Advertisement

Treatment policy for colonic laterally spreading tumors based on each clinicopathologic feature of 4 subtypes: actual status of pseudo-depressed type

Open AccessPublished:April 23, 2020DOI:https://doi.org/10.1016/j.gie.2020.04.033

      Background and Aims

      Laterally spreading tumors (LSTs) are originally classified into 4 subtypes. Pseudo-depressed nongranular types (LSTs-NG-PD) are gaining attention because of their high malignancy potential. Previous studies discussed the classification of nongranular (LST-NG) and granular types (LST-G); however, the actual condition or indication for endoscopic treatment of LSTs-NG-PD remains unclear. We aimed to compare the submucosal invasion pattern of LSTs-NG-PD with the other 3 subtypes.

      Methods

      A total of 22,987 colonic neoplasms including 2822 LSTs were resected endoscopically or surgically at Showa University Northern Yokohama Hospital. In these LSTs, 322 (11.4%) were submucosal invasive carcinomas. We retrospectively evaluated the clinicopathologic features of LSTs divided into 4 subtypes. In 267 LSTs resected en bloc, their submucosal invasion site was further evaluated.

      Results

      The frequency of LSTs in all colonic neoplasms was significantly higher in women (14.9%) than in men (11.0%). Rates of submucosal invasive carcinoma were .8% in the granular homogenous type (LSTs-G-H), 15.2% in the granular nodular mixed type (LSTs-G-M), 8.0% in the nongranular flat elevated type (LSTs-NG-F), and 42.5% in LSTs-NG-PD. Tumor size was associated with submucosal invasion rate in LSTs-NG-F and LSTs-NG-PD (P < .001). The multifocal invasion rate of LSTs-NG-PD (46.9%) was significantly higher than that of LSTs-G-M (7.9%) or LSTs-NG-F (11.8%). In LSTs-NG-PD, the invasion was significantly deeper (≥1000 μm) if observed in 1 site.

      Conclusions

      For LSTs-G-M and LSTs-NG-F that may have invaded the submucosa, en bloc resection could be considered. Considering that LSTs-NG-PD had a higher submucosal invasion rate, more multifocal invasive nature, and deeper invasion tendency, regardless if invasion was only observed in 1 site, than LSTs-NG-F, we should endoscopically distinguish LSTs-NG-PD from LSTs-NG-F and strictly adopt en bloc resection by endoscopic submucosal dissection or surgery for LSTs-NG-PD. (Clinical trial registration number: UMIN 000020261.)

      Graphical abstract

      Abbreviations:

      CI (confidence interval), EPMR (endoscopic piecemeal mucosal resection), ESD (endoscopic submucosal dissection), LST (laterally spreading tumor), LST-G (laterally spreading tumor-granular), LST-G-H (laterally spreading tumor-granular-homogeneous), LST-G-M (laterally spreading tumor-granular-nodular mixed), LST-NG (laterally spreading tumor-nongranular), LST-NG-F (laterally spreading tumor-nongranular-flat elevated), LST-NG-PD (laterally spreading tumor-nongranular-pseudo-depressed)
      Endoscopic treatment techniques typified by EMR and endoscopic submucosal dissection (ESD) have progressed in recent years.
      • Kudo S.
      Endoscopic mucosal resection of flat and depressed types of early colorectal cancer.
      • Tanaka S.
      • Oka S.
      • Chayama K.
      Colorectal endoscopic submucosal dissection: present status and future perspective, including its differentiation from endoscopic mucosal resection.
      • Tamegai Y.
      • Saito Y.
      • Masaki N.
      • et al.
      Endoscopic submucosal dissection: a safe technique for colorectal tumors.
      • Saito Y.
      • Uraoka T.
      • Matsuda T.
      • et al.
      Endoscopic treatment of large superficial colorectal tumors: a case series of 200 endoscopic submucosal dissections (with video).
      • Saito Y.
      • Fukuzawa M.
      • Matsuda T.
      • et al.
      Clinical outcome of endoscopic submucosal dissection versus endoscopic mucosal resection of large colorectal tumors as determined by curative resection.
      The long-term outcome of ESD, which is considered to be more radical, is also excellent.
      • Patel N.
      • Patel K.
      • Ashrafian H.
      • et al.
      Colorectal endoscopic submucosal dissection: systematic review of mid-term clinical outcomes.
      ,
      • Shigita K.
      • Oka S.
      • Tanaka S.
      • et al.
      Long-term outcomes after endoscopic submucosal dissection for superficial colorectal tumors.
      Soetikno et al
      • Soetikno R.M.
      • Kaltenbach T.
      • Rouse R.V.
      • et al.
      Prevalence of nonpolypoid (flat and depressed) colorectal neoplasms in asymptomatic and symptomatic adults.
      pointed out that nonpolypoid colorectal neoplasms were relatively common lesions diagnosed during routine colonoscopy and had a greater association with carcinoma compared with polypoid neoplasms. To avoid piecemeal EMR, ESD has often been adopted in the treatment of nonpolypoid colorectal neoplasms represented by laterally spreading tumors (LSTs), and feasible results have been reported in several studies.
      • Tanaka S.
      • Oka S.
      • Chayama K.
      Colorectal endoscopic submucosal dissection: present status and future perspective, including its differentiation from endoscopic mucosal resection.
      • Tamegai Y.
      • Saito Y.
      • Masaki N.
      • et al.
      Endoscopic submucosal dissection: a safe technique for colorectal tumors.
      • Saito Y.
      • Uraoka T.
      • Matsuda T.
      • et al.
      Endoscopic treatment of large superficial colorectal tumors: a case series of 200 endoscopic submucosal dissections (with video).
      • Saito Y.
      • Fukuzawa M.
      • Matsuda T.
      • et al.
      Clinical outcome of endoscopic submucosal dissection versus endoscopic mucosal resection of large colorectal tumors as determined by curative resection.
      • Patel N.
      • Patel K.
      • Ashrafian H.
      • et al.
      Colorectal endoscopic submucosal dissection: systematic review of mid-term clinical outcomes.
      • Shigita K.
      • Oka S.
      • Tanaka S.
      • et al.
      Long-term outcomes after endoscopic submucosal dissection for superficial colorectal tumors.
      Thus, most patients with LSTs are generally considered suitable for endoscopic resection.
      LST was originally proposed by Kudo et al
      • Kudo S.
      Endoscopic mucosal resection of flat and depressed types of early colorectal cancer.
      ,
      • Kudo S.
      Early colorectal cancer.
      • Kudo S.E.
      • Takemura O.
      • Ohtsuka K.
      Flat and depressed types of early colorectal cancers: from East to West.
      • Kudo Se
      • Lambert R.
      • Allen J.I.
      • et al.
      Nonpolypoid neoplastic lesions of the colorectal mucosa.
      and defined as flat type neoplasms that spread extensively and circumferentially along the colonic wall and were larger than 10 mm in diameter. LSTs are classified broadly into granular (LST-G) and nongranular types (LST-NG) by endoscopic morphology. The former is further subclassified into homogeneous (LST-G-H) and nodular mixed types (LST-G-M) and the latter into flat elevated (LST-NG-F) and pseudo-depressed types (LST-NG-PD) (Fig. 1; Supplementary Figs. 1-6, available online at www.giejournal.org).
      • Kudo S.
      Early colorectal cancer.
      This concept of LST is currently used worldwide, and a number of articles related to LST have been published.
      • Kudo Se
      • Lambert R.
      • Allen J.I.
      • et al.
      Nonpolypoid neoplastic lesions of the colorectal mucosa.
      • Kim B.C.
      • Chang H.J.
      • Han K.S.
      • et al.
      Clinicopathological differences of laterally spreading tumors of the colorectum according to gross appearance.
      • Zhao X.
      • Zhan Q.
      • Xiang L.
      • et al.
      Clinicopathological characteristics of laterally spreading colorectal tumor.
      • Agarwal A.
      • Garimall S.
      • Colling C.
      • et al.
      Incidence and risk factors of advanced neoplasia after endoscopic mucosal resection of colonic laterally spreading lesions.
      • Soliman H.
      • Brieau B.
      • Guillaumot M.A.
      • et al.
      Invasive pit pattern, macronodule and depression are predictive factors of submucosal invasion in colorectal laterally spreading tumours from a Western population.
      Figure thumbnail gr1
      Figure 1Endoscopic images of each subtype of laterally spreading tumor (LST). A, LST-G-H, LST granular homogeneous type. B, LST-G-M, LST granular nodular mixed type. C, LST-NG-F, LST nongranular flat elevated type. D, LST-NG-PD, LST nongranular pseudo-depressed type.
      LSTs-NG-PD have been reported to have a high malignancy potential among the 4 subtypes,
      • Zhao X.
      • Zhan Q.
      • Xiang L.
      • et al.
      Clinicopathological characteristics of laterally spreading colorectal tumor.
      ,
      • Oka S.
      • Tanaka S.
      • Kanao H.
      • et al.
      Therapeutic strategy for colorectal laterally spreading tumor.
      • Kim K.O.
      • Jang B.I.
      • Jang W.J.
      • et al.
      Laterally spreading tumors of the colorectum: clinicopathologic features and malignant potential by macroscopic morphology.
      • Myung D.S.
      • Kweon S.S.
      • Lee J.
      • et al.
      Clinicopathological features of laterally spreading colorectal tumors and their association with advanced histology and invasiveness: an experience from Honam province of South Korea: a Honam Association for the Study of intestinal Diseases (HASID).
      have clinicopathologic features similar to depressed-type (0–IIc) lesions in terms of depressed surface formation, and show a high submucosal invasion rate at small sizes.
      • Kudo Se
      • Lambert R.
      • Allen J.I.
      • et al.
      Nonpolypoid neoplastic lesions of the colorectal mucosa.
      Although Oka et al,
      • Oka S.
      • Tanaka S.
      • Kanao H.
      • et al.
      Therapeutic strategy for colorectal laterally spreading tumor.
      Kim et al,
      • Kim K.O.
      • Jang B.I.
      • Jang W.J.
      • et al.
      Laterally spreading tumors of the colorectum: clinicopathologic features and malignant potential by macroscopic morphology.
      Zhao et al,
      • Zhao X.
      • Zhan Q.
      • Xiang L.
      • et al.
      Clinicopathological characteristics of laterally spreading colorectal tumor.
      and Myung et al
      • Myung D.S.
      • Kweon S.S.
      • Lee J.
      • et al.
      Clinicopathological features of laterally spreading colorectal tumors and their association with advanced histology and invasiveness: an experience from Honam province of South Korea: a Honam Association for the Study of intestinal Diseases (HASID).
      have pathologically confirmed and reported a high submucosal invasion rate in LST-NG-PD, they did not evaluate the exact depth of submucosal invasion. Additionally, distinguishing the submucosal invasion pattern of LSTs-NG-PD from that of LSTs-NG-F has not been reported.
      Considering the aforementioned issues, a study that has pathologically confirmed the feature in detail according to the 4 LST subtypes has not been conducted yet. Thus, we aimed to elucidate the actual submucosal invasion pattern of LSTs-NG-PD comparing with other subtypes using a large number of LST specimens.

