Therapeutics

Early Improvements in Individual Symptoms The Journal of Clinical Pharmacology

2016, 56(9) 1111–1119
to Predict Later Remission in Major 
C 2016, The American College of

Clinical Pharmacology
DOI: 10.1002/jcph.710
Depressive Disorder Treated
With Mirtazapine

Kei Funaki, MD1 , Shinichiro Nakajima, MD, PhD1,2,3 , Takefumi Suzuki, MD, PhD1,4 ,
Masaru Mimura, MD, PhD1 , and Hiroyuki Uchida, MD, PhD1,2

Abstract
Few studies, to our knowledge, have examined whether early improvements in individual, instead of overall, depressive symptoms predict remission
in major depressive disorder (MDD). This post hoc analysis used data from 194 patients with MDD enrolled in a 6-week double-blind, placebo-
controlled, randomized trial of mirtazapine, to identify improvements in specific individual depressive symptoms in the early phase that are associated
with subsequent remission. Trajectories of individual depressive symptoms over 6 weeks were compared between remitters and nonremitters. Early
improvement was defined as a ࣙ20% decrease in the Hamilton Rating Scale for Depression 17 items (HAM-D17) total score in weeks 1 and 2, and
remission was defined as a HAM-D17 final score of ࣘ7. Reliability parameters were calculated for early improvements in predicting later remission.
Whether improvement in each of the HAM-D17 symptoms in weeks 1 or 2 predicted remission was examined, using binary logistic regression analyses.
As a result, improvements in weeks 1 and 2 were associated with sensitivity of 0.82 and 0.99 and specificity of 0.54 and 0.44, respectively, in predicting
remission in week 6. Improvements in insomnia late (P = .04) and insight (P = .007) in week 1 and somatic symptoms general (P = .002) and insight (P =
.04) in week 2 were associated with remission in week 6. In conclusion, early improvements in insight, insomnia late, and somatic symptoms general,
as well as overall depressive symptoms, may serve as specific clinical indicators of subsequent remission in patients with MDD receiving mirtazapine.

Keywords
antidepressant, depression, early improvement, mirtazapine, remission, prediction

Antidepressants play an important role in the treatment
for major depressive disorder (MDD), especially when
illness severity is moderate to severe.1,2 Nevertheless,
50% to 60% of people with depression fail to show
sufficient treatment response with currently available
antidepressants.3 It is therefore important to forecast at
the earliest possible stages of the illness which antide-
pressant is effective in an effort to reduce the suffering
of patients. It has already been identified that early
improvement or response with antidepressants serves
to predict later remission regardless of the baseline
severity.4–8 However, most of the studies thus far have
only focused on overall depressive symptoms, concen-
trating on the sum of the scores in the representative
rating scales (ie, the Hamilton Rating Scale for De-
pression 17 items [HAM-D17]9 and the Montgomery-
Åsberg Depression Rating Scale10 ). It is crucial to
Treatment Alternatives to Relieve Depression study.11
They noted that improvements with citalopram in sad
mood, low energy, feeling slowed down, restlessness,
and negative self-view in week 2 were associated with
remission in week 14. We are unaware of another report
that has examined trajectories in specific depressive
symptoms or symptom clusters to effectively predict
remission in MDD.
Mirtazapine is classified as a noradrenergic and
specific serotonergic antidepressant (NaSSA) that
1 Department
of Neuropsychiatry, Keio University School of Medicine,
Tokyo, Japan
2 Geriatric Psychiatry Division, Centre for Addiction and Mental Health,
Toronto, ON, Canada
3 Department of Psychiatry, University of Toronto, ON, Canada
4 Department of Psychiatry, Inokashira Hospital, Tokyo, Japan

