Key words
depression - human - ketamine - treatment-resistant
Introduction
Ketamine is a dissociative anesthetic that antagonizes glutamatergic
N-methyl-D-aspartate (NMDA) receptors. Clinical trials have consistently
demonstrated that ketamine has rapid-acting and robust antidepressant effects in
patients with major depressive disorder (MDD) and bipolar depression (BD) [1]
[2]
[3]
[4]. However, retrospective chart reviews have
reported that only 18% to 45% of patients with MDD or BD respond to
acute intravenous ketamine therapy in real-world clinical settings [5]
[6].
Therefore, predicting patients with MDD or BD who are likely to benefit from
intravenous ketamine treatment is clinically relevant.
Several predictors of the efficacy of intravenous ketamine therapy have been
reported. Rong et al. (2018) reviewed 12 studies to identify potential predictors of
a successful response to intravenous ketamine infusion in patients with
treatment-resistant depression (TRD) [7]. In
their analysis, patients with MDD and BD with a family history of alcohol use
disorder showed a greater improvement in the total scores of the Montgomery-Asberg
Depression Scale (MADRS) after ketamine infusion than in patients without these
factors [8]
[9]. Moreover, higher body mass index was correlated with a greater
reduction in the 17-item Hamilton Depression Rating Scale (HAM-D-17) total score at
1 d and 230 min after ketamine infusion in patients with MDD and BD [10]. Other studies have reported the
association of following factors with better response to ketamine treatment: lower
number of treatment failures and lower baseline severity of illness [11], younger age and no history of
neuromodulation treatment [12], lower baseline
intelligence quotient [13], lower pretreatment
working memory function [14], lower
adiponectin [15], higher brain-derived
neurotrophic factor (BDNF) after ketamine administration [16], single nucleotide polymorphism (SNP) in
the Val/Val BDNF allele at rs6265 [17], and smaller left hippocampal volume [18]. However, these studies had several limitations, such as the study
design heterogeneity (e. g., a combination of open-label and randomized
placebo-controlled trials, including both MDD and BD). Recently, Alnefeesi et al.
(2022) conducted a systematic review and meta-analysis of studies evaluating the
real-world clinical effectiveness of ketamine in patients with TRD [19]. They reported that the mean number of
failed antidepressants and depressive symptomatology scores at baseline were
negatively correlated with remission rates, whereas the mean age was positively
correlated with symptom improvement in the meta-regression analysis. However, other
clinically relevant variables, such as age at onset and dissociative symptoms, have
not been thoroughly investigated. While one study reported that earlier onset age
was predictive of favorable treatment response after ketamine administration in
patients with TRD [20], the findings from
previous studies examining the association between age of onset and response to
antidepressant treatment in patients with MDD have been inconsistent [21]
[22]
[23]
[24]
[25]
[26]
[27]
[28].
In addition, the lack of agreement among previous studies is also the case for the
association between dissociative symptoms and treatment response to ketamine
infusion [29]
[30]
[31]
[32]
[33].
To fill this gap in the literature and examine the factors associated with treatment
response, we conducted a post-hoc analysis of data from a placebo-controlled,
double-blind, randomized controlled trial that assessed the acute efficacy of
ketamine infusion versus an active placebo in patients with TRD. We hypothesized
that dissociative symptoms and age of onset would be related to treatment response
in this homogeneous diagnostic group.
Methods
We conducted a post-hoc analysis of the data from a clinical trial (NCT01920555). A
detailed explanation of this trial is described elsewhere [2]. Briefly, all enrolled patients experienced
a current depressive episode lasting at least eight weeks and were diagnosed with
MDD according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth
Edition-Text Revision. The Structured Clinical Interview for the DSM-IV Patient
Edition supported the diagnosis of MDD. Additionally, all patients had TRD, which
was defined as a subjectively unsatisfactory response (i. e., less than
50% improvement in depression symptoms) to at least two adequate treatment
courses during the current depressive episode, including the current
antidepressants. At both the screening and baseline visits, all patients were
required to have a MADRS [34] total score
of≥20. In total, 99 eligible patients were randomly assigned to one of five
40-min infusion groups in a 1:1:1:1:1 ratio: a single dose of ketamine
0.1 mg/kg (n+=+18), ketamine
0.2 mg/kg (n+=+20), ketamine
0.5 mg/kg (n+=+22), ketamine
1.0 mg/kg (n+=+20), and midazolam
0.045 mg/kg (active placebo) (n+=+19). The
primary endpoint assessments were performed over 3 days, and all patients were
followed for 30 days. At each visit, the study clinicians used the Hamilton
Depression Rating Scale 6 items (HAM-D-6) as the primary outcome in all patients
(i. e., days 0, 1, 3, 5, 7, 14, and 30). All patients also underwent
assessment using the HAM-D-17 at baseline. The Clinician-Administered Dissociative
States Scale (CADSS) [35] was used to assess
dissociative symptoms 5 min before and 40, 80, and 120 min after ketamine infusion.