      Methods

      Patients and clinical data

      In this retrospective study, we enrolled consecutive patients with LSTs who underwent endoscopic or surgical resection at Showa University Northern Yokohama Hospital from January 2007 to June 2017. Written informed consent was obtained from all patients before initial colonoscopy, EMR, ESD, or surgery. This informed consent contained the agreement to use the clinical data or pathologic specimen for retrospective studies. Epithelial colorectal neoplasms resected by endoscopically or surgically were included in this study (Fig. 2). Advanced carcinoma (T2 or more) was excluded, and the remaining LSTs with adenoma to T1 carcinoma were evaluated in terms of clinicopathologic features in each LST subtype described below (Fig. 1). In all T1 LSTs, several lesions were excluded because of Lynch syndrome, ulcerative colitis, and inadequate or impossible pathologic evaluation because of specimen damage. The clinicopathologic features of the remaining T1 LSTs were also evaluated. Subsequently, after excluding the lesions resected by endoscopic piecemeal mucosal resection (EPMR) for accurate pathologic examination, the T1 LSTs resected en bloc were extracted as the main study subjects and then divided into 4 subtypes to evaluate each pathologic feature, especially the depth and number of submucosal invasion site.
      Figure thumbnail gr2
      Figure 2Flow diagram of study process. LST, Laterally spreading tumor LST-G-H, LST granular homogeneous type; LST-G-M, LST granular nodular mixed type; LST-NG-F, LST nongranular flat elevated type; LST-NG-PD, LST nongranular pseudo-depressed type.
      When we calculated the frequency of LSTs, all epithelial neoplastic lesions (adenoma, intramucosal carcinoma, and submucosal carcinoma) resected endoscopically or surgically in our hospital were included as subjects. The macroscopic morphology of the lesions, age and sex of the patients, tumor size, location, pit pattern, Kudo degree of invasion, depth of invasion, and treatment method were reviewed retrospectively from the electronic health record system.

      Ethical approval

      This study was approved by the institutional review board of Showa University Northern Yokohama Hospital (Approval number: 1510-14) and was registered in the University Hospital Medical Network Clinical Trials Registry (clinical trials registry number: UMIN000020261).

