Commercial

There are two endoscopic resection procedures commonly used for diagnostic and potentially therapeutic purposes for pre-cancer and early cancers of the digestive tract: endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD). Although EMR has historically been the primary minimally invasive technique, ESD provides potential solutions to the limitations of EMR. Primarily, ESD can be used for larger lesions. Because EMR is limited by the size of snare, larger lesions are unable to be removed in one piece, both superficially and deep. This makes it difficult for an accurate pathohistological assessment, which can lead to cancer remaining in the body, cancer recurrence, and unnecessary surveillance/treatment.

ESD allows for the removal of larger lesions because of the use of dissecting knives (rather than a snare) and deeper removal of tissue (submucosa rather than mucosa). However, ESD is not without limitations. First, this technique requires a significant level of expertise, particularly in parts of the digestive tract that have thinner tissue layers. Second, the expertise is required due to the inherent risk of bleeding, perforation, and stricture (primarily seen with ESD in the esophagus), among other challenges associated with endoscopic procedures. Third, ESD has longer procedure times compared to EMR. Fourth, as with EMR, there is a risk associated with improper triage to ESD, where surgery and/or lymph node dissection would have been more appropriate. Nevertheless, ESD provides an endoscopic resection technique for larger lesions that can be both diagnostic and therapeutic.

PROFESSIONAL SOCIETY GUIDELINES

ENDOSCOPIC SUBMUCOSAL DISSECTION OF ESOPHAGEAL SQUAMOUS CELL CARCINOMA AND ESOPHAGEAL DYSPLASIA

The National Comprehensive Cancer Network (NCCN) Esophageal and Esophagogastric Junction Cancers guideline (2026) describes the dual utility of endoscopic resection (both ESD and EMR) for diagnostic and therapeutic purposes. The guideline does not provide explicit preference for one technique over the other for diagnostic purposes. However, after a diagnostic endoscopy and biopsies have concluded that early stage esophageal squamous neoplasia does not contain high-risk features (i.e., nonulcerated, not poorly differentiated, no lymphatic invasion), a preference for EMR or ESD is based on size. If the lesion is 15 mm or greater, ESD is preferred. If the lesion is less than 15 mm, then ESD or EMR can be used (although ESD may be preferred if the lesion is >10 mm or the morphology is suspicious for deep submucosal invasion).

• Esophageal squamous dysplasia or early-stage esophageal squamous cell carcinoma (ESCC); well differentiated, nonulcerated; without submucosal invasion; measuring greater than 15 mm: ESD over EMR is suggested.
• Esophageal squamous dysplasia or early-stage ESCC; well differentiated, nonulcerated; without submucosal invasion; measuring 15 mm or less: no recommendation is made for or against either ESD or EMR.

In 2023, the Japan Esophageal Society (JES) published an esophageal cancer practice guideline with the following recommendations (no differentiation between squamous cell carcinoma [SCC] and adenocarcinoma):

• T1a-EP/LPM/MM non-circumferential: EMR or ESD
• T1a-EP/LPM whole-circumferential, 5 cm or less: ESD

In 2022, the European Society of Gastrointestinal Endoscopy (ESGE) published a guideline for ESD of superficial gastrointestinal lesions, with the following ESCC recommendations:

• ESGE recommends ESD as the treatment of choice for most esophageal squamous cell and gastric (or junctional non-Barrett) superficial lesions, mainly to provide en bloc potentially curative resection with accurate pathological staging. Strong recommendation, moderate quality of evidence.
• ESGE suggests that ESD might also be considered for en bloc resection of noncircumferential clinically staged T1a-m3/T1b-sm1 or for circumferential clinically staged T1a-m1/m2 esophageal SCC. Weak recommendation, moderate-quality evidence.

In 2020, the Korean Society of Gastrointestinal Endoscopy (KSGE) published a clinical practice guideline for endoscopic resection of early gastrointestinal cancer, recommending that:

• ESD rather than EMR for en bloc and curative resection of ESCC confined to the mucosa (grade of recommendation, strong; level of evidence, moderate). Additionally recommended is consideration of ESD first, regardless of the lesion size because ESD has higher en bloc and curative resection rates than EMR for ESCC confined to the mucosa, and has complication risks within an acceptable range. If the operator lacks experience with ESD, then EMR could be considered for only lesions smaller than 10 mm.

