeal endoscopic mucosal resection, or endoscopic submucosal dissection) or surgery to eliminate the late stage of colorectal cancer.However, diagnosis and subtypes of LSTs are complicated because LSTs grow along the surface of the intestine with a low vertical axis that extends laterally along the luminal wall. According to their morphological features, LSTs can be classified into2types and4subtypes, i.e., granular type (LST-G), including homogeneous G-type and nodular mixed G-type; non-granular type (LST-NG), including flat elevated NG-type and a pseudo-depressed NG-type. Previous studies have suggested distinct genetic characteristics exist between LST-G and LST-NG types.and that there are distinct incidences for the development of submucosal invasive carcinoma among these four subtypes. Furthermore, previous studies have shown distinct clinicopathological LSTs characteristics (for example, morphology, colorectal localization, histopathological feature, and incidence of submucosal invasion) in different populations.In China, because of their flat morphology and low awareness among colonoscopists, many flat adenomas are missed during colonoscopy although they are often visible. Therefore, flat adenomas are not only difficult to detect, but also easy to miss. This may be why only a limited number of studies of the four LSTs subtypes are available.LSTs subtypes are morecommonly associated with histopathology of lesions. Traditionally, the term "adenoma"has included tubular adenoma, tubulevillousadenoma, and villous adenoma. However, another distinct form of adenoma has recently been described,"serrated adenoma"-distinguishedfrom traditional adenoma. Most previous studieson LSTs included exclusively "traditional" adenomas. A few studies investigating LSTs have included serrated adenomas.There are Various differences exist between the proximal and distal colon, that could predispose the tumours originating at these sites to develop along different pathways. We speculate that there are difference between the proximal colon and the distal colon with regarding to LSTso In this study, we retrospectively collected LSTs case histories and investigated their clinicopathological characteristics in a Chinese population. Materials and methodsStudy populationThe study population included38,050(male:female was21291:16759) consecutive patients who underwent complete colonoscopy in The Endoscopy Center of Nanfang Hospital between January2001and August2011.Among this cohort of patients, a total of259LST patients were found. The written informed consents were obtained from all the patients before endoscopic procedure. For those patients with multiple lesions, we randomly selected one lesion per multiple lesion patient. Moreover,20cases were excluded due to lack of histopathological data; thus, a total of239lesions from239patients were included for statistical analysis.1. Comparison of the four LSTs subtypesThe LST weredivided into LST-G-H (homogeneous type), LST-G-MIX (nodular mixed type), LST-NG-F (flat type) and LST-NG-PD (pseudodepressed type) type lesions. Clinical and histopathological parameters were compared among the four subtypes.Binary logistic regression analysis was used to analyze risk factors of development of LSTs to submucosal carcinoma.2.Histopathological evaluation of LST lesionsHistopathology include traditional adenomas and serrated adenomas, the endoscopic feature and advanced histology of each of these pathology types were investigated. We used Binary logistic regression analysis investigate the associations between clinical variables and the presence of advanced histology in LSTs.3. The comparsion of LSTs lesions between proximal colon and distal colonThe colon were divided into proximal colon (caecum,ascending colon, transverse colon)and distal colon (descending colon, sigmoid,rectum) relative to the splenic flexure. We investigated clinicopathological characteristics of LSTs between proximal colon and distal colon.Furthermore, the endoscopic characteristics and pathological characteristics of the same LSTs Subtype were compared between proximal colon and distal colon. 4. Endoscopic resections in laterally spreading tumors granular typeWe classified the tumors as LST-G-Hor LST-G-NM subtypes.We analyzed clinicopathological characteristics and submucosal invasion rates for both subtypes, and we determined the incidence of submucosal invasions associated with each subtype.Statistical analysesQuantitative data were expressed as mean±standard deviation and analyzed using the Student’s t-test or one-way analysis of variance. Qualitative data were expressed using ratios and analyzed by the Pearson chi-square test or Fisher’s exact test. Kruskal-wallis H test was used for data with nonparametrics. Binary logistic regression analysis was used to analyze risk factors of development of LSTs to advanced histology lesions. All data were analyzed using SPSS17.0for Windows (SPSS, Chicago, IL, USA) and a p value<0.05was considered to be statistically significant.Results1.Comparison of the four LSTs subtypesSubtype evaluation revealed239lesions of LSTs, including62(25.9%) cases of homogeneouss G-type,99(41.4%) cases of nodular mixed G-type,71(29.7%) cases of flat elevated NG-type,7(2.9%) cases of pseudo-depressed NG-type. Of these239lesions,86(36.0%) lesions were HGIN, and24(10.0%) lesions were submucosal invasive carcinoma. Furthermore, lesions of nodular mixed G-type has an average size of43.0±22.1mm, which was much larger than that of the other three subtypes (P=0.001). However, there was