Prediction of treatment response in depressive and anxious adolescents and development of an integrated depression-anxiety model = 우울·불안 청소년의 항우울제 치료반응 예측과 우울-불안 통합 모형 구축

국문 초록 (Abstract)

서론: 청소년기 우울증은 높은 유병률을 지닌 질환으로 청소년의 일상생활 영역에 다양한 부정적 영향을 미친다. 청소년기 우울증의 치료에는 가장 효과적인 치료로 알려진 우울증 인지행동치료와 항우울제를 병용한 치료법으로도 청소년기 우울증 관해율이 60%에 지나지 않는다. 따라서 우울증 치료효과를 높이는 것이 필수적이며, 이러한 과정에는 자주 동반 이환되는 불안장애에 대한 고려가 중요하다. 우울증과 불안장애가 동반 이환되는 경우에는 치료 반응률이 낮아지고 질병의 고통 기간과 재발률이 증가된다. 현재 정신과적 진단체계로는 우울장애와 불안장애는 다른 질병으로 분류된다. 하지만 실제 임상 현장에서 우울 및 불안 증상은 종종 다양한 중증도로 함께 공존한다. 따라서 본 연구에서는 우울과 불안 청소년의 다양한 치료 이후의 종단적 결과를 분류하기 위해 우울증상과 불안증상을 함께 고려하고자 한다. 치료에 따른 다른 종단적 궤적을 가진 그룹을 분류하고, 관심 영역으로 편도체와 해마를 사용하여 그룹별 신경회로의 차이점을 조사하는 것을 목표로 한다. 또한 머신 러닝 접근 방식을 사용하여 치료 반응을 판별 및 예측하고 결과적으로는 우울증과 불안의 통합모형을 제시하고자 한다.
방법: 12세에서 17세 사이의 우울증 환아 95명(14.8 ± 1.6세, 여성 65명)이 연구에 포함되었다. 이들은 우울증이 있는 청소년의 항우울제 반응과 자살률의 바이오마커를 개발하는 연구의 일부에 해당한다. 모든 연구 참여자는 8주간 항우울제(에시탈로프람) 치료를 받았고, 우울증상은 Children’s Depression Inventory, Beck Depression Scale, Children’s Depression Rating Scale-Revised로, 불안증상은 Screen for Child Anxiety Related Emotional Disorders 척도로 평가하였다. 뇌 영상 데이터는 0주차와 8주차 방문에 수집하였다. 우울과 불안을 종속변수로 사용하여 다양한 궤적을 가진 하위집단을 분류하기 위해 다변량 잠재계층분석을 수행하였다. 공변량은 FACES와 각 방문의 상호작용, ETI 및 PVS으로 설정하였다. 이후 하위 그룹 간의 뇌 용적, resting state functional connectivity (RSFC)와 항우울제의 종단 치료 효과를 비교했다.
결과: 종단적 궤적분석으로 3개 그룹 모델이 본 연구 데이터에 가장 적합한 것으로 확인했다. 8주 간의 항우울제 치료에서 그룹 1(연구 참여자 중 31.03%)의 우울과 불안 증상이 빠르게 감소하는 경향을 보였고 (Fast-decreasing group, FDG), 그룹 2 (32.18%)는 변화가 없었으며 (No-change group, NCG), 그룹 3 (36.78%)은 서서히 감소하는 경향을 보였다 (Slow-decreasing group, SDG). 본 연구의 region of interest (ROI)인 편도체와 해마의 치료 전 부피 비교에서는 그룹간 차이가 없었고, 치료전후 부피 비교에서도 시간에 따른 3개의 하위 그룹간 유의미한 차이는 없었다.
