Related Experiment Video
Updated: Jun 12, 2025
Generation of CAR T Cells for Adoptive Therapy in the Context of Glioblastoma Standard of Care
Published on: February 16, 2015
Conditional survival and changing risk profile in patients with gliosarcoma
Lei Xu1, Zhihao Yang1, Huawei Chen1
1Department of Neurosurgery, Shaoxing Central Hospital, The Central Affiliated Hospital, Shaoxing University, Shaoxing, China.
Background:
Conditional survival (CS) considers the duration since the initial diagnosis and can provide supplementary informative insights. Our objective was to evaluate CS among gliosarcoma (GSM) patients and develop a CS-incorporated nomogram to predict the conditional probability of survival.
Methods:
This retrospective study using the Surveillance, Epidemiology, and End Results (SEER) database included patients with GSM between 2000 and 2017. The CS was defined as the probability of surviving additional y years after already surviving for x years. The formula utilized for CS was: CS(y|x) = S(y + x)/S(x), where S(x) denotes the overall survival at x years. Univariate Cox regression, best subset regression (BSR) and the least absolute shrinkage and selection operator (LASSO) were used for significant prognostic factors screening. Following this, backward stepwise multivariable Cox regression was utilized to refine predictor selection. Finally, a novel CS-integrated nomogram model was developed and we also employed diverse evaluation methods to assess its performance.
Results:
This study included a total of 1,015 GSM patients, comprising 710 patients in training cohort and 305 patients in validation cohort. CS analysis indicated a gradual increase in the probability of achieving a 5-year survival, ascending from 5% at diagnosis to 13, 31, 56, and 74% with each subsequent year survived after 1, 2, 3, and 4 years post-diagnosis, respectively. Following variable screening through univariate Cox regression, BSR, and LASSO analysis, five factors-age, tumor stage, tumor size, radiotherapy, and chemotherapy-were ultimately identified for constructing the CS-nomogram model. The performance of the nomogram model was validated through discrimination and calibration assessments in both the training and validation cohorts. Furthermore, we confirmed that the effectiveness of the CS-nomogram in stratifying GSM patient risk status.
Conclusion:
This nationwide study delineated the CS of patients diagnosed with GSM. Utilizing national data, a CS-nomogram could provide valuable guidance for patient counseling during follow-up and risk stratification.