      Endoscopic procedure

      Bowel preparation involved the administration of 2 to 3 L of polyethylene glycol solution in the morning before the procedure. Colonoscopy was performed using a video endoscopic system (EVIS, Lucera Spectrum; Olympus Co, Tokyo, Japan) with a CF-H260AZI, PCF-Q260AZI, or CF-HQ290ZI optical magnification colonoscope (Olympus).
      We used .40% indigo carmine dye spraying of the lesion to determine LST subtype and assess the detailed shape, presence or absence of depressed area, and large nodule. Magnifying chromoendoscopy was subsequently performed in the presence of indigo carmine, which was replaced with .05% crystal violet staining as necessary. Evaluation of the pit pattern was according to the Kudo classification.
      • Kudo S.
      Endoscopic mucosal resection of flat and depressed types of early colorectal cancer.
      ,
      • Kudo Se
      • Lambert R.
      • Allen J.I.
      • et al.
      Nonpolypoid neoplastic lesions of the colorectal mucosa.
      ,
      • Kudo S.
      • Rubio C.A.
      • Teixeira C.R.
      • et al.
      Pit pattern in colorectal neoplasia: endoscopic magnifying view.
      Based on the results of the above evaluation, we made a treatment decision method (EMR, EPMR, ESD, or surgery).

      LST subtypes

      LSTs were finally classified according to the definition from the Kyoto Workshop, based on previous reports by Kudo et al.
      • Kudo S.
      Early colorectal cancer.
      • Kudo S.E.
      • Takemura O.
      • Ohtsuka K.
      Flat and depressed types of early colorectal cancers: from East to West.
      • Kudo Se
      • Lambert R.
      • Allen J.I.
      • et al.
      Nonpolypoid neoplastic lesions of the colorectal mucosa.
      LST-G-H was defined as an epithelial tumor principally developing laterally and forming a cluster of granules, and LST-G-M had 1 or more nodules on the base of the granules. LST-NG-F was an epithelial tumor principally developing laterally and not forming a cluster of granules, and LST-NG-PD had an ill-defined and groove-like appearance depression called pseudo-depressed.
      • Kudo S.
      Early colorectal cancer.
      • Kudo S.E.
      • Takemura O.
      • Ohtsuka K.
      Flat and depressed types of early colorectal cancers: from East to West.
      • Kudo Se
      • Lambert R.
      • Allen J.I.
      • et al.
      Nonpolypoid neoplastic lesions of the colorectal mucosa.
      These subtypes corresponded to the Paris classification type 0-IIa, type 0-IIa+Is or 0-Is+IIa, type 0-IIa, and type 0-IIa+IIc or 0-IIc+IIa. Three endoscopists (T.I., T.H., and H.M.) included as coauthors consensually performed the classification of LST subtypes.

      Pathologic evaluation

      All resected specimens were retrieved and immediately fixed in 10% buffered formalin solution for 24 to 48 hours and subsequently cut into parallel 2- to 3-mm-thick sections. The specimens were examined by 1 pathologist (S.H.), who is an expert on GI neoplasms, based on the World Health Organization criteria
      • Hamilton S.R.B.F.
      • Boffetta P.
      • Ilyas M.
      • et al.
      WHO classification of tumours of the digestive system
      Carcinoma of the colon and rectum.
      and the Japanese Society for Cancer of the Colon and Rectum (JSCCR).
      • Watanabe T.
      • Itabashi M.
      • Shimada Y.
      • et al.
      Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines 2010 for the treatment of colorectal cancer.
      Pathologic examinations were initially performed using hematoxylin and eosin staining, and desmin staining (Dako North America Inc, Carpinteria, Calif, USA) was added for specimens that were supposed to show submucosal invasion. The vertical depth of submucosal invasion was measured according to the Japanese Society for Cancer of the Colon and Rectum guidelines. The degree of submucosal invasion was categorized according to the Kudo classification (Fig. 3) into 3 stages based on the vertical depth of invasion: sm1 (into the upper third of the submucosal layer), sm2 (middle third), and sm3 (lower third) in surgically resected specimens. In endoscopically resected specimens, the resected submucosal layer was vertically divided into 2. Submucosal invasion within the upper layer was considered as sm1, deeper invasion as sm2, and a positive vertical margin as sm3. Sm1 was subclassified into sm1a, sm1b, and sm1c based on the ratio of the horizontal extent of the submucosal invasion. Sm1a referred to a ratio of less than .25; sm1b, ratio of between .25 and .5; and sm1c, ratio greater than .5.
      • Kudo S.
      Endoscopic mucosal resection of flat and depressed types of early colorectal cancer.
      In the present study, sm1a or sm1b lesions were defined as slightly invasive and sm1c, sm2, and sm3 lesions as massively invasive.
      Figure thumbnail gr3
      Figure 3Kudo classification of the degree of submucosal invasion according to its vertical and horizontal extension. sm1a, Carcinoma invasion limited to the upper third of the submucosal layer and the ratio of B to A is less than 1/4; sm1b, the ratio of B to A is between 1/4 and 1/2; sm1c, the ratio of B to A is greater than 1/2; sm2, carcinoma invasion extending to the middle third of the layer; sm3, carcinoma invasion extending up to the bottom layer.

      Evaluation of the submucosal invasion site

      These evaluations were conducted retrospectively to include 267 specimens by 1 pathologist (S.H.) and 1 endoscopist (T.I.) comparing the pathologic findings and endoscopic images. We mainly focused on the number of submucosal invasion sites: 1 site invasion or multifocal invasion.
      For evaluation of submucosal invasion site, we used hematoxylin and eosin–stained and desmin-immunostained sections. Multifocal submucosal invasion was defined as the existence of 2 or more invasion sites that interposed through the muscularis mucosae into the submucosa. Particular attention was provided to avoid double counting of the submucosal invasion sites between adjacent sections. Three typical examples are shown in Figure 4.
      Figure thumbnail gr4ai
      Figure 4Cases and schematic diagrams of submucosal invasion style. A, Laterally spreading tumor granular nodular mixed type in the sigmoid colon. B, Crystal violet staining. VI low-grade pit pattern. C, Resected specimen after endoscopic submucosal dissection (ESD). A split was made on the nodule (yellow line). D and E, Submucosal invasion was observed at 1 site on the maximal nodule, as pathologically assessed using H&E staining and desmin immunostaining. F, Schema of submucosal invasion site in this case. G, Laterally spreading tumor nongranular flat elevated type in the sigmoid colon. H, Crystal violet staining. VI low-grade pit pattern. I, Resected specimen after ESD. A split was made on the yellow line. J and K, Submucosal invasion was observed at 1 site, as pathologically assessed using H&E staining and desmin immunostaining. L, Schema of submucosal invasion site in this case. M, Laterally spread tumor nongranular pseudo-depressed type in the transverse colon. N, Crystal violet staining. VI low-grade pit pattern. O, Resected specimen after ESD. A split was made on the yellow line. P and Q, Submucosal invasion was observed at 4 sites (multifocally), as pathologically assessed using HE staining and desmin immunostaining. R, Schema of submucosal invasion site in this case.
      Figure thumbnail gr4jr
      Figure 4Cases and schematic diagrams of submucosal invasion style. A, Laterally spreading tumor granular nodular mixed type in the sigmoid colon. B, Crystal violet staining. VI low-grade pit pattern. C, Resected specimen after endoscopic submucosal dissection (ESD). A split was made on the nodule (yellow line). D and E, Submucosal invasion was observed at 1 site on the maximal nodule, as pathologically assessed using H&E staining and desmin immunostaining. F, Schema of submucosal invasion site in this case. G, Laterally spreading tumor nongranular flat elevated type in the sigmoid colon. H, Crystal violet staining. VI low-grade pit pattern. I, Resected specimen after ESD. A split was made on the yellow line. J and K, Submucosal invasion was observed at 1 site, as pathologically assessed using H&E staining and desmin immunostaining. L, Schema of submucosal invasion site in this case. M, Laterally spread tumor nongranular pseudo-depressed type in the transverse colon. N, Crystal violet staining. VI low-grade pit pattern. O, Resected specimen after ESD. A split was made on the yellow line. P and Q, Submucosal invasion was observed at 4 sites (multifocally), as pathologically assessed using HE staining and desmin immunostaining. R, Schema of submucosal invasion site in this case.

