Journal of Psychiatric Research 46 (2012) 667e674

Contents lists available at SciVerse ScienceDirect

Journal of Psychiatric Research

journal homepage: www.elsevier.com/locate/psychires

Apathy in currently nondepressed patients treated with a SSRI for a major

depressive episode: Outcomes following randomized switch to either duloxetine
or escitalopram
Joel Raskin a, *, Thomas George b, Renee E. Granger c, Nadia Hussain a, George Weizhong Zhao c,
Lauren B. Marangell d, e
a
Lilly Research Laboratories, Eli Lilly Canada, 3650 Danforth Ave, Toronto, ON, Canada M1N 2E8
b
North West Specialist Centre, Everton Park, QLD, Australia
c
Development Centre of Excellence Asia-Pacific, Eli Lilly Australia Pty Ltd, Macquarie Park, NSW, Australia
d
Eli Lilly and Company, Lilly USA, LLC, Indianapolis, IN, USA
e
University of Texas Health Science Center School of Medicine, Houston, TX, USA

a r t i c l e i n f o a b s t r a c t

Article history: Apathy in the context of treated major depressive disorder (MDD) is a common but understudied
Received 10 August 2011 symptom. This multicenter, double-blind, randomized study investigated whether switching from
Received in revised form a selective serotonin reuptake inhibitor (SSRI) to a serotonin-norepinephrine reuptake inhibitor (SNRI),
6 February 2012
compared with switching to another SSRI, improved apathy symptoms in patients who had been treated
Accepted 17 February 2012
with a SSRI for MDD for !3 months, were no longer depressed (Montgomery-Åsberg Depression Rating
Scale [MADRS] total score "15), and continued to have apathy (Apathy Evaluation Scale e Clinician rated
Keywords:
version [AES-C] total score >30). Following 8 weeks of treatment, both the duloxetine (SNRI, 244
Apathy
Major depressive disorder
patients) and escitalopram (SSRI, 239 patients) groups significantly improved from baseline on the AES-C
Duloxetine hydrochloride total score (least squares mean change [standard error]: duloxetine #13.9 [0.54]; escitalopram #13.5
Escitalopram [0.54], both P < 0.001), and on the secondary apathy, depression, and functional outcomes. There were
Norepinephrine no significant differences between the two groups on any measure, including AES-C total score (least
Serotonin squares mean difference [95% confidence interval]: #0.4 [#1.87 to 1.10], P ¼ 0.612; primary objective).
There was a significant within-group improvement in apathy in the subgroup who received escitalopram
before and during the study. There were few differences in safety between the two groups. This study did
not support the hypothesis that switching from a SSRI to a SNRI has a beneficial effect on apathy
symptoms. However, given the study limitations, it is possible that more specific targeting of the
noradrenergic pathway would be of benefit.
! 2012 Elsevier Ltd. All rights reserved.

1. Introduction respectively, of 161 patients who had completed a course of treat-

Residual symptoms in the context of “successful” antidepressant
treatment for major depressive disorder (MDD) are increasingly
recognized as a frequent occurrence (McClintock et al., 2011; Fava
et al., 2006). These symptoms are relevant because of the acute
discomfort felt by the patient and because they predict a greater
likelihood of poor outcomes overall, including relapse (Paykel et al.,
1995) and treatment discontinuation (Demyttenaere and Jaspers,
2008; Bolling and Kohlenberg, 2004). Bolling and Kohlenberg
(2004) reported apathy and loss of ambition in 18.6% and 16.1%,
* Corresponding author. Tel.: þ1 416 699 7260; fax: þ1 416 699 7352.
E-mail addresses: Raskin_Joel@lilly.com, joel.raskin@lilly.com (J. Raskin).
ment for depression with a selective serotonin reuptake inhibitor
(SSRI). In a cross-sectional study, Fava et al. (2006) reported apathy in
32.5% of 117 depressed patients who were in remission or partial
remission following at least 3 months of antidepressant treatment for
MDD. Residual symptoms may be a remnant of the original depres-
sion, a side effect of the antidepressant, or perhaps both. While apathy
may occur in the context of treatment with any antidepressant, most
reports (all small in number) have been in the context of SSRI treat-
ment, with speculation that this effect is treatment-emergent
(Marangell et al., 2002; Demyttenaere and Jaspers, 2008).
Apathy has been associated with reduced activity of the
noradrenergic system (Birkmayer and Riederer, 1989; Gold and
Chrousos, 1999). In addition, noradrenergic activity is modulated
by serotonin (5-hydroxytryptamine, 5-HT) such that unopposed

