30.Wk2

DEPRESSION AND ANXIETY 30:787–791 (2013)

The Cutting Edge BIOMARKERS IN PEDIATRIC DEPRESSION

Uma Rao, M.D., is Professor of Psychiatry at Meharry Medical College and Vanderbilt Uni- versity. She serves as the Director of the Center for Molecular and Behavioral Neuroscience and holds the Endowed Chair in Brain and Behavior Research at Meharry Medical College. Her work is focused on the interactions between neurobiological and psychosocial factors in predicting the onset and longitudinal clinical course of depressive and addictive disorders in adolescents. She is also involved in translational intervention studies for these disorders. She has received awards from the American Academy of Child and Adolescent Psychiatry as well as other community organizations.

Depression is a leading cause of morbidity and mortal- ity in youngsters. Elevated risk for the disorder begins in the early teens and continues to rise in a linear fashion throughout adolescence, with lifetime rates estimated to range from 15 to 25% by late adolescence. Numerous studies have documented that early depressive episodes persist or recur into adult life along with ongoing psy- chosocial difficulties. A better understanding of the eti- ology and pathophysiology of pediatric depression will be helpful in the development and implementation of more effective primary and secondary preventive strate- gies, thereby allowing such youth to achieve their full potential as adults.

There is a general consensus that depression results from complex interactions between multiple genetic and environmental factors. Endophenotypes or biomarkers help target the underlying mechanisms. The biomarkers also can be used to strengthen classifications of clinical phenotypes, or to differentiate possible biological subtypes that may, in turn, have different clinical or treatment profiles. In order to better characterize the po- tential biomarkers associated with pediatric depression, a summary of the literature on adult depression will be provided; in contrast to the wealth of information avail- able in adults, empirical data in youngsters are limited by relatively modest sample sizes in far fewer studies.

BIOMARKERS IN ADULT DEPRESSION

Although there is no clear single biomarker associ- ated with depression, there is mounting evidence of

∗Correspondence to: Uma Rao, Center for Molecular and Behav- ioral Neuroscience, Meharry Medical College, 1005 Dr. D.B. Todd Jr. Boulevard, Nashvile, TN 37208. E-mail: [email protected]

DOI 10.1002/da.22171 Published online in Wiley Online Library (wileyonlinelibrary.com).

contributing factors, including sleep, neuroendocrine, inflammatory, metabolic, neurotrophic, and neural networks.[1–5]

SLEEP There are a number of reasons to consider the regula-

tion of sleep as an essential component for understanding the pathophysiology and treatment of depression.[2, 6] There is a significant overlap in the control of sleep and mood regulation. Sleep complaints are commonly associated with depression and form an essential crite- rion of the diagnosis. Developmental influence(s) on the rates of depression and maturational changes in sleep regulation also imply a close connection between de- pressive disorders and sleep regulation. Depression and sleep “abnormalities” often co-segregate among family members. Certain sleep markers have been detected in healthy individuals at high familial risk for depression and they were associated with the development of de- pression during prospective follow-up. Sleep alterations often persist beyond the clinical episode of depression and increase the vulnerability to relapse or recurrence. Sleep changes also predict treatment response, and many antidepressant treatments impact sleep.

Although no single sleep marker is specifically associ- ated with depression, a constellation of sleep changes has been observed. The most reliable sleep macroarchitec- tural changes associated with major depression include sleep continuity disturbances (e.g., delayed sleep onset and decreased sleep efficiency), earlier onset of rapid eye movement (REM) sleep, increased REM activity and REM density, increased amount of REM sleep, and diminished slow-wave sleep (SWS).[7] Specifically, individuals experiencing major depression and those in remission exhibit increased REM density and shortened SWS, as do persons at high familial risk. Therefore, this combination of sleep features may represent a genetic biomarker of depression. Further, SWS appears to be even shorter during remission following a depressive

C© 2013 Wiley Periodicals, Inc.

