Fig. 3

Methylation diagnostic model development and validation for lymph node detection in EGC. a, b ROC curves of the 3-marker methylation model in the training and test set, respectively. c The sensitivity, specificity, and accuracy of the model were determined by the cutoff value (0.2327) in the training and test set in the model development cohort and independent validation cohort, respectively. d, e LNM risk score of the 3-marker methylation model between LNM+ EGC samples (n = 36) and LNM− EGC samples (n = 115) in the training and test set. The dotted line shows the cutoff value (0.2327) to distinguish LNM+ from LNM− samples. f LNM risk score of the 3-marker methylation model between LNM+ samples (n = 32) and LNM− EGC samples (n = 98) in the validation cohort. g ROC curves of the 3-marker methylation model in the validation cohort. h LNM risk score of the 3-marker methylation model in different clinicopathological features including the ulcerative type, differentiation, invasional depth, and lymphovascular invasion status. The data are shown as median the interquartile range. Statistical significance was assessed using an unpaired t test (two-tailed). *p < 0.05, **p < 0.01, ***p < 0.001