Multisite Tractometry Study Reveals Microstructural Abnormalities Along Tracts in Bipolar Disorder

570 

Abstract Submission 

Leila Nabulsi¹, Bramsh Chandio¹, Genevieve McPhilemy², Fiona Martyn², Gloria Roberts³, Brian Hallahan², Udo Dannlowski⁴, Tilo Kircher⁵, Benno Haarman⁶, Philip Mitchell³, Colm McDonald², Dara Cannon², Ole Andreassen⁷, Christopher Ching⁸, Paul Thompson⁹, ENIGMA Bipolar Disorder Working Group¹

¹University of Southern California, Los Angeles, CA, ²University of Galway, Galway, Galway, ³Discipline of Psychiatry and Mental Health, University of New South Wales, Sydney, New South Wales, Sydney, New South Wales, ⁴Institute for Translational Psychiatry, Münster, North Rhine Westphalia, ⁵Department of Psychiatry and Psychotherapy, University of Marburg, Marburg, Hesse, ⁶Department of Psychiatry, University Medical Center Groningen, University of Groningen, Groningen, Groningen, Groningen, ⁷NORMENT, Oslo, Norway, ⁸Imaging Genetics Center, Mark and Mary Stevens Neuroimaging & Informatics Institute, USC, Los Angeles, CA, ⁹Imaging Genetics Center, Keck School of Medicine of University of Southern California, Los Angeles, CA

Leila Nabulsi, PhD  
University of Southern California
Los Angeles, CA

Bramsh Chandio  
University of Southern California
Los Angeles, CA

Genevieve McPhilemy, PhD  
University of Galway
Galway, Galway

Fiona Martyn, PhD  
University of Galway
Galway, Galway

Gloria Roberts  
Discipline of Psychiatry and Mental Health, University of New South Wales, Sydney, New South Wales
Sydney, New South Wales

Brian Hallahan, MD  
University of Galway
Galway, Galway

Udo Dannlowski  
Institute for Translational Psychiatry
Münster, North Rhine Westphalia

Tilo Kircher  
Department of Psychiatry and Psychotherapy, University of Marburg
Marburg, Hesse

Benno Haarman  
Department of Psychiatry, University Medical Center Groningen, University of Groningen, Groningen
Groningen, Groningen

Philip Mitchell  
Discipline of Psychiatry and Mental Health, University of New South Wales, Sydney, New South Wales
Sydney, New South Wales

Colm McDonald, PhD  
University of Galway
Galway, Galway

Dara Cannon, PhD  
University of Galway
Galway, Galway

Ole Andreassen  
NORMENT
Oslo, Norway

Christopher Ching  
Imaging Genetics Center, Mark and Mary Stevens Neuroimaging & Informatics Institute, USC
Los Angeles, CA

Paul Thompson, PhD  
Imaging Genetics Center, Keck School of Medicine of University of Southern California
Los Angeles, CA

ENIGMA Bipolar Disorder Working Group  
University of Southern California
Los Angeles, CA

Methods to more finely map the brain circuitry alterations associated with bipolar disorder (BD) hold promise for identifying new potential biomarkers that could one day support more impactful treatments for BD and related mental illness. Prior diffusion MRI studies have reported subtle abnormalities in the brain's white matter (WM) microstructure by analyzing data within specific regions of interest (Favre 2019). However, there is a need for a more detailed 3D spatial assessment of microstructural differences along WM tracts in BD. A significant advancement would involve integrating tractometry data from multiple scanning sites to better detect subtle effects and improve replication of findings across larger international datasets. In this multisite study using diffusion MRI tractometry, we applied the BUndle ANalytics (BUAN) method (Chandio 2020), which offers a sophisticated analytic approach to tractography. BUAN enables the extraction and mapping of fiber bundles, facilitating the detection and visualization of microstructural irregularities across 3D representations of fiber tracts.

Diffusion-weighted 3D brain MRI scans of 148 participants diagnosed with BD (age: 36.8+13.5 y) and 259 psychiatrically-healthy controls (age: 37.7+13.7 y), were drawn from 6 independent scanning sites. Scans were corrected for subject motion and eddy-current distortions (Leemans 2009). Deterministic (non-tensor) constrained spherical deconvolution (CSD) was used to account for crossing fibers within voxels (Jeurissen 2014; ExploreDTI), and fractional anisotropy (FA) was calculated at each voxel. Individual whole-brain tractograms underwent streamline-based registration to a bundle atlas template in MNI space (Yeh 2018). Bundle extraction was performed using the auto-calibrated version of RecoBundles and a standard WM tract atlas (Yeh 2018; Chandio 2020). A tract profile for each extracted bundle was generated using the BUAN software package, whereby each profile consisted of 100 segments per subject (Chandio 2020). By treating variations across scanning sites and imaging protocols as random effects, we investigated microstructural abnormalities along WM tracts between BD and controls.

Significant differences between BD and controls were detected in mean FA across several WM tracts (F1). Lower FA values were detected in fronto-limbic, interhemispheric, and posterior pathways in the BD group relative to controls; specifically, in localized regions of the cingulum and the fornix, and in regions within the corpus callosum (middle portion and forceps minor) and the fronto-parietal tract. Several long-range intra-hemispheric WM tracts exhibited lower FA in BD compared to controls in localized segments – specifically, in the extreme capsule, the arcuate, uncinate, inferior-fronto occipital, medial- and middle-longitudinal fasciculi. Higher FA was observed in posterior bundles in BD relative to controls. The incremental inclusion of additional sites progressively enhanced the detection of group differences, as revealed by quantile-quantile plots.

Using an advanced along-tract analytic method, we conducted fine-scale spatial mapping of regional WM microstructure differences in BD, relative to controls, integrating data across sites in the largest such study to date. By integrating tractography and anatomical information, BUAN captured unique effects along WM tracts, offering more anatomical specificity than region-of-interest based analyses, and providing valuable insights into anatomical variations that may assist in disease classification. The tracts implicated here connect regions with important functional roles in the regulation of emotions, motivation, decision-making, and cognitive control, all impaired in BD. Effect sizes for each tract increased with the incremental inclusion of more samples. These findings advance our understanding of the neural underpinnings of BD and offer valuable leads for refining disease classification frameworks.

Psychiatric (eg. Depression, Anxiety, Schizophrenia) ¹

Diffusion MRI Modeling and Analysis

White Matter Anatomy, Fiber Pathways and Connectivity ²

Diffusion MRI

MRI

Psychiatric Disorders

Tractography

White Matter

Other - Bipolar Disorder; Diffusion-weighted MRI