Plasma BDNF concentrations and the antidepressant effects of six ketamine infusions in unipolar and bipolar depression

Introduction

Accumulating evidence suggests that glutamatergic abnormalities are associated with the pathophysiology of mood disorders (Yüksel & Öngür, 2010). Numerous early studies had consistently reported that an antagonist of glutamatergic N-methyl-D-aspartate (NMDA) receptors ketamine at subanesthetic doses could result in fast-acting and sustained antidepressant effects in individuals suffering from unipolar and bipolar depression (Na & Kim, 2021; Phillips et al., 2020). For example, ketamine’s repeated administration had quick and enduring antidepressant and antisuicidal effects in depressed patients (Kryst et al., 2020).

The precise mechanisms underlying subanesthetic intravenous ketamine’s antidepressant actions are still incompletely understood (Rong et al., 2018). A recent animal study found that blockade of NMDA receptors increased the induction of α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptor expression in models of depression, and subsequent activation of the mammalian target of rapamycin (mTOR) pathway was needed for the rapid and robust antidepressant action of ketamine (Li et al., 2010). Growing evidence implicated neurotrophic factors, such as brain-derived neurotrophic factor (BDNF), played an important in the pathophysiology of mood disorders (Duman, 2004; Duman & Monteggia, 2006). BDNF is a key protein in facilitating and supporting memory growth and neuronal survival (Leal, Bramham & Duarte, 2017). Rapid and transient upregulation of BDNF reversed or blocked atrophy and cell loss in patients with depression, and it may be a critical component in subanesthetic intravenous ketamine’s antidepressant actions (Haile et al., 2014).

In general, BDNF plays a role in the pathophysiology of schizophrenia (Singh et al., 2020) and mood disorders (Molendijk et al., 2014; Sagud et al., 2016). For example, several studies found that individuals suffering from depression had lower serum BDNF concentrations and pBDNF concentrations than that of healthy subjects (Molendijk et al., 2014; Sagud et al., 2016) and recovered after successful antidepressant therapy (Brunoni, Lopes & Fregni, 2008; Polyakova et al., 2015). Central and peripheral BDNF is positively correlated with the response and remission of antidepressant treatment (Lee & Kim, 2010). Notably, Kurita et al. (2012) reported that remitted than nonremitted depressed patients appeared to have higher pBDNF concentrations, and these concentrations were associated with the Montgomery-Asberg Depression Rating Scale (MADRS) scores.

BDNF as a predictor of ketamine’s antidepressant efficacy in individuals suffering from treatment-refractory depression (TRD) has been investigated, but with inconsistent findings. For example, several open-label studies on ketamine and BDNF found a negative association of the increase in BDNF following a single ketamine infusion with the severity of depression (Cornwell et al., 2012; Duncan et al., 2013). Another study found that BDNF did not mediate single subanesthetic intravenous ketamine’s antidepressant efficacy (Machado-Vieira et al., 2009). However, no studies had been published to examine the relationship of pBDNF concentrations and serial subanesthetic intravenous ketamine infusions’ antidepressant efficacy in Chinese individuals suffering from depression.

The present study was performed to examine the correlation of pBDNF concentrations and six subanesthetic intravenous ketamine’s antidepressant efficacy (0.5 mg/kg) administered thrice weekly over two weeks in Chinese individuals suffering from unipolar and bipolar depression. In this study, we hypothesized that serial intravenous subanesthetic ketamine would increase pBDNF concentrations, and baseline pBDNF concentrations would be associated with ketamine’s antidepressant efficacy in individuals suffering from depression.

Methods

Results

Ninety-four individuals (aged 18 to 62 years) with unipolar or bipolar depression who provided a baseline blood sample were enrolled. Of these patients, 81.9% (77/94) fulfilled the diagnostic criteria of TRD. Baseline pBDNF concentrations with a mean value of 10.1 ng/ml, ranged from 0.9 to 27.2 ng/ml.

Correlation of BDNF and MADRS scores

Correlation analyses showed significant associations between pBDNF concentrations at baseline and MADRS scores at 13 d and 26 d in depressed patients (all p s<0.05; Table 3). The significant association of pBDNF concentrations at baseline and MADRS scores at 13 d (t = −2.011, p = 0.047) and 26 d (t = −2.398, p = 0.019) remained in multiple regression analysis. Similar results were found in patients with TRD (Table S2).

