Background: At present, there is no unified treatment for the evaluation and management of gastric low-grade intraepithelial neoplasia (LGIN) all over the world.
Methods: Patients who were Helicobacter pylori eradicated, with low-grade gastric intraepithelial neoplasia were gathered.Several demographic and clinicopathological characteristics were described and analyzed retrospectively by LASSO regression analysis and multivariable logistic regression. Then the predictive nomogram was established. C-index, the area under the receiver operating characteristic curve (AUC) , calibration plot and decision curve analysis (DCA) were used to evaluate the accuracy and reliability of the model.
Results: A total of 309 patients with LGIN were included, divided into training groups and validation groups randomly. LASSO regression analysis and multivariable logistic regression showed that six variables, gender, size, location, border line, number and erosion were independent risk factors for progression of gastric LGIN. The nomogram model displayed good discrimination with a C-index of 0.765 (95% confidence interval: 0.702–0.828). High C-index value of 0.768 could still be reached in the internal validation. The accuracy and reliability of the model was also verified by the AUC of 0.764 in the training group and 0.757 in the validation group. The calibration curve showed the model was in good agreement with the actual results as well. Decision curve analysis suggested that the predictive nomogram had clinical utility.
Conclusions: A predictive nomogram model was successfully established and proved to identify high-risk groups with possible pathologic upgrade in patients with gastric LGIN. It suggested that after identifying high-risk patients, strengthening follow-up or endoscopic treatment may benefit in improving the detection rate or reducing the incidence of gastric cancer, which providing a reliable basis for the treatment of LGIN.