Putting It All Together: Indication for Endoscopic Mucosal Resection and Endoscopic Submucosal Dissection in the Stomach

Yutaka Tomizawa, MD, MSc and Joo Ha Hwang, MD, PhD

Introduction

Endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) are applied for resection of early gastric cancers (EGCs) confined to the superficial (mucosa and submucosa) layers of the gastrointestinal tract. The technique of ESD was invented by Japanese endoscopists to enable en bloc margin-negative resection of EGC while avoiding invasive surgery. En bloc resection allows for detailed histopathologic assessment that is necessary to confirm curative resection. Although ESD has now become widely available and the standard of care for the treatment of EGCs in Asia, it is important to recognize the indications and limitations of ESD so that appropriate patient selection is made and alternatives (eg, surgery) are considered.

Preprocedural Endoscopic Evaluation

A thorough examination of the surface characteristics of the lesion (eg, vascular morphology and pit pattern) is a prerequisite. The key component is to assess the depth of neoplastic involvement. High-definition optical endoscope with Narrow Band Imaging (Olympus) and/or chromoendoscopy with indigo carmine with near-focus visualization/magnification can all be used to assess resectability. Determining the depth of submucosal involvement (minimal submucosal invasive cancer of < 500 µm from the muscularis mucosa [SM1] vs deeper submucosal invasive cancer of > 500 µm from the muscularis mucosa [SM2]) is of critical importance because deeper submucosal invasion (SMI) (ie, SM2) precludes ESD for curative resection (described in detail later in this chapter). If deep invasion (submucosa or deeper) is suspected based on prior pathology or endoscopic evaluation, then endoscopic ultrasound (EUS) examination should be performed. Deeper submucosal layer invasion carries a very low likelihood of achieving an R0 resection (negative horizontal and deep resection margins) as nearly all of these lesions will show lymphatic spread.

Table 7-1. Indication of Endoscopic Submucosal Dissection for Gastric Neoplasm6

Reprinted with permission from Spring Japan KK.

Principles of Endoscopic Resection (Endoscopic Mucosal Resection/Endoscopic Submucosal Dissection)

Endoscopic resection should be considered for lesions with negligible risk of lymph node metastasis. The large volume of surgical experience for the management of EGC in Japan clearly demonstrated that the presence of lymph node metastasis is the most important prognostic factor and that tumor depth is associated with lymph node metastasis.^1,2 Estimation of depth of invasion should be evaluated based on the macroscopic morphology (ie, Paris classification) and endoscopic features (eg, margin elevation, ulceration, enlarged folds, vascular patterns, and pit patterns).² EUS could be considered to further ascertain the depth of invasion. However, the accuracy of EUS in assessing the depth of invasion of EGC was reported to be 71% to 78% and not sufficient enough to be recommend in all the cases.^3,4 EMR is generally indicated for removal of superficial lesions smaller than 20 mm at its largest diameter. However, piecemeal EMR is sometimes necessary even for lesions greater than 10 mm. Hence, the main advantage of ESD over EMR is the ability to achieve complete en bloc resection of large lesions, irrespective of size, avoiding piecemeal resection and its associated risk of local recurrence.⁵ According to the current Japanese Gastric Cancer Treatment Guidelines,⁶ the absolute indications of ESD for EGC are as follows (Table 7-1):

Differentiated-type adenocarcinoma without ulcerative findings (UL[–])
Depth of invasion confined to the mucosa (clinically diagnosed as T1a)
Largest diameter ≤ 20 mm.

Based on these criteria, endoscopic resection is considered curative when all of the following conditions are fulfilled: en bloc resection, lesion size ≤ 2 cm, differentiated-type histology, pT1a, negative horizontal margin (HM0), negative vertical margin (VM0), and no lymphovascular infiltration.

Table 7-2. Frequency of Lymph Node Metastases in Early Gastric Cancer1

Reprinted with permission from Spring Japan KK.

A meta-analysis of 9 retrospective studies with 3548 lesions (ESD 1495; EMR 2053) demonstrated that the rate of en bloc resection in the ESD group was significantly higher than that in the EMR group (odds ratio [OR] 9.69; 95% confidence interval [CI], 7.74% to 12.13%), as was the total histologically complete resection rate (OR 5.66; 95% CI, 2.92% to 10.96%). ESD also had a lower recurrence rate (OR 0.10; 95% CI, 0.06% to 0.18%); however, the mean time required for resection was longer for ESD than EMR (weighted mean difference 59.4; 95% CI, 16.8% to 102.0%). In the aggregate, multiple studies have shown that ESD results in a higher rate of en bloc, R0, and curative resection rates and a lower rate of local recurrence with an acceptable safety profile for EGC when compared with EMR.^7–10

