White Matter Microstructure and its Relation to Clinical Features of Obsessive-Compulsive Disorder: Findings from the ENIGMA OCD Working Group

Importance Microstructural alterations in cortico-subcortical connections are thought to be present in Obsessive-Compulsive Disorder (OCD). However, prior studies have yielded inconsistent findings, perhaps because small sample sizes provided insufficient power to detect subtle abnormalities. Objective To investigate microstructural white matter alterations and their relation to clinical features in the largest dataset of adult and pediatric OCD to date. Design, Setting, and Participants In this cross-sectional case-control magnetic resonance study, we investigated diffusion tensor imaging metrics from 700 adult patients and 645 adult controls, as well as 174 pediatric patients and 144 pediatric controls across 19 sites participating in the ENIGMA-OCD Working Group. Main Outcomes and Measures We extracted measures of fractional anisotropy (FA) as main outcome, and mean diffusivity, radial diffusivity, and axial diffusivity as secondary outcomes for 25 white matter regions. We meta-analyzed patient-control group differences (Cohen’s d) across sites, after adjusting for age and sex, and investigated associations with clinical characteristics. Results Adult OCD patients showed significant FA reduction in the sagittal stratum (d=-0.21, z=-3.21, p=0.001) and posterior thalamic radiation (d=-0.26, z=-4.57, p<0.0001). In the sagittal stratum only, lower FA was associated with a younger age of onset (z=2.71, p=0.006), longer duration of illness (z=-2.086, p=0.036) and a higher percentage of medicated patients in the cohorts studied (z=-1.98, p=0.047). No significant association with symptom severity was found. Pediatric OCD patients did not show any detectable microstructural abnormalities compared to matched controls. Conclusions and Relevance Microstructural alterations in projection and association fibers to posterior brain regions were found in adult OCD, and related to disease course and medication status. Such results are relevant to models positing deficits in connectivity as a crucial mechanism in OCD. KEY POINTS Question Do patients with Obsessive-Compulsive Disorder (OCD) show white matter microstructural alterations, and are these alterations related to clinical features? Findings Data from 19 sites of the ENIGMA-OCD Consortium were included, involving 700 adult patients and 645 adult controls, 174 pediatric patients and 144 pediatric controls. Diffusion tensor imaging data were meta-analyzed using a harmonized data processing and analysis protocol. Adult, but not pediatric, patients showed alterations in the sagittal stratum and posterior thalamic radiation; sagittal stratum differences were associated with clinical features. Meaning Microstructural abnormalities found in adult but not in the pediatric cohort, are related to illness duration and medication status.

magnetic resonance study, we investigated diffusion tensor imaging metrics from 700 adult patients and 645 adult controls, as well as 174 pediatric patients and 144 pediatric controls across 19 sites participating in the ENIGMA-OCD Working Group.

Main Outcomes and Measures:
We extracted measures of fractional anisotropy (FA) as main outcome, and mean diffusivity, radial diffusivity, and axial diffusivity as secondary outcomes for 25 white matter regions. We meta-analyzed patient-control group differences (Cohen's d) across sites, after adjusting for age and sex, and investigated associations with clinical characteristics.
Results: Adult OCD patients showed significant FA reduction in the sagittal stratum (d=-0.21, z=-3.21, p=0.001) and posterior thalamic radiation (d=-0.26, z=-4.57, p<0.0001). In the sagittal stratum only, lower FA was associated with a younger age of onset (z=2.71, p=0.006), longer duration of illness (z=-2.086, p=0.036) and a higher percentage of medicated patients in the cohorts studied (z=-1.98, p=0.047). No significant association with 8 symptom severity was found. Pediatric OCD patients did not show any detectable microstructural abnormalities compared to matched controls.

Conclusions and Relevance: Microstructural alterations in projection and
association fibers to posterior brain regions were found in adult OCD, and related to disease course and medication status. Such results are relevant to models positing deficits in connectivity as a crucial mechanism in OCD.

INTRODUCTION
Abnormalities in cerebral white matter (WM) are relevant to models of anomalous brain circuitry that posit deficits in connectivity in obsessivecompulsive disorder (OCD) 1,2 . OCD has a childhood onset in over 50% of all cases, and most childhood-onset OCD cases persist into adulthood 3 .
Diffusion tensor imaging (DTI) allows the study of WM at the microstructural level through the analysis of intrinsic, three-dimensional diffusion properties of water within brain tissues 4 . Prior DTI studies in OCD [5][6][7] suggest that microstructural alterations are present in a number of WM areas. However, results across studies are inconsistent, with contrasting or conflicting effects of OCD on DTI metrics 8 . Sources of heterogeneity may include methodological factors (e.g., imaging acquisition and data processing), clinical characteristics, and variations in demographic or socioeconomic factors. More importantly, sample size variations may impact reported findings, as small studies may have insufficient power to detect subtle alterations 9 . Brain imaging consortia offer new opportunities, pooling data and findings from around the world to achieve an appropriate sample size.
The OCD working group of the Enhancing Neuro-Imaging Genetics through Meta-Analysis (ENIGMA) consortium 10 , is one such collaboration. Previous findings from the working group focused on subcortical and cortical brain grey matter abnormalities, using subcortical volumes, cortical thickness and surface area quantification algorithms. An initial analysis of data from 3,589 individuals showed distinct subcortical volume abnormalities in adults (smaller hippocampal and larger pallidal volumes) and unmedicated children (larger thalamic volume) with OCD 11 . The second study focused on cortical  13 and the Child Yale-Brown Obsessive-Compulsive Scale (CY-BOCS) 14 to assess symptom severity. These scales are clinician-rated, 10-item scales, with each item rated from 0 (no symptoms) to 4 (extreme symptoms; total range, 0 to 40), with separate subtotals for severity of obsessions and compulsions. Tables 1 and 2 show the demographic and clinical characteristics of the participants from each site.
All local IRBs approved the use of measures extracted from completely anonymized data.