0022-3956/$ e see front matter ! 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jpsychires.2012.02.010
668 J. Raskin et al. / Journal of Psychiatric Research 46 (2012) 667e674
serotonergic manipulation may decrease noradrenergic activity
(Segal, 1979), possibly via stimulation of 5-HT 2C receptors (Gobert
et al., 2000). Therefore, increasing noradrenergic activity by
switching to a serotonin-norepinephrine reuptake inhibitor (SNRI)
is hypothesized to result in a greater improvement in apathy than
switching to another SSRI.
At present, there is no consensus on treating apathy in the
context of MDD. In order to assess the acute benefits of switching to
a SNRI, the current study was conducted in patients with apathy
who were nondepressed following at least 3 months of treatment
with a SSRI for a major depressive episode. The primary objective
was to compare the mean baseline to endpoint change in the
Apathy Evaluation Scale e Clinician rated version (AES-C) total
score between patients receiving duloxetine or escitalopram after 8
weeks of treatment. Secondary objectives included comparison
between the two treatment groups of mean changes from baseline
scores and/or longitudinal profiles in a variety of apathy, depres-
sion, and functional measures, and comparison of the safety and
tolerability of duloxetine and escitalopram.
2. Materials and methods
2.1. Study information
This was a multicenter, double-blind, randomized study con-
ducted in eight countries (Australia, Canada, China, Italy, Mexico,
Korea, Russia, Taiwan) between September 2009 and December
2010. The study was conducted in accordance with the latest
version of the Declaration of Helsinki and Good Clinical Practice
guidelines. The study design was reviewed and approved by
appropriate institutional ethics review boards. Written informed
consent was obtained from the study participants after the nature
of the study procedures had been fully explained. The study was
registered at www.clinicaltrials.gov (NCT00985504).
2.2. Eligibility criteria
Eligible patients were outpatients aged !18 years with a histor-
ical diagnosis of MDD, as defined by the Diagnostic and Statistical
Manual of Mental Disorders, 4th edition (DSM-IV). Patients were
required to have received treatment with a SSRI (escitalopram,
sertraline, paroxetine, or citalopram) for MDD for at least 3 months
before randomization, and to no longer be experiencing an episode
of MDD, based on a Montgomery-Åsberg Depression Rating Scale
(MADRS) total score "15 and Item 1 (apparent sadness) score of <2
at the start and end of the Screening Period (Visits 1 and 2). Patients
were required to have an AES-C total score >30 at Visits 1 and 2. A
negative urine test for pregnancy at the time of enrollment and
agreement to use a reliable method of birth control was required for
females of child-bearing potential. Exclusion criteria included:
a previous lack of response to an adequate trial of duloxetine or
escitalopram within the past 12 months; any current or historical
DSM-IV diagnosis of mania, bipolar disorder, treatment-resistant
depression (defined as failure of two antidepressant trials of
adequate dose and duration), or psychosis; current suicide risk, as
assessed by the Mini International Neuropsychiatric Interview and
Columbia Suicide Severity Rating Scale; pregnant or breast-feeding;
and abnormal thyroid stimulating hormone concentration.
2.3. Study design
Eligible patients were randomly assigned (1:1) to double-blind
(double-dummy) duloxetine or escitalopram treatment. Randomi-
zation was achieved with an interactive voice response system
(IVRS) and was stratified by site and previous SSRI treatment
(escitalopram or other SSRI). The study consisted of four periods
that included 8 weeks of treatment after randomization followed
by an optional tapering period. Patients remained on their current
SSRI treatment during the Screening Period while their eligibility
for the study was evaluated. During the Acute Therapy Period,
patients were enrolled and randomized to switch from their
current SSRI treatment to duloxetine 60 mg/day or escitalopram
10 mg/day in a double-blinded manner for 1 week. During the 7-
week Optimization Period, patients who experienced a worsening
in their depressive symptoms (based on the clinical judgment of
the investigator) could have their doses optimized (duloxetine
120 mg/day or escitalopram 20 mg/day), at the discretion of the
investigator. This was followed by a 2-week Optional Taper Period,
during which patients could taper off their study treatment to
reduce the likelihood of discontinuation-emergent adverse events.
2.4. Study assessments
Investigator-assessed instruments included the AES-C (Marin
et al., 1991), the Rothschild Scale for Antidepressant Tachyphy-
laxis (RSAT) (Rothschild, 2008), the MADRS (Montgomery and
Åsberg, 1979), and the Clinical Global Impression of Severity (CGI-
Severity) scale (Guy, 1976). The AES-C (18 items) assesses cognitive,
behavioral, emotional, and other symptoms of apathy based on
verbal and nonverbal information provided by the patient. In the
current study, residual apathy was defined as an AES-C total score
of >30. The RSAT (7 items) assesses symptoms of apathy or
decreased motivation among depressed patients who have ach-
ieved symptomatic remission with antidepressant treatment. The
instrument consists of 6 self-report items (energy level, motivation
and interest, cognitive functioning, weight gain, sleep, and sexual
functioning) and affect (assessed by the interviewer). An RSAT score
of !7 has been hypothesized to reflect antidepressant tachyphy-
laxis (Rothschild, 2008). The MADRS (10 items) measures severity
of depressive mood symptoms. The MADRS Item 8 assesses the
patient’s inability to feel (through evaluation of their interest in
their surroundings or activities that normally give pleasure) and
their ability to react with adequate emotion to circumstances or
people. The CGI-Severity scale is used in the presence of the patient
to record the severity of illness at the time of assessment.
The patient-reported instruments included the Patient’s Global
Impression of Improvement (PGI-Improvement) scale (Guy, 1976),
the Massachusetts General Hospital Cognitive and Physical Func-
tioning Questionnaire (CPFQ) (Fava et al., 2009), and the Sheehan
Disability Scale (SDS) (Sheehan et al., 1998). The PGI-Improvement
scale rates the improvement of symptoms since the previous visit.
The CPFQ (7 items) evaluates cognitive and physical well being. This
questionnaire assesses motivation, wakefulness, energy, focus,
word-finding difficulty, and mental acuity. The SDS assesses the
functional impact of symptoms on several inter-related domains.
Patients are also asked to provide their impression of how many
days were lost or rendered unproductive during the past week
because of their symptoms.
2.5. Study outcomes
The primary endpoint was the comparison of the mean baseline
to endpoint change in AES-C total score between patients
randomized to switch treatment to duloxetine or escitalopram for 8
weeks. The secondary endpoints included comparison between the
duloxetine and escitalopram treatment groups on: mean changes
from baseline scores and longitudinal profiles for RSAT total score,
MADRS total and Item 8 scores, CGI-S rating scale, and PGI-I rating
scale; mean changes from baseline scores for CPFQ total score and
SDS total score; the proportion of patients who relapsed (defined as
 J. Raskin et al. / Journal of Psychiatric Research 46 (2012) 667e674
MADRS total score !16) and the time to relapse; and serious
adverse events (SAEs), treatment-emergent adverse events
(TEAEs), discontinuations due to adverse events, and vital signs.
2.6. Statistical analysis
Assuming an effect size of 0.28, the study was estimated to have
approximately 86% power for the primary objective with 240
patients in each treatment group using a two-sided test at an alpha
level of 0.05. Based on the estimation that approximately 4% of
patients would not have a postbaseline measure of AES-C, the
planned sample size was 500 patients. Patients who were switched
directly from escitalopram to escitalopram formed a control
subgroup. In addition, all patients who had received escitalopram
for at least 3 months before the study were excluded in a sensitivity
analysis of the primary objective. Efficacy analyses included all
patients with a baseline and at least one postbaseline observation.
Safety analyses included all randomized patients who received at
least one dose of study drug. The primary objective was assessed
using a mixed-effects model with repeated measures (MMRM)
approach. Mean changes from baseline in secondary efficacy
measures were analyzed by MMRM or by analysis of covariance
(ANCOVA). Effect sizes (ES) were calculated for apathy measures
using Cohen’s d. The percentages of patients in the two treatment
Fig. 1. Patient disposition.
669
groups who met the relapse criterion were compared at endpoint
using a last observation carried forward approach and the Fisher’s
exact test. Time to relapse was analyzed using the KaplaneMeier
method. The incidence of SAEs, TEAEs, and discontinuations due
to adverse events was compared between the two treatment
groups using Fisher’s exact test. Mean changes from baseline to last
observation for vital signs were assessed using an MMRM
approach. Statistical analyses were conducted using SAS Version
9.1.3 (SAS Institute, Cary, N.C.).
3. Results
3.1. Patient disposition
In total, 483 patients met the eligibility criteria and were
randomized in the study, with 244 and 239 patients randomized to
the duloxetine and escitalopram treatment groups, respectively
(Fig. 1). The majority of the randomized patients (407/483 patients,
84.3%) completed 8 weeks of treatment (Fig. 1).
3.2. Patient characteristics
Patient characteristics are summarized in Table 1. The median
age of the patients enrolled in the study was 45.0 years and
670 J. Raskin et al. / Journal of Psychiatric Research 46 (2012) 667e674

