I now have comprehensive content from multiple authoritative textbooks. Let me compile the full answer.
Endoscopic Resection for Early Gastric Carcinoma (EGC)
Early gastric cancer (EGC) is defined as adenocarcinoma confined to the mucosa or submucosa, regardless of lymph node status. Endoscopic resection offers curative-intent treatment while preserving the stomach - its major advantage over gastrectomy.
Modalities
Two primary techniques are used:
1. Endoscopic Mucosal Resection (EMR)
Figure: EMR by strip biopsy. (1) Saline injected into submucosa to elevate the lesion; (2) the mound is grasped with forceps and the snare placed at its base; (3) snare positioned; (4) electrosurgical current applied and the mucosal lesion is removed. - Sabiston Textbook of Surgery
- Technique: Saline injection into the submucosal layer lifts the lesion, which is then encircled with a snare or suction cap and excised with electrocautery.
- Methods: Cap-assisted (lesion sucked into transparent cap then snared), ligation-assisted (band creates pseudopolyp then excised), or injection-assisted.
- Perforation rate: Low; bleeding ~15% (usually controllable endoscopically).
- Key limitation: Piecemeal resection is common for larger lesions, which raises recurrence risk. En bloc resection is preferred.
2. Endoscopic Submucosal Dissection (ESD)
- Technique: Electrocautery marks the lesion borders; epinephrine + indigo carmine is injected submucosally to hydrodissect; an insulation-tipped (IT) knife dissects a submucosal plane deep to the tumor for complete en bloc removal; bleeding controlled with electrocautery throughout.
- Key advantage: Enables en bloc resection of larger tumors and those with limited submucosal invasion, regardless of tumor size, location, or ulceration status.
- En bloc resection rate: 92.7-98.0% across large series; a meta-analysis confirms ESD superior to EMR for en bloc rates.
- More technically demanding than EMR, with a higher risk of perforation.
- Predominantly performed in East Asian centers; adoption in Western centers remains limited due to the steep learning curve.
Indications
Standard (Absolute) Criteria
(Box 87.2, Sabiston Textbook of Surgery)
| Criterion | Requirement |
|---|
| Histological type | Intestinal-type (differentiated) adenocarcinoma |
| Depth | Confined to mucosa (Tis or T1a) |
| Lymphovascular invasion | Absent |
| Ulceration | None |
| Size | < 2 cm |
Risk of lymph node metastasis in patients meeting standard criteria: approximately 0.2%.
Expanded (Extended) Criteria
(Japanese guidelines / Clinical Gastrointestinal Endoscopy)
ESD allows curative resection for "extended indications" if the lesion is resected en bloc with negative margins, no lymphovascular invasion, and meets any one of the following:
- > 2 cm, differentiated type, intramucosal, without ulceration
- ≤ 3 cm, differentiated type, intramucosal, with ulceration (UL+)
- ≤ 2 cm, undifferentiated type, intramucosal, without ulceration
- ≤ 3 cm, differentiated type, SM1 invasion (< 500 µm into submucosa)
Risk of lymph node metastasis at expanded criteria: ~0.7% (vs 0.2% for standard criteria). Expanded criteria are not widely adopted in Western centers.
The JCOG 0607 trial validated ESD for expanded indications (excluding undifferentiated type): 5-year OS was 97.0% (95% CI: 95.0-98.2%), exceeding the pre-specified threshold, establishing ESD as a standard treatment for EGC within expanded indications.
Pre-procedural Assessment
- Endoscopic ultrasound (EUS): Critical for T staging - the first/second hypoechoic layers = mucosa; third hyperechoic = submucosa; fourth hypoechoic = muscularis propria. EUS distinguishes T1a (mucosal) from T1b (submucosal) disease.
- Magnifying NBI: The MESDA-G algorithm (Magnifying Endoscopy Simple Diagnostic Algorithm for EGC) guides optical diagnosis.
- Paris classification: Classifies superficial lesion morphology (0-I polypoid; 0-IIa/b/c flat; 0-III excavated).
- CT staging: To exclude lymph node involvement and distant metastasis before planning endoscopic vs surgical approach.
Post-resection Pathological Assessment and Decision-Making
After resection, the specimen is assessed for:
- Completeness: Lateral and vertical margins
- Depth of invasion: Mucosal (M) vs. submucosal (SM1 < 500 µm; SM2 ≥ 500 µm)
- Histological type: Differentiated vs. undifferentiated
- Lymphovascular invasion: ly(+) or ly(-)
- Ulceration: Present or absent
Management by Resection Result
| Scenario | Action |
|---|
| En bloc, negative margins, no LVI, meets standard/expanded criteria | Curative - surveillance only |
| Positive lateral margins | Repeat endoscopic therapy or close surveillance |
| Positive vertical margins, LVI, or SM2+ invasion | Refer for gastrectomy + lymphadenectomy |
Outcomes vs. Surgery
Meta-analyses show endoscopic resection is associated with:
- Higher rates of incomplete resection and local recurrence compared with conventional surgery
- No significant difference in 5-year overall survival when selection criteria are appropriately applied
- (Sabiston Textbook of Surgery)
A 2025 systematic review comparing ESD vs. EMR (
Zheng & Xu, Front Med, PMID 41393136) reconfirms ESD superiority for en bloc and R0 resection rates.
Complications
| Complication | EMR | ESD |
|---|
| Bleeding | ~15% | ~5-7% (delayed bleeding a known risk) |
| Perforation | Low (< 1%) | Higher than EMR (1-5%) |
| Stricture | Rare in stomach | Rare in stomach |
A 2026 meta-analysis (
Chen et al., PMID 41896436) identified risk prediction models for delayed bleeding after gastric ESD, noting anticoagulant use, lesion size, and specimen size as key predictors.
Complications including perforation and post-ESD bleeding are generally manageable endoscopically (clips, coagulation), avoiding the need for surgery in most cases.
Technical Innovations (2025-2026)
- Traction-assisted ESD: A 2026 meta-analysis (Zhang et al., Surg Endosc, PMID 41731140) confirms traction devices improve ESD efficacy and safety (faster dissection, reduced perforation).
- Pocket-creation method (PCM): A 2025 meta-analysis (Song et al., Saudi Med J, PMID 40335108) found PCM reduces procedure time and improves en bloc resection rates vs. conventional ESD.
Summary
| Feature | EMR | ESD |
|---|
| Best for | Lesions < 2 cm, no ulcer | Larger, ulcerated, or SM1 lesions |
| En bloc rate | Lower | 92-98% |
| Recurrence | Higher (piecemeal risk) | Lower |
| Complexity | Lower | Higher (steep learning curve) |
| Perforation risk | Lower | Higher |
| West adoption | Moderate | Limited |
Key principle: Any resection with positive vertical margins, lymphovascular invasion, or SM2 submucosal invasion should prompt referral for gastrectomy with lymphadenectomy, as these features correlate with meaningful lymph node metastasis risk that exceeds the safety threshold for endoscopic cure.
Sources: Sabiston Textbook of Surgery (p. 1803-1805); Clinical Gastrointestinal Endoscopy, Expert Consult 3e (pp. 487-492); Current Surgical Therapy 14e (p. 69); recent systematic reviews PMID 41393136, 41731140, 40335108, 41896436.