Documentos de Académico
Documentos de Profesional
Documentos de Cultura
Metabolism
www.metabolismjournal.com
A R T I C LE I N FO
AB S T R A C T
Keywords:
Allostatic load
population health. They facilitate tapping meaningfully into the complex, non-linear
Mood disorders
interactions that affect the brain and multiple systems of the body and promote adaptation
Hippocampus
or, when dysregulated, they can accelerate disease processes. This has demanded a
Amygdala
Prefrontal cortex
adequately describe and predict how an individual embedded in a particular social and
physical environment, and with a unique genotype and set of lifetime experiences, will fare
in terms of health and disease risk, as well as how that individual will respond to an
intervention. Yet, at the same time, single biomarkers can have a predictive or diagnostic
value when combined with carefully designed longitudinal assessment of behavior and
disease related to stress. Moreover, the methods of brain imaging, themselves the reflection
of the complexity of brain functional architecture, have provided new ways of diagnosing,
and possibly differentiating, subtypes of depressive illness and anxiety disorders that are
precipitated or exacerbated by stress. Furthermore, postmortem assessment of brain
biomarkers provides important clues about individual vulnerability for suicide related to
depression and this may lead to predictive biomarkers to better treat individuals with
suicidal depression. Once biomarkers are available, approaches to prevention and
treatment should take advantage of the emerging evidence that activating brain plasticity
together with targeted behavioral interventions is a promising strategy.
2015 Elsevier Inc. All rights reserved.
1.
Introduction
M ET AB O LI S M CL IN I CAL AN D EX PE RI ME N TAL 6 4 ( 2 01 5 ) S 2 S 10
2.
S3
S4
3.
How can the biomarkers collected in living
subjects be analyzed and used as a predictive tool?
For assessing and using the allostatic load battery, the
simplest approach for prediction of outcomes was to divide
values for each marker into quartiles and score anyone with
markers in the most extreme quartile (e.g., highest quartile
for negative markers like blood pressure and IL-6; lowest
quartiles for positive markers like IGF-1 or DHEA). For 10
markers, for example, the highest score would be 10 and the
lowest, zero [23]. In later work, a metafactor model of AL as
an aggregate measure of six underlying latent biological
subfactors was found to fit the data, with the metafactor
structure capturing 84% of variance of all pair-wise associations
among biological subsystems [17,18]. Another method of
analysis involves recursive partitioning [27] which was used to
identify a set of pathways, composed of combinations of
different biomarkers, that were associated with and predictive
of a high-risk of mortality over the 12-year period. Of the 13
biomarkers examined, almost all entered into one or more highrisk pathways although combinations of neuroendocrine and
immune markers appeared frequently in high-risk pathways in
4.
M ET AB O LI S M CL IN I CAL AN D EX PE RI ME N TAL 6 4 ( 2 01 5 ) S 2 S 10
pressure, cholesterol levels, HDL and LDL, which do not tap into
all the important mediators and health-related indices but
which can be part of a broader array of biomarkers. This has
resulted in commercial and non-commercial efforts such as
ALLOSTATIX (http://www.allostatix.com/) and Optimal Health
and Prevention Research Foundation (http://optimalhealth.org/
index.php/about/), respectively, to apply this type of multifactorial analysis to programs intended to monitor and help
individuals improve health behaviors and overall health,
while at the same time attempting to use distortions of the
allostatic load profile (e.g. high inflammation or low cortisol or
high inflammation plus high cortisol, indicating glucocorticoid
resistance; high sympathetic and low parasympathetic tone) as
not only as predictors of disease risk, but also predictors of the
most effective treatment/prevention strategy. Time will tell if
these approaches, such as in Howard County, Maryland (http://
www.healthyhowardmd.org/healthy-howard), can catch on
and become useful. Such programs emphasize prevention
strategies using health coaches (http://en.wikipedia.org/wiki/
Health_coaching) and other methods that get individuals
involved in promoting a healthy lifestyle for themselves and
others, and the allostatic load battery may be able to provide a
longitudinal index of the success of such interventions.
5.
The diagnostic and predictive value of
structural and functional brain imaging as a biomarker
The brain is the central organ of stress and adaptation to
stressors via activity of brain regions that control behavior.
These brain regions also regulate systemic physiology
through autonomic, neuroendocrine, immune and metabolic
activity. The dysregulation of these interactions, and particularly the dysregulation of neural function, is a signature of
brain and body pathophysiology [11]. Both structural and
functional brain imaging are becoming useful as biomarkers.
Structural imaging of the brain (MRI) has been used to
assess the effects of regular physical activity and demonstrate
its ability to enlarge the hippocampus due to structural
remodeling, including neurogenesis [31]. Previous work by
the same group had shown that regular aerobic physical
activity increased blood flow in the prefrontal and parietal
cortices, measured by functional imaging, and enhanced
executive function [32,33]. Using longitudinal MRI of the
brains of individuals with anxiety disorders, mindfulness
based stress reduction was shown to decrease amygdala
volume in those patients who showed reduced chronic
anxiety [34]. Amygdala enlargement and hyperactivity are
features of chronic anxiety [35]. A similar approach was used
to show the efficacy of cognitive behavioral therapy for
chronic fatigue, showing increased gray matter volume of
the medial prefrontal cortex via MRI in those individuals
whose symptoms had improved with treatment [36].