Expanded Endoscopic Submucosal Dissection Criteria

The traditional ESD criteria for EGC were developed to ensure successful en bloc resection in lesions with negligible risk of lymph node metastasis. However, the stringent criteria for patient selection raised the concern that this conservative approach would possibly result in unnecessary surgery for additional patients with EGC.¹¹ As such, subsequent accumulating data further demonstrated expanded tumor-factor categories associated with a very low likelihood of lymph node metastasis (Table 7-2).^1,12–14 As the ESD technique gained proficiency with regards to resection techniques, the criteria for ESD were expanded to lesions previously considered unsuitable, including the following:

Tumors clinically diagnosed as T1a
Differentiated-type, UL(–), but > 20 mm in diameter
Differentiated-type, UL(+), and < 30 mm in diameter
Differentiated-type, minute SMI (SM1, cancer invasion into the upper third of the submucosa), and < 30 mm in diameter
Undifferentiated-type, UL(–), and < 20 mm in diameter

The large-scale data in Japan of 1161 patients with EGC (1332 lesions) treated by ESD for the criteria for absolute or expanded indications without additional gastrectomy reported that en bloc resection and R0 resection rates were 99.0% and 97.4% (P = .867), and 96.4% and 93.4% (P = .736), in the absolute and expanded indication groups, respectively. The 5-year overall survival and recurrence-free rates were 93.7% and 99.8%, and 90.5% and 98.9%, in the absolute and the expanded indication groups, respectively, with no significant differences between the groups for either outcome. Delayed bleeding rates were significantly higher in the expanded indication group, albeit all cases were successfully managed conservatively.15 An additional study from South Korea has validated the clinical effectiveness of ESD for the expanded criteria.¹⁶

Undifferentiated cancer is known to have an increased risk for lymph node metastasis due to the high probability of microlymphovascular invasion. In an analysis of 3843 patients consisting of 2163 (56.3%) intramucosal cancers and 1680 (43.7%) submucosal invasive cancers who underwent gastrectomy with lymph node dissection for solitary undifferentiated EGC demonstrated that lesion characteristics of size > 20 mm, lymphovascular involvement, and SMI were all independent risk factors for lymph node metastasis (P < .001).¹⁴ In the same study, none of the 310 intramucosal cancers ≤ 20 mm in size, with no lymphovascular involvement or ulcerative findings were associated with lymph node metastases (95% CI 0% to 0.96%). Indeed, retrospective studies from expert referral centers have shown favorable long-term outcomes with low local recurrence and distant metastasis^17,18 and have supported that curative resection is associated with excellent disease-free clinical outcome. Furthermore, another recent study revealed that the 5-year disease-specific survival rates with curative ESD for absolute indications, with curative ESD for expanded indications of differentiated type, and with curative ESD for expanded indications of undifferentiated type, were 99.9%, 99.9%, and 100%, respectively.¹⁹

Endoscopic Submucosal Dissection Versus Surgery

Perioperative clinical factors to be considered when deciding treatment modality (eg, EMR/ESD or surgery) include the likelihood of lymph node metastasis, patient comorbidities, the possibility of early and late adverse events, costs, and expected disease-free survival. The decision as to treatment modality of choice should be determined on an individualized basis by a multidisciplinary team. For lesions that meet the absolute ESD criteria as per the established guidelines, current consensus is that ESD should be the standard of care as to avoid unnecessary surgical intervention. Although no randomized, controlled study has been conducted, a propensity score-matched retrospective analysis has demonstrated that the overall survival was similar for patients with EGC undergoing ESD vs surgery. Furthermore, ESD was associated with lower complication rates and shorter hospital stay.²⁰

Surgery has been traditionally the treatment of choice for lesions within the expanded ESD criteria, particularly when these were initially introduced. A multicenter retrospective study from South Korea of patients who were treated with ESD or surgical resection within expanded criteria from 2006 to 2008 showed a higher rate of 5-year cancer recurrence and a lower rate of disease-free survival in the ESD group compared to the surgical resection group (P = .001).²¹ A more recent study evaluating the clinical outcome for the expanded indication of differentiated-type early gastric neoplasm treated by ESD demonstrated that ESD is not inferior to surgery.²² A propensity-matched analysis of 308 patients with expanded-indication of differentiated-type EGC confirmed that the 5-year overall survival rate was higher in the ESD group than in the surgery group (97.1% vs 85.8%; P = .01) and significantly fewer adverse events were associated with ESD than with surgery (6.8% vs 28.4%; P < .01) with no disease-specific mortality observed in either group.²³ Hence, appropriate lesion selection for expanded ESD criteria can lead to noninferior overall survival with ESD.

Table 7-3. Relationship Between Depth of Invasion and Lymph Node Metastasis26