      Reproducibility study in LST-NG

      Although the LSTs were divided into 4 subtypes mentioned above, LSTs-NG-F and LSTs-NG-PD seemed difficult to distinguish. Therefore, we conducted a pilot reproducibility study of endoscopic morphology. The kappa statistic was used to calculate inter- and intraobserver agreement using 30 endoscopic images of LSTs-NG-F or LSTs-NG-PD comparing reproducibility according to the degree of endoscopy experience (specifically between K.M. and Y.M. [both with 3 years of experience in GI endoscopy] as a comparison of beginners and also between M.M. and K.W. [with 15 and 18 years of experience in GI endoscopy, respectively] as a comparison of experts). A kappa of 0 was considered as “poor” agreement; .00 to .20, “slight” agreement; .21 to .40, “fair” agreement; .41 to .60, “moderate” agreement; .61 to .80, “substantial” agreement; and .80 to 1.00, “almost perfect” agreement.

      Statistical analyses

      In the analysis of LSTs, the Fisher exact test or χ2 test was used in the comparison of frequency, location, treatment method, T1 carcinoma rate, and submucosal invasion site, and P < .05 was considered statistically significant. In the comparison among the 4 subtypes of LSTs, Bonferroni correction was performed, and P < .008 (.05/6) was considered statistically significant. One-way analysis of variance with Bonferroni correction was used in the comparison of LST size, and P < .008 was considered statistically significant. The Cochran-Armitage test for trend was used to examine the association between the size of LSTs and submucosal invasion.

      Results

      A total of 22,987 colonic neoplasms including 2822 LSTs were continuously resected endoscopically or surgically. In these 2822 LSTs, 322 (11.4%) were submucosal invasive carcinomas. Of these 322 T1 LSTs, Lynch syndrome (n = 4), ulcerative colitis (n = 1), and inadequate or impossible pathologic evaluation (n = 6) were excluded. Among the remaining 311 T1 LSTs, 44 lesions resected by EPMR were further excluded. Finally, 267 T1 LSTs resected en bloc were extracted as the main study subjects and were divided into 4 subtypes shown in Figure 2.

      Clinicopathologic characteristics of LSTs and treatment method

      The frequency of LSTs in all epithelial colon tumors including advanced carcinomas treated in our hospital was 12.3% (2822/22,987) (Table 1). The frequency was significantly higher in women (14.9% [1132/7579]) than in men (11.0% [1690/15,408]) (P < .001). Of 2822 LSTs, 1766 (62.6%) were located in the right colon segment. Moreover, 187 of 640 LSTs-G-M (29.2%) were localized in the rectum, and the proportion in the rectum was significantly higher than that of the other subtypes (vs LST-G-H, 14.5%, P < .001; vs LST-NG-F, 5.2%, P < .001; vs LST-NG-PD, 6.7%, P < .001). The mean size (35 mm in diameter) of LSTs-G-M was significantly larger than that of the other subtypes (vs LST-G-H, 22 mm, P < .001; vs LST-NG-F, 20 mm, P < .001; vs LST-NG-PD, 20 mm, P < .001). The submucosal invasive (T1 carcinoma) rates and submucosal massive invasive (≥1000 μm) rates of LST-G-H, LST-G-M, LST-NG-F, and LST-NG-PD were .8% (4/518), 15.2% (97/640), 8.0% (113/1410), and 42.5% (108/254) and .4% (2/518), 12.0% (77/640), 4.1% (58/1410), and 20.5% (52/254), respectively. The Kudo degree of submucosal invasion and the depth of invasion showed a similar tendency. The T1 carcinoma rate and submucosal massive invasive rate of LSTs-NG-PD were significantly higher than those of the other 3 subtypes (P < .001).
      Table 1Overview of LSTs from January 2007 to June 2017
      LST-G-H

      (n = 518)
      LST-G-M

      (n = 640)
      LST-NG-F

      (n = 1410)
      LST-NG-PD

      (n = 254)
      Total

      (n = 2822)
      Mean age, y (range)69.5 (54-85)65.5 (42-93)70.0 (44-93)70.5 (40-94)69.5 (40-94)
      Sex
       Male246 (47.5)345 (53.9)912 (64.7)187 (73.6)1690 (59.9)
       Female272 (52.5)295 (46.1)498 (35.3)67 (26.4)1132 (40.1)
      Frequency of LSTs in neoplastic lesions
       All (neoplastic lesions, 22,987)518 (2.3)640 (2.8)1410 (6.1)254 (1.1)2822 (12.3)
       Male (neoplastic lesions, 15,408)246 (1.6)345 (2.2)912 (5.9)187 (1.2)1690 (11.0)
       Female (neoplastic lesions, 7579)272 (3.6)295 (3.9)498 (6.6)67 (.9)1132 (14.9)
      Location
       Right-sided colon369 (71.2)268 (41.9)966 (68.5)163 (64.2)1766 (62.6)
       Left-sided colon74 (14.3)185 (28.9)371 (26.3)74 (29.1)704 (24.9)
       Rectum75 (14.5)187 (29.2)73 (5.2)17 (6.7)352 (12.5)
      Mean size, mm (range)22 (10-163)35 (10-187)20 (10-91)20 (10-65)20 (10-187)
      Pathology
      Adenoma or intramucosal carcinoma514 (99.2)543 (84.8)1297 (92.0)146 (57.5)2500 (88.6)
       Submucosal carcinoma4 (.8)97 (15.2)113 (8.0)108 (42.5)322 (11.4)
      Massive invasion
      Kudo’s degree of invasion (sm1c/sm2/sm3)2 (.4)75 (11.7)66 (4.7)59 (23.2)202 (7.2)
      Depth of invasion (≥1000 μm)2 (.4)77 (12.0)58 (4.1)52 (20.5)189 (6.7)
      Treatment method
       EMR203 (39.2)85 (13.3)621 (44.0)51 (20.1)960 (34.0)
       EPMR141 (27.2)138 (21.6)227 (16.1)30 (11.8)536 (19.0)
       ESD164 (31.7)397 (62.0)547 (38.8)151 (59.4)1259 (44.6)
       Surgery10 (1.9)20 (3.1)15 (1.1)22 (8.7)67 (2.4)
      Values are n (%) unless otherwise defined.
      LST, Laterally spreading tumor; LST-G-H, LST granular homogeneous type; LST-G-M, LST granular nodular mixed type; LST-NG-F, LST nongranular flat elevated type; LST-NG-PD, LST nongranular pseudo-depressed type, right-sided (colon, cecum, ascending and transverse colon), left-sided (colon, descending and sigmoid colon); neoplastic lesions, epithelial colorectal tumor (adenoma, intramucosal carcinoma, and submucosal carcinoma) resected at our hospital; EPMR, endoscopic piecemeal mucosal resection; ESD, endoscopic submucosal dissection.
      Table 2 shows the association between the size and rate of T1 carcinoma in 322 T1 LSTs. LSTs-G-H still had low T1 carcinoma rate regardless of tumor size. The T1 carcinoma rate of LSTs-G-M did not increase as the tumor size increased (P = .199), whereas that of LSTs-NG-F and LSTs-NG-PD increased (P < .001). LSTs-NG-PD 10 to 19 mm in diameter and LSTs-NG-PD ≥20 mm in diameter showed a 26.4% occurrence and a 54.1% (80/148) occurrence of submucosal invasion, respectively.
      Table 2Association between the size and rate of submucosal invasion in each laterally spreading tumor subtype
      10-1920-29Size (mm)

