Predicting lymph node metastasis in early gastric cancer patients: development and validation of a model.

To develop and validate a model to predict possibility of lymph node metastasis (LNM) in early gastric cancer. An LNM prediction model was developed by logistic regression based on the demographics or characteristics of the tumor (N = 746) and then internally and externally validated (N = 126). Four variables, lymphovascular invasion, differentiated types, diameter of tumor and T stage were screened into the model. The area under the receiver-operating characteristic curve of the model was 0.861 (95% CI: 0.851-0.864) in internal validation and 0.911 (95% CI: 0.848-0.974) in the validation set. The model shows excellent discrimination and calibration performance, and is potential to be a useful clinical model to predict the risk of LNM in early gastric cancer.