<<

Brain Health

Precision biomarkers for disorders based BMJ: first published as 10.1136/bmj.m3618 on 9 October 2020. Downloaded from on brain imaging Identification of biomarkers could facilitate earlier diagnosis and better treatment, say Runsen Chen and colleagues

ood disorders are a global are associated with abnormalities in both clinical symptoms, while patients with public health problem the structure and function of specific brain these diagnoses present with heterogene- because of their high prev- circuits. Interest is growing in developing ous symptoms.17 alence, chronicity, and precision biomarkers for mood disorders Clinical symptoms may be shared recurrence throughout the based on a deeper understanding of their across mood disorders or with psychiatric lifespan as well as increased risk of mor- biological bases by integrating multilevels disorders. For example, major depressive M1-3 tality. They also impose a heavy eco- of data, such as brain imaging, clinical disorder is commonly associated with nomic burden on society because of lost symptoms, and cognitive behaviours.11 12 anxiety symptoms, but these also occur in work productivity (occupational disabil- Structural and functional magnetic and schizophrenia. Recent ity) and increased use of health services.4 resonance imaging (MRI) can detect subtle studies have used brain imaging and Early diagnosis and effective treatment are deficits in brain structure and function. For transdiagnostic approaches to characterise therefore essential. example, structural MRI data from more the neural basis of the shared symptoms.17-19 Mood disorders are characterised than 20 cohorts worldwide showed that, Resting state fMRI data showed that by a significant change in a person’s compared with healthy controls, patients patients with major depressive disorder, state of mood and include two main with major depressive disorder had thinner bipolar disorder, and schizophrenia had subtypes: depressive disorders (major cortical grey matter in the orbitofrontal common brain functional deficits.18 More depressive episode and ) and cortex, cingulate cortex, and insula,13 generally, recent studies proposed a general bipolar disorders (hypomania, , while patients with bipolar disorder had psychopathology factor (p factor) to describe or —that is, cycling between thinner cortical grey matter in the left a shared vulnerability to a broad range depressed and manic states). In 2015, pars opercularis, left fusiform gyrus, and of symptoms across mental disorders,20 over 300 million people were living with left rostral middle frontal cortex (fig 1).14 and a higher p factor was associated with http://www.bmj.com/ major depressive disorder worldwide, Using resting state functional MRI (fMRI), diminished activation of the frontal pole, representing 4.4% of the global Satterthwaite and colleagues found that the anterior cingulate cortex, and anterior insula population.5 A world mental health survey severity of major depressive disorder was during executive tasks.19 The imbalance in reported that the lifetime and 12 month associated with diminished connectivity these brain circuits is believed to confer prevalences of bipolar disorders in the between the amygdala and frontal areas, vulnerability to mood disorders as well as general population were 2.4% and 1.5%, including both the dorsolateral prefrontal other mental disorders, which could be 6 15 respectively. cortices and anterior cingulate cortex. the underlying mechanism of the shared on 1 October 2021 by guest. Protected copyright. Mood disorders are associated with Studies using task based fMRI showed that symptoms. Other specific brain circuits a widespread cognitive dysfunction, patients with mood disorders exhibited related to factors such as may involving higher-order executive function,7 atypical neural responses to emotional therefore give rise to specific diagnoses such reward processing,8 and emotional processing in the medial prefrontal cortex, as major depressive disorder. regulation (fig 1, box 1).10 These deficits amygdala, and insula, as well as decreased Heterogeneity is also a problem in the neural responses to emotional regulation current diagnostic system. Mood disorders in the dorsolateral prefrontal cortex.16 are increasingly viewed not as a unitary Key Messages Multimodal brain imaging therefore but as a heterogeneous clinical offers the potential to identify biomarkers syndrome that encompasses multiple • Mood disorders are among the lead- for mood disorders that could improve subtypes with distinct pathophysiological ing causes of disability across all age diagnosis and treatment. deficits. Interest is therefore growing in groups parsing the neurobiological heterogeneity of • High quality brain imaging data is Diagnostic confusion mood disorders.17 Drysdale and colleagues providing a better understanding of Current diagnostic systems, including the subdivided major depressive disorder into brain mechanisms underlying shared Diagnostic and Statistical Manual of Men- four discrete biotypes using a clustering symptoms, heterogeneity, and atypi- tal Disorder and International Classifica- approach, each type defined by distinct cal development of mood disorders tion of , were developed based on patterns of dysfunctional connectivity in the • Such data could allow development clinical symptoms and signs. The acquired limbic and frontostriatal systems.21 Rather of precision biomarkers for mood dis- diagnostic categories do not align with the than dividing mood disorders into discrete orders underlying psychopathology or predict categories, dimensional approaches are • The benefits of a cognitive neuropsy- treatment response. The diagnostic cat- able to parse them into several dimensions chological model could be harnessed egories of major depressive disorder and using brain imaging, with each dimension to predict treatment outcomes bipolar disorder, for example, share many representing loadings onto symptoms.17 the bmj | BMJ 2020;371:m3618 | doi: 10.1136/bmj.m3618 1 Brain Health