Although complex and believed to measure cerebral blood
flow, functional brain imaging (fMRI) signatures have been
used to predict antidepressant outcome [25] and to subtype
anxiety disorders [24]. Patients with post-traumatic stress
disorder (PTSD) showed hyperactivity during processing of
emotions in the amygdala, parahippocampal gyrus, insula,
inferior parietal lobule, mid-cingulate, and precuneus, while
S5
patients with social anxiety disorder also showed hyperactivity in the amygdala, parahippocampal gyrus, and insula, but
also in the fusiform gyrus, globus pallidus, inferior frontal
gyrus, and superior temporal gyrus. In contrast, patients with
specific phobia displayed hyperactivity in the amygdala and
insula, as in the other disorders, but also in fusiform gyrus,
substantia nigra and mid-cingulate. Thus, hyperactivity was
observed in all three disorders in only two structuresthe
amygdala and the insula, and the other areas of activation
begin to provide a unique signature for each disorder.
Further differentiation was found during fear conditioning
of healthy subjects and subjects with different anxiety
disorders [24]. In healthy individuals, fear conditioning
increased activity in the amygdala and bilateral insula.
Compared to healthy subjects, less activation was seen only
in PTSD, specifically in the inferior occipital gyrus, ventromedial prefrontal cortex, rostral anterior cingulate cortex,
parahippocampal gyrus, lingual gyrus, dorsal amygdala and
anterior hippocampus, orbitofrontal cortex, putamen, middle
occipital gyrus, dorsomedial prefrontal cortex, dorsal anterior
cingulate cortex, and mid-cingulate. In fact, symptom severity
of PTSD was related to degree of hypoactivation in the medial
prefrontal cortex. Hyperactivation in the amygdala and
insular cortices of patients was found more frequently in
social anxiety disorder and specific phobia than in PTSD.
However, hypoactivation of rostral anterior cingulate cortex,
dorsal anterior cingulate cortex, and ventromedial prefrontal
cortex was seen more frequently in PTSD than in either social
anxiety disorder or specific phobia, and thalamic hypoactivity
was also more commonly seen in PTSD than either social
anxiety disorder or specific phobia [24]. Unfortunately, the
complexity of these patterns of activation and variability
between subjects with the particular disorder make it difficult
to use this as a routine diagnostic procedure [37].
In spite of the potential variability of brain imaging as
noted above, a new study using positron imaging tomography
(PET) for brain glucose metabolism has provided clues as to
brain regions, the activity of which may predict response or
lack of response to antidepressant therapies [25]. Brain
glucose metabolism was measured with PET prior to treatment randomization to either escitalopram oxalate or cognitive behavior therapy for 12 weeks. Patients who did not remit
on completion of their phase 1 treatment were offered
enrollment in phase 2 comprising an additional 12 weeks of
treatment with combination escitalopram and cognitive
behavior therapy [25]. What these authors found were both
positive and negative predictors of remission. Of six limbic
and cortical regions that were identified, the right anterior
insula showed the most robust discriminant properties across
groups in which insula hypometabolism (relative to wholebrain mean) was associated with remission to cognitive
behavior therapy and poor response to escitalopram, while
insula hypermetabolism was associated with remission to
escitalopram and poor response to cognitive behavior therapy. The insula is also a brain region where direct intracranial
brain stimulation has been shown to relieve depression [38].
Another use of imaging involves the diagnosis and
treatment of social anxiety disorders with cognitive behavioral therapy [39]. Response to angry vs neutral and emotional
vs neutral faces was used to probe individuals before CBT.
S6
According to the authors, Pretreatment responses significantly predicted subsequent treatment outcome of patients
selectively for social stimuli and particularly in regions of
higher-order visual cortex. Combining the brain measures
with information on clinical severity accounted for more than
40% of the variance in treatment response and substantially
exceeded predictions based on clinical measures at baseline.
Prediction success was unaffected by testing for potential
confounding factors such as depression severity at baseline.
Such successes indicate that proper use of brain imaging
biomarkers will continue to be very useful in clinical practice.
6.
Measuring functional activity of single
mediators and interpreting the results
In addition to the methods for dealing with complexity of
interacting mediators via the allostatic load battery, there are
measures of certain mediators that have predictive validity by
themselves, and a number of examples will illustrate this. The
first two examples concern cortisol, and the third involves
inflammatory mediators. First, low normal cortisol has been
linked to increased risk for PTSD in both animal models and
humans after trauma, and this has led to the use of glucocorticoids as a preventative treatment for the development of
PTSD symptoms [4043]. This highlights the important adaptive
and protective function of cortisol which is often measured and
interpreted in terms of negative consequences of stress. Indeed,
cortisol has many positive functions that include facilitating,
during acute stress, acquired immune responses [44] and
promoting plasticity at the synaptic structural level [45,46].
