Journal of Affective Disorders 141 (2012) 308–314

Contents lists available at SciVerse ScienceDirect

Journal of Affective Disorders

j o u rn al ho m ep ag e: w w w . e ls e v i e r . c o m / l oca te/ jad

Research report

Effect of celecoxib add-on treatment on symptoms and serum IL-6

concentrations in patients with major depressive disorder: Randomized
double-blind placebo-controlled study
Seyed-Hesameddin Abbasi a, b, Fahimeh Hosseini a, Amirhossein Modabbernia c,
Mandana Ashrafi c, Shahin Akhondzadeh c,⁎
a
Family Health Research Center, Iranian Petroleum Industry Health Research Institute, NIOC Central Hospital, Tehran, Iran
b
Tehran Heart Center, Tehran University of Medical Sciences, Tehran, Iran
c
Psychiatric Research Center, Roozbeh Hospital, Tehran University of Medical Sciences, Tehran, Iran

a r ticl e i nf o a b s t r a c t

Article history: Background: It has been proposed that the mechanism of the antidepressant effect of celecoxib
Received 5 February 2012 is linked to its anti-inflammatory action and particularly its inhibitory effect on pro-
Received in revised form 26 March 2012 inflammatory cytokines (e.g. interleukin-6(IL-6)). We measured changes in serum IL-6
Accepted 27 March 2012
concentrations and depressive symptoms following administration of celecoxib in patients
Available online 18 April 2012
with major depressive disorder (MDD).
Methods: In a randomized double-blind placebo-controlled study, 40 patients with MDD and
Keywords:
Hamilton Depression Rating Scale—17 items (Ham-D) score ≥ 18 were randomly assigned to
Add-on
either celecoxib (200 mg twice daily) or placebo in addition to sertraline (200 mg/day) for
Celecoxib
6 weeks. Outcome measures were serum IL-6 concentrations at baseline and week 6, and
COX-2 inhibitor
Ham-D scores at baseline and weeks 1, 2, 4, and 6.
IL-6
Major depressive disorder Results: The celecoxib group showed significantly greater reduction in serum IL-6 concentra-
Sertraline tions (mean difference (95%CI) = 0.42(0.30 to 0.55) pg/ml, t(35)= 6.727, P b 0.001) as well as
Ham-D scores (mean difference (95%CI)= 3.35(1.08 to 5.61), t(38)= 2.99, P = 0.005) than the
placebo group. The patients in the celecoxib group experienced more response (95%) and
remission (35%) than the placebo group (50% and 5%, P = 0.003 and 0.04 respectively).
Baseline serum IL-6 levels were significantly correlated with baseline Ham-D scores (r= 0.378,
P = 0.016). Significant correlation was observed between reduction of Ham-D scores and
reduction of serum IL-6 levels at week 6 (r= 0.673, P b 0.001).
Limitations: We did not measure other inflammatory biomarkers.
Conclusions: We showed that the antidepressant activity of celecoxib might be linked to its
capability of reducing IL-6 concentrations. Moreover, supporting previous studies we showed
that celecoxib is both safe and effective as an adjunctive antidepressant (Registration number:
IRCT138903124090N1).
© 2012 Elsevier B.V. All rights reserved.

1. Introduction

Due to incomplete response to conventional antidepres-

* Corresponding author at: Psychiatric Research Center, Roozbeh Psychi-
atric Hospital, Tehran University of Medical Sciences, South Kargar Street,
Tehran 13337, Iran. Tel.: + 98 21 88281866; fax: + 98 21 55419113.
E-mail address: s.akhond@neda.net (S. Akhondzadeh).
sants in patients with major depressive disorder (MDD),
several augmentative strategies have been developed for the
treatment of these patients in recent decade (Abolfazli et al.,
2011; Akhondzadeh et al., 2009; Fava, 2009; Muller et al.,
2006). One of the suggested add-on treatments is celecoxib, a