disorder had a greater grey matter Striatum volume contraction and less white matter BMJ: first published as 10.1136/bmj.m3618 on 9 October 2020. Downloaded from expansion in the prefrontal cortex during Thalamus development.24 Xia and colleagues found Prefrontal that mood symptoms were related to cortex dysconnectivity within the frontoparietal system and the pattern of dysconnectivity was strengthened from childhood to adulthood.25 Based on these findings, scientists increasingly conceptualise mood disorders as neurodevelopmental disorders.26 An understanding of the atypical brain development in patients with Anterior mood disorders could allow identification cingulate cortex of a biologically informed time of onset for Amygdala mood disorders, which might be earlier than the onset of clinical symptoms. Dense sampling shows particular Executive function Hippocampus Emotional regulation promise for precise characterisation of the Reward processing atypical development of brain functional organisation in patients with mood disorders. Typical fMRI acquisitions last Fig 1 | Brain regions underlying executive function, emotional regulation, and reward 5-10 minutes, and may be adequate for processing. The figure was made with reference from Wang and Olson9 characterising group average functional organisation. However, such short acquisitions have limited reliability for Dimensional approaches can account Accurate identification of the time of onset describing an individual’s brain functional for the continuous spectrum from health of a disorder is essential for achieving this 27 28 organisation in detail. Recent studies to disease by including people at risk of goal. Clinical symptoms of mood disorders have shown individual variations in the illness and considering diagnosable cases usually begin in youth, with the mean age topography of brain organisation that are as an extreme phenotype. Consistent with of onset being 14.9 years for major depres- 22 obscured in group level organisation and these efforts, the US National Institute of sive disorder and 20.2 years for bipolar http://www.bmj.com/ 11 12 23 can only be reliably detected using dense Mental Health research domain criteria disorders. Therefore, youth—defined 27 29 30 fMRI scanning data. Dense fMRI framework invited the scientific and clinical as ranging from childhood to early adult- protocols acquire a much longer time series communities to integrate brain imaging, hood—is a critical period for diagnosis and of fMRI data on one person, which can be cognition, and clinical symptoms to develop implementation of suitable interventions. acquired by repeated scanning in multiple a novel biologically informed nosology and Youth is also a period of considerable sessions. Such protocols reveal specific precision biomarkers for mood disorders. brain development, and atypical brain topography details for individuals that are

