Glutamate levels in antipsychotic-free, first-episode psychosis: a 7-tesla MRS study across five brain regions

Disrupted glutamatergic signaling across cortical and subcortical circuits is increasingly recognized as a critical player in the pathophysiology of psychosis. Using ultra-high-field (7 Tesla) proton MRS, we quantified glutamate levels across five brain regions in antipsychotic-free first-episode psychosis (FEP). 47 patients (22.4 ± 3.4 years; 25 female) and 21 healthy controls (23.5 ± 2.6 years; 12 female) underwent 7 T MRS with voxels placed in the anterior cingulate cortex, dorsolateral prefrontal cortex, basal ganglia, hippocampus, and thalamus. In the primary analyses we tested region-wise group differences in glutamate concentrations using ANCOVA. Exploratory analyses examined inter-regional glutamate correlations and associations with symptom severity. Regional glutamate concentrations did not differ between groups (all p > 0.2), and a nominally significant association between basal ganglia glutamate and negative symptoms did not survive correction for multiple comparisons. Exploratory analyses revealed that FEP patients, in contrast to controls, showed significant inter-regional glutamate correlations across six of ten region pairs (r = 0.3–0.4; p = 0.01–0.04), spanning cortical-cortical, cortical-subcortical, and subcortical-subcortical connections. Four of these correlations were associated with positive symptom severity (p = 0.03–0.05). While hypothesized group differences in regional glutamate concentrations were not confirmed, antipsychotic-free FEP patients exhibited widespread inter-regional glutamate correlations, a pattern of network-level co-regulation absent in HC. This pattern of network-level dysconnectivity may reflect a glutamatergic substrate for the positive symptoms of psychosis, with potential implications for new treatment strategies. These exploratory findings suggest that glutamatergic dysregulation in early psychosis may be better captured by inter-regional covariance than by concentration differences alone.