0165-0327/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.jad.2012.03.033
 S.-H. Abbasi et al. / Journal of Affective Disorders 141 (2012) 308–314
cyclooxygenase-2 (COX-2) inhibitor (Muller, 2010). COX-2 is
an enzyme which is implicated in prostaglandin E2 (PGE2)
production. Several animal and human studies have shown
the add-on antidepressant effects of COX-2 inhibitors
(Akhondzadeh et al., 2009; Chen et al., 2010; Muller et al.,
2006; Myint et al., 2007; Nery et al., 2008). Moreover, several
lines of evidence suggest beneficial effect of COX-2 inhibitors
on other psychiatric disorders such as schizophrenia
(Akhondzadeh et al., 2007; Bresee et al., 2006; Muller et al.,
2004a,2004b; Rapaport et al., 2005). The mechanism under-
lying antidepressant effect of these drugs is still unclear.
COX-2 inhibitors are capable of reducing inflammatory
cytokines in several tissues (Alhan et al., 2004; Bukata et al.,
2004; Chen et al., 2011; Liu et al., 2011b; Tipton et al., 2003).
Thus, it has been suggested that these drugs might exert their
antidepressant effect through inhibition of proinflammatory
cytokines particularly Interleukin (IL)-6 and IL-1 (Muller et
al., 2009).
Growing body of evidence suggests that inflammation is
implicated in the pathophysiology of MDD (Dinan, 2009;
Dowlati et al., 2010; Muller, 2010; Muller et al., 2011). Sickness
behavior which is a result of inflammatory activation, shares
many clinical features such as anhedonia, anorexia, irritability,
and mild cognitive problems with MDD (Dantzer and Kelley,
2007). Several studies have shown an elevation of proinflam-
matory cytokines (particularly IL-6 and tumor necrosis factor
(TNF-α)) in patients with MDD (Janssen et al., 2010). A recent
meta-analysis has confirmed the activation of proinflammatory
cytokines in these patients (Dowlati et al., 2010). IL-6 is one of
the most widely studied cytokines in patients with MDD
(Glaser et al., 2003; Liu et al., 2011a; Muller et al., 2009, 2011;
Pace et al., 2006; Prather et al., 2009). In addition to elevation of
this cytokine in patients with MDD, relation of IL-6 concentra-
tion to severity of depression (Hannestad et al., 2011; Prather
et al., 2009; Su et al., 2009), a shift in circadian rhythm (Alesci
et al., 2005), and a reduction in its concentration in response to
antidepressants (Basterzi et al., 2005; Frommberger et al.,
1997; Sluzewska et al., 1995) have been shown in several
studies. PGE2 (the main product of COX-2) can stimulate the
production of IL-6 and has a close association with inflamma-
tion and sickness behavior (Hinson et al., 1996; Muller et al.,
2009). A role for PGE2 has been suggested in patients with
MDD (Leonard, 2001). PGE2 concentration increases in body
fluids of patients with MDD and decreases following antide-
pressant treatment (Calabrese et al., 1986; Linnoila et al.,
1983). Muller et al. (2009) suggested that the link between
inflammation and depression can be explained by direct effect
of cytokines on neurotransmitter metabolism, or activation of
indoleamine 2,3-dioxygenase (IDO) and subsequent increase
in serotonin metabolism and production of quinolinic acid (a
neurotoxic compound). IDO activity is stimulated by proin-
flammatory cytokines and is decreased by anti-inflammatory
cytokines (Carlin et al., 1987; Musso et al., 1994). Taken
together, these findings suggest that a hypothetical antide-
pressant mechanism for celecoxib can be its effect on IL-6
expression and/or levels. To date, no study to our knowledge
has addressed this issue in the clinical setting.
The present study aimed to assess the effect of celecoxib
add-on therapy on depressive symptoms and serum IL-6
concentration in patients with MDD. We hypothesized that
the reduction in both depressive symptoms and IL-6 levels
309
should be more pronounced in patients who received
celecoxib. To link IL-6 to the antidepressant mechanism of
celecoxib, we also hypothesized that the change of IL-6 over
the course of the trial should be correlated with the change in
depressive symptoms.
2. Methods
2.1. Participants
Men and women aged 18 to 50 years, with a diagnosis of
MDD (based on DSM-IV-TR criteria), with Hamilton Depres-
sion Rating Scale—17 items (Hamilton, 1960) (Ham-D) score
of ≥ 18 and with a score of at least 2 on item 1 of Ham-D were
included. Wash-out periods of 1 month for fluoxetine and
1 week for other antidepressants were required prior to
entry. Exclusion criteria were psychosis, other diagnoses in
DSM Axes I and II, use of anti-depressant drugs in the last
month, electroconvulsive therapy in the last two months,
substance abuse or dependence within one year, high risk of
suicide, thyroid disorder, cardiovascular disorders, pregnancy
and lactation. Screened participants gave history, and under-
went physical and electrocardiographic examination to rule
out the presence of any of the exclusion criteria. Institutional
Review Board (IRB) of Tehran University of Medical Sciences
(Grant No: 12967) approved the study. The trial was
conducted in line with the latest revision of the Declaration
of Helsinki. All participants and their legally authorized
representatives signed an informed consent form.
2.2. Trial design
This was a double-center, randomized, double-blind,
placebo-controlled, and parallel-group study conducted in
Tehran, Iran. Patients were assessed in the outpatient clinics
of Roozbeh psychiatric hospital (a tertiary referral center
affiliated with Tehran University of Medical Sciences) and
National Iranian Oil Company (NIOC) central hospital from
June 2010 to September 2011.
2.3. Interventions
The patients were randomly assigned to receive either
celecoxib 200 mg twice daily or placebo twice daily (with the
same taste and shape as celecoxib) for 6 weeks. All partici-
pants received sertraline 200 mg/day during the course of the
trial. Sertraline was administered half dose in the first week.
2.4. Outcomes
Ham-D was used to assess depressive symptoms at
baseline and at weeks 1, 2, 4, and 6. Ten ml of venous blood
was drawn from each patient for measurement of IL-6 at
baseline and week 6. The blood was transferred into tubes
without anticoagulants and centrifuged for 15 min at a speed
of 250 g. All sera samples were stored at −70 °C before use.
IL-6 concentrations in the sera were determined by the
