Prospective associations between structural brain development and onset of depressive disorder during adolescence and emerging adulthood

Prospective associations between structural brain development and onset of depressive disorder during adolescence and emerging adulthood | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Prospective associations between structural brain development and onset of depressive disorder during adolescence and emerging adulthood Sarah Whittle, Divyangana Rakesh, Julian Simmons, Orli Schwartz, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4267037/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Brain structural alterations are consistently reported in depressive disorders, yet it remains unclear whether these alterations reflect a pre-existing vulnerability or are the result of psychopathology. We aimed to investigate prospective adolescent neurodevelopmental risk markers for depressive disorder onset, using data from a fifteen-year longitudinal study. A risk-enriched community sample of 161 adolescents who had no history of depressive disorders participated in neuroimaging assessments conducted during early (age 12), mid (age 16) and late adolescence (age 19). Onsets of depressive disorders were assessed for the period spanning early adolescence through emerging adulthood (post-baseline, ages 12 to 27). Forty-six participants (28 female) experienced a first episode of a depressive disorder during the follow-up period; eighty-three participants (36 female) received no mental disorder diagnosis. Joint modelling was used to investigate whether brain structure (subcortical volume, cortical thickness and surface area) or age-related changes in brain structure were associated with the risk of depressive disorder onset. Analyses revealed that age-related increases in a) amygdala volume (hazard ratio [HR] 3.01, p FDR 0.036), and b) thickness of temporal (parahippocampal [HR 3.73, p 0.004] and fusiform gyri [HR 4.14, p 0.003]), insula (HR 4.49, p 0.024) and occipital (lingual gyrus, HR 4.19, p 0.013) regions were associated with the onset of depressive disorder. Findings suggest that relative increases in amygdala volume and temporal, insula, and occipital cortical thickness across adolescence may reflect disturbances of normative brain development, predisposing some individuals to depression. This raises the possibility that prior findings of grey matter decreases in clinically depressed individuals may instead reflect alterations that are caused by disorder-related factors. Biological sciences/Neuroscience Health sciences/Biomarkers/Predictive markers Neurodevelopment adolescence emerging adulthood depression neuroimaging cortical thickness Figures Figure 1 Figure 2 Figure 3 Introduction A dramatic rise in the incidence of depressive disorders occurs during adolescence and emerging adulthood (i.e., age 10-29 years). Approximately 20% of individuals experience a first-onset depressive disorder during adolescence, and ~23% experience a first-onset depressive disorder during the third decade of life 1 . As depressive disorders are the leading cause of global disease burden in these age groups 2 , an improved understanding of the etiological pathways to their onset may provide new information to drive the development of strategies to ameliorate deleterious outcomes. It has been suggested that neurodevelopmental alterations during adolescence are likely to be involved in the etiology of depression 3, 4 , and may underpin deficits in socioemotional functioning that are thought to predispose individuals to developing depression 5 . Brain structural alterations are well documented in individuals with depressive disorders, although they differ based on age, suggesting that neurodevelopmental alterations may differ for those who develop depression in youth versus later in life. For example, meta-analyses have revealed reduced hippocampal volume 6 and thinner cortex in frontal and temporal cortical regions (e.g., rostral anterior cingulate cortex [ACC], orbitofrontal cortex [OFC], inferior temporal and fusiform gyri) in adults with Major Depressive Disorder (MDD), in addition to the insula in adult patients with a first episode of MDD 7 . On the other hand, adolescents with MDD have been found to primarily exhibit reduced surface area of frontal (precentral gyrus), inferior parietal and occipital (i.e., lingual gyrus) regions 7 , in addition to smaller caudate and putamen volume 8 . While there is evidence for structural brain alterations present early in the course of depressive disorder 9 , however, it is unclear whether these changes occur post-disorder onset as a result of disorder-related factors or represent pre-existing differences. Relatively few studies have investigated neuroanatomical changes prior to disorder onset in young people; such study types are required to more confidently draw conclusions about risk mechanisms. Results from these studies have been largely inconsistent. Regarding subcortical structures, for example, larger amygdala volumes (or greater amygdala volume growth) have been associated with subsequent MDD onset in some studies 10, 11 , while other studies have reported null findings 12, 13 . One study found that smaller hippocampal and nucleus accumbens volumes predicted subsequent MDD 11, 14 , but null effects have been observed for other subcortical structures 10-12, 15 . Regarding cortical regions, both larger 16 and smaller/thinner rostral ACC volume/thickness 15, 17 have been prospectively associated with onset of MDD in risk-enriched samples of adolescents and young adults. Other research has failed to find prospective associations for the structure of this region 10-12 . Smaller OFC volume/thickness was found to predict MDD onset in two adolescent studies 15, 17 , but these findings have not been replicated in other work 16 . Single studies have found prospective associations of thinner parahippocampal 12 and insula cortex 17 , greater increases in thickness of precentral gyrus and inferior frontal gyrus (IFG) over time 12 , with the onset of MDD in adolescents. There are a number of limitations with the existing research, some of which may have contributed to inconsistent findings. First, many studies did not account for drop-out over time, and/or investigated a limited age range with relatively short follow-ups (1.5-6 years), meaning that potential subsequent diagnoses may be missed. Second, many studies have not considered sex differences, despite some evidence that these may be important 11, 15 . Third, no prospective studies have investigated surface area, despite the fact that a meta-analysis identified only surface area (i.e., not thickness) alterations in adolescent MDD 7 . Finally, few studies have investigated whether the developmental changes in brain structure are associated with depression risk, despite knowledge that risk pathways to mental health problems likely involve alterations to brain developmental trajectories 18 . In the current study, we aimed to address these limitations, and build on our prior work with the longitudinal Orygen Adolescent Development Study. Our previous work in this sample investigated early adolescent neuroanatomical predictors of depressive disorder onset during adolescence 11, 16 . In this prior work, we investigated whether subcortical volumes and prefrontal cortical volumes and thickness during early and mid-adolescence predicted onset of MDD by age 19. However, we did not capture the peak age for depressive disorder onset, which extends into the third decade of life 1 , nor did we investigate brain development across the whole adolescent period, or account for drop-out over time (potentially biasing findings). We thus extended this work here by examining longitudinal adolescent neuroanatomical predictors of depressive disorder onset throughout the second and third decades of life , and we utilize appropriate statistical techniques that reduce bias due to drop-out. We also focused here on an extended set of regions (across frontal, temporal, parietal, occipital and insula cortical regions, in addition to the amygdala, hippocampus and striatal regions) that have been most strongly implicated in prior MDD meta-analyses and prospective studies of structural grey matter predictors of depressive disorder onset. Also, we investigated both cortical thickness and surface area, in addition to subcortical volume. Given prior mixed findings, we did not make firm hypotheses about the direction of effects (i.e., smaller vs. larger structure predicting disorder onset), however, based on prior findings 7 we tentatively hypothesized that neuroanatomical predictors may differ according to metric (surface area vs. thickness). In secondary analyses, we also investigated whether brain structural predictors differed by gender. Methods Participants Participants were a community sample of 161 adolescents (78 females) recruited from schools across Melbourne, Australia, as part of a longitudinal cohort study (the Orygen Adolescent Development Study). The aim of the study was to examine risk and protective factors for common mental health disorders during adolescence. To maximize variability in such factors, 2,453 early adolescents were screened on key affective temperaments known to promote risk and resilience for psychopathology. A subsample of 415 adolescents was selected to participate in further assessments. Equal numbers of males and females were selected from the following ranges of scores on the Effortful Control and Negative Affectivity dimensions of the Early Adolescent Temperament Questionnaire–Revised 19 : 0–1, 1–2, 2–2.5, and > 2.5 standard deviations above and below the mean, resulting in an oversampling of those with high and low temperamental risk for psychopathology. Exclusion criteria for study entry included current or past depressive, substance use, or eating disorders according to the Schedule for Affective Disorders and Schizophrenia for School-Age Children–Present and Lifetime Version (K-SADS-PL) 20 ; chronic illness; language or learning disabilities; and use of medications known to affect nervous system functioning. Adolescents were invited to participate in five waves of assessments spanning approximately ages 12 to 27. Of the selected 415 adolescents, 245 agreed to participate in at least the first wave of the study and met inclusion criteria. At each wave, comprehensive demographic data were collected, in addition to diagnostic interviews used to determine current and past DSM-IV axis I disorders since the date of last interview (except at wave 1, when lifetime disorder was assessed). At waves 1, 3 and 4 (approximate ages 12, 17, and 19, respectively), participants underwent structural MRI scans. The final sample of 161 represented those participants who participated in brain imaging during at least one MRI wave (and had useable MRI data, see below) and had interview data at that same or a subsequent wave (Table 1 ). 