developmental trajectories are related on 1 October 2021 by guest. Protected copyright. both reliable and reproducible and could be High quality imaging data for early to clinical symptoms and cognitive useful in identifying precision biomarkers identification dysfunction in patients with mood for mood disorders.31 Precision biomarkers need to be detected disorders.26 For example, longitudinal Recently, Cui and colleagues used 27 at an early stage in the progression of mood high resolution MRI showed that relative minute fMRI acquisitions, which is much disorders to facilitate early intervention. to healthy youths, those with bipolar longer than traditional acquisitions, to show that the topography of brain functional organisation is refined Box 1: Key terms during youth.32 Such characterisation of • Higher order executive dysfunction—Includes difficulties in planning, organisation, and typical brain development in youth is a inhibiting distracting thoughts or behaviours. prerequisite for understanding differences • Reward processing dysfunction—Reduced ability to learn or modulate behaviours as a in brain development in mood disorders. function of reinforcement Results also showed that the individual • —Includes difficulties in down regulating negative emotions and variation in brain functional organisation excessive emotional reactivity predicted executive function,32 which is • Precision biomarkers—A system of diagnosis and treatment evaluation for mood disorders typically impaired in mood disorders. based on an understanding of their biological basis These findings suggest that the topo­ • General psychopathology factor (p factor)—A factor that describes a shared vulnerability to a graphy of functional organisation broad range of symptoms across mental disorders revealed by dense fMRI scanning could • Dense fMRI scanning—Obtaining large quantities of functional MRI data in a single person, be helpful in understanding the atypical such as through many hours of scanning neurodevelopmental trajectories of mood • Cognitive neuropsychological model—In the early stage of treatments the disorders. However, dense fMRI scanning primary effect of is to modify negative biases in the processing of emotional data are still lacking for healthy youths as information well as for youths with mood disorders,

