S151. HIPPOCAMPAL CONNECTIVITY AND VOLUME AS PREDICTORS OF TREATMENT RESPONSE – A REPLICATION STUDY IN TWO PSYCHOSIS COHORTS

BACKGROUND: Approximately 30% of patients with schizophrenia do not improve with antipsychotic drug (APD) treatment and 60% show sub-optimal response. Converging lines of evidence point to hippocampal dysfunction in schizophrenia. It is thought that hippocampal dysfunction spreads across hippocampal subfields and to cortical regions by way of long-range efferent projections. Our prior studies have shown altered hippocampal regional cerebral blood flow in unmedicated patients and normalization after APD treatment. Meta-analyses show reduced hippocampal volume in first episode psychosis (FEP) patients. We evaluated resting state hippocampal functional connectivity (hFC) as well as hippocampal and hippocampal subfield volumes as predictors of treatment response (TR) in two cohorts of patients with a psychosis spectrum disorder. All patients were subsequently treated with an APD for 6 weeks. METHODS: Cohort 1 consisted of 55 medication-naïve first episode psychosis (FEP) subjects (36 male; mean age 24.18 years). Cohort 2 consisted of 42 unmedicated patients with schizophrenia (SZ) (31 male; mean age 27.9 years). FEP were scanned on a Siemens MAGNETOM Prisma MRI scanner using a 20 channel head coil. Anatomical scans were acquired via T1-weighted and T2-weighted images. Two 6-minute resting state scans were acquired in opposing phase encoding directions (A > P and P > A). SZ were scanned on a Siemens MAGNETOM Allegra MRI scanner with a circularly polarized transmit/receive head coil. Anatomical scans were acquired via a T1-weighted sequence. Resting state scans were acquired with a single 5-minute gradient recalled echo-planar imaging sequence. For both datasets, resting state data were preprocessed in the CONN toolbox (version 18a). We used the left hippocampus as a seed region to create whole brain seed-to-voxel correlation maps for each subject. Regression analyses were then performed to assess the relationship between resting state connectivity and TR (% change in BPRS positive score from (A) baseline to (B) after 6 weeks of APD: (((B-A)/A)*-100). Analyses were corrected using voxel (p < 0.05, uncorrected) and cluster level correction (p < 0.05, FDR corrected). Age, sex, and framewise displacement were used as covariates of no interest. T1 and T2 weighted images were preprocessed using FreeSurfer 6.0. Freesurfer’s hippocampus subfield segmentation module was used to calculate left and right subfield volumes. SPSS 25 was used to regress hippocampal subfield volumes on TR. Age and estimated total intracranial volume (eTIV) were included as covariates of no interest. RESULTS: In both cohorts greater hFC to the cuneus and precuneus was predictive of better TR, as was greater hFC to the fusiform gyrus, medial prefrontal cortex (PFC) and anterior cingulate cortex in cohort 2. Reduced hFC connectivity to the angular gyrus in supramarginal gyrus and temporal pole in cohort 1 as well as the orbitofrontal cortex and dorsolateral PFC in cohort 2 were also predictive of better TR. Results from the stepwise regression showed that neither right nor left whole hippocampal volume, or subfield volumes, significantly predict TR for either cohort. DISCUSSION: In two patient cohorts, we observed a similar pattern where increased hFC to the cuneus and precuneus was predictive of better response to APD. Furthermore, the lack of a significant predictive value of hippocampal volumes in predicting TR was replicated in each cohort. The replicability of these findings, particularly in a cohort of medication-naïve FEP provides potential biological patterns useful in determining initial response to APD medication in patients with a psychosis spectrum disorder.