Predictors of staging accuracy, pathologic nodal involvement, and overall survival for cT2N0 carcinoma of the esophagus

OBJECTIVE: Clinical T2N0 esophageal carcinoma is a heterogenous disease frequently complicated by inaccurate staging. Incorrect staging may lead to suboptimal treatment for patients with unidentified local-regionally advanced disease. Therapeutic options for these patients remain controversial. We sought to evaluate the outcomes of patients with cT2N0 who underwent esophagectomy as either primary therapy or after neoadjuvant treatment. METHODS: This was a multi-institutional collaboration of 26 high-volume esophageal centers. Patients with complete staging who underwent elective resection from 2002 to 2012 were included. Three treatment groups were identified; primary esophagectomy, preoperative chemotherapy, and preoperative chemoradiation (CXRT). Pretreatment variables were explored for independent predictors of long-term outcomes. The primary esophagectomy subgroup was evaluated for stage migration. RESULTS: In total, 767 patients were evaluated; 35% (268) had preoperative therapy (195 CXRT, 73 chemotherapy). Staging accuracy was 14% (70/499), with pT < 2 identified in 45% (222) and pN > 0 in 39% (195). Preoperative treatment modality (none, CXRT, chemotherapy) was not identified as a predictor of outcome (median survival 63, 70, 71 months, respectively, P = .956). Longitudinal tumor length >3.25 cm was predictive of pN+ for the primary esophagectomy cohort as well as adenocarcinoma histology only (odds ratio 2.2 and 2.4, respectively, P < .001). CONCLUSIONS: Current treatment options for patients with cT2N0M0 do not reveal a comparative survival advantage to preoperative therapy. Pretreatment tumor length can identify a subgroup of patients at risk for understaging (pN+). The incidence of overstaging suggests that organ-sparing approaches (endoscopic resection) may play a future role in appropriately selected patients.