Whole-brain analysis에서는 3개의 그룹간 종단적 치료효과 차이 분석에서는 그룹 1에서 치료 전후 ROI와 RSFC의 유의미한 차이를 보이는 cluster를 도출해냈다. 이는 오른쪽 편도체와 보조 운동 영역(supplementary motor area, SMA) 간의 연결 그리고 좌측 postcentral gyrus과 rolandic operculum(ROL)으로 확장된 cluster를 포함한다. 해당 RSFC는 그룹 1에서는 치료 후 감소하는 형태를 보였다. 왼쪽 편도체와는 왼쪽 postcentral gyrus의 RSFC에서 유의미한 결과를 보였고, 오른쪽 해마와는 SMA의 RSFC, 왼쪽 해마와는 오른쪽 lingual gyrus의 RSFC가 유의미했다. 그룹 1은 치료 전과 치료 후 사이에 RSFC의 감소를 보인 반면, 그룹 2와 3은 해당 영역에서 치료 전과 후 사이에 RSFC 차이를 보이지 않았다. 더 나아가, 본 연구에서는 그룹 1과 3를 비교했을 때 치료 전후 다른 종단적 치료 효과를 보이는 cluster를 도출해냈다. 해당 cluster는 우측 편도체와 우측 post/pre-central gyrus, 양측 SMA, 우측 postcentral gyrus, 우측 calcarine gyrus와의 연결을 포함했으며, 왼쪽 편도체와 우측 middle frontal gyrus, 좌측 inferior parietal lobule, 우측 postcentral gyrus의 연결을 포함했다. 우측 해마와는 양측 SMA, 우측 postcentral gyrus의 연결이 유의미했다. 더 나아가 functional ROI를 사용한 ROI 분석을 진행하였고 치료 전 그룹 1과 그룹 3의 baseline RSFC를 비교하였을 때, 우측 편도체와의 연결성에서는 유의미한 종단적 치료 효과를 보이는 5개의 영역이 도출되었다. 이는 양측 postcentral gyrus, 우측 precentral gyrus, 양측 calcarine gyrus에 해당한다. 해당 connectivity는 모두 유의미한 그룹 (그룹 1, 2, 3) x 시간 (0주차, 8주차 방문) 상호작용을 보였고, 그룹 1의 RSFC는 치료 후에 감소하는 패턴을 보였다.
결론: 본 논문은 청소년기의 우울증과 불안 증상을 함께 고려하여 신경회로에서 종단적 치료 효과를 평가한 최초의 연구로서 중요한 임상적 의미를 갖는다. 본 논문에서는 FDG에서 치료전 시각, 청각 및 체성 감각 기능과 관련된 cluster와 오른쪽 편도체의 RSFC이 증가되어 있으며, 치료 후에는 해당 RSFC가 감소되는 것을 보여주었다. 해당 결과는 우울/불안 청소년의 특징인 외부 자극에 대한 과각성과 과민성을 SSRI 약물로 감소시킬 수 있음을 시사한다. 또한 FDG에서 초반에 감소되어 있던 왼쪽 편도체와 집행 기능 및 주의력에 관여하는 영역이 있는 RSFC가 치료 후에는 증가하는 결과를 살펴볼 수 있었다. 이는 왼쪽 편도체에 대한 top-down process는 강화되어 우울하고 불안한 청소년의 인지 기능이 향상된 것을 보여준다고 이해할 수 있다. 본 연구에서 밝힌 결과를 일반화하기 위해서는 더 많은 환아를 대상으로 하여 더 긴 추적 기간 동안 대상을 평가한 추가 연구가 필요하다.