788 Rao

episode, suggesting its role as both a genetic marker as well as a biological scar of the disorder.

NEUROENDOCRINE SYSTEM(S) There has been considerable interest in the

hypothalamic–pituitary–adrenal (HPA) system, consis- tent with the hypothesis that depression is linked to al- tered responses to stress.[4, 5] Depression is associated with higher basal corticotropin-releasing hormone and cortisol secretion.[8] Increased HPA activity in depres- sion is due, in part, to altered feedback inhibition of the HPA axis by endogenous glucocorticoids. It has been proposed that elevated cortisol in patients with de- pression is a compensatory mechanism in response to decreased glucocorticoid receptor function and expres- sion in the brain. Increased cortisol secretion frequently persists during remission and increases the risk for re- lapse/recurrence of depression.[3] HPA-axis hyperactiv- ity also was observed in unaffected individuals at familial risk for depression and predicted the onset of depression, suggesting that it may be a genetic vulnerability marker of depression.[3, 9]

INFLAMMATORY MARKERS Evidence suggests that inflammation may have a criti-

cal role in the pathophysiology of depression.[1, 3, 4] Clin- ical studies demonstrated that patients with depression have elevated blood/serum levels of inflammatory mark- ers, including proinflammatory cytokines. Cytokine ac- tivation produces sickness behaviors, which share fea- tures with depression. Moreover, chronic stress exposure produces changes in immune function that may influ- ence the pathophysiology of depression. Consistent with these findings, inhibiting proinflammatory cytokine sig- naling in patients with inflammatory disorders, as well as in patients with depression, improves mood and facili- tates antidepressant treatment response. Although most of the data on the association between depression and inflammatory markers are cross-sectional, several lines of research indicate that the link between inflammation and depression is likely bidirectional.[3]

METABOLIC FUNCTION Evidence indicates a bidirectional relationship be-

tween depression and metabolic dysregulation.[3, 4] Circulating hormones, such as leptin and ghrelin, relay information regarding peripheral energy homeostatic levels to the brain. Leptin receptors are expressed in the limbic system, and leptin also has been shown to affect hippocampal and cortical structures through its actions on neurogenesis, axon growth, synaptogensis, and den- dritic morphology. Low levels of leptin are associated with depressive behaviors, and chronic stress exposure decreases serum leptin. Consistent with these results, acute leptin administration produces antidepressant ef- fects and increases the expression of brain-derived neu- rotrophic factor (BDNF) in the hippocampus. By con-

trast, chronic stress exposure increases serum ghrelin levels. Calorie restriction produces antidepressant ef- fects, which are mediated by ghrelin.[4] Hence, leptin and ghrelin may serve as putative biomarkers for de- pression in general, or in depressed patients with altered metabolic function.

NEUROTROPHIC FACTORS Chronic stress exposure, which can precipitate or

exacerbate depressive episodes, alters the expression of neurotrophic (growth) factors. By contrast, antide- pressant treatment enhances trophic factor expression and neuroplasticity.[4] Among the various neurotrophic factors, BDNF has been best investigated in both pre- clinical and clinical studies. BDNF is transcribed at rel- atively high levels and expressed in several peripheral tissues, and plasma/serum BDNF levels might be de- rived from these tissues as well as the brain. Although the functional significance of plasma/serum BDNF is not known, recent studies in animals suggest that peripheral growth factors (including BDNF) can enter the brain and produce both behavioral and cellular responses. Several studies have reported reduced plasma/serum BDNF lev- els in patients with depression, and that antidepressant treatment normalizes these levels.[4, 5, 10]

NEURAL NETWORKS A growing body of unimodal structural and functional

neuroimaging studies found alterations in frontolim- bic and frontostriatal circuits in depression.[5, 11] A few investigations utilizing multimodal methods have sug- gested that structural alterations are related to functional changes, while others reported functional alterations in the absence of structural changes.[11] Alterations in neu- ral structure/function also have been associated with the treatment of depression (as predictors of clinical re- sponse and/or as changes in response to antidepressant treatments).[12]

