Blood Biological Markers for Prediction of Escitalopram Response in Patients with Major Depressive Disorder: Preliminary Study

Catecholamines, neurotrophic factors and cytokine play important role for pathophysiology of major depressive disorder (MDD). We previously reviewed plasma levels of catecholamines predict the response to treatment with antidepressants [1]. Briefly, paroxetine, a selective serotonin reuptake inhibitor (SSRI) had better response to the patients with higher plasma levels of 3-methoxy-4-hydroxyphenylglycol (MHPG), a major metabolite of norepinephrine, whereas, milnacipran, a serotonin, norepinephrine reuptake inhibitor (SNRI), had better response to those who with lower MHPG levels [2]. However, no changes in plasma MHPG switching paroxetine to milnacipran between the responders to nonresponders with milnacipran treatment in MDD patients [3]. Serum levels of brain-derived neurotrophic factor (BDNF) decrease in MDD patients compared with healthy control [4]. A negative correlation was found between serum BDNF levels and Hamilton Rating Scale for Depression (HAMD) [1]. Antidepressants increase plasma BDNF levels in MDD patients [4]. We also reported that treatment with paroxetine, a SSRI and milnacipran, a SNRI, increased plasma levels of brain-derived neurotrophic factor (BDNF) in responders to those antidepressants [5]. From these findings into account, serum BDNF is one of candidate for biological markers of severity of depressive state. On the other hand, interleukin 6 (IL 6) and tumor necrotic factor-alpha (TNF-alpha) in MDD patients were higher than those in healthy control [6]. We reported that serum IL 6 levels were lower in MDD patients compared with sex, and age-matched healthy controls [7]. It is however controversial that serum levels of BDNF or IL 6 predict response to treatment with antidepressants. A report demonstrated patients with higher serum BDNF had better response [8], other did not [7]. We reported that responders to SSRI were significantly higher serum IL 6 levels than nonresponders suggesting serum IL 6 might be a predictor for SSRI response in MDD patients [9]. Kotan et al. [10] reported that serum vascular endothelial growth factor (VEGF) was related with MDD. *Corresponding author: Reiji Yoshimura, Department of Psychiatry, University of Occupational and Environmental Health, 1-1 Iseigaoka, Yahatanishi-ku, Kitakyushu, Fukuoka8078555, Japan, Tel: +81936917253; Fax: +81936924894; E-mail: yoshi621@med.uoeh-u.ac.jp


Introduction
Catecholamines, neurotrophic factors and cytokine play important role for pathophysiology of major depressive disorder (MDD). We previously reviewed plasma levels of catecholamines predict the response to treatment with antidepressants [1]. Briefly, paroxetine, a selective serotonin reuptake inhibitor (SSRI) had better response to the patients with higher plasma levels of 3-methoxy-4-hydroxyphenylglycol (MHPG), a major metabolite of norepinephrine, whereas, milnacipran, a serotonin, norepinephrine reuptake inhibitor (SNRI), had better response to those who with lower MHPG levels [2]. However, no changes in plasma MHPG switching paroxetine to milnacipran between the responders to nonresponders with milnacipran treatment in MDD patients [3]. Serum levels of brain-derived neurotrophic factor (BDNF) decrease in MDD patients compared with healthy control [4]. A negative correlation was found between serum BDNF levels and Hamilton Rating Scale for Depression (HAMD) [1]. Antidepressants increase plasma BDNF levels in MDD patients [4]. We also reported that treatment with paroxetine, a SSRI and milnacipran, a SNRI, increased plasma levels of brain-derived neurotrophic factor (BDNF) in responders to those antidepressants [5]. From these findings into account, serum BDNF is one of candidate for biological markers of severity of depressive state. On the other hand, interleukin 6 (IL 6) and tumor necrotic factor-alpha (TNF-alpha) in MDD patients were higher than those in healthy control [6]. We reported that serum IL 6 levels were lower in MDD patients compared with sex, and age-matched healthy controls [7]. It is however controversial that serum levels of BDNF or IL 6 predict response to treatment with antidepressants. A report demonstrated patients with higher serum BDNF had better response [8], other did not [7]. We reported that responders to SSRI were significantly higher serum IL 6 levels than nonresponders suggesting serum IL 6 might be a predictor for SSRI response in MDD patients [9]. Kotan et al. [10] reported that serum vascular endothelial growth factor (VEGF) was related with MDD.

Blood Biological Markers for Prediction of Escitalopram Response in Patients with Major Depressive Disorder: Preliminary Study
Masato Yamana, Kiyokazu Atake, Asuka Katsuki, Hikaru Hori and Reiji Yoshimura*

Department of Psychiatry, University of Occupational and Environmental Health, Japan
In the present study, we investigated between the response to escitalopram, a SSRI and serum levels of BDNF, cytokines (interleukin 2 (IL2), IL6, TNF-alpha), vascular endothelial growth factor (VEGF), and plasma levels of MHPG and homovanillic acid (HVA), a major metabolite of dopamine, in MDD patients.