다국어 초록 (Multilingual Abstract)

Introduction: Adolescent depression is a problematic disease with high prevalence, affecting the daily life of adolescents. Notably, cognitive-behavioral therapy and antidepressant medication have a remission rate of approximately 60% in the treatment of adolescent depression, which is a severe drawback. Therefore, to improve the efficacy of depression treatment, it is crucial to take the co-morbid anxiety disorder into account. Co-morbid depressive and anxiety disorder lowers the treatment response rate and increases the disease suffering period and recurrence rate. Depressive disorder and anxiety disorder are classified as symptomatically different diseases by the current diagnostic system. However, depressive and anxiety symptoms often co-exist with various severity in a clinical setting. Therefore, we considered depression and anxiety to categorize the different longitudinal outcomes of depressive and anxious adolescents. Based on classifying subgroups with different trajectories, I sought to examine the differences in neural circuits by using the amygdala and hippocampus as my regions of interest (ROI). Furthermore, using a machine learning approach, we aimed to discriminate and predict the treatment response and eventually establish the relationship between depression and anxiety in adolescent depression.
Method: A total of 95 depressed participants aged 12 to 17 (14.8 ± 1.6 years, 65 girls) were included in the study. They were a part of the study developing the biomarkers of the antidepressant response and suicidality in adolescents with depression. All participants were treated with escitalopram in an 8-week open-label trial. The depressive and anxiety symptoms were measured by the Children’s Depression Inventory, Beck Depression Inventory, Children’s Depression Rating Scale-Revised, and Screen for Child Anxiety Related Emotional Disorders. Brain imaging data were collected at baseline (0-week) and 8-week. First, multivariate latent class analysis was performed to classify subgroups with different trajectories using depression and anxiety as the dependent variable. Covariates were the interaction of family cohesion and adaptability, childhood trauma, peer-victimization, and visits. Then, we compared the brain volume, baseline functional connectivity, and longitudinal treatment effect among the subgroups. Furthermore, using a machine learning approach, we integrated the socio-demographic, clinical data, resting-state functional connectivity (RSFC), and brain volume data of the depression and anxiety participants to predict the healthy control vs. depressive and different treatment trajectories.
Results: Trajectory analysis revealed that a 3-class model fitted my data best. Depression and anxiety of Subgroup 1 (31.0% of participants) showed a fast-decreasing tendency (fast-decreasing group), while Subgroup 2 (32.2% of participants) showed no change (no-change group) and Subgroup 3 (36.8% of participants) with a slow decreasing pattern (slow-decreasing group). A baseline volume comparison of my ROI showed no difference between the three subgroups. Longitudinal volume comparison showed no significant group (three subgroups) x time (Visit 1 and Visit 6) interactions. In the region of interest (ROI) analysis using functional ROI, resting-state function connectivity comparison between Subgroup 1 and Subgroup 3 at baseline was conducted. Five regions connected with amygdala showed a significant longitudinal treatment effect on RSFC changes in Subgroup 1: Both postcentral gyrus, right precentral gyrus, and both calcarine gyrus. In the whole-brain analysis (seed to voxel-wise functional connectivity), we investigated the clusters showing significant longitudinal effects of treatment on RSFC in the subgroup 1. We identified the connection between the right amygdala and (1) supplementary motor area (SMA), (2) left postcentral gyrus, and (3) rolandic operculum (ROL) was significant. The left amygdala showed significant RSFC with (1) left post and (2) precentral gyrus, right hippocampus with (1) both SMA, and left hippocampus with (1) right lingual gyrus. Subgroup 1 showed a decrease in RSFC between pre-treatment and post-treatment in the connectivity in my findings whereas Subgroups 2 and 3 did not show RSFC differences between pre- and post-treatment in the clusters. Furthermore, clusters showing the different longitudinal effects of treatment on RSFC between Subgroups 1 and 3 were calculated. The regions that had a significant correlation with the right amygdala were the (1) right post/precentral gyrus, (2) both SMA, (3) left post and precentral gyrus, (4) right postcentral gyrus extending to the superior parietal lobule, and (5) right calcarine gyrus. RSFC between the left amygdala, (1) right middle frontal gyrus extending to superior frontal gyrus, (2) left inferior parietal lobule, and (3) right postcentral gyrus was significant. The right hippocampus presented significant correlation with (1) right postcentral gyrus, and (2) both SMA. Subgroup 1 showed decreasing pattern of RSFC in the right amygdala connectivity with the regions identified, whereas, the connectivities of left amygdala had increasing trend of RSFC.
Using machine learning method, we found that the clinical data predicted the healthy control vs. patients with best accuracy (micro-average F1=1). In predicting SSRI treatment response, combining clinical data and RSFC presented the highest accuracy (micro-average F1=1), followed by combining clinical, RSFC, and volume model (micro-average F1=0.9).
Conclusion: My research is the first to evaluate the longitudinal antidepressant treatment effects on neural circuits in adolescent depression by considering depression and anxiety symptoms. My results showed that the fast-decreasing group had increased functional connectivity in the right amygdala with the regions of visual, auditory, and somatosensory function at baseline and the functional connectivity decreased after the SSRI treatment. The results suggest that the hypervigilance and hypersensitivity to external stimuli, the characteristics of depressive and anxious youth, may be decreased with SSRI medication. Additionally, increased RSFC with the left amygdala and regions involved in executive function and attention may indicate strengthened top-down regulation on the left amygdala, which, in turn, may have improved the cognitive function in depressive and anxious adolescents. Further studies are needed with a larger sample size and a more extended follow-up period to generalize my findings of elucidated neural circuits of depression and anxiety.