Discussion

This is the first study to determine pBDNF concentrations after six subanesthetic intravenous ketamine in Chinese individuals suffering from unipolar and bipolar depression and to investigate the correlation of pBDNF concentrations at baseline and six subanesthetic intravenous ketamine’s antidepressant efficacy. The following main findings included: (1) ketamine increased pBDNF at 13 d and 26 d compared to baseline; (2) responders/remitters had significantly higher baseline pBDNF concentrations than nonresponders/nonremitters; (3) MADRS scores showed significant improvement at both time points across the total sample compared to baseline; (4) baseline pBDNF concentrations were related with MADRS scores; and (5) additional analysis of patients with TRD also found that pBDNF concentrations were related with the antidepressant outcome of ketamine in patients with TRD.

Consistent with an animal study after single ketamine infusion (Pytka et al., 2018), our study demonstrated that ketamine increased pBDNF concentrations after six ketamine infusions. Although nonresponders/nonremitters had significantly lower pBDNF concentrations at baseline than responders/remitters, repeated ketamine infusions failed to significantly increase pBDNF concentrations in responders/remitters when compared to nonresponders/nonremitters. Similarly, a previous study found no changes in pBDNF concentrations in individuals suffering from TRD after completion of an intravenous infusion of ketamine compared to baseline (Machado-Vieira et al., 2009). However, Haile et al. found that pBDNF concentrations were significantly increased following a single ketamine infusion in responders compared to nonresponders (Haile et al., 2014). Therefore, these findings should be confirmed by randomized controlled trials.

The observed rapid reduction in MADRS scores lasted up to 2 weeks, replicating the previous findings (Rasmussen et al., 2013; Shiroma et al., 2014). However, the primary objective of this study is to examine the association of baseline pBDNF concentrations and six subanesthetic intravenous ketamine’s antidepressant efficacy. Several studies examined the association of pBDNF concentrations with the antidepressant response of a single infusion of ketamine, but these findings are inconsistent (Haile et al., 2014; Lee & Kim, 2010). For instance, one study reported that pBDNF concentrations were related with the severity of depression (Haile et al., 2014). However, Machado-Vieira et al.’s study reported a negative finding on the association of pBDNF concentrations and ketamine’s antidepressant efficacy (Machado-Vieira et al., 2009).

Notably, several animal studies reported that increased hippocampal and cortical BDNF expression can partly accounting for ketamine’s antidepressant-like efficacy (Autry et al., 2011; Réus et al., 2011). pBDNF concentrations were lower in individuals suffering from depression compared to healthy controls (Kishi et al., 2017; Munno et al., 2013) and increased after receiving antidepressants (Munno et al., 2013; Polyakova et al., 2015), electroconvulsive therapy (Luan et al., 2020; Piccinni et al., 2009), and repeated transcranial magnetic stimulation (Yukimasa et al., 2006). Therefore, neurotrophic factors, such as BDNF, might be involved in ketamine’s antidepressant mechanism. Notably, BDNF is implicated in the regulation of synaptic plasticity, including the synaptic recruitment of AMPA receptors. Growing studies indicate that synaptic plasticity is altered in individuals with depression (Machado-Vieira, Zarate Jr & Manji, 2006; Schloesser et al., 2008; Zarate Jr, Singh & Manji, 2006), and ketamine’s antidepressant efficacy may be attributed to the synaptic potentiation of neural circuits mediated by increased AMPA-to-NMDA glutamate receptors (Maeng & Zarate Jr, 2007).

The following limitations should be acknowledged. First, the participants continued receiving previous medications and lacked a washout period during the study, which may have affected pBDNF concentrations. However, the combination of ketamine and other antidepressants for individuals with depression is increasingly being used in the real-world clinical setting (Shiroma et al., 2014). Second, the sample size was small in the current study. Third, the possible impact of subjective evaluation was inevitable due to lack of a control group. Fourth, some comprehensive analyses, such as the mediating and moderating effect analysis, were not conducted in this study. Finally, brain BDNF concentrations and other key neurobiological mediators, such as mTOR, were not directly measured. However, BDNF crosses the blood–brain barrier, and pBDNF concentrations are closely correlated with cortical BDNF concentrations, and likely reflect brain BDNF concentrations (Pillai et al., 2010; Poduslo & Curran, 1996).

In conclusion, this preliminary study suggests that baseline pBDNF concentrations appeared to be correlated with ketamine’s antidepressant efficacy in Chinese patients with depression.

Supplemental Information