Corticolimbic network perturbations in early psychosis: an arterial spin labelling and [11C]Ro15-4513 positron emission tomography study

Background Hippocampal hyperactivity in early psychosis may result from excitation-inhibition imbalance within corticolimbic regions. Preclinical evidence suggests that positive allosteric modulation of hippocampal inhibitory γ-aminobutyric acid receptors (α5-GABAAR) attenuates hippocampal hyperactivity, striatal hyperdopaminergia, and psychosis-relevant behaviours. Here, we investigated whether hippocampal hyperactivity and α5-GABAAR network covariance are perturbed in people at clinical high risk for psychosis (CHR-P) and with first-episode psychosis (FEP). Method Twenty-four individuals at CHR-P, 24 healthy controls (HC), and 10 FEP, underwent simultaneous PET-MRI to quantify regional cerebral blood flow (rCBF) with arterial spin labelling and α5-GABAAR availability with [11C]Ro15-4513 PET. Individual deviations in corticolimbic rCBF and α5-GABAAR covariance were calculated using a perturbation analysis approach based on a HC-derived reference. Permutation tests and Pearson's correlations assessed group differences and symptom relationships. Results For rCBF, absolute deviations were greater at hippocampal edges (F=5.763, p=.006) in CHR-P (p=.016) and FEP (p=.002). For α5-GABAAR, absolute deviations did not differ from HC across the network (p=.145) or at hippocampal edges (p=.245). However, negative covariance deviations (reduced covariance vs HC) were greater for rCBF (F=4.006, p=.032) and α5-GABAAR across the network (F=6.265, p=.026), and for hippocampal edges for rCBF (F=4.471, p=.015; CHR-P) and α5-GABAAR (F=3.14, p=.047; CHR). In CHR-P, rCBF deviations correlated with positive symptoms (r=-.516, p=.001); in FEP, α5-GABAAR deviations correlated with general symptoms (r=.722, p=.018). No rCBF-α5-GABAAR covariance correlations were observed. Conclusions Early psychosis involves alterations in corticolimbic network organization in both neural activity and α5-GABAAR availability, driven by negative deviations. Individualized covariance perturbation provides a promising approach for detecting network-level dysfunction psychosis.