1 INTRODUCTION

Mood dysregulation is a central component of many psychiatric disorders (Aldao et al., 2010). In offspring of individuals with bipolar disorder (BD), mood dysregulation is associated with an increased risk of incipient mania (Geller et al., 2001; M. K. Singh et al., 2007). Previous magnetic resonance imaging (MRI) studies of mood dysregulation have focused mainly on structural alterations in local brain regions and seed-based functional connectivity, which can fail to capture the complex alterations in brain network connectivity that support higher cognitive and affective processes. Alterations at the whole-brain connectome level have been previously reported in individuals with BD and depression (Manelis et al., 2016; W. Zhang et al., 2020), but not in mood-dysregulated youth with familial risk for BD.

Antidepressant medications are the most common treatment for depression and anxiety in youth (Baumer et al., 2006; Cotton et al., 2016). However, these treatments can precipitate adverse effects, including accelerating the onset of mania or hypomania (Reichart & Nolen, 2004; Strawn et al., 2014). Studies evaluating psychological treatment options in mood-dysregulated youth, especially those with a bipolar parent, are urgently needed as an initial step toward establishing alternative interventions that can prevent the later emergence of mania.

Mindfulness-based cognitive therapy (MBCT) is an evidence-based manualised therapy initially developed for adult mood disorders that combines mindfulness training with the characteristics of cognitive behavioural therapy (Hofmann et al., 2010; Williams et al., 2000). The protocol has recently been applied to children and youth with anxiety disorders with promising results (Kim et al., 2009). Coelho et al. reported that one of the significant effects of MBCT is to improve emotion regulation (Coelho et al., 2007). Additionally, Strawn et al. demonstrated that MBCT alters brain activation, mainly in subcortical and emotion-related areas, in anxious youth who are at risk for BD (Strawn et al., 2016). However, little is known about the treatment's effects on whole-brain topological properties.

The use of graph theory modelling has become an informative strategy for evaluating the brain connectome in neuropsychiatric disorders (Sporns et al., 2005). In graph theory approaches, a brain network is represented as a group of nodes connected by edges. Brain regions are considered as nodes, with edges represented as the anatomical connection or correlation between two nodes. Multiple metrics are computed at the global and nodal level to characterise the brain connectome as discussed in Supporting Information, including network properties such as path length (Lp) which reflects the number of connections linking pairs of nodes (Watts & Strogatz, 1998), and global network efficiency (Eglob), which measures how efficiently information can been transferred between nodes at the global level (Latora & Marchiori, 2001).

Disruptions in topological organisation have been found in disorders associated with mood dysregulation, including BD (Manelis et al., 2016), social anxiety disorder (Jacob et al., 2019; Xing et al., 2017), and major depressive disorder (Z. Wang, Yuan, et al., 2016; J. Zhang et al., 2011), including abnormalities at both the global and nodal levels. Furthermore, most connectome studies have focused on resting-state functional magnetic resonance imaging (fMRI) and diffusion tensor imaging, and there are few studies using this approach to examine brain structural morphological networks.

Studies of brain networks using grey matter measures from structural MRI have been informative in characterising interregional morphological associations based on structural covariance in grey matter volume and cortical thickness (Alexander-Bloch et al., 2013; He et al., 2007). Alterations in these associations have been observed in adult patients with mood disorders (Bassett et al., 2008; M. K. Singh, Chang, et al., 2013). However, most previous studies have concentrated on group-level analyses, which limited their application to the investigation of variability in brain structure and its relation to psychiatric symptoms and change after treatment in individual patients (Kong et al., 2015). Tools to construct morphological networks in individual patients have been developed by Tijms et al. (2012) and Kong et al. (2015). These methods have since been applied in studies of psychiatric disorders (Niu et al., 2018a, 2018b; W. Zhang et al., 2020). To date, few studies have examined the impact of psychological treatments on the brain structural network in paediatric populations based on grey matter networks.