detect individual depressive symptoms whose improve-
ments predict later remission for each patient in daily
clinical practice. For example, Sakurai et al examined
whether trajectories of individual depressive symptoms
over time could predict remission in the acute phase
in MDD by analyzing the data from the Sequenced
Submitted for publication 29 October 2015; accepted 18 January 2016.
Corresponding Author:
Shinichiro Nakajima, MD, PhD, Department of Neuropsychiatry, Keio
University School of Medicine, 35, Shinanomachi, Shinjuku-ku, Tokyo,
160–8582, Japan
Email: shinichiro_nakajima@hotmail.com
1112
presumably acts by antagonizing adrenergic alpha 2-
autoreceptors and alpha 2-heteroreceptors in addition
to blocking both serotonin 5-hydroxytryptamine (5-
HT)2 and 5-HT3 receptors.12 In detail, antagonism
of presynaptic alpha 2-adrenergic autoreceptors is
suggested to increase norepinephrine release, whereas
blockade of alpha 2-heteroreceptors, contained in sero-
tonergic neurons, is assumed to enhance the release of
serotonin. In addition, by blocking 5-HT2, particularly
subtypes 2A and 2C, and 5-HT3 receptors, mirtaza-
pine increases serotonin release, which is expected to
result in enhancement of 5-HT1A-mediated seroton-
ergic transmission.13,14 This dual mode of action may
plausibly be responsible for mirtazapine’s rapid onset
of action. The recent Cochrane Database Systematic
Review indicated that mirtazapine was found to be
superior to other antidepressants during the acute
phase of MDD treatment and that it could also have a
faster onset of action than selective serotonin reuptake
inhibitors.15
Considering the unique pharmacological profile of
mirtazapine, which is different from mechanisms of
action of other antidepressants (eg, serotonin and
norepinephrine reuptake inhibitors), to inhibit sero-
tonin or noradrenaline transporters, the trajectories of
specific symptoms in patients may differ from those
of other classical or newer antidepressants. Therefore,
in this study, we evaluated how individual depressive
symptoms initially progress over time between remitters
and nonremitters using the data from a 6-week double-
blind, placebo-controlled, randomized trial of mirtaza-
pine. Alternatively, the objective of this analysis was
to examine the usefulness of early treatment improve-
ments (in weeks 1 and 2) in depressive symptomatology
in predicting remission in week 6 among patients with
MDD evaluated individually by symptoms.
Methods
Study Design
Following a complete description of the study, partici-
pants provided written informed consent at enrollment
in the original study, which was approved by the in-
stitutional review board of all participating sites. This
post hoc analysis used data that were made completely
anonymous; thus, ethical approval was not sought for
this study. This study analyzed data from a multicenter
randomized, double-blind, placebo-controlled 6-week
phase 2 trial of mirtazapine conducted from November
2004 to December 2005 in Japan.16 Study inclusion
criteria were: (1) aged 20–75 years old; (2) primary
diagnosis of MDD according to the Diagnostic and
Statistical Manual of Mental Disorders, Fourth Edition
(DSM-IV)17 ; (3) a baseline total score of 18 or higher
on the HAM-D17; and (4) capability of providing
The Journal of Clinical Pharmacology / Vol 56 No 9 2016
informed consent. Patients were excluded if they were
treated by another antidepressant that was effective
for them, were treated with electroconvulsive therapy
before this trial, or were pregnant or lactating. Patients
were also excluded if they had already received mirtaza-
pine for the current episode, were treated for diabetes,
had a psychiatric comorbidity, showed significant sui-
cidal ideation (ie, score of 3 or higher on item 11 of
the HAM-D17), had a history of epilepsy, had a history
of treatment with mood stabilizers (ie, lithium, sodium
valproate, or carbamazepine) within 2 weeks, or had a
significant neurological or general medical condition.
The participants were randomized to 1 of the
following 4 treatment arms: mirtazapine 15 mg/day
(15 mg/day for 6 weeks), 30 mg/day (15 mg/day for
the first week and 30 mg/day for another 5 weeks),
45 mg/day (15 mg/day for the first week, 30 mg/day
for the second week, and 45 mg/day for another
4 weeks), and a placebo group. The study drugs were
all administered once daily before bedtime. The HAM-
D17 was completed at baseline and weekly through
week 6. Study physicians, participants, and raters were
all blinded throughout the study.
Statistical Analysis
In accordance with previous work of this sort, early
improvement was defined as a ࣙ20% decrease in HAM-
D17 total scores in weeks 1 and 2, and remission
was defined as a final HAM-D17 score ࣘ 7.18 Base-
line sociodemographic and clinical characteristics were
compared between remitters and nonremitters by the
Mann-Whitney test and the chi-square test or Fisher’s
exact test for continuous and categorical variables,
respectively. Sensitivity, specificity, positive predictable
value (PPV), and negative predictable value (NPV) of
early improvement in predicting remission in week 6
were each calculated. Then scores for individual symp-
toms on the HAM-D17 throughout the 6 weeks were
extracted. Average trajectories of mean scores over
time for each symptom in mirtazapine groups were
estimated by means of 2 data sets: (1) observed case
(OC) analysis, which included all participants with
nonmissing observations at each time; and (2) last
observation carried forward (LOCF) analysis, in which
the score at the previous visit was adopted in case
of premature attrition. Binary logistic regression anal-
yses were performed to evaluate the association be-
tween individual symptom(s) in the early phase and
remission at treatment exit. In this analysis, remission
in week 6 was the dependent variable. We employed
baseline sociodemographic and clinical characteristics
that were statistically different between the 2 groups
and the changes of each individual symptom from
week 0 to week 1 or 2 as the independent variables.
A P < .05 was considered statistically significant
Funaki et al 1113

Table 1. Baseline Sociodemographic and Clinical Characteristics

Characteristics Remitters (n = 89) Nonremitters (n = 105) Statistics

Age (y), mean ± SD (range) 38.7 ± 10.9 (20–64) 40.1 ± 12.0 (21–74) U = 4438 Z = –0.60 P = .55a
Women, n (%) 47 (52.8) 54 (51.4) df = 1 χ2 = 0.04 P = .85b
Race
Japanese (%) 89 (100.0) 105 (100.0)
Weight (kg), mean ± SD (range) 58.9 ± 10.5 (31.8–98.0) 60.0 ± 13.7 (38.5–119.0) U = 4641 Z = –0.08 P = .94a
Outpatient, n (%) 89 (100) 102 (97.1) P = .25c
DSM-IV-TR diagnosis
Single episode, n (%) 57 (64.0) 74 (70.5) df = 1 χ2 = 0.91 P = .34b
Recurrent episodes, n (%) 32 (36.0) 31 (29.5)
DSM-IV-TR severity
Mild, n (%) 11 (12.3) 8 (7.6) P = .54c
Moderate, n (%) 74 (83.1) 90 (85.7)
Severe, n (%) 4 (4.5) 7 (6.7)
Number of depressive episodes
First time, n (%) 57 (64.0) 74 (70.5) P = .22c
Second time and more, n (%) 23 (25.8) 27 (25.7)
Others, n (%) 9 (10.1) 4 (3.8)
Presence of pretreatment antidepressant, n (%) 7 (7.9) 26 (24.8) P = .002c
Presence of concomitant disease, n (%) 37 (41.6) 48 (45.7) df = 1 χ2 = 0.34 P = .56b
HAM-D17 total score at baseline, mean ± SD (range) 21.7 ± 3.0 (18–30) 23.5 ± 4.0 (18–36) U = 3496.5 Z = –3.03 P = .002a
Duration of this episode (mo)
<1, n (%) 3 (3.4) 1 (1.0) P = .22c
1 – ࣘ6, n (%) 48 (54.5) 51 (49.0)
6 – ࣘ12, n (%) 13 (14.8) 11 (10.6)
>12, n (%) 24 (27.3) 41 (39.4)
mean ± SD (range) 11.3 ± 18.3 (0–114) 15.0 ± 22.1 (0–132) U = 3743 Z = –2.18 P = .03a
Duration from first episode (mo), mean ± SD (range) 39.2 ± 57.6 (0–360) 41.9 ± 65.2 (1–360) U = 4309.5 Z = –0.34 P = .73a

