Risk assessment of lymph node metastasis in early gastric cancer: Establishment and validation of a Seven-point scoring model

Background: Treatment options for early gastric cancer have evolved toward achieving accurate evaluation of lymph node metastasis. This study aimed to investigate risk factors of lymph node metastasis in patients with early gastric cancer and establish a risk score model to guide the selection of optimal treatment. Methods: The clinicopathological characteristics of 351 patients with early gastric cancer from January 2016 to December 2018 were reviewed retrospectively. On the basis of the independent risk factors determined by multivariate binary logistic regression analysis, we established a risk score model for predicting lymph node metastasis and then verified it. The receiver operating characteristic curves were plotted using the test and validation sets. The area under the receiver operating characteristic curve was used to assess the discriminant ability of the model. Results: Lymph node metastasis was observed in 10.5% (37/351) of early gastric cancer cases. Patients with early gastric cancer were grouped based on the independent risk factors for lymph node metastasis (tumor size, depth, histological type, and lymphovascular involvement) determined by multivariate analysis. A 7-point risk score model was established to predict the risk of lymph node metastasis. The area under the receiver operating characteristic curve in the development and validation sets were 0.839 (95% confidence interval, 0.769%-0.910%) and 0.820 (95% confidence interval, 0.711%-0.930%), respectively. Conclusion: A feasible risk score model for lymph node metastasis was established to guide the optimal treatment of patients with early gastric cancer early gastric cancer. (C) 2021 Elsevier Inc. All rights reserved.

Automated summary

This study established a risk score model for lymph node metastasis to guide the optimal treatment of patients with early gastric cancer. The model includes independent risk factors such as tumor size, depth, histological type, and lymphovascular involvement.