Effectiveness and Safety of Vortioxetine for the Treatment of Major Depressive Disorder in the Real World: A Systematic Review and Meta-Analysis

Abstract

Introduction

Major depressive disorder (MDD) is a highly prevalent and burdensome condition. This study aims to evaluate the effectiveness, tolerability, and safety of vortioxetine in treating MDD based on real-world data.

Methods

A systematic search of 8 electronic databases was performed from inception until October 2022 to identify real-world studies, excluding randomized controlled trials. We conducted subgroup, meta-regression, sensitivity analyses, publication bias, and quality assessments using the random-effects model. The effects were summarized by rates or standardized mean difference (SMD) with 95% confidence interval (CI).

Results

Of the 870 records identified, 11 studies (3139 participants) and 10 case reports or series were eligible for inclusion. Vortioxetine significantly relieved depression symptoms as assessed by both patients (SMD = 2.25, 95% CI = 1.60−2.89) and physicians (SMD = 3.73, 95% CI = 2.78−4.69). Cognitive function (SMD =1.86, 95% CI = 1.11−2.62) and functional disability (SMD =1.71, 95% CI = 1.14−2.29) were similarly markedly improved. Subgroup and meta-regression analyses showed that geographic location and medication regimen (whether combined with other antidepressants) were crucial factors influencing effectiveness (in terms of depression severity and cognitive function), potentially contributing to significant heterogeneity. The estimated response and remission rates were 66.4% (95% CI = 51.2%−81.5%) and 58.0% (95% CI = 48.9%−67.1%), respectively. Vortioxetine was well tolerated, with a pooled dropout rate of 3.5% (95% CI = 1.8%−5.8%), and the most common adverse event was nausea, with an estimated rate of 8.9% (95% CI = 3.8%−15.8%).

Limitations

The study has some limitations, including significant heterogeneity and limited evidence for some outcomes.

Conclusions

Vortioxetine is effective, well tolerated, and safe for treating MDD in clinical practice, with significant improvements observed in depressive severity, cognitive function, and functioning. Future studies should directly compare vortioxetine with other antidepressants in real-world settings to further evaluate its clinical utility.

INTRODUCTION

Major depressive disorder (MDD) is a highly prevalent psychiatric disorder (Herrman et al., 2022) affecting approximately 5% of adults globally (World Health Organization, 2021). Over the past 30 years, depression has remained a leading cause of years lived with disability, contributing significantly to the global disease burden (GBD 2015 Disease and Injury Incidence and Prevalence Collaborators, 2016; Malhi and Mann, 2018; GBD 2015 Disease and Injury Incidence and Prevalence Collaborators, 2022). The pervasiveness of MDD and its negative impact on morbidity, disability, and mortality underscore the pressing need for efficacious treatment programs (Cuijpers et al., 2014). Although there are numerous therapies available for MDD, the critical challenge lies in their implementation in real-world settings (Cipriani and Tomlinson, 2019; Cuijpers et al., 2020).

Different classes of antidepressants typically have been used to treat MDD, such as selective serotonin reuptake inhibitors, serotonin noradrenaline reuptake inhibitors, tricyclic antidepressants, and others (Cipriani et al., 2018). Despite this, there is still considerable debate surrounding their efficacy and effectiveness (Ioannidis, 2008; Cipriani et al., 2018), particularly given that approximately 50% of patients do not achieve remission with first-line therapy (Caldiroli et al., 2021). Consequently, the need to discover novel antidepressants with superior effectiveness and tolerability in treating MDD is crucial.

The novel multimodal antidepressant vortioxetine modulates several neurotransmitter systems by directly modulating serotonin receptor and inhibiting the serotonin transporter (Bang-Andersen et al., 2011; Mork et al., 2012). Many randomized controlled trials (RCTs) under ideal conditions and meta-analyses based on RCTs have confirmed that vortioxetine’s efficacy in adults is comparable with most other antidepressants (Thase et al., 2016; Cipriani et al., 2018; Inoue et al., 2018; Nishimura et al., 2018; Wagner et al., 2018; Zheng et al., 2019; Inoue et al., 2020; Thase et al., 2022; Zhang et al., 2022). Additionally, vortioxetine has shown significant benefits in treating cognitive symptoms of depression (Mahableshwarkar et al., 2015; Huang et al., 2022). Although RCTs are the most reliable source of evidence for guiding treatment decisions, real-world research can challenge their findings because controlled settings and patients recruited for RCTs may significantly differ from those in routine clinical practice. Consequently, real-world evidence is equally necessary for providing information on therapeutic effectiveness and safety predictions. In this study, we aimed to conduct a meta-analysis of real-world data, excluding idealized RCTs, to evaluate the effectiveness, tolerability, and safety of vortioxetine in treating patients with MDD.

METHODS

This meta-analysis has been registered on PROSPERO (CRD42022362161) and adhered to the guidelines in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement (Page et al., 2021).

Search Strategy and Study Selection

PubMed, Embase, the Cochrane Central Register of Controlled Trials, Web of Science, PsycINFO, ProQuest Dissertations, CINAHL (Cumulative Index to Nursing and Allied Health Literature), and LiLACS (Literatura Latino Americana em Ciências da Saúde) were systematically searched to identify literature published from inception to October 2, 2022. We used a combination of medical subject headings and free terms based on the following 3 words and their synonyms: depression, vortioxetine, and real-world study. The supplementary materials contain detailed search strategies for each database.

