The goal of this study is to develop robust tools for acquiring and analyzing in vivo MR metabolic data in patients with recurrent glioma that will be integrated with state of the art MR imaging methods to provide improved metrics for serially monitoring response to therapy. This addresses an important clinical problem, which is confounding the evaluation of novel treatments and making it difficult to make informed decisions about patient care. For subjects with high grade glioma, there is ambiguity between morphological changes that are caused by recurrent tumor and treatment effects such as gliosis and edema. Based on results from the current cycle of this R01, metabolites of interest for mapping out the true extent of recurrent tumor are myo-inositol (myo-I), N-acetylasparate (NAA), creatine (Cr) and choline (Cho). For subjects with an original diagnosis of low grade glioma who have a lesion which is increasing in size, the critical issue is to determine when they have undergone transformation to a more malignant phenotype that requires aggressive therapy. From our ex vivo analysis of tissue samples, additional metabolites of interest in this case are myo-I, glutamate, glutamine, glycine, glutathione and 2-hydroxyglutarate (2HG). In Specific Aims 1 and 2 we will optimize and then evaluate the test-retest accuracies of automated short TE 3D MRSI, as well as single voxel spectral editing and 2-D COSY sequences that will be used to detect in vivo levels of metabolites with complex coupling patterns and overlapping resonances. This will be done for both 3T and 7T scanners in order to determine which method and field strength will provide the most definitive results for patient studies. In Specific Aim 3 we will apply methods optimized for detecting 2HG to patients with recurrent low grade glioma to identify metabolic parameters associated with time to progression and overall survival. In Specific Aim 4, we will perform an analysis in patients with recurrent hig grade glioma using the most reliable strategy for obtaining short TE 3D MRSI data. The results from this study will provide new metrics for assessing response to therapy and for selecting alternative treatments. The objective of this study is to improve the management of patients with recurrent glioma using novel MR metabolic imaging methods that can distinguish treatment effects from tumor and are able to predict clinical outcome. This is critically important for making decisions about when and how to treat patients with novel therapies, which can provide ambiguous findings using conventional anatomic imaging.