Prognostic models for predicting overall and cancer-specific survival of patients with initially diagnosed metastatic cervical squamous cell carcinoma: A study based on SEER database

Abstract

Cervical squamous cell carcinoma (CSCC) is the most common histological type of cervical cancer (CC). And mCSCC is the end stage of CSCC. The aim of this study was to develop prognostic nomograms that provide better predictions for overall survival (OS) and cancer-specific survival (CSS) in mCSCC patients. Data from patients with initially diagnosed mCSCC were extracted from the Surveillance, Epidemiology, and End Results (SEER) database between 2004 and 2015. The nomograms for OS and CSS were constructed based on Cox regression analysis. The validation of the newly established nomograms was evaluated by concordance index (C-index), calibration curves, and decision curve analyses (DCAs). A total of 2198 patients with mCSCC were included and randomly split into training (n = 1539) and validation (n = 659) cohorts in a 7:3 ratio. Multivariate analyses revealed that the prognostic variables significantly related to the OS and CSS were marital status, T stage, brain metastasis, lung metastasis, tumor size, number of positive lymph nodes, chemotherapy, and radiotherapy. The nomograms were constructed based on these factors. The C-index value of the nomograms for predicting OS and CSS was 0.714 and 0.683, respectively. The calibration curves of the nomograms showed good consistency between nomogram prediction and actual survival for both OS and CSS, and the DCAs showed great clinical usefulness of the nomograms. The mCSCC patients were classified into low- and high-risk groups based on the scores from the nomograms. In the validation cohort, mCSCC patients with low-risk had much higher OS and CSS than those with high-risk. We constructed nomograms for predicting the OS and CSS of patients with initially diagnosed mCSCC. Our models had satisfactory predictive performance and could be useful in survival prediction for mCSCC.