92% of the sample were of Central European genetic ancestry, and 8% were of Chinese ancestry. At each wave, written informed consent was obtained from all adolescents (and their parent or guardian when a participant was under 18 years of age). The Human Research Ethics Committee at the University of Melbourne (Australia) approved this study. Table 1 Final MRI and interview wave for individuals included in the analytic sample. MRI Interview W1 W3 W4 W1 9 0 0 W2 16 3 0 W3 8 12 0 W4 6 24 19 W5 3 13 48 Note: For participants that developed a depressive disorder, the final MRI timepoint is that prior to their disorder onset. W = Wave. MRI At wave 1, MRI scans were performed on a 3-T GE scanner (repetition time = 36 ms, echo time = 9 ms, flip angle = 35°, field of view = 20 cm 2 ) to obtain 124 T 1 -weighted contiguous slices (voxel dimensions = 0.4883×0.4883×1.5 mm). Wave 3 and 4 scans were conducted on a 3-T Siemens scanner (repetition time = 1900 ms, echo time = 2.24 ms; flip angle = 9°, field of view = 23 cm 2 ), producing 176 T 1 -weighted contiguous 0.9 mm thick slices (voxel dimensions = 0.9 mm 3 ). An interscanner reliability assessment showed no interscanner bias (described previously 21 , 22 ). Measurement of region of interest structure Hippocampus and amygdala volumes and mean thickness and surface area of cortical regions were estimated at waves 1, 3 and 4 using the longitudinal stream in FreeSurfer, version 5.1. Cortical regions previously found to be altered (with strongest effect sizes) in meta-analytic work, or significantly predictive of depressive disorder onset during adolescence were investigated. These regions were defined using the Desikan-Killiany atlas 23 , and included: frontal (medial OFC, lateral OFC, rACC, pars orbitalis, pars triangularis, pars opercularis, precentral gyrus), insula, temporal (parahippocampal gyrus, inferior temporal gyrus, fusiform gyrus), parietal (inferior parietal cortex) and occipital (lingual gyrus) regions. All segmentations were visually inspected by trained researchers blind to participant characteristics. Corrections to the grey matter-white matter surface were made to the majority of images (the most consistent issue was underestimation of temporal pole grey matter), and corrected images were re-run through the analysis pipeline. ROIs from the left and right hemisphere were averaged given there were no hypotheses regarding laterality of effects. Diagnostic Interviews At the first four waves, the K-SADS-PL was administered to assess current and past DSM-IV axis I disorders. At wave 5, the Structured Clinical Interview for DSM - 5 was administered, and interviewers probed to assess current and past DSM-IV axis I disorders so that diagnostic criteria were consistent across waves. During the follow-up period (from age 12 to 27), 46 (28.6%, 28 female) of the 161 included participants had a first onset of a depressive disorder (see Fig. 1 ). Thirty-eight had a first onset of MDD, whereas eight had a first onset of another depressive disorder (i.e., persistent depressive disorder, adjustment disorder with depressed mood, depressive disorder not otherwise specified [NOS]). Eighteen of the depressive disorder cases experienced the onset of another disorder prior to their depressive disorder onset (12 experienced an anxiety disorder, four an alcohol or substance use disorder, two a behavioral disorder [i.e., oppositional defiant disorder], and one an eating disorder [eating disorder NOS]). Of the 115 participants who did not experience a depressive disorder during the follow-up period, 32 experienced another form of psychopathology (14 experienced an alcohol or substance use disorder, 10 experienced an anxiety disorder, and 12 experienced a behavioral disorder [e.g., oppositional defiant disorder]). These 32 individuals were excluded from analyses given that their diagnoses were substantially different to the comorbid disorders experienced by the depressive disorder cases (i.e., predominantly alcohol/substance use and behavioral disorders versus predominantly anxiety disorders). Eighty-three participants did not develop any disorder during the follow-up period, and were included as control participants in analyses. Statistical Analysis Joint models 24 , conducted using RStudio version 1.2.5042 and the package JMbayes2 24 , were used to investigate the association between brain structure and the onset of depressive disorder. Joint models allow investigation of the association between longitudinal changes of a continuous variable (sub-model 1) with time-to-event data (sub-model 2, see Fig. 2 ). In the current study, joint models allowed modelling the longitudinal brain structure data and simultaneously investigate its association with onset of depressive disorder. Within the joint models, sub-model 1 examined longitudinal changes in brain structure using a linear mixed effects (LME) model with a fixed effect of time, and random intercept and time slope (nonlinear time effects were not estimated due to concerns of overfitting nonlinear slopes with three time points 25 ). The time-to-event sub-model 2 was built using the structure of the Cox proportional hazards regression model (where the measure of effect is the hazard rate, or risk of depressive disorder onset at time t , given that depressive disorder onset has not occurred up to time t ). The joint modelling framework accounts for right-censored data (e.g., drop-out over time), and allows for imbalanced data. Therefore, we were able to use all MRI data for each participant, whether that be from one, two or three time points. The joint models were estimated within a Bayesian framework using Markov chain Monte Carlo (MCMC) algorithms 24 . In Bayesian estimation, inference is based on the distribution over the parameter space (full posterior distribution) instead of maximum likelihood. The parameter values from the sub-models of the joint model were given as prior information for the full posterior distribution. Three chains were set for the MCMC, leading to three distributions of the possible values for each model parameter. The models provided the posterior mean estimates as hazard ratios (HR) and 95% credible intervals (CI). The Bayesian CIs represent the interval in which the population parameter lies with a given probability. Two sets of joint models were run using 1) a “current value” association structure, whereby onset of depressive disorder at a given point in time t (as determined by interview) is predicted by (interpolated) brain structure at the same time point t , and 2) a slope association structure, whereby onset of depressive disorder at a given point in time t is predicted by the brain structure slope (i.e., age-related change in brain structure) at time t (given only linear time effects were modelled, slope was constant across time). These two sets of joint models were run separately for each brain region (and for cortical regions each brain metric [thickness, surface area]). See Supplementary Information for example model code. Covariates included in all analyses included the time-fixed variables gender (male/female), baseline socioeconomic status (SES) and intracranial volume (ICV; included for subcortical volume and cortical surface area ROIs for “current value” models only). SES was assessed using a measure of income-to-needs, based on reported family income relative to the relevant Australian poverty line for household size. Variables were checked for univariate outliers, and any values outside the 1st and 99th percentiles were winsorized (using the package RobustHD). Surface area and volume variables were standardized prior to entry into models. Exploratory analyses investigated whether gender moderated associations (note that participants reported on their “gender” [male/female] and data on biological sex versus gender identity was not collected). For all models, the Gelman–Rubin diagnostic was used to check the similarity of the obtained distributions for the parameters, and thus assess the convergence of the MCMC sampler. The diagnostic values less than 1.1 were considered as convergence. In addition, density plots and traceplots were visually inspected. Correction for multiple comparisons was performed using a false discovery rate (FDR) of p < 0.05 (where corrections are applied within each group of models [thickness value, thickness slope, surface area value, surface area slope, subcortical volume value, subcortical volume slope, thickness value * gender, etc.]). Supplementary Analyses In sensitivity analyses, we re-ran significant models from primary analyses, with the inclusion of additional covariates: baseline (self-reported) depressive and anxiety symptoms, and baseline (parent-reported) pubertal status (using the Centre for Epidemiological Studies Depression Scale-Revised 26 , Beck Anxiety Inventory 27 , and Pubertal Development Scale 28 , respectively). Missing data for these variables (ranging from 3–9%) were imputed using multiple imputation (using the R package mice 29 , with 50 imputations). Attrition Analysis There were no differences between the final sample (N = 161) and both the screening sample (N = 2,453) and the selected sample (N = 415) on socioeconomic status, affective temperament scores at screening, or gender (p’s > 0.05). Results Brain structure predicting concurrent risk of depressive disorder onset (“current value” association structure) At any given time point, (the current value of) cortical thickness or surface area or subcortical volume was not associated with depressive disorder onset for any ROIs (see Tables S2, S4 and S6 in Supplementary Information). Age-related change in brain structure predicting risk of depressive disorder onset (slope association structure) Cortical thickness . Positive age-related change (i.e., relatively increased slope) of thickness in the fusiform gyrus, parahippocampal gyrus, lingual gyrus, and insula cortex was associated with increased risk of depressive disorder onset (see Table 2 , Fig. 3 and Table S1 ). Table 2 Output of models where slope of brain ROI was significant in predicting onset of depressive disorder. Predictor variables HR 95% CI p pFDR Parahippocampal thickness 3.73 1.36,23.30 0.0003 0.004 Fusiform thickness 4.14 1.54,16.96 0.0004 0.003 Lingual thickness 4.19 1.40,30.72 0.003 0.013 Insula thickness 4.49 1.26,51.93 0.006 0.024 Amygdala volume 3.01 1.23,20.12 0.005 0.036 Note: HR for slopes represent HR for a 1 SD change in random age slope. CI = confidence interval, HR = hazard ratio. Cortical surface area. Slope of surface area was not associated with depressive disorder onset for any ROIs (see Table S3). Subcortical volume. Positive age-related change (i.e., relatively increased volume) of the amygdala was associated with increased risk of depressive disorder onset (see Table 2 and S5). Moderating effects of gender There were no regions where gender significantly moderated the association between current value or slope of structural metric and onset of depressive disorder (see Tables S7-S12). Supplementary analyses The addition of baseline depressive and anxiety symptoms and pubertal status did not alter the pattern of significant findings (see Table S13). Discussion In the current prospective study of neurodevelopmental predictors of depressive disorder onset in youth, age-related change in temporal, insula, and occipital cortical, and subcortical structure were associated with the onset of depressive disorder. Specifically, more positive age-related changes in thickness of temporal, occipital, and insula cortices, and amygdala volume, were associated with increased odds of disorder onset. Notably, age-related change, but not current value (i.e., value at the time of depression onset), of cortical and subcortical grey matter was associated with the risk of depression onset. The lack of findings for the current value of cortical structure is inconsistent with some other studies of cross-sectional cortical structure in young people prior to depression onset 12 , 15 , 17 . However, some other studies have also found null effects (e.g., 10 ). Our findings suggest that during periods of marked brain development, such as adolescence, alterations to the trajectory of brain development are particularly important in the pathogenesis of depressive disorders. Relatively increased age-related thickening of cortical regions predicting depression onset is consistent with some prior research 11 . In addition, several studies that have assessed brain structure in individuals early in the course of their disorder have reported increases in cortical grey matter 30 – 32 . As such, in contrast to consistent findings of structural decreases in individuals with established MDD, it may be that increases in cortical thickness are present prior to and early in the course of depressive disorder, reflecting a disturbance of normal brain development. Whether these structural changes reflect genetic mechanisms or occur as a result of early experience is unclear, although we note that our prior work has shown exposure to adverse environments is associated with similar patterns of relatively increased age-related change in cortical structure 33 . Our findings may also be interpreted in relation to theories that propose protracted development of cortical (particularly prefrontal) regions, combined with accelerated development of subcortical regions during adolescence, to be associated with depression risk. Popular theories suggest that these changes may underlie