Endoscopic Resection for Gastric Muscularis Propria-Derived Tumors: Clinical Outcomes from a Stepwise EFTR...

This study aimed to evaluate the clinical outcomes of endoscopic resection for gastric muscularis propria-derived subepithelial lesions, including gastrointestinal stromal tumors, within a stepwise implementation program for endoscopic full-thickness resection (EFTR), and to identify factors associated with pathological R1 resection and technical difficulty.

This retrospective single-center study included consecutive patients who underwent endoscopic muscularis dissection (EMD), endoscopic resection with one-port placement (EROPP), or pure EFTR between January 2017 and January 2026. These techniques were introduced sequentially but selected according to lesion characteristics, technical feasibility, and institutional experience. Clinicopathological characteristics, procedural outcomes, pathological resection status, adverse events, and long-term outcomes were analyzed. Technical difficulty was assessed using a composite definition incorporating prolonged treatment time, major adverse events, pathological R1 resection, and the use of multiple over-the-scope clips.

Seventy patients were included: 6 underwent EMD, 35 EROPP, and 29 pure EFTR. The median tumor size was 22 mm. Endoscopic complete resection was achieved in all patients, and pathological R0 resection in 62/70 (88.6%). Pathological R0 resection rates were 83.3%, 82.9%, and 96.5% for EMD, EROPP, and pure EFTR, respectively, with no significant difference among the three groups (P = 0.140). The median resection and closure times were 35 and 18 min, respectively. One major adverse event occurred: delayed perforation requiring endoscopic treatment. Larger tumor size was significantly associated with both pathological R1 resection and technical difficulty. During a median follow-up of 47 months for the entire cohort, no local recurrence or distant metastasis was observed.

Endoscopic resection for selected gastric muscularis propria-derived tumors within a stepwise implementation program was associated with favorable long-term clinical outcomes. Tumor size was the principal factor associated with technical difficulty and pathological R1 resection. Pure EFTR provided favorable outcomes in appropriately selected lesions, while EROPP remained a complementary option for selected technically challenging lesions. These findings provide practical information for institutions considering the introduction of EFTR and support lesion-based procedural selection according to technical feasibility and institutional expertise.