no statistically significant difference between size of flat elevated NG-type and pseudo-depressed NG-type lesions (P=0.705). Moreover,29(46.8%) lesions of homogeneous G-type and38(53.5%) lesions of flat elevated NG-type were localized at the proximal colon.83(83.8%) lesions of nodular mixed G-type and6(85.7%) lesions of pseudo-depressed NG-type were localized at the distal colon. The size of homogeneous G-type LSTs and flat elevated NG-type lesions with advanced histology were markedly larger than those of low-grade and non-cancerous lesions. However, this was not observed in the nodular mixed G-type lesions. Moreover, if the size of flat elevated NG-type lesions was greater than20mm and the size of homogeneous G-type was greater than30mm, the risk of advanced histology was markedly increased. Noticeably, no diffidence was detected between the lesions of homogeneous G-type and flat elevated NG-type with regard to the histopathological features. LST-NG-PD was good independent predictors of submucosal carcinoma.2. Histopathological evaluation of LST lesionsHistopathological evaluation revealed239lesions of LSTs,194cases of traditional adenomas, including42(17.6%) cases of tubular adenoma,86(36.0%) cases of tubular villous adenoma,45(18.8%) cases of villous adenomas;42(17.6%) cases of serrated adenomas. Of these239lesions,101(42.3%) lesions were LGIN,86(36.0%) lesions were HGIN, and24(10.0%) lesions were submucosal invasive carcinoma. There were distinct endoscopic LST characteristics (for example, morphology, colorectal localization, size) between serrated adenomas-LSTs and traditional adenomas-LSTs. The incidence of advanced histology in serrated adenomas-LSTs was found to be lower than in traditional adenomas-LSTs (21.4%vs44.5%, P=0.006). However, there was no statistically significant difference in canceration rate between serrated adenomas-LSTs and traditional adenomas-LSTs (6.7%vs.10.8%,P=0.404):In the multivariate analysis, Malignancy potential of histology was associated with nodular mixed G-type [OR=2.7,95%CI (1.3-5.7); P=0.009], flat elevated NG-type [OR=2.7,95%CI (1.2-6.2); P=0.017],LST-NG-PD[OR=6.6,95%CI (1.1-40.9); P=0.043], Distal colon[OR=2.6,95%CI (1.3-4.9); P=0.005], Tumor size≥30mm [OR=2.7,95%CI (1.4-5.1); P=0.004].However, age and gender were irrelevant.3. The comparsion of LSTs lesions between proximal colon and distal colonThe average size of the LSTs lesions at the proximal colon was significantly smaller than in the distal colon (37.8±22.6mm vs.22.2±11.2mm,P<0.001). The incidence of developing advanced histology of the LSTs lesions at the distal colon was significantly higher in the proximal colon(56.8%vs.26.2%,P=0.001).There is no difference found between the proximal colon and the distal colon with regard to incidence of submucosal invasion. The average size of these four LSTs subtypes at the distal colon were significantly bigger than in the promixal colon (LST-G-H:38.0±24.1mm vs.24.5+12.4mm, P=0.005;LST-G-NM:45.6±22.7mmvs.29.4±12.3mm, P=0.007; LST-NG-F:21.8±10.5mmvs.17.4±7.3mm,P=0.014;). The incidence of homogeneous G-type LSTs and flat elevated NG-type lesions with advanced histology in distal colon were markedly higher than those in the promixal (LST-G-H:42.4%vs.10.3%,P=0.005; LST-NG-F:54.5%vs.26.3%,P=0.015) However, this was not observed in the nodular mixed G-type Iesions62.7%vs.50%,P=0.343). In proximal colon,a binary logistic regression analysis revealed that lesion subtype, size≥30mm were all independent predictors of advanced histology in proximal colon. However, only size of LSTs lesions≥30mm was an independent predictor of advanced histology, the lesion subtype were irrelevant.4. Endoscopic resections in laterally spreading tumors granular typeWe evaluated the histopathological data from62LST-G-Hand99LST-G-MX lesions with diameters of10-19mm (14.9%),20-29mm (23.6%),30-39mm (21.7%),40-49mm (9.9%),or≥50mm (29.8%).Submucosal invasions were observed in2(3.2%) LST-G-H and16(14.1%) LST-G-NM lesions. Dameter of LST-G-H lesions with submucosal invasion were≥60mm. In LST-G-MX lesions, the submucosal invasion incidences (within a tumor-size category) were as follows:25%(10-19mm),9.1%(20-29mm),7.4%(30-39mm), and23.1%(40-49mm) or18.2%(≥50mm),respectively.The submucosal invasion incidences were irrelevant with lesion size.Conclusion1. Chinese LSTs can be divided into four different subtypes, which show distinct clinicopathological characteristics. Noticeably, no difference was detected between the LST-G-H and LST-NG-F lesions regarding Clinicopathological characteristics (except diameter). LST-NG-PD was good independent predictors of submucosal carcinoma. 2. There were distinct LST clinicopathological characteristics between serrated adenomas-LSTs and traditional adenomas-LSTs. Serrated adenoma-LSTs has a lower potential for the development of malignancy than traditional adenomas-LSTs. Morphology, location, and tumor size≥30mm all independent predictors of advanced histology.3. In terms of LSTs, there were distinct clinicopathological characteristics between proximal colon and distal colon.Furthermore, The same LSTs Subtype, there were distinct in endoscopic characteristics or pathological characteristics between proximal colon and distal colon.4. Our results indicated that, for LST-G-H lesions, EPMR(endoscopic piecemeal resections) would be acceptable due to the low risk of submucosal invasion. For LST-G-NM lesions, particularly those with diameters≥20mm, en bloc removal in an endoscopic resection(ESD) is preferable for sufficient histological evaluation.