sandwich enzyme immunoassay method, using Cytelisa
human IL-6 kit (Cytimmune Sciences, Maryland). All samples
were tested in the same run, which also included a set of
standards that were measured in duplicate. The amount of IL-
310 S.-H. Abbasi et al. / Journal of Affective Disorders 141 (2012) 308–314
6 (pg/ml) in each sample was calculated using the standard
curve.
The primary outcome was the difference in Ham-D score
reduction from baseline to week 6 between the celecoxib and
the placebo groups. The other primary outcome was the
difference in IL-6 concentration change from baseline to week
6 between the two study groups. Score change (at each visit),
response rates (at least 50% reduction in the Ham-D score at
weeks 4 and 6), and remission rates (Ham-D score ≤7) at the
end of the study were also compared between the two study
groups (McIntyre, 2010). Side effects were systematically
recorded at each visit using a side-effect checklist. This was
further augmented by self-reports of the patients.
2.5. Sample size
To calculate sample size, we assumed a clinically signif-
icant difference of 3.5 on the Ham-D, a standard deviation of
3.5, a two-sided significance of 5%, and a power of 80%. A
sample size of 32 (16 per group) was calculated, and
assuming a 20% drop-out rate a final sample size of 40 was
achieved.
2.6. Randomization, allocation concealment, and blinding
To randomize the patients in a 1:1 ratio (blocks of four), a
computer random number generator was used. Allocation
concealment was achieved using sequentially numbered,
opaque, and sealed envelopes. Random allocation and rating
of the participants were done by separate persons. The
participants, the physician who referred them, the psychia-
trists who assessed the patients and prescribed the drug, and
the statistician were blind to allocation.
2.7. Statistical analysis
We used IBM SPSS Statistic 19.0.0 (IBM Co.) to analyze the
data. Categorical variables were reported as number (percent-
age) and continuous variables were reported as mean
(±standard deviation, SD). Two-factor repeated measure
analysis of variance (ANOVA) was used to assess the effect of
treatment and time-treatment interaction and to compare the
score change between the two groups. The result of Green-
house–Geisser's correction was reported whenever Mauchly's
test of sphericity was significant. To compare the score change
from baseline to each time point between the two groups,
independent sample t-test was used and Cohen's d was
calculated. To compare the IL-6 concentration at each time
point and its change between the two groups as well as
between the responder/remitters and non-responders/non-
remitters, independent sample t-test was used. Paired t-test
was used for comparison of IL-6 pre- and post-treatment levels
in each group. Correlation of IL-6 with Ham-D scores was
determined using Pearson's correlation coefficient. To compare
the proportions (responders and remitters) between the two
groups, Pearson's chi-square and Fisher's exact test were used.
Cohen's d effect sizes were determined by dividing the mean
difference of the two groups at each time point by their pooled
standard deviation. Prediction of Ham-D score reduction,
response, and remission were done using regression analysis.
All analyses were based on the intention-to-treat sample and
were performed using last observation carried forward (LOCF)
procedure. A P value of b 0.05 was considered statistically
significant in all analyses.
3. Results
3.1. Participants
Sixty-nine potentially eligible candidates were screened, 29
of which did not meet the inclusion criteria. Forty eligible
patients were randomly assigned to either celecoxib plus
sertraline (n=20) or placebo plus sertraline (n=20), and 37
patients completed the study (Fig. 1). Table 1 shows baseline
data of the participants. Baseline Ham-D scores was not
significantly different between the two groups (mean± SD
for the celecoxib group= 22.2± 1.7, for the placebo
group= 21.3 ± 1.9, P=0.121). Baseline serum IL-6 concentra-
tions were similar between the two groups (2.79± 0.76 pg/ml
in celecoxib group vs. 2.78 ± 0.72 pg/ml in placebo group,
P= 0.966). None of the patients received any other psychotro-
pic medication (including benzodiazepines) during the study
course.
3.2. Clinical outcomes
Two-factor repeated measure ANOVA showed significant
effect for time-treatment interaction indicating that the
behavior of two treatment groups was not similar during the
course of the trial [Greenhouse–Geisser's correction: F(1.803,
68.507)=5.287, P=0.009]. The effect was also significant for
time [Greenhouse–Geisser's correction: F(1.803, 68.507)=
314.286, Pb 0.001], and nearly significant for treatment [F(1,
38)= 3.023, P= 0.09]. After adjusting for baseline score as
covariate in the repeated measure analysis, the effect of
treatment became significant [F(1, 37)=5.383, P= 0.026]
(Fig. 2). Moreover, score reduction from baseline was signifi-
cantly greater in the celecoxib group than the placebo group at
any given time-point with effect sizes ranging from 0.68 to 0.95
(Table 2).
Eleven (55%) patients in the celecoxib group and three
(15%) patients in the placebo group responded to treatment
by week 4 (Χ 2(1)= 7.033, P = 0.008). The response rate at
week 6 was 95% in the celecoxib group compared with 50% in
the placebo group (Fisher's exact test P = 0.003). Moreover,
seven (35%) patients in the celecoxib group experienced
remission by week 6 compared with only one (5%) patient in
the placebo group (Fisher's exact test P = 0.04).
3.3. Serum IL-6 concentrations
Serum IL-6 concentrations at the end of the study were
2.16 ± 0.60 pg/ml in the celecoxib group and 2.56 ± 0.64 pg/
ml in the placebo group. Paired t-test showed a significant
reduction in the serum IL-6 concentrations at the end of the
sixth week in both groups [celecoxib group: t(18)= 11.366,
P b 0.001, placebo group: t(17)= 11.926, Pb 0.001]. However,
the celecoxib group experienced significantly more reduction
in serum level of this cytokine than the placebo group
[celecoxib group (mean± SD reduction)= 0.66 ± 0.25, pla-
cebo group (mean± SD reduction)= 0.24 ± 0.08, t(35)=
6.727, P b 0.001]. Baseline IL-6 was significantly correlated
 S.-H. Abbasi et al. / Journal of Affective Disorders 141 (2012) 308–314 311

