Background: Endoscopic submucosal dissection (ESD) is a widely accepted treatment option for superficial esophageal squamous neoplasms (SSN) in Asia and is supported by US society guidelines. Outcomes data on ESD for SSN in the Western setting is limited to a few small, single-center studies with a lack of data coming from North America.
Aim: To evaluate ESD for SSN outcomes across various centers in North America.
Methods: This is a retrospective analysis of patients who underwent ESD for SSN at 13 academic centers in North America. Primary outcomes were rate of en bloc, R0, and curative resection, adverse events, local recurrence, lymph node (LN) and distant metastasis, and death and disease-related survival. Curative resection was defined as en bloc, R0 resection of a well to moderately differentiated lesion without LVI in M1/M2 (absolute criteria) and M3/SM1 (expanded criteria).
Results: 146 patients [median age 69, 40% female] underwent ESD for clinical SSN [median tumor size 30 mm (IQR 20-50 mm)]. Most lesions were in the mid esophagus (54%). 25 (17%) patients received prior endoscopic treatment. Histologically, the lesions included 14 (9.5%) LGD, 28 (18.9%) HGD, 52 (35.2%) intramucosal cancer, 50 (34.3%) invasive cancer. In ESCC group, 60 (41%) were M1/M2, 38 (26%) were M3/SM1, and 30 (14%) were SM2 or deeper. En bloc resection rates for all lesions, LGD/HGD, and ≤SM1 ESCC were 98%, 98%, and 97%, respectively. R0 resection rates for all lesions, LGD/HGD, and ≤SM1 ESCC were 76.7%, 83.3%, and 85.9%. Curative resection rates for all lesions, LGD/HGD, and ≤SM1 ESCC were 58.2%, 76.2%, and 79.5%. Curative resection rates were 83.3% for M1/M2 and 78.3% for M3/SM1. There were 2 (1.4%) cases of intraprocedural perforation and 1 (0.7%) case of severe intraprocedural bleeding which were managed endoscopically. Delayed perforation occurred in 1 (0.7%) case and was managed conservatively. Delayed bleeding occurred in 2 (1.4%) cases and was completely treated endoscopically. Stricture formation occurred in 17 (11.6%) cases. 111 (76%) patients had follow-up endoscopy. The median follow-up time was 16 months (IQR 7-32). In the LGD/HGD group, local recurrence occurred in 3 (10%) cases and there was no ESCC-related death. For the ESCC group, local recurrence occurred in 5/59 (8.5%): 2/50 (4%) in the curative and 3/9 (33.3%) in the non-curative cases (p=0.02). LN and distal metastasis occurred 4/59 (6.8%): 1/50 (2%) in the curative and 3/9 (33.3%) in the non-curative cases (p=0.009). 2/59 (3.4%) ESCC patients died from ESCC.
Conclusions: This large multicenter study from North America on ESD in patients with SSN demonstrates a high rate of complete resection with acceptable safety profiles. Risk of local recurrence and metastasis is low after curative ESD. Altogether, ESD appears to be a highly effective treatment for SSN in a Western setting.

<b>Table 1. </b>Patient and Procedure Characteristics

Table 1. Patient and Procedure Characteristics

Table 2. Procedure Outcomes

Jean-Christophe Rwigema

Appearances