목차 (Table of Contents)

• I. Introduction １
• 1. Depression and Anxiety Disorder in Adolescents １
• 2. Current Depression and Anxiety Disorder Diagnosis System and Limitations ２
• 3. Multivariate latent class analysis to classify depressive and anxious youth ４
• 4. Neural Circuits in Adolescents with Depression and Anxiety disorder ５
• 5. Neural Circuit Changes after Depression and Anxiety Treatments ６
• 6. Machine learning approaches to discriminate and predict the treatment response ８
• II. Research Questions & Hypothesis ９
• 1. Predicting treatment response in Adolescents with Depression and Anxiety disorder ９
• 2. Neural Circuits Related to Treatment Response in Adolescents with Depression and Anxiety Disorder １０
• 3. Machine learning modeling to predict treatment response １１
• 4. Hypotheses １１
• III. Methods １２
• 1. Participants １２
• 2. Procedures １３
• 3. Measures １４
• 4. Data Analyses １９
• IV. Results ２２
• 1. Trajectory analysis, demographic data ２２
• 2. MRI analyses ２３
• 3. Machine learning classification ２６
• V. Discussion ２６
• 1. Summary of the results ２６
• 2. Subgroup characteristics ２７
• 3. Brain regions and functional connectivity in subgroups ２８
• 4. Machine learning predicting (1) HC vs. patients and (2) Subgroup 1 vs. 2 vs. 3 ３４
• 5. Clinical implications ３５
• 6. Limitations and Further Studies ３６
• VI. Conclusions ３６
• Bibliography ３８
• 국문 초록 ５１

참고문헌 (Reference)

1. Machine learning in medicine, Rajkomar A , Dean J , Kohane I, 380 ( 14 ) :1347 58, , 2019

2. Emotion circuits in the brain, LeDoux , J. E., 23 ( 1 ) :155 84 ., , 2000

3. Depressed adolescents grown up, Weissman MM , Wolk S , Goldstein RB , Moreau D , Adams P , Greenwald S , et al, 281 ( 18 ) :1707 13 ., , 1999

4. The constructio n system of the brain, Hassabis D , Maguire EA ., 364 ( 1521 ) :1263 71 ., , 2009

5. Follow up and family study of anxious depression, Clayton PJ , Grove WM , Coryell W , Keller M , Hirschfeld R , Fawcett J, 148 ( 11 ) :1512 7, , 1991

6. Development and validation of the penn state worry questionnai re, Meyer TJ , Miller ML , Metzger RL , Borkovec TD, 28 ( 6 ) :487 95, , 1990

7. B rain gray matter deficits in patients with chronic primary insomnia, Joo EY , Noh HJ , Kim J S , Koo DL , Kim D , Hwang KJ , et al, 36 ( 7 ) :999 1007, , 2013

8. Normative and reliability data for the Children 's Depression Inventory, Smucker MR , Craighead WE , Craighead LW , Green BJ, 14 ( 1 ) :25 39 ., , 1986