Fig. 1. Flow diagram of the trial.

with baseline Ham-D scores (r= 0.378, P = 0.016). The
reduction in IL-6 levels showed a high correlation with
reduction in Ham-D scores at week 6 (r= 0.673, P b 0.001)
(Fig. 3). This correlation was present in both groups of the
patients [celecoxib group: r= 0.493, P= 0.03, placebo group:
r= 0.500, P = 0.03]. The patients who were categorized as
responder at week 6 showed significantly greater reduction
in their serum IL-6 levels than the non-responders [re-
sponders (mean± SD)= 0.52 ± 0.29 pg/ml, non-responders
(mean± SD)= 0.21 ± 0.06 pg/ml, t(35)= 3.017, P = 0.005].
Similar results were obtained when comparing remitters and
non-remitters [remitters (mean± SD)= 0.68 ± 0.29 pg/ml,
non-remitters (mean± SD)= 0.39 ± 0.25 pg/ml, t(35)=
2.811, P = 0.008]. In linear regression analysis, change in
the IL-6 concentration positively predicted Ham-D reduction
in all subjects (adjusted R-squared, aR2 = 0.437, Pb 0.001) as
well as in the celecoxib (aR2 = 0.200, P = 0.031) and the
placebo (aR2 = 0.202, P = 0.035) groups. Number of subjects
with positive or negative outcomes in each group was too
few to allow running logistic regression separately for each
study group. Considering all participants in logistic
regression analysis, reduction of serum IL-6 concentration
(pg/l) significantly and positively predicted response (Odds
ratio, OR = 1.0125, 95% confidence intervals, 95%CI= 1.0020
to 1.0230, P = 0.019) and remission (OR= 1.0036,
95%CI= 1.0005 to 1.0066, P = 0.023).

Table 1
Baseline characteristics of the patients.

Variable Sertraline + Sertraline +

celecoxib (n= 20) placebo (n= 20)

Gender, F(%) 7(35%) 6(30%)

Age, mean± SD 35.1 ± 8.0 34.2 ± 6.9
Number of previous 3.7 ± 0.8 3.6 ± 0.9
episodes, mean± SD
Duration of illness, 2.4 ± 0.9 2.7 ± 1.0
months, mean± SD
Weight, kilograms, 67.5 ± 11.6 66.5 ± 11.1
mean± SD
Baseline Ham-D score, 22.2 ± 1.7 21.3 ± 1.9
mean± SD
Fig. 2. Results of two-factor repeated measure ANOVA for the effect of
Baseline IL-6 levels, 2.79 ± 0.76 2.78 ± 0.72
treatment on Hamilton Depression Rating Scale. Values are mean± SEM.
pg/ml, mean± SD
*P b 0.05, **P b 0.01. P values are results of independent t-test.
312 S.-H. Abbasi et al. / Journal of Affective Disorders 141 (2012) 308–314

Table 2
Comparison of score changes from baseline between the two groups.

Week Celecoxib + sertraline Placebo+ sertraline Mean difference t(38) P value Cohen's d
group (mean± SD) group (mean± SD) (95% confidence interval) (95% confidence interval)

Week 1 3.75 ± 2.02 2.50 ± 1.43 1.25(0.13 to 2.37) 2.255 0.030 0.71(0.07 to 1.35)
Week 2 7.40 ± 2.23 5.55 ± 2.06 1.85(0.47 to 3.22) 2.720 0.010 0.86(0.20 to 1.50)
Week 4 10.30 ± 3.31 8.30 ± 2.49 2.00(0.12 to 3.87) 2.158 0.037 0.68(0.04 to 1.32)
Week 6 13.40 ± 3.88 10.05 ± 3.15 3.35(1.08 to 5.61) 2.994 0.005 0.95(0.29 to 1.60)