BIOMARKERS IN PEDIATRIC DEPRESSION

Although there are far fewer studies with modest sam- ple sizes in pediatric depression, many of the biomarkers described in adult depression also have been observed in pediatric depression. For example, reduced REM la- tency and REM density and hypercortisolemia not only have been observed during a depressive episode, but also in unaffected youth at high familial risk for depression, and these biomarkers predicted the subsequent onset of depression.[9] A bidirectional relationship between depression and inflammatory markers was observed during prospective follow-up in unaffected adolescents who were at high risk for depression based on familial or cognitive vulnerability, particularly in those who experienced childhood adversity.[13] Depression was associated with inflammatory and metabolic markers in

Depression and Anxiety

The Cutting Edge: Biomarkers in Pediatric Depression 789

youth with diabetes.[14] Reduced neurotrophic factors (including BDNF) have been reported in pediatric depression.[15] Alterations in frontolimbic and fron- tostriatal circuits also have been observed in structural and functional neuroimaging studies of early-onset depression, with some studies reporting these changes in unaffected youth at high familial risk.[16]

The advantage of studying pediatric samples is that we can identify premorbid (vulnerability) markers in high- risk populations so that they can be targeted for preven- tive interventions and reduce the economic and social burden associated with depression. Even individuals with the disorder are in the early course of illness and will be ideal for identifying markers associated with treatment resistance or poor prognosis. Also, early-onset illness is highly familial and is associated with a chronic recur- rent course, possibly representing a unique subtype that could be targeted for identifying genetic and epigenetic correlates.[17]

INTEGRATIVE SUMMARY OF BIOMARKERS IN DEPRESSION

Given that depression is a heterogeneous disorder, it is unlikely that any given biomarker has a high degree of sensitivity and specificity to make it clinically use- ful. Hence, the development of panels that aim to pro- file a diverse array of biomarkers to provide coverage of multiple biological abnormalities that contribute to the heterogeneity of depression, and its response to treat- ment, is a potentially promising approach. Consistent with this theme, a combination of biomarkers seem to aggregate more consistently in some patients but not in others. Such clustering may be more closely related to the etiology of a depression subtype and, in turn, could lead to more effective, etiologically based treatments for subgroups of patients.[3, 4]

Although it appears that the above-described biomarkers are diverse, they are interrelated. For ex- ample, sleep regulation is not only linked to mood, but also to cognitive, endocrine, immune, and metabolic functions. Inflammatory markers, including cytokines, regulate neuroendocrine function, and reciprocally glu- cocorticoids have inhibitory effects on inflammation. Both HPA and immune systems affect metabolic func- tion, and vice versa. Activation of inflammatory path- ways and reduced leptin and glucocorticoid receptor function within the brain are believed to contribute to decreased neurotrophic support and altered glutamate release/reuptake, as well as oxidative stress, leading to excitotoxicity and loss of glial elements, consistent with neuropathologic findings that characterize depression.

Genetic variants have a moderating influence on the association between biomarkers and depression phe- notypes and/or treatment response. For example, in different clinical studies, the Val66Met BDNF polymor- phism (met-allele) was associated with higher depression

severity, lower serum BDNF levels, smaller hippocam- pal volume or abnormal hippocampal activity during an episodic memory task, but better response to antidepres- sant treatment, compared to the Val-allele.[5] An emerg- ing literature suggests that stress exposure induces epi- genetic mechanisms, such as histone modifications and deoxyribonucleic acid methylation, and alters the expres- sion of many of these biomarkers. Therefore, tracking genetic variants and epigenetic changes might compli- ment biomarker panels.[1, 4, 5]