2 doi: 10.1136/bmj.m3618 | BMJ 2020;371:m3618 | the bmj Brain Health

and the neuroscience and

Start of Biological changes BMJ: first published as 10.1136/bmj.m3618 on 9 October 2020. Downloaded from communities should concentrate on this antidepressant Downstream neuroadaptive effects effort. treatment As well as dense fMRI scanning data Delay onset Monoamine DNA RNA neurotransmission proteins on individuals, data from large samples are also essential to characterise how BDNF Glutamate mood disorders differ from the typical mTOR brain developmental trajectory. The 5-HT1A (serotonin) research community has publicly receptor desensitisation released several youth datasets with large sample sizes, such as the Philadelphia Early in treatment 33 (hours to days) neurodevelopmental cohort (box 2). The ? Philadelphia cohort includes young people representing a spectrum from a healthy to a diseased state and is therefore a great Change in emotional bias Improved mood symptoms resource for identification of the time of onset of mood disorders. Future efforts Fig 2 | Antidepressants produce positive biases in emotional processing from the start of should collect datasets with both a large treatment. This is an automatic effect which does not affect subjective mood state immediately sample size and dense fMRI scanning to (BDNF=brain derived neurotropic factor, mTOR=mammalian target of rapamycin) better characterise the time of onset. disorder before producing any effects on of novel drugs in both patients and healthy Early biomarkers of treatment outcomes mood. volunteers. Neural responses could also Identifying early biomarkers of treatment Interestingly, studies have consistently serve as an early biomarker of treatment outcomes is vital for evaluating treatment shown that antidepressant induced early efficacy and improve understanding of strategies for mood disorders. A recent changes in neural response to emotional mechanisms of action, including those theory of antidepressant drug action—the processing tasks predict treatment underlying drugs targeting N-methyl-D- cognitive neuropsychological model—sug- outcomes.37 A recent study assessed aspartate receptors (such as ketamine37). gests that drugs do not act primarily on changes in neural response to emotional Candidate drugs are typically screened mood but rather modify biases in the cog- facial expressions before and after 7 using animal models, which have low nitive processing of emotional information days of treatment with predictive validity, and a lack of efficacy in the brain (fig 2).34 35 fMRI has been used in 35 patients with major depressive in translation to humans is the main http://www.bmj.com/ to study changes in the brain associated disorder who had not previously had reason for failure in drug development.39 with positive or negative biases in emo- drug treatment.38 Based on the criterion Ensuring that new drugs affect the core tional processing that occur when patients of 50% reduction in depressive symptoms measures of emotional or cognitive with mood disorders and healthy volun- at the end of six weeks’ treatment, 22 function during the early phase of drug teers take antidepressants. For instance, patients were classified as responders to development could reduce costs of failed 36 Godlewska and colleagues found that, the treatment and 13 as non-responders. clinical trials. A better understanding of compared with placebo, treatment with Compared with the non-responders, the cognitive neuropsychological effects the antidepressant escitalopram for 7 days responders showed decreased neural of psychopharmacological drugs and using on 1 October 2021 by guest. Protected copyright. normalised neural responses in patients activity to fearful versus happy faces in the neural assessment to predict treatment with major depressive disorder by reduc- amygdala, insula, anterior and posterior response may in future allow better ing the response of the amygdala to nega- cingulate cortices, bilateral supramarginal patient stratification for tive stimuli (fearful facial expressions). gyri, and thalamus. These results suggest treatment. Importantly, the neural changes seen that early changes in neural responses during the early stages of antidepressant to fearful faces are predictive of clinical Conclusion treatment were independent of subjective response to treatment for major depressive Development and validation of precision mood changes, suggesting that antide- disorder. biomarkers for mood disorders is urgently pressant drugs act directly on neural func- Studies of neural responses could needed for early diagnosis and treatment tions that are relevant to major depressive therefore be useful in testing the efficacy evaluation. However, the overlap and heterogeneity in mood disorders impede this progress. Brain imaging, which can detect brain structural and functional Box 2: Open access brain developmental datasets in youth changes, is one promising way to solve • Philadelphia neurodevelopmental cohort: https://www.ncbi.nlm.nih.gov/projects/gap/cgi- this problem and identify biomarkers that bin/study.cgi?study_id=phs000607.v3.p2 can cut across different facets of disorders • Adolescent Brain Cognitive Development: https://abcdstudy.org of mood and emotion. This will require a • Human Connectome Project Development: https://humanconnectome.org/study/hcp- combination of brain imaging, cognitive lifespan-development neuroscience, experimental modelling, • Pediatric Imaging, Neurocognition, and Genetics (PING) Data Repository: http://pingstudy. and computational techniques (box 3). ucsd.edu/ We encourage the close collaborations • Healthy Brain Network Biobank: https://data.healthybrainnetwork.org/main.php between a diverse research community • IMAGEN: http://www.imagen-europe.com/ for this endeavour. the bmj | BMJ 2020;371:m3618 | doi: 10.1136/bmj.m3618 3 Brain Health

Psychiatry 2006;19:34-9. doi:10.1097/01.

