Individualized functional connectome identifies treatment-response biomarkers in patients with depression

Functional connectivity studies implicate circuit dysfunction in depression. However, group-level analyses have not yielded reliable neuroimaging biomarkers for treatment response. Given inter-subject heterogeneity in brain organization, individualized network mapping may better capture the neural correlates of clinical state and treatment response. However, how antidepressant therapies modulate individualized connectivity patterns remains unclear. This study aimed to investigate whether personalized functional connectomes can identify treatment-response biomarkers associated with antidepressant-related symptom improvement, and to elucidate the underlying neural mechanisms in depression. We employed individualized brain parcellation to map subject-specific functional regions in 34 depression patients. Using these personalized networks, we identified cortical connectivity biomarkers that tracked longitudinal changes in functional connections following antidepressant treatments. The performance of these symptom-tracking biomarkers was evaluated via cross-validation and compared directly with conventional group-level atlas-based approaches. Change of Hamilton Depression Scale (HAMD) scores was associated with dysconnectivity between key large-scale networks, particularly the default mode network (DMN), dorsal attention network (dATN), and frontoparietal control network (FPN). A subset of connections between these networks tracked improvement in depressive symptoms after treatment. The individualized analysis showed a significant association with symptom changes, whereas the group-level atlas-based analysis did not. While individualized parcellation systematically reduced the absolute strength of between-network connections, these refined signals demonstrated stronger clinical relevance. Our findings demonstrate that alterations in functional connectivity among the DMN, dATN, and FPN constitute candidate treatment-response biomarkers associated with clinical symptom improvement in depression. Functional connectivity derived from personalized parcellation may serve as a candidate treatment-agnostic biomarker of antidepressant treatment response, highlighting the importance of moving toward individualized neuroimaging approaches to better understand and track clinical outcomes in depression. However, treatment-specific and externally validated biomarkers remain to be established. This study was registered with the Chinese Clinical Trial Registry (ChiCTR, www.chictr.org.cn , ChiCTR-IOR-15007551) on 10 December 2015.