ENDOSCOPIC SUBMUCOSAL DISSECTION OF EARLY STAGE ESOPHAGEAL ADENOCARCINOMA AND BARRETT ESOPHAGUS DYSPLASIA​

The NCCN's Esophageal and Esophagogastric Junction Cancers guideline (2026) describes the dual utility of endoscopic resection (both ESD and EMR) for diagnostic and therapeutic purposes. The guideline does not provide explicit preference for one technique over the other for diagnostic purposes. However, after a diagnostic endoscopy and biopsies have concluded that early-stage esophageal high-grade dysplasia (HGD)/adenocarcinoma has a visible lesion present and does not contain high-risk features (i.e., nonulcerated, not poorly differentiated, no lymphatic invasion), a preference for EMR or ESD is based on size. If the lesion is 2 cm or greater, ESD is preferred. If the lesion is less than 2 cm, ESD or EMR can be used (although ESD may be preferred if the morphology is suspicious for deep submucosal invasion). If there is either high-grade dysplasia or early carcinoma, there is no preference for ESD or EMR.

In 2024, the American Gastroenterological Association (AGA) published a clinical practice guideline on endoscopic eradication therapy (EET) of Barrett esophagus and related neoplasm. The guideline suggests the use of EMR or ESD for individuals with Barrett esophagus with neoplastic lesions undergoing endoscopic resection. The guideline suggests the following implementation considerations:

• Individuals suspected of having T1 esophageal adenocarcinoma (EAC) should be referred for consideration of EET.
• Endoscopic resection is the test of choice over endoscopic ultrasound for distinguishing EAC from HGD and for staging depth of invasion in early cancer.
• The vast majority of neoplastic lesions may be managed with EMR rather than ESD.
• Individuals with large bulky neoplastic lesions or lesions highly suspicious of at least T1b invasion (e.g., those with depressed, Paris IIc or IIa+c lesions) and deemed candidates for endoscopic resection might benefit from ESD over EMR.
• Individuals with previously failed EMR might benefit from ESD.
  

As mentioned previously, ASGE published a summary of recommendations for ESD for the management of early esophageal and gastric cancers. ASGE provides the following conditional, low quality of evidence recommendations:

• Early-stage EAC (T1) or Barrett nodular dysplasia; well differentiated, nonulcerated; measuring over 20 mm: ESD over EMR is suggested. 
• Early-stage EAC (T1) or Barrett nodular dysplasia; well differentiated, nonulcerated; measuring 20 mm or less: no recommendation is made for or against ESD or EMR.​​

JES published an esophageal cancer practice guideline in 2023. They provided the following recommendations (not differentiated between SCC and adenocarcinoma):

• T1a-EP/LPM/MM non-circumferential: either EMR or ESD 
• T1a-EP/LPM whole-circumferential, 5 cm or less​: ESD​

ESGE published a guideline for ESD of superficial gastrointestinal lesions. They provided the following recommendations for Barrett esophagus–associated lesions:

• ESGE recommends using EMR for visible lesions 20 mm or less with low probability of submucosal invasion (Paris type 0-IIa, 0-IIb) and for larger or multifocal benign (dysplastic) lesions. Strong recommendation, high-quality evidence.
• For Barrett esophagus–associated lesions, ESGE suggests using ESD for lesions suspicious for submucosal invasion (Paris type 0-Is, 0-IIc), for malignant lesions larger than 20 mm, and for lesions in scarred/fibrotic areas. Weak recommendation, low-quality evidence.

In 2020, the Japan Gastroenterological Endoscopy Society (JGES) published a guideline outlining endoscopic ESD and EMR for esophageal cancer. JGES provides the following explicit recommendation of ESD over EMR. ESD was strongly recommended over EMR for the radical treatment of superficial EAC suitable for endoscopic resection.

