Vortioxetine: a review of efficacy, safety and tolerability with a focus on cognitive symptoms in major depressive disorder

Abstract

Introduction: Vortioxetine is a pharmacodynamically novel antidepressant that exerts effects on various neurotransmitters including serotonin, noradrenaline, dopamine, glutamate, histamine and acetylcholine. Its efficacy in the symptomatic management of major depressive disorder (MDD) has been established in several short- and long-term trials. Vortioxetine has also demonstrated independent pro-cognitive effects in adults with MDD.

Areas covered: This report provides a concise review of the pharmacology, efficacy and safety of vortioxetine as they pertain to cognition.

Expert opinion: The significant impact of cognitive dysfunction in MDD has achieved increased consideration among researchers over the past decade. Vortioxetine is the first antidepressant agent to demonstrate meaningful clinical efficacy in the improvement of cognition in adults with MDD, independent of improvement in affective symptomatology. These results provide the impetus for further study into the potential pro-cognitive effects of vortioxetine in other conditions wherein cognitive dysfunction is prominent.

Keywords:

• antidepressant
• cognition
• major depressive disorder
• neurotransmitter
• vortioxetine

1. Introduction

During the past two decades, several classes of mechanistically dissimilar antidepressants have received approval by the US FDA for the symptomatic treatment of a major depressive episode (MDE) as part of major depressive disorder (MDD). Notwithstanding the availability of many pharmacological options for MDD, results from both efficacy and effectiveness studies indicate that the majority of individuals with MDD have insufficient therapeutic outcomes including, but not limited to, ongoing clinically significant depressive symptoms and/or treatment-limiting adverse effects [1]. These symptoms can include not only depressed mood, but also somatic complaints and cognitive impairment. The foregoing unmet needs provide the impetus to discover, develop and receive regulatory approval for genuinely novel agents for individuals with mood disorders. Vortioxetine is a pharmacodynamically novel antidepressant categorized as ‘multi-modal’ that exerts effects on serotonin (5-HT), norepinephrine (NE), dopamine, amino acid (e.g., glutamate and γ-aminobutyric acid; GABA), histamine (HA) and cholinergic systems (Box 1). Vortioxetine’s pharmacodynamic profile was predictive of not only antidepressant effects but also independent cognitive benefits in adults with MDD [2].

Box 1. Drug summary.

Drug name (generic)	Vortioxetine
Phase (for indication under discussion)	Phase III
Indication (specific to discussion)	Treatment of major depressive disorder
Pharmacology description/mechanism of action	Serotonin transporter inhibition and serotonin receptor modulation (agonist at 5-HT1A; partial agonist at 5-HT1B; antagonist at 5-HT1D, 5-HT3, 5-HT7)
Route of administration	Oral
Chemical structure	1-(2-((2,4-Dimethylphenyl)thio)phenyl)piperazine
Pivotal trial(s)	[19-31,37,38]

Pharmaprojects – copyright to Citeline Drug Intelligence (an Informa business). Readers are referred to Pipeline (http://informa-pipeline.citeline.com) and Citeline (http://informa.citeline.com).

Vortioxetine’s efficacy in mitigating depression symptom severity across the adult age range has been established, and pivotal trials are either published and/or in press. Moreover, several narrative reviews have summarized the extant literature pertaining to vortioxetine’s efficacy in the symptomatic treatment of MDD. Herein we seek to succinctly review vortioxetine’s pharmacological profile, as well as its safety, tolerability and efficacy in the domain of cognition in adults with MDD.

2. Pharmacokinetics

Following oral administration, vortioxetine is slowly absorbed with an absolute bioavailability of 75%. Peak plasma concentrations of 9 – 33 ng/mL were reached in 7 – 11 h following multiple administrations of 5 – 10 mg/day of vortioxetine. Food does not appear to affect the absorption of vortioxetine and, as such, vortioxetine may be administered without regard to food intake [3]. Moreover, vortioxetine is highly protein-bound (98 – 99%), and it is distributed extensively to peripheral tissue, as evidenced by its relatively large volume of distribution (mean Vd = 2600 L) [3,4].

Vortioxetine is extensively metabolized via multiple cytochrome P450 isozymes, including CYP2D6, CYP3A4/5, CYP2C19, CYP2C9, CYP2A6, CYP2C8 and CYP2B6. CYP2D6 metabolism constitutes the primary metabolic pathway, converting vortioxetine to its major, pharmacologically inactive metabolite [3,5]. The enzymes alcohol dehydrogenase, aldehyde dehydrogenase and aldehyde oxidase also play a role in vortioxetine metabolism [6]. Furthermore, based on in vitro data, neither vortioxetine nor its metabolites exhibit clinically significant inhibition of any of the cytochrome P450 isozymes or the p-glycoprotein transporter [3]. Approximately two thirds of vortioxetine inactive metabolites are excreted in the urine, with the rest being excreted in feces. As vortioxetine is extensively metabolized by the liver, very small amounts of unchanged parent drug are excreted in the urine [3].

3. Pharmacodynamics

Vortioxetine has a unique pharmacological profile and a multimodal mechanism of action that acts on multiple neurotransmitter systems. In addition to increasing synaptic serotonin levels via inhibition of the serotonin transporter (SERT), vortioxetine interacts with and modulates the effects of several serotonin receptors. Namely, it acts as an agonist at 5-HT1A, a partial agonist at 5-HT1B and an antagonist at 5-HT1D, 5-HT3 and 5-HT7 [3,4,7]. The foregoing serotonergic actions are hypothesized to be relevant to the downstream modulation of other neurotransmitters implicated in mood and cognitive dysfunction, including NE, acetylcholine (ACh) and glutamate [7-10]. In preclinical studies, vortioxetine was found to increase extracellular levels of 5-HT, NE, ACh, HA and dopamine in brain regions relevant to mood and cognition [8,9,11].