Statistically significant (P < .05) values are highlighted in bold.

a
Mann-Whitney U test.
b
Chi-square test.
c
Fisher’s exact test.

Table 2. Number of Patients Who Failed to Continue Throughout the Study

Baseline Week 1 Week 2 Week 3 Week 4 Week 5 Week 6

Total number of dropouts (%) 0 (0.0) 1 (0.5) 5 (2.6) 12 (6.2) 23 (11.9) 31 (16.0) 33 (17.0)
Total number of evaluable subjects (%) 194 (100) 193 (99.5) 189 (97.4) 182 (93.8) 171 (88.1) 163 (84.0) 161 (83.0)

(2 sided). Statistical analyses were performed with IBM Table 3a. Numbers of Those With Improvement in Week 1 and Those
SPSS Statistics version 22.0. With Remission in Week 6

Week 1 Remission Nonremission

Results
HAM-D17 improvement ࣙ 20% 73 48 121
Baseline Characteristics HAM-D17 improvement< 20% 16 57 73
Sociodemographic and clinical characteristics of 194
89 105 194
participants are summarized in Table 1. Participants
were Japanese outpatients aged about 40 years on
average with a nearly equal male-female ratio. Two-
thirds had had a first depressive episode. The severity of lower HAM-D17 total score at baseline, and had a
episodes was moderate in more than 80% of our sample. shorter duration of the current depressive episode than
The dropout rate was 17.0% in week 6. The reasons did nonremitters (Table 1).
for dropout were mainly withdrawal of consent (about
40%), adverse events (about 20%), and worsening of Trajectories of Individual Symptoms
depression or suicide score in HAM-D17 (about 20%); Tables 3a and 3b summarize how many showed
see Table 2. There were 89 remitters (45.9%). Remitters improvement in week 1 and in week 2 and how many
received more pretreatment antidepressants, showed a were in remission in week 6, respectively. Figure 1a,b
1114 The Journal of Clinical Pharmacology / Vol 56 No 9 2016

Table 3b. Numbers of Those With Improvement in Week 2 and Those highly sensitive predictor of remission; improvement
With Remission in Week 6 in week 1 was associated with a sensitivity of 0.82, a
Week 2 Remission Nonemission specificity of 0.54, a PPV of 0.60, and an NPV of 0.78
in predicting remission status in week 6. The sensitivity,
HAM-D17 improvement ࣙ 20% 88 59 147
specificity, PPV, and NPV of improvement in week 2 in
HAM-D17 improvement < 20% 1 46 47
predicting remission status in week 6 were 0.99, 0.44,
89 105 194 0.60, and 0.98, respectively.
HAM-D17, Hamilton Rating Scale for Depression 17 items.
In the binary logistic regression analysis with LOCF,
improvements in weeks 1 and 2 in the following symp-
toms in the HAM-D17 were significantly associated
depicts the trajectories of HAM-D17 scores for with remission at treatment exit: insomnia late and
improvers and nonimprovers in week 1 and in week 2 in insight in week 1 and somatic symptoms general and
the mirtazapine group and in the whole placebo group, insight in week 2 (Table 4). In addition, associations
respectively. Early improvement was identified as a between remission and baseline HAM-D17 total score

25

20

15 Mirtazapine early improver

group at week 1
HAM-D17
Mirtazapine early nonimprover
10 group at week 1
Placebo group

(a)

25

20

15 Mirtazapine early improver

group at week 2
HAM-D17
Mirtazapine early non-improver
10 group at week 2
Placebo group

(b)
Figure 1. (a) Trajectories of HAM-D17 scores for improvers and nonimprovers in week 1 in mirtazapine group and whole placebo group.
(b) Trajectories of HAM-D17 scores for improvers and nonimprovers in week 2 in mirtazapine group and whole placebo group.
Funaki et al 1115

Table 4. Association Between Early Improvements in Individual Symptoms in the HAM-D17 and Subsequent Remission in Patients With MDD

Score Changes From Baseline (Range)