reduced capacity for regulation of behavior and emotion, promoting risk for disorders such as depression 34 , 35 . Our finding of increased age-related change in amygdala volume predicting depression onset may be partly consistent with these neurodevelopmental theories. This finding was also consistent with our prior investigation of early adolescent brain development changes predicting depression onset by late adolescence (although this prior finding was in females only). Increased age-related change in amygdala volume across adolescence may underlie alterations in emotion reactivity and stress regulation 36 , 37 , predisposing adolescents and emerging adults to depression. Further research is needed to test this theory. Findings of age-related change in temporal, occipital and insula regions associated with risk for depression onset, however, suggest that the aforementioned theories of neurodevelopmental risk for depression may be overly simplistic in their sole focus on prefrontal and subcortical regions 3 . Structural changes in the temporal, occipital and insula regions implicated here have been implicated in several prior studies of depression and depression risk 7 , 12 , 17 . All of these regions are implicated in social and emotional processing. The lingual and fusiform gyri are involved in the perceptual processing of facial emotion expressions 38 . The parahippocampal gyrus is involved in context appraisal 39 . The insula is suggested to be involved in visceral information processing, interoception and subjective emotional experience, including of social emotions such as shame and empathy 40 , 41 . Alterations in the activity of these regions have been observed in several studies of depression 42 – 45 . As such, it is possible that altered developmental trajectories of these regions may underlie altered socio-emotional processing, which may create risk for depression onset 46 . Cortical thickness, but not surface area changes were implicated in depression onset. The interpretation of this finding is unclear, although speculatively, it could indicate specific processes underlying depression risk. It has been suggested that the development of cortical thickness is dependent on local or intrinsic factors 47 , such as synaptic pruning 48 . As such, it may be that alterations in such processes are primarily implicated in the pathogenesis of depression. Further work, however, is needed to understand the mechanisms driving changes in thickness that influence vulnerability to depression. Several limitations of this work should be considered when interpreting findings. First, gender was not found to moderate any associations; however, we may not have had adequate power to detect gender differences. Second, brain imaging was restricted to the adolescent period, and as such, brain development during emerging adulthood was interpolated. Future work is needed with parallel brain imaging and diagnostic assessments across the full developmental period. Third, it is possible that brain structural predictors of depression onset might change across development. While we did not address this possibility, it may be a valuable future direction of this work. Fourth, it is unclear whether the implicated brain changes are relevant to depressive disorder onset specifically, or may be associated with the development of psychopathology more generally. Further research is needed to address this question. To conclude, in a fifteen-year longitudinal study of adolescent risk factors for mental disorder onset, we found that relative age-related increases in the thickness of temporal, insula and occipital brain regions, in addition to amygdala volume, were associated with risk for first onset of depressive disorder. Speculatively, these neurodevelopmental patterns may relate to alterations in social and emotional functioning that precede depression onset. Our findings suggest that increases in cortical thickness are present prior to the onset of depressive disorders, reflecting disturbances of normal brain development predisposing to depression. Decreases in structures commonly observed in extant studies of depression might occur subsequent to onset, because of disorder-related factors. Further research is needed to explore these hypotheses. Declarations Acknowledgments. This study was funded by the Colonial Foundation, the National Health and Medical Research Council (NHMRC; Australia; Program Grant 350241), and the Australian Research Council (ARC; Discovery Grants DP0878136 and DP109 2637). Conflicts of Interest : Nicholas Allen has an equity interest in and received salary from Ksana Health Inc. No Ksana Health services or products were used in the current research project. No other authors report financial relationships with commercial interests. References Rohde P, Lewinsohn PM, Klein DN, Seeley JR, Gau JM. Key characteristics of major depressive disorder occurring in childhood, adolescence, emerging adulthood, and adulthood. Clinical Psychological Science 2013; 1 (1) : 41-53. Gore FM, Bloem PJ, Patton GC, Ferguson J, Joseph V, Coffey C et al. Global burden of disease in young people aged 10–24 years: a systematic analysis. The Lancet 2011; 377 (9783) : 2093-2102. Rakesh D, Allen NB, Whittle S. Balancing act: Neural correlates of affect dysregulation in youth depression and substance use–A systematic review of functional neuroimaging studies. Developmental cognitive neuroscience 2020; 42: 100775. Casey BJ, Getz S, Galvan A. The adolescent brain. Developmental review 2008; 28 (1) : 62-77. Arnarson EÖ, Matos AP, Salvador C, Ribeiro C, de Sousa B, Edward Craighead W. Longitudinal study of life events, well-being, emotional regulation and depressive symptomatology. Journal of Psychopathology and Behavioral Assessment 2016; 38: 159-171. Schmaal L, Veltman DJ, van Erp TG, Sämann P, Frodl T, Jahanshad N et al. Subcortical brain alterations in major depressive disorder: findings from the ENIGMA Major Depressive Disorder working group. Molecular psychiatry 2016; 21 (6) : 806-812. Schmaal L, Hibar D, Sämann PG, Hall G, Baune B, Jahanshad N et al. Cortical abnormalities in adults and adolescents with major depression based on brain scans from 20 cohorts worldwide in the ENIGMA Major Depressive Disorder Working Group. Molecular psychiatry 2017; 22 (6) : 900-909. Kim JH, Suh S-i, Lee H-J, Lee J-h, Lee M-S. Cortical and subcortical gray matter alterations in first-episode drug-naïve adolescents with major depressive disorder. Neuroreport 2019; 30 (17) : 1172. Ramezani M, Johnsrude I, Rasoulian A, Bosma R, Tong R, Hollenstein T et al. Temporal-lobe morphology differs between healthy adolescents and those with early-onset of depression. NeuroImage: Clinical 2014; 6: 145-155. Nickson T, Chan S, Papmeyer M, Romaniuk L, Macdonald A, Stewart T et al. Prospective longitudinal voxel-based morphometry study of major depressive disorder in young individuals at high familial risk. Psychological medicine 2016; 46 (11) : 2351-2361. Whittle S, Lichter R, Dennison M, Vijayakumar N, Schwartz O, Byrne ML et al. Structural brain development and depression onset during adolescence: a prospective longitudinal study. American Journal of Psychiatry 2014; 171 (5) : 564-571. Papmeyer M, Giles S, Sussmann JE, Kielty S, Stewart T, Lawrie SM et al. Cortical thickness in individuals at high familial risk of mood disorders as they develop major depressive disorder. Biological psychiatry 2015; 78 (1) : 58-66. Shapero BG, Chai XJ, Vangel M, Biederman J, Hoover CS, Whitfield-Gabrieli S et al. Neural markers of depression risk predict the onset of depression. Psychiatry Research: Neuroimaging 2019; 285: 31-39. Little K, Olsson CA, Youssef GJ, Whittle S, Simmons JG, Yücel M et al. Linking the serotonin transporter gene, family environments, hippocampal volume and depression onset: A prospective imaging gene× environment analysis. Journal of abnormal psychology 2015; 124 (4) : 834. Vulser H, Lemaitre H, Artiges E, Miranda R, Penttilä J, Struve M et al. Subthreshold depression and regional brain volumes in young community adolescents. Journal of the American Academy of Child & Adolescent Psychiatry 2015; 54 (10) : 832-840. Little K, Olsson CA, Whittle S, Youssef G, Byrne ML, Simmons JG et al. Association between serotonin transporter genotype, brain structure and adolescent-onset major depressive disorder: a longitudinal prospective study. Translational Psychiatry 2014; 4 (9) : e445-e445. Foland-Ross LC, Sacchet MD, Prasad G, Gilbert B, Thompson PM, Gotlib IH. Cortical thickness predicts the first onset of major depression in adolescence. International Journal of Developmental Neuroscience 2015; 46: 125-131. Rakesh D, Whittle S, Sheridan MA, McLaughlin KA. Childhood socioeconomic status and the pace of structural neurodevelopment: accelerated, delayed, or simply different? Trends in Cognitive Sciences 2023. Revision of the early adolescent temperament questionnaire. Proceedings of the Poster presented at the 2001 biennial meeting of the society for research in child development, Minneapolis, Minnesota 2001. Citeseer. Kaufman J, Birmaher B, Brent D, Rao U, Flynn C, Moreci P et al. Schedule for affective disorders and schizophrenia for school-age children-present and lifetime version (K-SADS-PL): initial reliability and validity data. Journal of the American Academy of Child & Adolescent Psychiatry 1997; 36 (7) : 980-988. Vijayakumar N, Allen NB, Youssef G, Dennison M, Yücel M, Simmons JG et al. Brain development during adolescence: A mixed‐longitudinal investigation of cortical thickness, surface area, and volume. Human brain mapping 2016; 37 (6) : 2027-2038. Dennison M, Whittle S, Yücel M, Vijayakumar N, Kline A, Simmons J et al. Mapping subcortical brain maturation during adolescence: evidence of hemisphere‐and sex‐specific longitudinal changes. Developmental science 2013; 16 (5) : 772-791. Desikan RS, Ségonne F, Fischl B, Quinn BT, Dickerson BC, Blacker D et al. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage 2006; 31 (3) : 968-980. Rizopoulos D, Papageorgiou G, Miranda Afonso P. JMbayes2: extended joint models for longitudinal and time-to-event data. R package version 03-0, ed 2022. King KM, Littlefield AK, McCabe CJ, Mills KL, Flournoy J, Chassin L. Longitudinal modeling in developmental neuroimaging research: Common challenges, and solutions from developmental psychology. Developmental cognitive neuroscience 2018; 33: 54-72. Eaton WW, Muntaner C, Smith C, Tien A, Ybarra M. Center for epidemiologic studies depression scale: Review and revision. The use of psychological testing for treatment planning and outcomes assessment 2004 : p363-p377. Beck AT, Epstein N, Brown G, Steer R. Beck anxiety inventory. Journal of consulting and clinical psychology 1993. Brooks-Gunn J, Warren MP, Rosso J, Gargiulo J. Validity of self-report measures of girls' pubertal status. Child development 1987 : 829-841. van Buuren S, Groothuis-Oudshoorn K, Robitzsch A, Vink G, Doove L, Jolani S. Package ‘mice’. Computer software 2015; 20 . Qiu L, Lui S, Kuang W, Huang X, Li J, Zhang J et al. Regional increases of cortical thickness in untreated, first-episode major depressive disorder. Translational psychiatry 2014; 4 (4) : e378-e378. Reynolds S, Carrey N, Jaworska N, Langevin LM, Yang X-R, MacMaster FP. Cortical thickness in youth with major depressive disorder. BMC psychiatry 2014; 14 (1) : 1-8. van Eijndhoven P, van Wingen G, Katzenbauer M, Groen W, Tepest R, Fernández G et al. Paralimbic cortical thickness in first-episode depression: evidence for trait-related differences in mood regulation. American Journal of Psychiatry 2013; 170 (12) : 1477-1486. Whittle S, Simmons JG, Dennison M, Vijayakumar N, Schwartz O, Yap MB et al. Positive parenting predicts the development of adolescent brain structure: a longitudinal study. Developmental cognitive neuroscience 2014; 8: 7-17. Somerville LH, Jones RM, Casey B. A time of change: behavioral and neural correlates of adolescent sensitivity to appetitive and aversive environmental cues. Brain and cognition 2010; 72 (1) : 124-133. Casey B, Jones RM, Somerville LH. Braking and accelerating of the adolescent brain. Journal of Research on Adolescence 2011; 21 (1) : 21-33. Dedovic K, Duchesne A, Andrews J, Engert V, Pruessner