Establishment of Risk Prediction Model and Nomogram for Lymph Node Metastasis of Cervical Cancer: Based on SEER Database

Objective: To predict the risk factors of lymph node metastasis in cervical cancer by using large sample clinical data, and to construct and verify the nomogram for predicting lymph node metastasis. Methods: A total of 5940 patients with cervical cancer from 2004 to 2015 in the National Cancer Institute Surveillance Epidemiology and End Results database were retrospectively screened and randomly assigned to training group (n = 4172) and validation group (n = 1768). Multivariate Logistic regression analysis was used, and the optimal model was selected according to AIC or BIC and likelihood ratio test, and a nomogram was drawn. The accuracy and robustness of the prediction model were evaluated in three aspects: discrimination, calibration and clinical net benefit. Results: The prediction model based on race, tumor tissue differentiation degree, tumor histopathological type, distant metastasis of tumor, tumor diameter and other risk factors was successfully established and a nomogram was constructed. The AUCs of training group and validation group were: 0.736 and 0.714, respectively. And the p-values of the Hosmer-Lemeshow test were 0.28 and 0.11, respectively. The calibration curve was in good agreement with the ideal curve. It had high accuracy and applicability after internal verification. Conclusion: A prediction model is constructed based on the risk factors of lymph node metastasis of cervical cancer. The nomogram has a good effective prediction and can provide a theoretical basis for clinicians to assess the disease quickly before surgery.