PT  - JOURNAL ARTICLE
AU  - Xiaohua Qian
AU  - Hua Tan
AU  - Wei Chen
AU  - Weiling Zhao
AU  - Michael D. Chan
AU  - Xiaobo Zhou
TI  - Radiogenomics-based Risk Prediction of Glioblastoma Multiforme with Clinical Relevance
AID  - 10.1101/350934
DP  - 2018 Jan 01
TA  - bioRxiv
PG  - 350934
4099  - http://biorxiv.org/content/early/2018/06/20/350934.short
4100  - http://biorxiv.org/content/early/2018/06/20/350934.full
AB  - GBM is the most common and aggressive primary brain tumor. Although the TMZ-based radiochemotherapy improves overall GBM patients’ survival, it also increases the frequency of false positive post-treatment magnetic resonance imaging (MRI) assessments for tumor progression. Pseudoprogression is a treatment-related reaction with an increase in contrast-enhancing lesion size at the tumor site or resection margins which mimics tumor recurrence on MRI. Accurate and reliable prognostication of GBM progression is urgently needed in the clinical management of GBM patients. Clinical data analysis indicates that the patients with PsP had superior overall and progression-free survival rates. In this study, we aimed to develop a prognostic model to evaluate tumor progression potential of GBM patients following standard therapies. We applied a dictionary learning scheme to obtain imaging features of GBM patients with PsP or TTP from the Wake dataset. Based on these radiographic features, we then conducted radiogenomics analysis to identify the significantly associated genes. These significantly associated genes were then used as features to construct a 2YS logistic regression model. GBM patients were classified into low-and high-survival risk groups based on the individual 2YS scores derived from this model. We tested our model using an independent TCGA dataset and found that 2YS scores were significantly associated with the patients’ overall survival. We further used two cohorts of the TCGA data to train and test our model. Our results show that 2YS scores-based classification results from the training and testing TGCA datasets were significantly associated with the overall survival of patients. We also analyzed the survival prediction ability of other clinical factors (gender, age, KPS, normal cell ratio) and found that these factors were not related or weakly correlated with patients’ survival. Overall, our studies have demonstrated the effectiveness and robustness of the 2YS model in predicting clinical outcomes of GBM patients after standard therapies.