9. A national survey on the prevalence and risk factors of child sexual abuse, Han I Y , Lee Y W , Yoo S K , Park M S , Kim J S., 60 ( 2 ) :131 53 ., , 2008

10. Attentional bias in generalized anxiety disorder versus depressive disorder, Mogg , K. , & Bradley , B. P., 29 ( 1 ) :29 45, , 2005

11. Irritable mood and the diagnostic and statistical manual of mental disorders, Safer DJ, 3 ( 1 ) :1 4 ., , 2009

12. Dynamic modulation of amygdala hippocampal connectivity by emotional arousal ., Fastenrath M , Coynel D , Spalek K , Milnik A , Gschwind L , Roozendaal B , et al ., 34 ( 42 ) :13935 47 ., , 2014

13. Prevalence and develo pment of psychiatric disorders in childhood and adolescence, Costello EJ , Mustillo S , Erkanli A , Keeler G , Angold A, 60 ( 8 ) :837 44, , 2003

14. Arabic version of the Children 's Depressi on Inventory : Reliability and validity, Ghareeb GA , Beshai J., 18 ( 4 ) :323 6, , 1989

15. From th reat to fear : the neural organization of defensive fear systems in humans, Mobbs D , Marchant JL , Hassabis D , Seymour B , Tan G , Gray M , et al, 29 ( 39 ) :12236 43 ., , 2009

16. Comorbidity of anxiety and depression in children and a dolescents : 20 years after, Cummings CM , Caporino NE , Kendall PC, 140 ( 3 ) :816, , 2014

17. Anxiety , depression , impulsivity , and brain structure in children and adolescents, Merz EC , He X , Noble KG, 20:243 51, , 2018

18. Correspondence betwee n adolescent report and parent report of psychiatric diagnostic data, Cantwell DP , Lewinsohn PM , Rohde P , Seeley JR, 36 ( 5 ) :610 9, , 1997

19. Brain activity in adolescent major depressive disorder before and after fluoxetine treatment, Tao R , Calley CS , Hart J , Mayes TL , Nakonezn y PA , Lu H , et al, 169 ( 4 ) :381 8, , 2012

20. The effects of adverse childhood expe riences on internalizing versus externalizing outcomes, Muniz CN , Fox B , Miley LN , Delisi M , Cigarran GP , Birnbaum A ., 46 ( 4 ) :568 89, , 2019

21. Threat-related attentional bias in anxious and nonanxious individuals : a meta-analytic study, Bar-Haim , Y. , Lamy , D. , Pergamin , L. , Bakermans-Kranenburg , M. J. , & van Ijzendoorn , M. H., 133 ( 1 ) :1 ., , 2007

22. When fear is near : threat imminence elicits prefrontal periaqueductal gray shifts in humans ., Mobbs D , Petrovic P , Marchant JL , Hassabis D , Weiskopf N , Seymour B , et al ., 317 ( 5841 ) :1079 83 ., , 2007

23. `` AFNI : software for analysis and visualization of functional magnetic resonance neuroimages, Cox , R. W., 29 ( 3 ) :162 73 ., , 1996

24. Latent class analysis of child behavior checklist anxiety/depression in children and adolescents, Wadsworth ME , Hudziak JJ , Heath AC , Achenbach TM, 40 ( 1 ) :106 14, , 2001

25. The early trauma inventory self report short form : psychometric properties of the Korean version, Jeon J R , Lee E H , Lee S W , Jeong E g , Kim J H , Lee D , et al ., 9 ( 3 ) :229 ., , 2012

26. Abnormal functional connectivity of the amygdala is associated with depression in Parkinson 's disease, Hu X , Song X , Yuan Y , Li E , Liu J , Liu W , et al, 30 ( 2 ) :238 44, , 2015

27. The indirect effect of peer problems on adolescent depression through nucleus accumbens volume alteration ., Lee KH , Yoo JH , Lee J , Kim SH , Han JY , Hong S B , et al, 10 ( 1 ) :1 9 ., , 2020