Table 1
Patient characteristics.

Characteristic Duloxetine (N ¼ 244) Escitalopram (N ¼ 239) Overall (N ¼ 483)

Median age, years 45.1 45.0 45.0
Minimum, maximum 18.7, 73.9 19.2, 78.0 18.7, 78.0
Gender, n (%)
Female 187 (76.6) 179 (74.9) 366 (75.8)
Male 57 (23.4) 60 (25.1) 117 (24.2)
Ethnicity, n (%)
African/African American 0 (0) 1 (0.4) 1 (0.2)
Asian 90 (36.9) 82 (34.3) 172 (35.6)
Caucasian 120 (49.2) 119 (49.8) 239 (49.5)
Native American/Alaskan Native 34 (13.9) 37 (15.5) 71 (14.7)
Native Hawaiian/Pacific Islander 1 (0.4) 0 (0) 1 (0.2)
Mean (SD) BMI, kg/m2 25.6 (5.56) 26.8 (7.03) 26.2 (6.36)
Previous treatment, n (%)
Escitalopram 67 (27.5) 60 (25.1) 127 (26.3)
Other SSRI 177 (72.5) 179 (74.9) 356 (73.7)
Median age at first MDD episode, years (range) 34.0 (6.0e72.4) 34.7 (8.0e76.6) 34.2 (6.0e76.6)
Previous episodes of MDD
Experienced one or more, n (%) 175 (71.7) 182 (76.2) 357 (73.9)
Number of previous episodes, median (range) 1.0 (0e63.0) 2.0 (0e12.0) 1.0 (0e63.0)
Median duration of current episode, months (range) 5.0 (0e204.0) 6.0 (0e360.0) 5.8 (0e360.0)
Median time between last/current episode, months (range) 12.0 (0e360.0) 17.0 (0e252.0) 14.0 (0e360.0)

BMI ¼ body mass index; MDD ¼ major depressive disorder; SD ¼ standard deviation; SSRI ¼ selective serotonin reuptake inhibitor.