A MULTIDIMENSIONAL BIOLOGICAL MODEL OF

DEPRESSION Based on the multidimensional concept, Schneider

et al. proposed an exemplary biological model of depression (see Fig. 1).[5] This model consists of three levels: a neuronal network level (assessed with structural and functional neuroimaging methods as well as with neurophysiological methods), a molecular systems level (assessed using proteomic, lipidomic, and tran- scriptomic approaches), and a genetic/epigenetic level. Additionally, environmental factors are considered as an important factor within this model, as they interact with neurobiological systems on all three levels. Various constellations of gene–gene and gene–environment interactions can lead to imbalances within and across these three major pathways, culminating in depressive disorder. On the other hand, such interactions also could have buffering effects and promote resilience in at-risk individuals.

Consistent with this theme, a recent study identi- fied 26 candidate blood transcriptomic markers from genome-wide analysis of two animal models, represent- ing the genetic and environmental (stress-related), eti- ology of depression.[18] They applied these markers in a sample of adolescents with depression (n = 14) and con- trols with no disorder (n = 14). A panel of 11 blood mark- ers differentiated depressed and control groups. Addi- tionally, a separate but partially overlapping panel of 18 transcripts distinguished depressed youth with or with- out comorbid anxiety. These data should be replicated in larger samples before these markers can be applied clinically.

A multidimensional approach, as described in Fig. 1, requires large sample sizes to obtain adequate power to define severity and identifiable subtypes of depression. Results obtained from the biomarker panels will need to be reproducible and standardized such that clear asso- ciations between these biosignatures and clinical sub- types are readily apparent. Some approaches, such as neuroimaging and sleep electroencephalography, are not practical or cost-effective in routine clinical practice. In- stead, ambulatory measures such as near-infrared spec- troscopy might be more suitable.[19]

Depression and Anxiety

790 Rao

Figure 1. A biological model of depression (adapted from Schneider et al.[ 5 ]).

CLINICAL APPLICATION OF BIOMARKERS FOR DEPRESSION Ridge Diagnostics developed a blood-based diagnos-

tic test, Major Depressive Disorder Score (MDDScore). This multianalyte immunoassay detects four major bi- ological pathways: inflammation, HPA axis, metabolic, and eurochemical pathways. A mathematical algorithm provides an MDDScore, which predicts the probabil- ity of depression with high sensitivity and specificity.[20] The MDDScore is yet to be compared across laborato- ries and tested on a large population of patients with a va- riety of psychiatric disorders before the test is standard- ized and available for use in routine clinical practice. The Food and Drug Administration has approved genotyp- ing tests for common variants of drug metabolism genes (e.g., cytochrome P450). These tests help the physi- cian to select an appropriate antidepressant drug for a given patient, as differences in clearance, half-life, and peak blood concentrations are controlled by genetic vari- ability in drug metabolism. They are especially helpful in alerting to possible adverse effects and to optimize dose.[1]

CONCLUSIONS Currently, there is no clear biomarker profile associ-

ated with depression for clinical use. However, based on a combination of preclinical and clinical studies, a bio- chemical profile has emerged that can be tested in clin- ical populations. In the meantime, existing information

on genetic biomarkers and the availability of genotyping tests concerning genetic control of drug metabolism and associated toxicity can help the physician in selecting a safe antidepressant for a given patient. A truly personal- ized medicine approach for depression will be achieved only when the biomarker assays are widely available and can be considered to be cost-effective diagnostic tests. Research directed toward the discovery of biomarkers associated with depression and its response to treatment is of the utmost importance in this endeavor.

Acknowledgments. This work was supported, in part, by grants from the National Institutes of Health (R01 MH068391, G12 RR003032/MD007586, UL1 RR024975/TR000445, and U54 RR026140/ MD007593), and by the Endowed Chair in Brain and Behavior Research at Meharry Medical College.

REFERENCES 1. Miller DB, O’Callaghan JP. Personalized medicine in ma-

jor depressive disorder—opportunities and pitfalls. Metabolism 2013;62(Suppl 1):S34–S39.

