ABSTRACT Schizophrenia (SZ) is a highly debilitating neuropsychiatric disorder of young adulthood onset and a leading cause of disability worldwide. While treatments delivered at early stages of the disorder may be effective at reducing psychosis or altering the course of the disease, there are currently no biomarkers capable of identifying subjects at ultra-high risk (UHR) for psychosis who are likely to convert to the full clinical syndrome; or those in early stages of SZ who are likely to respond to treatment and would be good candidates for available proactive, symptomatic or future disease-modifying treatments; or those who would not respond and can be spared unnecessary medication exposure. The lack of these vitally important biomarkers provides a compelling rationale for the present multidisciplinary, transnational research project, which aims to develop and validate highly promising noninvasive and objective proton magnetic resonance spectroscopy (1H MRS)-based biomarkers for assessing conversion risk in UHR subjects and monitoring treatment response in first-episode psychosis (FEP) patients. In support of the viability of this overall objective is a large body of data, reported by the applicants and others, that show (a) that levels of glutamate (Glu) and -aminobutyric acid (GABA) – respectively, the major excitatory and inhibitory amino acid neurotransmitter systems - are abnormally elevated in medication-naïve and unmedicated UHR, FEP and chronic SZ cohorts; (b) that the effect of treatment with antipsychotic medications may be to lower or normalize brain levels of both Glu and GABA; and (c) that Glu elevations at baseline in UHR subjects may be a reliable predictor of conversion to full psychosis. To investigate the potential of these in vivo brain Glu and GABA abnormalities to serve as biomarkers of conversion risk in UHR subjects and of treatment response in early-stage SZ, the applicants propose to use 1H MRS to measure Glu and GABA levels in the largest cohorts of medication-naïve UHR and FEP subjects to date, at baseline and at conversion or at 2 years in UHR subjects; and at baseline and following 4 weeks of antipsychotic treatment in FEP subjects. The hypotheses to be tested are that (1) baseline elevations of Glu and GABA in UHR subjects will predict the risk of conversion to full psychosis; that (2) both GABA and Glu will be elevated at baseline in FEP subjects and decrease or normalize with second-generation antipsychotic treatment; and that (3) levels of both GABA and Glu will correlate positively with clinical measures in both groups at baseline and or negatively or not with treatment response in FEP subjects. If successful, the proposed studies have the potential to establish 1H MRS measures of brain GABA and Glu as predictors of conversion risk and biomarkers of treatment response or non-response, support the exploration of novel glutamate- or GABA-based treatments that may be more effective and present lower risks, and...