Improvement in Week 1 Remitters Nonremitters P Odds Ratio 95%CI

Depressed mood 0.85 ± 0.87 (–1 to 3) 0.50 ± 0.85 (–1 to 3) .93 1.02 0.61–1.72
Feelings of guilt 0.62 ± 0.86 (–1 to 3) 0.31 ± 0.63 (–1 to 2) .48 1.21 0.71–2.05
Suicide 0.44 ± 0.58 (0 to 2) 0.27 ± 0.52 (–1 to 2) .63 1.19 0.58–2.45
Insomnia early 0.57 ± 0.88 (–2 to 2) 0.44 ± 0.96 (–2 to 2) .57 1.13 0.74–1.74
Insomnia middle 0.74 ± 0.86 (–1 to 2) 0.43 ± 0.84 (–2 to 2) .57 1.15 0.71–1.88
Insomnia late 0.83 ± 0.86 (–1 to 2) 0.42 ± 0.91 (–2 to 2) .04a 1.65 1.02–2.67
Work and activities 0.42 ± 0.74 (–1 to 3) 0.10 ± 0.71 (–3 to 2) .85 1.06 0.59–1.90
Retardation: psychomotor 0.37 ± 0.59 (–1 to 2) 0.15 ± 0.53 (–1 to 2) .15 1.68 0.83–3.41
Agitation 0.37 ± 0.59 (–1 to 2) 0.18 ± 0.65 (–2 to 2) .43 1.30 0.68–2.48
Anxiety (psychological) 0.72 ± 0.81 (–1 to 3) 0.29 ± 0.94 (–2 to 3) .29 1.32 0.79–2.19
Anxiety somatic 0.51 ± 0.79 (–1 to 3) 0.24 ± 0.60 (–1 to 2) .28 1.34 0.79–2.24
Somatic symptoms (gastrointestinal) 0.31 ± 0.60 (–1 to 2) 0.24 ± 0.63 (–2 to 2) .78 0.92 0.50–1.69
Somatic symptoms general 0.45 ± 0.64 (–1 to 2) 0.17 ± 0.56 (–2 to 2) .07 1.81 0.95–3.45
Genital symptoms 0.15 ± 0.51 (–1 to 2) 0.05 ± 0.40 (–2 to 1) .71 1.19 0.48–2.92
Hypochondriasis 0.25 ± 0.68 (–2 to 2) 0.14 ± 0.77 (–2 to 3) .94 0.98 0.55–1.76
Loss of weight 0.27 ± 0.64 (1 to 2) 0.22 ± 0.71 (–2 to 2) .88 1.04 0.60–1.83
Insight 0.19 ± 0.47 (–1 to 2) –0.03 ± 0.38 (–1 to 1) .007a 3.49 1.42–8.60

Score Changes From Baseline (Range)

Improvement in Week 2 Remitters Nonremitters P Odds Ratio 95%CI

Depressed mood 1.37 ± 0.96 (–1 to 3) 0.72 ± 1.01 (–2 to 3) .26 1.31 0.82–2.11
Feelings of guilt 0.83 ± 0.82 (–1 to 2) 0.43 ± 0.81 (–2 to 3) .26 1.35 0.81–2.25
Suicide 0.69 ± 0.76 (0 to 2) 0.35 ± 0.62 (–1 to 2) .21 1.49 0.80–2.80
Insomnia early 0.69 ± 0.91 (–1 to 2) 0.57 ± 0.97 (–2 to 2) .25 1.30 0.83–2.05
Insomnia middle 0.82 ± 0.87 (–1 to 2) 0.63 ± 0.74 (–1 to 2) .80 1.08 0.60–1.94
Insomnia late 0.97 ± 0.83 (–1 to 2) 0.71 ± 0.94 (–2 to 2) .05 1.68 1.00–2.84
Work and activities 0.92 ± 0.84 (–1 to 3) 0.26 ± 0.84 (–2 to 2) .08 1.58 0.95–2.63
Retardation: psychomotor 0.55 ± 0.62 (–1 to 2) 0.32 ± 0.58 (–1 to 2) .52 1.30 0.58–2.90
Agitation 0.51 ± 0.64 (–1 to 2) 0.30 ± 0.79 (–2 to 2) .86 1.06 0.56–2.01
Anxiety (psychological) 1.13 ± 1.05 (–2 to 3) 0.62 ± 1.09 (–2 to 3) .65 1.11 0.71–1.74
Anxiety somatic 0.82 ± 0.95 (–2 to 3) 0.44 ± 0.73 (–1 to 3) .27 1.32 0.81–2.17
Somatic symptoms (gastrointestinal) 0.44 ± 0.62 (–1 to 2) 0.36 ± 0.65 (–2 to 2) .36 0.73 0.37–1.44
Somatic symptoms general 0.66 ± 0.69 (–1 to 2) 0.11 ± 0.54 (–1 to 2) .002a 3.05 1.52–6.11
Genital symptoms 0.30 ± 0.55 (–1 to 2) 0.09 ± 0.57 (–2 to 2) .36 1.47 0.65–3.30
Hypochondriasis 0.44 ± 0.67 (–1 to 2) 0.19 ± 0.91 (–2 to 4) .62 1.14 0.68–1.91
Loss of weight 0.38 ± 0.73 (–1 to 2) 0.36 ± 0.77 (–2 to 2) .24 1.44 0.79–2.64
Insight 0.26 ± 0.53 (–1 to 2) –0.02 ± 0.44 (–1 to 1) .04a 2.54 1.05–6.13

CI, confidence interval; HAM-D17, Hamilton Rating Scale for Depression 17 items; MDD, major depressive disorder; SD, standard deviation.
Binary logistic regression analyses were performed with the last observation carried forward analysis. Values are shown as mean ± SD.
a
Significant P (ࣘ .05).

were found (P = .004 for week 1 analysis and P < .001
for week 2 analysis), whereas there were no associations
between remission and pretreatment antidepressants
(P = .22, and P = .47, respectively) or duration of
the current depressive episode (P = .42, and P = .81,
respectively). The OC analysis showed similar results
(data available on request). Trajectories of individual
symptoms of HAM-D17 in remitters and nonremitters
are summarized in Table 5.
Discussion
To our knowledge, this is the first study to exam-
ine whether early improvements with mirtazapine in
depressive symptomatology may serve as a clinically
useful marker for remission during the acute-phase
treatment of MDD. First of all, early improvement in
overall depressive symptoms strongly predicted later re-
mission with mirtazapine, which is compatible with the
preceding studies.6,8 Second, improvements in weeks
1 and 2 in the following symptoms in the HAM-
D17 were associated with remission status in week
6: insomnia late and insight in week 1 and somatic
symptoms general and insight in week 2. Finally, apart
from our main interest, the baseline HAM-D17 total
score predicted remission at the end point.
Predicting pharmacologic response in MDD at the
earliest possible opportunity is important to minimize
 1116