Box 3: Future directions in precision biomarkers for mood disorders yco.0000191500.46411.00 BMJ: first published as 10.1136/bmj.m3618 on 9 October 2020. Downloaded from • Understand the neural substrate underlying shared symptoms using transdiagnostic 11 Insel TR. The NIMH research domain criteria (RDoC) project: precision medicine for psychiatry. Am J approach, which cut across the diagnostical categories Psychiatry 2014;171:395-7. doi:10.1176/appi. • Identify biologically informed biotypes of mood disorder and their distinct brain deficits ajp.2014.14020138 • Characterise how the brain development of patients with mood disorders deviates from the 12 Insel T, Cuthbert B, Garvey M, et al. Research domain criteria (RDoC): toward a new classification typical trajectory in youth framework for research on mental disorders. Am J • Use brain imaging data to predict the emergence of mood disorders and treatment outcome Psychiatry 2010;167:748-51. doi:10.1176/appi. • Advance the cognitive neuropsychological model with improved experimental designs to ajp.2010.09091379 better evaluate the treatment outcomes of psychopharmacological therapies 13 Schmaal L, Hibar DP, Sämann PG, et al. Cortical abnormalities in adults and adolescents with major based on brain scans from 20 cohorts worldwide in the ENIGMA Major Depressive Disorder 2 We thank Ying Wang for help with the figures. Department of Psychiatry, University of Oxford, Working Group. Mol Psychiatry 2017;22:900-9. Oxford, UK doi:10.1038/mp.2016.60 Contributors and sources: GW is the president 3Department of Psychiatry, University of Pennsylvania, 14 Hibar DP, Westlye LT, Doan NT, et al. Cortical of Beijing Anding Hospital, Capital Medical Philadelphia, PA 19104, USA abnormalities in bipolar disorder: an MRI analysis of 6503 individuals from the ENIGMA bipolar disorder University, and the director of the National Clinical 4 Research Center for Mental Disorders, China. Oxford Health NHS Foundation Trust, Warneford working group. Mol Psychiatry 2018;23:932-42. RC is a mental health researcher, specialising Hospital, Oxford, UK doi:10.1038/mp.2017.73 in psychopharmacology, emotion, and clinical Correspondence to: G Wang 15 Satterthwaite TD, Cook PA, Bruce SE, et al. psychology. ZC is a postdoctoral fellow investigating [email protected] Dimensional depression severity in women with typical and atypical brain development using brain major depression and post-traumatic stress disorder imaging and computational techniques. LC is a correlates with fronto-amygdalar hypoconnectivty. researcher within the experimental medicine and Mol Psychiatry 2016;21:894-902. doi:10.1038/ adult mental health themes, part of the Oxford mp.2015.149 Health Biomedical Researcher Centre. TS is director This is an Open Access article distributed in 16 Malhi GS, Byrow Y, Fritz K, et al. Mood of the lifespan informatics and neuroimaging accordance with the Creative Commons Attribution disorders: neurocognitive models. Bipolar centre at the University of Pennsylvania Perelman Non Commercial (CC BY-NC 4.0) license, which Disord 2015;17(Suppl 2):3-20. doi:10.1111/ School of Medicine. CH is the director of the permits others to distribute, remix, adapt, build bdi.12353 psychopharmacology and emotional research upon this work non-commercially, and license 17 Kaczkurkin AN, Moore TM, Sotiras A, Xia CH, laboratory at the University of Oxford. GW is a their derivative works on different terms, provided Shinohara RT, Satterthwaite TD. Approaches to guarantor of the article. RC and ZC drafted the article. the original work is properly cited and the use is defining common and dissociable neurobiological All authors contributed to critical revision of the non-commercial. See: http://creativecommons.org/ deficits associated with psychopathology in youth. article and approved the final manuscript. RC and licenses/by-nc/4.0/. Biol Psychiatry 2020;88:51-62. doi:10.1016/j. ZC contributed equally to this article and share first biopsych.2019.12.015 authorship. 18 Ma Q, Tang Y, Wang F, et al. Transdiagnostic dysfunctions in brain modules across patients Competing interests: We have read and understood with schizophrenia, bipolar disorder, and major http://www.bmj.com/ the BMJ policy on declaration of interests and have depressive disorder: a connectome-based study. 