Abstract
BACKGROUND: Endoscopic resection has emerged as a treatment option for T1a esophageal cancer, but the impact of age on patient selection for surgery versus endoscopic resection has not been well studied. We hypothesized that endoscopic resection would be associated with improved survival compared with surgery in older patients with early esophageal cancer and worse survival in younger patients. METHODS: The National Cancer Database was used to identify patients with cT1aN0M0 esophageal cancer (2010-2015) treated with endoscopic resection or esophagectomy. The relationship between age and treatment effect on survival was modeled with an interaction term in a Cox proportional hazards regression. The primary outcome was overall survival. RESULTS: A total of 831 patients met study criteria: A total of 448 patients (54%) underwent endoscopic resection, and 383 patients (46%) underwent esophagectomy. In a multivariable Cox model, the interaction term between patient age and type of treatment was nonsignificant (P = .11), suggesting that age did not influence the effect of endoscopic resection compared with surgery on survival. In 285 propensity score-matched patients receiving endoscopic resection or surgery, surgery was associated with similar survival compared with endoscopic resection (hazard ratio, 1.40; 95% confidence interval, 0.97-2.03). CONCLUSIONS: Endoscopic resection was associated with similar survival compared with surgery in patients with cT1a esophageal cancer regardless of age. Endoscopic resection can be considered for patients at low risk of nodal involvement across all age groups as an alternative to surgery for T1a esophageal cancer.