3.1 Serotonin transporter occupancy

It is postulated that selective serotonin reuptake inhibitors (SSRIs) yield symptom relief in depression primarily through high occupancy inhibition of the SERT. Based on positron emission tomography studies in human subjects, SERT occupancy rates of 80% occur at recommended therapeutic doses of SSRIs and serotonin and norepinephrine reuptake inhibitors (SNRIs) [12]. In contrast, vortioxetine can exhibit considerably lower SERT occupancy rates than SSRIs and SNRIs. Clinically relevant vortioxetine daily doses of 5, 10 and 20 mg exhibit occupancy rates of approximately 50, 65 and 80%, respectively [3,13,14]. Vortioxetine’s ability to exert significant clinical effect at lower SERT occupancy rates than traditional SERT inhibitors strongly suggests that, unlike SSRIs, additional downstream and non-serotonergic mechanisms are also involved in mediating vortioxetine’s antidepressant and domain-specific effects on cognition [13].

3.2 Effects at serotonin receptors

Vortioxetine exerts pleiotropic effects across serotonergic receptor populations. Its agonist effects at 5-HT1A receptors accelerate the desensitization of inhibitory presynaptic 5-HT1A autoreceptors while simultaneously activating postsynaptic 5-HT1A receptors [7,9]. Concurrent administration of pindolol, a 5-HT1A receptor partial agonist, with an SSRI accelerates the onset of SSRI therapeutic effects [9]. The aforementioned evidence recapitulates vortioxetine’s multiple actions on serotonergic receptors and transporters and may contribute to the potentiation of this agent’s antidepressant effects.

Vortioxetine’s antagonism at the 5-HT3 receptor has several possible physiological effects. For example, it is proposed that 5-HT3 antagonism can increase NE and ACh release, as mediated by GABA interneurons [7,9]. Specifically, it is thought that the 5-HT3 receptor exerts inhibitory effects over NE and ACh release, and its antagonism, therefore, disinhibits the inhibitory tonus. Convergent evidence indicates that the 5-HT3 antagonist ondansetron, while not independently affecting basal levels of extracellular 5-HT, can potentiate increases in 5-HT levels when combined with the SSRI citalopram [9]. Consequently, increased extracellular NE and ACh levels may underlie antidepressant and pro-cognitive effects.

3.3 Interactions between serotonin receptors and glutamate

In addition to its described effects on SERT and serotonin receptors, vortioxetine is believed to exert its therapeutic effect via serotonin-mediated modulation of glutamate neurotransmission. Several glutamatergic agents (e.g., ketamine, tianeptine) have exhibited antidepressant or antidepressant-like effects in clinical and preclinical trials, but the specific receptor-level actions necessary to elicit antidepressant activity are unclear, and this activity may be seen with both positive and negative glutamate modulation, depending on the compound [2]. That is to say that antidepressant-like effects have been observed in certain compounds that increase glutamate neurotransmission as well as those that decrease glutamate signaling [2]. While the specific association between in vivo glutamate modulation and antidepressant effects is not fully elucidated, it is accepted that glutamate features notably in cognition, with increased glutamate neurotransmission being associated with pro-cognitive effects [2,15]. For example, glutamate is necessary and implicated in memory fixation and encoding of new information. However, unchecked glutamatergic activity can worsen cognitive function via its inherent excitotoxicity, posing a challenge for the development of pro-cognitive glutamatergic agents. Instead, an indirect potentiation of the glutamate system at physiologically relevant brain regions is hypothesized to provide a better option for drug targets [2].

Several of the serotonin receptors (e.g., 5-HT1A, 5-HT1B, 5-HT3 and 5-HT7) targeted by vortioxetine are known to modulate glutamate transmission by interacting either with gluatamate receptors or via other mediators (e.g., GABA interneurons) and, therefore, it is postulated that indirect, downstream changes to glutamate transmission contribute to the medication’s empirically instantiated observations of cognitive benefits [2].

Table 1 summarizes the proposed effects and significance of vortioxetine’s actions at different receptors [2,7-10].

Table 1. Proposed effects and significance of vortioxetine action at various receptors [2,7-10].

Receptor/protein	Affected neurotransmitter(s)	Vortioxetine action	Significance/proposed clinical effect
SERT	5-HT	Inhibition	Improve mood via increased synaptic serotonin levels [7]
5-HT1A	5-HT; glutamate	Agonist	Accelerate vortioxetine action via presynaptic autoreceptor desensitization and postsynaptic activation [7-10] Anxiolytic effects [7]
5-HT1B	5-HT; glutamate	Partial agonist	Improve cognition [2]
5-HT1D	5-HT	Antagonist	Potentiate increase in synaptic serotonin levels [10]
5-HT3	5-HT; glutamate; NE; Ach	Antagonist	Improve cognition [2,7]
5-HT7	5-HT; glutamate	Antagonist	Improve mood [2,7,9] Improve cognition [2]

5-HT: Serotonin; Ach: Acetylcholine; NE: Norepinephrine; SERT: Serotonin transporter.

4. Therapeutic efficacy

Several reviews of vortioxetine preclinical and clinical data analyzing the therapeutic efficacy of vortioxetine have been previously published [10,16]. As well, two recent meta-analyses assessed the efficacy of vortioxetine in randomized controlled trials [17,18]. Herein we provide a brief overview of the efficacy of vortioxetine in treating adult patients with depression, highlighting vortioxetine’s clinically significant effects on cognition.

