Normative brain variability, clinical dimension, and molecular‐transcriptomic correlates of bipolar risk in depressed patients

AIM: Bipolar disorder (BD) is commonly misdiagnosed as major depressive disorder (MDD) due to depression often preceding manic symptoms. Yet, neurobiological mechanisms underlying emotional states transition in depressed patients remain largely unexplored. METHODS: Leveraging a lifespan normative model from a large healthy cohort (N = 1262), we quantified the structural or functional brain variability for 389 depressed patients (179 MDD, 138 BD, 72 transition to BD [tBD]). The MDD and tBD patients were followed up for 6~13 years. A dimensional approach was employed to dissect the neuroimaging variability across different clinical dimensions in MDD and BD patients, represented as the transdiagnostic covariation modes between clinical risk factors for emotional states transition and brain structural or functional variability. RESULTS: Two covariation modes were identified: Mode 1, tied to earlier age of onset, exhibited reduced activity in limbic/subcortical networks and increased activity in dorsal attention, executive control networks, which facilitated differentiating BD from MDD; Mode 2, associated with the retardation symptom, revealed gray matter atrophy in default mode, limbic and subcortical networks, whose structural pattern identified tBD from MDD. Multifaceted genetic landscape underpinning the structural pattern in Mode 2 suggested dopamine (specifically DRD2-related genetic risk) showed a significant association with the structural deficits in reward circuit, covary with the changes of retardation symptom. CONCLUSION: Our findings aid in better understanding the underlying neurobiological mechanisms of emotional states transition and clinically subtle symptom changes in early-stage BD patients who have never experienced a mania/hypomania episode, providing important target information for early intervention in BD.