3.4. Adverse events
No patient, in either group, experienced severe adverse
events. Nine side-effects were recorded during the course of
the trial. There was no significant difference in the frequency
of the side effects between the two groups (Table 3).
4. Discussion
The present study showed superiority of celecoxib over
placebo in the treatment of patients with MDD. In line with
our hypothesis, we documented that celecoxib is capable of
reducing of IL-6 concentration. Importantly, we also showed
that IL-6 reduction was in close association with reduction of
depressive symptoms, further strengthening this hypothesis
that celecoxib might exert part of its antidepressant effect
through reduction of this cytokine. Several studies have
shown the importance of IL-6 in depression. Su et al. (2009)
studied 188 twins for depressive symptoms and inflamma-
tory markers and showed that IL-6 was strongly correlated
with depressive symptoms even after controlling for other
factors. Interestingly, they also found significant genetic
correlation between IL-6 and depression. Several other
studies on IL-6 also showed increased IL-6 concentrations,
its relation to the severity of depressive symptoms, and its
exaggerated response to stress in patients with MDD (Glaser
et al., 2003; Kiecolt-Glaser et al., 2007; Pace et al., 2006;
Prather et al., 2009). It has been shown that plasma IL-6 has a
positive correlation with 5-hydroxyindoleacetic acid/serotonin
ratio (increased serotonin metabolism) in all brain regions
in omega-3-deficient mice (McNamara et al., 2010). Thus,
IL-6 might be capable of increasing serotonin metabolism
and causing neurotoxicity (as a result of IDO activation)
both of which are important components of MDD (Muller
et al., 2009). On the other hand, celecoxib can act as both
an antidepressant and a neuroprotectant (Chen et al., 2010;
McNamara et al., 2010). Taken together, these findings
strongly suggest that celecoxib exerts part of its antidepres-
sant mechanism through its anti-inflammatory properties and
specifically through inhibition of production of pro-
inflammatory cytokines. Of note, in our study the patients
who only received sertraline also showed a significant
decrease in their IL-6 levels. This later finding along with
evidence of anti-inflammatory effect of other antidepressants
further supports this notion that many antidepressants might
exert part of their antidepressant effects through inhibition
of inflammatory pathways (Hannestad et al., 2011;
Sluzewska et al., 1995). It has been suggested that the effect
of celecoxib on IL-6 might be exerted through the regulation
of IL-6 gene expression or reduction of PGE2 (El-Ghazaly et
al., 2010; Hinson et al., 1996). In addition to inhibition of IL-6,
celecoxib might affect the central nervous system (CNS) in
other ways. This drug may also inhibit the expression of other
cytokines which are involved in sickness behavior (Muller et
al., 2009). In one study, this drug was capable of reducing
blood PGE2, IL-1beta, and corticosterone concentrations and
increasing nerve growth factor (NGF) expression in olfactory
bulbectomized rats (an animal model of MDD) (Song et al.,
2009). Furthermore, treatment with celecoxib can prevent
dysregulation of the hypothalamic–pituitary–adrenal axis
and glucocorticoid receptor function (two key features of
MDD) possibly through reduction in PGE2 and IL-6 produc-
tion (Casolini et al., 2002; Hu et al., 2005; Humphreys et al.,
2006).
COX-2 inhibitors not only show antidepressant proper-
ties, but also seem to be promising neuroprotectants and
antipsychotics (Akhondzadeh et al., 2007; Minghetti, 2004;
Muller, 2010; Rapaport et al., 2005). The mechanism of their
CNS effects (other than antidepressant effect), although not
completely understood, can be explained at least in part, by
their anti-inflammatory properties. For example, in one study
on the effect of celecoxib add-on treatment in patients with

Table 3
Frequency of adverse events in the two groups.