Data used in this post-hoc analysis were derived from patients who received a single
therapeutic dose of ketamine infusion (i. e., 0.5 or
1.0 mg/kg) and received an assessment with the HAM-D-17 [36]
[37]
at baseline and HAM-D-6 at both baseline and day 3. We selected these two dose
groups because the original study demonstrated significantly superior antidepressant
efficacy in the 0.5 and 1.0 mg/kg groups than that of the active
placebo. The treatment response in the current study was defined as a
≥50% reduction in the HAM-D-6 total score [38]
[39]
on day 3 than that of the baseline score, which was consistent with the original
trial. A local institutional review board approved the protocols, and all patients
provided written informed consent to participate in this trial. Ethical approval was
not sought for this study, which used anonymous data.
Logistic regression analysis was conducted to examine factors associated with
treatment response on day 3; the following explanatory variables were included: age,
age of onset, sex, HAM-D-6 total scores at baseline, and CADSS total scores
40 min after ketamine infusion. Moreover, multiple regression analysis was
performed using the same explanatory variables to examine the factors associated
with the percentage change in HAM-D-6 total scores on day 3 from baseline.
P<0.05 was considered statistically significant (two-tailed). Statistical
analyses were performed using R, version 4.1.0.
Results
Among the patients who participated in the original clinical trial, 31 (ketamine
0.5 mg/kg, n=17; 1.0 mg/kg, n=14; 13
women; mean age 48.4 years) were included in this post-hoc analysis. The demographic
and clinical characteristics of the participants are summarized in [Table 1].
Table 1 Demographic and clinical characteristics of
patients.
|
Characteristics (N = 31)
|
Mean±SD or N (%)
|
|
Age (year)
|
48.4±10.9
|
|
Age of onset (year)
|
22.6±12.4
|
|
Duration of illness (year)
|
25.8±15.3
|
|
Number of hospitalizations
|
0.2±0.4
|
|
Number of nonresponding antidepressants
|
3.0±1.2
|
|
HAM-D-6 total score at baseline
|
12.7±1.8
|
|
HAM-D-6 total score on day 3
|
6.8±4.3
|
|
Change in HAM-D-6 total score on day 3
|
−6.0±4.3
|
|
HAM-D-17 sleep symptoms at baseline
|
4.6±1.9
|
|
HAM-D-17 core emotional symptoms at baseline
|
8.2±2.2
|
|
HAM-D-17 atypical symptoms at baseline
|
4.4±1.9
|
|
CADSS total score at baseline
|
0.0±0.2
|
|
CADSS total score at 40 min
|
18.8±15.3
|
|
CADSS total score at 80 min
|
0.2±0.4
|
|
CADSS total score at 120 min
|
0.1±0.3
|
|
CADSS amnesia symptoms at 40 min
|
2.5±2.5
|
|
CADSS depersonalization symptoms at 40 min
|
5.8±5.3
|
|
CADSS derealization symptoms at 40 min
|
9.0±7.3
|
|
Sex, Male
|
18 (58.1)
|
|
Past treatment with TMS
|
1 (0.03)
|
|
Concurrent psychotherapy
|
13 (41.9)
|
|
Treatment responder
|
14 (45.2)
|
Abbreviations: CADSS, Clinician-Administered Dissociative States Scale;
HAM-D-6, Hamilton Rating Scale for Depression, 6-item version; HAM-D-17,
Hamilton Rating Scale for Depression, 17-item version; SD, standard
deviation; TMS, Transcranial Magnetic Stimulation.
Logistic regression analysis revealed a positive correlation between the age of onset
and the treatment response on day 3 (β=0.08, p=0.037), as
shown in [Table 2]. However, treatment
response to ketamine was not significantly related to age, sex, baseline HAM-D-6
total score, and CADSS total score 40 min after ketamine infusion. Multiple
regression analysis showed that no factors significantly correlated with the rate of
change in the HAM-D-6 total score on day 3 ([Table
3]). All variance inflation factors of the explanatory variables included
in these analyses were below five, suggesting the absence of multicollinearity among
these variables.
