ObjectiveThis study aimed to determine the prognostic factors for disease-specific survival (DSS) of glioblastoma (GBM) and establish a corresponding effective nomogram for clinical prediction.Methods This study was based on Surveillance, Epidemiology, and End Results database between 2004 and 2015. Kaplan-Meier survival analysis was used to evaluate the effect of various prognostic factors on DSS. Lasso regression was used to determine the independent prognostic factors of DSS and multivariate cox regression analysis was performed correspondingly. Additional restricted cubic spline cox regression was used to analyze the trend of the risk effect (hazard ratio) of continuous variables on DSS. Based on the multivariate cox regression model, a nomogram was established to predict DSS. ResultsThe average age at diagnosis of all enrolled patients was 59.8±12.2 years, of which 40.5% were women and 59.5% were men. Lasso regression analysis showed that age at diagnosis, sex, marital status, race, tumor size, primary site, laterality, surgery, radiotherapy and chemotherapy, radiotherapy sequence with surgery, and year of diagnosis were independent prognostic factors for DSS. Multivariate cox regression analysis showed that elderly, males, unmarried status, larger tumors were all risk factors for DSS. Restricted cubic spline cox regression showed that the risk of death from GBM was significantly increased for the elderly, especially older than 75 years. When the tumor was smaller than 75mm, an increasing risk linearly was associated with DSS, but the risk effect remained constant after 75mm. Constructing the nomogram to predict the DSS probability of 1-, 3- and 5-year respectively, and its good predictive performance was proved by the calibration curve.ConclusionThe advanced age was one of the significant risk factors for GBM. How the change of tumor size affected DSS needed further study and discussion. The established nomogram was robust in predicting 1-, 3-, and 5- year DSS.