approximately three-quarters of the enrolled patients (75.8%) were on a higher than minimum dose of SSRI. During the study, 87/244
female. Before the study, 87 patients (18.0%) had received cit- patients (35.7%) in the duloxetine treatment group had a dose
alopram, 127 patients (26.3%) escitalopram, 130 patients (26.9%) escalation to 120 mg/day and 93/239 patients (38.9%) in the esci-
paroxetine, 134 patients (27.7%) sertraline, and 5 patients (1.0%) had talopram treatment group had a dose escalation to 20 mg/day.
received other drugs (fluoxetine hydrochloride, 2; fluvoxamine, 1;
metoclopramide, 1; reboxetine, 1). The median age at the first MDD 3.4. Clinical outcomes
episode was 34.2 years. Approximately three-quarters of patients
(73.9%) had experienced one or more episodes of MDD, and the The mean baseline to endpoint change in AES-C total score after 8
median (range) number of episodes was 1.0 (0e63.0). There was weeks of treatment was not statistically significantly different
clinically significant apathy in the study population at baseline between the duloxetine and escitalopram treatment groups (Table 3,
and patients were nonsyndromic but symptomatic for MDD: the Fig. 2). The least squares (LS) mean difference (95% confidence
mean AES-C total score (standard deviation [SD]) was 46.3 (7.97) interval [CI]) was #0.4 (#1.87 to 1.10, P ¼ 0.612, ES #0.05) (Table 3).
and the mean MADRS total score (SD) was 10.4 (3.59) (Table 2). There were improvements in apathy (AES-C, RSAT, MADRS Item 8),
Patient characteristics and apathy, depression, and functioning mood (MADRS), and functioning (CGI-S, PGI-I, CPFQ, SDS) (Table 3)
were similar in the duloxetine and escitalopram treatment groups in both the duloxetine and escitalopram treatment groups after 8
at baseline (Tables 1and 2). weeks of treatment. There were no statistically significant differ-
ences between the duloxetine and escitalopram treatment groups
for any of the secondary apathy, depression, or functional outcomes
3.3. Dose escalations after 8 weeks of treatment (Table 3).
There was a significant within-group improvement in AES-C
Before the study, 80/241 patients (33.2%) in the duloxetine total score in the group of patients who switched directly from
treatment group, 89/236 patients (37.7%) in the escitalopram escitalopram to escitalopram (escitalopram to escitalopram
treatment group, and 30/58 patients (51.7%) in the group of patients subgroup, n ¼ 57) over the 8-week treatment period. The LS mean
who switched directly from escitalopram to escitalopram had been change over the treatment period (as an average of all visits)
was #6.6 (standard error [SE] 0.85, P < 0.001). There was no
Table 2
statistically significant difference between the escitalopram to
Apathy, depression, and functioning at baseline.
escitalopram subgroup and the subgroup who switched from
Scalea Scale Duloxetine Escitalopram Overall escitalopram to duloxetine (P ¼ 0.600). In addition, there was no
range (N ¼ 244) (N ¼ 239) (N ¼ 483)
statistically significant difference between the subgroups who
Mean (SD) Mean (SD) Mean (SD) switched to escitalopram or duloxetine from another SSRI (i.e., not
AES-C total 18e72 46.3 (7.82) 46.3 (8.14) 46.3 (7.97) escitalopram, P ¼ 0.954). The sensitivity analysis, which removed
RSAT total 0e28 10.7 (4.92) 10.5 (4.95) 10.6 (4.93) all patients who had received escitalopram for at least 3 months
MADRS total 0e60 10.6 (3.49) 10.3 (3.68) 10.4 (3.59)
MADRS item 8 0e6 1.8 (1.10) 1.8 (1.15) 1.8 (1.13)
before the study, had no significant effect on the difference
CGI-S 1e7 3.1 (0.91) 3.0 (0.91) 3.1 (0.91) between the two treatment groups in the mean baseline to
CPFQ total 7e42 24.8 (5.73) 24.7 (5.83) 24.7 (5.78) endpoint change in AES-C total score after 8 weeks of treatment (LS
SDS total 0e30 15.3 (6.76) 14.6 (6.39) 15.0 (6.58) mean difference [95% CI]: 0.1 [#1.70 to 1.80], P ¼ 0.954).
AES-C ¼ Apathy Evaluation Scale e Clinician rated version; CGI-S ¼ Clinical Global An ad hoc analysis showed that the mean baseline to endpoint
Impression of Severity; CPFQ ¼ Cognitive and Physical Functioning Questionnaire; change in AES-C total score after 8 weeks of treatment was greater
MADRS ¼ Montgomery-Åsberg Depression Rating Scale; RSAT ¼ Rothschild Scale in the group of patients who were on the minimum SSRI dose
for Antidepressant Tachyphylaxis; SD ¼ standard deviation; SDS ¼ Sheehan
Disability Scale.
before the study compared with the group on a higher than
a
For each of the scales, the higher the score, the more severe the condition under minimum SSRI dose. The LS mean (SE) for the group on the
assessment. minimum SSRI dose before the study was #14.5 (0.51) and the LS
 J. Raskin et al. / Journal of Psychiatric Research 46 (2012) 667e674 671

Table 3
Change in apathy, depression, and functioning from baseline to 8 weeks.

Scalea Duloxetine (N ¼ 244) Escitalopram (N ¼ 239) Difference Effect sizeb P-value

LS mean (SE) LS mean (SE) LS mean (95% CI) Week 8c Treatmentd Treatment by visite
AES-C total #13.9 (0.54) #13.5 (0.54) #0.4 (#1.87 to 1.10) #0.05 0.612 0.674 0.964
RSAT total #5.5 (0.26) #5.0 (0.26) #0.5 (#1.22 to 0.20) #0.15 0.157 0.500 0.085
MADRS total #4.2 (0.32) #4.1 (0.33) #0.1 (#0.96 to 0.82) NC 0.880 0.866 0.360
MADRS item 8 #1.0 (0.06) #0.9 (0.06) #0.1 (#0.27 to 0.06) #0.09 0.224 0.101 0.061
CGI-S #0.9 (0.06) #0.9 (0.06) 0.1 (#0.09 to 0.22) NC 0.410 0.552 0.840
PGI-If 2.6 (0.08) 2.6 (0.08) 0 (#0.19 to 0.27) NC 0.723 0.792 0.170
CPFQ total #7.0 (0.34) #6.9 (0.35) #0.1 (#0.97 to 0.87) #0.01 0.910 0.910 N/A
SDS total #7.6 (0.40) #7.7 (0.41) 0.1 (#0.96 to 1.22) NC 0.821 0.821 N/A

