A mega-analysis of functional connectivity and network changes in youth major depression

Presented During:

Tuesday, June 25, 2024: 12:00 PM - 1:15 PM
COEX  
Room: Hall D 2  

Poster No:

554 

Submission Type:

Abstract Submission 

Authors:

Nga Yan Tse1, Aswin Ratheesh2, Robin Cash1, Andrew Zalesky1

Institutions:

1Department of Psychiatry, The University of Melbourne, Carlton, Victoria, 2Orygen, Parkville, Victoria

First Author:

Nga Yan Tse  
Department of Psychiatry, The University of Melbourne
Carlton, Victoria

Co-Author(s):

Aswin Ratheesh  
Orygen
Parkville, Victoria
Robin Cash  
Department of Psychiatry, The University of Melbourne
Carlton, Victoria
Andrew ZALESKY, PhD  
Department of Psychiatry, The University of Melbourne
Carlton, Victoria

Introduction:

Major depressive disorder (MDD) has a lifetime prevalence of 11.1-14.6% [1,2] and represents the leading cause of disability due to mental health conditions for young people aged 10-24 years worldwide [3,4]. Functional neuroimaging can delineate the neural substrates of psychiatric, cognitive, and neurological disorders and potentially provide targets for treatment [5-8]. Youth MDD research however lag behind that in adults where existing resting-state functional MRI (rs-fMRI) studies have yielded inconsistent findings [9]. Further, mega-analysis, involving compilation of independent cohort datasets and offering unprecedented advantages of improved generalizability and increased statistical power, has remained unexplored in youth MDD.

Methods:

Here, we conducted the first mega-analysis of functional connectivity (FC) changes in youth MDD, encompassing 810 raw, unprocessed T1 and rs-fMRI images collated from 7 international sites (n=440 youths with MDD and 370 healthy controls aged between 12-25 years). Standardized fMRIprep pre-processing, quality control, and COMBAT harmonization were completed for all fMRI data. Whole-brain FC (connectomes) were mapped for each individual based on the Schaefer Yeo 7-network 400 functional atlas. Using impartial, whole-brain-based statistical inference termed network-based statistic (NBS) [10], mega-analyses of between-group and symptom severity-related connectivity differences were conducted at the scale of functional connections and canonical networks. All NBS analyses were adjusted for age and sex and corrected for family-wise error at p<.05. Linear support vector machines with ridge regularization and leave-one-site-out cross validation were applied to build predictive models of diagnostic status and symptom severity (as measured by the Montgomery-Asberg Depression Rating Scale). Accuracy was tested on unseen datasets kept out of the model training to ensure robustness to inter-site variability. Supplementary analyses were conducted to ensure findings are not biased by global signal regression, atlas choice, or potential head motion differences.

Results:

Our analyses consistently implicated core nodes of the default mode network (DMN) in youth MDD, particularly the rostral/subgenual anterior cingulate, medial prefrontal cortex, posterior cingulate, and precuneus (Fig. 1-2). Altered connectivity of components of the limbic, and dorsal (DAN) and ventral attentional (VAN) networks also tended to emerge, localizing to the orbitofrontal cortex, insula, striatum, and intraparietal sulcus/superior parietal cortex (Fig. 1-2). Strikingly, these regions have been implicated as rich-club nodes in past literature and consistently demonstrated a higher level of hubness in our analyses, supporting extensive earlier studies reporting hub involvement in early psychopathology development.

Critically, individual variation in FC within these networks of regions was significantly associated with depression symptom severity (r=-.58 and r=.65 for hypo- and hyper-connected regions; both p<.001), supporting the clinical importance of these connectivity alterations. Consistently, our machine learning analysis further demonstrated that these FC features provided capacity to classify diagnostic status and depression severity in a generalizable and robust fashion beyond site-specific confounds with good accuracy (averaged cross-validated AUC=73% and r=.63).
Supporting Image: Fig1.png
   ·Fig.1
Supporting Image: Fig2.png
   ·Fig.2
 

Conclusions:

Our data-driven, connectome-wide FC and machine learning analyses converge to implicate robust involvement of hub regions within the DMN, DAN, VAN, and limbic network. Adolescence, coinciding with a protracted period of significant plastic changes and psychosocial transitions, represents a unique window of increased vulnerability to altered hub development. This may in turn confer risks for altered network dynamics and discoordination of a myriad of processes centred on the attentional, affective, and introspective systems, and ultimately early emergence of youth MDD.

