White and gray matter alterations in bipolar I and bipolar II disorder subtypes compared with healthy controls – exploring associations with disease course and polygenic risk

Patients with bipolar disorder (BD) show alterations in both gray matter volume (GMV) and white matter (WM) integrity compared with healthy controls (HC). However, it remains unclear whether the phenotypically distinct BD subtypes (BD-I and BD-II) also exhibit brain structural differences. This study investigated GMV and WM differences between HC, BD-I, and BD-II, along with clinical and genetic associations. N = 73 BD-I, n = 63 BD-II patients and n = 136 matched HC were included. Using voxel-based morphometry and tract-based spatial statistics, main effects of group in GMV and fractional anisotropy (FA) were analyzed. Associations between clinical and genetic features and GMV or FA were calculated using regression models. For FA but not GMV, we found significant differences between groups. BD-I patients showed lower FA compared with BD-II patients (ptfce-FWE = 0.006), primarily in the anterior corpus callosum. Compared with HC, BD-I patients exhibited lower FA in widespread clusters (ptfce-FWE < 0.001), including almost all major projection, association, and commissural fiber tracts. BD-II patients also demonstrated lower FA compared with HC, although less pronounced (ptfce-FWE = 0.049). The results remained unchanged after controlling for clinical and genetic features, for which no independent associations with FA or GMV emerged. Our findings suggest that, at a neurobiological level, BD subtypes may reflect distinct degrees of disease expression, with increasing WM microstructure disruption from BD-II to BD-I. This differential magnitude of microstructural alterations was not clearly linked to clinical and genetic variables. These findings should be considered when discussing the classification of BD subtypes within the spectrum of affective disorders.

Multi-dimensional scaling (MDS) components were estimated on genotype data to adjust for population stratification.More specifically, MDS components were estimated based on the identity-by-state distance matrix of all individuals included in this study by using the eigendecomposition-based algorithm in PLINK (version 1.9).According to the scree plot, the first three MDS components were included as covariates in the following analyses.
PRS were calculated with PRS-CS 6 in PLINK based on a previous genome-wide association study (GWAS) for bipolar disorder. 7Following a study by Kalman and colleagues, 8 we estimated PRS weights with PRS-CS-auto (φ=1.29×10-4).Linear regression analyses were performed between PRS and gray and white matter using SPM and FSL, respectively.As covariates, age, sex, scanner settings (Marburg Body-Coil pre, Marburg Body-Coil post, Münster) and the first three MDS components were included.

DTI data acquisition
Fifty-six axial slices with no gap were measured with an isotropic voxel size of 2.5x2.5x2.5mm³(TE=90ms, TS=7300ms), using a GRAPPA acceleration factor of 2. Five non-diffusionweighted (DW) images (b=0s/mm2) and 2x30 DW images with a b-value of 1000s/mm² were acquired.For quality assurance of the data, the open-source software DTIPrep (Oguz et al., 2014) was used with default options.In case of artifacts, individual images of a given participant were eliminated, and a participant was excluded from further analyses if more than 20% of images were affected.On average, the included participants had 64.11 images (SD=1.40,range: 56-65).

DTI data preprocessing
0][11] The DW images were corrected for head motion and eddy current induced distortions using "eddy" from FSL, 12 and b-vectors were rotated accordingly.After removal of non-brain tissue using the Brain Extraction Tool (BET) in FSL, 13 the first b0 image from each participant was used as reference for alignment.

DTI data analysis
Analysis of DTI data was performed using TBSS, 14 a technique designed to reduce registration misalignment.Using FMRIB's non-linear registration tool, all FA images were aligned to the FMRIB58 FA template brain in 1×1×1 mm³ Montreal Neurological Institute (MNI) standard space.All registered FA images were averaged and a threshold of 0.2 was applied to create a WM skeleton representing the centers of the tracts common to all participants.Each participant's aligned FA data were then projected onto the mean skeleton mask by searching for maximum FA values perpendicular to the local skeleton direction.

Calculation of DTI metrics
FA is a measure of the directionality of water diffusion on a scale from 0 (indicating isotropic diffusion) to 1 (indicating completely anisotropic diffusion).It is calculated as the normalized variance of the three eigenvalues.MD represents average water diffusion and is calculated as the mean of the three eigenvalues.AD and RD are specific measures of diffusivity parallel and perpendicular to the principal direction of the axonal fibres, respectively.AD is equivalent to the first eigenvalue, whereas RD is calculated as the average of the second and third eigenvalues. 15,16
The most affected tract was consistently the forceps minor (Table S2).

b) Associations with clinical variables and polygenic risk within all BD patients
For RD, MD, and AD, BD patients showed a negative association with the time since first psychiatric hospitalization (RD: ptfce-FWE =.006, k=45300 voxels; MD: ptfce-FWE =.003, k=56027; AD: ptfce-FWE =.009, k=10951).Furthermore, AD showed a negative association with body mass index (ptfce-FWE =.032, k=440) (Table S1 and S2).However, only the association between MD and time since first psychiatric hospitalization survived Bonferroni correction.Beyond that, no significant associations emerged between clinical and genetic features and AD, MD, or RD.

Table S1 :
Cluster sizes and MNI coordinates of the peak voxel of all significant clusters, derived with the "cluster" tool implemented in FSL.(see separate excel file)

Table S2 :
Anatomical regions comprising the significant effects of the analyses based on the "JHU White-Matter Tractography Atlas".(see separate excel file)

Table S4 :
Difference between BD-I and BD-II in fractional anisotropy and radial diffusivity, corrected for different types of medication.(see separate excel file)

Table S5 :
Associations between clinical variables and diffusion tensor imaging metrics in BD-I and BD-II separately (see separate excel file)

Table S6 :
Associations between clinical variables and GMV in BD-I and BD-II separately (see separate excel file)

Table S3 .
Results of the exploratory whole-brain analyses of gray matter volumes, pairwise comparisons between HC, BD-I and BD-II, conducted at p<.001, uncorrected, with a threshold of k=50.