論文の公開元へKynurenic acid is a potential overlapped biomarker between diagnosis and treatment response for depression from metabolome analysis
概要
www.nature.com/scientificreports
OPEN
Kynurenic acid is a potential
overlapped biomarker
between diagnosis and treatment
response for depression
from metabolome analysis
Hisayuki Erabi1, Go Okada1, Chiyo Shibasaki1, Daiki Setoyama2, Dongchon Kang2,
Masahiro Takamura1, Atsuo Yoshino1, Manabu Fuchikami1, Akiko Kurata1, Takahiro A. Kato3,
Shigeto Yamawaki1 & Yasumasa Okamoto1*
Since optimal treatment at an early stage leads to remission of symptoms and recovery of function,
putative biomarkers leading to early diagnosis and prediction of therapeutic responses are desired.
The current study aimed to use a metabolomic approach to extract metabolites involved in both
the diagnosis of major depressive disorder (MDD) and the prediction of therapeutic response for
escitalopram. We compared plasma metabolites of MDD patients (n = 88) with those in healthy
participants (n = 88) and found significant differences in the concentrations of 20 metabolites.
We measured the Hamilton Rating Scale for Depression (HRSD) on 62 patients who completed
approximately six-week treatment with escitalopram before and after treatment and found that
kynurenic acid and kynurenine were significantly and negatively associated with HRSD reduction.
Only one metabolite, kynurenic acid, was detected among 73 metabolites for overlapped biomarkers.
Kynurenic acid was lower in MDD, and lower levels showed a better therapeutic response to
escitalopram. Kynurenic acid is a metabolite in the kynurenine pathway that has been widely accepted
as being a major mechanism in MDD. Overlapping biomarkers that facilitate diagnosis and prediction
of the treatment response may help to improve disease classification and reduce the exposure of
patients to less effective treatments in MDD.
The pathomechanism of major depressive disorder (MDD) remains largely unknown. Although there have been
many studies to identify biomarkers for the diagnosis of MDD, there are currently no diagnostic biomarkers that
are routinely used in clinical practice1. Furthermore, although there is a wide variety of treatment options for
MDD, only approximately 40% of MDD patients achieve remission after initial t reatment2.
Selective serotonin reuptake inhibitors (SSRIs) are commonly used as first-line treatment for MDD3. They are
thought to increase the extracellular availability of the neurotransmitter serotonin by limiting its reabsorption
into presynaptic cells, increasing serotonin levels in the synaptic cleft, and making it available for postsynaptic
receptor binding4. However, only about half to two-thirds of patients respond to SSRIs, requiring weeks of treatment before an optimal therapeutic response is achieved5. The effects of antidepressants vary significantly from
person to person, and finding the right drug at the right dose requires trial and e rror6. Clinical manifestations
are insufficient to guide appropriate treatment options, and it is essential to develop biomarkers of MDD that
will lead to the diagnosis of MDD or to predict response to treatment.
1
Department of Psychiatry and Neurosciences, Graduate School of Biomedical Sciences, Hiroshima University,
1‑2‑3 Kasumi, Minami‑ku, Hiroshima 734‑8551, Japan. 2Department of Clinical Chemistry and Laboratory
Medicine, Graduate School of Medical Sciences, Kyushu University, 3‑1‑1 Maidashi Higashi‑Ku, Fukuoka 812‑8582,
Japan. 3Department of Neuropsychiatry, Graduate School of Medical Sciences, Kyushu University, 3‑1‑1 Maidashi
Higashi‑Ku, Fukuoka 812‑8582, Japan. *email: oy@hiroshima‑u.ac.jp
Scientific Reports |
(2020) 10:16822
| https://doi.org/10.1038/s41598-020-73918-z
1
Vol.:(0123456789)
www.nature.com/scientificreports/
HC (n = 88)
MDD (n = 88)
p-value
Gender (male: female)
40/48
46/42
0.366
Age (years)
41.3 ± 12.7
42.9 ± 12.2
0.415
HRSD score at baseline
19.3 ± 5.0
Episode (single: recurrent)
43/45
Other mental disorder (yes: no)
34/54
Use of benzodiazepines at baseline (yes: no)
13/75
Table 1. Demographic and clinical characteristics of the patients with major depressive disorder (MDD) and
healthy controls (HC).
Biomarkers can be classified into four types as follows: diagnostic, predictive, prognostic, and therapeutic
response. Diagnostic biomarkers can detect disease early and indicate future onset by non-invasive methods.
