White Matter Indices of Medication Response in Major Depression: A Diffusion Tensor Imaging Study
Journal Articles
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
BACKGROUND: While response to antidepressants in major depressive disorder is difficult to predict, characterizing the organization and integrity of white matter in the brain with diffusion tensor imaging (DTI) may provide the means to distinguish between antidepressant responders and nonresponders. METHODS: DTI data were collected at 6 sites (Canadian Biomarker Integration Network in Depression-1 [CAN-BIND-1 study]) from 200 (127 women) depressed and 112 (71 women) healthy participants at 3 time points: at baseline, 2 weeks, and 8 weeks following initiation of selective serotonin reuptake inhibitor treatment. Therapeutic response was established by a 50% reduction of symptoms at 8 weeks. Analysis on responders, nonresponders, and control subjects yielded 4 scalar metrics: fractional anisotropy and mean, axial, and radial diffusivity. Region-of-interest analysis was carried out on 40 white matter regions using a skeletonization approach. Mixed-effects regression was incorporated to test temporal trends. RESULTS: The data acquired at baseline showed that axial diffusivity in the external capsule, which overlaps the superior longitudinal fasciculus, was significantly associated with medication response. Regression analysis revealed further baseline differences of responders compared with nonresponders in the cingulum regions, sagittal stratum, and corona radiata. Additional group differences relative to control subjects were seen in the internal capsule, posterior thalamic radiation, and uncinate fasciculus. Most effect sizes were moderate (near 0.5), with a maximum of 0.76 in the cingulum-hippocampus region. No temporal changes in DTI metrics were observed over the 8-week study period. CONCLUSIONS: Several DTI measures of altered white matter specifically distinguished medication responders and nonresponders at baseline and show promise for predicting treatment response in depression.
