International Journal of Neuropsychopharmacology (2014), 17, 491–495.

© CINP 2013 TRENDS AND

doi:10.1017/S1461145713001107
PERSPECTIVES

Clinical staging: a necessary step in the

development of improved animal models
of mood disturbance?

Frederick Rohan Walker1,2*, Morgan H. James1,2, Ian B. Hickie3 and Patrick D. McGorry4
1
School of Biomedical Sciences and Pharmacy, University of Newcastle, Callaghan, NSW, Australia
2
Centre for Translational Neuroscience and Mental Health Research, Hunter Medical Research Institute, NSW, Australia
3
Brain and Mind Research Institute, The University of Sydney, Camperdown, NSW, Australia
4
Department of Psychiatry, ORYGEN Youth Health Research Centre, University of Melbourne, VIC, Australia

Abstract
Recently, it has been suggested that the clinical staging approach be considered a serious alternative framework
for conceptualising mood related psychopathology. The fundamental difference between clinical staging and the
now dominant categorical diagnostic framework is that the entire illness trajectory becomes relevant, as opposed
to simply the end-stage. The concept of disease trajectory has significant implications for animal models of psy-
chopathology, and particularly for animal models of depression. This article will introduce and discuss the impli-
cations of the clinical staging approach for those undertaking research using animal models of mood disturbance.

Received 11 April 2013; Reviewed 13 May 2013; Revised 26 August 2013; Accepted 26 August 2013;
First published online 16 October 2013
Key words: Animal models of depression, clinical staging, depression, mood disorders.