In the current study, we explored differences in the topological organisation of individual-level morphological brain networks between mood-dysregulated youth with familial risk for BD and HC at baseline, and evaluated brain network changes after 12 weeks of MBCT for children (MBCT-C) in the mood-dysregulated group. We predicted that there would be altered global topological properties in individual morphological brain networks in mood dysregulated individuals compared with HC. Furthermore, given reports of focal abnormities in emotion-related areas in mood disorders, notably in the limbic system (Drevets et al., 2008; Sitoh & Tien, 1997), combined with our previous report that MBCT-C has a significant effect on connectivity of the limbic system of the functional brain network (Qin et al., 2021), we predicted alterations in nodal topological properties in this circuitry. Based on the reported positive effects of MBCT on the activation of brain structures (Cotton et al., 2016), we hypothesised that there would be positive effects of MBCT-C in reducing the identified topological brain alterations evident at baseline.

2 MATERIALS AND METHODS 2.1 Participants

The study included 10 mood-dysregulated individuals with a familial risk for BD and 15 demographically matched healthy individuals, all between 10 and 17 years of age. Mood-dysregulated at-risk participants met the following inclusion criteria: 1) at least one biological parent with bipolar I disorder, confirmed with the Structured Clinical Interview for DSM-IV-present/lifetime (SCID-P/L), 2) Children's Depression Rating Scale-Revised (CDRS-R) score >28, Young Mania Rating Scale (YMRS) score >12, or Emotion Regulation Checklist (ERC) score <28 (Shields & Cicchetti, 1997). At recruitment, mood dysregulated youth agreed to participate in at least 75% of treatment sessions. Four patients took psychotropic medications [antidepressants (n = 2) and psychostimulants (n = 2)] at stable therapeutic doses throughout the study.

Exclusion criteria were as follows: 1) previously documented diagnosis of mental retardation or an IQ < 70; 2) lifetime DSM-IV diagnosis of BD or any psychotic disorder; 3) previous participation in a mindfulness-based treatment; 4) a substance use disorder (except nicotine or caffeine) within the past 3 months; 5) clinically judged to be at risk for suicide, or a baseline CDRS-R suicide score >3; 6) change in psychopharmacological treatment within 30 days prior to screening; 7) psychotherapy within 2 months prior to screening or initiating psychotherapy during study participation; 8) contraindication to an MRI scan; and 9) history of neurological disorder, head trauma resulting in loss of consciousness for >10 min, or any unstable medical illness. All methods and procedures in this study were approved by the University of Cincinnati Institutional Review Board, and youth participants and their legal guardians provided written, informed assent and consent, respectively. All the mood-dysregulated youths included in analyses completed the MBCT-C protocol that included weekly group sessions, regular home practice, and the core curriculum of formal mindfulness practices (Cotton et al., 2016).

2.2 Graph theory metric calculation

After the raw image data was collected, Statistical Parametric Mapping (SPM) software was used for preprocessing. Individual structural brain networks were constructed based on information regarding regional morphological distribution using a completely automated, data-driven method (H. Wang, Jin, et al., 2016). First, the whole brain was divided into 90 cortical and subcortical regions of interest each representing a network node- using the automated anatomical labelling (AAL) atlas. Next, the edges of the network were defined as the similarity of their morphological features between two regions. This process resulted in a 90 × 90 weighted correlation matrices for each subject. The graph theoretical topological properties were performed using GRETNA software (Wang et al., 2015). Seven global network metrics (clustering coefficient Cp, characteristic path length Lp, normalised clustering coefficient γ, normalised characteristic path length λ, and small-worldness σ, local efficiency Eloc and global efficiency Eglob) and three nodal network metrics (nodal degree, nodal efficiency, and nodal betweenness) were calculated for each subject. The detailed methods for calculating graph theory topological properties and their interpretation are presented in Supporting Information.

2.3 Statistical analysis

The area under the curve (AUC) was calculated for each network metric over the sparsity (S) range from 0.05 < S < 0.40 with an interval of 0.01. The AUC values provide a summary scalar of the network topology representation independent of a single threshold selection. This index has been shown to be sensitive for in detecting topological changes in brain networks (Achard & Bullmore, 2007; He et al., 2009; J. Zhang et al., 2011).