      30-39
      40-49≥50P valueTotal
      LST-G-H0/185 (0)1/153 (.7)1/83 (1.2)1/42 (2.4)1/55 (1.8).060
      Cochran-Armitage test for trend.
      4/518 (.8)
      LST-G-M4/60 (6.7)19/161 (11.8)31/148 (20.9)20/103 (19.4)23/168 (13.7).199
      Cochran-Armitage test for trend.
      97/640 (15.2)
      LST-NG-F28/683 (4.1)40/499 (8.0)26/148 (17.6)11/53 (20.8)8/27 (29.6)<.001
      Cochran-Armitage test for trend.
      113/1410 (8.0)
      LST-NG-PD28/106 (26.4)52/97 (53.6)22/37 (59.5)4/10 (40.0)2/4 (50.0)<.001
      Cochran-Armitage test for trend.
      108/254 (42.5)
      Values are n/N (%).
      LST, Laterally spreading tumor; LST-G-H, LST granular homogeneous type; LST-G-M, LST granular nodular mixed type; LST-NG-F, LST nongranular flat elevated type; LST-NG-PD, LST nongranular pseudo-depressed type.
      Cochran-Armitage test for trend.
      Table 3 shows the clinicopathologic features and treatment methods of 311 T1 LSTs (11 LSTs excluded). Seventy-five of 95 T1 LSTs-G-M (78.9%) showed an invasion depth ≥1000 μm. Focusing on the Kudo degree of submucosal invasion, the most lesions exhibiting sm1a invasion were observed among submucosal slight invasive carcinomas. Considering that LSTs by definition have a large width, more LSTs were considered to be sm1a among submucosal slight invasion carcinomas according to the Kudo degree of submucosal invasion focusing on the horizontal extent. T1 LSTs-NG-PD were treated by ESD (65.4%) or surgery (18.3%).
      Table 3Clinicopathologic features and treatment methods of T1 LSTs
      LST-G-H

      (n = 3)
      LST-G-M

      (n = 95)
      LST-NG-F

      (n = 109)
      LST-NG-PD

      (n = 104)
      Total

      (n = 311)
      Mean tumor size, mm (range), mm39 (20-70)36 (10-187)26 (10-65)22 (10-62)28 (10-187)
      Location
       Right-sided colon0 (0)29 (30.5)62 (56.9)62 (59.6)156 (50.2)
       Left-sided colon3 (100)29 (30.5)32 (29.4)32 (29.4)93 (29.9)
       Rectum0 (0)37 (38.9)15 (13.8)15 (13.8)62 (19.9)
      Pit pattern
       VN/VI high-grade1 (33.3)40 (42.1)28 (25.7)35 (33.7)104 (33.4)
       Others2 (66.7)55 (57.9)81 (74.3)69 (66.3)207 (66.6)
      Kudo degree of invasion
       sm1a1 (33.3)21 (22.1)44 (40.4)46 (44.2)112 (36.0)
       sm1b0 (0)1 (1.1)2 (1.8)0 (0)3 (1.0)
       sm1c0 (0)1 (1.1)0 (0)3 (2.9)4 (1.3)
       sm22 (66.7)64 (67.4)52 (47.7)48 (46.2)166 (53.4)
       sm30 (0)8 (8.4)11 (10.1)7 (6.7)26 (8.4)
      Depth of invasion
       <1000 μm1 (33.3)20 (21.1)54 (49.5)52 (50.0)127 (40.8)
       ≥1000 μm2 (66.7)75 (78.9)55 (50.5)52 (50.0)184 (59.2)
      Treatment method
       EMR0 (0)7 (7.4)18 (16.5)9 (8.7)34 (10.9)
       EPMR1 (33.3)19 (20.0)16 (14.7)8 (7.7)44 (14.1)
       ESD0 (0)58 (61.1)66 (60.6)68 (65.4)192 (61.7)
       Surgery2 (66.7)11 (11.6)9 (8.3)19 (18.3)41 (13.2)
      En bloc resection rate2 (66.7)76 (80.0)93 (85.3)96 (92.3)267 (85.6)
      Values are n (%) unless otherwise defined.
      LST, Laterally spreading tumor; LST-G-H, LST granular homogeneous type; LST-G-M, LST granular nodular mixed type; LST-NG-F, LST nongranular flat elevated type; LST-NG-PD, LST nongranular pseudo-depressed type, right-sided (colon, cecum, ascending and transverse colon), left-sided colon (descending and sigmoid colon); EPMR, endoscopic piecemeal mucosal resection; ESD, endoscopic submucosal dissection.

