Development and validation of a nomogram for predicting pelvic lymph node metastasis and prognosis in patients with cervical cancer

OBJECTIVE: Cervical cancer (CC) is one of the main causes of death among gynecological malignancies. Patients with CC with lymph node metastasis (LNM) have poor prognoses. We investigated the risk factors and prognosis of LNM in patients with CC patients using data from the SEER database. METHODS: We collected the information of cervical cancer patients registered in SEER database from 2010 to 2015. The dataset was divided into a training set and a validation set at a 7:3 ratio. LASSO regression analysis was used to evaluate risk factors for LNM in patients with CC. Using the results, we established a nomogram prediction model. C-index, ROC curves, calibration curves, decision curve analysis, and clinical impact curves were used to evaluate the prediction performance of the model. RESULTS: We included 14,356 patients with CC in the analysis. Among these, 3997 patients were diagnosed with LNM. A training set (10,050 cases) and a validation set (4306 cases) were used for the following analysis. We established nomogram LNM prediction models for the patients with T(1-2)-stage CC. The C-indices for the internal and external validations of the prediction models were 0.758 and 0.744, respectively. In addition, we established a prognostic nomogram for all CC patients with LNM, and the internal and external validation C-indices were 0.763 and 0.737. CONCLUSION: We constructed a quantitative and visual predictive nomogram that predicted prognosis of patients with LNM in CC to provide clinicians with a reference for diagnosis and treatment.