Clinical staging: implications for animal models of the clinical staging approach is that the entire illness
trajectory becomes relevant, as opposed to simply the
The recent release of the 5th edition of Diagnostic and
end-stage – which remains the primary focus of existing
Statistical Manual of Mental Disorders (DSM-5) has pro-
diagnostic approaches (McGorry et al., 2006). The clinical
voked considerable debate on the strengths and weak-
aim of the staging concept is to guide the deployment
nesses of our traditional approaches to clinical and
of stage-appropriate (and often early-phase) interventions
research diagnoses. One of the strongest themes to
that have the potential to slow or halt progression to more
emerge is the view that existing diagnostic criteria are
advanced stages.
fundamentally limited by three major factors. First, they
By its nature, a clinical staging approach endorses a
are based largely on symptoms presented only by those
more dimensional view of psychiatric illness, adding the
with severe or persisting illness. While these ‘end-stage’
variables of time course and disease progression to the
phenotypes are representative of a subset of patients,
traditional dimension of illness severity. A significant
they may have very limited application outside of
feature of the staging approach is the recognition that
highly-specialised clinical settings. Second, little con-
the initial stage of major psychotic and mood disorders
sideration is given to those early and often sub-threshold
is phenotypically similar (Hafner et al., 2005). Indeed,
symptoms that are more characteristic of the onset
because of the degree of phenotypic overlap it has been
phase rather than the persistent course of an illness.
suggested that the initial phase of illness be considered
Third, the validity of the proposed links between the vari-
as a ‘pluripotential’ rather than disease-specific ‘at risk’
ous symptom-sets and any underlying or differentiating
stage (McGorry, 2011). While the clinical utility of adopt-
pathophysiologies remains largely unresolved (McGorry
ing this approach is being evaluated (McGorry, 2008),
et al., 2006).
the broad implications that it has for development of
An alternative view proposed independently by
relevant animal models for mood disturbance need to
McGorry et al. (2006) and Insel (2007) contends that a
be considered. To date, most animal models of mood
clinical staging approach is essential for both enhanced
disturbance have been designed without giving serious
clinical practice and the development of more meaning-
consideration to neither what stage of illness has been
ful clinical and basic research. A fundamental difference
modelled nor the fact that different phenomena may
emerge at different stages of illness. Indeed, a poor fit
between the stage of illness and the stage captured by
* Address for correspondence: Dr F. R. Walker, Room MS306, Medical
the animal model may have contributed to the clinical
Sciences Building, University of Newcastle, Callaghan, NSW 2308,
Australia.
failure of several compounds that had exhibited signifi-
Tel.: +61 2 4921 5012 Fax: +61 2 4921 5141 cant promise during pre-clinical testing (Cryan and
Email: Rohan.Walker@newcastle.edu.au Slattery, 2007).
492 F. R. Walker et al.
Animal models of mood disturbance – what stage are
they at?
There are now a large number of experimental models
of depression in rodents. The central feature of almost
all of these models is their reliance on repeated exposure
to aversive experiences. Amongst the most commonly
used models are: repeated social defeat (Berton et al.,
2006; Tsankova et al., 2006), repeated social disruption
(Herzog et al., 2009), repeated corticosterone adminis-
tration (Gregus et al., 2005; Zhao et al., 2008), repeated
restraint stress (Tynan et al., 2010; Hinwood et al., 2012)
and chronic mild stress (Bortolato et al., 2007; Goshen
et al., 2008). Each of these models in turn can be combined
with developmentally earlier interventions to model early
life and adulthood interactions (Hodgson et al., 2001;
Gardner et al., 2005). Many of these models are considered,
in the presence of appropriately supportive data, to pro-
vide valid approximations of the human condition as
they produce a symptom set that is highly analogous
to that seen individuals with diagnosable depression.
Although all symptoms are not routinely measured in all
experiments, there are numerous instances in the literature
where any of the aforementioned models can be shown
to produce robust levels of anhedonia, psychomotor re-
tardation, working memory deficits, anorexia and disturb-
ances in sleep (Willner, 1997, 2005; Yan et al., 2010).
Despite the undeniable ability of standard pre-clinical
models of mood psychopathology to elicit what are rou-
tinely referred to as ‘depression-like’ symptoms, it is not
totally clear that this interpretation is as straightforward
as it seems. Specifically, some interpretational uncertainty
arises because numerous clinical studies have demon-
strated that many symptoms present in patients that are
diagnosed with depression (anhedonia, sleep disturb-
ance, irritability, and cognitive disturbance) are also
present in the prodromal (pre-diagnosis) stage of the
condition.
The concept of consistent prodromal symptoms is not
new, and has been extensively investigated, principally
due to interest in using prodromal symptoms to predict
the emergence of subsequent illness (Hetrick et al., 2008;
Kovacs and Lopez-Duran, 2010; Szklo-Coxe et al., 2010;
Fusar-Poli et al., 2013). Particularly notable amongst this
literature is Fava’s (1990) early study highlighting the
consistency with which general anxiety, loss of interest,
impaired work performance, fatigue and insomnia
appear in individuals that subsequently go on to develop
depression. If features such as anhedonia are present in
the pre-diagnosis and the post-diagnosis phase of major
depression, the critical question for those using standard
pre-clinical models and measuring anhedonic behaviour
is: which phase of the disorder is this measured symptom
actually occurring in?
The mere presence of certain behavioural changes
is not the only reason that current animal models are
considered to be approximations of the diagnosable
condition. Many working within the field consider it to
be critical that animals exposed to a particular model
also respond in a manner that approximates the way in
which patients respond to medication in clinical settings.
For example, the validity of a given animal model of
depression is predicated not only on its ability to elicit
‘depression-like symptoms’, such as anhedonia, but on
whether animals display a marked reduction in this and
other symptoms when treated with a selective serotonin
reuptake inhibitor or serotonin-noradrenalin reuptake
inhibitor (or another accepted antidepressant). Consistent
with this thinking, there are now many dozens of articles
attesting to the fact that the behavioural disturbances
exhibited by animals who have been subjected to a stan-
dard model of depression exhibit a marked reduction in
these symptoms following treatment with a conventional
antidepressant, such as a selective serotonin reuptake
inhibitor (Christiansen et al., 2011; Razzoli et al., 2011;
Gumuslu et al., 2013).
Significantly, however, we could not find any clinical
literature describing the use of anti-depressants during
the pre-diagnosable early stage of what would become
major depression. Therefore it is not clear that a reduction
in symptoms when treated with an accepted antidepress-
ant is in actual fact a measure of end-stage illness. It is
possible that such a reversal may also occur in early
stage illness. To be certain that the behaviour(s) exhibited
by an animal following exposure to a standard experimen-
tal model of depression are genuinely analogous to those
seen in individuals with depression, evidence should
also be available demonstrating that the clinical and pre-
clinical pre-diagnosis phase of the disorder do not
respond in a similar way to antidepressants. Obtaining
such evidence at the present time represents a major chal-
lenge, as metrics capable of effectively discriminating
between stage progressions have yet to be developed.
The need to develop indices of stage progression
The development and introduction of metrics or decision
matrices that would allow researchers to determine how
far a disorder had progressed would be enormously ben-
eficial. Perhaps the best outcome would be that each
stage, from moderate behavioural perturbation through
to a diagnosable condition, could be uniquely and categ-
orically discriminated from the next, or in lieu of this, a
probabilistic estimate could be ascribed to stage pro-
gression. Unfortunately, without extensive experimental
work it is difficult to know what the best indicators of
stage progression will be. Clearly, evaluating the extent
to which commonly used and well understood metrics
(e.g. sucrose preference) show graduated, intensity-
dependent changes is an obvious place to begin investi-
gations. Ideally, if sucrose preference did change in a
graduated intensity-dependent manner, it may be poss-
ible to ascribe particular stages to certain changes in
preference scores.