1 Vieta E, Berk M, Schulze TG, et al. Bipolar disorders. the following interests to declare: GW declares Schizophr Bull 2020;46:699-712. doi:10.1093/ Nat Rev Dis Primers 2018;4:18008. doi:10.1038/ funding from the National Key Research and schbul/sbz111 nrdp.2018.8 Development Program of China (2016YFC1307200; 19 Shanmugan S, Wolf DH, Calkins ME, et al. Common 2 Belmaker RH, Agam G. Major depressive disorder. 2017YFA0505700), the Capital’s Funds for Health and dissociable mechanisms of executive system N Engl J Med 2008;358:55-68. doi:10.1056/ Improvement and Research (2018-1-2121), and dysfunction across psychiatric disorders in youth. Am NEJMra073096 the Capital’s Science and Technology Talent Project J Psychiatry 2016;173:517-26. doi:10.1176/appi. 3 Carvalho AF, Firth J, Vieta E. Bipolar disorder. (Z181100006318009). LP and CH are supported by ajp.2015.15060725 N Engl J Med 2020;383:58-66. doi:10.1056/ the NIHR Oxford Health Biomedical Research Centre. 20 Caspi A, Houts RM, Belsky DW, et al. The p factor: one on 1 October 2021 by guest. Protected copyright. NEJMra1906193 The views expressed are those of the authors and general psychopathology factor in the structure of 4 Simon GE. Social and economic burden of mood not necessarily those of the NHS, the NIHR, or the psychiatric disorders?Clin Psychol Sci 2014;2:119- disorders. Biol Psychiatry 2003;54:208-15. Department of Health. CH receives consultancy fees 37. doi:10.1177/2167702613497473 doi:10.1016/S0006-3223(03)00420-7 from p1vital, Lundbeck, Sage Therapeutics, and J&J. 21 Drysdale AT, Grosenick L, Downar J, et al. 5 World Health Organization. Depression and other The funders had no role in the design or conduct of Resting-state connectivity biomarkers define common mental disorders: global health estimates. the study, collection, management, analysis, or the neurophysiological subtypes of depression. Nat World Health Organization; 2017. https://apps.who. interpretation of the data. Med 2017;23:28-38. doi:10.1038/nm.4246 int/iris/bitstream/handle/10665/254610/WHO- 22 Lewinsohn PM, Clarke GN, Seeley JR, Rohde P. Major MSD-MER-2017.2-eng.pdf Provenance and peer review: Commissioned; depression in community adolescents: age at onset, 6 Merikangas KR, Jin R, He J-P, et al. Prevalence externally peer reviewed. episode duration, and time to recurrence. J Am and correlates of bipolar in Acad Child Adolesc Psychiatry 1994;33:809-18. This article is part of a series launched at the Chinese the world mental health survey initiative. Arch doi:10.1097/00004583-199407000-00006 Stroke Association annual conference on 10 October Gen Psychiatry 2011;68:241-51. doi:10.1001/ 23 Morken G, Vaaler AE, Folden GE, Andreassen OA, 2020, Beijing, China. Open access fees were funded archgenpsychiatry.2011.12 Malt UF. Age at onset of first episode and time to by the National Science and Technology Major Project. 7 Cotrena C, Branco LD, Shansis FM, Fonseca RP. treatment in in-patients with bipolar disorder. Br J The BMJ peer reviewed, edited, and made the decision Executive function impairments in depression Psychiatry 2009;194:559-60. doi:10.1192/bjp. to publish these articles. and bipolar disorder: association with bp.108.054452 Runsen Chen, researcher1,2 functional impairment and quality of life. J Affect 24 Najt P, Wang F, Spencer L, et al. Anterior cortical 3 Disord 2016;190:744-53. doi:10.1016/j. Zaixu Cui, postdoctoral fellow development during adolescence in bipolar disorder. jad.2015.11.007 2,4 Biol Psychiatry 2016;79:303-10. doi:10.1016/j. Liliana Capitão, research coordinator 8 Panchal P, Kaltenboeck A, Harmer CJ. Cognitive biopsych.2015.03.026 Gang Wang, professor1 emotional processing across mood disorders. 25 Xia CH, Ma Z, Ciric R, et al. Linked dimensions of 3 CNS Spectr 2019;24:54-63. doi:10.1017/ Theodore D Satterthwaite, associate professor psychopathology and connectivity in functional brain S109285291800130X networks. Nat Commun 2018;9:3003. doi:10.1038/ Catherine Harmer, professor of cognitive 9 Wang Y, Olson IR. The original social network: 2,4 s41467-018-05317-y neuroscience white matter and social cognition. Trends 1 26 Insel TR. Mental disorders in childhood: National Clinical Research Center for Mental Cogn Sci 2018;22:504-16. doi:10.1016/j. shifting the focus from behavioral symptoms Disorders, Beijing Key Laboratory of Mental Disorders, tics.2018.03.005 to neurodevelopmental trajectories. Beijing Anding Hospital, and Advanced Innovation 10 Leppänen JM. Emotional information processing JAMA 2014;311:1727-8. doi:10.1001/ Center for Human Brain Protection, Capital Medical in mood disorders: a review of behavioral jama.2014.1193 University, Beijing, China and neuroimaging findings. Curr Opin

