Resting-state fMRI signals identify bipolar risk features in depressed individuals and their relation to cognitive function.

BACKGROUND

Emerging evidence suggests that individuals with major depressive disorder (MDD) at risk for bipolar disorder (BD) may represent a distinct subtype within the mood disorder spectrum. However, the neurocognitive and neurofunctional characteristics of this group remain underexplored.

METHODS

Resting-state fMRI and cognitive assessments were obtained from currently depressed participants. BD risk factors were identified by Hypomania Checklist-32 (HCL-32) scores ≥14 and/or a first-degree family history of BD. 139 BD, 181 MDD+ (with BD risk), and 138 MDD- (without BD risk) patients were identified. Neuroimaging analyses included the amplitude of low-frequency fluctuations (ALFF) and seed-based functional connectivity (FC). Correlational analyses were conducted to explore the relationships among clinical characteristics, neuroimaging indices, and cognitive performance.

RESULTS

Compared to MDD-, MDD+ demonstrated increased ALFF in the right postcentral gyrus and stronger intra-regional and inter-regional FC with the left supramarginal gyrus. Relative to BD, MDD+ showed reduced ALFF in the left putamen and weakened FC between with the right precuneus. MDD+ also outperformed both MDD- and BD in processing speed and cognitive flexibility. In MDD+, ALFF in the left putamen correlated with processing speed. In BD, ALFF in the left cuneus correlated positively with mania severity and negatively with family history. However, brain-behavior associations did not survive correction for multiple comparisons.

CONCLUSIONS

MDD+ demonstrated distinct cognitive and neural patterns, supporting the notion of a unique depressive subtype. These findings emphasize the importance of refined phenotyping to facilitate early recognition and personalized treatment strategies.