Mendelian randomization analyses uncover causal relationships between brain structural connectome and risk of psychiatric disorders

AIMS: Growing evidence suggests abnormalities of brain structural connectome in psychiatric disorders, but the causal relationships remain underexplored. Therefore, elucidating the causality is critical for deciphering the neurobiological underpinnings of mental illnesses. METHODS: We conducted bidirectional two-sample Mendelian randomization (MR) analyses to investigate the causal links between 206 white-matter connectivity phenotypes (n = 26,333, UK Biobank) and 13 major psychiatric disorders (n = 14,307 to 1,222,882). RESULTS: Forward MR analyses identified causal effects of five genetically predicted white-matter structural connectivity phenotypes on six psychiatric disorders, with associations being significant or suggestive. For instance, the increase in structural connectivity between the left-hemisphere frontoparietal control network and right-hemisphere default mode network was significantly causally associated with decreased autism spectrum disorder risk, while elevated structural connectivity between the right-hemisphere frontoparietal control network and hippocampus was significantly causally linked to lower risk of both anorexia nervosa and cannabis use disorder. Reverse MR analyses revealed significantly or suggestively causal relationships between the risk of two psychiatric disorders and four different white-matter structural connectivity phenotypes. For example, the heightened susceptibility to anorexia nervosa was found to be significantly causally associated with diminished structural connectivity between the left-hemisphere visual network and pallidum. CONCLUSIONS: These findings offer new insights into the cause of psychiatric disorders and highlight potential biomarkers for early detection and prevention at the brain structural connectome level.