LAUSR

Abstract Background The N-methyl-D-aspartate receptor (NMDAR) hypofunction model of schizophrenia suggests that dysfunction of these receptors could underlie the brain functional abnormalities characterising these patients. In NMDAR encephalitis (NMDARE), which holds clinical similarities with schizophrenia, autoantibodies target NMDARs, predominantly located in the hippocampal region, leading to disrupted glutamatergic transmission. One study so far has described abnormal connectivity between the medial temporal lobe and posterior default mode network regions in patients with NMDARE (Peer et al., 2017), however no study so far has examined brain functional correlates of schizophrenia and NMDARE comparatively. Methods Patients with schizophrenia (N=16) and with NMDARE shortly after clinical stabilisation (N=15) were recruited within a tertiary setting, and compared with age and sex-matched healthy volunteers (N=20). All individuals were scanned with a 3T Siemens scanner including a functional resting state sequence, during which individuals were instructed to view a fixation cross. A seed-based analysis was performed using a spherical seed of 4mm located in the left hippocampus (MNI coordinates x =-32, y=-24, x=-14). Only grey matter voxels were considered to obtain the seed average signal. BOLD signal was preprocessed including slice timing, motion correction, spatial smoothing, and frequency filtering, and regressed taken into account movement parameters (rigid transformation, frame displacement, and DVARS). Statistics was performed between the correlation maps including age and sex as covariates. Family-wise error (FWE) was used to correct for multiple comparisons. Results Seed to voxel analyses revealed connectivity between the seed and the bilateral thalamus, parahippocampus, precuneus, posterior cingulate, right hippocampus and lateral temporal regions (threshold Fisher’s z > 0.28). We first examined differences in connectivity between both patient groups combined and healthy volunteers, which revealed greater connectivity in patients than in controls between the left hippocampus and the left inferior parietal, postcentral and posterior cingulate gyri (pFWE< 0.05). When examining patient groups individually, patients with schizophrenia continued to exhibit significantly greater connectivity between the left hippocampus and left inferior parietal and postcentral region (pFWE< 0.05) and at near trend level in the posterior cingulate (pFWE= 0.15). NMDARE patients also presented near trend level increased connectivity in the posterior cingulate and postcentral gyri (pFWE= 0.10). There were no differences in functional connectivity of the left hippocampus between patients with schizophrenia and with NMDARE. Discussion To our knowledge, this is the first study to compare patients with schizophrenia with patients with NMDARE using measures of brain function. We found that, for both conditions, the left hippocampus showed greater connectivity with areas belonging to the posterior default mode network. Connectivity between these regions has been associated with psychotic symptoms in schizophrenia (Lefebvre et al., 2016). Our findings suggest that this alteration could also underlie neuropsychiatric symptoms in NMDARE, and provides further evidence that NMDAR dysfunction may underpin the pathophysiology of schizophrenia.