2. Palagini L, Baglioni C, Ciapparelli A, et al. REM sleep dys- regulation in depression: state of the art. Sleep Med Rev 2013.

3. Penninx BW, Milaneschi Y, Lamers F, Vogelzangs N. Under- standing the somatic consequences of depression: biological mech- anisms and the role of depression symptom profile. BMC Med 2013;11:129.

Depression and Anxiety

The Cutting Edge: Biomarkers in Pediatric Depression 791

4. Schmidt HD, Shelton RC, Duman RS. Functional biomarkers of depression: diagnosis, treatment, and pathophysiology. Neu- ropsychopharmacology 2011;36(12):2375–2394.

5. Schneider B, Prvulovic D, Oertel-Knochel V, et al. Biomarkers for major depression and its delineation from neurodegenerative disorders. Prog Neurobiol 2011;95(4):703–717.

6. Rao U. Sleep disturbances in pediatric depression. Asian J Psychi- atr 2011;4(4):234–247.

7. Pillai V, Kalmbach DA, Ciesla JA. A meta-analysis of electroen- cephalographic sleep in depression: evidence for genetic biomark- ers. Biol Psychiatry 2011;70(10):912–919.

8. Stetler C, Miller GE. Depression and hypothalamic-pituitary- adrenal activation: a quantitative summary of four decades of re- search. Psychosom Med 2011;73(2):114–126.

9. Rao U, Hammen CL, Poland RE. Risk markers for depression in adolescents: sleep and HPA measures. Neuropsychopharmacol- ogy 2009;34(8):1936–1945.

10. Hashimoto K. Brain-derived neurotrophic factor as a biomarker for mood disorders: an historical overview and future directions. Psychiatry Clin Neurosci 2010;64(4):341–357.

11. Hamilton JP, Chen MC, Gotlib IH. Neural systems ap- proaches to understanding major depressive disorder: an intrin- sic functional organization perspective. Neurobiol Dis 2013;52: 4–11.

12. Fu CH, Steiner H, Costafreda SG. Predictive neural biomarkers of clinical response in depression: a meta-analysis of functional

and structural neuroimaging studies of pharmacological and psy- chological therapies. Neurobiol Dis 2013;52:75–83.

13. Miller GE, Cole SW. Clustering of depression and inflammation in adolescents previously exposed to childhood adversity. Biol Psy- chiatry 2012;72(1):34–40.

14. Hood KK, Lawrence JM, Anderson A, et al. Metabolic and in- flammatory links to depression in youth with diabetes. Diabetes Care 2012;35(12):2443–2446.

15. Pallavi P, Sagar R, Mehta M, et al. Serum neurotrophic factors in adolescent depression: gender difference and correlation with clinical severity. J Affect Disord 2013.

16. Hulvershorn LA, Cullen K, Anand A. Toward dysfunctional con- nectivity: a review of neuroimaging findings in pediatric major depressive disorder. Brain Imaging Behav 2011;5(4):307–328.

17. Talati A, Weissman MM, Hamilton SP. Using the high-risk family design to identify biomarkers for major depression. Philos Trans R Soc Lond B Biol Sci 2013;368(1615):20120129.

18. Pajer K, Andrus BM, Gardner W, et al. Discovery of blood tran- scriptomic markers for depression in animal models and pilot val- idation in subjects with early-onset major depression. Transl Psy- chiatry 2012;2:e101.

19. Takizawa R, Fukuda M, Kawasaki S, et al. Neuroimaging-aided differential diagnosis of the depressive state. Neuroimage 2013.

20. Souslova T, Marple TC, Spiekerman AM, Mohammad AA. Per- sonalized medicine in Alzheimer’s disease and depression. Con- temp Clin Trials 2013.

Depression and Anxiety

Copyright of Depression & Anxiety (1091-4269) is the property of John Wiley & Sons, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.