All within-group P-values from t test on LS mean change were <0.001. All models were adjusted by including the fixed effect of investigator and baseline measure (except PGI).
AES-C ¼ Apathy Evaluation Scale e Clinician rated version; ANCOVA ¼ analysis of covariance; CGI-S ¼ Clinical Global Impression of Severity; CI ¼ confidence interval;
CPFQ ¼ Cognitive and Physical Functioning Questionnaire; LS ¼ least squares; MADRS ¼ Montgomery-Åsberg Depression Rating Scale; MMRM ¼ mixed-effects model
repeated measures; N/A ¼ not applicable; NC ¼ not calculated; PGI-I ¼ Patient Global Impression of Improvement; RSAT ¼ Rothschild Scale for Antidepressant Tachyphylaxis;
SDS ¼ Sheehan Disability Scale; SE ¼ standard error.
a
For each of the scales, a negative LS mean value denotes improvement on the scale.
b
Effect sizes calculated using Cohen’s d.
c
Between group P-values from t test on LS mean difference based on MMRM or ANCOVA (CPFQ and SDS only).
d
Main effect F test for treatment in MMRM or ANCOVA model.
e
Main effect F test for interaction term (treatment by visit) in MMRM over the 8-week treatment period.
f
PGI-I scores are presented as means rather than mean change from baseline values.

mean (SE) for the group on a higher than minimum SSRI dose
was #12.7 (0.70); the LS mean difference (95% CI) was #1.84 (#3.57
to #0.11, P ¼ 0.037).
3.5. Relapse
The relapse criterion (MADRS total score !16) was met by 55/
480 patients (11.5%) in the study population, and 29/243 patients
(11.9%) and 26/237 patients (11.0%) in the duloxetine and escitalo-
pram treatment groups, respectively. The minimum and maximum
time to relapse was 4 and 81 days, respectively. The median time to
relapse could not be calculated using the KaplaneMeier procedure.
The relapse criterion was met by 10/59 patients (17.0%) in the
escitalopram to escitalopram subgroup.
3.6. Safety
duloxetine and escitalopram treatment groups (Table 4). There was 1
death (suicide) in the escitalopram treatment group. Approximately
half of the patients in the study experienced one or more TEAE; the
most commonly reported TEAEs were nausea, headache, con-
stipation, and insomnia (Table 4). More patients in the duloxetine
compared with the escitalopram treatment group experienced
TEAEs of constipation (9.4% vs 3.8%) and anxiety (3.7% vs 0.4%).
Statistically significantly greater increases in heart rate and
diastolic blood pressure were observed in the duloxetine compared
with the escitalopram treatment group after 8 weeks of treatment.
The LS mean change (SE) in heart rate was 3.6 (0.51) and 1.0 (0.52)
in the duloxetine (N ¼ 212) and escitalopram (N ¼ 207) treatment
groups, respectively (LS mean difference [95% CI]: 2.7 [1.26e4.08],
P < 0.001). The LS mean change (SE) in diastolic blood pressure was
1.4 (0.49) and #0.4 (0.50) in the duloxetine (N ¼ 212) and escita-
lopram (N ¼ 207) treatment groups, respectively (LS mean differ-
ence [95% CI]: 1.8 [0.43e3.11], P ¼ 0.010).

The rates of SAEs, discontinuations due to adverse events, or

discontinuations due to lack of efficacy were similar in the
Table 4
Summary of adverse events and most common treatment-emergent adverse events
(!5%).

Adverse event Duloxetine Escitalopram Overall P-valuea

(N ¼ 244) (N ¼ 239) (N ¼ 483)

n (%) n (%) n (%)

SAEsb 3 (1.2) 3 (1.3) 6 (1.2) N/A
Deaths 0 (0) 1 (0.4) 1 (0.2) N/A
Discontinuations due 10 (4.1) 13 (5.4) 23 (4.8) 0.528
to an adverse event
Discontinuations due 5 (2.0) 3 (1.3) 8 (1.7) 0.724
to lack of efficacy
!1 TEAE 132 (54.1) 100 (41.8) 232 (48.0) 0.008
Nausea 26 (10.7) 28 (11.7) 54 (11.2) N/Ac
Headache 16 (6.6) 20 (8.4) 36 (7.5) N/Ac
Constipation 23 (9.4) 9 (3.8) 32 (6.6) N/Ac
Insomnia 17 (7.0) 11 (4.6) 28 (5.8) N/Ac