Table 2 Logistic regression analysis to predict responses on
day 3
|
β
|
Std error
|
t-value
|
p-value
|
|
(Intercept)
|
−3.85
|
4.37
|
−0.88
|
0.38
|
|
Age (year)
|
0.05
|
0.04
|
1.22
|
0.22
|
|
Age of onset (year)
|
0.08
|
0.04
|
2.09
|
0.04
|
|
Sex, Male
|
−0.88
|
0.93
|
−0.95
|
0.34
|
|
HAM-D-6 total score at baseline
|
−0.02
|
0.25
|
−0.09
|
0.93
|
|
CADSS total score at 40 min
|
−0.01
|
0.03
|
−0.19
|
0.85
|
Bold letters indicate p<0.05. Abbreviations: CADSS,
Clinician-Administered Dissociative States Scale; HAM-D-6, Hamilton Rating
Scale for Depression, 6-item version; Std error, standard error.
Table 3 Multivariable regression analysis to predict the rate
of change in the HAM-D-6 total score on day 3 from
baseline
|
β
|
Std error
|
t-value
|
p-value
|
|
(Intercept)
|
−0.19
|
0.62
|
−0.30
|
0.77
|
|
Age (year)
|
−0.01
|
0.01
|
−2.1
|
0.83
|
|
Age of onset (year)
|
0.01
|
0.01
|
1.95
|
0.06
|
|
Sex, Male
|
0.01
|
0.13
|
0.05
|
0.96
|
|
HAM-D-6 total score at baseline
|
0.00
|
0.04
|
−0.01
|
0.99
|
|
CADSS total score at 40 min
|
0.00
|
0.00
|
0.05
|
0.96
|
Abbreviations: CADSS, Clinician-Administered Dissociative States Scale;
HAM-D-6, Hamilton Rating Scale for Depression, 6-item version; Std error,
standard error.
Discussion
In this post-hoc analysis, we examined the clinical variables associated with the
treatment response to intravenous ketamine therapy in patients with TRD. Our study
used data from a placebo-controlled, double-blind, randomized controlled trial that
assessed the acute efficacy of ketamine infusion compared to an active placebo. In
patients who received 0.5 mg/kg or 1.0 mg/kg
ketamine infusion, the age of onset was positively correlated with treatment
response according to the HAM-D-6 total score change 3 d after ketamine infusion.
That is, a later onset of illness was associated with a better treatment response.
This finding suggests that the onset of depression may reflect the treatment
response of patients with TRD to intravenous ketamine therapy.
The findings of previous studies were inconsistent regarding the associations between
age of onset and response to antidepressant treatment in patients with MDD [21]
[22]
[23]
[24]
[25]
[26]
[27]
[28].
In contrast to our result, Chen and colleagues (2021) demonstrated that earlier age
of onset was one of the predictors of good treatment response 2 d after a single
0.5 mg/kg intravenous ketamine administration in 73 patients with
TRD [20]. The discrepancy between their
findings and our result may partly be due to the differences in the study design.
Chen et al. included patients who were resistant to treatment for both MDD and BD,
and conducted a regression tree analysis using binary classification. They
demonstrated that an earlier age of onset was predictive of good treatment response
among patients who had a current depressive episode for 24 months or less and a
baseline HAM-D-17 total score of 23 or less; however, the present age was not
included as a covariate in their analysis. Thus, to clarify the relationship between
age at onset and treatment response after ketamine infusion, future studies should
adjust for the age in a uniform diagnostic group.
The worse treatment response associated with an earlier onset of illness may be
because MDD with an earlier age of onset may be related to chronic depression.
According to the DSM-5, chronic depression is defined as depression that has
persisted for at least two years, [40].
Compared to non-chronic depression, chronic depression is associated with an earlier
age of onset and more frequent episodes of depression [41]
[42].
Moreover, patients with chronic depression often do not respond to pharmacotherapy
and psychotherapy [43]
[44]
[45]
and need higher dosages to achieve improvement [44]. In the present study, it may be possible that MDD patients with an
earlier age of onset had characteristics of chronic depression and were more
resistant to ketamine treatment than other MDD patients.
As another explanation for the association between an earlier onset of illness and a
worse treatment response is that an earlier age at onset is associated with a longer
duration of illness in MDD [27]
[46]
[47].
Previous studies have found that a longer duration of illness is associated with a
lower treatment response to antidepressant treatment in patients with MDD [21]
[48]
[49]. Thus, the higher
resistance to ketamine treatment in early-onset patients with TRD in the present
study may possibly be attributed to the longer duration of illness. Preclinical and
clinical evidence support impaired glutamatergic pathways in patients with MDD [50]
[51].
Some post-mortem studies have reported that compared to controls, patients with MDD
have decreased expression of glutamatergic
α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptor
subunit in the perirhinal cortex [52], cornu
ammonis (CA)1, and dentate gyrus [53].