Alternatively, in terms of developing stage metrics
capable of differentiating prodromal-like behaviour
from diagnosable-like behaviour, it may be useful to
look at the approach taken with clinical diagnosis.
One of the key aspects of clinically diagnosing the pres-
ence of major depression is to identify the continued pres-
ence of certain debilitating symptoms for a given period
of time (DSM-IV-TR; APA, 2000). At present, the vast
majority of animal models assess symptom severity either
during the application of aversive experience or immedi-
ately after its termination. We could find very few studies
that tracked how persistent the depression-like symptoms
were following the cessation of aversive experiences.
While the suggestion may carry some anthropomorphic
bias, it does not seem unreasonable to expect that if a
diagnosable condition is being modelled then symptoms
should not rapidly resolve following the removal of the
aversive stimulus. If rapid symptom dissipation was
observed using standard models it may be taken to
suggest that they elicit more prodromal-like behaviour.
While speculative, a straightforward response to this situ-
ation would be to employ repeat cycles of aversive experi-
ence, arguably providing a more accurate representation
of the life history of those that do eventually develop
depression (Kendler et al., 1993; Chapman et al., 2004).
One of the most significant issues in terms of moving
forward with the development of metrics capable of
detecting modest behavioural disturbances is sample
size. Much of the work undertaken thus far using animal
models of mood disturbance has made use of group sizes
of between 6 and 10 animals per group. This modest
number reflects the fact most studies have only been pow-
ered to detect relatively large between-group differences.
Given the typical variability observed in studies of animal
behaviour, it is likely that sample size will have to be
increased significantly in order to robustly identify
small differences in behaviour that may occur during
early phases of disease progression.
Whatever ‘stage-metrics’ are eventually employed,
hopefully they will confirm, as many have long assumed,
that existing animal models of psychopathology, and par-
ticularly animal models of depression, are reasonable
models of the end-stage clinical condition. We must, how-
ever, be open to the possibility that existing models may
represent earlier stages of disease progression than we
had originally anticipated.
Are new models required or can existing models be
repurposed to accommodate the staging concept?
Application of the staging concept to preclinical model-
ling of mood disorders raises questions about what
models should be used. Can existing models of mood
disturbance, such as the repeated restraint stress model,
be pressed into service by simply evaluating change
earlier in the environmental insult process? If not, it
may be the case that specific early (or later) stage models
Clinical staging for animal models 493
may be more appropriate, models that focus more on the
application of specific risk or time-limited factors (e.g.
acute stress, sleep deprivation, shifting circadian timing,
exposure to infection) or manipulation of quite specific
developmental, pharmacological or other biological
pathways. The challenge, however, inherent with the
introduction of ‘early-stage’ models is that the emergent
symptoms may only be loosely associated (pheno-
typically) with major depression in humans. Further
complicating matters, and mirroring clinical reality,
early-stage models could produce symptoms of mood
disturbance that may in fact be linked equally to other
psychiatric syndromes (McGorry, 2011).
Conclusion
Perhaps the greatest utility of animal models to date has
been their role in testing the potential therapeutic capacity
of novel pharmacological agents. It is now apparent that a
range of compounds that appeared to have therapeutic
value in animal models of depression has not translated
well into the clinic (Cryan and Slattery, 2007). There are
a variety of possible explanations for this phenomenon,
but one that has not been widely considered may well
be a poor fit between the stage of illness in most human
subjects (i.e. chronic or recurrent illness) and the stage
