Studies suggest that abnormal elevations in extracellular glutamate can act as a pathogenic driver of schizophrenia, including in the hippocampus a brain region that neuroimaging studies suggest might be affected first and foremost. If this formulation is correct, then high glutamate should be considered a molecular target for drug discovery and agents that reduce extracellular glutamate should be an effective intervention. Pomaglumetad methionil (which we will call ?POMA?) is one such agent, because as an agonist of presynaptic metabotropic glutamate 2/3 receptors this class of drug has been found to reduce glutamate release. Nevertheless, trials that have used POMA have to date shown little efficacy in patients with schizophrenia. Despite these failures, we believe that, based on the accumulative evidence, high glutamate is indeed a valid target and that two reasons might account for these initial failures. First, as these trials did not use a reliable brain biomarker of glutamate elevations there was no readout of ?target engagement?, and thus it remains possible that these trials were false negatives. Second, and more importantly, schizophrenia is now thought to start in a prodromal phase before the onset of psychosis, and there is evidence to suggest that the prodromal stage might represent a unique time-window for therapeutic interventions, and so it possible that the drug was given at the wrong stage of the disease. Pathophysiological support for these conclusions comes from our studies (Schobel et al, Neuron, 2013) in which we applied fMRI indicators of metabolism and volumetric MRI to patients in prodromal stages of the disease who then progressed to the psychotic stage, and to a mouse model of the disease. Collectively, these studies suggested that glutamate is a driver of hippocampal dysfunction, that fMRI measures are sensitive to glutamate elevations, and that glutamate-reducing drugs can ameliorate hippocampal dysfunction. We also have completed preliminary studies using magnetic resonance spectroscopy, in which we show that hippocampal glutamate is elevated in prodromal stages of the disease. Based on the results of these and other studies, we have recently proposed a mechanistic model of disease progression (Small, Neuron, 2014), which predicts that because of its distinct pathophysiological features the prodromal stage of disease is a unique time-window that is most amenable to glutamate-reducing interventions. This mechanistic model motivates this proposal in which we hypothesize that: 1) MRI based measures of hippocampal metabolism and glutamate can be used as in vivo biomarkers of target engagement when using a glutamate-reducing intervention. 2) Glutamate-reducing interventions will be most effective when administered to prodromal patients, and only when they are shown to engage their target. The R61 phase of this proposal is designed to test the first hypothesis, and the R33 phase is designed to the second.

Public Health Relevance

While elevations in glutamate have been implicated in the pathophysiology schizophrenia, to date glutamate- reducing interventions have shown no reliable clinical efficacy. We hypothesize that these failures reflect the fact that biomarkers were not used to determine ?target engagement?, and that the interventions were administered in relatively late stages of the illness. This proposal is designed to test this hypothesis, where in the R61 phase we will validate neuroimaging biomarkers aimed at measuring target engagement of a glutamate-reducing intervention, and in the R33 phase we will use the biomarkers to test the clinical efficacy of a glutamate-reducing intervention in early stage patients.

National Institute of Health (NIH)
National Institute of Mental Health (NIMH)
Project #
Application #
Study Section
Special Emphasis Panel (ZMH1)
Program Officer
Hillefors, MI
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Columbia University (N.Y.)
Schools of Medicine
New York
United States
Zip Code