Safety and Efficacy of Underwater Endoscopic Mucosal Resection for Type I Gastric Neuroendocrine Tumors Measuring 5-9 mm

Study design This is a prospective, single-centre, single-arm cohort study evaluating UEMR for type I gastric NETs. Outcomes will be assessed at the lesion level, with statistical methods accounting for clustering of multiple lesions within individual patients.

Study population Inclusion criteria Patient-level

• Age ≥18 years
• Fit for endoscopic therapy and sedation
• Written informed consent Lesion-level

3.Endoscopically visible gastric neuroendocrine tumor 4.Size 5-9 mm (endoscopic estimation) 5.Clinical and/or histologic features consistent with type I gastric NET 6.No endoscopic features suggestive of deep invasion 7.Technically amenable to snare resection Exclusion criteria

Lesion-level

• Lesions <5 mm (managed with surveillance in routine practice)
• Lesions ≥10 mm
• Type II or type III gastric NET
• Evidence of muscularis propria invasion on EUS
• Distant metastasis or nodal disease on DOTANOC PET-CT
• Prior attempted resection at the same site with significant fibrosis
• Non-lifting lesion due to scarring

Patient-level

• Uncorrectable coagulopathy
• Inability to manage antithrombotic therapy as per guidelines
• Pregnancy
• Inability to provide informed consent

Pre-procedure evaluation

• Detailed clinical assessment and laboratory work-up as per institutional protocol
• Endoscopic documentation of:
• Lesion size, location, morphology
• Number of lesions
• Background mucosal atrophy
• EUS performed for lesions ≥8 mm, atypical morphology, or suspected deeper invasion
• DOTANOC PET-CT performed where clinically indicated to exclude metastatic disease

Intervention: Underwater EMR technique

• Standard diagnostic gastroscope
• Minimal insufflation followed by complete deflation of the gastric lumen
• Lumen filled with normal saline to achieve complete submersion of the lesion
• Optional marking for flat or ill-defined lesions
• Snare capture with a 1-2 mm margin under water
• Resection using standardized electrosurgical settings
• Defect management (clips or coagulation) at endoscopist discretion
• Retrieval of intact specimen
• Specimen orientation and pinning on cork for histopathologic assessment

Histopathologic assessment:

All specimens will be evaluated by experienced gastrointestinal pathologists.

Reported parameters:

• Tumor size
• Tumor grade and Ki-67 index
• Depth of invasion
• Lateral and vertical margins
• Lymphovascular invasion
• Resection status classified as:
• R0: negative lateral and vertical margins
• R1: microscopic margin positivity
• Rx: indeterminate margins

Outcome measures Primary outcome

• R0 resection rate per lesion Secondary outcomes
• En bloc resection rate
• Immediate bleeding (requiring endoscopic hemostasis)
• Delayed bleeding (within 14 days requiring medical attention)
• Perforation
• Procedure time (scope insertion to removal)
• Residual disease at scar on surveillance endoscopy (3-6 months)
• Metachronous lesions during follow-up

Follow-up protocol

• Clinical follow-up at 24-72 hours and at 14 days
• Surveillance endoscopy at 3-6 months post-resection
• Further surveillance as per institutional protocol for type I gastric NETs

Inclusion criteria

Patient-level

• Age ≥18 years
• Fit for endoscopic therapy and sedation
• Written informed consent

Exclusion criteria

Lesion-level

• Endoscopically visible gastric neuroendocrine tumor
• Size 5-9 mm (endoscopic estimation)
• Clinical and/or histologic features consistent with type I gastric NET
• No endoscopic features suggestive of deep invasion
• Technically amenable to snare resection