A nonparametric permutation test (http://www.mathworks.com) was used to identify significant differences in the AUCs of the whole-brain network metrics and nodal characteristics of the morphological network between HC and mood-dysregulated at-risk participants at baseline, and between pre- and post-treatment scans in the mood-dysregulated group. In this analysis, the primary focus was on global brain metrics. Nodal level analyses were conducted for exploratory purposes. To correct for multiple comparisons, we adopted the false discovery rate (FDR) procedure to preserve a significance level of .05 (Genovese et al., 2002). Regression analysis using anatomic metrics that differentiated mood-dysregulated and control participants at baseline was used to predict improvements in rating scores [(Emotion Regulation Checklist, ERC) and (Child and Adolescent Mindfulness Measure, CAMM)] after MBCT-C. When significant changes from baseline to post-treatment scans in the mood-dysregulated group were identified in network metrics, partial correlations using age and sex as covariates were performed to evaluate relationships between changes in these network metrics and changes in rating scores (ERC and CAMM).

3 RESULTS 3.1 Demographic and clinical comparisons

Twenty-five participants were evaluated (10 mood dysregulated at-risk participants, age: 14.6 ± 1.8, 4 boys; 15 HC, age: 15.1 ± 1.0, 7 boys). There were no statistically significant differences in age (p = .45) or sex (p = .74) between mood-dysregulated and HC participants. ERC parental ratings of mood-dysregulated at-risk participants after MBCT-C tended to improve, but these changes were not significant (p = .10). CAMM (p = .90) scores for the mood-dysregulated at-risk participants did not change after MBCT-C in this sample.

3.2 Alterations in global brain network properties

At baseline, mood-dysregulated at-risk participants exhibited significantly increased Eglob (p = .041) and decreased Lp (p = .041) relative to HC (Figure 1). Following weeks of MBCT-C treatment, the whole-brain network parameter Eglob was significantly reduced (p = .029), and Lp was significantly increased (p = .032). These significant changes reflected a significant normalisation (shift toward values of the HC group) of both global pretreatment neuroanatomic alterations (FDR corrected) (Figure 3a).

Global topological attribute differences in mood-dysregulated participants before MBCT-C. Global topological attribute differences in the mood-dysregulated group before MBCT-C relative to HC. AUC, area under the curve; Cp, clustering coefficients; Eglob, global efficiency; HC, healthy control; Lp, path length; MBCT-C, mindfulness-based cognitive therapy for children; mood dysregulated_baseline, mood-dysregulated youth at risk for bipolar disorder before MBCT-C; mood dysregulated_follow-up, mood-dysregulated youth at risk for bipolar disorder after MBCT-C

3.3 Alterations in nodal brain network properties

Abnormal nodal network connectivity attributes were also identified before treatment in mood-dysregulated participants at-risk for BD compared to HC. Mood dysregulated individuals exhibited significant increases in the right hippocampus, left putamen, and right temporal pole, and decreases in the left inferior frontal gyrus, left median cingulate, left paracingulate gyri, and left postcentral gyrus (Table 1 and Figure 2). There were no significant changes in nodal features in the mood-dysregulated group after treatment, although with our sample size and the large number of metrics examined we were not adequately powered to detect such effects.

TABLE 1. Brain regions with between-group differences in network node topological properties Nodeb p valuec Nodal degree Nodal betweenness Nodal efficiency Mood dysregulated baseline >HC Hippocampus R .004 .024 .003 Putamen L .024 .028 .005 Temporal pole: middle temporal gyrus R <.001 .002 <.001 Mood dysregulated baseline <HC Inferior frontal gyrus L .006 .006 .011 Median cingulate and paracingulate gyrus L .010 .003 .019 Calcarine fissure and surrounding cortex R .014 .041 .024 Postcentral gyrus L <.001 .013 <.001 Abbreviations: R, L, right, left hemisphere; HC, health control; mood dysregulated baseline, mood-dysregulated youth with a familial risk for bipolar disorder before treatment. a Regions are listed above if there were significant between-group differences in at three nodal centrality parameters. b All the brain regions are from AAL (automated anatomical labelling). c The p value threshold taken as .05. Nodal topological attribute and network connectivity differences in mood-dysregulated group relative to HC before MBCT-C. The results are presented using BrainNet software (http://www.nitrc.org/projects/bnv). CAL, calcarine fissure and surrounding cortex; DCG, median cingulate and paracingulate gyri; HIP, hippocampus; IFGoperc, inferior frontal gyrus, opercular part; L, left; PoCG, temporal pole: middle temporal gyrus; PUT, putamen; R, right; TPOmid, temporal pole: middle temporal gyrus 3.4 Relationship between graph metrics and rating scale scores