Table 5. Trajectories of Individual Symptoms of HAM-D17 in Remitters and Nonremitters (LOCF)

Baseline Week 1 Week 2 Week 6

Mean ± SD (Range) Remitters Nonremitters Remitters Nonremitters Remitters Nonremitters Remitters Nonremitters

Depressed mood 2.6 ± 1.1 (1–4) 2.7 ± 0.8 (0–4) 1.7 ± 1.0 (0–4) 2.2 ± 1.0 (0–4) 1.2 ± 1.0 (0–4) 2.0 ± 1.0 (0–4) 0.4 ± 0.9 (0–2) 1.6 ± 1.1 (0–4)
Feelings of guilt 1.5 ± 0.9 (0–3) 1.5 ± 0.8 (0–4) 0.9 ± 0.9 (0–3) 1.2 ± 0.9 (0–3) 0.7 ± 0.8 (0–3) 1.1 ± 0.9 (0–3) 0.2 ± 0.5 (0–3) 0.9 ± 0.8 (0–3)
Suicide 0.9 ± 0.8 (0–2) 1.0 ± 0.8 (0–2) 0.5 ± 0.7 (0–2) 0.8 ± 0.8 (0–3) 0.2 ± 0.6 (0–2) 0.7 ± 0.8 (0–3) 0.0 ± 0.2 (0–1) 0.5 ± 0.8 (0–3)
Insomnia early 1.1 ± 0.8 (0–2) 1.4 ± 0.8 (0–2) 0.5 ± 0.8 (0–2) 1.0 ± 0.9 (0–2) 0.4 ± 0.7 (0–2) 0.8 ± 0.9 (0–2) 0.2 ± 0.5 (0–2) 0.7 ± 0.9 (0–2)
Insomnia middle 1.2 ± 0.8 (0–2) 1.3 ± 0.7 (0–2) 0.5 ± 0.7 (0–2) 0.8 ± 0.7 (0–2) 0.4 ± 0.6 (0–2) 0.6 ± 0.7 (0–2) 0.1 ± 0.4 (0–2) 0.6 ± 0.7 (0–2)
Insomnia late 1.3 ± 0.8 (0–2) 1.2 ± 0.8 (0–2) 0.4 ± 0.7 (0–2) 0.8 ± 0.9 (0–2) 0.3 ± 0.5 (0–2) 0.5 ± 0.7 (0–2) 0.1 ± 0.3 (0–1) 0.5 ± 0.7 (0–2)
Work and activities 2.3 ± 1.1 (1–4) 2.7 ± 1.0 (0–4) 1.9 ± 1.1 (0–4) 2.6 ± 1.0 (1–4) 1.4 ± 1.1 (0–4) 2.4 ± 1.0 (1–4) 0.5 ± 1.1 (0–2) 1.9 ± 1.0 (0–4)
Retardation: psychomotor 1.0 ± 0.7 (0–2) 1.2 ± 0.7 (0–3) 0.7 ± 0.7 (0–2) 1.0 ± 0.7 (0–3) 0.5 ± 0.6 (0–2) 0.9 ± 0.7 (0–2) 0.1 ± 0.3 (0–1) 0.7 ± 0.7 (0–2)
Agitation 0.9 ± 0.8 (0–2) 0.8 ± 0.8 (0–2) 0.5 ± 0.8 (0–2) 0.6 ± 0.7 (0–3) 0.4 ± 0.7 (0–2) 0.4 ± 0.8 (0–3) 0.0 ± 0.2 (0–1) 0.5 ± 0.8 (0–3)
Anxiety (psychological) 2.3 ± 1.0 (0–4) 2.3 ± 0.8 (0–4) 1.6 ± 1.0 (0–4) 2.0 ± 0.9 (0–4) 1.2 ± 1.0 (0–3) 1.7 ± 1.0 (0–4) 0.3 ± 0.5 (0–2) 1.3 ± 1.0 (0–4)
Anxiety somatic 1.7 ± 0.8 (0–3) 1.7 ± 0.7 (0–3) 1.2 ± 0.8 (0–3) 1.4 ± 0.8 (0–3) 0.9 ± 0.8 (0–3) 1.2 ± 0.8 (0–3) 0.4 ± 0.5 (0–2) 1.1 ± 0.9 (0–3)
Somatic symptoms (gastrointestinal) 0.7 ± 0.6 (0–2) 0.9 ± 0.6 (0–2) 0.3 ± 0.5 (0–2) 0.7 ± 0.6 (0–2) 0.2 ± 0.5 (0–2) 0.5 ± 0.6 (0–2) 0.0 ± 0.3 (0–1) 0.5 ± 0.6 (0–2)
Somatic symptoms general 1.4 ± 0.6 (0–2) 1.4 ± 0.6 (0–2) 1.0 ± 0.6 (0–2) 1.3 ± 0.6 (0–2) 0.8 ± 0.6 (0–2) 1.3 ± 0.6 (0–2) 0.3 ± 0.5 (0–2) 1.1 ± 0.6 (0–2)
Genital symptoms 1.1 ± 0.8 (0–2) 1.2 ± 0.8 (0–2) 0.9 ± 0.8 (0–2) 1.2 ± 0.8 (0–2) 0.8 ± 0.8 (0–2) 1.2 ± 0.8 (0–2) 0.5 ± 0.8 (0–2) 1.0 ± 0.8 (0–2)
Hypochondriasis 0.8 ± 0.8 (0–3) 1.0 ± 0.9 (0–4) 0.6 ± 0.7 (0–2) 0.9 ± 0.9 (0–3) 0.4 ± 0.6 (0–2) 0.8 ± 0.9 (0–4) 0.1 ± 0.3 (0–1) 0.6 ± 0.9 (0–4)
Loss of weight 0.5 ± 0.7 (0–2) 0.8 ± 0.8 (0–2) 0.2 ± 0.6 (0–2) 0.6 ± 0.8 (0–2) 0.1 ± 0.5 (0–2) 0.4 ± 0.7 (0–2) 0.0 ± 0.3 (0–1) 0.5 ± 0.7 (0–2)
Insight 0.5 ± 0.5 (0–2) 0.3 ± 0.5 (0–1) 0.3 ± 0.4 (0–1) 0.3 ± 0.5 (0–2) 0.2 ± 0.4 (0–1) 0.3 ± 0.5 (0–2) 0.1 ± 0.2 (0–1) 0.3 ± 0.5 (0–2)

