Development and Validation of Nomogram Model to Predict Cancer-Specific Survival and Overall Survival in Adult Patients with Malignant Glioma: A Population-Based Analysis

Abstract Malignant glioma (MG) is the most common primary central nervous system malignancy; it is highly invasive and has a poor prognosis. Accurate and effective evaluation of prognostic factors is of great clinical significance for individualized treatment, prognosis and follow-up of MG patients. This study aimed to develop and validate a nomogram model to predict cancer-specific survival (CSS) and overall survival (OS) in adult MG patients. The data of adult MG patients were downloaded from the Surveillance, Epidemiology, and End Results (SEER) database. Univariate and multivariate Cox regression models were used to determine independent risk factors for the prognosis of adult MG patients. Nomograms were developed to predict the CSS and OS of adult MG patients based on a multivariate Cox regression model. Furthermore, calibration curves were used to verify the consistency of the nomogram model. The consistency index (C-index) and the area under the subject operating characteristic curve (AUC) values were used to test for the models’ accuracy and discrimination, respectively. Decision curve analysis (DCA) was used to evaluate the clinical applicability of the nomogram models. Subsequently, a risk score was calculated for each patient based on the nomogram. ROC was used to find the optimal cut-off value, and all patients were divided into the high-risk group or the low-risk group. The Kaplan-Meier (K-M) curve of the high-risk group and the low-risk group was drawn, and the survival difference between both two groups was tested by the Log-rank test. The differences in survival with respect to various surgical procedures and sequence number (SN) were analyzed. Finally, a computer and mobile calculators were designed to facilitate the use of the nomograms. In total, 37474 adult MG patients were included from 2004 to 2019. Univariate and multivariate Cox regression models revealed that age, marital status, race, tumor site, laterality, histology, stage, surgery, chemotherapy, radiotherapy and SN were independent risk factors for predicting CSS and OS in adult patients with MG. The calibration curve demonstrated that the model had good consistency. The C-index and AUC verified the discrimination of the model, and the DCA revealed that the nomogram model had superior clinical applicability to the stage model of the SEER database. The models can help doctors and patients to make better clinical decisions.