Predictive biomarkers allow the identification of patients who are likely to benefit from therapy. Prognostic biomarkers provide information about disease course and outcome. Therapeutic response biomarkers predict the
effect of therapeutic intervention and can be used as selection of efficacious antidepressant for MDD with some
biological features7,8. Since optimal treatment at an early stage leads to remission of symptoms and recovery of
function9, putative biomarkers leading to early diagnosis and prediction of therapeutic responses are desired.
In recent years, metabolomics approaches have attracted attention due to new possibilities for biomarkers.
Metabolites are the final phenotype and are thought to be influenced by genetic and environmental factors and
associated with disease pathology. The metabolomic mass spectrometry-based approach is a means of exhaustively searching for changes in vivo metabolites that are unpredictable from previous knowledge using unbiased
techniques8,10. Recent advances in analytical chemistry have made this approach possible.
Various biological fluids, such as urine, plasma, and cerebrospinal fluid, have been analyzed. Blood in particular is easy and less invasive to obtain. One of the first metabolomics studies using blood in the field of MDD
diagnosis was conducted by Paige et al. (2007), who analyzed approximately 800 metabolites in plasma in three
groups as follows: people with depression, people in remission, and control participants. The depression group
showed a significant overall decrease in gamma-aminobutyric acid (GABA) and medium-chain fatty acid l evels11.
Since then, many studies have been conducted on MDD metabolites12–15. According to MacDonald et al. (2019),
nine diagnostic biomarkers have been identified in plasma of MDD patients, with glutamate and alanine showing up-regulation, and myo-inositol, GABA, phenylalanine, creatine, methionine, oleic acid, and tryptophan
showing down-regulation16.
Kaddurah-Daouk et al. (2011) conducted one of the first metabolomic studies of treatment prediction for
MDD. This study demonstrated the potential of metabolomics to provide information on the early efficacy of
sertraline17 and another study (2013) reported that good therapeutic outcomes of MDD were associated with low
levels of branched-chain amino acids18. Zhu et al. (2013) demonstrated that high pretreatment levels of 5-methoxytryptamine were associated with sertraline r esponsiveness19. Rotroff et al. (2016) demonstrated that none
of the baseline metabolomes was significantly associated with treatment response to ketamine, esketamine, or
placebo20. For citalopram and escitalopram treatment response, Bhattacharyya et al. (2019) suggested that higher
baseline serotonin and 3-methoxy-4-hydroxyphenylglycol levels were associated with better responses to S SRIs21.
Although the knowledge about diagnostic biomarkers of MDD is expanding through this metabolomic
approach, predictive biomarkers of antidepressant therapy are not enough and require further study. Besides, if
a biomarker capable of simultaneously performing MDD diagnosis and treatment prediction can be established,
treatment for MDD can be introduced more simply and efficiently. Based on the above points of view, the present
study aims to extract markers that overlap diagnostic biomarkers of MDD and predictive biomarkers of treatment of escitalopram from many metabolites obtained by metabolomics. Escitalopram was selected because
it has a highly selective, dose-dependent inhibitory effect on the serotonin transporter, is highly effective and
well-tolerated, and the initial dosage is effective for treating depression22.
Results
Sample demographics. Demographic information including gender, age, as well as HRSD score, comorbidity of other mental disorders, and use of benzodiazepines for the study participants are shown in Table 1.
MDD participants and healthy controls (HC) did not differ significantly in gender (42/88 (48%) female vs.
48/88 (55%) female, p = 0.37, chi-square test), or age (42.9 ± 12.2 vs. 41.3 ± 12.7, p = 0.42, two-sample t-test) at
baseline. The average baseline HRSD rating was 19.3 ± 5.0. Forty-three patients (48.9%) had the first episode, and
34 patients (38.6%) had other psychiatric disorders. Thirteen patients (14.8%) were receiving benzodiazepines.
As for treatment response data, we excluded 26 patients due to withdrawal (n = 9), change to or combination
with other antidepressants (n = 12), or discontinuation of escitalopram (n = 5). Thus, these data were obtained
in 62 patients after approximately 6 weeks of treatment with escitalopram. Subjects were evaluated for at least
6 weeks and up to 8 weeks, and the mean duration of treatment was 45.9 ± 4.5 days. The initial dose was 5–10 mg
and the dose after 6 weeks were 12.8 ± 5.1 mg (maximum total dose 13.7 mg/day). The clinical features are shown
in Table 2. Follow-up HRSD was recorded for all 62 patients, and there was a significant difference in the HRSD
scores recorded at baseline and follow-up (p < 0.001, paired-sample t-test). ...
この論文で使われている画像
