Human Brain Mapping. 2015 Aug;36(8):2915-27.
Sub-hubs of Baseline Functional Brain Networks Are Related to Early Improvement Following Two-week Pharmacological Therapy for Major Depressive Disorder
Yuedi Shen1, Jiashu Yao2, Xueyan Jiang3,4, Lei Zhang2, Luoyi Xu2, Rui Feng2, Liqiang Cai2, Jing Liu2, Jinhui Wang3,4, Wei Chen2,5.
1The Affiliated Hospital of Hangzhou Normal University, Hangzhou Normal University, Hangzhou, China.
2Department of Psychiatry, Sir Run Run Shaw Hospital, Zhejiang University School of Medical and the Collaborative Innovation Center for Brain Science, Hangzhou, China.
3Center for Cognition and Brain Disorders, Hangzhou Normal University, Hangzhou, China.
4Zhejiang Key Laboratory for Research in Assessment of Cognitive Impairments, China.
5Key Laboratory of Medical Neurobiology of Chinese Ministry of Health, Zhejiang University School of Medicine, Hangzhou, China
Accumulating evidence suggests that early improvement after two-week antidepressant treatment is predictive of later outcomes of patients with major depressive disorder (MDD); however, whether this early improvement is associated with baseline neural architecture remains largely unknown. Utilizing resting-state functional MRI data and graph-based network approaches, this study calculated voxel-wise degree centrality maps for 24 MDD patients at baseline and linked them with changes in the Hamilton Rating Scale for Depression (HAMD) scores after two weeks of medication. Six clusters exhibited significant correlations of their baseline degree centrality with treatment-induced HAMD changes for the patients, which were mainly categorized into the posterior default-mode network (i.e., the left precuneus, supramarginal gyrus, middle temporal gyrus and right angular gyrus) and frontal regions. Receiver operating characteristic curve and logistic regression analyses convergently revealed excellent performance of these regions in discriminating the early improvement status for the patients, especially the angular gyrus (sensitivity and specificity of 100%). Moreover, the angular gyrus was identified as the optimal regressor as determined by stepwise regression. Interestingly, these regions possessed higher centrality than others in the brain (p < 10-3) although they were not the most highly connected hubs. Finally, we demonstrated a high reproducibility of our findings across several factors (e.g., threshold choice, anatomical distance and temporal cutting) in our analyses. Together, these preliminary exploratory analyses demonstrate the potential of neuroimaging-based network analysis in predicting the early therapeutic improvement of MDD patients and have important implications in guiding earlier personalized therapeutic regimens for possible treatment-refractory depression.
KEYWORDS: centrality; default-mode; depression; early improvement; fMRI; treatment
Major depressive disorder (MDD) is the most common unipolar affective disorder characterized by significant suffering, high morbidity and mortality rates, and psychosocial functional impairments. Despite considerable advances in the development of pharmacological treatment for MDD, clinical studies have reported high rates of non-remission even after several courses of standard antidepressant treatments. This raises an important question: can we know in advance whether a patient will respond to antidepressant therapy?
In this regard, several previous studies have shown that the early improvement status of a MDD patient after two-week antidepressant treatment is predictive of later treatment outcome. Patients with MDD who fail to show early improvement have little chance of stable response or remission (Szegedi et al., 2009) and could benefit from timely switching of treatment strategies (Nakajima et al., 2011). Here, we further question whether the prediction could be shifted to an earlier time, that is, before the treatment begins.
To this end, we utilized resting-state functional connectivity MRI, a noninvasive neuroimaging technique that does not need participants to engage in cognitive activities, to examine the relationship between intrinsic brain network architecture and changes in the Hamilton Rating Scale for Depression scores after two weeks of medication. As it turns out, a specific set of parietal, frontal and temporal regions were identified whose baseline functional connectivity were significantly correlated with short-term treatment-induced HAMD changes for the patients (Figure 1A). Further examinations showed that these regions were densely interconnected (Figure 1B) and possessed high degree centralities (Figure 1C), indicating that they belonged to a common functional module and had hub-like characteristics.
