Widespread white matter microstructural differences in schizophrenia across 4322 individuals: results from the ENIGMA Schizophrenia DTI Working Group

The regional distribution of white matter (WM) abnormalities in schizophrenia remains poorly understood, and reported disease effects on the brain vary widely between studies. In an effort to identify commonalities across studies, we perform what we believe is the first ever large-scale coordinated study of WM microstructural differences in schizophrenia. Our analysis consisted of 2359 healthy controls and 1963 schizophrenia patients from 29 independent international studies; we harmonized the processing and statistical analyses of diffusion tensor imaging (DTI) data across sites and meta-analyzed effects across studies. Significant reductions in fractional anisotropy (FA) in schizophrenia patients were widespread, and detected in 20 of 25 regions of interest within a WM skeleton representing all major WM fasciculi. Effect sizes varied by region, peaking at (d=0.42) for the entire WM skeleton, driven more by peripheral areas as opposed to the core WM where regions of interest were defined. The anterior corona radiata (d=0.40) and corpus callosum (d=0.39), specifically its body (d=0.39) and genu (d=0.37), showed greatest effects. Significant decreases, to lesser degrees, were observed in almost all regions analyzed. Larger effect sizes were observed for FA than diffusivity measures; significantly higher mean and radial diffusivity was observed for schizophrenia patients compared with controls. No significant effects of age at onset of schizophrenia or medication dosage were detected. As the largest coordinated analysis of WM differences in a psychiatric disorder to date, the present study provides a robust profile of widespread WM abnormalities in schizophrenia patients worldwide. Interactive three-dimensional visualization of the results is available at www.enigma-viewer.org.

components), and 19 lateralized regions (left and right are averaged to obtain bilateral FA) (Supplementary material). The overall average FA values were calculated by averaging values for the entire white matter skeleton.
ENIGMA-DTI QA/QC protocol consists of visual inspection of the images before and after registration to the ENIGMA template, as well as calculating the average skeleton projection distance. The distance of voxel projection to the ENIGMA skeleton can assess the registration quality between individual images and ENIGMA-DTI template. Higher projection distance may indicate problems with aligning individual brain to the template. After ROI extraction, histograms of FA and diffusivity measures are computed for each ROI.

Abbreviation
Full tract name Supplementary Note 2: Covarying for core and periphery FA To tease apart regional WM effects from the global differences, post-hoc analysis on a sub-sample of available data from 17 cohorts (1,361 healthy controls and 1,226 schizophrenia patients) was carried out in individual tract ROIs as above but additionally covarying for 1) average FA across the entire skeleton, 2) average FA within the core of the skeleton within which ROIs are defined, or 3) the average FA within the periphery of the skeleton, areas on the skeleton not included in the ROIs of the JHU atlas. Core FA was calculated by computing the weighted average of all non-overlapping ROIs. Periphery FA was calculated by subtracting weighted core FA (N core voxels/N periphery voxels x core FA) from the weighted average FA voxels (N average FA voxels/N periphery FA voxels x averageFA).

Supplementary Note 3: Meta-analysis description
Using this a random-effects inverse-variance weighted meta-analysis, results from statistical tests performed at each individual site were combined and the overall statistical effect size was calculated as the weighted average of the effects from the individual cohorts as determined by the standard error of the effect.

Supplementary Note 4: Power analysis
A post-hoc power analysis, conducted using G*Power v3.1, revealed that the current sample of 1,984 individuals with schizophrenia and 2,391 healthy controls achieved 80% power to detect Cohen's d effect sizes as small as d=0.085 at the standard alpha level of p<0.05 (two-tailed), and 80% power to detect Cohen's d effect sizes of d=0.12 at the study's Bonferroni-corrected threshold of p<0.002. Across all ROIs, we estimated N80: the total number of samples required, per group, to achieve 80% power to detect group differences using a t-test at the threshold of p<0.05 (two-tailed). N80 ranged from 104-13,581 (See supplementary table 5).

Supplementary Note 5: Diffusivity analysis
Available diffusivity images, including mean, radial and axial diffusivity, were skeletonized and relevant ROI information was extracted according to the ENIGMA-DTI template. A random effects, inverse variance weighted meta-analysis was conducted to combine results. A random effects, inverse variance weighted meta-analysis was conducted to combine results.

Supplementary Note 6: Sex-Specific Effect Sizes: Male / Female Cases vs Controls
We observed significant differences in effect size when analyzing males and females separately, with females showing significantly larger effects for decreased FA (paired t=-5.21, p=0.0001).
For females, 20 of the 25 ROIs showed significantly lower FA for patients at the p=0.002 significance threshold (See supplementary table 6). For males, 14 of the 25 ROIs showed significantly decreased FA for patients at the p=0.002 significance threshold (See  supplementary table 7).

Supplementary Note 7: Acknowledgements and conflicts of interest
The ENIGMA-Schizophrenia working group gratefully acknowledges support from the NIH BD2K award, U54EB020403.