Expanding endoscopic resection in the rectum: knife-assisted resection (KAR) and submucosal tunneling endoscopic resection (STER) for subepithelial tumors beyond neuroendocrine tumors, a multicenter retrospective cohort study

Le résumé fourni par la source

Aims To demonstrate the feasibility and real-world effectiveness of KAR and STER techniques for endoscopic resection of rectal subepithelial tumors, excluding neuroendocrine tumors (NETs). Table 1 Patients’ Demographics and Clinical Indications for Resection Variable n (%) Type of Hospital a Private: 10 (18.2%); Public: 42 (76.4%); Not specified: 3 (5.4%) Type of Hospital b University: 17 (30.9%); Non-university: 27 (49.1%); Not specified: 11 (20%) Gender Male: 33 (60.0%); Female: 22 (40.0%) Age, years, median (range) 60 (27–87) ASA score I: 24 (43.6%); II: 28 (50.9%); III: 3 (5.5%) Charlson Comorbidity Index, median (range) 2 (0–4) Indication for endoscopic resection Malignant potential: 19 (34.5%); Diagnostic resection: 17 (30.9%); Symptom-related: 19 (34.5%) Symptoms None: 32 (58.2%); Obstruction: 7 (12.7%); Pain: 8 (14.5%); Bleeding: 8 (14.5%) Methods A retrospective, multicenter study was conducted between January 2017 and October 2025. Data were collected using a standardized case report form (CRF) distributed to all participating centers. Descriptive and comparative analyses were performed to evaluate key outcomes, namely en bloc resection and adverse events, using R software (version 4.4.2). Results Fifty-five patients were included (30 males, 55%), with a median age of 60 years (range 27–87), from 11 centers across 9 countries ([ Table 1 ]). Diagnostic pre-resection sampling was available in 26 cases (47.3%), most commonly obtained with EUS-FNA/FNB (41.8%). The main indications for resection were malignant potential (34.5%) and symptoms ( 34.5%) ([ Table 1 ]). On EUS, the layer of origin was the muscularis propria in 27 cases (49.1%), the submucosa in 18 (32.7%), and the muscularis mucosa in 4 (7.3%). The majority of lesions demonstrated a partially (49.1%) or mainly endoluminal (38.2%) growth pattern, while a mainly extraluminal component was noted in 5 cases (9.1%). KAR was performed in 33 patients (60.0%), STER in 20 (36.4%), and 2 patients (3.6%) underwent knife-assisted hybrid techniques (circumferential incision with snare resection or full-thickness resection using a knife). Lesion size, location, distance from the dentate line, and growth pattern were comparable between groups, whereas muscularis propria origin was more frequent in STER [12/20 (60%)] than in KAR [14/33 (42%)], p=0.005. The median distance from the dentate line was 6.0 cm in both groups, while the median tumor size was 20 mm (range 10–58) for KAR and 29 mm (range 9–64) for STER (p=0.07). During dissection, the plane was purely submucosal in 18 cases (32%), intramuscular in 30 (55%), and extended into the perirectal fat in 7 (13%). All tumors were removed en bloc, with no septic complications. The predominant histology was GIST (34.5%), followed by leiomyoma (21.8%), lipoma (16.4%), and schwannoma (9.1%). Most GISTs were very low or low risk, and five patients received adjuvant Tyrosine Kinase Inhibitor therapy. The median procedure duration was 52.5 minutes (range 15–181). Two adverse events were recorded (3.6%), neither directly attributable to the endoscopic intervention: one urinary tract infection and one ischemic stroke requiring ICU admission. The median hospital stay was 1 day (range 0–7). Two early recurrences in GIST cases were recorded, one managed endoscopically and the other under surveillance [ 1 ] [ 2 ] [ 3 ]. Conclusions KAR and STER appear safe and effective for rectal subepithelial tumor resection. High en bloc rates, low complications, and short hospital stay support their minimally invasive role. Prospective studies are needed to confirm these retrospective findings. Publication History Article published online: 06 June 2026 © 2026. European Society of Gastrointestinal Endoscopy. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.