Functional Network Dysconnectivity in Schizophrenia Independent of Medication and Illness Duration

537 

Abstract Submission 

Kyle Jensen^1,2, Adithya Ram Ballem^1,2, Pablo Andrés-Camazón^3,1, Shalaila Haas^4,1, Jiayu Chen^1,2, Zening Fu^1,2, Vince Calhoun^1,2, Armin Iraji^1,2

¹Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS), Atlanta, GA, ²Georgia State University, Atlanta, GA, ³Hospital General Universitario Gregorio Marañón, Madrid, Spain, ⁴Icahn School of Medicine at Mount Sinai, New York, NY

Kyle Jensen  
Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS)|Georgia State University
Atlanta, GA|Atlanta, GA

Adithya Ram Ballem  
Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS)|Georgia State University
Atlanta, GA|Atlanta, GA

Pablo Andrés-Camazón  
Hospital General Universitario Gregorio Marañón|Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS)
Madrid, Spain|Atlanta, GA

Shalaila Haas, PhD  
Icahn School of Medicine at Mount Sinai|Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS)
New York, NY|Atlanta, GA

Jiayu Chen  
Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS)|Georgia State University
Atlanta, GA|Atlanta, GA

Zening Fu  
Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS)|Georgia State University
Atlanta, GA|Atlanta, GA

Vince Calhoun  
Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS)|Georgia State University
Atlanta, GA|Atlanta, GA

Armin Iraji  
Tri-institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS)|Georgia State University
Atlanta, GA|Atlanta, GA

Recent efforts in neuroscience seek to re-define schizophrenia (SZ) as a disorder of the brain by identifying abnormalities in functional circuitry. Symptoms of SZ (e.g., psychosis) are believed to be caused by disruptions in cortical-subcortical-cerebellar circuitry [1]. Resting-state functional magnetic resonance imaging (rs-fMRI) has been used to investigate and identify abnormal functional network connectivity (FNC) linked to SZ [2]. Specifically, previous studies have highlighted patterns of SZ-related aberrant cerebello-thalamo-cortical connectivity (CTCC) in networks involving the motor cortices, thalamic and subthalamic nuclei, the cerebellum, and the prefrontal cortex [2], [3]. However, these results are likely influenced by various confounds including changes in the brain over time due to cortical atrophy [4] as well as the effects of medication usage [5]. The current study seeks to disentangle the effects of medication and duration of illness through a data-driven approach to examine whole-brain FNC in individuals with SZ.

rs-fMRI as well as demographic and clinical features from a large multisite dataset (18 individual sites aggregated from multiple prior studies) was used in the current analyses. In order to investigate the impact of medication and duration of illness, participants with SZ were excluded from the analysis if they lacked this information. Our clinical dataset included 1288 participants (729 Male; mean age 34.96 ± 11.73 years; 45.89% Caucasian, 22.05% African American, 26.63% Asian, 5.43% Other), with 362 individuals with SZ (242 Male; mean age 39.10 ± 12.11 years; mean duration of illness 18.32 ± 12.26 years; mean chlorpromazine equivalence dosage (CPZ) 398.12 ± 332.72 mg), and 926 individuals considered typical controls (487 Males; mean age 33.34 ± 11.18 years). We applied multivariate-objective optimization ICA with reference (MOO-ICAR) and our recently developed multi-spatial-scale intrinsic connectivity networks (ICNs) optimized from 100K+ human participants [6] to identify 105 subject-specific ICNs from preprocessed rs-fMRI data. Postprocessing was performed on the ICN time courses, including steps to remove additional noise effects through detrending, despiking, regression of head motion, and band-pass filtering [0.01-0.15 Hz]. A subject-specific functional network connectivity (FNC) matrix was calculated by a Pearson correlation between all pairs of the 105 ICNs. We conducted a statistical case/control comparison to identify differences in FNC between individuals with SZ and controls. For each FNC, a general linear model was applied, correcting for age, sex, race, imaging site, head motion, CPZ, and duration of illness. We adjusted for multiple comparisons using the false discovery rate (FDR) correction.

Widespread disruptions in functional circuitry were observed across the whole brain in individuals with SZ, indicated by statistically significant case/control differences in visual (VIS), cerebellar (CB), temporal (TM), subcortical (SC), sensorimotor (SM), and higher cognitive (HC) domains (see Fig 1).

The patterns of dysconnectivity in functional circuitry observed between CB, SC, and cortical ICNs lends further insight into a CTCC model of SZ (See Fig 2). Furthermore, while medication and duration of illness remain pervasive confounds in SZ research, our analytical approach enabled us to observe many patterns of dysconnectivity associated with SZ which were unrelated to these effects. An increased understanding of the neural underpinnings of SZ will undoubtedly improve diagnosis and treatment outcomes.

Psychiatric (eg. Depression, Anxiety, Schizophrenia) ¹

Connectivity (eg. functional, effective, structural)

fMRI Connectivity and Network Modeling ²

Task-Independent and Resting-State Analysis

BOLD fMRI

Cerebellum

Cortex

FUNCTIONAL MRI

Psychiatric

Psychiatric Disorders

Schizophrenia

Sub-Cortical

Thalamus

Other - Multi-spatial-scale networks; Medication