28. Stress of home life and gender role socializations , family cohesion , and symptom s of anxiety and depression, Anyan F , Hjemdal O ., 58 ( 5 ) :548 64, , 2018

29. Altered effective connectivity network of the amygdala in socia l anxiety disorder : a resting state FMRI study, Liao W , Qiu C , Gentili C , Walter M , Pan Z , Ding J , et al, 5 ( 12 ) : e15238, , 2010

30. Clinical practice guideline for the assessment and treatment of children and adolescents with anxiety disorders, Walter HJ , Bukstein OG , Abright AR , Keable H , Ramtekkar U , Ripperger Suhler J , et al, 59 ( 10 ) :1107 24, , 2020

31. Major depression in the national comorbidity survey adolescent supplement : Prevalence , correlates , and treatment, Avenevoli S , Swendsen J , He J P , Burstein M , Merikangas KR ., 54 ( 1 ) :37 44. e2 ., , 2015

32. Emotion processing , reappraisal , and craving in alcohol dependence : A functional magnetic resonance imaging study, Jansen JM , van den Heuvel OA , van der Werf YD , De Wit SJ , Veltman DJ , Van Den Brink W , et al, 10:227, , 2019

33. Lifetime comorbidity among anxiety disorders and between anxiety disorders and oth er mental disorders in adolescents, Lewinsohn PM , Zinbarg R , Seeley JR , Lewinsohn M , Sack WH, 11 ( 4 ) :377 94, , 1997

34. Remission and recovery in the Treatment f or Adolescents with Depression Study ( TADS ) : acute and long term outcomes, Kennard BD , Silva SG , Tonev S , Rohde P , Hughes JL , Vitiello B , et al, 48 ( 2 ) :186 95 ., , 2009

35. Using support vector machine to identify i maging biomarkers of neurological and psychiatric disease : a critical review, Orru G , Pettersson Yeo W , Marquand AF , Sartori G , Mechelli A ., 36 ( 4 ) :1140 52 ., , 2012

36. Brain structural abnormalities in emotional regulation and sensory processing regions associated with anxious depression ., Peng W , Jia Z , Huang X , Lui S , Kuang W , Sweeney JA , et al, 94:109676, , 2019

37. Prospective associations between the family environment , family cohesion , and psychiatric symptoms among adolescent girls, White J , Shelton KH , Elgar FJ, 45 ( 5 ) :544 54, , 2014

38. R esting state functional connectivity of the ventral attention network in children with a history of depression or anxiety, Sylvester CM , Barch DM , Corbetta M , Power JD , Schlaggar BL , Luby JL ., 52 ( 12 ) :1326 36. e5 ., , 2013

39. Functional Neuroimaging of Anxiety: A Meta-analysis of Emotional Processingin PTSD, Social Anxiety Disorder, and Specific Phobia, Etkin, A. & Wager, T. D., 164 ( 10 ) :1476 88 ., , 2007

40. Mood c ongruent amygdala responses to subliminally presented facial expressions in major depression : associations with anhedonia, Stuhrmann A , Dohm K , Kugel H , Zwanzger P , Redlich R , Grotegerd D , et al ., 38 ( 4 ) :249 58, , 2013

41. Nucleus accum bens volume as a predictor of anxiety symptom improvement following CBT and SSRI treatment in two independent samples, Burkhouse KL , Jimmy J , Defelice N , Klumpp H , Ajilore O , Hosseini B , et al ., 45 ( 3 ) :561 9 ., , 2020

42. The RDoC framework : facilitating transition from ICD/DSM to dimensional approaches that integrate neuroscience and psychopathology, Cuthbert BN, 13 ( 1 ) :28 35, , 2014

43. Cortical thickness and subcortical structure volume abnormalities in patients with major depression with and without anxious symptoms, Zhao K , Liu H , Yan R , Hua L , Chen Y , Shi J , et al, 7 ( 8 ) : e00754 ., , 2017