captured by the animal model.
Fortunately, a firm platform for clinical research
in major mood disorders based on the clinical staging
approach has now been established (McGorry et al.,
2006). The implications of the staging model for the sub-
sequent development of relevant animal models of
depression, and potentially other mood disorders, are
profound. It will be necessary, first, to state what stage
of illness (and observed behaviour or physiological
change) is produced in response to a given risk factor
and, second, what types of therapeutic strategies are rel-
evant to the various stages of illness progression.
Importantly, without the development, introduction and
evaluation of stage metrics it will be difficult to fully
understand how representative the information is that is
currently being generated from existing animal models
of mood disturbance. Ultimately, continued development
of the staging concept in the field of pre-clinical models
of mood disorders has the potential to increase the chance
of successful translation of therapies from the lab to
personalised clinical practice.
Acknowledgments
This work was supported by project grants from
the National Health and Medical Research Council of
Australia and the Hunter Medical Research Institute.
Statement of Interest
F.R.W. currently receives research support from the
National Health and Medical Research Council of
494 F. R. Walker et al.
Australia (NHMRC). M.H.J. has no interests to declare.
I.B.H. Sits on the National Advisory Council on Mental
Health, which reports to the Federal Minister for Mental
Health, is a member of BUPA Australia and a Headspace
board member. I.B.H. was previously chief executive
officer and clinical adviser of Beyondblue, an Australian
National Depression Initiative. He has led projects for
health professionals and the community supported by
governmental, community agency and drug industry part-
ners (Wyeth, Eli Lily, Servier, Pfizer, Astra-Zeneca) for the
identification and management of depression and anxiety.
He has served on advisory boards convened by the drug
industry in relation to specific antidepressants, including
nefazodone, duloxetine, and desvenlafaxine and has par-
ticipated in a multi-centre clinical trial of agomelatine
effects on sleep architecture in depression. I.B.H. is also
supported by a National Health and Medical Research
Council Australian Medical Research Fellowship. He is
a participant in a family-practice-based audit of sleep
disturbance and major depression, supported by Servier,
the manufacturers of agomelatine. Research studies
done by I.B.H. are mainly funded by NHMRC project
and programme grants.
P.M. Currently receives research support from the
NHMRC and the Colonial Foundation.
References
American Psychiatric Association (2000) Diagnostic and
statistical manual of mental disorders, 4th edn, Text Revision.
American Psychiatric Publishing. Washington, DC.
Berton O, McClung CA, Dileone RJ, Krishnan V, Renthal W,
Russo SJ, Graham D, Tsankova NM, Bolanos CA, Rios M,
Monteggia LM, Self DW, Nestler EJ (2006) Essential role of
BDNF in the mesolimbic dopamine pathway in social defeat
stress. Science 311:864–868.
Bortolato M, Mangieri RA, Fu J, Kim JH, Arguello O, Duranti A,
Tontini A, Mor M, Tarzia G, Piomelli D (2007)
Antidepressant-like activity of the fatty acid amide hydrolase
inhibitor URB597 in a rat model of chronic mild stress. Biol
Psychiatry 62:1103–1110.
Chapman DP, Whitfield CL, Felitti VJ, Dube SR, Edwards VJ,
Anda RF (2004) Adverse childhood experiences and the risk
of depressive disorders in adulthood. J Affect Disord
82:217–225.
Christiansen SH, Olesen MV, Wortwein G, Woldbye DP
(2011) Fluoxetine reverts chronic restraint stress-induced
depression-like behaviour and increases neuropeptide Y
and galanin expression in mice. Behav Brain Res 216:585–591.
Cryan JF, Slattery DA (2007) Animal models of mood disorders:
recent developments. Curr Opin Psychiatry 20:1–7.
Fava GA, Grandi S, Canestrari R, Molnar G (1990) Prodromal
symptoms in primary major depressive disorder. J Affect
Disord 19:149–152.
Fusar-Poli P, Yung AR, McGorry P, van Os J (2013) Lessons
learned from the psychosis high-risk state: towards
a general staging model of prodromal intervention.
Psychol Med 1–8. doi: 10.1017/S0033291713000184.
