Nomogram for predicting para-aortic lymph node metastases in patients with cervical cancer

Some patients with cervical cancer may benefit from prophylactic extended-field radiotherapy (EFRT). However, the indications for prophylactic EFRT were unclear. In this study, we constructed a nomogram to predict para-aortic lymph node (PALN) metastases in cervical cancer. Between January 2011 and June 2017, 1903 patients with stage IA-IVA cervical cancer received definitive radiotherapy. Patients who were diagnosed during 2011-2015 were assigned to a model development cohort (n = 1193) and others were assigned to a validation cohort (n = 710). Possible predictors were analysed using logistic regression models with model development cohort. A nomogram based on this work was constructed and validated. A total of 130 patients (6.8%) suffered from PALN metastases, with 81 patients (6.8%) in the model development cohort and 49 patients (6.9%) in the validation group. Multivariate analysis of the model development cohort demonstrated that histology, tumour size, bilateral pelvic lymph node (PLN) metastases, common iliac lymph node metastases and PLN convergence or muscle involvement were significant predictive factors of PALN metastases. The nomogram including these five factors showed good predictive accuracy (concordance index 0.916, 95% confidence interval, CI 0.881-0.952) in the model development cohort. In the validation cohort, the nomogram showed good discrimination (concordance index 0.949, 95% CI 0.911-0.988) and the predicted probability was close to the actual observed outcome. We have developed a robust tool that is able to predict PALN metastases in patients with cervical cancer. The nomogram could help physicians to decide whether prophylactic EFRT should be performed for patients with cervical cancer patients.