ENDOSCOPIC SUBMUCOSAL DISSECTION OF GASTRIC LESIONS SUGGESTIVE OF EARLY ADENOCARCINOMA

The NCCN's Gastric Cancer guideline (2026) describes the dual utility of endoscopic resection (both ESD and EMR) for diagnostic and therapeutic purposes. The guideline does not provide explicit preference for one technique over the other for diagnostic purposes. However, after the appropriate diagnostic workup, an early-stage gastric adenocarcinoma with a favorable pathology (i.e., well- or moderately differentiated, intestinal-type) and not suggestive of deep submucosal invasion, ESD is the preferred endoscopic technique, for lesions 2 cm or less.

​ASGE published a summary of recommendations for ESD for the management of early esophageal and gastric cancers. ASGE provided the following conditional, low quality of evidence recommendations:

• Well- or moderately differentiated early-stage gastric adenocarcinoma; nonulcerated, intestinal type; measuring 20–30 mm: ESD over EMR is suggested. 
• Well- or moderately differentiated early-stage gastric adenocarcinoma; nonulcerated, intestinal type; measuring less than 20 mm; no recommendation for or against ESD or EMR is made. ​

ESGE published a guideline for ESD of superficial gastrointestinal lesions. ESGE provided the following gastric lesion recommendations:

• ESD for differentiated gastric lesions clinically staged as dysplastic or as intramucosal carcinomas (of any size if not ulcerated and ≤ 30 mm if ulcerated), with EMR being an alternative for Paris 0-IIa lesions of size 10 mm or less with low likelihood of malignancy. Strong recommendation, moderate-quality evidence.
• Gastric adenocarcinomas that are 30 mm or less, submucosal (sm1), and well-differentiated, or 20 mm or less, intramucosal, and poorly differentiated type, both without ulcerative findings, can be considered for ESD, although the decision should be individualized. Weak recommendation, low-quality evidence.

In 2021, the JGES, in collaboration with the Japanese Gastric Cancer Association (JGCA), published a guideline outlining ESD and EMR for early gastric cancer. They provide the following absolute indications for ESD, with a strong recommendation, based on moderate evidence:

• Nonulcerated, clinically intramucosal (cT1a), differentiated-type carcinomas with a long diameter greater than 2 cm
• Ulcerated, cT1a, differentiated-type carcinomas with a long diameter of 3 cm or less
• Non-ulcerated, cT1a, nondifferentiated-type carcinomas with a long diameter of 2 cm or less ​

ENDOSCOPIC SUBMUCOSAL DISSECTION OF COLORECTAL LESIONS

In 2024, ESGE published a colorectal polypectomy and EMR guideline update. ESGE provides the following recommendations:

• ESD may also be suggested as an alternative for removal of large nonpedunculated adenomatous polyps of sized 20  mm or greater in selected cases and in high-volume centers. Weak recommendation, low quality of evidence.
• En bloc resection techniques, such as en bloc EMR, ESD, endoscopic intermuscular dissection, endoscopic full-thickness resection (EFTR), or surgery should be the techniques of choice in cases with suspected superficial invasive carcinoma, which otherwise cannot be removed en bloc by standard polypectomy or EMR. Strong recommendation, moderate quality of evidence.

• ESGE recommends polypectomy and/or EMR (en bloc or piecemeal) as the treatment of choice for most superficial colorectal lesions. Strong recommendation, high-quality evidence.
• ESGE suggests that ESD should be considered for en bloc resection of colorectal (but particularly rectal) lesions with suspicion of limited submucosal invasion (demarcated depressed area with irregular surface pattern or a large protruding or bulky component, particularly if the lesions are >20 mm), or for lesions that otherwise cannot be completely removed by snare-based techniques. Weak recommendation, moderate-quality evidence.

In 2022, the American Society of Colon and Rectal Surgeons (ASCRS) published clinical practice guidelines for the management of colon cancer. ASCRS provided the following recommendations:

• For individuals with a "malignant polyp," either endoscopic excision or oncological resection may be appropriate depending largely on polyp histopathological features and completeness of excision. Grade of recommendation: strong recommendation based on moderate-quality evidence, 1B.
• A low-risk malignant pedunculated or sessile polyp may be defined as one with well- or moderately differentiated cancer, no lymphovascular invasion or perineural invasion, no or low tumor budding, a negative resection margin, and less than 1 mm submucosal invasion depth. Endoscopic excision is generally considered definitive treatment for these malignant polyps as the risk of residual disease in the colon wall or cancerous lymph nodes is negligible.