Disorders of the Nervous System:

Neurodevelopmental/ Early Life (eg. ADHD, autism)
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 1

Lifespan Development:

Early life, Adolescence, Aging

Modeling and Analysis Methods:

Connectivity (eg. functional, effective, structural) 2
fMRI Connectivity and Network Modeling

Keywords:

DISORDERS
Emotions
FUNCTIONAL MRI
PEDIATRIC
Psychiatric
Psychiatric Disorders
Other - Major depressive disorder; functional connectivity; functional network; youth depression

1|2Indicates the priority used for review

Provide references using author date format

[1] Kessler, R.C., Bromet, E.J. (2013), ‘The epidemiology of depression across cultures’, Annu Rev Public Health,vol. 34, pp. 119-138. doi:10.1146/annurev-publhealth-031912-114409
[2] Kovess-Masfety, V., Alonso, J., Angermeyer, M., et al (2013), ‘Irritable mood in adult major depressive disorder: Results from the world mental health surveys’, Depress Anxiety, vol. 30, no. 4, pp. 395-406. doi:10.1002/da.22033
[3] Gore, F.M., Bloem, P.J.N., Patton, G.C., et al (2011), ‘Global burden of disease in young people aged 10-24 years: A systematic analysis’, The Lancet, vol. 377, no. 9783, pp. 2093-2102. doi:10.1016/S0140-6736(11)60512-6
[4] Vos, T., Allen, C., Arora, M., et al (2016), ‘Global, regional, and national incidence, prevalence, and years lived with disability for 310 diseases and injuries, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015’, The Lancet, vol. 388, no. 10053, pp. 1545-1602. doi:10.1016/S0140-6736(16)31678-6
[5] Menon, V. (2011), ‘Large-scale brain networks and psychopathology: A unifying triple network model’, Trends Cogn Sci, vol. 15, no. 10, pp. 483-506. doi:10.1016/j.tics.2011.08.003
[6] Bressler, S.L., Menon, V. (2010), ‘Large-scale brain networks in cognition: emerging methods and principles’, Trends Cogn Sci, vol. 14, no. 6, pp. 277-290. doi:10.1016/j.tics.2010.04.004
[7] Siddiqi, S.H., Taylor, S.F., Cooke, D., Pascual-Leone, A., George, M.S., Fox, M.D. (2020), ‘Distinct symptom-specific treatment targets for circuit-based neuromodulation’, American Journal of Psychiatry, vol. 177, no. 5, pp. 435-446. doi:10.1176/appi.ajp.2019.19090915
[8] Siddiqi, S.H., Schaper, F.L.W.V.J., Horn, A., et al (2021), ‘Brain stimulation and brain lesions converge on common causal circuits in neuropsychiatric disease’, Nat Hum Behav, vol. 5, no. 12, pp. 1707-1716. doi:10.1038/s41562-021-01161-1
[9] Tse, N.Y., Ratheesh, A., Ganesan, S., Zalesky, A., Cash, R.F.H. (2023), ‘Functional dysconnectivity in youth depression: Systematic review, meta‐analysis, and network-based integration’, Neurosci Biobehav Rev, vol. 153, no. 105394. doi:10.1016/j.neubiorev.2023.105394
[10] Zalesky, A., Fornito, A., Bullmore, E.T. (2010), ‘Network-based statistic: Identifying differences in brain networks’, Neuroimage, vol. 53, no. 4, pp. 1197-1207. doi:10.1016/j.neuroimage.2010.06.041