Sertraline+ celecoxib Sertraline+ placebo Adverse events

3(15%) 4(20%) Anxiety

1(5%) 1(5%) Tremor
3(15%) 3(15%) Sweating
3(15%) 2(10%) Abdominal pain
1(5%) 1(5%) Insomnia
2(10%) 2(10%) Decrease appetite
5(25%) 6(30%) Increased appetite
1(5%) 1(5%) Nausea
Fig. 3. Correlation of changes in Hamilton Depression Rating Scale and
1(5%) 1(5%) Headache
changes in the IL-6 concentrations between week 0 and week 6.
 S.-H. Abbasi et al. / Journal of Affective Disorders 141 (2012) 308–314
schizophrenia, reduction in CD19 + lymphocytes was corre-
lated with a decrease in negative symptoms of these patients,
suggesting that the anti-inflammatory effect of celecoxib may
explain its antipsychotic properties (Muller et al., 2004b).
In line with previous studies, we showed that celecoxib
resulted in greater score reduction, and response and remission
rates than placebo without causing any notable adverse events
(Akhondzadeh et al., 2009; Muller et al., 2006; Nery et al.,
2008). Prior to this study, celecoxib had been used as add-on to
fluoxetine and reboxetine in patients with MDD. The present
study extends the use of celecoxib add-on to sertraline, another
widely used antidepressant. These findings further confirm
safety and efficacy, and augmentative capability of celecoxib in
patients with MDD.
5. Limitations
Although the present study provided evidence for the
antidepressant mechanism of celecoxib, it certainly had some
limitations. First, because we did not plan to measure other
inflammatory markers in our study we had to select the most
obvious choice (IL-6) to our work. Second, one might argue
that the correlation of IL-6 with depressive symptoms in the
present study may simply show an epiphenomenon rather
than a causal association. However as noted before, the relation
of pro-inflammatory cytokines to important pathophysiologi-
cal changes in MDD (increased IDO activity, alteration in
serotonin metabolism, and neurotoxicity) is beyond a simple
epiphenomenon. Third, because we used a fixed dose of
celecoxib in our study, the effect of other dosages of this drug
is still unclear. In addition, although some studies suggest that
celecoxib might be associated with increased cardiac adverse
events (White et al., 2007), to date, this drug has been safe in
the neuropsychiatric setting. This might be due to small sample
sizes of studies or because of small dose and short duration of
drug administration. Larger controlled studies may address this
important issue.
6. Conclusion
The findings of our study together with previous studies
provided evidence for involvement of IL-6 in the antidepres-
sant mechanism of celecoxib. Furthermore, in line with
previous studies, we showed that celecoxib add-on is an
effective and safe adjunct for treatment of patients with MDD.
(Trial registration number: IRCT138903124090N1 URL: http://
www.irct.ir/searchresult.php?id=4090&number=1).
Role of funding source
This study is supported by a grant from Tehran University of Medical
Sciences (grant no: 12967).
Conflict of interest
All authors declare that they have no conflicts of interest.
Acknowledgments
This study was supported by a grant from Tehran University of Medical
Sciences to Professor Shahin Akhondzadeh (grant number: 12967).
313
References
Abolfazli, R., Hosseini, M., Ghanizadeh, A., Ghaleiha, A., Tabrizi, M.,
Raznahan, M., Golalizadeh, M., Akhondzadeh, S., 2011. Double-blind
randomized parallel-group clinical trial of efficacy of the combination
fluoxetine plus modafinil versus fluoxetine plus placebo in the
treatment of major depression. Depression and Anxiety 28, 297–302.
Akhondzadeh, S., Tabatabaee, M., Amini, H., Ahmadi Abhari, S.A., Abbasi, S.H.,
Behnam, B., 2007. Celecoxib as adjunctive therapy in schizophrenia: a
double-blind, randomized and placebo-controlled trial. Schizophrenia
Research 90, 179–185.
Akhondzadeh, S., Jafari, S., Raisi, F., Nasehi, A.A., Ghoreishi, A., Salehi, B.,
Mohebbi-Rasa, S., Raznahan, M., Kamalipour, A., 2009. Clinical trial of
adjunctive celecoxib treatment in patients with major depression: a
double blind and placebo controlled trial. Depression and Anxiety 26,
607–611.
Alesci, S., Martinez, P.E., Kelkar, S., Ilias, I., Ronsaville, D.S., Listwak, S.J.,
Ayala, A.R., Licinio, J., Gold, H.K., Kling, M.A., Chrousos, G.P., Gold, P.W.,
2005. Major depression is associated with significant diurnal eleva-
tions in plasma interleukin-6 levels, a shift of its circadian rhythm, and
loss of physiological complexity in its secretion: clinical implications.
Journal of Clinical Endocrinology and Metabolism 90, 2522–2530.
Alhan, E., Kalyoncu, N.I., Ercin, C., Kural, B.V., 2004. Effects of the celecoxib
on the acute necrotizing pancreatitis in rats. Inflammation 28, 303–
309.
Basterzi, A.D., Aydemir, C., Kisa, C., Aksaray, S., Tuzer, V., Yazici, K., Goka, E.,
2005. IL-6 levels decrease with SSRI treatment in patients with major
depression. Human Psychopharmacology 20, 473–476.
Bresee, C.J., Delrahim, K., Maddux, R.E., Dolnak, D., Ahmadpour, O.,
