Versus ESD: Pros and Cons

 

Early esophageal squamous cell cancer


Early esophageal adenocarcinoma


Absolute indication for endoscopic resection


T1 m1–m2


T1 m1–m3


T1 sm1 invasion cutoff for endoscopic resection


200 μm depth into submucosa


500 μm depth into submucosa


Relative indication for endoscopic resection


T1 m3–sm1 without histologic risk factors (good to moderate differentiation, no LV invasion, and radical vertical margin)


T1sm1 without histologic risk factors (good to moderate differentiation, no LV invasion, absence of tumor budding, and radical resection)


Preferred endoscopic resection technique


ESD


EMR




Early Esophageal Adenocarcinoma (EAC) or Barrett’s-Related Dysplasia


Barrett’s esophagus is the only identifiable premalignant condition for EAC. BE is characterized by the replacement of stratified squamous esophageal mucosa with metaplastic intestinal-type columnar epithelium in distal esophagus. The estimated annual risk of BE progressing to adenocarcinoma is 0.1% to 0.5% and increases to around 5–10% per year if HGD is present. The progression of BE to EAC is believed to be stepwise from intestinal metaplasia to low-grade dysplasia, high-grade dysplasia, intramucosal cancer, and finally invasive EAC [2325].


Visible lesions on endoscopy are usually classified using the Paris classification (protrude, flat, and excavated). Macroscopic appearance of lesions was shown to correlate with the grade and degree of mucosal/submucosal invasion, in a prospective study by Pech et al. Completely flat lesions (Paris type 0–IIb) had no risk of submucosal involvement. On the contrary, slightly elevated (Paris type 0–IIa) lesions had a 9% risk and protruded lesions (Paris type Is and 0–Ip) had 25–26% risk of submucosal invasion [26]. While there is no risk of nodal involvement in high-grade dysplasia, it is also very low in intramucosal cancer with a reported rate of 0–2%. This makes endoscopic resection the treatment of choice for these early lesions. Although a bit controversial, superficial submucosal cancer (sm1, depth of invasion ≤500 μm) with low-risk features (lack of lymphovascular invasion, tumor budding, or poor differentiation) may also be amenable to endoscopic resection [27, 28].


Esophageal Squamous Cell Carcinoma (ESCC)


ESCC is the predominant subtype in the Middle East, Africa, and Asia with abuse of alcohol and tobacco being the most common risk factors [20, 29]. Unlike Barrett’s-related neoplasia, early ESCC has a higher rate of nodal metastasis even when confined to the mucosa. The risk of LN invasion in T1 m3–T1sm1 has been reported to be as high as up to 15%, and therefore these lesions are considered relative indications for curative endoscopic resection as long as there are good or moderate differentiation and absence of lymphovascular invasion. Furthermore, the cutoff for depth of invasion for sm1 lesion is 200 μm. Moreover, in ESCC, submucosal glands can harbor epithelial squamous neoplasia that extends from the luminal epithelial layer, and therefore en bloc resection of early ESCC is highly recommended [30, 31].


Endoscopic Resection


This can be accomplished by either EMR or ESD.


Endoscopic Mucosal Resection


En bloc resection is generally possible by EMR for lesions less than 20 mm in diameter. For larger lesions, piecemeal technique is usually required. The two common EMR techniques for resection of esophageal lesions are the cap (lift-suck-cut) technique and the multiband mucosectomy (ligate and cut) technique. In both techniques, the borders of the lesion should ideally be marked with cautery, prior to resection.


The cap technique requires the use of a transparent cap (straight or oblique) attached to the tip of the endoscope. First, the lesion is lifted with submucosal saline injection and then sucked into the cap creating a pseudopolyp . This is then captured by a snare that is pre-positioned along the rim of the cap and resected by electrocautery. These steps can be sequentially repeated for larger lesions that need piecemeal resection.


The multiband mucosectomy technique on the other hand does not require submucosal lifting and uses a modified banding apparatus similar to that used for variceal banding. The identified lesion is sucked into the cap, and a rubber band is released creating a pseudopolyp that is subsequently resected using a hexagonal snare and electrocautery. The banding and resection can be repeated to remove larger lesion in a piecemeal fashion taking precautions not to leave bridges or islands of neoplastic tissues in between the resected areas. Both the cap and band techniques are comparable in effective piecemeal resection and complete eradication of neoplasia, but the former is more time consuming and requires a higher skill level [32, 33].


Endoscopic Submucosal Dissection


En bloc resection of early esophageal cancers by ESD offers the advantage over piecemeal resection due to the superior ability to assess the depth and lateral extent of invasion. ESD can achieve en bloc resection for lesions larger than 20 mm. After marking the margins of the lesion with cautery, submucosal saline injection is performed to lift the lesion. Then a circumferential incision is made using an electrosurgical knife followed by dissection of the submucosa under direct endoscopic visualization until the entire lesion is removed in one piece. As in the colon, ESD is technically more challenging and carries higher rate of complications when compared to EMR in the esophagus [34].


