A prognostic model guides surgical resection in cervical squamous cell carcinoma

BACKGROUND: To explore the independent risk factors of cervical squamous cell carcinoma and establish a Nomogram model to predict the prognosis of patients. METHODS: We randomly divided the total data of patients with cervical squamous cell carcinoma from 2010 to 2015 obtained from the SEER database and cleaned them into training and verification cohorts. The Cox proportional hazard regression model was used to perform univariate and multivariate analyses on the three cohorts of data including the total data. After the intersection, the independent factors and their nomograms with statistical significance were obtained, and the degree of differentiation and calibration between predicted results and real values were obtained by using C-index and calibration map respectively. In addition, the ROC curve was used for correction and evaluation, and the 1-, 3- and 5-year overall and specific survival rates of patients were finally predicted. RESULTS: We found age, surgical condition of the primary site and tumor size were all independent factors of cervical cancer. The high-risk survival rates of patients at 1, 3 and 5 years were 77.7%, 48.6% and 36.4%, respectively. We determined that minimally invasive hysterectomy and uterine-preserving surgery (UPS) have a better survival rate for early (stage I) tumors or tumor diameter less than 20 mm. For the late (stage III–IV) or tumor diameter greater than 20 mm, auxiliary open hysterectomy after radiotherapy, and requires careful evaluation of the postoperative residual tumor is the best policy. CONCLUSIONS: The constructed nomograms could predict overall survival with good performance, and guide surgical resection in cervical squamous cell carcinoma.