Rapaport, M.H., 2006. The effects of celecoxib augmentation on
cytokine levels in schizophrenia. The International Journal of Neurop-
sychopharmacology 9, 343–348.
Bukata, S.V., Gelinas, J., Wei, X., Rosier, R.N., Puzas, J.E., Zhang, X., Schwarz, E.M.,
Song, X.Y., Griswold, D.E., O'Keefe, R.J., 2004. PGE2 and IL-6 production by
fibroblasts in response to titanium wear debris particles is mediated
through a Cox-2 dependent pathway. Journal of Orthopaedic Research 22,
6–12.
Calabrese, J.R., Skwerer, R.G., Barna, B., Gulledge, A.D., Valenzuela, R., Butkus,
A., Subichin, S., Krupp, N.E., 1986. Depression, immunocompetence, and
prostaglandins of the E series. Psychiatry Research 17, 41–47.
Carlin, J.M., Borden, E.C., Sondel, P.M., Byrne, G.I., 1987. Biologic-response-
modifier-induced indoleamine 2,3-dioxygenase activity in human
peripheral blood mononuclear cell cultures. Journal of Immunology
139, 2414–2418.
Casolini, P., Catalani, A., Zuena, A.R., Angelucci, L., 2002. Inhibition of COX-
2 reduces the age-dependent increase of hippocampal inflammatory
markers, corticosterone secretion, and behavioral impairments in the
rat. Journal of Neuroscience Research 68, 337–343.
Chen, C.Y., Tzeng, N.S., Chen, Y.C., 2010. Maintenance therapy of celecoxib for
major depression with mimicking neuropsychological dysfunction.
General Hospital Psychiatry 32 (647), e7–e9.
Chen, L., Li, D.Q., Zhong, J., Wu, X.L., Chen, Q., Peng, H., Liu, S.Q., 2011. IL-17RA
aptamer-mediated repression of IL-6 inhibits synovium inflammation in
a murine model of osteoarthritis. Osteoarthritis and Cartilage 19, 711–
718.
Dantzer, R., Kelley, K.W., 2007. Twenty years of research on cytokine-
induced sickness behavior. Brain, Behavior, and Immunity 21, 153–160.
Dinan, T.G., 2009. Inflammatory markers in depression. Current Opinion in
Psychiatry 22, 32–36.
Dowlati, Y., Herrmann, N., Swardfager, W., Liu, H., Sham, L., Reim, E.K., Lanctot,
K.L., 2010. A meta-analysis of cytokines in major depression. Biological
Psychiatry 67, 446–457.
El-Ghazaly, M.A., Nada, A.S., El-Hazek, R.M., Khayyal, M.T., 2010. Effect of
selective COX-2 inhibitor, celecoxib on adjuvant-induced arthritis
model in irradiated rats. International Journal of Radiation Biology 86,
1079–1087.
Fava, M., 2009. Augmentation and combination strategies for complicated
depression. The Journal of Clinical Psychiatry 70, e40.
Frommberger, U.H., Bauer, J., Haselbauer, P., Fraulin, A., Riemann, D., Berger,
M., 1997. Interleukin-6-(IL-6) plasma levels in depression and schizo-
phrenia: comparison between the acute state and after remission.
European Archives of Psychiatry and Clinical Neuroscience 247, 228–
233.
Glaser, R., Robles, T.F., Sheridan, J., Malarkey, W.B., Kiecolt-Glaser, J.K.,
2003. Mild depressive symptoms are associated with amplified and
prolonged inflammatory responses after influenza virus vaccination in
older adults. Archives of General Psychiatry 60, 1009–1014.
Hamilton, M., 1960. A rating scale for depression. Journal of Neurology,
Neurosurgery, and Psychiatry 23, 56–62.
314 S.-H. Abbasi et al. / Journal of Affective Disorders 141 (2012) 308–314
Hannestad, J., Dellagioia, N., Bloch, M., 2011. The effect of antidepressant
medication treatment on serum levels of inflammatory cytokines: a
meta-analysis. Neuropsychopharmacology 36, 2452–2459.
Hinson, R.M., Williams, J.A., Shacter, E., 1996. Elevated interleukin 6 is
induced by prostaglandin E2 in a murine model of inflammation:
possible role of cyclooxygenase-2. Proceedings of the National Academy
of Sciences of the United States of America 93, 4885–4890.
Hu, F., Wang, X., Pace, T.W., Wu, H., Miller, A.H., 2005. Inhibition of COX-2 by
celecoxib enhances glucocorticoid receptor function. Molecular Psychi-
atry 10, 426–428.
Humphreys, D., Schlesinger, L., Lopez, M., Araya, A.V., 2006. Interleukin-6
production and deregulation of the hypothalamic–pituitary–adrenal axis in
patients with major depressive disorders. Endocrine 30, 371–376. Janssen,
D.G., Caniato, R.N., Verster, J.C., Baune, B.T., 2010. A psychoneur-
oimmunological review on cytokines involved in antidepressant treat-
ment response. Human Psychopharmacology 25, 201–215.
Kiecolt-Glaser, J.K., Belury, M.A., Porter, K., Beversdorf, D.Q., Lemeshow, S.,
Glaser, R., 2007. Depressive symptoms, omega-6:omega-3 fatty acids, and
inflammation in older adults. Psychosomatic Medicine 69, 217–224.
Leonard, B.E., 2001. Changes in the immune system in depression and
dementia: causal or co-incidental effects? International Journal of
Developmental Neuroscience 19, 305–312.
Linnoila, M., Whorton, A.R., Rubinow, D.R., Cowdry, R.W., Ninan, P.T., Waters,
R.N., 1983. CSF prostaglandin levels in depressed and schizophrenic
patients. Archives of General Psychiatry 40, 405–406.
LIU, Y., HO, R.C., MAK, A., 2011a. Interleukin (IL)-6, tumour necrosis factor alpha
(TNF-alpha) and soluble interleukin-2 receptors (sIL-2R) are elevated in
patients with major depressive disorder: a meta-analysis and meta-
regression. Journal of Affective Disorders. doi:10.1016/j.jad.2011.08.003