EMR Versus ESD in Early Esophageal Cancer


The goal of using ESD in removing superficial esophageal cancers is to ensure an en bloc resection, which is optimal for histopathologic evaluation, with the aim of curative resection. ESD does confer a higher rate of R0 resection when compared to EMR in both subtypes of early esophageal cancer [31, 35]. However, there was no significant difference seen in complete remission of BE-related neoplasia at 3 months in one study. In this randomized control trial from Germany by Terheggen et al. comparing EMR versus ESD in BE, R0 resection was achieved more frequently in the ESD group (58.8%) versus 11.7% in the EMR group. However, there was no difference in complete remission from neoplasia at 3 months or during the follow-up period of the study (23.1 ± 6.4 months). ESD had a higher rate of adverse events, but that was not statistically significant [35]. Therefore, the advantages of ESD do not appear to culminate into clinically impactful difference as any residual BE and related neoplasia after EMR can be treated with adjunctive modalities like radiofrequency ablation . ESD may have an edge over EMR in selective situations like large lesions with higher likelihood of submucosal invasion and those with bulky intraluminal component that may be difficult to capture in a band or cap [36].


On the other hand in early ESCC, ESD is preferred as it offers higher rates of en bloc curative R0 resections and lower rates of local recurrence. A retrospective cohort study from Japan on 300 cases comparing ESD to EMR in early ESCC reported a 100% rate of en bloc resection in the ESD group compared to 53.3% in the EMR group. Subsequently there was lower local recurrence rate in the ESD group versus EMR 0.9% vs. 9.8%, respectively [31].


In the esophagus too, ESD has higher rates of perforation when compared to EMR. A meta-analysis comparing ESD and EMR for resection of superficial esophageal cancers reported a 4% perforation rate in ESD compared to 1.3% for EMR [34]. Many of the perforations can be usually managed endoscopically and do not require surgical intervention. Pneumomediastinum is another complication of ESD, which is not uncommon and seen in up to 30%. This usually resolves within 24 hours as carbon dioxide is used for insufflation [37].


EMR, on the other hand, has a higher risk of esophageal stenosis, up to 26%. The risk of stenosis is higher with longer treated segments and with more circumferential area of resection [38, 39].


As discussed earlier, in the colon section, ESD remains more expensive than EMR, requiring more time and costlier equipment as well being technically more challenging. Furthermore, the complexity of ESD may be more obvious in the esophagus especially in BE-related neoplasia due to the limited endoscopic working space, fibrotic submucosa, angulations in the distal esophagus, and movement due to respiration, motility, and heartbeat [36].


Early Gastric Cancer


Gastric cancer is the fourth most common cancer and the leading cause of cancer-related mortality worldwide. Gastric cancer is more prevalent in the countries of East Asia, Eastern Europe, and South America compared to Europe and North America [40]. Because gastric cancer is more prevalent in those countries, especially Japan, national screening programs have been developed for early detection and endoscopic resection techniques for early gastric cancer have become extremely refined and sophisticated. Endoscopic resection offers a less morbid and less expensive alternative compared to surgery for early gastric cancer.


The Japanese Gastric Cancer Association (JGCA) has recommended criteria for endoscopic resection of EGC. Based on these criteria, to be amenable for endoscopic resection, the tumor had to be differentiated-type adenocarcinoma without ulcerative findings, mucosal based (i.e., invading the lamina propria or muscularis mucosae) with a diameter of 2 cm or less. With the development of ESD in the late 1990s in Japan, allowing for en bloc resection of larger lesions, the JGCA updated their guidelines introducing “expanded indications” for endoscopic resection of EGC [41] (refer to Table 10.2 for JGCA absolute and expanded indication).


Table 10.2

JGCA absolute and relative criteria for endoscopic resection of EGC [41]













Absolute criteria


A differentiated-type adenocarcinoma without ulcerative findings of which the depth of invasion is clinically diagnosed as T1a (invades the lamina propria or muscularis mucosae) and the largest diameter of lesion ≤2 cm


Expanded criteria


(ESD should be employed, not EMR)


Depth of invasion is clinically diagnosed as T1a (invades the lamina propria or muscularis mucosae)


AND


(a) Differentiated-type, without ulcerative findings, but >2 cm in diameter


(b) Differentiated-type, with ulcerative findings, and ≤3 cm in diameter


(c) Undifferentiated-type, without ulcerative findings, and ≤2 cm in diameter

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

May 2, 2020 | Posted by in GASTOINESTINAL SURGERY | Comments Off on Versus ESD: Pros and Cons

Full access? Get Clinical Tree

Get Clinical Tree app for offline access