44. Latent class analysis of anxiety and depressive symptoms of the Youth Self Report in a general population sample o f young adolescents, van Lang ND , Ferdinand RF , Ormel J , Verhulst FC, 44 ( 6 ) :849 60, , 2006

45. Psychometric properties and factor structure of t he Korean version of the screen for child anxiety related emotional disorders ( SCARED ), Shin J , Kim KM , Lee KH , Hong S B , Lee J , Choi C H , et al, 20 ( 1 ) :1 12, , 2020

46. Investigating distinct and common abnormalities of resting state functional connectivity in depression , anxiety , and their comorbid states, Pannekoek J , Van der Werff S , Van Tol M , Veltman D , Aleman A , Zitman F , et al ., 25 ( 11 ) :1933 42 ., , 2015

47. Lesions in the right Rolandic operculum are associated with self rating affective and apathetic depressive symptoms for post stroke patients, Sutoko S , Atsumori H , Obata A , Funane T , Kandori A , Shimonaga K , et al ., ; 10 ( 1 ) :1 10 ., , 2020

48. Stimulated left DLPFC nucleus accumbens functional connectivity predicts the anti depression and anti anxiety effects of rTMS for depression, Du L , Liu H , Du W , Chao F , Zhang L , Wang K , et al, 7 ( 11 ) :1 10, , 2018

49. Development and Application of a Machine Learning Approach to Assess Sho rt term Mortality Risk Among Patients With Cancer Starting Chemotherapy, Elfiky AA , Pany MJ , Parikh RB , Obermeyer Z ., 1 ( 3 ) : e180926 e ., , 2018

50. Antidepre ssant treatment normalizes hypoactivity in dorsolateral prefrontal cortex during emotional interference processing in major depression ., Fales CL , Barch DM , Rundle MM , Mintun MA , Mathews J , Snyder AZ , et al, 112 ( 1 3 ) :206 11 ., , 2009

51. Systematic review and meta analysis : early treatment resp onses of selective serotonin reuptake inhibitors in pediatric major depressive disorder, Varigonda AL , Jakubovski E , Taylor MJ , Freemantle N , Coughlin C , Bloch MH ., 54 ( 7 ) :557 64 ., , 2015

52. Understanding vulnerability for depression from a cognitive neuroscience perspective : A reappraisal of attentional factors and a new conceptual framework, De Raedt R , Koster EH, 10 ( 1 ) :50 70, , 2010

53. Long term outcome of adolescent depression initially resistant to selective serotonin reuptake inhibitor treatment : a follow up study of the TORDIA sample, Vitiello B , Emslie G , Clarke G , Wagner KD , Asarnow JR , Keller MB , et al, 71 ( 3 ) : 10732 ., , 2010

54. Comparative effectiveness and safety of cognitive behavioral therapy and pharmacotherapy for childhood anxiety disorders : a systematic review and meta analysis, Wang Z , W hiteside SP , Sim L , Farah W , Morrow AS , Alsawas M , et al, 171 ( 11 ) :1049 56, , 2017

55. Common abnormalities and disorder specific compensation during implicit regulation of emotional processing in generalized anxiety and major depressive disor ders, Etkin A , Schatzberg AF ., 168 ( 9 ) :968 78, , 2011

56. Clini cal response and risk for reported suicidal ideation and suicide attempts in pediatric antidepressant treatment : a meta analysis of randomized controlled trials, Bridge JA , Iyengar S , Salary CB , Barbe RP , Birmaher B , Pincus HA , et al, 297 ( 15 ) :1683 96 ., , 2007

57. Single nucleotide po lymorphism of the FKBP5 gene and childhood maltreatment as predictors of structural changes in brain areas involved in emotional processing in depression, Tozzi L , Carballedo A , Wetterling F , McCarthy H , O'keane V , Gill M , et al ., 41 ( 2 ) :487 97 ., , 2016