Subjects and Methods
Forty-eight patients diagnosed MDD using DSM-IV-TR were enrolled in the present study. The demographics data were shown in Table 1. All participants were treated with only escitalopram. The severity of the patients' depressive symptoms was evaluated using the 17-items of Hamilton Rating Scale for Depression (HAMD17) at each week. The end point of the follow up period was 8 weeks. The patients who decreased their HAMD17 scores 50 % or more were defined as responders. Others were defined as nonresponders. The blood was drawn once just before starting escitalopram treatment. Their serum levels of BDNF, IL 2, IL 6, TNF-alpha, and VEGF were assayed by ELISA using Milliplex MAP Kit (HCYTOMAG-60K and HNDG3MAG-36K) on a Milliplex Analyzer 4.2 MAGPIX machine (Millipore) according to the manufacturer's instructions. Analyses included IL-2, IL-6, TNF-alpha, VEGF, and BDNF. Their plasma levels of metabolites of catecholamine were measured using high-performance liquid chromatography with electrochemical detection. The study protocol was approved by the Ethics Committee of the University of Occupational and Environmental Health. Written informed consent was obtained from all participants. Twenty-seven patients of 48 MDD patients (56.25%) completed the present study. The reasons for drop out the present study were the protocol violation (15 patients), moved to other hospitals (4 patients), or the study withdrawal (22 patients).

Response to escitalopram treatment
No significant differences were observed in demographic background between the responders and the nonresponders to escitalopram ( Table 1). The response rate at 8 weeks with escitalopram treatment was 41.7%. The dose of escitalopram at week was 16.2 ± 6.2 (mean ± standard deviation).

Plasma levels of catecholamine metabolites and response to escitalopram treatment
Plasma MHPG levels at starting escitalopram in the responders were significantly higher than those in nonresponders (p = 0.0032). No difference was found in plasma HVA levels between the responders and nonresponders (p = 0.5981) (Figure 1). We further calculated cutoff value by using Receiver Operating Characteristic Curve (ROC), and found the cut-off value was 8.0 ng/ml (sensitivity; 90.0%: specificity; 85.7%) (Figure 2).

Serum levels of BDNF and VEGF and response to escitalopram
No differences were observed regarding serum levels of BDNF and VEGF at starting escitalopram between the responders and nonresponders (p = 0.2763, p = 0.3029) (Figure 3).

Serum levels of cytokines and response to escitalopram
No differences were observed regarding serum levels of IL 2, IL 6, and TNF-alpha at starting escitalopram between the responders and nonresponders (p = 0.4577, p = 0.4873, p = 0.8330) (Figure 4).

Discussion
The main finding of the present study was plasma MHPG level at starting escitalopram predicted its response in patients with MDD. Patients whose plasma MHPG 8ng/ml or more was prone to better response. Whereas, plasma HVA level, serum levels of BDNF, VEGF, IL 2, IL 6 were not related to the response to escitalopram. The results of the present study were partially in accordance with our previous studies regarding plasma MHPG level. On the other hand, serum IL 6 level was not different between the responders and nonresponders. We recently reported that higher serum IL 6 was associated with response to SSRI. The discrepancy between the present result and our previous one remains unknown. The patients who enrolled in our previous study was treated with only sertraline or paroxetine, but not escitalopram. The deference in the SSRI may one of the reasons. We investigated baseline serum BDNF levels and the response to several antidepressants in MDD patients, also found that no association was observed between serum BDNF and response to antidepressants treatment [1,11]. The result in the present study was reconfirmed in patients with escitalopram treatment. Ventriglia et al. [12] reported that baseline serum VEGF levels did not predict the response to SSRIs. In addition, the authors also demonstrated that treatment with SSRIs did not alter VEGF levels in MDD patients. The result regarding VEGF in the present study was not contradict Ventriglia's finding     [12]. The weakest point in the present study was lacked in data at after escitalopram treatment. In other words, we predicted the response to escitalopram treatment at only one point. It remains unknown from the present data that treatment with escitalopram changes the biological markers or not, which is associated with the improvement of depressive symptoms or not. To elucidate this question, we should accumulate larger samples and follow up biological markers. We are now starting longitudinal study for biological markers. Furthermore, we should also combine genetic, proteolytic, and neuroimaging data with our preliminary findings. This process will lead to find a robust biological marker to predict response to escitalopram in MDD. In conclusion, plasma MHPG level at baseline could predict the response to escitalopram in MDD patients. Its cut-off value was 8 ng/ml.