      Difference in submucosal invasion pattern among the 4 subtypes of T1 LSTs

      Table 4 shows the number of submucosal invasion site in each T1 LST subtype. Seventy of 76 T1 LSTs-G-M (92.1%) invaded the submucosa at 1 site. Of these, in 63 of 70 lesions (82.9%), the invasion site was found only under the maximal nodule. Moreover, of these 63 lesions, 7 (9.2%) were found only outside under the maximal nodule and 6 (7.9%) were found multifocally (2 or more sites of invasion). In T1 LSTs-NG-F, 82 of 93 lesions (88.2%) showed only 1 site invasion, and 11 of 93 lesions (11.8%) invaded the submucosa multifocally. In LSTs-NG-PD, of the 96 lesions, 51 (53.1%) invaded only 1 site and 45 (46.9%) showed multifocal invasion. The rate of multifocal invasion in LSTs-NG-PD was significantly higher than that in LSTs-NG-F (P < .001) or LSTs-G-M (P < .001).
      Table 4Number of submucosal invasion sites in each T1 LST subtype
      LST-G-H (n = 2)LST-G-M (n = 76)LST-NG-F (n = 93)LST-NG-PD (n = 96)Total (n = 267)
      No. of submucosal invasion sites (%)
      one2 (100)70 (92.1)82 (88.2)51 (53.1)205 (76.8)
      Two or more (multifocal invasion)0 (0)6 (7.9)11 (11.8)45 (46.9)62 (23.2)
      a, P = 1; b, P = .450; c, P < .001; d, P = 1; e, P < .001; f, P = .498, Fisher exact test with Bonferroni correction.
      Table 5 shows the association between the size and rate of multifocal invasion in each T1 LST subtype. Only LSTs-NG-PD showed a tendency to increase the risk of multifocal invasion as tumor size increased (P = .004).
      Table 5Association between the size and rate of multifocal invasion in each T1 LST subtype
      10-1920-29Size (mm)

      30-39
      40-49≥50P value
      Cochran-Armitage test for trend.
      Total
      LST-G-H0/0 (0)0/0 (0)0/1 (0)0/0 (0)0/1 (0)0/2 (0)
      LST-G-M0/3 (0)1/13 (7.7)1/24 (4.2)1/17 (5.9)3/19 (15.8).2496/76 (7.9)
      LST-NG-F1/20 (5.0)5/34 (14.7)3/23 (13.0)2/10 (20.0)0/6 (0).58411/93 (11.8)
      LST-NG-PD5/25 (20.0)23/43 (53.5)13/22 (59.1)3/4 (75.0)1/2 (50.0).00445/96 (46.9)
      Values are n/N (%).
      LST, Laterally spreading tumor; LST-G-H, LST granular homogeneous type; LST-G-M, LST granular nodular mixed type; LST-NG-F, LST nongranular flat elevated type; LST-NG-PD, LST nongranular pseudo-depressed type.
      Cochran-Armitage test for trend.
      Table 6 shows the comparison of depth and number of submucosal invasion sites between LSTs-NG-F and LSTs-NG-PD. Thirty-four of 51 T1 LSTs-NG-PD (66.7%) with 1 site invasion showed ≥1000 μm invasion depth, whereas 29 of 45 T1 LSTs-NG-PD (66.4%) with multifocal invasion showed <1000 μm invasion depth. In LSTs-NG-PD, the invasion was significantly deeper (≥1000 μm) if it was 1 site (P = .002) and significantly shallower (<1000 μm) if it was multifocal. In contrast to LSTs-NG-PD, no statistically significant difference was observed in LSTs-NG-F (P = .248).
      Table 6Comparison of depth and number of submucosal invasion site between LST-NG-F and LST-NG-PD
      T1 LST-NGSubmucosal invasion siteP value
      One siteTwo or more sites
      LST-NG-F82 (88.2)11 (11.8)
       <1000 μm37 (45.1)7 (63.6).248
      Fisher exact test.
       ≥1000 μm45 (54.9)4 (36.4)
      LST-NG-PD51 (53.1)45 (46.1)
       <1000 μm17 (33.3)29 (64.4).002
      Fisher exact test.
       ≥1000 μm34 (66.7)16 (35.6)
      Values are n (%).
      LST, Laterally spreading tumor; LST-NG-F, LST nongranular flat elevated type; LST-NG-PD, LST nongranular pseudo-depressed type.
      Fisher exact test.

      Reliability for endoscopic distinction between LSTs-NG-F and LSTs-NG-PD

      Among the beginner endoscopists, the kappa value of the interobserver agreement was .87 (95% confidence interval [CI], .69-1.00) and of the intraobserver agreement .87 (95% CI, .69-1.00) for K.M. and 1.00 (95% CI, 1) for Y.M. Among the experts, the kappa value of interobserver agreement was .93 (95% CI, .81-1.00) and of intraobserver agreement 1.00 (95% CI, 1) for M.M. and 1.00 (95% CI, 1) for K.W. Therefore, almost perfect agreements were recognized not only among the experts but also among the beginners. Based on the above results, it can be concluded that the endoscopic reproducibility of LST-NG-F or LST-NG-PD has been guaranteed.

