Development and Validation of an Individualized Nomogram for Predicting Overall Survival in Patients With Typical Lung Carcinoid Tumors.

摘要: Objective We aim to develop and validate an effective nomogram prognostic model for patients with typical lung carcinoid tumors using a large patient cohort from the Surveillance, Epidemiology, End Results (SEER) database. Materials methods Data between 2010 2015 were selected SEER database retrospective analysis. Univariate multivariate Cox analysis was performed clarify independent factors. Next, formulated predict probability of 3- 5-year overall survival (OS). Concordance indexes (c-index), receiver operating characteristic calibration curves used evaluate model. The randomly divided into training validation cohort. A established based on results indicated that age, sex, T stage, N surgery, bone metastasis variables OS. All these factors, except included in predicting internally externally validated c-indexes 0.787 0.817, respectively. For 3-year prediction, showed areas under curve cohorts 0.824 0.795, area 0.812 0.787, plots good agreement. Conclusion brings us closer personalized medicine maximization predictive accuracy prediction OS tumors.