4.1 Efficacy results – MDD – affective symptomatology

Table 2 summarizes results of key short-term (6 – 8-week), randomized, double-blind, placebo-controlled trials assessing the efficacy of vortioxetine for MDD. In all trials, patients at baseline had a current MDE with moderate to severe depression, as determined by Montgomery-Åsberg Depression Rating Scale (MADRS) score. Vortioxetine daily doses ranged from 1 to 20 mg, with 10 mg being most commonly studied. In six trials, an active reference was used: venlafaxine 225 mg/day [19] or duloxetine 60 mg/day [20-24]. The primary efficacy measure was the change from baseline in either MADRS total score [19,21,23,24] or the 24-item Hamilton Rating Scale for Depression (HAMD-24) [20,22,25,26]. Five trials were considered positive in that vortioxetine separated from placebo at all doses studied [19,20,23,25] or one of the doses [24].

Table 2. Efficacy of vortioxetine versus placebo in short-term, randomized, double-blind placebo-controlled trials in major depressive disorder.

Study (n*)	Methodology	Population	Primary efficacy measure (related to depression)	Statistical method for primary efficacy analysis	Intervention/comparator (n*)	Mean change from baseline (in primary measure score)
Alvarez et al. [19] (425)	Randomized, 6-week, double-blind, fixed-dose, placebo-controlled, active-referenced multi-country (non-US)	Adults with current MDE MADRS ≥ 30 at baseline	Change from baseline to week 6 in MADRS total score	LOCF, ANCOVA	1. Vortioxetine 5 mg/day (108) 2. Vortioxetine 10 mg/day (100) 3. Venlafaxine 225 mg/day; titrated up from 75 mg/day (112) 4. Placebo (105)	(-20.4)^‡ (-20.2)^‡ (-20.9)^‡ (-14.5)
Katona et al. [20] (446)	Randomized, 8-week, double-blind, fixed-dose, placebo-controlled, active-referenced multi-country (including US)	Elderly adults (age ≥ 65 years) with current MDE MADRS ≥ 26 at baseline	Change from baseline to week 8 in HAMD-24 total score	LOCF, ANCOVA	1. Vortioxetine 5 mg/day (154) 2. Duloxetine 60 mg/day (147) 3. Placebo (145)	(-13.7)^§ (-15.8)^‡ (-10.3)
Baldwin et al. [21] (755)	Randomized, 8-week, double-blind, fixed-dose, placebo-controlled, active-referenced multi-country (non-US)	Adults with current MDE MADRS ≥ 26 at baseline	Change from baseline to week 8 in MADRS total score	LOCF, ANCOVA	1. Vortioxetine 2.5 mg/day (155) 2. Vortioxetine 5 mg/day (155) 3. Vortioxetine 10 mg/day (151) 4. Duloxetine 60 mg/day (149) 5. Placebo (145)	(-16.2) (-16.5) (-16.3) (-16.8) (-14.8)
Mahableshwarkar et al. [22] (597)	Randomized, 8-week, double-blind, fixed-dose, placebo-controlled, active-referenced single country (US)	Adults with current MDE MADRS ≥ 22 at baseline	Change from baseline to week 8 in HAMD-24 total score	LOCF, ANCOVA	1. Vortioxetine 2.5 mg/day (146) 2. Vortioxetine 5 mg/day (153) 3. Duloxetine 60 mg/day (149) 4. Placebo (149)	(-12.04) (-11.08) (-13.47)^¶ (-10.5)
Boulenger et al. [23] (604)	Randomized, 8-week, double-blind, fixed-dose, placebo-controlled, active-referenced multi-country (non-US)	Adults with current MDE MADRS ≥ 26 at baseline CGI-S ≥ 4 at baseline	Change from baseline to week 8 in MADRS total score	MMRM	1. Vortioxetine 15 mg/day (149) 2. Vortioxetine 20 mg/day (151) 3. Duloxetine 60 mg/day (146) 4. Placebo (158)	(-17.2)^‡ (-18.8)^‡ (-21.2)^‡ (-11.7)
Mahableshwarkar et al. [24] (591)	Randomized, 8-week, double-blind, fixed-dose, placebo-controlled, active-referenced single country (US)	Adults with current MDE MADRS ≥ 26 at baseline CGI-S ≥ 4 at baseline	Change from baseline to week 8 in MADRS total score	MMRM	1. Vortioxetine 15 mg/day (145) 2. Vortioxetine 20 mg/day (147) 3. Duloxetine 60 mg/day (146) 4. Placebo (153)	(-14.3) (-15.57)^# (-16.9)** (-12.83)
Henigsberg et al. [25] (503)	Randomized, 8-week, double-blind, fixed-dose, placebo-controlled multi-country (non-US)	Adults with current MDE MADRS ≥ 26 at baseline	Change from baseline to week 8 in HAMD-24 total score	MMRM	1. Vortioxetine 1 mg/day (124) 2. Vortioxetine 5 mg/day (129) 3. Vortioxetine 10 mg/day (122) 4. Placebo (128)	(-14.82)** (-15.42)** (-16.23)** (-11.3)
Jain et al. [26] (578)	Randomized, 6-week, double-blind, fixed-dose, placebo-controlled single country (US)	Adults with current MDE MADRS ≥ 30 at baseline	Change from baseline to week 6 in HAMD-24 total score	LOCF, ANCOVA	1. Vortioxetine 5 mg/day (292) 2. Placebo (286)	(-14.61) (-13.87)

*Based on full analysis set (unless otherwise stated).

^‡p < 0.0001 vs placebo.

^§p = 0.0011 vs placebo.

^¶p = 0.005 vs placebo.

^#p = 0.023.

**p < 0.001 vs placebo.

ANCOVA: Analysis of covariance; CGI-S: Clinical global impression – severity; HAMD-24: Hamilton rating scale for depression (24 items); LOCF: Last observation carried forward; MADRS: Montgomery-Åsberg depression rating scale; MDE: Major depressive episode; MMRM: Mixed model for repeated measures.

The inclusion criteria were similar for these trials, but one trial of note limited its patient population to elderly adults [20]. This is significant as there are few placebo-controlled clinical trials that assess antidepressant use for unipolar MDD in an elderly population [20].

