GABAA receptor availability in clinical high-risk and first-episode psychosis: a [11C]Ro15-4513 positron emission tomography study

Disrupted gamma-aminobutyric acid (GABA) neurotransmission may contribute to the pathophysiology of schizophrenia. Reductions in hippocampal GABAergic neurons have been found in schizophrenia, and increased hippocampal perfusion has been described in schizophrenia and in people at clinical high-risk for psychosis (CHRp). We have also found decreases in hippocampal GABAA receptors containing the α5 subunit (GABAARα5) in a well-validated neurodevelopmental rat model of relevance for schizophrenia. Positive allosteric modulation of these receptors in the hippocampus using a specific compound was shown to reverse the behavioural and neurophysiological phenotypes of this model. However, whether GABAARα5 availability is dysregulated in the psychosis spectrum at the regional or network levels is unknown. We addressed this issue by using [11C]Ro15-4513 and positron emission tomography (PET) in 22 individuals at CHRp, 10 people with a first-episode psychosis (FEP) and 23 healthy controls (HC). We quantified GABAARα5 availability in the hippocampus and across the brain, and employed a perturbation covariance method to assess individual molecular covariance deviations in CHRp and FEP groups compared to the HC group. Hippocampal GABAARα5 availability was not significantly different between groups (F(2,50) = 0.25, p = 0.78). However, network analysis identified significant deviations in GABAARα5 covariance between groups, both across all regions (all p < 0.001, pairwise Cohen’s d = 0.07–0.5) and relative to the hippocampus (all p < 0.001, pairwise Cohen’s d = 0.01–0.67). These findings suggest that individuals at clinical high-risk for psychosis and people with early psychosis may show alterations to the brain-wide organisation of the GABAARα5 system, rather than changes at a regional level.