Betweenness in the right temporal pole at baseline predicted changes in CAMM scores after treatment (r = −.669, p = .034 uncorrected). Changes from baseline to endpoint in Lp were positively correlated with improvements in ERC scores (r = .866, p = .005 significant with, FDR correction) (Figure 3b).

Longitudinal differences in mood-dysregulated group before and after MBCT-C and correlational analyses. (a) Global topological attribute differences in the mood-dysregulated group before and after MBCT-C. (b) Relationship between the change of global topological measure Lp and ERC clinical rating scores treating age and gender as covariates; regression analysis between CAMM in the mood-dysregulated group after MBCT-C. CAMM, Child and Adolescent Mindfulness Measure; ERC change, the change of emotion regulation checklist clinical rating scale after MBCT-C; HC, healthy control; Lp change, the change of path length before and after MBCT-C; MBCT-C, mindfulness-based cognitive therapy for children; mood dysregulated_baseline, mood-dysregulated youth at risk for bipolar disorder before MBCT-C; mood dysregulated_follow-up, mood-dysregulated youth at risk for bipolar disorder after MBCT-C; right_temporal_pole_bi, the betweenness of the right temporal pole

4 DISCUSSION

In the present study, although there were only 10 cases assessed pre and post, no patient control group (i.e., mood dysregulated youth who did not get MBCT), and no significant changes in either ERC or CAMM that can be attributed to the treatment, however, we demonstrated that at baseline, relative to HC, the mood-dysregulated at-risk youth showed a shift toward a randomised brain network organisation (If the probabilities of having the short- and long range connections are the same, this kind of network is called a random network) characterised by higher Eglob and lower Lp in their structural networks. Similar findings have been observed in other neuropsychiatric disorders, including major depressive disorder (Borchardt et al., 2016; J. Zhang et al., 2011), Alzheimer's disease (Stam et al., 2005) and psychotic disorders (Lynall et al., 2010; W. Zhang et al., 2020); connectome networks in individuals with these disorders have also been shown to be shifted toward the randomised network organisation (Fang et al., 2012; M. K. Singh, Kesler, et al., 2013). Our findings suggest that MBCT-C therapy may effectively reduce these alterations in individuals with mood dysregulation who have a family history of BD.

The human brain is a dynamically interconnected system with two main organisational principles: segregation, reflected by Cp and Eloc, and integration, reflected by shorter Lp and Eglob (Bullmore & Sporns, 2012). The optimal balance between these features supports efficiency in higher brain functions. Increased Eglob and decreased Lp in the mood-dysregulated at-risk participants indicate reductions in segregation and integration of brain structure. Eglob reflects the efficiency of a parallel system, where all the nodes in the network concurrently exchange packets of information. The abnormally elevated Eglob observed in our results may reflect an overactivation of the cerebral cortex leading to abnormal changes in the brain's information processing (Latora & Marchiori, 2001). J. Zhang et al. (2011) reported that patients with major depressive disorder showed significantly lower Lp that controls, similar to our findings, the shortening of the length of the morphological network pathways may limit long-distance functional integration in brain networks. An altered balance between segregation and integration in the brain connectome leads to less modularised information processing (The ability to integrate different modules in the network decreases) and reduced fault tolerance of brain networks (Latora & Marchiori, 2001). Therefore, our findings at baseline reflect a less optimal topological organisation in brain networks characterised by the increased Eglob brings information transfer disorder and reduced Lp constraints for long distance functional integration.

The limbic system is well known to modulate emotion expression via both its intrinsic components and its outputs to extralimbic brain regions (Bartlett et al., 2017; Jeganathan et al., 2018; Sitoh & Tien, 1997). Our observations in limbic and paralimbic systems, including effects in hippocampus, temporal pole, and cingulate and paracingulate cortex suggest that alterations in this circuitry may play a significant role in the neurobiology of mood dysregulation via impact on cognition and emotional reactivity. This interpretation is consistent with previous studies that reported abnormal functional connectivity in the limbic system in individuals with depression (Achard & Bullmore, 2007), which has been correlated with disturbances in emotion regulation (Redlich et al., 2015).