Second, failure to habituate cortisol responsiveness to a
repeated public speaking challenge is indicative of low self
esteem and poor locus of control and this was linked to a smaller
hippocampal volume [4749]. This assessment requires repeated
testing under a defined stressor and is of particular interest
because a smaller hippocampus and low self esteem are both
risk factors associated with PTSD [50,51]. By the same token,
repeated sampling of cortisol around the day is diagnostic for
whether an individual has a flat diurnal rhythm that is a factor in
such conditions as depression and burnout [52,53]. A flat cortisol
diurnal rhythm, as shown in animal models, is associated with a
sluggish turning-on and turning-off of the cortisol response to
stress, which is maladaptive [54,55]. Both of these measures of
cortisol emphasize a basic principle of allostasis, namely, that
efficiently turning on a stress sensitive mediator, in this case a
cortisol response, when needed, and efficiently turning it off
again when not needed, is a key to successful adaptation with
minimal allostatic load. Moreover, a normal diurnal cortisol
rhythm is important to program many behaviors and brain
functions, including an efficient HPA stress response [54].
Third, inflammatory markers have been very informative
when measured in the context of studies of early life adversity
(abuse, neglect by parents or caregivers in childhood). Along
with telomere shortening in the case of exposure to violence
[56], increased levels of inflammatory cytokines and Creactive protein (CRP), along with obesity as a secondary
outcome, are seen to persist into adulthood along with
increased incidence of depression, substance abuse, cardiovascular disease and diabetes [2,30,57,58]. Some hopeful
7.
Molecular biomarkers for translation and
reverse translation
There are also biomarkers that are only available after the
fact, e.g., as correlates of depression with suicide as determined post mortem in rigorously collected and preserved
human brain samples. These may provide a better understanding of the etiology of depressive illness and lead to
preventative treatments. For depression that led to suicide,
low levels of fibroblast growth factor 2 (FGF2) and FGF receptor
2 were found in the anterior cingulate (ACC) or dorsolateral
prefrontal cortex (DLPFC), along with reduced levels of FGF1,
FGFR2, and FGFR3 and elevated levels of FGF9 and FGF12
[60,61]. Similar changes were found in other brain regions,
such as hippocampus where FGFR2 was reduced even as FGF2
itself was elevated, and animal models that show depressiveand anxiety-like behaviors also show similar patterns. Indeed, administration of FGF2 in adults, or neonatally in those
animal models, protects against these behaviors [60,62].
Furthermore, using the same rigorously collected brain
samples as in the FGF studies, a pathway analysis revealed
alterations of serotonergic signaling in the DLPFC and
alternations of glucocorticoid signaling in the ACC and NAcc
[63]. Also deficient in suicide brains was the 5-HT2A gene in
the DLPFC, previously associated both with suicide and mood
disorders. Moreover, in the ACC, 6 metallothionein genes
were down-regulated in suicide (MT1E, MT1F, MT1G, MT1H,
MT1X, MT2A) and three were down-regulated in the NAcc in
suicide (MT1F, MT1G, MT1H). Thus, 5-HT2A and members of
the metallothionein subfamilies MT 1 and 2, which are downregulated in suicide completers, are normally protective
factors following stress and glucocorticoid stimulation [63].
In order to fully utilize the power of such post-mortem
biomarker studies in prevention of suicide and better treatment
of depression, humanized animal models with alleles of
genes that represent vulnerability for depression, such as the
BDNF val66met alleles, can provide important insights into
plasticity of neural architecture in relation to anxiety-like
behavior and the efficacy of antidepressant medications [64,65].
8.
M ET AB O LI S M CL IN I CAL AN D EX PE RI ME N TAL 6 4 ( 2 01 5 ) S 2 S 10
S7
9.
This article has examined how biomarkers are being used for
diagnosis and treatment of stress-related mental and physical health disorders. Moreover, studies on postmortem brain
tissue are showing that certain gene products can be useful to
better understand brain mechanisms, hopefully leading to
treatment of suicidal depression. This article has also noted
that, in some cases, single biomarkers can be very informative
when measured in the context of careful longitudinal
behavioral assessment. Examples include the use of cortisol
in relation to diagnosing the physiological consequence of
low self-esteem and vulnerability for PTSD, and for PTSD's
preventative treatment. Furthermore, inflammatory cytokines,
obesity and telomere length are useful biomarkers in relation to
the impact of adverse childhood experiences [56,57].
Yet, even if only one or just a few biomarkers are
measured, it should be kept in mind that in all of these
pathophysiological conditions, there are perturbations of the
whole network of allostasis, which is indicative of systemic
allostatic load affecting many body systems and functions
concurrently. For multi-systems interactions of biological
mediators of stress and adaptation, it is important, therefore,
S8
[15]
[16]
[17]
[18]
[19]
REFERENCES
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
M ET AB O LI S M CL IN I CAL AN D EX PE RI ME N TAL 6 4 ( 2 01 5 ) S 2 S 10
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
S9
S10
[64]
[65]
[66]
[67]
[68]
[69]
[70]
[71]
[72]
[73]
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
[86]
[87]
[88]
[89]
[90]
[91]
[92]
[93]