French Colorectal ESD Cohort in Experts Centers

R0 Resection rate according to the definition of of the European Society of Gastrointestinal Endoscopy.

Recurrence during the first endoscopic follow-up will be defined by the presence of adenoma or adenocarcinoma at the resection scar, whether visible or not, and confirmed by systematic biopsies of the resection scar.

Monobloc resection is defined as resection of the lesion in a single piece.

Monobloc resection exclusively by ESD is defined as resection of the lesion in a single piece using submucosal dissection only, without the use of a diathermic loop (hybrid technique).

Curative resection is defined according to the recommendations of the European Society of Digestive Endoscopy as a monobloc R0 dissection without any negative anatomopathological criteria (well-differentiated lesion, no emboli, no budding > 1, submucosal infiltration < 1 mm).

Optimal dissection is defined as exclusive R0 dissection without perforation, with a resection speed > 20 mm2/min.

• Per-procedural perforation: breach of the colonic musculature with visualization of the peritoneal cavity.
• Post-procedural perforation (≤ 30 days): abdominal pain associated with fever or inflammatory syndrome and presence of extraluminal air on abdominal-pelvic CT scan, in the absence of per-procedural perforation.
• Per-procedural hemorrhage: considered a complication only if it leads to interruption of the procedure.
• Post-procedural hemorrhage: rectal bleeding or melena requiring hospitalization (or prolonged hospitalization) or endoscopic hemostasis.

Curative endoscopic resection without surgery is defined by the absence of adenoma or adenocarcinoma at the resection scar after 36 months of follow-up, regardless of the number of endoscopic treatments required.

A metachronous lesion is defined as the presence of a new superficial colorectal lesion during one of the endoscopic checks, a lesion not visualized during one of the previous examinations.

Any colorectal surgery will be taken into account, whether due to failure of the endoscopic procedure, recurrence, a complication of the procedure, or an anatomopathological reason.

Recurrence at 36 months will be defined by the presence of adenoma or adenocarcinoma at the resection scar, whether visible or not, and confirmed by systematic biopsies of the resection scar.

The histological prediction of resected lesions will be established using validated classifications (Paris, SANO, NICE, KUDO, JNET, CONECTT). It will be compared with the definitive histological results to assess their sensitivity, specificity, and diagnostic accuracy within the cohort.

Oncological, technical, and complication outcomes will be analyzed according to the annual volume of the centers (low volume = < 50 ESDs per year; intermediate volume = between 50 and 100 procedures per year; high volume = > 100 procedures per year).

Oncological, technical, and complication outcomes will be analyzed according to the colonic or rectal location of the lesion.

Oncological, technical, and complication outcomes will be analyzed according to the traction system used for the procedure, matching lesions according to difficulty criteria validated by the literature.

The learning curve of trainees will be used to evaluate oncological, technical, and complication outcomes using the LC CUSUM method.

A difficulty score predicting success (R0 without perforation) will be created by performing a multivariate analysis according to the TRIPOD GUIDELINES using a derivation cohort and validated on a derivation cohort.