4 doi: 10.1136/bmj.m3618 | BMJ 2020;371:m3618 | the bmj Brain Health

27 Laumann TO, Gordon EM, Adeyemo B, et al. 32 Cui Z, Li H, Xia CH, et al. Individual variation in Psychol Med 2012;42:2609-17. doi:10.1017/

Functional system and areal organization of functional topography of association networks S0033291712000591 BMJ: first published as 10.1136/bmj.m3618 on 9 October 2020. Downloaded from a highly sampled individual human brain. in youth. Neuron 2020;106:340-353.e8. 37 Harmer CJ, Duman RS, Cowen PJ. How do Neuron 2015;87:657-70. doi:10.1016/j. doi:10.1016/j.neuron.2020.01.029 antidepressants work? New perspectives for neuron.2015.06.037 33 Satterthwaite TD, Elliott MA, Ruparel K, refining future treatment approaches. Lancet 28 Zuo XN, Xu T, Milham MP. Harnessing reliability for et al. Neuroimaging of the Philadelphia Psychiatry 2017;4:409-18. doi:10.1016/S2215- neuroscience research. Nat Hum Behav 2019;3:768- neurodevelopmental cohort. 0366(17)30015-9 71. doi:10.1038/s41562-019-0655-x Neuroimage 2014;86:544-53. doi:10.1016/j. 38 Godlewska BR, Browning M, Norbury R, Cowen PJ, 29 Gordon EM, Laumann TO, Gilmore AW, et al. Precision neuroimage.2013.07.064 Harmer CJ. Early changes in emotional processing functional mapping of individual human brains. 34 Harmer CJ, Goodwin GM, Cowen PJ. Why do as a marker of clinical response to SSRI treatment Neuron 2017;95:791-807.e7. doi:10.1016/j. antidepressants take so long to work? A cognitive in depression. Transl Psychiatry 2016;6:e957. neuron.2017.07.011 neuropsychological model of antidepressant doi:10.1038/tp.2016.130 30 Braga RM, Buckner RL. Parallel interdigitated drug action. Br J Psychiatry 2009;195:102-8. 39 Pringle A, Harmer CJ. The effects of drugs on human distributed networks within the individual doi:10.1192/bjp.bp.108.051193 models of emotional processing: an account of estimated by intrinsic functional connectivity. 35 Zghoul T, Cowen PJ, Harmer CJ. A perspective: antidepressant drug treatment. Dialogues Clin Neuron 2017;95:457-471.e5. doi:10.1016/j. from the serotonin hypothesis to cognitive Neurosci 2015;17:477-87. neuron.2017.06.038 neuropsychological approaches. The serotonin 31 Gratton C, Kraus BT, Greene DJ, et al. Defining system. Elsevier, 2019: 95-104. individual-specific functional neuroanatomy for 36 Godlewska BR, Norbury R, Selvaraj S, Cowen PJ, precision psychiatry. Biol Psychiatry 2020;88:28-39. Harmer CJ. Short-term SSRI treatment normalises Cite this as: BMJ 2020;371:m3618 doi:10.1016/j.biopsych.2019.10.026 amygdala hyperactivity in depressed patients. http://dx.doi.org/10.1136/bmj.m3618 http://www.bmj.com/ on 1 October 2021 by guest. Protected copyright.

the bmj | BMJ 2020;371:m3618 | doi: 10.1136/bmj.m3618 5