Regression analysis demonstrated a nominally significant ability of MBCT-C levels or outcomes to be predicted based on pretreatment MRI values. Our study provides preliminary suggestive evidence that alterations in the integration of the right temporal pole may predict a positive response to MBCT-C for mood-dysregulated participants, but this needs to be confirmed in future clinical trials.

Comparison of the mood-dysregulated group before and after MBCT-C showed that Eglob decreased and Lp increased at follow-up. The identified changes in the brain connectome indicated increased long-distance connections and local efficiency of network functions. These findings indicate increased capacity for longer distance connections to influence brain function, which suggestive evidence indicates may be related to increased modulation of emotional reactions. Larger studies to determine whether these changes reflect or facilitate clinical improvement following MBCT-C are needed.

These findings are generally consistent with previous observations indicating that mindfulness therapy can normalise brain morphological features such as cortical thickness, which would likely increase processing efficiency and connectivity (Grant et al., 2010), as well as local brain activation patterns (Cotton et al., 2016; Zeidan et al., 2011). MBCT-C increased Lp values in this study, a reflection of increased long-distance connections in the topological organisation of the morphological network. Connection distance or length reflects the potential information flow between more distant regions of the whole-brain network (Sporns & Zwi, 2004), while the Lp value is a measure of the global efficiency of the whole-brain network. A longer path length between nodes suggests a greater efficiency of information transmission between distant brain regions. This change in Lp was related to changes in ERC, suggesting a clinical relevance (r = .866, p = .005).

MBCT-C decreased Eglob values in the current study, indicating that MBCT-C can help to balance the brain's ability to specialise in various types of information processing, thus improving the ability of mood-dysregulated at-risk participants to process emotions and cognition. Thus, after treatment, youth with mood dysregulation showed a reduction in local specialisation and increased broader connectome integration, both shifts toward levels shown by healthy controls. These findings suggest that MBCT-C may be beneficial to patients by modifying the information processing of mood-dysregulated at-risk participants in a manner that improves the information transfer capability between remote brain regions and reduces a heightened level of local specialisation. Such changes could enhance emotion regulation and increase the capacity for complex cognitive functions, two changes that may underlie the therapeutic benefit of MBCT-C treatment.

Our study has certain limitations that need to be considered when interpreting the findings. First, the sample size was small, which limits statistical power to detect more modest differences between groups and changes over time. This is especially relevant considering the need to correct for multiple corrections in statistical analyses when testing for changes in nodal features. As such, this study should be considered preliminary in its findings. Second, the lack of a follow-up assessment in the HC group makes it difficult to confirm that identified brain changes in the mood dysregulated group are a consequence of intervention. While these brain morphological changes were consistent with findings reported in previous mindfulness studies in other populations, conclusions drawn from the current research should be treated with caution. Third, the clinical relevance of our imaging findings remains uncertain. Therapeutic effects on behaviour were variable, and significant group-level improvements in rating scale scores was not identified in our small sample. Furthermore, broader questions such as whether improvement in MR features may precede behavioural change, and whether the identified brain alterations at baseline are related to risk for BD, or if their improvement with treatment might reflect a reduction in later risk for BD, remain questions for future research. The nature and time course of clinical and brain changes associated with MBCT-C and related forms of psychotherapy is another important area of investigation requiring larger datasets and longer-term follow up in clinical trials. Fourth, the brain morphological network extraction method used in this study was based on dividing the brain into 90 brain regions; different template parcellations could yield different findings. Fifth, the regression analysis of relationships between pretreatment anatomic features and improvements in the level of mindfulness after treatment only identified effects that were nominally significant. Thus, those observations in our small sample study need to be considered suggestive.