In addition to the short-term efficacy trials, several medium- and long-term investigations have been conducted. Boulenger et al. conducted a placebo-controlled, double-blind, fixed-dose study to evaluate the efficacy of vortioxetine in the prevention of MDD relapse [27]. Patients recruited from 17 countries were eligible if they had a current MDE with moderate to severe depression at baseline (MADRS score ≥ 26). After a 12-week open-label period of treatment with vortioxetine 5 or 10 mg, 400 patients who had a remission of their MDE (as defined by MADRS score ≤ 10) at both weeks 10 and 12 were randomized to either continue vortioxetine (n = 206) according to the dose they were previously receiving or switch to placebo (n = 194). The primary efficacy endpoint was time to relapse of MDD within the first 24 weeks post-randomization. Using a Cox proportional hazard model, time to relapse was significantly longer in the vortioxetine group (p = 0.035). Secondary analyses of MADRS score, 17-item HAMD score, and Clinical Global Impression–Severity (CGI-S) scale also favored vortioxetine over placebo.

Two open-label extension studies assessed the long-term effectiveness and safety of vortioxetine in MDD. After completing a lead-in study [21], 535 patients entered the first of the two extension trials [28] and received vortioxetine doses of 2.5 – 10 mg/day. Effectiveness was measured with MADRS total score. After 52 weeks, MADRS mean total score had decreased by 8 points from the open-label baseline, with response to treatment in 94% of patients (63% at entry into the study) and remission in 83% of patients (42% at entry) (observed case rates). The last observation carried forward (LOCF) remission and response rates were 71.2 and 84.3%, respectively. Similar to the lead-in study, nausea and headache were the most common adverse events (AEs) in the trial. Approximately 8% of patients withdrew from the study because of AEs. The second long-term study [29] was an extension of two short-term trials [22,25]. In total, 834 patients entered the extension study in which they received vortioxetine 2.5 – 10 mg/day. At the final visit, the HAMD-24 mean total score had decreased by 7.9 points from the open-label baseline of the extension study. The response and remission rates assessed at the final visit were 51 and 55.6%, respectively. The rate of withdrawal secondary to AEs was 6%.

Based on efficacy results from short- and long-term trials, vortioxetine was approved for the treatment of MDD in adults at daily doses of 5, 10, 15 and 20 mg. The recommended starting daily dose is 10 mg for adults and 5 mg for elderly adults (i.e., those over 65 years of age) [3].

In the first active comparator MDD efficacy trial to be reported, Montgomery et al. randomized 501 adult patients with depression and an inadequate response to SSRI or SNRI therapy to receive either vortioxetine (10 – 20 mg daily) or the serotonergic/melatonergic antidepressant agomelatine (25 – 50 mg daily) in a double-blind manner for 12 weeks [30]. Vortioxetine was significantly superior to agomelatine in the primary efficacy outcome (change from baseline of MADRS total score at 8 weeks; p < 0.01) as well as most secondary outcomes related to mood, anxiety and productivity (p < 0.05 or lower for most secondary efficacy variables). Secondary efficacy and assessment measures included the CGI-S, CGI-Improvement (CGI-I) and Hamilton Anxiety Rating Scale, as well as the patient-reported outcomes from the Sheehan Disability Scale and Depression and Family Functioning Scale. There were no differences noted in treatment-emergent AEs between the two groups.

In a randomized, double-blind study, Wang et al. also assessed the efficacy of vortioxetine in MDD versus an active comparator, the SNRI venlafaxine [31]. Four hundred and forty-three patients with moderate to severe depression (as indicated by MADRS and CGI-S scores) were randomized to receive 8 weeks of therapy with a fixed daily dose of either vortioxetine 10 mg or venlafaxine 150 mg. The primary endpoint of mean change from baseline of MADRS total score at week 8 was -19.4 for vortioxetine and -18.2 for venlafaxine (difference = -1.2 points; 95% confidence interval = -3.03 to +0.63). Since the non-inferiority margin was +2.5 MADRS score points, the primary efficacy analysis demonstrated non-inferiority of vortioxetine to venlafaxine. Non-inferiority was also noted in the secondary efficacy analyses, which included both clinician-rated assessments and patient-reported outcomes. Treatment-emergent AEs were generally comparable between vortioxetine- and venlafaxine-treated patients, although significantly more patients in the venlafaxine group withdrew from the study as compared to the vortioxetine group.

4.2 Efficacy results – pro-cognitive effects in preclinical models

Various preclinical studies have investigated the pro-cognitive effects of vortioxetine [10]. In an in vitro hippocampal pyramidal cell study, vortioxetine was found to potentiate excitatory synaptic transmission and plasticity associated with learning and memory. Escitalopram, meanwhile, did not show any such pro-cognitive features [32]. Utilizing behavioral models of cognition (e.g., novel object recognition tests), vortioxetine achieved memory enhancements in rats. Additionally, vortioxetine was noted to increase ACh and HA levels in the medial prefrontal cortex. This is significant from a cognitive perspective as ACh is associated with learning and memory and HA is implicated in attention, vigilance and memory [11]. In several studies, rats were treated with a tryptophan hydroxylase inhibitor in order to deplete central 5-HT and consequently impair cognition. Vortioxetine demonstrated a reversal of this induced impairment in rats as assessed by the novel object recognition and Y-maze spontaneous alteration (spatial memory) tests [33,34]. The SSRI escitalopram and SNRI duloxetine, however, did not reverse any deficits [34]. Further, the beneficial effects in the object recognition task were maintained with chronic vortioxetine administration, but no improvement was noted in performance in the spontaneous alternation test [33]. With acute and subchronic vortioxetine administration, Wallace et al. also showed reversal of induced cognitive impairment in rats using the attentional set-shifting test as a measure of reversal learning [35]. In the same study, chronic vortioxetine administration prevented stress-induced learning deficits [35]. Based on these and other findings, it is hypothesized that the cognitive benefits of vortioxetine may be attributed to its action at various serotonin receptors, rather than its SERT inhibition. Further, preclinical outcomes provide mechanistic models for the pro-cognitive effects observed in clinical studies of vortioxetine.