JC. The brain and the stress axis: the neural correlates of cortisol regulation in response to stress. Neuroimage 2009; 47 (3) : 864-871. Phelps EA. Human emotion and memory: interactions of the amygdala and hippocampal complex. Current opinion in neurobiology 2004; 14 (2) : 198-202. Kesler ML, Andersen AH, Smith CD, Avison MJ, Davis CE, Kryscio RJ et al. Neural substrates of facial emotion processing using fMRI. Cognitive Brain Research 2001; 11 (2) : 213-226. Rudy JW. Context representations, context functions, and the parahippocampal–hippocampal system. Learning & memory 2009; 16 (10) : 573-585. Uddin LQ, Nomi JS, Hébert-Seropian B, Ghaziri J, Boucher O. Structure and function of the human insula. Journal of clinical neurophysiology: official publication of the American Electroencephalographic Society 2017; 34 (4) : 300. Pulcu E, Lythe K, Elliott R, Green S, Moll J, Deakin JF et al. Increased amygdala response to shame in remitted major depressive disorder. PloS one 2014; 9 (1) : e86900. Suh JS, Minuzzi L, Cudney LE, Maich W, Eltayebani M, Soares CN et al. Cerebral cortical thickness after treatment with desvenlafaxine succinate in major depressive disorder. NeuroReport 2019; 30 (5) : 378-382. Lee JS, Kang W, Kang Y, Kim A, Han K-M, Tae W-S et al. Alterations in the occipital cortex of drug-naïve adults with major depressive disorder: A surface-based analysis of surface area and cortical thickness. Psychiatry Investigation 2021; 18 (10) : 1025. Li X, Wang J. Abnormal neural activities in adults and youths with major depressive disorder during emotional processing: a meta-analysis. Brain imaging and Behavior 2021; 15: 1134-1154. Delvecchio G, Fossati P, Boyer P, Brambilla P, Falkai P, Gruber O et al. Common and distinct neural correlates of emotional processing in bipolar disorder and major depressive disorder: a voxel-based meta-analysis of functional magnetic resonance imaging studies. European Neuropsychopharmacology 2012; 22 (2) : 100-113. Allen NB, Badcock PB. The social risk hypothesis of depressed mood: evolutionary, psychosocial, and neurobiological perspectives. Psychological bulletin 2003; 129 (6) : 887. Wierenga LM, Langen M, Oranje B, Durston S. Unique developmental trajectories of cortical thickness and surface area. Neuroimage 2014; 87: 120-126. Jeon T, Mishra V, Ouyang M, Chen M, Huang H. Synchronous changes of cortical thickness and corresponding white matter microstructure during brain development accessed by diffusion MRI tractography from parcellated cortex. Frontiers in Neuroanatomy 2015; 9: 158. Additional Declarations The authors have declared there is NO conflict of interest to disclose Supplementary Files SupplementaryMaterialMP.docx Supplementary Information Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4267037","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":292259510,"identity":"3a82a5a9-216d-4ebe-92d2-180d7d94bfc1","order_by":0,"name":"Sarah Whittle","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIie3OPQrCMBTA8edSl0LXJ4K9QqRQBHuYdNEl7VIQB8GCoIs4FzyFCM7VQKceIODgBToogjiJiY4SPzaH/IfkZfiRB2Ay/WU25AAdcOSoBipv8pnkFKGRPgh+R0ARkj9eXxA3i7b8dMWWJ6LtTgwxhvpkgzDiWkJETNVini8HzkpMwC4GCMUb0mREkXAjGOHRFMMUmY9g6Ym7fJLxOlPkJolbSXLTE9g/CSWoSKp+sX2sTd8sdqhIXvawnZUV4azAxLJ7SSdc9PWLzUvvOAwC15kx78xGQezU+UocL139Yi9Z6qA/AJPJZDK9dgdPklfkz0OZCwAAAABJRU5ErkJggg==","orcid":"","institution":"University of Melbourne","correspondingAuthor":true,"prefix":"","firstName":"Sarah","middleName":"","lastName":"Whittle","suffix":""},{"id":292259511,"identity":"584ffe53-fa9f-45e6-9d58-8df2a0905e09","order_by":1,"name":"Divyangana Rakesh","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Divyangana","middleName":"","lastName":"Rakesh","suffix":""},{"id":292259512,"identity":"5f1e1a13-27cf-4753-b0d4-727096cf8463","order_by":2,"name":"Julian Simmons","email":"","orcid":"https://orcid.org/0000-0002-7228-1847","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Julian","middleName":"","lastName":"Simmons","suffix":""},{"id":292259513,"identity":"afcc0431-7ff5-4bef-b700-cf866a4a4e63","order_by":3,"name":"Orli Schwartz","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Orli","middleName":"","lastName":"Schwartz","suffix":""},{"id":292259514,"identity":"288b1123-5887-4bd3-a771-7a20afe19db8","order_by":4,"name":"Nandita Vijayakumar","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Nandita","middleName":"","lastName":"Vijayakumar","suffix":""},{"id":292259515,"identity":"5ea02ea7-722d-4288-b2b5-4f45a1bda617","order_by":5,"name":"Nicholas Allen","email":"","orcid":"https://orcid.org/0000-0002-1086-6639","institution":"University of Oregon","correspondingAuthor":false,"prefix":"","firstName":"Nicholas","middleName":"","lastName":"Allen","suffix":""}],"badges":[],"createdAt":"2024-04-15 03:55:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4267037/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4267037/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":55339246,"identity":"e89a4e76-ee4b-4368-8e7b-35502c47232e","added_by":"auto","created_at":"2024-04-26 01:47:06","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":39596,"visible":true,"origin":"","legend":"\u003cp\u003eCumulative number of depressive disorder onsets in males and females, from wave 1 (age = 12; years since baseline = 0) to wave 5 (age = 27; years since baseline = 16).\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4267037/v1/2a3a87918fcfed3cbc3b4552.jpg"},{"id":55339248,"identity":"f4df35f7-852c-479c-919d-175cd4a786b0","added_by":"auto","created_at":"2024-04-26 01:47:06","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":12963,"visible":true,"origin":"","legend":"\u003cp\u003eThe joint model.\u003c/p\u003e\n\u003cp\u003eNote: A illustrative representation. The top panel depicts the hazard process, which describes how the hazard function of an event (depressive disorder) changes over time. In the bottom panel the blue dots depict the longitudinal observations (brain structure) and the black line depicts the underlying longitudinal process (brain development). The red arrow depicts the slope of the longitudinal process (because we only investigated linear age effects here, the slope is the same across time). The joint model associates the hazard function at any particular time point \u003cem\u003et\u003c/em\u003e, denoted by the vertical dashed line, with the value (blue circle) or slope (red arrow) of the longitudinal process at the same time point.\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4267037/v1/eccd78fc838376696954329a.jpg"},{"id":55339247,"identity":"fecdb184-52ec-40eb-b1c3-16c142d73a07","added_by":"auto","created_at":"2024-04-26 01:47:06","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":21322,"visible":true,"origin":"","legend":"\u003cp\u003eRegions (and hazard ratios) where positive age-related change in thickness predicted depressive disorder onset.\u003c/p\u003e","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4267037/v1/96a2ca5e114d91a23c0f9086.jpg"},{"id":58675244,"identity":"06292235-e25e-43f2-aa1f-d2039de47bf8","added_by":"auto","created_at":"2024-06-19 15:36:21","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":690692,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4267037/v1/b4327779-780d-4577-bc5a-5488239deeaf.pdf"},{"id":55339250,"identity":"feed8853-7bdf-4de4-a90b-115c7bdac519","added_by":"auto","created_at":"2024-04-26 01:47:06","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":776938,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Information\u003c/p\u003e","description":"","filename":"SupplementaryMaterialMP.docx","url":"https://assets-eu.researchsquare.com/files/rs-4267037/v1/35e40ee69a465deb4ba22b04.docx"}],"financialInterests":"The authors have declared there is \u003cb\u003eNO\u003c/b\u003e conflict of interest to disclose","formattedTitle":"Prospective associations between structural brain development and onset of depressive disorder during adolescence and emerging adulthood","fulltext":[{"header":"Introduction","content":"\u003cp\u003eA dramatic rise in the incidence of depressive disorders occurs during adolescence and emerging adulthood (i.e., age 10-29 years). Approximately 20% of individuals experience a first-onset depressive disorder during adolescence, and ~23% experience a first-onset depressive disorder during the third decade of life\u003csup\u003e1\u003c/sup\u003e. As depressive disorders are the leading cause of global disease burden in these age groups\u003csup\u003e2\u003c/sup\u003e, an improved understanding of the etiological pathways to their onset may provide new information to drive the development of strategies to ameliorate deleterious outcomes.\u003c/p\u003e\n\u003cp\u003eIt has been suggested that neurodevelopmental alterations during adolescence are likely to be involved in the etiology of depression\u003csup\u003e3, 4\u003c/sup\u003e, and may underpin deficits in socioemotional functioning that are thought to predispose individuals to developing depression\u003csup\u003e5\u003c/sup\u003e. Brain structural alterations are well documented in individuals with depressive disorders, although they differ based on age, suggesting that neurodevelopmental alterations may differ for those who develop depression in youth versus later in life. For example, meta-analyses have revealed reduced hippocampal volume\u003csup\u003e6\u003c/sup\u003e and thinner cortex in frontal and temporal cortical regions (e.g., rostral anterior cingulate cortex [ACC], orbitofrontal cortex [OFC], inferior temporal and fusiform gyri) in adults with Major Depressive Disorder (MDD), in addition to the insula in adult patients with a first episode of MDD\u003csup\u003e7\u003c/sup\u003e. On the other hand, adolescents with MDD have been found to primarily exhibit reduced surface area of frontal (precentral gyrus), inferior parietal and occipital (i.e., lingual gyrus) regions\u003csup\u003e7\u003c/sup\u003e, in addition to smaller caudate and putamen volume\u003csup\u003e8\u003c/sup\u003e. While there is evidence for structural brain alterations present early in the course of depressive disorder\u003csup\u003e9\u003c/sup\u003e, however, it is unclear whether these changes occur post-disorder onset as a result of disorder-related factors or represent pre-existing differences.\u003c/p\u003e\n\u003cp\u003eRelatively few studies have investigated neuroanatomical changes prior to disorder onset in young people; such study types are required to more confidently draw conclusions about risk mechanisms. Results from these studies have been largely inconsistent. Regarding subcortical structures, for example, larger amygdala volumes (or greater amygdala volume growth) have been associated with subsequent MDD onset in some studies\u003csup\u003e10, 11\u003c/sup\u003e, while other studies have reported null findings\u003csup\u003e12, 13\u003c/sup\u003e. One study found that smaller hippocampal and nucleus accumbens volumes predicted subsequent MDD\u003csup\u003e11, 14\u003c/sup\u003e, but null effects have been observed for other subcortical structures\u003csup\u003e10-12, 15\u003c/sup\u003e. Regarding cortical regions, both larger\u003csup\u003e16\u003c/sup\u003e and smaller/thinner rostral ACC volume/thickness\u003csup\u003e15, 17\u003c/sup\u003e have been prospectively associated with onset of MDD in risk-enriched samples of adolescents and young adults. Other research has failed to find prospective associations for the structure of this region\u003csup\u003e10-12\u003c/sup\u003e. Smaller OFC volume/thickness was found to predict MDD onset in two adolescent studies\u003csup\u003e15, 17\u003c/sup\u003e, but these findings have not been replicated in other work\u003csup\u003e16\u003c/sup\u003e. Single studies have found prospective associations of thinner parahippocampal\u003csup\u003e12\u003c/sup\u003e and insula cortex\u003csup\u003e17\u003c/sup\u003e, greater \u003cem\u003eincreases\u003c/em\u003e in thickness of precentral gyrus and inferior frontal gyrus (IFG) over time\u003csup\u003e12\u003c/sup\u003e, with the onset of MDD in adolescents.