N/A ¼ not applicable; SAE ¼ serious adverse event; TEAE ¼ treatment-emergent

adverse event.
a
P-values from Fisher’s exact test.
b
3 patients in the duloxetine treatment group experienced 3 SAEs (depression,
cholelithiasis, and asthma) and 3 patients in the escitalopram treatment group
experienced 4 SAEs (suicide attempt, insomnia, melaena, and syncope).
c
Fig. 2. Change in AES-C total score in the duloxetine and escitalopram treatment Statistical significance for individual TEAEs are not reported because of the
groups over the 8-week treatment period. AES-C ¼ Apathy Evaluation Scale e Clinician issues associated with multiplicity and the suitability of the sample size for such
rated version; CI ¼ confidence interval; LS ¼ least squares. comparisons.
672 J. Raskin et al. / Journal of Psychiatric Research 46 (2012) 667e674
4. Discussion
To our knowledge, this is the largest controlled study designed
to evaluate apathy outcomes in patients with MDD. However,
several limitations must be considered. The results of this study are
only applicable to patients who were treated with a SSRI for MDD
for 3 or more months and continued to experience apathy in the
absence of syndromic depression. The current study does not
address if apathy is a residual symptom from the initial MDD or
a treatment-emergent symptom of SSRIs. Further, the study was
designed to allow patients to enter the trial from a SSRI at any
approved dose. This approach is important for generalizability and
feasibility, but the variability in treatment may have contributed to
the inability to find a difference between groups. However, post hoc
sensitivity analyses did not suggest that one or another SSRI
subgroup had a markedly different response than the group as
a whole (see Supplementary Table). Of note, improvements in
apathy were observed in patients in the escitalopram to escitalo-
pram subgroup, who had not switched antidepressant. These
patients had received escitalopram before the study and were
randomized to remain on escitalopram during the study. Apathy
improvements in this subgroup suggest that nonspecific effects of
time and participation in a study were relevant factors for both
groups.
The SNRI (duloxetine) treatment group showed improvements
in apathy during the study, as shown by statistically significant and
clinically significant within-group improvements in the AES-C total
score and MADRS Item 8. However, there was a similar degree of
improvement in patients in the escitalopram treatment group, who
had not switched class of antidepressant (i.e., continued to receive
treatment with a SSRI during the study). Although this was
a randomized controlled study, the lack of a placebo arm prevents
a full assessment of the effect of participation in the trial on
treatment outcomes, although the escitalopram to escitalopram
subgroup was included to allow for some interpretation. Partici-
pation in the trial may have improved outcomes in the following
ways: (i) patients knew they were receiving an active treatment
and that its effect on apathy was being assessed, and this may have
resulted in higher expectations of therapy; (ii) participation may
have resulted in improved treatment adherence; and (iii) partici-
pation required actions that may have counteracted apathy to some
extent, for example, leaving the house for study visits and engaging
with new individuals. As patients were not asked to guess treat-
ment assignment, the extent of blinding cannot be fully assessed.
There may also have been a nonspecific effect of time if the duration
of treatment before study entry had not been sufficient to treat the
patient’s depression fully. In this case, continuing antidepressant
treatment, regardless of the class of drug, may have been sufficient
to improve depression and, consequently, apathy. Although
patients were nonsyndromic for MDD, a mean MADRS total score of
10.4 at baseline suggests many patients still had moderate MDD
symptoms. Moreover, the numerically higher relapse rate in the
escitalopram to escitalopram subgroup compared with the primary
treatment groups implies that patients’ symptoms of depression
may not have been under good control in this subgroup, despite
being treated for at least 3 months. Given the above factors,
a randomized treatment period longer than 8 weeks may have
allowed differences between the two antidepressants, if any, to be
revealed.
The requirement for patients to have received treatment with
a SSRI for at least 3 months before the study was intended to
minimize symptoms of depression and allow for the appearance
of SSRI treatment-emergent effects. As there was no prospective
run-in to the study, it is not possible to state the cause of apathy
with certainty. The baseline assessments suggest that the study
population was more depressed than expected (mean MADRS
total score 10.4 [SD 3.59]). Other evidence that the patients may
have been more depressed at study entry was the relapse rate
(11.5% for the overall study population), which was higher than
expected for a stable population over 8 weeks. The relapses were
spread out over the study period, which suggests that the relapse
rate was not related to a dosing issue; if patients were switched to
too low a dose of antidepressant, the relapses would be expected
to occur in the first week or two of treatment. However, the
changing doses of SSRI that patients received before and during
the study may also have had an effect on apathy. For example,
a patient with apathy on a high-dose SSRI before the study may
have had an improvement in apathy on receiving a lower dose in
the study (Hoehn-Saric et al., 1990; Garland and Baerg, 2001;
Barnhart et al., 2004). Conversely, a patient on a low-dose SSRI
before the study may have received a higher dose in the study,
which could have improved the patient’s depression and, as
a result, improved their apathy. Analysis of the change in AES-C
total score by SSRI dose before the study suggested that there
was an improvement in apathy in both the group of patients who
had been on a higher than minimum SSRI dose before the study
and the group of patients who had been on the minimum SSRI
dose. In total, 35.4% of patients entered the study on higher doses
of SSRIs and a similar proportion (37.3%) had their antidepressant
dose optimized during the study.
There have been few assessments of apathy outcomes in
