FUNCTIONAL AND STRUCTURAL CONNECTIVITY IN SUBJECTS AT HIGH RISK FOR PSYCHOSIS AS A POSSIBLE BIOMARKER FOR THEIR TRANSITION TO SCHIZOPHRENIA – A COMBINED EEG AND DTI STUDY
Background Altered gamma-band connectivity has been shown across all stages of schizophrenia (SZ) and for persons at High Risk for Psychosis (HRP). Oscillations in the gamma-band frequency range are suggested to play a crucial role for both local and long-range synchronization within the brain during perceptual and cognitive processes. Structural connectivity in the brain is provided by white matter (WM) tracts connecting these areas. Reduced WM integrity is also a feature of SZ that has often been described. The aim of this study is to investigate the relationship between altered gamma-band connectivity and WM structural connectivity in Clinical High Risk for Psychosis (HRP) patients using Electroencephalography (EEG) and Diffusion Tensor Imaging (DTI) to achieve a better understanding of which factors could predict a transition to SZ considering functional and structural connectivity. The identification of those biomarkers would be a further step toward understanding the causes of SZ and could contribute to a more individualized and targeted therapy to achieve a better outcome in patients.
Methods The study includes a total of 26 HRP patients, 7 converted to schizophrenia after one year (HRP-C) and 13 healthy controls (HC). High density EEG (64 channels) was recorded during the performance of a cognitively demanding auditory choice reaction task. Source estimation and analysis of the functional gamma-band connectivity was done using eLORETA. We defined Regions of Interest (ROIs) that were known to be generators of the auditory evoked Gamma-Band Response (aeGBR) from a previous combined EEG-fMRI study of this population – the dorsal anterior cingulate cortex (dACC) and the bilateral auditory cortices. MRI data acquisition was done with a 3T Siemens scanner (Trio). We generated whole-brain 2-Tensor-Tractography for all subjects in order to analyze the integrity of the underlying WM fiber bundles implicated in this functional network, namely the Cingulum Bundle (CB), the Medial Longitudinal Fasciculus (MdLF), the Arcuate Fasciculus (AF) and the Extreme Capsule (EC).The primary outcome measure is the fractional anisotropy of the tissue (FAt), which is corrected for free water (FW) in the brain.
Results Our preliminary results show a lower aeGBR in HRP-C patients compared to HC with the HRP-NC at an intermediate level. We found longer reaction times for HRP compared to HC. The study is presently on-going. Based on previous work in this sample, we hypothesize that the CB, MdLF, AF and EC of the HRP-C will exhibit decreased FAt and increased FW compared to HC.
Discussion The combination of EEG with its high temporal resolution and DTI with its high spatial resolution allows for detailed insight into the possible disturbances in HRP patients compared to HC. Alterations in both modalities have been shown before, however, in this study we aim to link functional connectivity alterations in the previously described networks to their structural correlate to achieve a better insight into their interaction. Taken together, the quantification of those possible biomarkers might help to detect a transition to SZ earlier and thus allow an early intervention.