\u003c/p\u003e\n\u003cp\u003eThere are a number of limitations with the existing research, some of which may have contributed to inconsistent findings. First, many studies did not account for drop-out over time, and/or investigated a limited age range with relatively short follow-ups (1.5-6 years), meaning that potential subsequent diagnoses may be missed. Second, many studies have not considered sex differences, despite some evidence that these may be important\u003csup\u003e11, 15\u003c/sup\u003e. Third, no prospective studies have investigated surface area, despite the fact that a meta-analysis identified only surface area (i.e., not thickness) alterations in adolescent MDD\u003csup\u003e7\u003c/sup\u003e. Finally, few studies have investigated whether the developmental changes in brain structure are associated with depression risk, despite knowledge that risk pathways to mental health problems likely involve alterations to brain developmental trajectories\u003csup\u003e18\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eIn the current study, we aimed to address these limitations, and build on our prior work with the longitudinal Orygen Adolescent Development Study. Our previous work in this sample investigated early adolescent neuroanatomical predictors of depressive disorder onset during adolescence\u003csup\u003e11, 16\u003c/sup\u003e. In this prior work, we investigated whether subcortical volumes and prefrontal cortical volumes and thickness during early and mid-adolescence predicted onset of MDD by age 19. However, we did not capture the peak age for depressive disorder onset, which extends into the third decade of life\u003csup\u003e1\u003c/sup\u003e, nor did we investigate brain development across the whole adolescent period, or account for drop-out over time (potentially biasing findings). We thus extended this work here by examining longitudinal adolescent neuroanatomical predictors of depressive disorder onset throughout the second \u003cem\u003eand third\u003c/em\u003e decades of life\u003cem\u003e, \u003c/em\u003eand we utilize appropriate statistical techniques that reduce bias due to drop-out. We also focused here on an extended set of regions (across frontal, temporal, parietal, occipital and insula cortical regions, in addition to the amygdala, hippocampus and striatal regions) that have been most strongly implicated in prior MDD meta-analyses and prospective studies of structural grey matter predictors of depressive disorder onset. Also, we investigated both cortical thickness and surface area, in addition to subcortical volume. Given prior mixed findings, we did not make firm hypotheses about the direction of effects (i.e., smaller vs. larger structure predicting disorder onset), however, based on prior findings\u003csup\u003e7\u003c/sup\u003e we tentatively hypothesized that neuroanatomical predictors may differ according to metric (surface area vs. thickness). In secondary analyses, we also investigated whether brain structural predictors differed by gender.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003eParticipants\u003c/h2\u003e \u003cp\u003eParticipants were a community sample of 161 adolescents (78 females) recruited from schools across Melbourne, Australia, as part of a longitudinal cohort study (the Orygen Adolescent Development Study). The aim of the study was to examine risk and protective factors for common mental health disorders during adolescence. To maximize variability in such factors, 2,453 early adolescents were screened on key affective temperaments known to promote risk and resilience for psychopathology. A subsample of 415 adolescents was selected to participate in further assessments. Equal numbers of males and females were selected from the following ranges of scores on the Effortful Control and Negative Affectivity dimensions of the Early Adolescent Temperament Questionnaire\u0026ndash;Revised\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e: 0\u0026ndash;1, 1\u0026ndash;2, 2\u0026ndash;2.5, and \u0026gt;\u0026thinsp;2.5 standard deviations above and below the mean, resulting in an oversampling of those with high and low temperamental risk for psychopathology. Exclusion criteria for study entry included current or past depressive, substance use, or eating disorders according to the Schedule for Affective Disorders and Schizophrenia for School-Age Children\u0026ndash;Present and Lifetime Version (K-SADS-PL)\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e; chronic illness; language or learning disabilities; and use of medications known to affect nervous system functioning.\u003c/p\u003e \u003cp\u003eAdolescents were invited to participate in five waves of assessments spanning approximately ages 12 to 27. Of the selected 415 adolescents, 245 agreed to participate in at least the first wave of the study and met inclusion criteria. At each wave, comprehensive demographic data were collected, in addition to diagnostic interviews used to determine current and past DSM-IV axis I disorders since the date of last interview (except at wave 1, when lifetime disorder was assessed). At waves 1, 3 and 4 (approximate ages 12, 17, and 19, respectively), participants underwent structural MRI scans. The final sample of 161 represented those participants who participated in brain imaging during at least one MRI wave (and had useable MRI data, see below) and had interview data at that same or a subsequent wave (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). 92% of the sample were of Central European genetic ancestry, and 8% were of Chinese ancestry. At each wave, written informed consent was obtained from all adolescents (and their parent or guardian when a participant was under 18 years of age). The Human Research Ethics Committee at the University of Melbourne (Australia) approved this study.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eFinal MRI and interview wave for individuals included in the analytic sample.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eMRI\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInterview\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eW1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eW3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eW4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eW1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eW2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eW3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eW4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eW5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e48\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eNote: For participants that developed a depressive disorder, the final MRI timepoint is that prior to their disorder onset.\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eW\u0026thinsp;=\u0026thinsp;Wave.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eMRI\u003c/h2\u003e \u003cp\u003eAt wave 1, MRI scans were performed on a 3-T GE scanner (repetition time\u0026thinsp;=\u0026thinsp;36 ms, echo time\u0026thinsp;=\u0026thinsp;9 ms, flip angle\u0026thinsp;=\u0026thinsp;35\u0026deg;, field of view\u0026thinsp;=\u0026thinsp;20 cm\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e) to obtain 124 T\u003csub\u003e1\u003c/sub\u003e-weighted contiguous slices (voxel dimensions\u0026thinsp;=\u0026thinsp;0.4883\u0026times;0.4883\u0026times;1.5 mm). Wave 3 and 4 scans were conducted on a 3-T Siemens scanner (repetition time\u0026thinsp;=\u0026thinsp;1900 ms, echo time\u0026thinsp;=\u0026thinsp;2.24 ms; flip angle\u0026thinsp;=\u0026thinsp;9\u0026deg;, field of view\u0026thinsp;=\u0026thinsp;23 cm\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e), producing 176 T\u003csub\u003e1\u003c/sub\u003e-weighted contiguous 0.9 mm thick slices (voxel dimensions\u0026thinsp;=\u0026thinsp;0.9 mm\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e). An interscanner reliability assessment showed no interscanner bias (described previously\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eMeasurement of region of interest structure\u003c/h2\u003e \u003cp\u003eHippocampus and amygdala volumes and mean thickness and surface area of cortical regions were estimated at waves 1, 3 and 4 using the longitudinal stream in FreeSurfer, version 5.1. Cortical regions previously found to be altered (with strongest effect sizes) in meta-analytic work, or significantly predictive of depressive disorder onset during adolescence were investigated. These regions were defined using the Desikan-Killiany atlas\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e, and included: frontal (medial OFC, lateral OFC, rACC, pars orbitalis, pars triangularis, pars opercularis, precentral gyrus), insula, temporal (parahippocampal gyrus, inferior temporal gyrus, fusiform gyrus), parietal (inferior parietal cortex) and occipital (lingual gyrus) regions.\u003c/p\u003e \u003cp\u003eAll segmentations were visually inspected by trained researchers blind to participant characteristics. Corrections to the grey matter-white matter surface were made to the majority of images (the most consistent issue was underestimation of temporal pole grey matter), and corrected images were re-run through the analysis pipeline. ROIs from the left and right hemisphere were averaged given there were no hypotheses regarding laterality of effects.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eDiagnostic Interviews\u003c/h2\u003e \u003cp\u003eAt the first four waves, the K-SADS-PL was administered to assess current and past DSM-IV axis I disorders. At wave 5, the Structured Clinical Interview for DSM\u003cem\u003e-\u003c/em\u003e5 was administered, and interviewers probed to assess current and past DSM-IV axis I disorders so that diagnostic criteria were consistent across waves.\u003c/p\u003e \u003cp\u003eDuring the follow-up period (from age 12 to 27), 46 (28.6%, 28 female) of the 161 included participants had a first onset of a depressive disorder (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Thirty-eight had a first onset of MDD, whereas eight had a first onset of another depressive disorder (i.e., persistent depressive disorder, adjustment disorder with depressed mood, depressive disorder not otherwise specified [NOS]). Eighteen of the depressive disorder cases experienced the onset of another disorder prior to their depressive disorder onset (12 experienced an anxiety disorder, four an alcohol or substance use disorder, two a behavioral disorder [i.e., oppositional defiant disorder], and one an eating disorder [eating disorder NOS]). Of the 115 participants who did not experience a depressive disorder during the follow-up period, 32 experienced another form of psychopathology (14 experienced an alcohol or substance use disorder, 10 experienced an anxiety disorder, and 12 experienced a behavioral disorder [e.g., oppositional defiant disorder]). These 32 individuals were excluded from analyses given that their diagnoses were substantially different to the comorbid disorders experienced by the depressive disorder cases (i.e., predominantly alcohol/substance use and behavioral disorders versus predominantly anxiety disorders). Eighty-three participants did not develop any disorder during the follow-up period, and were included as control participants in analyses.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eJoint models\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e, conducted using RStudio version 1.2.5042 and the package JMbayes2\u003csup\u003e24\u003c/sup\u003e, were used to investigate the association between brain structure and the onset of depressive disorder. Joint models allow investigation of the association between longitudinal changes of a continuous variable (sub-model 1) with time-to-event data (sub-model 2, see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). In the current study, joint models allowed modelling the longitudinal brain structure data and simultaneously investigate its association with onset of depressive disorder. Within the joint models, sub-model 1 examined longitudinal changes in brain structure using a linear mixed effects (LME) model with a fixed effect of time, and random intercept and time slope (nonlinear time effects were not estimated due to concerns of overfitting nonlinear slopes with three time points\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e). The time-to-event sub-model 2 was built using the structure of the Cox proportional hazards regression model (where the measure of effect is the hazard rate, or risk of depressive disorder onset at time \u003cem\u003et\u003c/em\u003e, given that depressive disorder onset has not occurred up to time \u003cem\u003et\u003c/em\u003e). The joint modelling framework accounts for right-censored data (e.g., drop-out over time), and allows for imbalanced data. Therefore, we were able to use all MRI data for each participant, whether that be from one, two or three time points.