patients with MDD to date. The suggestion that apathy may be
a treatment-emergent effect of SSRI treatment arose from reports
of apathy in children, adolescents, and adults receiving a SSRI for
various conditions, including depression (Hoehn-Saric et al., 1990,
1991; Garland and Baerg, 2001; Barnhart et al., 2004). The apathy
occurred after weeks or months of otherwise effective treatment
and could be alleviated by reducing the dose of SSRI or stopping
treatment. Preliminary studies supported this proposed effect of
SSRI treatment and suggested that switching from a SSRI to a dual
reuptake inhibitor may improve apathy symptoms. A retrospective
analysis of antidepressant tachyphylaxis across antidepressant
classes found lower rates in patients receiving dual reuptake
inhibitors (venlafaxine and tricyclic antidepressants) compared
with those receiving SSRIs (Posternak and Zimmerman, 2005). A
case control study found that elderly depressed patients treated
with SSRIs may have more apathy symptoms at discharge
compared with those treated with non-SSRI antidepressants
(Wongpakaran et al., 2007). In addition, a study conducted by
Marangell et al. (2002) found a significant reduction in apathy
symptoms after the addition of olanzapine to SSRI treatment in
patients with MDD who had achieved remission. The rationale for
adding olanzapine to SSRI treatment was based on the suggestion
that increased levels of dopamine may improve apathy, and olan-
zapine may reverse the inhibition of dopamine release in the
frontal lobe that arises from chronic serotonergic stimulation by
SSRIs. The current study was intended to address the lack of large
controlled studies comparing antidepressants in treating residual
or antidepressant-induced apathy. While there were no statistically
significant differences between duloxetine and escitalopram on
apathy and while the conclusions that can be drawn from the study
are limited, this study (i) suggests that improving depressive
symptoms can improve symptoms of apathy; (ii) provides infor-
mation on apathy symptoms in a large group of patients with MDD
receiving antidepressant treatment; and (iii) provides valuable data
for the design of future studies of apathy in patients with MDD.
An issue to consider for future studies investigating apathy is the
choice of scale. There are few scales for assessing apathy and those
that are available are not well known to physicians. The AES treats
apathy as a psychological dimension that may be evaluated in
 J. Raskin et al. / Journal of Psychiatric Research 46 (2012) 667e674
patients whose apathy characterizes their overall clinical state as
well as in those in whom apathy is a symptom of another
syndrome, such as depression (Marin et al., 1991). The clinician
rated version of the AES (AES-C) was chosen as the primary means
of assessing apathy in this study. Also used were the RSAT
(Rothschild, 2008), which was designed to assess symptoms of
antidepressant tachyphylaxis in patients who do not meet the
criteria for a relapse or recurrence of an episode of major depres-
sion and includes “motivation and interest” and “energy level”
items, and the MADRS Item 8 (Montgomery and Åsberg, 1979),
which assesses the patient’s inability to feel. Significant within-
group improvements in apathy symptoms were observed in both
treatment groups on each of these scales.
There were few differences in safety and tolerability observed
between the duloxetine and escitalopram treatment groups;
however, the study was not powered for the safety endpoints. More
patients reported one or more TEAE in the duloxetine treatment
group than the escitalopram treatment group (132 vs 100) over the
8-week treatment period. However, patients switching from a SSRI
to another SSRI would be expected to have fewer side effects than
those switching to another class (SNRI). In addition, approximately
25% of patients in the escitalopram treatment group had already
been receiving treatment with escitalopram before the study and,
therefore, would be expected to experience even fewer adverse
events compared with patients switching from other SSRIs. Overall,
duloxetine-treated patients more commonly experienced
treatment-emergent anxiety or constipation, and experienced
greater increases in heart rate and diastolic blood pressure
compared with escitalopram-treated patients, as expected given
that duloxetine inhibits the reuptake of norepinephrine (Stahl et al.,
2005). The safety findings in the duloxetine treatment group were
similar to those observed in other clinical trials of duloxetine and
are what would be expected of a SNRI (Hudson et al., 2005; Thase
et al., 2005; Gahimer et al., 2007; Norman and Olver, 2010).
In conclusion, this study did not detect a difference between
duloxetine and escitalopram on apathy, depression, or functional
outcomes after 8 weeks of treatment in nonsyndromal depressed
patients with residual apathy; there were significant improvements
in all outcomes in both treatment groups. Patients in the escitalo-
pram to escitalopram subgroup also experienced improvements in
apathy over the 8-week treatment period. Given the limitations
discussed above, this study does not support our hypothesis that
switching from a SSRI to a SNRI alleviates apathy. Duloxetine has
effects on both serotonin and norepinephrine reuptake. While
speculative, it is possible that a noradrenergic agent without
concomitant serotonergic effects would be effective i.e., the sero-
tonergic component of duloxetine might have been problematic
with regard to alleviating apathy. Further study of the natural
course of apathy in patients with MDD and the relative benefits of
different treatments is warranted.
Role of funding source
This study (F1J-CR-HMGM) was sponsored by Eli Lilly and
Company. Eli Lilly was involved in the study design, data collection,
data analysis, and preparation of the manuscript. In compliance
with the Uniform Requirements for Manuscripts, established by the
International Committee of Medical Journal Editors, the sponsor of
this study did not impose any impediment, directly or indirectly, on
the publication of the study’s results. The authors acknowledge the
independent medical writing assistance provided by Justine
Southby, PhD, and Serina Stretton, PhD, of ProScribe Medical
Communications, funded from an unrestricted financial grant from
Eli Lilly Australia. ProScribe’s services complied with international
