Endoscopy 2019; 51(04): 296-297
DOI: 10.1055/a-0790-8472
© Georg Thieme Verlag KG Stuttgart · New York

Evolution of endoscopic resection for early esophageal squamous cell carcinoma in Western countries

Referring to Berger A et al. p. 298–306
Hajime Isomoto
Division of Medicine and Clinical Science, Department of Multidisciplinary Internal Medicine, Tottori University Faculty of Medicine, Yonago, Japan
› Author Affiliations
Further Information

Publication History

Publication Date:
18 March 2019 (online)

Few data exist in the Western literature about long-term outcomes following endoscopic resection for esophageal squamous cell carcinoma (SCC). Endoscopic mucosal resection (EMR) has been a well-established method for early esophageal SCCs confined to the mucosa; however, its major limitation lies in the difficulty of achieving en bloc resection for larger lesions [1] [2] [3]. EMR has been performed in a piecemeal manner, leading to increased risks of post-procedural local recurrence [1] [2]. Endoscopic submucosal dissection (ESD) allows en bloc resection regardless of tumor size and permits accurate pathological assessment of the risk of lymph node metastasis (LNM) risk [1] [2]. Hence it has been widely used for esophageal SCC in Eastern countries. In recent meta-analyses of Eastern comparative studies of esophageal ESD and EMR, ESD showed considerably higher rates of en bloc and curative resection [1] [2]. However, Western experience of ESD remains scarce owing to the lower incidence of esophageal SCC and the lesser capability of using ESD to treat early disease and dysplastic lesions than in Eastern countries. Santos-Antunes et al. reported adequate curative resection rates in the biggest Western ESD series; nevertheless this still included only a small number of patients with esophageal tumors [4]. Thus, in Western countries accumulation of data with larger cohorts is needed regarding the feasibility of ESD for esophageal SCC.

In this issue of Endoscopy, Berger et al. compare EMR and ESD for long-term outcome and oncological clearance in the treatment of superficial esophageal SCC, based on one of the largest European multicenter evaluations [5]. Follow-up data were obtained from five tertiary referral hospitals in France. These data showed that, for esophageal SCC after ESD, the recurrence-free survival rates at 5 years for tumor invasion depths < m3 and ≥ m3 (m3 tumors involving the muscularis mucosa and also submucosal tumors) were 100 % and 88.2 %, respectively. In contrast the corresponding rates for EMR were 79.5 % and 60.2 %, respectively. Again, Berger et al. confirmed that resection by EMR, tumor infiltration depth of m3 or deeper, noncurative resection, and lack of complementary chemoradiotherapy were reasonably independent risk factors associated with recurrence after endoscopic resection for superficial esophageal SCC [5].

“Berger et al. confirmed that resection by EMR, tumor infiltration depth of m3 or deeper, noncurative resection, and lack of complementary chemoradiotherapy were independent risk factors for recurrence after endoscopic resection for superficial esophageal SCC.”

Squamous cell carcinoma remains the most common subtype of esophageal cancer in Asia [1]. The risk of LNM mainly depends on tumor infiltration depth along with lymphovascular involvement [2]. The LNM rates for m1 and m2 tumors are nil and an acceptable ≤ 3 %, respectively, whereas the LNM rate reaches 10 % for m3 tumors (those involving the muscularis mucosa). Thus, endoscopic resection is the preferred treatment modality for m1 and m2 disease, according to the recommendations of the European Society of Gastrointestinal Endoscopy (ESGE) guidelines [3]. It is not unusual for even extensive lesions, including circumferentially spreading SCC, to remain confined to the esophageal mucosa (m1 – m2 disease) [6]. For such cases, ESD has already been indicated in Japan since 2008, and this practice has now spread throughout Eastern countries and not only in tertiary centers.

Limitations such as retrospective study design and small numbers of subjects enrolled in each group have been reported; however, these results are rather equivalent to previous Eastern studies [1] [2]. The ESGE guidelines also recommended ESD as the interventional option for esophageal SCCs > 10 mm [3]. According to Berger et al., the recurrence-free survival rate at 5 years was 100 % in the EMR group with small tumors < 10 mm [5]. In other words, esophageal EMR is still suitable for smaller SCCs because it is a safe, quickly performed, and straightforward procedure. In a recent comparative study from China, for esophageal SCC < 15 mm, EMR was associated with the same levels of clinical success as ESD [7]. Even when operators have sufficient experience in Eastern countries, esophageal ESD has long procedure times and high rates of complications, including life-threatening perforation [1] [2] [7].

In the study by Berger et al., with the largest Western multicenter series of ESD for esophageal SCC, the proportions for infiltration depth were 64.2 % for m1 – m2, 16.2 % for m3, and 19.5 % for submucosal invasive tumors [5], suggesting that preoperative assessment for the suitability of resection was not feasible. The relatively high incidence of m3 or submucosal SCC at entry could reflect the use of endoscopic ultrasound (EUS) for the evaluation of tumor infiltration depth. In recent decades in Western countries, including France, esophageal EUS has commonly been performed for preoperative assessment even for superficial malignancies [5]. While EUS is an indispensable mainstay for T and N staging in advanced disease, its role for superficial malignancy remains controversial.

Magnifying endoscopy with narrow band imaging (NBI) is widely used in Eastern countries, and the Japanese Esophageal Society proposed a classification system to optimize diagnostic accuracy in decisions for endoscopic resection [1] [8]. The morphology of intrapapillary capillary loops is classified into type A or B based on the presence of abnormalities including weaving, dilatation, irregular caliber, and diversity in shape. Type B microvessels are further subclassified into B1 – B3 based on the size of the abnormal intrapapillary capillary loops and whether a loop-like appearance is preserved [1] [8]. The prospective multicenter study using this classification showed that the sensitivity and positive predictive value of the B1 type for m1 – m2 tumors were rather high at 97.5 % and 92.4 %, respectively [8], exceeding the accuracy of EUS. In addition, nonmagnifying NBI endoscopy has been considered to be a reliable alternative method for the screening of early esophageal SCC or dysplasia. Although chromoendoscopy with iodine staining remains the gold standard method [1], incorporation of the NBI-based endoscopic evaluations would achieve more accurate preoperative diagnosis regarding tumor infiltration depth, and thus more appropriate endoscopic resection of esophageal SCCs.

Post-ESD esophageal strictures were reported as the main delayed complication in the study by Berger et al. [5]. This problem remains unresolved even in the Eastern experience, particularly in cases of aggressive esophageal ESD for widespread SCC, although endoscopic balloon dilation, steroid administration, or the innovative technique of oral epithelial cell sheet engraftment [1] [2] [6] have been applied to limit post-ESD stricture.

ESD is being increasingly adopted in Western countries, and from these early stages of Western implementation, esophageal ESD could be a safe and efficient method when performed for more carefully selected cases in tertiary settings.