4.3 Efficacy results – pro-cognitive effects in clinical trials

In a randomized, placebo-controlled study in 24 healthy volunteers, vortioxetine 10 mg did not negatively affect driving, psychomotor or cognitive performance with acute or chronic administration. Meanwhile, mirtazapine 30 mg, the active comparator, demonstrated cognitive and psychomotor impairment after single dose administration [36]. These results provided support that vortioxetine does not adversely impact cognition.

Several clinical studies sought to determine the extent of vortioxetine’s independent pro-cognitive effects (Table 3). One randomized clinical trial specifically assessed the cognitive effects of vortioxetine in depressed patients as a primary endpoint [37]. This double-blind, fixed-dose, placebo-controlled study randomized adult patients with a current MDE and MADRS score ≥ 26 to receive vortioxetine 10 mg/day, vortioxetine 20 mg/day or placebo for a period of 8 weeks. The primary efficacy measure was the change from baseline to 8 weeks in a composite score that incorporated scores from two neuropsychological tests: the Digit Symbol Substitution Test (DSST; assessing information-processing speed, executive function, and attention) and the Rey Auditory Verbal Learning Test (RAVLT; assessing learning and memory). Included in this analysis were 193, 204 and 194 patients from the vortioxetine 10 mg/day, vortioxetine 20 mg/day and placebo groups, respectively. When compared to placebo, both doses of vortioxetine were found to be statistically significantly superior (p < 0.001) in terms of the primary endpoint. In secondary analyses evaluating individual neuropsychological tests as well as patient-reported measures, vortioxetine at both doses also showed superiority versus placebo. Interestingly, the effect of vortioxetine on cognition was maintained even when corrected for the effect on depressive symptoms.

Table 3. Cognitive efficacy of vortioxetine versus placebo in short-term, randomized, double-blind placebo-controlled trials in major depressive disorder.

Study (n*)	Methodology	Population	Efficacy measure (related to cognition)	Statistical method for efficacy analysis	Intervention/comparator (n*)	Mean change from baseline (in efficacy measure score)
Katona et al. [20] (446)	Randomized, 8-week, double-blind, fixed-dose, placebo-controlled, active-referenced	Elderly adults (age ≥ 65 years) with current MDE MADRS ≥ 26 at baseline	Predefined exploratory analyses: change from baseline to week 8 in DSST – Number of correct symbols, RAVLT (acquisition), and RAVLT (delayed recall)	LOCF, ANCOVA	1. Vortioxetine 5 mg/day (154) 2. Duloxetine 60 mg/day (147) 3. Placebo (145)	DSST: 2.79^‡; RAVLT (acquisition): 1.14^‡; RAVLT (recall): 0.47^‡ DSST: 0.77; RAVLT (acquisition): 1.41^§; RAVLT (recall): 0.64^§ not reported
McIntyre et al. [37] (591)	Randomized, 8-week, double-blind, fixed-dose, placebo-controlled	Adults with current MDE MADRS ≥ 26 at baseline	Primary efficacy measure: change from baseline to week 8 in composite z-score (weighted sum of z-scores in the DSST and RAVLT)	MMRM	1. Vortioxetine 10 mg/day (193) 2. Vortioxetine 20 mg/day (204) 3. Placebo (194)	0.13^¶ 0.1^¶ (-0.24)
Mahableshwarkar et al. [38] (529)	Randomized, 8-week, double-blind, flexible dose, placebo-controlled, active-referenced	Adults with current MDE MADRS ≥ 26 at baseline DSST < 70 at baseline	Primary efficacy measure: change from baseline to week 8 in DSST – Number of correct symbols	MMRM	1. Vortioxetine 10–20 mg/day (175)** 2. Duloxetine 60 mg/day (187) 3. Placebo (167)	4.60^# 4.06 2.85

*Based on full analysis set.

^‡p < 0.05 vs placebo.

^§p < 0.01 vs placebo.

^¶p < 0.001 vs placebo.

^#p = 0.019 vs placebo.

**All patients randomized to vortioxetine received 10 mg for the first week.

ANCOVA: Analysis of covariance; DSST: Digit symbol substitution test; LOCF: Last observation carried forward; MADRS: Montgomery-Åsberg depression rating scale; MDE: Major depressive episode; MMRM: Mixed model for repeated measures; RAVLT: Rey auditory verbal learning test.

Mahableshwarkar et al. conducted a randomized placebo-controlled, duloxetine-referenced trial specifically assessing the effects of vortioxetine on cognition in patients with depression and self-reported cognitive dysfunction [38]. Vortioxetine 10 – 20 mg daily was superior to placebo in change from baseline of DSST score (p < 0.05), the primary endpoint, as well as perceived deficit questionnaire (PDQ; p = 0.001) and CGI-I (p < 0.05) scores. A significant improvement in functional capacity of vortioxetine-treated patients was also demonstrated by way of increased University of San Diego Performance-Based Skills Assessment scores. Notably, the reference antidepressant, duloxetine, did not separate from placebo in DSST or University of San Diego Performance-Based Skills Assessment scores, though it showed a statistically significant change in PDQ and CGI-I scores.

In secondary analyses of their study of elderly patients with MDE, Katona et al. also assessed effects of vortioxetine on cognitive function [20]. Vortioxetine was statistically superior (p < 0.05) to placebo in terms of improvement on the DSST and the RAVLT. This trial included duloxetine as an active reference and, compared to placebo, duloxetine showed an improvement in the RAVLT but not the DSST. These results, as well as those of Mahableshwarkar et al. [38], are consistent with results of a previous trial assessing effect of duloxetine on cognition in elderly adults, where duloxetine demonstrated superiority to placebo in improving scores for a modified RAVLT but not the DSST [39]. This indicates that vortioxetine may exert its positive effect on more cognitive domains than duloxetine.