guidelines for Good Publication Practice (GPP2).
673
Role of contributors
JR, REG, and LM were involved in study design, NH was involved
in data collection, GWZ was involved in the statistical analysis, and
TG was a Principal Investigator. All authors participated in the
interpretation of the study results, and in the drafting, critical
revision, and approval of the final version of the manuscript.
Conflicts of interest
JR, REG, GWZ, and LM are full-time employees of, and NH is
contractually employed by, Eli Lilly and Company. JR and LM own
stock in Eli Lilly and Company. TG has no conflicts of interest to
disclose.
Acknowledgments
The authors acknowledge Richard Walton (Eli Lilly) for his
assistance in reviewing the manuscript. The authors would also like
to thank the investigators, their staff, and their patients for their
participation in this study.
Appendix. Supplementary material
Supplementary material related to this article can be found
online at doi:10.1016/j.jpsychires.2012.02.010
References
Barnhart WJ, Makela EH, Latocha MJ. SSRI-induced apathy syndrome: a clinical
review. Journal of Psychiatric Practice 2004;10:196e9.
Birkmayer W, Riederer P. Understanding the neurotransmitters: keys to the work-
ings of the brain. Blau K, trans.. New York, NY: Springer-Verlag Wien; 1989
Bolling MY, Kohlenberg RJ. Reasons for quitting serotonin reuptake inhibitor
therapy: paradoxical psychological side effects and patient satisfaction.
Psychotherapy and Psychosomatics 2004;73:380e5.
Demyttenaere K, Jaspers L. Review: bupropion and SSRI-induced side effects.
Journal of Psychopharmacology 2008;22:792e804.
Fava M, Graves LM, Benazzi F, Scalia MJ, Iosifescu DV, Alpert JE, et al. A cross-
sectional study of the prevalence of cognitive and physical symptoms during
long-term antidepressant treatment. Journal of Clinical Psychiatry 2006;67:
1754e9.
Fava M, Iosifescu DV, Pedrelli P, Baer L. Reliability and validity of the Massachusetts
general hospital cognitive and physical functioning questionnaire. Psycho-
therapy and Psychosomatics 2009;78:91e7.
Gahimer J, Wernicke J, Yalcin I, Ossanna MJ, Wulster-Radcliffe M, Viktrup L.
A retrospective pooled analysis of duloxetine safety in 23,983 subjects. Current
Medical Research and Opinion 2007;23:175e84.
Garland EJ, Baerg EA. A motivational syndrome associated with selective serotonin
reuptake inhibitors in children and adolescents. Journal of Child and Adolescent
Psychopharmacology 2001;11:181e6.
Gobert A, Rivet JM, Lejeune F, Newman-Tancredi A, Adhumeau-Auclair A, Nicolas JP,
et al. Serotonin2C receptors tonically suppress the activity of mesocortical
dopaminergic and adrenergic, but not serotonergic, pathways: a combined
dialysis and electrophysical analysis in the rat. Synapse 2000;36:205e21.
Gold PW, Chrousos GP. The endocrinology of melancholic and atypical depression:
relation to neurocircuitry and somatic consequences. Proceedings of the
Association of American Physicians 1999;111:22e34.
Guy W. ECDEU assessment manual for psychopharmacology, revised. DHEW
Publication ADM-76-338. Rockville (MD): US Department of Health, Education,
and Welfare; 1976.
Hoehn-Saric R, Lipsey JR, McLeod DR. Apathy and indifference in patients on flu-
voxamine and fluoxetine. Journal of Clinical Psychopharmacology 1990;10:
343e5.
Hoehn-Saric R, Harris GJ, Pearlson GD, Cox CS, Machlin SR, Camargo EE.
A fluoxetine-induced frontal lobe syndrome in an obsessive compulsive patient.
Journal of Clinical Psychiatry 1991;52:131e3.
Hudson JI, Wohlreich MM, Kajdasz DK, Mallinckrodt CH, Watkin JG, Martynov OV.
Safety and tolerability of duloxetine in the treatment of major depressive
disorder: analysis of pooled data from eight placebo-controlled clinical trials.
Human Psychopharmacology 2005;20:327e41.
Marangell LB, Johnson CR, Kertz B, Zboyan HA, Martinez JM. Olanzapine in the
treatment of apathy in previously depressed participants maintained with
selective serotonin reuptake inhibitors: an open-label, flexible dose study.
Journal of Clinical Psychiatry 2002;63:391e5.
674 J. Raskin et al. / Journal of Psychiatric Research 46 (2012) 667e674
Marin RS, Biedrzycki RC, Firinciogullari S. Reliability and validity of the apathy
evaluation scale. Psychiatry Research 1991;38:143e62.
McClintock SM, Husain MM, Wisniewski SR, Nierenberg AA, Stewart JW,
Trivedi MH, et al. Residual symptoms in depressed outpatients who respond by
50% but do not remit to antidepressant medication. Journal of Clinical
Psychopharmacology 2011;31:180e6.
Montgomery SA, Åsberg M. A new depression scale designed to be sensitive to
change. British Journal of Psychiatry 1979;134:382e9.
Norman TR, Olver JS. Continuation treatment of major depressive disorder: is there
a case for duloxetine? Journal of Drug Design, Development and Therapy 2010;
4:19e31.
Paykel ES, Ramana R, Cooper Z, Hayhurst H, Kerr J, Barocka A. Residual symptoms
after partial remission: an important outcome in depression. Psychological
Medicine 1995;25:1171e80.
Posternak MA, Zimmerman M. Dual reuptake inhibitors incur lower rates of
tachyphylaxis than selective serotonin reuptake inhibitors: a retrospective
study. Journal of Clinical Psychiatry 2005;66:705e7.
Rothschild AJ. The Rothschild Scale for Antidepressant Tachyphylaxis: reliability and
validity. Comprehensive Psychiatry 2008;49:508e13.
Segal M. Serotonergic innervation of the locus coeruleus from the dorsal raphe and its
action on responses to noxious stimuli. Journal of Physiology 1979;286:401e15.
Sheehan DV, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, et al. The Mini-
International Neuropsychiatric Interview (M.I.N.I.): the development and vali-
dation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10.
Journal of Clinical Psychiatry 1998;59(Suppl. 20):22e33.
Stahl SM, Grady MM, Moret C, Briley M. SNRIs: their pharmacology, clinical efficacy,
and tolerability in comparison with other classes of antidepressants. CNS
Spectrums 2005;10:732e47.
Thase ME, Tran PV, Wiltse C, Pangallo BA, Mallinckrodt C, Detke MJ. Cardiovascular
profile of duloxetine, a dual reuptake inhibitor of serotonin and norepinephrine.
Journal of Clinical Psychopharmacology 2005;25:132e40.
Wongpakaran N, van Reekum R, Wongpakaran T, Clarke D. Selective serotonin
reuptake inhibitor use associates with apathy among depressed elderly: a case-
control study. Annals of General Psychiatry 2007;6:7.