Moreover, animal studies have demonstrated that exposure to chronic stress causes a
reduction in AMPA receptor subunit expression in these brain regions [54]. Furthermore, the number of apical dendrite
spines of CA1 and CA3 pyramidal cells was progressively reduced over 3 w in a
depression model consisting of chronically and unpredictably stressed mice [55]. The antidepressant effects of ketamine are
assumed to occur via the activation of AMPA receptors and synaptogenic signaling
pathways [51]. The long course of the disease
may have impaired glutamatergic signaling transmission, disturbed neuroplasticity,
and developed ketamine resistance. However, the relationship between illness
duration and AMPA receptor expression in patients with MDD has not been clarified,
since no technique to measure AMPA receptors in living humans is available.
Recently, [11C] K-2, the first positron emission tomography tracer that
specifically binds to AMPA receptors, was developed [56]
[57]
[58]
[59].
Future studies are warranted to investigate the relationship between the treatment
response to ketamine and AMPA receptor expression before and after ketamine infusion
in the context of neural plasticity.
This study found no association between the total CADSS score 40 min after ketamine
administration and treatment response 3 d after ketamine infusion. The relationship
between the dissociative side effects of ketamine and antidepressant effects has
been examined in several previous studies, but their findings are inconsistent [29]
[30].
For example, in a single-blind study of 10 patients with MDD, no association was
found between the maximum change in the CADSS total score and the change in the
HAM-D-17 total score at any time point after administering a single dose of ketamine
at a subanesthetic dose [32]. However,
Luckenbaugh et al. (2014) studied 108 patients with TRD and found that a higher
CADSS total score 40 min after ketamine administration was correlated with a
percentage decrease in the HAM-D-17 total score 7 d and 230 min after ketamine
infusion [33]. Later, Phillips et al. (2019)
conducted a randomized, double-blind crossover trial in which participants received
a single dose of 0.5 mg/kg ketamine infusion over 40 min. They found
that the increase in the CADSS total score 40 min after ketamine administration
correlated with the MADRS total score reduction at 24 h post-infusion [31]. The discrepancy between these findings and
our result may possibly be due to the small sample size and lack of power of the
present study. Additionally, the CADSS may not have fully captured the dissociative
symptoms of ketamine because it was initially developed to assess dissociative
symptoms of post-traumatic stress disorder [35]. To elucidate the relationship between treatment response and
dissociative symptoms after ketamine administration, further studies should include
larger sample sizes and utilize more sensitive and specific rating scales that
measure dissociative symptoms after ketamine administration.
The strengths of this study were that the original trial had a robust study design,
and the study participants consisted of a uniform diagnostic group (i. e.,
TRD), resulting in high-quality data. However, our study had several limitations.
First, this was a post-hoc analysis and was not originally designed to investigate
the predictors of treatment response after ketamine infusion. Second, the sample
size was small. Third, the study evaluated depressive symptoms only 3 d after
ketamine administration and did not assess them at other time points. Fourth, the
data on the duration of the index episode were not available; therefore we could not
analyze the effect of this factor on the treatment outcome. Finally, previous
studies reported that a positive family history of alcohol use disorder and body
mass index were associated with treatment response to ketamine; however, these
factors were not assessed in the current study.
Conclusions
A later onset of illness was associated with a better treatment response 3 d after
0.5 mg/kg or 1.0 mg/kg ketamine infusion in patients
with TRD. This finding suggests that patients with early-onset TRD may be
biologically distinct from those with late-onset TRD, and ketamine treatment may be
more beneficial in patients with late-onset TRD. The long course of the disease may
impair glutamatergic signal transmission and neuroplasticity; however, the
relationship between neuroplasticity and treatment response to ketamine infusion is
not fully understood. Further studies are needed to predict the response to
intravenous ketamine and elucidate the mechanism underlying the antidepressant
effect of ketamine.
Additional Information
Data used in this study were obtained from controlled access datasets distributed
by The National Institute of Mental Health (NIMH)-supported National Database
for Clinical Trials. The original dataset is available at the NIMH Data Archive
(https://nda.nih.gov/). The
identification number of the NIMH data repository study is #2166. NCT01920555
was supported by the NIMH contract HHSN271201100006I to the Massachusetts
General Hospital. The primary purpose of the present study was to examine the
factors associated with treatment response to intravenous ketamine in patients
with TRD. This article reflects the authors’ views and may not reflect
the opinions or views of the NCT01920555 study investigators or the NIMH.
Source of Funding
This study was supported by the Japan Society for the Promotion of Science
KAKENHI under grant numbers 22H03001 (H.U.) and 22K15793 (H.T.), Keio
Next-Generation Research Project Program (H.U.), SENSHIN Medical Research
Foundation (H.U.), and Japan Research Foundation for Clinical Pharmacology
(H.T.).
