AimsThe purpose of this study was to investigate the incidence, survival and prognostic factors of cervical cancer with lung metastasis at the initial diagnosis and to develop a visual nomogram to predict the prognosis of these patients. MethodsWe used the Surveillance, Epidemiology and End Results (SEER) database to screen patients diagnosed with cervical cancer from 2010 to 2015. After strict inclusion and exclusion, the chi-square test was used to evaluate the differences in the clinical characteristics of patients with cervical cancer, and then we used Kaplan–Meier method to perform survival analysis among cervical cancer patients with lung metastasis. Next, univariate and multivariate Cox proportional hazard regression models were used to estimate prognostic factors of these patients and we developed a visualized and novel nomogram to judge the prognosis. Results476 patients with lung metastasis and 12,016 patients without lung metastasis were included in this study. The incidence of lung metastasis was higher in unmarried white cervical cancer patients between the ages of 40 and 60, and grade III cervical squamous cell carcinoma patients were more likely to have lung metastasis. In addition, grade, surgery, radiotherapy, sequence of surgery and radiotherapy and chemotherapy were significantly related to the outcomes of cervical cancer patients with lung metastasis. Furthermore, our nomogram could predict the 3-year and 5-year overall survival (OS) of these patients. Finally, the AUC of 3-year OS and 5-year OS were confirmed to be 0.969 and 0.939 respectively by ROC curves, with good consistency. ConclusionsAge at diagnosis, race, marital status, and characteristics of the tumor can influence the incidence of lung metastasis in cervical cancer patients. Besides, grade, surgery, radiotherapy, sequence of surgery and radiotherapy and chemotherapy may deeply affect the prognosis of cervical cancer patients with lung metastasis. The nomogram built in this study may help clinicians to formulate individualized treatment strategies and encourage the development of more and more comprehensive and accurate predictive models.
Read full abstract