"Lateral Spreading Tumors" (LSTs) are dysplastic lesions whose protrusion within the lumens the colon is not more than twice as compared to the surrounding non-dysplastic mucosa.

They can be divided into two groups:

Granular type (LST-G) and Non Granular type (LST-NG) Endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) are currently the most used techniques to resect this type of lesions. Compared to other methods of tissue ablation, EMR allows to carry out the histological evaluation of the resected fragments and ESD of the lesion in toto ("en bloc") EMR is currently the most used technique for removal of LST, but for lesions of ≥ 30 mm the resection is performed "piecemeal", i.e. fragmentary. This can compromise an adequate histological evaluation of the lateral and deep margins of the lesion.

Colonic EMR (EMR-S) is usually performed using a polypectomy snare, after lifting the lesion from the underlying layers with a submucosal injection of liquid (EMR standard or "inject-and-cut"). The aspiration of the lesion inside a plastic cap preloaded on the tip of the colonoscope ("cap-assisted EMR" - EMR-C) is almost exclusively used for the treatment of gastric and esophageal lesions. Its use for lesions of the colon and duodenum has been reported in limited experiences The principal aim of this study is to evaluate the efficacy and the safety of the EMR-C for the removal of large colonic LST-G and LST-NG, comparing it with EMR-S.

Colorectal carcinoma (CRC) is the second cause of death for cancer in industrialized countries, with annual incidence and mortality of about one million and 500.000 case respectively.

It's well known that most of CRC follow the path adenoma-carcinoma: early diagnosis and endoscopic removal of colonic polyps has been proved to be useful in preventing cancer.

Most of colorectal polyps are smaller than 1 cm and can be successfully resected with a standard polypectomy. However, between 0.8% and 5% of patients develop sessile polyps or lesions larger than 20 mm, of which removal can be difficult, requiring high endoscopic experience.

Recent prospective studies report that 7%-36% of CRC have a flat or depressed morphology and are more likely to infiltrate the submucosa compared with polypoid ones.

A univariate analysis has proved that the size of the lesion is the only significant risk factor associated with malignant evolution.

Contrary to sessile polyps (SP) that are protruding lesions without a peduncle and whose base has almost the same dimension of the head, "Lateral Spreading Tumors" (LSTs) are dysplastic lesions whose protrusion within the lumen is not more than twice as compared to the surrounding non-dysplastic mucosa. According to Kudo classification they are larger than 1 cm in size, slightly elevated and extending laterally along the intestinal wall.

They can be divided into two groups (according to Paris Classification, 2005, updated for the colon in Kyoto Classification 2008):

• Granular type (LST-G) characterized by nodular aggregates and sub-classified into homogeneous (0-IIa according to Paris Classification) and mixed nodular (0-IIa, 0-Is + IIa, 0-II+ Is) subtypes.
• Non Granular type (LST-NG) characterized by a non nodular surface and sub-classified into elevated (0-IIa) and pseudo-depressed (0-IIa + 0-IIc, 0-IIc +0- IIa) subtypes.

The risk of developing cancer is different between the two types (57.7% in LST-NG vs 32.7% in LST-G). LST-NG are more likely to invade the submucosa compared to LST-G (14% vs 7%). Within the LST-G group, lesions with a mixed nodular morphology have a greater tendency to infiltrate the submucosa compared to the homogeneous ones.

Endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) are currently the most used techniques to resect this type of lesions. Compared to other methods of tissue ablation, EMR allows to carry out the histological evaluation of the resected fragments and ESD of the lesion in toto ("en bloc").

EMR allows the resection of superficial neoplasia of gastro-intestinal tract (GI) confined to the mucosa, in the absence of vascular and/or lymphatic invasion.

ESD compared with EMR allows to remove "en bloc" lesions ≥20 mm in size. It should be preferred for lesions with higher risk of invasiveness or when the removal of the deepest layers or of the whole submucosa is desired, despite the size of the lesion. However, ESD is a complex procedure which requires a long training period and it is associated with higher risk of perforation compared with EMR (6.2% vs 1.3%). Furthermore, ESD requires a longer execution time.

