Construction and validation of a risk-scoring model to predict lymph node metastasis in T1b–T2 esophageal cancer

Background Lymph node status is an important factor in determining preoperative treatment strategies for stage T1b–T2 esophageal cancer (EC). Thus, the aim of this study was to investigate the risk factors for lymph node metastasis (LNM) in T1b–T2 EC and to establish and validate a risk-scoring model to guide the selection of optimal treatment options. Methods Patients who underwent upfront surgery for pT1b–T2 EC between January 2016 and December 2022 were analyzed. On the basis of the independent risk factors determined by multivariate logistic regression analysis, a risk-scoring model for the prediction of LNM was constructed and then validated. The area under the receiver operating characteristic curve (AUC) was used to assess the discriminant ability of the model. Results The incidence of LNM was 33.5% (214/638) in our cohort, 33.4% (169/506) in the primary cohort and 34.1% (45/132) in the validation cohort. Multivariate analysis confirmed that primary site, tumor grade, tumor size, depth, and lymphovascular invasion were independent risk factors for LNM (all P < 0.05), and patients were grouped based on these factors. A 7-point risk-scoring model based on these variables had good predictive accuracy in both the primary cohort (AUC, 0.749; 95% confidence interval 0.709–0.786) and the validation cohort (AUC, 0.738; 95% confidence interval 0.655–0.811). Conclusion A novel risk-scoring model for lymph node metastasis was established to guide the optimal treatment of patients with T1b–T2 EC.