Metastatic Lymph Node Station Count and Risk of First Distant Metastasis After Neoadjuvant Therapy in Esophageal Squamous Cell Carcinoma.

Abstract

After neoadjuvant therapy for esophageal squamous cell carcinoma, pathologic nodal staging counts metastatic nodes without capturing their anatomic distribution and is susceptible to stage migration. We evaluated whether involved lymph node station count improves 3-year distant metastasis prediction beyond conventional nodal staging. In 557 patients with esophageal squamous cell carcinoma undergoing neoadjuvant therapy and McKeown esophagectomy with three-field lymph node dissection at a single center from 2015 to 2023, two Fine-Gray competing-risks models differed solely in the nodal variable: pathologic nodal category versus station count. Discrimination, calibration, and clinical utility were cross-validated at 3 years. Three-year cumulative incidence of first distant metastasis increased from 13.0% with 0 involved stations to 19.9%, 29.6%, and 41.4% with 1, 2, and 3 or more stations, respectively (Gray P < 0.001). Station count yielded higher discrimination (area under the curve difference 0.0221; 95% CI, 0.0092-0.0355) and better calibration (Brier score difference -0.0057; 95% CI, -0.0094 to -0.0021). Each additional involved station independently raised risk (subdistribution hazard ratio 1.460; 95% CI, 1.287-1.657; P < 0.001). Net benefit favored station count at threshold probabilities of approximately 0.22 to 0.32. After neoadjuvant therapy and three-field lymph node dissection for esophageal squamous cell carcinoma, station count provided a monotonic risk gradient and modestly improved distant metastasis prediction over conventional nodal staging. As an adjunct to conventional nodal staging, station count may refine postoperative risk stratification. External validation is required before routine clinical use.