Molecular studies of high-grade gliomas (HGGs) have highlighted the heterogeneity of these tumors, and have linked molecular signatures to their natural history and to differences in survival rates. The ability to identify such molecular subtypes of tumors is essential for guiding therapeutic advances. We report the development and validation of a robust risk-score model highly associated with the outcome of patients with newly diagnosed HGG. We considered a supervised approach to account for the WHO grade of malignity when deriving gene biomarkers associated with outcome. We performed a meta-analysis of HGGs microarray data sets (267 patients) to identify such biomarkers from a robust signature related to tumor aggressiveness. These biomarkers were used to construct a risk-score equation based on a Cox proportional hazards model. The model best associated with overall survival (OS) and with good discrimination (C-statistic) was based on the expression of 4 genes. Patients... were ranked according to their risk score and stratified into 2 groups. Low-risk score patients had a median OS longer than high-risk score patients (46.6 vs 11.7 months, P , .001). These results were validated on an independent microarray study of 59 patients. We performed RT-qPCR validation on an independent set of HGGs (194 patients) and compared the performances of our risk-score model with the prognostic value of currently admitted clinical and molecular risk factors. Two multivariate models were built, including age, treatment, grade, RTOG RPA classes, MGMT methylation status, and IDH1 mutational status; one with and one without the 4-gene expression risk score. These models were used to estimate the prognostic value of the gene expression risk score for 176 patients with complete data for all variables and for a subset of 105 patients treated with temozolomide chemoradiation. This analysis confirmed that both the mutations of IDH1 and the presence of MGMT promoter methylation were associated with a survival benefit (P , .01 in the whole cohort and P , .05 in the subset). It also showed that the 4-gene risk score was strongly associated with OS in these two groups, independently from clinical and molecular risk factors (P , .01). Each time, the model discrimination improved significantly with the addition of the 4-gene risk score (0.816 vs 0.846, P , .001 and 0.792 vs 0.822, P , .001, respectively), showing that it added beyond standard clinical parameters and beyond both the MGMT methylation status and the IDH1 mutational status. One explanation for the association between the 4-gene signature and clinical outcome is that it may detect the molecular fingerprints inherent to glioma aggressiveness. These results suggest the importance of this 4-gene signature as a stratification factor for future comparative therapeutic trials, though it needs to be further investigated in a prospective clinical study.