      Discussion

      In this study, we elucidated the clinicopathologic features of colonic LSTs divided into 4 subtypes using a large number of LST specimens treated at our hospital. We particularly focused on LSTs-NG-PD, which were proven to harbor high submucosal invasion rate even in smaller sizes and to show multifocal submucosal invasion.
      The frequency of LSTs in whole epithelial colorectal tumors excluding advanced carcinomas treated in our hospital was 12.3% and was significantly higher in women (1132/7579 [14.9%]) than in men (1690/15,408 [11.0%]). Kaku et al
      • Kaku E.
      • Oda Y.
      • Murakami Y.
      • et al.
      Proportion of flat- and depressed-type and laterally spreading tumor among advanced colorectal neoplasia.
      reported that the frequency of LST was 1.9% (62/3269); however, their sample size was limited, and there was few information about frequency. Therefore, examining the frequency of LSTs using large samples in this study was considered significantly important. The reasons for a higher proportion of LSTs in women could be because of women’s genetic background, presence of intestinal flora in women, or women’s hormonal involvement. Moreover, it might be a single-institution trend bias. Hence, investigating the actual situation at other healthcare facilities is required in the future. Regarding localization, LSTs were found more frequently in the right side of the colon (62.6%) in this study. Consistent with our previous studies, 47% to 79% of LSTs were located in the right side of the colon.
      • Kim B.C.
      • Chang H.J.
      • Han K.S.
      • et al.
      Clinicopathological differences of laterally spreading tumors of the colorectum according to gross appearance.
      ,
      • Soliman H.
      • Brieau B.
      • Guillaumot M.A.
      • et al.
      Invasive pit pattern, macronodule and depression are predictive factors of submucosal invasion in colorectal laterally spreading tumours from a Western population.
      ,
      • Kim K.O.
      • Jang B.I.
      • Jang W.J.
      • et al.
      Laterally spreading tumors of the colorectum: clinicopathologic features and malignant potential by macroscopic morphology.
      ,
      • Kaku E.
      • Oda Y.
      • Murakami Y.
      • et al.
      Proportion of flat- and depressed-type and laterally spreading tumor among advanced colorectal neoplasia.
      ,
      • Uraoka T.
      • Saito Y.
      • Matsuda T.
      • et al.
      Endoscopic indications for endoscopic mucosal resection of laterally spreading tumours in the colorectum.
      Examination of the localization by subclassification showed that LSTs-G-M were found more frequently in the rectum compared with other types: 29.2% of the whole LSTs-G-M and 38.9% of the T1 carcinomas of LSTs-G-M localized in the rectum.
      Among the T1 carcinomas of LST, LSTs-G-M showed a higher rate of ≥1000 μm (78.9%) or a massive invasion (76.8%) based on the Kudo classification. LSTs-G-M were suggested to have a deeper invasion tendency if they invaded the submucosal layer. Although T1 carcinomas of LSTs-G-M were considered to invade under their maximal nodule conventionally, 17.1% (13/76) of the T1 carcinomas of LSTs-G-M showed invasion outside under their maximal nodule or multifocal invasion based on our data. Imai et al
      • Imai K.
      • Hotta K.
      • Yamaguchi Y.
      • et al.
      Should laterally spreading tumors granular type be resected en bloc in endoscopic resections?.
      also examined the submucosal invasion pattern of LSTs-G-M and reported that the invasion under the maximal nodule was 69% (34/49) and the invasion outside under the maximal nodule or multifocal invasion was 31% (15/49). Therefore, LSTs-G-M should be resected en bloc as far as possible to allow a firm pathologic assessment.
      The submucosal invasion rate of LSTs-NG-PD was 42.5r% (108/254). It was 54.1% (80/148) when they became 20 mm or more in this study (Table 2). Similarly, the submucosal invasion rates of LSTs-NG-PD were 42.1% (16/38) by Oka et al,
      • Oka S.
      • Tanaka S.
      • Kanao H.
      • et al.
      Therapeutic strategy for colorectal laterally spreading tumor.
      36.4% (8/23) by Kim et al,
      • Kim K.O.
      • Jang B.I.
      • Jang W.J.
      • et al.
      Laterally spreading tumors of the colorectum: clinicopathologic features and malignant potential by macroscopic morphology.
      and 11.7% (4/34) by Myung et al.
      • Myung D.S.
      • Kweon S.S.
      • Lee J.
      • et al.
      Clinicopathological features of laterally spreading colorectal tumors and their association with advanced histology and invasiveness: an experience from Honam province of South Korea: a Honam Association for the Study of intestinal Diseases (HASID).
      Considering the lower submucosal invasion rate of LSTs-NG-PD at 40 to 49 mm and ≥50 mm than 20 to 39 mm, most LSTs-NG-PD would become T2 carcinomas when their size reached 40 mm or more, and the lesions, which are less likely to be T2 carcinoma genetically and have more lateral growth tendency, remain at 40 mm or more. The submucosal invasion rate of LSTs-NG-PD was significantly higher than that of LSTs-NG-F (8.0%) (Table 1), and the multifocal invasion rate of LSTs-NG-PD was also significantly higher than that of LSTs-NG-F (46.9% vs 11.8%) (Tables 4 and 5). Although Uraoka et al
      • Uraoka T.
      • Saito Y.
      • Matsuda T.
      • et al.
      Endoscopic indications for endoscopic mucosal resection of laterally spreading tumours in the colorectum.
      and Yamada et al
      • Yamada M.
      • Saito Y.
      • Sakamoto T.
      • et al.
      Endoscopic predictors of deep submucosal invasion in colorectal laterally spreading tumors.
      reported that the multifocal invasion rates of LSTs-NG are 25% (8/32) and 45% (72/159), respectively, this study is the first to reveal that the multifocal invasion rate of LSTs-NG-PD was significantly higher than that of LSTs-NG-F, indicating the importance of distinguishing them within LSTs-NG. Furthermore, most T1 LSTs-NG-PD (66.7%) with 1 site invasion showed an invasion depth ≥1000 μm, and 64.4% of the T1 LSTs-NG-PD with multifocal invasion showed a depth <1000 μm. Statistically significant differences were observed (P = .002) in T1 LSTs-NG-PD. However, no statistically significant differences were observed in T1 LSTs-NG-F (P = .248) (Table 6). Therefore, compared with LSTs-NG-F, LSTs-NG-PD tended to invade deeply at 1 site but showed shallow multifocal invasion. This result has never been reported and should be significantly considered when determining the indication of endoscopic treatment and when performing pathologic evaluation. As described above, LSTs-NG-PD have a higher submucosal invasion rate and tend to invade more multifocally than LSTs-NG-F. Hence, LSTs-NG-PD could have a more malignant nature than LSTs-NG-F. Distinguishing them endoscopically is important. The endoscopic identification of LSTs-NG-F and LSTs-NG-PD is also guaranteed by the validation test in this study.
      Based on the above considerations, the basic endoscopic treatment policy for each LST subtype would be described as follows. Almost all LSTs-G-H do not invade the submucosa. Therefore, EMR and EPMR (not ESD or surgery) could be adapted. Even if they invade the submucosal layer, pathologic evaluation is possible if they are snared to prevent the division of the submucosal invasion site. Most T1 carcinomas of the LSTs-G-M invaded the submucosa under the maximal nodule, and 17.1% of them showed invasion outside under their maximal nodule or multifocal invasion. The mean size (35 mm in diameter) of LSTs-G-M was significantly larger than that of the other subtypes; hence, LSTs-G-M are often considered to be difficult to resect en bloc by EMR. For these reasons, easy EMR or EPMR dividing the submucosal invasion site should be avoided, and en bloc resection by ESD could be adapted to achieve accurate pathologic diagnoses. Moreover, because LSTs-G-M were most frequently localized in the rectum, large LSTs-G-M in the rectum could be resected through ESD but not through initial surgery, considering the possibility of colostomy or several postoperative adverse events. LSTs-NG-F did not show a higher submucosal invasion rate or multifocal invasion rate. If they invaded the submucosa, almost all invasions were at 1 site. Therefore, several LSTs-NG-F could be cured by endoscopic treatment, such as EMR. However, because the tumor size was associated with an increased possibility of submucosal invasion and when LSTs-NG-F were greater than 30 mm in diameter, their submucosal invasion rate increased to 17.6%. Thus, en bloc resection by ESD could also be adopted if they are difficult to remove en bloc by EMR. LSTs-NG-PD had the highest T1 carcinoma rate, the highest multifocal invasion rate, and the most invasive nature, regardless if invasion was only observed in 1 site, among the 4 LST subtypes. Therefore, for LSTs-NG-PD, en bloc resection by ESD should always be adopted if they are difficult to resect en bloc by EMR. However, because LSTs-NG-PD greater than 20 mm in diameter might have already massively invaded the submucosal layer, it should be kept in mind that an initial surgery with lymph node dissection is necessary beyond the indications for endoscopic treatments.
      This study has some limitations. First, this was a single-institution investigation and retrospectively designed study. Consequently, a regional or institutional selection bias could exist. Second, T1 lesions resected by EPMR were excluded from the subjects of examination of submucosal invasion pattern to exclude inaccurate pathologic evaluation of invasion site. This could have resulted in selection bias in favor of aggressive lesions selected for en bloc endoscopic resection or surgical resection. Third, contrast evaluations based on magnifying chromoendoscopic finding (pit pattern) corresponding to each pathologic finding (submucosal invasion site) were not performed because the pit pattern on the submucosal invasion site of some lesions was not confirmed from the endoscopic finding information stored in the electronic health record system.
      In conclusion, we elucidated the clinicopathologic features of colonic LSTs divided into 4 subtypes. Each subtype had different and unique features, which requires a decision on treatment policy to suit each feature. In particular, although LSTs-NG-F and LSTs-NG-PD are considered as 1 subtype (LSTs-NG), we first reported that LSTs-NG-PD had a higher submucosal invasion rate even in small sizes, a more multifocal invasive nature, and a deeper invasion tendency, regardless if invasion was only observed in 1 site, than the LSTs-NG-F. Additionally, endoscopic identification is guaranteed with repeatability. Therefore, en bloc resection mainly by ESD for LSTs-NG-PD should always be adapted to allow a firm pathologic assessment. Moreover, LSTs-NG-PD greater than 20 mm in diameter often invaded massively; hence, an initial surgery might be necessary.