Gardner KL, Thrivikraman KV, Lightman SL, Plotsky PM,
Lowry CA (2005) Early life experience alters behavior during
social defeat: focus on serotonergic systems. Neuroscience
136:181–191.
Goshen I, Kreisel T, Ben-Menachem-Zidon O, Licht T,
Weidenfeld J, Ben-Hur T, Yirmiya R (2008) Brain interleukin-1
mediates chronic stress-induced depression in mice via
adrenocortical activation and hippocampal neurogenesis
suppression. Mol Psychiatry 13:717–728.
Gregus A, Wintink AJ, Davis AC, Kalynchuk LE (2005) Effect
of repeated corticosterone injections and restraint stress on
anxiety and depression-like behavior in male rats. Behav Brain
Res 156:105–114.
Gumuslu E, Mutlu O, Sunnetci D, Ulak G, Celikyurt IK,
Cine N, Akar F (2013) The effects of Tianeptine, Olanzapine
and Fluoxetine on the cognitive behaviors of unpredictable
chronic mild stress-exposed mice. Drug Res (Stuttg)
doi: 10.1055/s-0033-1347237.
Hafner H, Maurer K, Trendler G, an der Heiden W,
Schmidt M (2005) The early course of schizophrenia
and depression. Eur Arch Psychiatry Clin Neurosci
255:167–173.
Herzog CJ, Czeh B, Corbach S, Wuttke W, Schulte-Herbruggen O,
Hellweg R, Flugge G, Fuchs E (2009) Chronic social
instability stress in female rats: a potential animal model
for female depression. Neuroscience 159:982–992.
Hetrick SE, Parker AG, Hickie IB, Purcell R, Yung AR,
McGorry PD (2008) Early identification and intervention
in depressive disorders: towards a clinical staging model.
Psychother Psychosom 77:263–270.
Hinwood M, Morandini J, Day TA, Walker FR (2012) Evidence
that microglia mediate the neurobiological effects of chronic
psychological stress on the medial prefrontal cortex. Cereb
Cortex 22:1442–1454.
Hodgson DM, Knott B, Walker FR (2001) Neonatal endotoxin
exposure influences HPA responsivity and impairs tumor
immunity in Fischer 344 rats in adulthood. Pediatr Res
50:750–755.
Insel TR (2007) The arrival of preemptive psychiatry. Early Interv
Psychiatry 1:5–6.
Kendler KS, Kessler RC, Neale MC, Heath AC, Eaves LJ
(1993) The prediction of major depression in women:
toward an integrated etiologic model. Am J Psychiatry
150:1139–1148.
Kovacs M, Lopez-Duran N (2010) Prodromal symptoms and
atypical affectivity as predictors of major depression in
juveniles: implications for prevention. J Child Psychol
Psychiatry 51:472–496.
McGorry P (2011) Transition to adulthood: the critical period
for pre-emptive, disease-modifying care for schizophrenia
and related disorders. Schizophr Bull 37:524–530.
McGorry PD (2008) Is early intervention in the major psychiatric
disorders justified? Yes. BMJ 337:a695. doi: http://dx.doi.org/
10.1136/bmj.a695.
McGorry PD, Hickie IB, Yung AR, Pantelis C, Jackson HJ (2006)
Clinical staging of psychiatric disorders: a heuristic framework
for choosing earlier, safer and more effective interventions.
Aust N Z J Psychiatry 40:616–622.
Razzoli M, Carboni L, Andreoli M, Michielin F, Ballottari A,
Arban R (2011) Strain-specific outcomes of repeated social
defeat and chronic fluoxetine treatment in the mouse.
Pharmacol Biochem Behav 97:566–576.
Szklo-Coxe M, Young T, Peppard PE, Finn LA, Benca RM (2010)
Prospective associations of insomnia markers and symptoms

with depression. Am J Epidemiol 171:709–720.
Tsankova NM, Berton O, Renthal W, Kumar A, Neve RL,
Nestler EJ (2006) Sustained hippocampal chromatin regulation
in a mouse model of depression and antidepressant action.
Nat Neurosci 9:519–525.
Tynan RJ, Naicker S, Hinwood M, Nalivaiko E, Buller KM,
Pow DV, Day TA, Walker FR (2010) Chronic stress alters the
density and morphology of microglia in a subset of stress-
responsive brain regions. Brain Behav Immun 24:1058–1068.
Willner P (1997) Validity, reliability and utility of the
chronic mild stress model of depression: a 10-year
Clinical staging for animal models 495
review and evaluation. Psychopharmacology (Berl)
134:319–329.
Willner P (2005) Chronic mild stress (CMS) revisited: consistency
and behavioural-neurobiological concordance in the effects of
CMS. Neuropsychobiology 52:90–110.
Yan HC, Cao X, Das M, Zhu XH, Gao TM (2010)
Behavioral animal models of depression. Neurosci Bull
26:327–337.
Zhao Y, Ma R, Shen J, Su H, Xing D, Du L (2008) A mouse model
of depression induced by repeated corticosterone injections.
Eur J Pharmacol 581:113–120.