Effect of nodal status compared to tumor response after neoadjuvant radiation on outcomes for patients with esophageal cancer.

Abstract

147 Background: The CROSS trial established the role of neoadjuvant radiation in the treatment of esophageal adenocarcinoma (EAC). While response to radiation is an important factor in predicting long-term outcomes, the vast majority of patients succumb to systemic disease. The purpose of this study is to assess predictors of survival in patients with EAC following radiation therapy. Methods: All patients who underwent resection after radiation therapy for EAC at a single institution were retrospectively identified from January 2004 to December 2014. Patients who died within 30 days of surgery were excluded. Cox-proportional hazard analyses were performed to identify clinico-pathological factors associated with survival after surgery. Results: In the time period, 334 patients underwent esophagectomy for EAC. Univariable/multivariable analyses are shown in the table. The presence of a pathologic complete response (pCR) did not correlate to survival. The most important factors in predicting outcome were pre-operative albumin and initial lymph node stage by endoscopic ultrasound (EUS). Pre-treatment N0 patients had better survival than N1 patients (median survival 37.2 vs. 16.3 months, P < 0.0001). Patients who remained N0 after radiation had much better outcomes than those that either developed N1 disease after radiation or were initially staged as N1 (stayed N0, N = 126, median survival 52.0 months; N1→N0, N = 85, median survival 22.9 months; N0→N1, N = 30, median survival 15.3 months; persistent N1, N = 44, median survival 11.4 months; P < 0.0001). Conclusions: Pathologic response to radiation does not predict outcomes for patients with EAC. Patients with node positive EAC have poor outcomes even after neoadjuvant radiation therapy. These patients are at an increased risk of distant disease and should be offered additional systemic therapies prior to surgical resection. [Table: see text]