      Acknowledgments

      We express our deepest appreciation to Chiaki Nishimura for his advice on statistical analysis, to Nobunao Ikehara for his beneficial suggestion, and to all members of the Digestive Disease Center and the Department of Pathology, Showa University Northern Yokohama Hospital for their excellent assistance.

      References

        • Kudo S.
        Endoscopic mucosal resection of flat and depressed types of early colorectal cancer.
        Endoscopy. 1993; 25: 455-461
        • Tanaka S.
        • Oka S.
        • Chayama K.
        Colorectal endoscopic submucosal dissection: present status and future perspective, including its differentiation from endoscopic mucosal resection.
        J Gastroenterol. 2008; 43: 641-651
        • Tamegai Y.
        • Saito Y.
        • Masaki N.
        • et al.
        Endoscopic submucosal dissection: a safe technique for colorectal tumors.
        Endoscopy. 2007; 39: 418-422
        • Saito Y.
        • Uraoka T.
        • Matsuda T.
        • et al.
        Endoscopic treatment of large superficial colorectal tumors: a case series of 200 endoscopic submucosal dissections (with video).
        Gastrointest Endosc. 2007; 66: 966-973
        • Saito Y.
        • Fukuzawa M.
        • Matsuda T.
        • et al.
        Clinical outcome of endoscopic submucosal dissection versus endoscopic mucosal resection of large colorectal tumors as determined by curative resection.
        Surg Endosc. 2010; 24: 343-352
        • Patel N.
        • Patel K.
        • Ashrafian H.
        • et al.
        Colorectal endoscopic submucosal dissection: systematic review of mid-term clinical outcomes.
        Dig Endosc. 2016; 28: 405-416
        • Shigita K.
        • Oka S.
        • Tanaka S.
        • et al.
        Long-term outcomes after endoscopic submucosal dissection for superficial colorectal tumors.
        Gastrointest Endosc. 2017; 85: 546-553
        • Soetikno R.M.
        • Kaltenbach T.
        • Rouse R.V.
        • et al.
        Prevalence of nonpolypoid (flat and depressed) colorectal neoplasms in asymptomatic and symptomatic adults.
        JAMA. 2008; 299: 1027-1035
        • Kudo S.
        Early colorectal cancer.
        Igaku-Shoin, Tokyo, Japan1996
        • Kudo S.E.
        • Takemura O.
        • Ohtsuka K.
        Flat and depressed types of early colorectal cancers: from East to West.
        Gastrointest Endosc Clin North Am. 2008; 18 (xi): 581-593
        • Kudo Se
        • Lambert R.
        • Allen J.I.
        • et al.
        Nonpolypoid neoplastic lesions of the colorectal mucosa.
        Gastrointest Endosc. 2008; 68: S3-47
        • Kim B.C.
        • Chang H.J.
        • Han K.S.
        • et al.
        Clinicopathological differences of laterally spreading tumors of the colorectum according to gross appearance.
        Endoscopy. 2011; 43: 100-107
        • Zhao X.
        • Zhan Q.
        • Xiang L.
        • et al.
        Clinicopathological characteristics of laterally spreading colorectal tumor.
        Plos One. 2014; 9e94552
        • Agarwal A.
        • Garimall S.
        • Colling C.
        • et al.
        Incidence and risk factors of advanced neoplasia after endoscopic mucosal resection of colonic laterally spreading lesions.
        Int J Colorectal Dis. 2018; 33: 1333-1340
        • Soliman H.
        • Brieau B.
        • Guillaumot M.A.
        • et al.
        Invasive pit pattern, macronodule and depression are predictive factors of submucosal invasion in colorectal laterally spreading tumours from a Western population.
        United Eur Gastroenterol J. 2018; 6: 1569-1577
        • Oka S.
        • Tanaka S.
        • Kanao H.
        • et al.
        Therapeutic strategy for colorectal laterally spreading tumor.
        Dig Endosc. 2009; 21: S43-S46
        • Kim K.O.
        • Jang B.I.
        • Jang W.J.
        • et al.
        Laterally spreading tumors of the colorectum: clinicopathologic features and malignant potential by macroscopic morphology.
        Int J Colorectal Dis. 2013; 28: 1661-1666
        • Myung D.S.
        • Kweon S.S.
        • Lee J.
        • et al.
        Clinicopathological features of laterally spreading colorectal tumors and their association with advanced histology and invasiveness: an experience from Honam province of South Korea: a Honam Association for the Study of intestinal Diseases (HASID).
        Plos One. 2017; 12e0184205
        • Kudo S.
        • Rubio C.A.
        • Teixeira C.R.
        • et al.
        Pit pattern in colorectal neoplasia: endoscopic magnifying view.
        Endoscopy. 2001; 33: 367-373
        • Hamilton S.R.B.F.
        • Boffetta P.
        • Ilyas M.
        • et al.
        • WHO classification of tumours of the digestive system
        Carcinoma of the colon and rectum.
        International Agency for Research on Cancer, Lyon, France2010: 134-146
        • Watanabe T.
        • Itabashi M.
        • Shimada Y.
        • et al.
        Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines 2010 for the treatment of colorectal cancer.
        Int J Clin Oncol. 2012; 17: 1-29
        • Kaku E.
        • Oda Y.
        • Murakami Y.
        • et al.
        Proportion of flat- and depressed-type and laterally spreading tumor among advanced colorectal neoplasia.
        Clin Gastroenterol Hepatol. 2011; 9: 503-508
        • Uraoka T.
        • Saito Y.
        • Matsuda T.
        • et al.
        Endoscopic indications for endoscopic mucosal resection of laterally spreading tumours in the colorectum.
        Gut. 2006; 55: 1592-1597
        • Imai K.
        • Hotta K.
        • Yamaguchi Y.
        • et al.
        Should laterally spreading tumors granular type be resected en bloc in endoscopic resections?.
        Surg Endosc. 2014; 28: 2167-2173
        • Yamada M.
        • Saito Y.
        • Sakamoto T.
        • et al.
        Endoscopic predictors of deep submucosal invasion in colorectal laterally spreading tumors.
        Endoscopy. 2016; 48: 456-464

      Linked Article