Differences in white matter structural networks in family risk of major depressive disorder and suicidality: A connectome analysis.

Depression and suicide are leading global causes of disability and death and are highly familial. Family and individual history of depression and suicide are associated with neurobiological differences including decreased white matter connectivity. However, this has been shown for individual regions, not considering that changes in connectivity may reverberate throughout the brain network. Furthermore, suicide and depression are correlated, but distinct psychological constructs. We use graph theory models to account for the network structure of the brain with high levels of specialization and integration and examine whether they differ by family history of depression or of suicidality within a three-generation longitudinal family study with well-characterized clinical histories. Clinician interviews across three generations were used to classify family risk of depression and suicidality. Then, we created weighted network models using 108 cortical and subcortical regions of interest for 96 individuals using diffusion tensor imaging derived fiber tracts. Global and local summary measures (clustering coefficient, characteristic path length, and global and local efficiencies) and network-based statistics were utilized for group comparison of family history of depression and, separately, of suicidality, adjusted for personal psychopathology. Clustering coefficient and local efficiency were associated with concurrent clinical symptoms. Network-based statistics showed hypoconnected cortical-subcortical subnetworks in individuals with high family risk of depression (including prefrontal and putamen regions) and of suicidality (prefrontal, thalamus and precuneus), after controlling for personal psychopathology. Family history of depression and of suicidality are associated with hypoconnectivity between subcortical and cortical regions, suggesting brain-wide impaired information processing, even in those personally unaffected.