Treatment strategies in recurrent esophageal or junctional cancer.

Little evidence is available about survival rates in patients with recurrent disease after potentially curative surgery for esophageal or junctional cancer. Only in limited occasions, potentially curative salvage strategies are available. The aim of this study is to analyze survival rates and patterns of dissemination, and to identify independent prognostic factors in a consecutive series of patients who develop recurrent esophageal or junctional cancer. Between 1994 and 2015, patients who developed disease recurrence after neoadjuvant chemo(radio)therapy followed by radical esophagectomy for esophageal or junctional cancer were retrospectively analyzed. The Kaplan-Meier estimates were performed to calculate and compare overall survival between patients with different patterns of dissemination and to compare between different treatment strategies. Furthermore, univariate and multivariate Cox-regression analyses were performed to identify independent prognostic factors for post recurrence survival. In this study, we included 219 patients. The median overall survival of all included patients was 3.2 months (range: 0.0-101.1 months). The median overall survival in patients with exclusively locoregional recurrence (n = 23, 10.8%) was 4.9 months (range: 0.1- 55.6) and 2.9 months (range: 0.0-101.1) in patients who had distant metastases (n = 189, 89.2%), P = 0.003. Patients who received treatment aimed at complete tumor eradication (n = 28, 13.7%) had a median overall survival of 13.6 months (range: 1.1-101.1) and palliative treated patients (n = 94, 46.1%) of 4.7 months (range: 0.3-25.6), P < 0.001. In a selected group of patients survival of more than 20 months was achieved. Univariate and multivariate Cox-regression analysis showed that a higher age at the diagnosis of recurrent disease (hazard ratio: 1.087, P ≤ 0.001), an irradical resection of the primary tumor (hazard ratio: 3.355, P = < 0.001), the number of positive lymph nodes after neoadjuvant therapy (hazard ratios: ypN2 = 1.724 (P = 0.024) and ypN3 = 2.082 (P = 0.028) and the presence of a single hematogenous distant metastases (hazard ratio: 2.281, P = 0.003) or more than one hematogenous distant metastasis (hazard ratio: 2.385, P = 0.005) were associated with a shorter postrecurrence survival. The prognosis of patients who develop recurrent esophageal or junctional cancer is poor. In a selected group of patients however relatively long survival can be achieved. This offers new perspectives to improve treatment strategies and survival rates.