As cognitive dysfunction is often a symptom of MDD, the ability to distinguish between a medication’s indirect and direct effects on cognition is valuable. Path analysis is a statistical approach that aims to look at the effect of an independent variable on a dependent outcome while attempting to adjust for the effect of other variables that may be moderational and/or colinear. Path analyses of current data assessing vortioxetine’s cognitive benefits suggest that benefits may be independent of the medication’s antidepressant effects, meaning that vortioxetine contributed directly to cognitive improvement [20,37,38]. Although both vortioxetine and duloxetine have demonstrated significant improvement in the treatment of depressive symptoms, only vortioxetine appeared to have a substantial direct effect on cognition, as measured by DSST [20,38].

5. Safety evaluation

Vortioxetine was generally well tolerated in clinical trials. Patients were more likely to discontinue vortioxetine than placebo because of AEs, while discontinuation due to an AE was more common in the active reference groups (i.e., duloxetine and venlafaxine) [16]. Table 4 summarizes the AEs commonly encountered in clinical trials [40].

Table 4. Percentage of individuals with common adverse events (incidence of at least 5% in 1 clinical trial).

Adverse event	Placebo	Vortioxetine: short-term trials				Vortioxetine: long-term trials
		5 mg	10 mg	15 mg	20 mg	2.5* – 20 mg
Constipation	1 – 7.3	3 – 6.1	5 – 10.4	6 – 9.3	6	NR
Diarrhea	3.8 – 8.9	1.9 – 9.2	5.3 – 12.9	9.9 – 10.6	7 – 12.6	4.9 – 7.2
Dizziness	2.1 – 9.4	2.5 – 9	2.5 – 8.4	4.6 – 19.7	5.3 – 19.7	2.5 – 19.7
Dry mouth	3.2 – 9.6	2 – 10.5	2 – 15.6	3.3 – 7	6 – 8	NR
Fatigue	2 – 5.7	1.5 – 7.1	1.5 – 6	4	3.3	NR
Gastroenteritis	NR	1.9 – 5.4	1.9 – 5.4	NR	NR	5.4
Headache	7.6 – 24.8	10.2 – 21.3	12.3 – 25	16.6 – 19.7	16 – 19.7	12.3 – 19.7
Hyperhydrosis	0.7 – 3.8	NR	10	3.3	0	NR
Influenza	NR	1.1 – 6.9	1.1 – 6.9	7	7	5 – 6.9
Insomnia	4 – 4.8	1.5 – 7.1	1.5 – 7.1	9.9	9.9	1.5 – 9.9
Nasopharyngitis	4.1 – 8.6	7 – 10.8	2.6 – 10.8	12.7	3.3 – 12.7	9.8 – 12.7
Nausea	4.3 – 10.6	8.8 – 29.6	8.8 – 38	26.5 – 33.8	29.3 – 32	8.8 – 31
Sexual dysfunction – ASEX (male:female)	14:20	16:22	20:23	19:33	29:34	< 2 (overall)
Sinusitis	NR	NR	NR	NR	NR	7
Somnolence	2.1 – 4	2.5 – 5.9	NR	NR	NR	NR
Upper respiratory tract infection	NR	NR	NR	NR	NR	6.4
Vomiting	1 – 3.4	1.9 – 3.8	4.5 – 9	6 – 11.9	5.3 – 6	NR
Weight gain	NR	NR	NR	NR	NR	5.8

Adapted from Pearce and Murphy [40].

*Note that vortioxetine 2.5 mg daily is not an approved dose for major depressive disorder.

ASEX: Arizona sexual experiences scale; NR: Not reported.

5.1 Gastrointestinal effects

The most common AEs reported in trials were gastrointestinal in nature (e.g., nausea, constipation and vomiting). Citrome calculated numbers needed to harm for all vortioxetine doses pooled of 6 for nausea, 28 for vomiting and 64 for constipation [16]. Nausea occurred most frequently and appeared to be dose-related, with up to 32% of patients reporting this AE at a dose of vortioxetine 20 mg daily [16]. Although it was transient and mostly of mild to moderate severity, it was also the single most common reason for discontinuation of vortioxetine [3]. In the active-referenced randomized controlled trials, rates of nausea were generally comparable or lower than the reference agent (i.e., venlafaxine or duloxetine) [19-23].

5.2 Sexual dysfunction

Rates of spontaneously self-reported sexual dysfunction were low in randomized as well as open-label trials [16]. Several studies also prospectively assessed the presence of sexual AEs in participants using the Arizona Sexual Experiences (ASEX) scale, a reliable, objective measure of sexual dysfunction. This is particularly valuable as sexual dysfunction may be underreported though it is a very common AE associated with SSRIs that can often lead to nonadherence. In pooled analysis of seven placebo-controlled clinical trials, men had an ASEX incidence of treatment-emergent sexual dysfunction of 16, 20, 19 and 29% for vortioxetine daily doses of 5, 10, 15 and 20 mg, respectively, versus 14% incidence for placebo. Meanwhile, women had an ASEX incidence of treatment-emergent sexual dysfunction of 22, 23, 33 and 34% for daily doses of 5, 10, 15 and 20 mg, respectively, versus 20% incidence for placebo [3].

In an 8-week, randomized, double-blind trial, 447 patients with well-controlled MDD and treatment-emergent sexual dysfunction on SSRI therapy were randomized to switch to either vortioxetine or escitalopram (each starting at 10 mg daily with potential dose increases to 20 mg) [41]. The primary endpoint was a change from baseline in the Changes in Sexual Functioning Questionnaire Short-Form (CSFQ-14) total score at week 8. Compared to escitalopram, vortioxetine showed significantly greater improvement in CSFQ-14 score (p = 0.013). Secondary endpoints assessing clinical efficacy were comparable between the two groups. This study shows potential for vortioxetine to reverse sexual dysfunction caused by SSRI therapy while maintaining clinical efficacy.