Figure 1. Six regions are identified whose baseline degree centralities are significantly correlated with clinical response of MDD patients to two-week medications (A). Further analyses indicate that the regions show high correlations in their spontaneous brain activity and in their functional connectivity profiles over the whole brain (B), and exhibit hub-like characteristic of high degree centralities (C).
We then asked to what extent the above results were reproducible given the fact that functional MRI has relatively low SNRs and that there exist numerous choices during data preprocessing and network analysis. To address this issue, we conducted a battery of complementary analyses to validate our findings by examining as many optional factors as possible in the whole analysis process. Encouragingly, the ability of baseline functional network architecture to predict short-term antidepressant clinical outcome was found to be largely independent of choices of different analytical strategies.
Finally, we questioned a more practical problem from clinical and economic views that which region could serve as the final candidate to provide the strongest power in the prediction. A qualitative comparison among the identified regions revealed that the right angular gyrus exhibited the strongest predictive power, a convergent finding regardless of the analytical methods of correlation, receiver operating characteristic curve or logistic regression (Figure 2). This was further validated by a stepwise regression model. All these together with our reproducibility analysis collectively suggest that the angular gyrus may serve as a stable biomarker to predict early antidepressant response of MDD patients.
The importance of this study is self-evident. Using resting-state functional connectivity MRI, we provide convincing evidence for the first time that clinical response of MDD patients to short-term antidepressant therapy which is crucial to foreknow the final therapeutic outcome can be predicted. This suggests the potential of resting-state functional brain networks to serve as prognostic neuroimaging biomarkers to guide personalized treatment for MDD. It’s worth mentioning that the potential neuroimaging biomarkers identified in the current study appear to be robustly discovered when different imaging modalities (e.g., structural MRI) (Korgaonkar et al., 2015) or treatment regimens (e.g., electroconvulsive therapy) (van Waarde et al., 2015) are used. This gives us the belief that neuroimaging will be an important complementary tool in the near future to help personalized therapy for individial MDD patients. Now, it’s time to speed up the arrival of that day by strengthening global collaboration for researchers as diverse as neuroscientists, psychiatrists and data-mining experts.
Figure 2. Performances of the right angular gyrus in predicting early improvements of the MDD patients using correlation, receiver operating characteristic curve and logistic regression.
- Korgaonkar MS, Rekshan W, Gordon E, Rush AJ, Williams LM, Blasey C, Grieve SM (2015): Magnetic Resonance Imaging Measures of Brain Structure to Predict Antidepressant Treatment Outcome in Major Depressive Disorder. EBioMedicine 2:37-45.
- Nakajima S, Uchida H, Suzuki T, Watanabe K, Hirano J, Yagihashi T, Takeuchi H, Abe T, Kashima H, Mimura M (2011): Is switching antidepressants following early nonresponse more beneficial in acute-phase treatment of depression?: a randomized open-label trial. Prog Neuropsychopharmacol Biol Psychiatry 35:1983-9.
- Szegedi A, Jansen WT, van Willigenburg AP, van der Meulen E, Stassen HH, Thase ME (2009): Early improvement in the first 2 weeks as a predictor of treatment outcome in patients with major depressive disorder: a meta-analysis including 6562 patients. J Clin Psychiatry 70:344-53.
- van Waarde JA, Scholte HS, van Oudheusden LJ, Verwey B, Denys D, van Wingen GA (2015): A functional MRI marker may predict the outcome of electroconvulsive therapy in severe and treatment-resistant depression. Mol Psychiatry 20:609-614.
This study was supported by the Natural Science Foundation of China (Nos. 81272502, 81301284 and 81371490), Department of Science and Technology of Zhejiang Province (No. 2013C33186) and Zhejiang Provincial Natural Science Foundation of China (No. LZ13C090001).
Jinhui Wang, Ph.D.
Center for Cognition and Brain Disorders
Hangzhou Normal University Hangzhou, China