Appraisal of a revised lymph node classification system for esophageal squamous cell cancer.

Abstract

Node-positive patients with esophageal carcinoma constitute a heterogeneous population with a variable prognosis, which the current staging system insufficiently addresses. To that end, 863 patients with a curative resection for esophageal squamous cell carcinoma were analyzed to evaluate a useful and simple nodal classification system. Along with standard conventional clinicopathologic factors, data for metastatic lymph node (MLN) number, metastatic to examined LN ratio (MLN ratio), and MLN size were evaluated. The greatest microscopic dimension of the metastatic tumor inside the largest MLN (MLN size) was measured on histopathologic slides. Patients with MLNs were classified into n1 (< 9 mm) and n2 (> or = 9 mm) groups, according to size of MLNs (n-stage). The paratracheal LNs most frequently contained the largest MLN and among them the right recurrent laryngeal LNs were the most common site (81.8%). Patients were stratified into significant groups by all the nodal criteria. In multivariable analysis, MLN size n-stage and MLN ratio N-stage were the best independent predictors for disease-free and overall survival, respectively. In the disease-free survival, MLN ratio N-stage subcategories were divided into prognostic groups according to the n-stage. A combined nodal staging strategy combining the n-stage and N-stage had the strongest prognostic value and was used for the tumor-node-metastasis classification with distinct separation of patients into prognostic groups. Results of this study indicate that the MLN size may serve as an accurate metric to classify node-positive patients and a combination of the MLN ratio and size may have synergism in classifying node-positive patients into prognostically homogenous groups.