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe joint models were estimated within a Bayesian framework using Markov chain Monte Carlo (MCMC) algorithms\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. In Bayesian estimation, inference is based on the distribution over the parameter space (full posterior distribution) instead of maximum likelihood. The parameter values from the sub-models of the joint model were given as prior information for the full posterior distribution. Three chains were set for the MCMC, leading to three distributions of the possible values for each model parameter. The models provided the posterior mean estimates as hazard ratios (HR) and 95% credible intervals (CI). The Bayesian CIs represent the interval in which the population parameter lies with a given probability.\u003c/p\u003e \u003cp\u003eTwo sets of joint models were run using 1) a \u0026ldquo;current value\u0026rdquo; association structure, whereby onset of depressive disorder at a given point in time \u003cem\u003et\u003c/em\u003e (as determined by interview) is predicted by (interpolated) brain structure at the same time point \u003cem\u003et\u003c/em\u003e, and 2) a slope association structure, whereby onset of depressive disorder at a given point in time \u003cem\u003et\u003c/em\u003e is predicted by the brain structure slope (i.e., age-related change in brain structure) at time \u003cem\u003et\u003c/em\u003e (given only linear time effects were modelled, slope was constant across time). These two sets of joint models were run separately for each brain region (and for cortical regions each brain metric [thickness, surface area]). See Supplementary Information for example model code. Covariates included in all analyses included the time-fixed variables gender (male/female), baseline socioeconomic status (SES) and intracranial volume (ICV; included for subcortical volume and cortical surface area ROIs for \u0026ldquo;current value\u0026rdquo; models only). SES was assessed using a measure of income-to-needs, based on reported family income relative to the relevant Australian poverty line for household size. Variables were checked for univariate outliers, and any values outside the 1st and 99th percentiles were winsorized (using the package RobustHD). Surface area and volume variables were standardized prior to entry into models.\u003c/p\u003e \u003cp\u003eExploratory analyses investigated whether gender moderated associations (note that participants reported on their \u0026ldquo;gender\u0026rdquo; [male/female] and data on biological sex versus gender identity was not collected). For all models, the Gelman\u0026ndash;Rubin diagnostic was used to check the similarity of the obtained distributions for the parameters, and thus assess the convergence of the MCMC sampler. The diagnostic values less than 1.1 were considered as convergence. In addition, density plots and traceplots were visually inspected. Correction for multiple comparisons was performed using a false discovery rate (FDR) of p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 (where corrections are applied within each group of models [thickness value, thickness slope, surface area value, surface area slope, subcortical volume value, subcortical volume slope, thickness value * gender, etc.]).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eSupplementary Analyses\u003c/h2\u003e \u003cp\u003eIn sensitivity analyses, we re-ran significant models from primary analyses, with the inclusion of additional covariates: baseline (self-reported) depressive and anxiety symptoms, and baseline (parent-reported) pubertal status (using the Centre for Epidemiological Studies Depression Scale-Revised\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e, Beck Anxiety Inventory\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e, and Pubertal Development Scale\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e, respectively). Missing data for these variables (ranging from 3\u0026ndash;9%) were imputed using multiple imputation (using the R package \u003cem\u003emice\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e, with 50 imputations).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eAttrition Analysis\u003c/h2\u003e \u003cp\u003eThere were no differences between the final sample (N\u0026thinsp;=\u0026thinsp;161) and both the screening sample (N\u0026thinsp;=\u0026thinsp;2,453) and the selected sample (N\u0026thinsp;=\u0026thinsp;415) on socioeconomic status, affective temperament scores at screening, or gender (p\u0026rsquo;s\u0026thinsp;\u0026gt;\u0026thinsp;0.05).\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eBrain structure predicting concurrent risk of depressive disorder onset (\u0026ldquo;current value\u0026rdquo; association structure)\u003c/h2\u003e \u003cp\u003eAt any given time point, (the current value of) cortical thickness or surface area or subcortical volume was not associated with depressive disorder onset for any ROIs (see Tables S2, S4 and S6 in Supplementary Information).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eAge-related change in brain structure predicting risk of depressive disorder onset (slope association structure)\u003c/h2\u003e \u003cp\u003e \u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eCortical thickness\u003c/span\u003e. Positive age-related change (i.e., relatively increased slope) of thickness in the fusiform gyrus, parahippocampal gyrus, lingual gyrus, and insula cortex was associated with increased risk of depressive disorder onset (see Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eOutput of models where slope of brain ROI was significant in predicting onset of depressive disorder.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePredictor variables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95% CI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003epFDR\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eParahippocampal thickness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3.73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.36,23.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.0003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFusiform thickness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.54,16.96\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.0004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLingual thickness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.40,30.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.013\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInsula thickness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.26,51.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.006\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.024\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAmygdala volume\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.23,20.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.036\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eNote: HR for slopes represent HR for a 1 SD change in random age slope.\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eCI\u0026thinsp;=\u0026thinsp;confidence interval, HR\u0026thinsp;=\u0026thinsp;hazard ratio.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eCortical surface area.\u003c/span\u003e Slope of surface area was not associated with depressive disorder onset for any ROIs (see Table S3).\u003c/p\u003e \u003cp\u003e \u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eSubcortical volume.\u003c/span\u003e Positive age-related change (i.e., relatively increased volume) of the amygdala was associated with increased risk of depressive disorder onset (see Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and S5).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eModerating effects of gender\u003c/h2\u003e \u003cp\u003eThere were no regions where gender significantly moderated the association between current value or slope of structural metric and onset of depressive disorder (see Tables S7-S12).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eSupplementary analyses\u003c/h2\u003e \u003cp\u003eThe addition of baseline depressive and anxiety symptoms and pubertal status did not alter the pattern of significant findings (see Table S13).\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn the current prospective study of neurodevelopmental predictors of depressive disorder onset in youth, age-related change in temporal, insula, and occipital cortical, and subcortical structure were associated with the onset of depressive disorder. Specifically, more positive age-related changes in thickness of temporal, occipital, and insula cortices, and amygdala volume, were associated with increased odds of disorder onset.\u003c/p\u003e \u003cp\u003eNotably, age-related change, but not current value (i.e., value at the time of depression onset), of cortical and subcortical grey matter was associated with the risk of depression onset. The lack of findings for the current value of cortical structure is inconsistent with some other studies of cross-sectional cortical structure in young people prior to depression onset\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. However, some other studies have also found null effects (e.g., \u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e). Our findings suggest that during periods of marked brain development, such as adolescence, alterations to the trajectory of brain development are particularly important in the pathogenesis of depressive disorders.\u003c/p\u003e \u003cp\u003eRelatively \u003cem\u003eincreased\u003c/em\u003e age-related thickening of cortical regions predicting depression onset is consistent with some prior research\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. In addition, several studies that have assessed brain structure in individuals early in the course of their disorder have reported increases in cortical grey matter\u003csup\u003e\u003cspan additionalcitationids=\"CR31\" citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e. As such, in contrast to consistent findings of structural decreases in individuals with established MDD, it may be that \u003cem\u003eincreases\u003c/em\u003e in cortical thickness are present prior to and early in the course of depressive disorder, reflecting a disturbance of normal brain development. Whether these structural changes reflect genetic mechanisms or occur as a result of early experience is unclear, although we note that our prior work has shown exposure to adverse environments is associated with similar patterns of relatively increased age-related change in cortical structure\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOur findings may also be interpreted in relation to theories that propose protracted development of cortical (particularly prefrontal) regions, combined with accelerated development of subcortical regions during adolescence, to be associated with depression risk. Popular theories suggest that these changes may underlie reduced capacity for regulation of behavior and emotion, promoting risk for disorders such as depression\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. Our finding of increased age-related change in amygdala volume predicting depression onset may be partly consistent with these neurodevelopmental theories. This finding was also consistent with our prior investigation of early adolescent brain development changes predicting depression onset by late adolescence (although this prior finding was in females only). Increased age-related change in amygdala volume across adolescence may underlie alterations in emotion reactivity and stress regulation\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e, predisposing adolescents and emerging adults to depression. Further research is needed to test this theory.