Abstract The Cancer Genome Atlas Project (TCGA) has produced a wide array of omic data on glioblastoma multiforme (GBM) tumors, and has resulted in several works on the molecular subtyping of GBM tumors. Despite its richness, the TCGA dataset currently contains very limited data for low-grade gliomas and progressive GBMs, preventing analysis of progression models and prognostic biomarker discovery for these tumor types. A complementary dataset, the NCI Repository for Molecular Brain Neoplasia Data (REMBRANDT), includes a collection of samples for low-grade and progressive gliomas. This data contains some paired tumors that progressed to GBM and like TCGA, includes pathological and survival information. Our in silico study of low-grade and progressive gliomas has leveraged this complementary data to examine the progression of low-grade gliomas to GBM. The REMBRANDT dataset includes 176 samples with survival outcome data, including 101 grade IV GBM, 32 astrocytomas, and 43 oligodendrogliomas. Of the astrocytomas and oligodendrogliomas, there are 31 grade II samples and 31 grade III samples, with the remainder lacking grade data. We have investigated the gene expression profiles of the REMBRANDT samples to compare them to the established classes from TCGA and other studies. Candidate prognostic biomarkers have been identified to predict the progression of low-grade gliomas. We have found that the TCGA Proneural subtype predicts improved clinical outcome for low-grade oligodendrogliomas, with significantly better survival among all 176 low and high-grade samples (log rank test p = 1.27e-08). Moreover, analysis of 62 grade II and grade III samples determined that the Proneural class still has significantly better outcome (p= 0.0033). Further sub-analysis to control for oligodendrogliomas and grade, found that grade III oligodendrogliomas with the Proneural signature still had significantly improved survival (p < 0.05, n = 12). Stratification using the signatures developed by Li et al that define six oligodendrioma and glioma subtypes demonstrated that the OA subclass had significantly better survival (log rank test p = 9.14e-08). However, these signatures were not significant in multivariate analysis after controlling for grade and disease subtype. Thus, the gene expression patterns identified in the TCGA analysis of GMB has intrinsic prognostic value for low-grade oligondendriomas, and likely represents important differences in tumor biology with implications for treatment and therapy. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 113.