Outcomes of Endoscopic Submucosal Dissection vs Esophagectomy for T1 Esophageal Squamous Cell Carcinoma in a Real-World Cohort.

BACKGROUND & AIMS: Esophagectomy is the standard treatment for early-stage esophageal squamous cell carcinoma (EESCC), but patients who undergo this procedure have high morbidity and mortality. Endoscopic submucosal dissection (ESD) is a less-invasive procedure for treatment of EESCC, but is considered risky because this tumor frequently metastasizes to the lymph nodes. We aimed to directly compare outcomes of patients with EESCC treated with ESD vs esophagectomy.
METHODS: We performed a retrospective cohort study of patients with T1a-m2/m3, or T1b EESCCs who underwent ESD (n = 322) or esophagectomy (n = 274) from October 1, 2011 through September 31, 2016 at Zhongshan Hospital in Shanghai, China. The primary outcome was all-cause mortality at the end of follow up (minimum of 6 months). Secondary outcomes included operation time, hospital stay, cost, perioperative mortalities/severe non-fatal adverse events, requirement for adjuvant therapies, and disease-specific mortality and cancer recurrence or metastasis at the end of the follow up period.
RESULTS: Patients who underwent ESD were older (mean 63.5 years vs 62.3 years for patients receiving esophagectomy; P = .006) and a greater proportion was male (80.1% vs 70.4%; P = .006) and had a T1a tumor (74.5% vs 27%; P = .001). A lower proportion of patients who underwent ESD had perioperative mortality (0.3% vs 1.5% of patients receiving esophagectomy; P = .186) and non-fatal severe adverse events (15.2% vs 27.7%; P = .001)-specifically lower proportions of esophageal fistula (0.3% of patients receiving ESD vs 16.4% for patients receiving esophagectomy; P = .001) and pulmonary complications (0.3% vs 3.6%; P = .004). After a median follow-up time of 21 months (range, 6-73 months), there were no significant differences between treatments in all-cause mortality (7.4% for ESD vs 10.9%; P = .209) or rate of cancer recurrence or metastasis (9.1% for ESD vs 8.9%; P = .948). Disease-specific mortality was lower among patients who received ESD (3.4%) vs patients who patients who received esophagectomy (7.4%) (P = .049). In Cox regression analysis, depth of tumor invasion was the only factor associated with all-cause mortality (T1a-m3 or deeper vs T1a-m2: hazard ration, 3.54; P = .04).
CONCLUSION: In a retrospective study of patients with T1am2/m3 or T1b EESCCs treated with ESD (n = 322) or esophagectomy (n = 274), we found lower proportions of patients receiving ESD to have perioperative adverse events or disease specific mortality after a median follow up time of 21 months. We found no difference in overall survival or cancer recurrence or metastasis in patients with T1a or T1b ESCCs treated with ESD vs esophagectomy.