HAM-D17, Hamilton Rating Scale for Depression 17 items; LOCF, last observation carried forward; SD, standard deviation.
Individual symptoms in week 1 or 2 that are significantly associated with remission in week 6, are highlighted in bold.
The Journal of Clinical Pharmacology / Vol 56 No 9 2016
Funaki et al
the severity of both depressive symptoms and adverse
events such as increased appetite, hypersomnia, seda-
tion, and dry mouth that can be bothersome with mir-
tazapine treatment. Furthermore, our study noted that
the NPV was higher than the PPV in predicting sub-
sequent remission in our patients. This finding is com-
patible with a previous study suggesting a pessimistic
outlook for achieving response or remission among
patients without any improvement within the first 2
weeks of antidepressant treatment.8 Indeed, as few as
2.4% of the patients who had not yet improved after
2 weeks of treatment with mirtazapine subsequently
achieved remission in the present study. Therefore, these
findings suggest that early nonimprovement may be a
more reliable marker to predict subsequent remission
than early improvement in patients with MDD.
Only a few studies thus far paid attention to symp-
tomatic progress of individual symptoms or symptom
clusters over the course of antidepressant treatment
in an effort to predict the efficacy of antidepressants.
Sakurai et al reported improvements with citalopram
in decreased concentration, sad mood, feeling slowed
down, negative self-view, early-morning insomnia, low
energy, decreased general interest, and sleep-onset in-
somnia in the 16-item Quick Inventory of Depres-
sive Symptomatology, Clinician-Rating (QIDS-C16 ) in
week 2 were associated with remission in week 14.11
Our finding was partially congruent with this finding
in that somatic symptoms in general in the HAM-D17
and low energy and feeling slowed down in the QIDS-
C16 can both be interpreted as representing fatigue
and loss of energy. In addition, both these studies
are also consistent in that remission can be predicted
by improvement in insomnia. Direct comparisons of
the results, however, are challenging in light of di-
verse clinical settings and rating scales employed in
each study. Further, also challenging is the different
antidepressants studied; mirtazapine, classified as an
NaSSA, is claimed to be a more adrenergic agent than
citalopram, a selective serotonin reuptake inhibitor.
Side-effect profiles also differ between the two. Thus,
further research is needed to examine whether early im-
provements in individual depressive symptoms predict
subsequent remission in patients with MDD who are
treated with various antidepressants in different clinical
settings.
Clinical relevance of somatic symptoms cannot be
overlooked in patients with MDD,19 including general
fatigue, uncomfortable feeling of the pharynx, nonspe-
cific musculoskeletal complaints, and back pain. As
is compatible with our finding, it has been reported
that early improvement in somatic symptoms may be
a key factor to achievint remission in patients with
MDD.20,21 Previous studies have also noted that poor
sleep quality was associated with a poor response to
1117
treatments of depression,22–26 which is keeping with
our results. Furthermore, lack of insight is regarded as
one of the features of endogenous and core depressive
symptoms.27 Ghaemi et al reported on correlations
between improvement in impairment of insight and
good outcomes in 101 patients with affective and anx-
iety disorders.28 These findings suggest a utility of
insight in the successful management of the illness,
and further early improvement in lack of insight may
be associated with subsequent remission in patients
with MDD. Taking into consideration that impaired
insight may lead to poor adherence to antidepressant
treatment, impaired insight should be a target of active
intervention to achieve recovery in this population.
Limitations
The results of this study need to be interpreted in light
of several limitations. First, the generalizability of our
findings may be limited to Japanese patients with MDD
treated with mirtazapine. The unique pharmacological
profile of mirtazapine needs to be taken into careful
account. Further, the diagnosis of MDD according to
the DSM-IV criteria includes heterogeneous popula-
tions in terms of clinical features. Second, this study
is a post hoc analysis of a study lasting for only 6
weeks; the original trial was not designed to assess
trajectories of individual depressive symptoms but to
evaluate overall efficacy and tolerability of mirtazapine
in MDD in the short term. In addition, because of the
short-term design of the original study, we were not able
to adopt the definition of full remission that included
the duration of clinical stabilization (eg, 2 weeks or
longer).18 Third, the rating scale used in the study was
the exclusively HAM-D17, an objective rating scale;
other assessment scales, including subjective ones such
as the QIDS-SR,29 may have yielded different insights.
Fourth, LOCF assumes that patients will have the same