\u003c/p\u003e \u003cp\u003eFindings of age-related change in temporal, occipital and insula regions associated with risk for depression onset, however, suggest that the aforementioned theories of neurodevelopmental risk for depression may be overly simplistic in their sole focus on prefrontal and subcortical regions\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. Structural changes in the temporal, occipital and insula regions implicated here have been implicated in several prior studies of depression and depression risk\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. All of these regions are implicated in social and emotional processing. The lingual and fusiform gyri are involved in the perceptual processing of facial emotion expressions\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. The parahippocampal gyrus is involved in context appraisal\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. The insula is suggested to be involved in visceral information processing, interoception and subjective emotional experience, including of social emotions such as shame and empathy\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. Alterations in the activity of these regions have been observed in several studies of depression\u003csup\u003e\u003cspan additionalcitationids=\"CR43 CR44\" citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. As such, it is possible that altered developmental trajectories of these regions may underlie altered socio-emotional processing, which may create risk for depression onset\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eCortical thickness, but not surface area changes were implicated in depression onset. The interpretation of this finding is unclear, although speculatively, it could indicate specific processes underlying depression risk. It has been suggested that the development of cortical thickness is dependent on local or intrinsic factors\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e, such as synaptic pruning\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. As such, it may be that alterations in such processes are primarily implicated in the pathogenesis of depression. Further work, however, is needed to understand the mechanisms driving changes in thickness that influence vulnerability to depression.\u003c/p\u003e \u003cp\u003eSeveral limitations of this work should be considered when interpreting findings. First, gender was not found to moderate any associations; however, we may not have had adequate power to detect gender differences. Second, brain imaging was restricted to the adolescent period, and as such, brain development during emerging adulthood was interpolated. Future work is needed with parallel brain imaging and diagnostic assessments across the full developmental period. Third, it is possible that brain structural predictors of depression onset might change across development. While we did not address this possibility, it may be a valuable future direction of this work. Fourth, it is unclear whether the implicated brain changes are relevant to depressive disorder onset specifically, or may be associated with the development of psychopathology more generally. Further research is needed to address this question.\u003c/p\u003e \u003cp\u003eTo conclude, in a fifteen-year longitudinal study of adolescent risk factors for mental disorder onset, we found that relative age-related increases in the thickness of temporal, insula and occipital brain regions, in addition to amygdala volume, were associated with risk for first onset of depressive disorder. Speculatively, these neurodevelopmental patterns may relate to alterations in social and emotional functioning that precede depression onset. Our findings suggest that increases in cortical thickness are present prior to the onset of depressive disorders, reflecting disturbances of normal brain development predisposing to depression. Decreases in structures commonly observed in extant studies of depression might occur subsequent to onset, because of disorder-related factors. Further research is needed to explore these hypotheses.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments.\u003c/strong\u003e This study was funded by the Colonial Foundation, the National Health and Medical Research Council (NHMRC; Australia; Program Grant 350241), and the Australian Research Council (ARC; Discovery Grants DP0878136 and DP109 2637).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of Interest\u003c/strong\u003e: Nicholas Allen has an equity interest in and received salary from Ksana Health Inc. No Ksana Health services or products were used in the current research project. No other authors report financial relationships with commercial interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eRohde P, Lewinsohn PM, Klein DN, Seeley JR, Gau JM. Key characteristics of major depressive disorder occurring in childhood, adolescence, emerging adulthood, and adulthood. \u003cem\u003eClinical Psychological Science\u003c/em\u003e 2013; \u003cstrong\u003e1\u003c/strong\u003e(1)\u003cstrong\u003e: \u003c/strong\u003e41-53.\u003c/li\u003e\n\u003cli\u003eGore FM, Bloem PJ, Patton GC, Ferguson J, Joseph V, Coffey C\u003cem\u003e et al.\u003c/em\u003e Global burden of disease in young people aged 10\u0026ndash;24 years: a systematic analysis. \u003cem\u003eThe Lancet\u003c/em\u003e 2011; \u003cstrong\u003e377\u003c/strong\u003e(9783)\u003cstrong\u003e: \u003c/strong\u003e2093-2102.\u003c/li\u003e\n\u003cli\u003eRakesh D, Allen NB, Whittle S. Balancing act: Neural correlates of affect dysregulation in youth depression and substance use\u0026ndash;A systematic review of functional neuroimaging studies. \u003cem\u003eDevelopmental cognitive neuroscience\u003c/em\u003e 2020; \u003cstrong\u003e42: \u003c/strong\u003e100775.\u003c/li\u003e\n\u003cli\u003eCasey BJ, Getz S, Galvan A. The adolescent brain. \u003cem\u003eDevelopmental review\u003c/em\u003e 2008; \u003cstrong\u003e28\u003c/strong\u003e(1)\u003cstrong\u003e: \u003c/strong\u003e62-77.\u003c/li\u003e\n\u003cli\u003eArnarson E\u0026Ouml;, Matos AP, Salvador C, Ribeiro C, de Sousa B, Edward Craighead W. Longitudinal study of life events, well-being, emotional regulation and depressive symptomatology. \u003cem\u003eJournal of Psychopathology and Behavioral Assessment\u003c/em\u003e 2016; \u003cstrong\u003e38: \u003c/strong\u003e159-171.\u003c/li\u003e\n\u003cli\u003eSchmaal L, Veltman DJ, van Erp TG, S\u0026auml;mann P, Frodl T, Jahanshad N\u003cem\u003e et al.\u003c/em\u003e Subcortical brain alterations in major depressive disorder: findings from the ENIGMA Major Depressive Disorder working group. \u003cem\u003eMolecular psychiatry\u003c/em\u003e 2016; \u003cstrong\u003e21\u003c/strong\u003e(6)\u003cstrong\u003e: \u003c/strong\u003e806-812.\u003c/li\u003e\n\u003cli\u003eSchmaal L, Hibar D, S\u0026auml;mann PG, Hall G, Baune B, Jahanshad N\u003cem\u003e et al.\u003c/em\u003e Cortical abnormalities in adults and adolescents with major depression based on brain scans from 20 cohorts worldwide in the ENIGMA Major Depressive Disorder Working Group. \u003cem\u003eMolecular psychiatry\u003c/em\u003e 2017; \u003cstrong\u003e22\u003c/strong\u003e(6)\u003cstrong\u003e: \u003c/strong\u003e900-909.\u003c/li\u003e\n\u003cli\u003eKim JH, Suh S-i, Lee H-J, Lee J-h, Lee M-S. Cortical and subcortical gray matter alterations in first-episode drug-na\u0026iuml;ve adolescents with major depressive disorder. \u003cem\u003eNeuroreport\u003c/em\u003e 2019; \u003cstrong\u003e30\u003c/strong\u003e(17)\u003cstrong\u003e: \u003c/strong\u003e1172.\u003c/li\u003e\n\u003cli\u003eRamezani M, Johnsrude I, Rasoulian A, Bosma R, Tong R, Hollenstein T\u003cem\u003e et al.\u003c/em\u003e Temporal-lobe morphology differs between healthy adolescents and those with early-onset of depression. \u003cem\u003eNeuroImage: Clinical\u003c/em\u003e 2014; \u003cstrong\u003e6: \u003c/strong\u003e145-155.\u003c/li\u003e\n\u003cli\u003eNickson T, Chan S, Papmeyer M, Romaniuk L, Macdonald A, Stewart T\u003cem\u003e et al.\u003c/em\u003e Prospective longitudinal voxel-based morphometry study of major depressive disorder in young individuals at high familial risk. \u003cem\u003ePsychological medicine\u003c/em\u003e 2016; \u003cstrong\u003e46\u003c/strong\u003e(11)\u003cstrong\u003e: \u003c/strong\u003e2351-2361.\u003c/li\u003e\n\u003cli\u003eWhittle S, Lichter R, Dennison M, Vijayakumar N, Schwartz O, Byrne ML\u003cem\u003e et al.\u003c/em\u003e Structural brain development and depression onset during adolescence: a prospective longitudinal study. \u003cem\u003eAmerican Journal of Psychiatry\u003c/em\u003e 2014; \u003cstrong\u003e171\u003c/strong\u003e(5)\u003cstrong\u003e: \u003c/strong\u003e564-571.\u003c/li\u003e\n\u003cli\u003ePapmeyer M, Giles S, Sussmann JE, Kielty S, Stewart T, Lawrie SM\u003cem\u003e et al.\u003c/em\u003e Cortical thickness in individuals at high familial risk of mood disorders as they develop major depressive disorder. \u003cem\u003eBiological psychiatry\u003c/em\u003e 2015; \u003cstrong\u003e78\u003c/strong\u003e(1)\u003cstrong\u003e: \u003c/strong\u003e58-66.\u003c/li\u003e\n\u003cli\u003eShapero BG, Chai XJ, Vangel M, Biederman J, Hoover CS, Whitfield-Gabrieli S\u003cem\u003e et al.\u003c/em\u003e Neural markers of depression risk predict the onset of depression. \u003cem\u003ePsychiatry Research: Neuroimaging\u003c/em\u003e 2019; \u003cstrong\u003e285: \u003c/strong\u003e31-39.\u003c/li\u003e\n\u003cli\u003eLittle K, Olsson CA, Youssef GJ, Whittle S, Simmons JG, Y\u0026uuml;cel M\u003cem\u003e et al.\u003c/em\u003e Linking the serotonin transporter gene, family environments, hippocampal volume and depression onset: A prospective imaging gene\u0026times; environment analysis. \u003cem\u003eJournal of abnormal psychology\u003c/em\u003e 2015; \u003cstrong\u003e124\u003c/strong\u003e(4)\u003cstrong\u003e: \u003c/strong\u003e834.\u003c/li\u003e\n\u003cli\u003eVulser H, Lemaitre H, Artiges E, Miranda R, Penttil\u0026auml; J, Struve M\u003cem\u003e et al.\u003c/em\u003e Subthreshold depression and regional brain volumes in young community adolescents. \u003cem\u003eJournal of the American Academy of Child \u0026amp; Adolescent Psychiatry\u003c/em\u003e 2015; \u003cstrong\u003e54\u003c/strong\u003e(10)\u003cstrong\u003e: \u003c/strong\u003e832-840.\u003c/li\u003e\n\u003cli\u003eLittle K, Olsson CA, Whittle S, Youssef G, Byrne ML, Simmons JG\u003cem\u003e et al.\u003c/em\u003e Association between serotonin transporter genotype, brain structure and adolescent-onset major depressive disorder: a longitudinal prospective study. \u003cem\u003eTranslational Psychiatry\u003c/em\u003e 2014; \u003cstrong\u003e4\u003c/strong\u003e(9)\u003cstrong\u003e: \u003c/strong\u003ee445-e445.\u003c/li\u003e\n\u003cli\u003eFoland-Ross LC, Sacchet MD, Prasad G, Gilbert B, Thompson PM, Gotlib IH. Cortical thickness predicts the first onset of major depression in adolescence. \u003cem\u003eInternational Journal of Developmental Neuroscience\u003c/em\u003e 2015; \u003cstrong\u003e46: \u003c/strong\u003e125-131.\u003c/li\u003e\n\u003cli\u003eRakesh D, Whittle S, Sheridan MA, McLaughlin KA. Childhood socioeconomic status and the pace of structural neurodevelopment: accelerated, delayed, or simply different? \u003cem\u003eTrends in Cognitive Sciences\u003c/em\u003e 2023.\u003c/li\u003e\n\u003cli\u003eRevision of the early adolescent temperament questionnaire. \u003cem\u003eProceedings of the Poster presented at the 2001 biennial meeting of the society for research in child development, Minneapolis, Minnesota\u003c/em\u003e2001. Citeseer.\u003c/li\u003e\n\u003cli\u003eKaufman J, Birmaher B, Brent D, Rao U, Flynn C, Moreci P\u003cem\u003e et al.\u003c/em\u003e Schedule for affective disorders and schizophrenia for school-age children-present and lifetime version (K-SADS-PL): initial reliability and validity data. \u003cem\u003eJournal of the American Academy of Child \u0026amp; Adolescent Psychiatry\u003c/em\u003e 1997; \u003cstrong\u003e36\u003c/strong\u003e(7)\u003cstrong\u003e: \u003c/strong\u003e980-988.