Therefore, EMR is currently the most used technique for removal of LST, but for lesions of ≥ 30 mm the resection is performed "piecemeal", i.e. fragmentary. This can compromise an adequate histological evaluation of the lateral and deep margins of the lesion.

Piecemeal resection increases the risk of residual disease that ranges from 12% to 20% compared with 5% described after "en bloc" removal while the percentage of recurrence reported for polypoid lesions ≥ 20 mm is on average 25% (21) and reaches 55% in some studies.

Colonic EMR is usually performed using a polypectomy snare, after lifting the lesion from the underlying layers with a submucosal injection of liquid (EMR standard or "inject-and-cut"). The aspiration of the lesion inside a plastic cap preloaded on the tip of the colonoscope ("cap-assisted EMR" - EMR-C) is almost exclusively used for the treatment of gastric and esophageal lesions. Its use for lesions of the colon and duodenum has been reported in limited experiences.

The advantage of diagnostic "cap-assisted colonoscopy" (CAC) is the higher chance of reaching cecum even by less experienced endoscopists in a shorter time, with less pain for the patients and a better observation of the mucosa behind the folds and at the flexures. There are not enough concordant data about the percentage of missing lesions, especially if small in size (27, 28). The cap makes the position of the instrument more stable during standard "inject-and-cut" technique (EMR-S), and reduces execution time. However, the realization of the EMR-C for colonic lesions isn't reported (29).

The use of EMR-C in colon is controversial because of the risk of entrapping the muscular layer in the polypectomy snare with risk of perforation.

The advantage of using the cap is represented by the possibility to perform mucosectomy of lesions located in difficult positions (between haustra, near or involving the ileo-caecal valve), thanks to the improved visibility on the operative field.

Our group has reported a 4% of residual disease/recurrence rate, much lower than those reported by other authors who performed EMR-S. We had a perforation and bleeding rate of 0% and 7% respectively vs 0.4% and 11% as reported in literature with EMR-S.

More recently, a study of 134 lesions treated with EMR-C reported a recurrence rate of 1.8% on 82 lesions treated, with a mean of 4.2 months follow-up.

The principal aim of this study is to evaluate the efficacy and the safety of the EMR-C for the removal of large colonic granular and non-granular Lateral Spreading Tumors (LST-G, LST-NG), comparing it with EMR-S.

Patients with colorectal LST-G/NG ≥30 mm will be included. Patients who refuse endoscopic follow up will be excluded from the study. The total enrollment period will be 6 months Endoscopic evaluation in patients without invasive carcinoma will be performed at 3, 6 and 12 months, and then annually Follow-up period will last 12 months from the enrollment of the last patient.

Will be defined as:

Residual lesion: the presence of adenomatous tissue endoscopically visible at follow-up colonoscopies within the first year from EMR.

Recurrent lesion: the presence of adenomatous tissue endoscopically visible after 2 (at 3 and 6 months from EMR) previous negative colonoscopies.

Inclusion Criteria:

• Patients ≥18 years Colonic LST-G/NG () ≥ 30 mm in size. Patients able to undergo all follow up procedures as indicated in the protocol and to provide written informed consent, at least 48 hours before the intervention (reasoned consent).

Exclusion Criteria:

• Presence of sessile polyps Non polypoid lesions 0-III according to Paris Classification Lesions with suspicion of malignancy (rigidity, non-lifting lesions, mucosal fragility, ulceration) Patients unable to provide informed consent Patients with coagulopathy and INR >1.5 (not corrected with replacement therapy, such as enoxaparin).

Patients who have undergone previous attempt of lesion resection (residual disease, local recurrence).

Patients with histological diagnosis of submucosal invading neoplasia who will be sent to surgery and excluded from follow-up.