In summary, the current pilot study demonstrated that the brain structural network in mood-dysregulated youth with familial risk for BD showed abnormal increase in Eglob (an overactivation of the cerebral cortex) and decrease in Lp (long-distance functional integration is limited) compared to HC at baseline. After 12 weeks of MBCT-C, there was a significant and clinically relevant reduction in illness-related connectome alterations. The observed pattern of anatomic network connectome alterations indicates abnormalities in the long-distance top-down regulation of emotional processing in mood dysregulation, and that MBCT-C might enhance the control of emotional processing via enhanced integration of limbic systems into the broader brain connectome.

ACKNOWLEDGEMENTS

The authors of this report would like to thank the families who participated in this study.

CONFLICT OF INTEREST

Dr DelBello is on the lecture bureau for Sunovion, has received research support from Acadia, Allergan, Johnson and Johnson, Lundbeck, Alkermes, Otsuka, Pfizer, Sunovion, Supernus, Shire, and Takeda, has provided consultation or advising for services for Alkermes, Allergan, Axsome, Johnson and Johnson, Lundbeck, Medscape, CME Inc, Neuronetics. Dr Patino has received research support from Acadia, Allergan, Johnson and Johnson, Lundbeck, Alkermes, Otsuka, Pfizer, Sunovion, Supernus, Shire, and has provided consultation services for Boehringer Ingelheim. Dr Sweeney consults to VeriSci. All other authors declare that they have no competing interests.

AUTHOR CONTRIBUTIONS

Conceptualization, Methodology, Software, Writing-Original draft preparation, Visualization: Jing Yang. Conceptualization, Methodology, Writing-Reviewing and Editing: Du Lei. Methodology, Writing-Reviewing and Editing: Xueling Suo. Investigation, Writing-Reviewing and Editing: Maxwell J. Tallman. Methodology: Wenbin Li. Data curation: Kaitlyn M. Bruns. Data curation: Thomas J. Blom. Data curation: Luis Rodrigo Patino Duran. Methodology, Software: Kun Qin. Data curation: Sian Cotton. Conceptualization, Writing, Reviewing and Editing: John A. Sweeney. Conceptualization, Writing-Reviewing and Editing, Supervision: Qiyong Gong. Conceptualization, Investigation, Writing-Reviewing and Editing, Supervision: Melissa P. DelBello.

Data supporting the results of this study can be obtained from the corresponding authors upon reasonable request.

Filename Description eip13245-sup-0001-AppendixS1.docxWord 2007 document , 46.7 KB

Appendix S1. Supporting Information.

Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

REFERENCES

Achard, S., & Bullmore, E. (2007). Efficiency and cost of economical brain functional networks. PLoS Computational Biology, 3(2), e17. https://doi.org/10.1371/journal.pcbi.0030017 Aldao, A., Nolen-Hoeksema, S., & Schweizer, S. (2010). Emotion-regulation strategies across psychopathology: A meta-analytic review. Clinical Psychology Review, 30(2), 217– 237. https://doi.org/10.1016/j.cpr.2009.11.004 Alexander-Bloch, A. F., Vértes, P. E., Stidd, R., Lalonde, F., Clasen, L., Rapoport, J., Giedd, J., Bullmore, E. T., & Gogtay, N. (2013). The anatomical distance of functional connections predicts brain network topology in health and schizophrenia. Cerebral Cortex, 23(1), 127– 138. https://doi.org/10.1093/cercor/bhr388 Bartlett, A. A., Singh, R., & Hunter, R. G. (2017). Anxiety and epigenetics. Advances in Experimental Medicine and Biology, 978, 145– 166. https://doi.org/10.1007/978-3-319-53889-1_8 Bassett, D. S., Bullmore, E., Verchinski, B. A., Mattay, V. S., Weinberger, D. R., & Meyer-Lindenberg, A. (2008). Hierarchical organization of human cortical networks in health and schizophrenia. The Journal of Neuroscience, 28(37), 9239– 9248. https://doi.org/10.1523/jneurosci.1929-08.2008 Baumer, F. M., Howe, M., Gallelli, K., Simeonova, D. I., Hallmayer, J., & Chang, K. D. (2006). A pilot study of antidepressant-induced mania in pediatric bipolar disorder: Characteristics, risk factors, and the serotonin transporter gene. Biological Psychiatry, 60(9), 1005– 1012. https://doi.org/10.1016/j.biopsych.2006.06.010 Borchardt, V., Lord, A. R., Li, M., van der Meer, J., Heinze, H. J., Bogerts, B., Breakspear, M., & Walter, M. (2016). Preprocessing strategy influences graph-based exploration of altered functional networks in major depression. Human Brain Mapping, 37(4), 1422– 1442. https://doi.org/10.1002/hbm.23111 Bullmore, E., & Sporns, O. (2012). The economy of brain network organization. Nature Reviews. Neuroscience, 13(5), 336– 349. https://doi.org/10.1038/nrn3214 Coelho, H. F., Canter, P. H., & Ernst, E. (2007). Mindfulness-based cognitive therapy: Evaluating current evidence and informing future research. Journal of Consulting and Clinical Psychology, 75(6), 1000– 1005. https://doi.org/10.1037/0022-006x.75.6.1000 Cotton, S., Luberto, C. M., Sears, R. W., Strawn, J. R., Stahl, L., Wasson, R. S., Blom, T. J., & Delbello, M. P. (2016). Mindfulness-based cognitive therapy for youth with anxiety disorders at risk for bipolar disorder: A pilot trial. Early Intervention in Psychiatry, 10(5), 426– 434. https://doi.org/10.1111/eip.12216 Drevets, W. C., Price, J. L., & Furey, M. L. (2008). Brain structural and functional abnormalities in mood disorders: Implications for neurocircuitry models of depression. Brain Structure & Function, 213(1–2), 93– 118. https://doi.org/10.1007/s00429-008-0189-x Fang, P., Zeng, L. L., Shen, H., Wang, L., Li, B., Liu, L., & Hu, D. (2012). Increased cortical-limbic anatomical network connectivity in major depression revealed by diffusion tensor imaging. PLoS One, 7(9), e45972. https://doi.org/10.1371/journal.pone.0045972 Geller, B., Zimerman, B., Williams, M., Bolhofner, K., & Craney, J. L. (2001). Bipolar disorder at prospective follow-up of adults who had prepubertal major depressive disorder. The American Journal of Psychiatry, 158(1), 125– 127. https://doi.org/10.1176/appi.ajp.158.1.125 Genovese, C. R., Lazar, N. A., & Nichols, T. (2002). Thresholding of statistical maps in functional neuroimaging using the false discovery rate. NeuroImage, 15(4), 870– 878. https://doi.org/10.1006/nimg.2001.1037 Grant, J. A., Courtemanche, J., Duerden, E. G., Duncan, G. H., & Rainville, P. (2010). Cortical thickness and pain sensitivity in zen meditators. Emotion, 10(1), 43– 53. https://doi.org/10.1037/a0018334 He, Y., Chen, Z. J., & Evans, A. C. (2007). Small-world anatomical networks in the human brain revealed by cortical thickness from MRI. Cerebral Cortex, 17(10), 2407– 2419. https://doi.org/10.1093/cercor/bhl149 He, Y., Dagher, A., Chen, Z., Charil, A., Zijdenbos, A., Worsley, K., & Evans, A. (2009). Impaired small-world efficiency in structural cortical networks in multiple sclerosis associated with white matter lesion load. Brain, 132(Pt 12), 3366– 3379. https://doi.org/10.1093/brain/awp089 Hofmann, S. G., Sawyer, A. T., Witt, A. A., & Oh, D. (2010). The effect of mindfulness-based therapy on anxiety and depression: A meta-analytic review. Journal of Consulting and Clinical Psychology, 78(2), 169– 183. https://doi.org/10.1037/a0018555 Jacob, Y., Shany, O., Goldin, P. R., Gross, J. J., & Hendler, T. (2019). Reappraisal of interpersonal criticism in social anxiety disorder: A brain network hierarchy perspective. Cerebral Cortex, 29(7), 3154– 3167. https://doi.org/10.1093/cercor/bhy181 Jeganathan, J., Perry, A., Bassett, D. S., Roberts, G., Mitchell, P. B., & Breakspear, M. (2018). Fronto-limbic dysconnectivity leads to impaired brain network controllability in young people with bipolar disorder and those at high genetic risk. NeuroImage: Clinical,