5.3 Cardiovascular risk

Analysis of data from short-term trials showed that vortioxetine treatment did not lead to any clinically significant changes in blood pressure and heart rate [3]. Further, vortioxetine was not found to have a significant increase in QTc interval duration in a randomized, 2-week, four–treatment arm study of 10 and 40 mg doses with placebo and active (i.e., moxifloxacin) controls [3,42].

Serotonergic transmission and modulation play a role in hemostasis, and serotonin reuptake inhibitors have been associated with increased risk of bleeding [43]. Vortioxetine is not known to be independently physiologically active at 5-HT2A, the receptor subtype associated with platelet aggregation [44]. However, it may still contribute to hemostatic alterations via its SERT inhibition and consequent increase in extracellular serotonin. Therefore, while coagulopathies were not observed in vortioxetine clinical trials, it is recommended that the medication be used with caution in patients with bleeding disorders [3].

From current clinical trial data, vortioxetine appears to have a favorable safety and AE profile and, thus, potentially greater tolerability than conventional monoaminergic antidepressants.

6. Summary and conclusions

Vortioxetine is a new antidepressant with a multimodal mechanism of action, acting on the 5-HT1A, 5-HT1B, 5-HT1D, 5-HT3 and 5-HT7 receptors in addition to inhibiting SERT. These diverse actions contribute to vortioxetine’s clinical effects by modulating serotonin neurotransmission and other neurotransmitters associated with mood and cognition. The foregoing effects on the serotonergic system exist in concert with pleiotropic effects on other monoaminergic systems as well as on glutamate, GABA, ACh and HA.

Cognitive symptoms are often present in MDD, contributing to the condition’s incapacitating effects [45]. Thorough reviews of cognitive deficits in MDD as well as the neurocognitive tests available to assess cognitive performance have been published elsewhere [45,46]. Only a limited number of small trials have assessed the direct and independent effects of antidepressant therapy on cognitive function in adult patients with MDD, wherein cognition is the primary outcome of interest. In these trials, it is not clear whether any cognitive improvement noted is independent of improvement in mood symptoms [45]. The potential for vortioxetine to improve cognition independent of its antidepressant effects is of great interest, with three currently open studies designed to assess cognition as a primary efficacy measure (NCT02272517; NCT02279953; NCT02279966) and one assessing the association between cognitive symptoms and work productivity (NCT02332954).

Similar to other trials assessing efficacy of different antidepressants, most placebo-controlled studies of vortioxetine efficacy in MDD have focused on identifying symptomatic improvement (with a focus on mood) rather than symptomatic remission. With vortioxetine’s potential to enhance cognition independent of mood improvement, it would be valuable to assess its role in affecting remission.

While vortioxetine demonstrated a significant improvement versus placebo in depressive symptoms in most clinical trials, not all trials had positive results. Several hypotheses have been presented as to why this may be the case [47,48]. For one, placebo effect tends to be prominent in antidepressant trials, and this may have contributed to the lack of clinical significance in some trials as well as minimizing observed therapeutic effect in the positive trials. Interestingly, two of the trials in which vortioxetine did not separate from placebo were conducted exclusively in the US. In a third trial also carried out exclusively in the US, vortioxetine demonstrated only modest benefit versus placebo [24]. It is proposed that higher rates of medication noncompliance in US studies are responsible for these results [49]. According to population pharmacokinetic/pharmacodynamic data from the manufacturer, samples from study patients in the US were more likely to have nonquantifiable concentrations (below the lower limit of quantification) of vortioxetine as compared to samples from patients in Europe or the rest of the world. These results could not be explained by differences in demographic factors (e.g., weight, body mass index) and, instead, suggest that noncompliance resulted in the noted differences in pharmacokinetics [49]. Additionally, the statistical model used for analyzing the data may have affected the results and led to non-separation from placebo across all active arms in one trial [21]. When the mixed model for repeated measures was used rather than LOCF analysis of covariance, a significant difference was noted. When assessing antidepressant effects, there is no single model that would be necessarily better than others in all respects and, as such, potential differences in methods should be considered when analyzing trial data.

SSRIs and SNRIs are commonly used as first-line antidepressant agents as they are considered to have more favorable AE profiles compared to other antidepressants. Still, they exhibit substantially increased risk for certain highly undesirable AEs, such as sexual dysfunction, which may contribute to medication nonadherence. In all efficacy and tolerability trials, patients had the opportunity to report sexual dysfunction as an AE. However, as spontaneous self-reporting of this AE is subject to underreporting, the use of sexual functioning scales provided a better reflection of treatment-emergent sexual AEs. A previous meta-analysis of treatment-emergent sexual dysfunction (as measured by various methods) in antidepressants found that SSRIs and SNRIs had rates of sexual dysfunction ranging from 25 to 80%, with fluoxetine, paroxetine, citalopram, venlafaxine and sertraline all having rates greater than 70% [50]. Further, in a trial of patients with treatment-emergent sexual dysfunction on SSRI treatment, those who were switched to vortioxetine showed significant improvement in CSFQ-14 total score as compared to patients who were switched to escitalopram [41]. Based on available data, vortioxetine appears advantageous in that it may have a lower incidence of sexual dysfunction than SSRIs or SNRIs and may potentially even reverse SSRI-induced sexual dysfunction. However, more direct comparison studies would be useful.

It is hypothesized that the nausea observed with vortioxetine is a function of the antidepressant’s SERT inhibition and 5-HT1A stimulation [10]. Vortioxetine’s antagonism of 5-HT3 receptors may play a role in mitigating gastrointestinal discomfort, and lower rates of this AE were indeed noted in clinical trials as compared to traditional SERT inhibitors. However, nausea remains a prominent AE with vortioxetine and, as gastrointestinal AEs are to some extent dose-dependent [3,16], they may be minimized by using the lowest effective dose of vortioxetine.