Image Acquisition and Processing
Harmonized preprocessing, including brain extraction, eddy current correction, movement correction, echo-planar imaging-induced distortion correction and tensor fitting, was carried out at each site, using protocols and quality control pipelines provided by the ENIGMA-DTI working group (http://enigma.ini.usc.edu/protocols/dti-protocols/) and already employed to pool harmonized DTI analyses from around the world [15][16][17] .
Once tensors were estimated, each site conducted a harmonized image analysis for FA quantification using the ENIGMA-DTI protocol, consisting of the tract-based spatial statistics (TBSS) 18    influence of medication status was also explored through a mixed-effects sub-group analysis, comparing effect sizes in medicated (n=8) and unmedicated (n=3) patient cohorts.

RESULTS
Demographics and clinical characteristics of the participants in each site are shown in Tables 1 and 2. Tables 1 and 2 Table 3 indicates the five out of 25 regions with marginally lower FA in patients compared to controls. These are the genu of the corpus callosum, the posterior corona radiata, the posterior thalamic radiation (PTR), the sagittal stratum (SS) and the uncinate fasciculus (see Figure 1). Table 3

Pediatric Cohort
In the pediatric cohort, patients showed no detectable FA abnormalities in any of the regions studied (see Table 4 for statistical details).

DISCUSSION
In the largest coordinated meta-analysis of WM in OCD to date, we demonstrated specific regional WM alterations, in adults with OCD with lower FA in the posterior thalamic radiation (PTR) and sagittal stratum (SS). Metaregression indicated that lower FA in the SS is associated with younger age at onset, longer duration of illness, and being on medication, but not with symptom severity -implying that, as with cortical thickness and subcortical volumes in OCD, the observed alterations may be markers of the disorder.
Findings were significant across the combinations of datasets in the sensitivity analysis for the PTR, while for the SS the effect remained significant when solely medicated patients were considered. We did not find case-control differences in WM microstructure of pediatric subjects.
A role for cerebral WM and oligodendrocytes (the myelinating cells of the central nervous system) in the pathophysiology of many psychiatric disorders has been supported by growing research evidence 6,20 , suggesting abnormalities of myelination status as a possible pathogenic mechanism 21 .
Specifically, altered myelin-related maturational growth may explain the enhanced risk for psychiatric disorders during the transition from childhood to adulthood 22,23 , an age window of intense ongoing brain development 21 .
Although FA is a general measure of microstructure -including variation in regional myelination levels, such as axon demyelination or loss, myelin loss or increased extracellular space -it does not provide a physiologically specific explanation of WM abnormalities 24 . In our study, higher RD in the same bundles, although at a trend level, supports the hypothesis that lower FA reflects a disruption of myelin sheaths 25 , given that RD is a putative myelin marker 24 . The association between myelin degradation in the SS and 1 6 longer illness duration together with the absence of a detectable alteration in pediatric patients, suggests that neuroplastic changes are epiphenomena of prolonged symptomatology, since compulsively engaging in a particular behavior or cognitive process has been suggested to alter brain structure 26,27 . Moreover

8
Axon-derived signals, like glutamate release, may also play a role in the epigenetic regulation of the transcriptional apparatus required for myelination by oligodendrocytes 44 , and in the process of remyelination after damage.
Glutamate-related genes are promising candidates for OCD [45][46][47] , and neuroimaging-gene association studies in animal models reinforce the hypothesis of glutamate involvement in the pathophysiology of the disorder 48 .
Since abnormalities in glutamate neurotransmission and homeostasis have shown to be crucial for OCD onset 49 , glutamate excitotoxicity-induced damage of the myelin sheaths may further explain the decline in myelin integrity suggested here. The PTR and, to a lesser extent, the SS convey fibers principally from the thalamus, a region found to be consistently enlarged in unmedicated pediatric OCD samples 11  Since both the PTR and the SS convey projection fibers to the posterior part of the brain, our results support the idea that OCD involves abnormalities affecting a network of regions that is more extensive than commonly believed 1,52 . Both bundles project to posterior parietal, temporal and occipital cortices, and include many major association fibers (i.e., the inferior 1 9 longitudinal fasciculus and the inferior fronto-occipital fasciculus). Their altered microstructure may be related to the cognitive dysfunctions subtended by intra-hemispheric disconnection 2,53 .
Notably, in our study WM microstructural alterations in OCD were associated with age at scanning. Specifically, WM alterations were observed in the adult cohort only, and were associated with longer illness duration. These findings, which were unrelated to OCD symptom severity, complement previous evidence of differences between adult and pediatric OCD patients in brain morphological 5,12,54 and clinical 55 correlates.
There is evidence that the human brain's protracted myelination 56,57 underpins myelin vulnerability along a continuum from early to late stages of development and disease 58     and percentage of medicated patients across the 11 ENIGMA-OCD sites.
Sphere magnitude indicates sample size.