KSGE published a clinical practice guideline for endoscopic resection of early gastrointestinal cancer. KSGE provided the following recommendation:

• En bloc and histologically complete resection should be achieved for endoscopic treatment of a suspected or established early colorectal cancer. They recommended ESD for the treatment of endoscopically resectable early colorectal cancer that cannot be resected en bloc using EMR technique (grade of recommendation, strong; level of evidence, moderate).

In 2020, the US Multi-Society Task Force on Colorectal Cancer, which comprises the American College of Gastroenterology, the American Gastroenterological Association, and ASGE, published a recommendation guideline on endoscopic removal of colorectal lesions. The Task Force provided the following recommendation:

• The indications for colorectal ESD are relatively few, even at experienced centers, because most colorectal neoplasms are benign and can be resected using piecemeal EMR with minimal risk of recurrence. Large-sized (>20 mm in diameter) lesions that are indicated for endoscopic rather than surgical resection, and in which en bloc resection using inject-and-cut EMR is difficult, may be considered. These include lesions suspected to have submucosal invasion (i.e., large, depressed lesion or pseudodepressed LST-NG lesion), mucosal lesions with fibrosis, local residual early carcinoma after endoscopic resection, and nonpolypoid colorectal dysplasia in individuals with inflammatory bowel disease.

PEER-REVIEWED LITERATURE

ENDOSCOPIC SUBMUCOSAL DISSECTION OF ESOPHAGEAL SQUAMOUS CELL CARCINOMA AND ESOPHAGEAL DYSPLASIA

In 2022, Liu and Zhao conducted a systematic review and meta-analysis comparing ESD to esophagectomy (ESO) as treatment options for superficial squamous esophageal cancer. Studies of all designs were included, so long as they compared hospital stay duration, procedure duration, recurrence, complications, or survival (primary outcome). Sixteen studies were included in the analysis, comprising 13,262 individuals (ESD, N=5213; ESO, N=8049). Hospital stay and procedure duration were both significantly decreased for individuals who received ESD, while there was no difference in recurrence. Complications and all-cause mortality were both significantly decreased for individuals who received ESD, while short-term (<3 years) and long-term (>3 years) were not significantly different. Because there was no mention of staging, it is unclear whether these procedures were indicated for similar individuals. Additionally, the review is limited by the included study types and heterogeneous outcomes of the studies.

In 2020, Wang et al. conducted a meta-analysis comparing ESD to EMR. Only studies that compared these two procedures were included in the review. In eight of the nine studies, the average lesion size treated with ESD was larger than those treated with EMR. ESD had a significantly greater en bloc resection rate and complete resection rate, although both have high heterogeneity. Local recurrence rate was significantly lower in the ESD group, but there were significantly more complications. ESD procedure time was also significantly greater.

Yeh et al. (2020) investigated the long-term outcomes of ESD compared to surgery for superficial esophageal SCC via a systemic review and meta-analysis. Varying study designs were included. The primary outcomes were overall survival, disease-specific survival, and recurrence-free survival. Of the studies that compared ESD and ESO, 638 lesions were treated with ESD and 546 lesions were treated with ESO. There was nearly a significant difference in depth of invasion between the groups (P=0.057), but there was no difference in lesion size. Five-year overall survival, disease-specific survival, and recurrence-free survival were not significantly different.

The NCCN Esophageal and Esophagogastric Junction Cancers guideline recommends ESD or EMR for lesions less than 15 mm with no high-risk features (i.e., muscularis mucosa or submucosal invasion, poorly differentiated, and lymphovascular invasion). The guideline then clarifies that “ESD may be preferred for lesions >10 mm in size; morphology suspicious for deep submucosal invasion should be specifically considered for ESD." The tentative language in the clarification (i.e., “may be preferred") is worth noting. Further, while the clarification preferentially suggests ESD for the treatment lesions “suspicious of deep submucosal invasion," this is a contraindication for ESD altogether. This is in fact a high-risk feature, for which NCCN recommends neither ESD nor EMR. Beyond 15 mm, with no high-risk features, ESD is preferred. Thus, this 10- to 15-mm lesion size range is under question, of which there are insufficient data to support ESD over EMR, given (1) the increased complexity and associated risks of ESD, and (2) EMR is already established as the technique for this lesion size.