(Electronic publication ahead of print).
Liu, Y., Liu, A., Li, H., Li, C., Lin, J., 2011b. Celecoxib inhibits interleukin-6/
interleukin-6 receptor-induced JAK2/STAT3 phosphorylation in human
hepatocellular carcinoma cells. Cancer Prevention Research (Philadel-
phia, Pa.) 4, 1296–1305.
McIntyre, R.S., 2010. When should you move beyond first-line therapy for
depression? The Journal of Clinical Psychiatry 71 (Suppl. 1), 16–20.
McNamara, R.K., Jandacek, R., Rider, T., Tso, P., Cole-Strauss, A., Lipton, J.W.,
2010. Omega-3 fatty acid deficiency increases constitutive pro-
inflammatory cytokine production in rats: relationship with central
serotonin turnover. Prostaglandins, Leukotrienes, and Essential Fatty
Acids 83, 185–191.
Minghetti, L., 2004. Cyclooxygenase-2 (COX-2) in inflammatory and
degenerative brain diseases. Journal of Neuropathology and Experimen-
tal Neurology 63, 901–910.
Muller, N., 2010. COX-2 inhibitors as antidepressants and antipsychotics:
clinical evidence. Current Opinion in Investigational Drugs 11, 31–42.
Muller, N., Strassnig, M., Schwarz, M.J., Ulmschneider, M., Riedel, M., 2004a.
COX-2 inhibitors as adjunctive therapy in schizophrenia. Expert Opinion
on Investigational Drugs 13, 1033–1044.
Muller, N., Ulmschneider, M., Scheppach, C., Schwarz, M.J., Ackenheil, M.,
Moller, H.J., Gruber, R., Riedel, M., 2004b. COX-2 inhibition as a
treatment approach in schizophrenia: immunological considerations
and clinical effects of celecoxib add-on therapy. European Archives of
Psychiatry and Clinical Neuroscience 254, 14–22.
Muller, N., Schwarz, M.J., Dehning, S., Douhe, A., Cerovecki, A., Goldstein-
Muller, B., Spellmann, I., Hetzel, G., Maino, K., Kleindienst, N., Moller, H.J.,
Arolt, V., Riedel, M., 2006. The cyclooxygenase-2 inhibitor celecoxib has
therapeutic effects in major depression: results of a double-blind,
randomized, placebo controlled, add-on pilot study to reboxetine.
Molecular Psychiatry 11, 680–684.
Muller, N., Myint, A.M., Schwarz, M.J., 2009. The impact of neuroimmune
dysregulation on neuroprotection and neurotoxicity in psychiatric
disorders — relation to drug treatment. Dialogues in Clinical Neurosci-
ence 11, 319–332.
Muller, N., Myint, A.M., Schwarz, M.J., 2011. Inflammatory biomarkers and
depression. Neurotoxicity Research 19, 308–318.
Musso, T., Gusella, G.L., Brooks, A., Longo, D.L., Varesio, L., 1994. Interleukin-4
inhibits indoleamine 2,3-dioxygenase expression in human monocytes.
Blood 83, 1408–1411.
Myint, A.M., Steinbusch, H.W., Goeghegan, L., Luchtman, D., Kim, Y.K.,
Leonard, B.E., 2007. Effect of the COX-2 inhibitor celecoxib on
behavioural and immune changes in an olfactory bulbectomised rat
model of depression. Neuroimmunomodulation 14, 65–71.
Nery, F.G., Monkul, E.S., Hatch, J.P., Fonseca, M., Zunta-Soares, G.B., Frey, B.N.,
Bowden, C.L., Soares, J.C., 2008. Celecoxib as an adjunct in the treatment
of depressive or mixed episodes of bipolar disorder: a double-blind,
randomized, placebo-controlled study. Human Psychopharmacology 23,
87–94.
Pace, T.W., Mletzko, T.C., Alagbe, O., Musselman, D.L., Nemeroff, C.B., Miller,
A.H., Heim, C.M., 2006. Increased stress-induced inflammatory re-
sponses in male patients with major depression and increased early
life stress. The American Journal of Psychiatry 163, 1630–1633.
Prather, A.A., Rabinovitz, M., Pollock, B.G., Lotrich, F.E., 2009. Cytokine-
induced depression during IFN-alpha treatment: the role of IL-6 and
sleep quality. Brain, Behavior, and Immunity 23, 1109–1116.
Rapaport, M.H., Delrahim, K.K., Bresee, C.J., Maddux, R.E., Ahmadpour, O.,
Dolnak, D., 2005. Celecoxib augmentation of continuously ill patients
with schizophrenia. Biological Psychiatry 57, 1594–1596.
Sluzewska, A., Rybakowski, J.K., Laciak, M., Mackiewicz, A., Sobieska, M.,
Wiktorowicz, K., 1995. Interleukin-6 serum levels in depressed patients
before and after treatment with fluoxetine. Annals of the New York
Academy of Sciences 762, 474–476.
Song, C., Zhang, X.Y., Manku, M., 2009. Increased phospholipase A2 activity
and inflammatory response but decreased nerve growth factor expres-
sion in the olfactory bulbectomized rat model of depression: effects of
chronic ethyl-eicosapentaenoate treatment. Journal of Neuroscience 29,
14–22.
Su, S., Miller, A.H., Snieder, H., Bremner, J.D., Ritchie, J., Maisano, C., Jones, L.,
Murrah, N.V., Goldberg, J., Vaccarino, V., 2009. Common genetic
contributions to depressive symptoms and inflammatory markers in
middle-aged men: the Twins Heart Study. Psychosomatic Medicine 71,
152–158.
Tipton, D.A., Flynn, J.C., Stein, S.H., Dabbous, M., 2003. Cyclooxygenase-2
inhibitors decrease interleukin-1beta-stimulated prostaglandin E2 and
IL-6 production by human gingival fibroblasts. Journal of Periodontology
74, 1754–1763.
White, W.B., West, C.R., Borer, J.S., Gorelick, P.B., Lavange, L., Pan, S.X.,
Weiner, E., Verburg, K.M., 2007. Risk of cardiovascular events in patients
receiving celecoxib: a meta-analysis of randomized clinical trials. The
American Journal of Cardiology 99, 91–98.