Disease-free and overall survival in nonmetastatic esophageal or gastroesophageal junctional cancer after treatment with curative intent: A nationwide population-based study.

Abstract

246 Background: Among patients with potentially curable esophageal cancer (EC) or gastroesophageal junctional cancer (GEJC) treated with curative intent, survival remains poor and around half of these patients have disease recurrence within a few years. This study addresses the need for real-world data on disease-free survival (DFS) and overall survival (OS) in patients with EC or GEJC who underwent potentially curative treatment. Methods: Patients selected from the nationwide Netherlands cancer registry (NCR) had received a primary diagnosis of non-metastatic EC or GEJC (excluding patients with T4b tumors) in 2015 or 2016 and received treatment with curative intent. Curative intent was defined as receiving resection (with or without [neo]adjuvant therapy) or definitive chemoradiotherapy (dCRT) without surgery. DFS and OS were analysed using Kaplan-Meier curves with Log-Rank test from resection date or end of dCRT. A sub-analysis was performed for NCR patients selected to align with the population of the CheckMate-577 phase 3 study of adjuvant nivolumab, i.e. patients with non-cervical stage II/III disease, R0 resection and residual pathological disease after neoadjuvant CRT (nCRT) and surgery. Results: We identified 1916 patients of median age of 67 years and predominantly male (76%). The majority (79%) received surgery and 21% of patients received dCRT. In resected patients, 83% received nCRT, 10% neoadjuvant chemotherapy (with or without adjuvant CRT) and 7% received no (neo)adjuvant treatment. Compared to the resected group, the population receiving dCRT had significantly fewer males (65% vs 78%), a higher median age (72 vs 65 years) and worse performance status. Patients receiving dCRT significantly shorter median DFS (14.2 months) and OS (20.9 months) compared to resected patients (DFS: 26.4 months, p < 0.001; OS: 40.5 months, p < 0.001). The 1- and 3-year DFS probabilities were 68% and 44%, respectively, in resected patients, and 56% and 24%, respectively, in patients receiving dCRT. In patients receiving nCRT followed by surgery, the median DFS and OS were 25.2 and 38.0 months, respectively, and 1- and 3-year DFS probabilities were 67% and 43%, respectively. In the sub-analysis (n = 725) the median DFS and OS were 19.2 and 29.4 months, respectively, and the 1- and 3-year DFS rates were 62% and 36%, respectively. Conclusions: Although patients are treated with curative intent, a considerable amount of patients with non-metastatic EC or GEJC experienced recurrence within two years. Resected patients had a higher DFS and OS compared to patients receiving dCRT.