level of depression at the end of the study as when
they drop out. As a result, this imputation method
may not estimate treatment effect of drugs perfectly.
Fifth, we treated 3 different dose groups (15 mg/day,
30 mg/day, and 45 mg/day) altogether as a mirtazapine
group. Although it may have been ideal to include
dosage information in our analysis in light of possible
dose–response relationship of mirtazapine,30 no dif-
ference was found in the distribution of mirtazapine
doses in week 2 between remitters and nonremitters
by the Mann-Whitney test (P = .63). Considering that
the small sample size may have resulted in this null
finding, further research is clearly needed to enroll
participants with a size sufficiently large to divide them
based on the dosing schedule. Sixth, in light of wide
inter- and intraindividual pharmacokinetic variance of
mirtazapine,31,32 it may have been ideal to measure the
1118
plasma concentrations to elucidate optimal plasma
drug concentrations. Finally, we did not correct for mul-
tiple comparisons when conducting the binary logistic
regression on the association between early improve-
ments in individual symptoms and remission because
the purpose of this study was to describe if early
improvements in individual symptoms also predicted
remission. However, it would have been ideal to correct
for multiple comparisons.
Conclusion
The findings from the present study showed that early
improvement with mirtazapine was a highly sensitive
and robust predictor of remission in patients with
MDD, replicating previous findings with other an-
tidepressant medications. Taking into account that
early improvements in insomnia late, insight, and so-
matic symptoms general predicted subsequent remis-
sion, clinicians are advised to focus on these symptoms
in the earliest phase to facilitate remission in this
population. The data were limited to Asian patients
with nonpsychotic MDD who were treated with mir-
tazapine; further large-scale replication studies with
other antidepressants in different populations with
MDD are needed to determine individual symptoms or
symptom clusters in which early improvements serve as
a predictor for remission in patients with MDD.
Acknowledgments
The authors thank Meiji-Seika Pharma for sharing the data.
Declaration of Conflicting Interests
Dr. Funaki has received speaker’s honoraria from Dainip-
pon Sumitomo Pharma, Eisai, Otsuka Pharmaceutical, and
Yoshitomi Yakuhin and manuscript fees from Dainippon
Sumitomo Pharma within the past 3 years. Dr. Nakajima has
received a fellowship grant from the Japan Society for the
Promotion of Science and manuscript fees from Dainippon
Sumitomo Pharma and Kyowa Hakko Kirin within the past
3 years. Dr. Suzuki has received manuscript or speaker’s
fees from Astellas, Dainippon Sumitomo, Eli Lilly, Elsevier
Japan, Janssen, Meiji-Seika, Novartis, Otsuka, and Wiley
Japan within the past 3 years. Dr. Mimura has received
grants and/or speaker’s honoraria from Asahi Kasei Pharma,
Astellas, Daiichi-Sankyo, Dainippon Sumitomo Pharma,
Eisai, Eli Lilly, GlaxoSmithKline, Janssen Pharmaceutical,
Meiji-Seika Pharma, Mochida Pharmaceutical, MSD, No-
vartis Pharma, Otsuka Pharmaceutical, Pfizer, Shionogi,
Takeda, Mitsubishi Tanabe Pharma, and Yoshitomi Yakuhin
within the past 3 years. Dr. Uchida has received grants
from Astellas Pharmaceutical, Eisai, Otsuka Pharmaceutical,
GlaxoSmithKline, Shionogi, Dainippon-Sumitomo Pharma,
Eli Lilly, Mochida Pharmaceutical, Meiji-Seika Pharma,
and Yoshitomi Yakuhin and speaker’s honoraria from
The Journal of Clinical Pharmacology / Vol 56 No 9 2016
Otsuka Pharmaceutical, Eli Lilly, Shionogi, GlaxoSmithK-
line, Yoshitomi Yakuhin, Dainippon-Sumitomo Pharma,
Meiji-Seika Pharma, Abbvie, MSD, and Janssen Pharmaceu-
tical within the past 3 years.
Funding
Meiji-Seika Pharma provided phase 2 data for the study.
Role of the Funding Support
The present work was initiated, designed, and conducted by
the authors. Meiji-Seika Pharma was not involved in any of
the activities associated with the present work.
Contributors
Drs. Funaki, Nakajima, Suzuki, and Uchida led the study
design, conducted the literature review, and analyzed and in-
terpreted the data. Drs. Funaki, Nakajima, Suzuki, Mimura,
and Uchida prepared the manuscript. All authors contributed
to and approved submission of the current version of the
article.
References
1. National Institute for Health and Clinical Excellence. Depres-
sion: the treatment and management of depression in adults
(update). 2009.
2. American Psychiatric Association. Practice guideline for the
treatment of patients with major depressive disorder (third
edition). American Psychiatric Association. 2010:152.
3. Fava M. Diagnosis and definition of treatment-resistant depres-
sion. Biol Psychiatry. 2003;53(8):649–659.
4. Nakajima S, Suzuki T, Watanabe K, Kashima H, Uchida H.
Accelerating response to antidepressant treatment in depression:
a review and clinical suggestions. Progr Neuropsychopharmacol
Biol Psychiatry. 2010;34(2):259–264.
5. Katz MM, Tekell JL, Bowden CL, et al. Onset and early
behavioral effects of pharmacologically different antidepres-