\u003c/li\u003e\n\u003cli\u003eVijayakumar N, Allen NB, Youssef G, Dennison M, Y\u0026uuml;cel M, Simmons JG\u003cem\u003e et al.\u003c/em\u003e Brain development during adolescence: A mixed‐longitudinal investigation of cortical thickness, surface area, and volume. \u003cem\u003eHuman brain mapping\u003c/em\u003e 2016; \u003cstrong\u003e37\u003c/strong\u003e(6)\u003cstrong\u003e: \u003c/strong\u003e2027-2038.\u003c/li\u003e\n\u003cli\u003eDennison M, Whittle S, Y\u0026uuml;cel M, Vijayakumar N, Kline A, Simmons J\u003cem\u003e et al.\u003c/em\u003e Mapping subcortical brain maturation during adolescence: evidence of hemisphere‐and sex‐specific longitudinal changes. \u003cem\u003eDevelopmental science\u003c/em\u003e 2013; \u003cstrong\u003e16\u003c/strong\u003e(5)\u003cstrong\u003e: \u003c/strong\u003e772-791.\u003c/li\u003e\n\u003cli\u003eDesikan RS, S\u0026eacute;gonne F, Fischl B, Quinn BT, Dickerson BC, Blacker D\u003cem\u003e et al.\u003c/em\u003e An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. \u003cem\u003eNeuroimage\u003c/em\u003e 2006; \u003cstrong\u003e31\u003c/strong\u003e(3)\u003cstrong\u003e: \u003c/strong\u003e968-980.\u003c/li\u003e\n\u003cli\u003eRizopoulos D, Papageorgiou G, Miranda Afonso P. JMbayes2: extended joint models for longitudinal and time-to-event data. \u003cem\u003eR package version 03-0, ed\u003c/em\u003e 2022.\u003c/li\u003e\n\u003cli\u003eKing KM, Littlefield AK, McCabe CJ, Mills KL, Flournoy J, Chassin L. Longitudinal modeling in developmental neuroimaging research: Common challenges, and solutions from developmental psychology. \u003cem\u003eDevelopmental cognitive neuroscience\u003c/em\u003e 2018; \u003cstrong\u003e33: \u003c/strong\u003e54-72.\u003c/li\u003e\n\u003cli\u003eEaton WW, Muntaner C, Smith C, Tien A, Ybarra M. Center for epidemiologic studies depression scale: Review and revision. \u003cem\u003eThe use of psychological testing for treatment planning and outcomes assessment\u003c/em\u003e 2004\u003cstrong\u003e: \u003c/strong\u003ep363-p377.\u003c/li\u003e\n\u003cli\u003eBeck AT, Epstein N, Brown G, Steer R. Beck anxiety inventory. \u003cem\u003eJournal of consulting and clinical psychology\u003c/em\u003e 1993.\u003c/li\u003e\n\u003cli\u003eBrooks-Gunn J, Warren MP, Rosso J, Gargiulo J. Validity of self-report measures of girls\u0026apos; pubertal status. \u003cem\u003eChild development\u003c/em\u003e 1987\u003cstrong\u003e: \u003c/strong\u003e829-841.\u003c/li\u003e\n\u003cli\u003evan Buuren S, Groothuis-Oudshoorn K, Robitzsch A, Vink G, Doove L, Jolani S. Package \u0026lsquo;mice\u0026rsquo;. \u003cem\u003eComputer software\u003c/em\u003e 2015; \u003cstrong\u003e20\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eQiu L, Lui S, Kuang W, Huang X, Li J, Zhang J\u003cem\u003e et al.\u003c/em\u003e Regional increases of cortical thickness in untreated, first-episode major depressive disorder. \u003cem\u003eTranslational psychiatry\u003c/em\u003e 2014; \u003cstrong\u003e4\u003c/strong\u003e(4)\u003cstrong\u003e: \u003c/strong\u003ee378-e378.\u003c/li\u003e\n\u003cli\u003eReynolds S, Carrey N, Jaworska N, Langevin LM, Yang X-R, MacMaster FP. Cortical thickness in youth with major depressive disorder. \u003cem\u003eBMC psychiatry\u003c/em\u003e 2014; \u003cstrong\u003e14\u003c/strong\u003e(1)\u003cstrong\u003e: \u003c/strong\u003e1-8.\u003c/li\u003e\n\u003cli\u003evan Eijndhoven P, van Wingen G, Katzenbauer M, Groen W, Tepest R, Fern\u0026aacute;ndez G\u003cem\u003e et al.\u003c/em\u003e Paralimbic cortical thickness in first-episode depression: evidence for trait-related differences in mood regulation. \u003cem\u003eAmerican Journal of Psychiatry\u003c/em\u003e 2013; \u003cstrong\u003e170\u003c/strong\u003e(12)\u003cstrong\u003e: \u003c/strong\u003e1477-1486.\u003c/li\u003e\n\u003cli\u003eWhittle S, Simmons JG, Dennison M, Vijayakumar N, Schwartz O, Yap MB\u003cem\u003e et al.\u003c/em\u003e Positive parenting predicts the development of adolescent brain structure: a longitudinal study. \u003cem\u003eDevelopmental cognitive neuroscience\u003c/em\u003e 2014; \u003cstrong\u003e8: \u003c/strong\u003e7-17.\u003c/li\u003e\n\u003cli\u003eSomerville LH, Jones RM, Casey B. A time of change: behavioral and neural correlates of adolescent sensitivity to appetitive and aversive environmental cues. \u003cem\u003eBrain and cognition\u003c/em\u003e 2010; \u003cstrong\u003e72\u003c/strong\u003e(1)\u003cstrong\u003e: \u003c/strong\u003e124-133.\u003c/li\u003e\n\u003cli\u003eCasey B, Jones RM, Somerville LH. Braking and accelerating of the adolescent brain. \u003cem\u003eJournal of Research on Adolescence\u003c/em\u003e 2011; \u003cstrong\u003e21\u003c/strong\u003e(1)\u003cstrong\u003e: \u003c/strong\u003e21-33.\u003c/li\u003e\n\u003cli\u003eDedovic K, Duchesne A, Andrews J, Engert V, Pruessner JC. The brain and the stress axis: the neural correlates of cortisol regulation in response to stress. \u003cem\u003eNeuroimage\u003c/em\u003e 2009; \u003cstrong\u003e47\u003c/strong\u003e(3)\u003cstrong\u003e: \u003c/strong\u003e864-871.\u003c/li\u003e\n\u003cli\u003ePhelps EA. Human emotion and memory: interactions of the amygdala and hippocampal complex. \u003cem\u003eCurrent opinion in neurobiology\u003c/em\u003e 2004; \u003cstrong\u003e14\u003c/strong\u003e(2)\u003cstrong\u003e: \u003c/strong\u003e198-202.\u003c/li\u003e\n\u003cli\u003eKesler ML, Andersen AH, Smith CD, Avison MJ, Davis CE, Kryscio RJ\u003cem\u003e et al.\u003c/em\u003e Neural substrates of facial emotion processing using fMRI. \u003cem\u003eCognitive Brain Research\u003c/em\u003e 2001; \u003cstrong\u003e11\u003c/strong\u003e(2)\u003cstrong\u003e: \u003c/strong\u003e213-226.\u003c/li\u003e\n\u003cli\u003eRudy JW. Context representations, context functions, and the parahippocampal\u0026ndash;hippocampal system. \u003cem\u003eLearning \u0026amp; memory\u003c/em\u003e 2009; \u003cstrong\u003e16\u003c/strong\u003e(10)\u003cstrong\u003e: \u003c/strong\u003e573-585.\u003c/li\u003e\n\u003cli\u003eUddin LQ, Nomi JS, H\u0026eacute;bert-Seropian B, Ghaziri J, Boucher O. Structure and function of the human insula. \u003cem\u003eJournal of clinical neurophysiology: official publication of the American Electroencephalographic Society\u003c/em\u003e 2017; \u003cstrong\u003e34\u003c/strong\u003e(4)\u003cstrong\u003e: \u003c/strong\u003e300.\u003c/li\u003e\n\u003cli\u003ePulcu E, Lythe K, Elliott R, Green S, Moll J, Deakin JF\u003cem\u003e et al.\u003c/em\u003e Increased amygdala response to shame in remitted major depressive disorder. \u003cem\u003ePloS one\u003c/em\u003e 2014; \u003cstrong\u003e9\u003c/strong\u003e(1)\u003cstrong\u003e: \u003c/strong\u003ee86900.\u003c/li\u003e\n\u003cli\u003eSuh JS, Minuzzi L, Cudney LE, Maich W, Eltayebani M, Soares CN\u003cem\u003e et al.\u003c/em\u003e Cerebral cortical thickness after treatment with desvenlafaxine succinate in major depressive disorder. \u003cem\u003eNeuroReport\u003c/em\u003e 2019; \u003cstrong\u003e30\u003c/strong\u003e(5)\u003cstrong\u003e: \u003c/strong\u003e378-382.\u003c/li\u003e\n\u003cli\u003eLee JS, Kang W, Kang Y, Kim A, Han K-M, Tae W-S\u003cem\u003e et al.\u003c/em\u003e Alterations in the occipital cortex of drug-na\u0026iuml;ve adults with major depressive disorder: A surface-based analysis of surface area and cortical thickness. \u003cem\u003ePsychiatry Investigation\u003c/em\u003e 2021; \u003cstrong\u003e18\u003c/strong\u003e(10)\u003cstrong\u003e: \u003c/strong\u003e1025.\u003c/li\u003e\n\u003cli\u003eLi X, Wang J. Abnormal neural activities in adults and youths with major depressive disorder during emotional processing: a meta-analysis. \u003cem\u003eBrain imaging and Behavior\u003c/em\u003e 2021; \u003cstrong\u003e15: \u003c/strong\u003e1134-1154.\u003c/li\u003e\n\u003cli\u003eDelvecchio G, Fossati P, Boyer P, Brambilla P, Falkai P, Gruber O\u003cem\u003e et al.\u003c/em\u003e Common and distinct neural correlates of emotional processing in bipolar disorder and major depressive disorder: a voxel-based meta-analysis of functional magnetic resonance imaging studies. \u003cem\u003eEuropean Neuropsychopharmacology\u003c/em\u003e 2012; \u003cstrong\u003e22\u003c/strong\u003e(2)\u003cstrong\u003e: \u003c/strong\u003e100-113.\u003c/li\u003e\n\u003cli\u003eAllen NB, Badcock PB. The social risk hypothesis of depressed mood: evolutionary, psychosocial, and neurobiological perspectives. \u003cem\u003ePsychological bulletin\u003c/em\u003e 2003; \u003cstrong\u003e129\u003c/strong\u003e(6)\u003cstrong\u003e: \u003c/strong\u003e887.\u003c/li\u003e\n\u003cli\u003eWierenga LM, Langen M, Oranje B, Durston S. Unique developmental trajectories of cortical thickness and surface area. \u003cem\u003eNeuroimage\u003c/em\u003e 2014; \u003cstrong\u003e87: \u003c/strong\u003e120-126.\u003c/li\u003e\n\u003cli\u003eJeon T, Mishra V, Ouyang M, Chen M, Huang H. Synchronous changes of cortical thickness and corresponding white matter microstructure during brain development accessed by diffusion MRI tractography from parcellated cortex. \u003cem\u003eFrontiers in Neuroanatomy\u003c/em\u003e 2015; \u003cstrong\u003e9: \u003c/strong\u003e158.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Neurodevelopment, adolescence, emerging adulthood, depression, neuroimaging, cortical thickness","lastPublishedDoi":"10.21203/rs.3.rs-4267037/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4267037/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eBrain structural alterations are consistently reported in depressive disorders, yet it remains unclear whether these alterations reflect a pre-existing vulnerability or are the result of psychopathology. We aimed to investigate prospective adolescent neurodevelopmental risk markers for depressive disorder onset, using data from a fifteen-year longitudinal study.\u003cstrong\u003e \u003c/strong\u003eA risk-enriched community sample of 161 adolescents who had no history of depressive disorders participated in neuroimaging assessments conducted during early (age 12), mid (age 16) and late adolescence (age 19). Onsets of depressive disorders were assessed for the period spanning early adolescence through emerging adulthood (post-baseline, ages 12 to 27). Forty-six participants (28 female) experienced a first episode of a depressive disorder during the follow-up period; eighty-three participants (36 female) received no mental disorder diagnosis. Joint modelling was used to investigate whether brain structure (subcortical volume, cortical thickness and surface area) or age-related changes in brain structure were associated with the risk of depressive disorder onset. Analyses revealed that age-related increases in a) amygdala volume (hazard ratio [HR] 3.01, p\u003csub\u003eFDR\u003c/sub\u003e 0.036), and b) thickness of temporal (parahippocampal [HR 3.73, p 0.004] and fusiform gyri [HR 4.14, p 0.003]), insula (HR 4.49, p 0.024) and occipital (lingual gyrus, HR 4.19, p 0.013) regions were associated with the onset of depressive disorder. Findings suggest that relative increases in amygdala volume and temporal, insula, and occipital cortical thickness across adolescence may reflect disturbances of normative brain development, predisposing some individuals to depression. This raises the possibility that prior findings of grey matter decreases in clinically depressed individuals may instead reflect alterations that are caused by disorder-related factors.\u003c/p\u003e","manuscriptTitle":"Prospective associations between structural brain development and onset of depressive disorder during adolescence and emerging adulthood","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-04-26 01:47:01","doi":"10.21203/rs.3.rs-4267037/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"883e0f5d-56ec-4d49-a0c5-6689d61f18fb","owner":[],"postedDate":"April 26th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":30788798,"name":"Biological sciences/Neuroscience"},{"id":30788799,"name":"Health sciences/Biomarkers/Predictive markers"}],"tags":[],"updatedAt":"2024-06-19T15:28:14+00:00","versionOfRecord":[],"versionCreatedAt":"2024-04-26 01:47:01","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4267037","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4267037","identity":"rs-4267037","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}