To establish the clinical efficacy and safety profile of vortioxetine versus other antidepressants, it is important to have more studies featuring active comparators. Only two studies to date have used active comparators in assessing vortioxetine efficacy. In the first, vortioxetine was found to be superior in efficacy to agomelatine [30]. While these results are interesting, they are limited by the fact that agomelatine is not available in North America and is considered to be only a modestly effective agent [51]. More promising are the results from the second active comparator trial that demonstrated non-inferiority of vortioxetine versus venlafaxine, a commonly used SNRI [31]. However, because neither trial included a placebo arm, the significance of their results is limited.

Vortioxetine has demonstrated efficacy in the clinical trial environment and exerts positive effect on affective symptomatology and cognitive performance in MDD. The independent effects on cognition address the emerging domain-specific cognitive deficit in mood disorders and may signal further applications of this agent. Considering the modest therapeutic outcome of traditional antidepressant agents, vortioxetine’s multi-modal mechanism of action, ability to target multiple therapeutic domains of depressive pathology, robust empirical demonstration of efficacy and tolerable adverse effect profile characterize vortioxetine as a promising novel agent for the treatment of MDD.

7. Expert opinion

Drug development in mood disorders has adhered to a familiar formula for the past several decades. Specifically, the design of registration trials has primarily sought to determine the effect of an antidepressant on total depression symptom severity. During the past decade, the point has been pressed strongly that symptomatic remission in MDD is a critical symptomatic endpoint. There is an absence of compelling data that highly reproducible differences exist between antidepressants in their ability to achieve remission when using group-level data. The design of the clinical trials unfortunately has not provided sufficient differentiation of antidepressants particularly as it relates to their efficacy in prespecified symptoms/dimensions/domains.

It is well established that cognitive dysfunction is a trans-nosological domain-affecting disorders across the developmental trajectory (e.g., Attention Deficit Hyperactivity Disorder, schizophrenia and mild cognitive impairment). The pertinence of cognitive dysfunction in the foregoing disorders, as well as in MDD, is instantiated by highly replicated study results indicating that disturbances in the cognitive domain are the principle determinant of health outcomes in neuropsychiatric disorders from patient, provider and societal perspectives.

During the past decade, a shift toward the domain of cognition in MDD has occurred largely inspired by the public health observation that MDD is the leading cause of human capital cost among young people. Moreover, results from clinical trials, observational studies and effectiveness research indicate that, notwithstanding symptomatic remission, most individuals with MDD continue to report insufficient health outcomes.

The relevance of cognitive dysfunction from a research perspective is underscored by the recent Research Domain Criteria proposal from the National Institute of Mental Health. The Research Domain Criteria framework appropriately gives a particular priority to cognitive systems primarily because of its well-characterized neural substrates.

Vortioxetine is the first antidepressant that has demonstrated replicated evidence of efficacy in mitigating cognitive dysfunction across the adult age range in MDD, wherein cognitive function was the primary outcome of interest. In addition to demonstrating efficacy in the cognitive domain, vortioxetine, but not duloxetine, was able to differentiate from placebo in a composite cognitive measure (i.e., DSST) that contemporaneously measures executive function, processing speed and attention. Results from path analysis, subgroup analysis, as well as the temporal improvement in overall depression symptom severity and cognitive performance provide evidence against the hypothesis that the vortioxetine effect in the domain of cognition is ‘pseudo-specific’ (i.e., epiphenomenon). Along with demonstrating an independent effect on cognition, vortioxetine has demonstrated a differential effect on measures of functional outcome when compared to duloxetine.

It is unlikely that vortioxetine’s benefit in the cognitive domain among adults with MDD is specific to that population. It is more likely the case that the benefit in the cognitive domain may be observed in other mental disorders. An additional opportunity worth exploring would be to combine vortioxetine with other psychotropic agents with known beneficial effects in the domain of cognition (e.g., psychostimulants) as well as with psychosocial strategies that primarily target the cognitive domain (e.g., cognitive remediation).

The foregoing review has given tremendous weight to the conventional neurochemical targets of interest in antidepressant/pro-cognition drug discovery. There would be much to be learned mechanistically by evaluating vortioxetine’s effect on neural circuit/subcircuit connectivity and reciprocity. It would be particularly interesting to know at what dose vortioxetine exerts circuitry effects and, as well, what are the effects of combination treatments with vortioxetine on neural circuit outcomes.

The economic costs of MDD are well established. The principal determinant of costs in MDD is impaired role function, notably in the workplace. It would be propitious to evaluate cognitive and functional outcomes in adults with MDD who are gainfully employed and/or currently receiving disability payments to determine whether the improvement in cognition offered by vortioxetine translates into benefits in workplace performance (e.g., decreased presenteeism). Such an outcome would be highly significant to all stakeholders in MDD and would provide an additional line of compelling evidence for timely detection, treatment and prevention of MDD to assure optimal outcomes and reduce costs.

Declaration of interest

No funding was received for the preparation of this manuscript. RS McIntyre has received research grants from Stanley Medical Research Institute, National Alliance for Research on Schizophrenia and Depression (NARSAD), National Institutes of Mental Health and Canadian Institutes of Health Research. He has participated in advisory boards, received speaker’s fees and research grants from Eli Lilly, Lundbeck, Pfizer, Sunovion and Forest. Speaker fees have also been received from Merck, Otsuka and Takeda, with additional research grants from Shire and Bristol-Myers Squibb. He has also participated in advisory boards for Bristol-Myers Squibb, Shire, Merck Otsuka and Takeda. He has also received speaker's fees and grants from Janssen-Ortho and AstraZeneca. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.