sants and placebo in depression. Neuropsychopharmacology.
2004;29(3):566–579.
6. Szegedi A, Muller MJ, Anghelescu I, Klawe C, Kohnen R,
Benkert O. Early improvement under mirtazapine and parox-
etine predicts later stable response and remission with high
sensitivity in patients with major depression. J Clin Psychiatry.
2003;64(4):413–420.
7. Tadic A, Helmreich I, Mergl R, et al. Early improvement is
a predictor of treatment outcome in patients with mild ma-
jor, minor or subsyndromal depression. J Affect Disord. 2010;
120(1-3):86–93.
8. Szegedi A, Jansen WT, van Willigenburg AP, van der Meulen
E, Stassen HH, Thase ME. Early improvement in the first 2
weeks as a predictor of treatment outcome in patients with major
depressive disorder: a meta-analysis including 6562 patients. J
Clin Psychiatry. 2009;70(3):344–353.
9. Hamilton M. A rating scale for depression. J Neurol Neurosurg
Psychiatry. 1960;23:56–62.
10. Montgomery SA, Asberg M. A new depression scale designed to
be sensitive to change. Br J Psychiatry. 1979;134:382–389.
11. Sakurai H, Uchida H, Abe T, et al. Trajectories of individual
symptoms in remitters versus non-remitters with depression. J
Affect Disord. 2013;151(2):506–513.
Funaki et al
12. Anttila SA, Leinonen EV. A review of the pharmacological and
clinical profile of mirtazapine. CNS Drug Rev. 2001;7(3):249–
264.
13. Berendsen HH, Broekkamp CL. Indirect in vivo 5-HT1A-
agonistic effects of the new antidepressant mirtazapine. Psy-
chopharmacology. 1997;133(3):275–282.
14. Nakayama K, Sakurai T, Katsu H. Mirtazapine increases
dopamine release in prefrontal cortex by 5-HT1A receptor
activation. Brain Res Bull. 2004;63(3):237–241.
15. Watanabe N, Omori IM, Nakagawa A, et al. Mirtazapine versus
other antidepressive agents for depression. Cochrane Database
Syst Rev. 2011(12):CD006528.
16. Kinoshita T. A double-blind, placebo-controlled study of a new
antidepressant, mirtazapine, in depressed patients. Jpn. J. Clin.
Psychopharmacol. 2009;12(2):289–306.
17. American Psychiatric Association. Diagnostic and Statistical
Manual of Mental Disorders: DSM-IV. Washington, DC;
1994.
18. Frank E, Prien RF, Jarrett RB, et al. Conceptualization and
rationale for consensus definitions of terms in major depressive
disorder. Remission, recovery, relapse, and recurrence. Arch Gen
Psychiatry. 1991;48(9):851–855.
19. Corruble E, Guelfi JD. Pain complaints in depressed inpatients.
Psychopathology. 2000;33(6):307–309.
20. McIntyre RS, Konarski JZ, Mancini DA, et al. Improving out-
comes in depression: a focus on somatic symptoms. J Psychosom
Res. 2006;60(3):279–282.
21. Karp JF, Scott J, Houck P, Reynolds CF 3rd, Kupfer DJ,
Frank E. Pain predicts longer time to remission during treat-
ment of recurrent depression. J Clin Psychiatry. 2005;66(5):591–
597.
22. Casper RC, Katz MM, Bowden CL, Davis JM, Koslow SH,
Hanin I. The pattern of physical symptom changes in major
depressive disorder following treatment with amitriptyline or
imipramine. J Affect Disord. 1994;31(3):151–164.
1119
23. Dew MA, Reynolds CF 3rd, Houck PR, et al. Temporal profiles
of the course of depression during treatment. Predictors of
pathways toward recovery in the elderly. Arch Gen Psychiatry.
1997;54(11):1016–1024.
24. Pigeon WR, Hegel M, Unutzer J, et al. Is insomnia a perpetuat-
ing factor for late-life depression in the IMPACT cohort? Sleep.
2008;31(4):481–488.
25. Franzen PL, Buysse DJ. Sleep disturbances and depression:
risk relationships for subsequent depression and therapeutic
implications. Dialogues Clin Neurosci. 2008;10(4):473–481.
26. Manber R, Chambers AS. Insomnia and depression: a multi-
faceted interplay. Curr Psychiatry Rep. 2009;11(6):437–442.
27. Yen CF, Chen CC, Lee Y, Tang TC, Ko CH, Yen JY. Insight and
correlates among outpatients with depressive disorders. Compr
Psychiatry. 2005;46(5):384–389.
28. Ghaemi SN, Boiman E, Goodwin FK. Insight and outcome
in bipolar, unipolar, and anxiety disorders. Compr Psychiatry.
2000;41(3):167–171.
29. Rush AJ, Trivedi MH, Ibrahim HM, et al. The 16-Item Quick In-
ventory of Depressive Symptomatology (QIDS), clinician rating
(QIDS-C), and self-report (QIDS-SR): a psychometric evalua-
tion in patients with chronic major depression. Biol Psychiatry.
2003;54(5):573–583.
30. Ueno F, Nakajima S, Suzuki T, et al. Whether to increase
or maintain dosage of mirtazapine in early nonimprovers with
depression. J Clin Psychiatry. 2015;76(4):434–439.
31. Reis M, Prochazka J, Sitsen A, Ahlner J, Bengtsson F. Inter- and
intraindividual pharmacokinetic variations of mirtazapine and
its N-demethyl metabolite in patients treated for major depressive
disorder: a 6-month therapeutic drug monitoring study. Ther
Drug Monit. 2005;27(4):469–477.
32. Reis M, Aamo T, Spigset O, Ahlner J. Serum concentrations of
antidepressant drugs in a naturalistic setting: compilation based
on a large therapeutic drug monitoring database. Ther Drug
Monit. 2009;31(1):42–56.
Copyright of Journal of Clinical Pharmacology is the property of Wiley-Blackwell and its
content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.