ENDOSCOPIC SUBMUCOSAL DISSECTION OF COLORECTAL LESIONS

In 2024, Malik et al. conducted a systematic review and meta-analysis of ESD for colorectal dysplasia in individuals with inflammatory bowel disease. Twelve studies were included in the review, consisting of 291 lesions. The average lesion size ranged between 19 mm and 40.7 mm. En bloc, R0, and curative resections were achieved in 92.5%, 81.5%, and 48.9% of resections, respectively. Thirteen percent of individuals required additional surgery following ESD. Bleeding and perforation occurred in 7.7% and 5.3% of individuals, respectively.

Singh et al. (2023) conducted a systematic review and meta-analysis of colorectal ESD, focused on Western countries. Thirty-three studies were included in the review, totaling 4206 individuals. Lesions were summarized by Paris classification and histopathology. The average lesion size ranged between 2.3 cm and 6.8 cm. En bloc, R0, and curative resections were achieved in 84.6%, 75.6%, and 81.9% of lesions, respectively. Bleeding and perforation occurred in 4.1% and 5.5% of individuals, respectively. Five percent of individuals required additional surgery following ESD. A subgroup analysis revealed that high-volume centers (>40 procedures annually) performed better than low-volume centers, although only en bloc resection rate was significantly different.

Wang et al. (2023) conducted a systematic review and meta-analysis comparing ESD and EMR of colorectal adenoma and tumors. Twelve studies were included in the review. Lesions treated with EMR were an average 11 mm smaller than those treated with ESD and EMR operating time was 53 minutes shorter than ESD. The odds of an en bloc and R0 resection both favored ESD over EMR. Complications were higher among individuals treated with ESD, but recurrence was lower.

In 2022, Nishizawa et al. conducted a systematic review of long-term outcomes of ESD for colorectal epithelial neoplasms. Notably, studies that required additional surgery after ESD were excluded, which may bias both long- and short-term outcomes. Regardless, en bloc and curative resections were obtained 94.6% and 87.1% of the time. Overall survival and disease-specific survival at 5 years were 93% and 99%, respectively. Local recurrence was associated with incomplete resection, noncurative resection, and deep submucosal invasion.

Rotermund et al. (2022) conducted a systematic review and meta-analysis of recurrence rates after endoscopic resection of large colorectal polyps. The 12-month local recurrence rate decreased based on endoscopic resection technique, in the following order: EMR with no margin ablation, EMR with some margin ablation, EMR with all margins ablated, and ESD (lowest recurrence rate). Notably, no studies utilized ESD for polyps less than 20 mm. This review is limited by the few ESD studies that were included.

In 2022, Zeng et al. conducted a systematic review and meta-analysis of the efficacy and safety of ESD for dysplasia in individuals with ulcerative colitis. A total of 192 individuals had 203 dysplastic lesions, averaging 26.7 mm in size. The en bloc, R0 resection, and curative resection rates are as follows: 94%, 84%, and 81%, respectively. Bleeding and perforation prevalences were 8% and 6%, respectively.

In 2021, Vaganov et al. conducted a systematic review and meta-analysis comparing ESD and EMR of colorectal tumors. The review included eight studies and 1989 tumors. There was no statistical difference in tumor size treated by each procedure, although the average ESD-treated tumors were larger in all but one study. En bloc resection was significantly higher among ESD-treated tumors (91% vs. 48%), and R0 resection rate was significantly higher as well (85% vs. 40%). Bleeding was similar between the treatment groups, but perforation was higher among ESD-treated individuals. Local recurrence was higher in the EMR treatment group.

Lim et al. (2021) conducted a meta-analysis and meta-regression of ESD versus EMR of colorectal polyps. A total of 281,344 polyps from 21 studies were included. En bloc resection and R0 resection rates were higher among the ESD-treated polyps, with a significant increase in procedure time. Perforation in ESD-treated individuals was greater in the ESD group, but there was no difference in bleeding. Further, recurrence in the ESD group was lower than the EMR group. Right colonic polyps were correlated with an increased risk in perforation when treated with ESD, but no other covariate was correlated with any outcome.