We included all real-world studies that evaluated the effectiveness and safety of vortioxetine for the treatment of people with a primary diagnosis of MDD. Trials that recruited participants with secondary diagnoses of comorbid general psychiatric disorders, such as anxiety or binge eating disorder, were also included. Real-world studies primarily included prospective or retrospective observational studies, and single-arm open-label trials that nicely simulated real-world settings also qualified. Case reports and case series were eligible for qualitative analysis. We excluded studies that met the following criteria: (1) diagnosed treatment-resistant depression, bipolar depression, or combined with substance abuse; primary diagnoses other than MDD, enrolled participants with secondary depressive disorders (e.g., poststroke depression), or depressive symptoms associated with other diseases; (2) non-original investigations: reviews, meta-analyses, letters, editorials, commentaries, errata; animal or in vitro model experiments and cell-line studies; and (3) conference abstracts or studies for which the authors had been contacted but valid data were still unable to be extracted. When more than 1 article was from the same trial, the article containing the comprehensive data set was included to avoid data duplication. There were no language or publication date restrictions.

Two investigators (Zejun Li and Shouhuan Liu) independently screened titles and abstracts and assessed potentially eligible studies by reviewing the full text. Any disagreements were resolved by a third reviewer (Tieqiao Liu). Inter-rater reliability was assessed using the kappa score, which demonstrated a high degree of agreement (κ = .83). An overview of the literature selection process and the reasons for exclusion are presented in Figure 1.

Figure 1.

PRISMA 2020 flow diagram for identification of studies via databases.

Data Extraction and Quality Assessment

Two investigators (Zejun Li and Shouhuan Liu) independently extracted data and evaluated the methodological quality of eligible studies. Any disagreements were arbitrated by a third researcher (Tieqiao Liu). A standardized data abstraction form was created using Microsoft Excel to document the extracted information, including basic information (title, first author, publication date, and the country conducting the studies), general methodological details (study design, original inclusion or exclusion criteria, and length of follow-up), demographic information (mean age, gender, setting, and sample size), clinical characteristics (underlying disease conditions, previous treatment history, type of depression, and diagnostic manual), treatment strategy (dosage, treatment duration, and whether combined with other antidepressants), and outcome data (baseline and posttreatment depression severity score; type of scale, type of adverse events (AEs) and number of cases, number of posttreatment responses and remissions, and the number of people who withdraw for various reasons). The SD of the partial means was obtained from the confidence interval (CI) (Cumpston et al., 2019). If the study authors did not provide sufficient outcome information, they were contacted for clarification. Data extraction software (GetData Graph Digitizer 2.26) was used to extract data from the original literature.

To assess the quality of included studies, we used the modified risk of bias tool, which was originally designed for case series (Murad et al., 2018) and has been adopted in previous studies (Alnefeesi et al., 2021). The checklist used for this assessment is presented in Table 1.

Table 1.

Modified Quality Assessment Checklist Applied in the Present Study

Selection	Does the patient(s) represent(s) the whole experience of the investigator (center) or is the selection method unclear to the extent that other patients with similar presentation may not have been reported?
Ascertainment	Was the exposure adequately ascertained? Was the outcome adequately ascertained
Causality	Were other alternative causes that may explain the observation ruled out?
Reporting	Is the case(s) described with sufficient details to allow other investigators to replicate the research or to allow practitioners to make inferences related to their own practice?

Outcomes

The study aimed to determine the efficacy and safety of vortioxetine as a treatment for depression. This study’s primary outcomes included the effectiveness of vortioxetine, defined as improvement in depression severity (as assessed by self- or other-rated scales), cognitive function (as assessed by the Perceived Deficits Questionnaire-Depression), and functional disability (as assessed using the patient-reported Sheehan Disability Scale) from baseline to posttreatment. We also evaluated the safety of vortioxetine, including types of AEs and the incidence of the most common AEs. The second primary outcomes were (1) response and remission rates after treatment using specific criteria defined by each original study. Typically, the response rate was defined as at least a 50% improvement in depression severity after treatment compared with baseline, and remission rate was defined as, for example, Hamilton Rating Scale for Depression-17 ≤7, Montgomery-Asberg Depression Rating Scale (MADRS) ≤12, Patient Health Questionnaire-9 ≤9, or Beck Depression Inventory ≤9, Clinical Global Impression-Severity (CGI-S) ≤3, etc. We also evaluated acceptability (discontinuation of treatment for all reasons, defined as the proportion of patients who discontinued treatment for any reason at the end of treatment), tolerability (discontinuation of therapy due to AEs, defined as the proportion of patients discontinuing treatment due to AEs), and the change in CGI-S score from baseline to endpoint.

Data Synthesis and Analysis

This meta-analysis was performed using Stata17.0 software. Continuous variables were represented by the standardized mean difference (SMD) and the corresponding 95% CI. Pooled rate estimates and 95% CI were used to express acceptability, tolerability, the incidence of nausea, response, and remission rates. When the majority of actual event rates were <0.2, the estimates were calculated based on the transformed values of the double arcsine method and then back-transformed to the original rates. The I² statistic was used to indicate heterogeneity among the included studies. A random-effects meta-analysis was conducted when there was considerable statistical heterogeneity (I² > 50% or P < .1).

Subgroup analyses and meta-regression were performed to explore potential sources of heterogeneity using variables such as country, continent, study design, follow-up period, sample size, literature quality (risk of bias), whether combined with other antidepressants, and type of depression scale. Visual examinations of funnel plots and Egger’s test were used to assess potential publication bias. The trim-and-fill analysis was used to evaluate the effect of publication bias on the combined effect size estimates. Influence analysis, a type of sensitivity analysis, was performed to identify the impact of individual studies on the pooled estimates. P < .05 was considered statistically significant in this study.

RESULTS

Search Results and Study Characteristics

A total of 870 records were systematically retrieved, with 1 additional record found from another source. After screening titles and abstracts, 62 potentially eligible reports were assessed in full text. Ultimately, 11 studies (involving 3139 participants) were included in the quantitative analysis (Chin et al., 2018; Chokka et al., 2019; Liguori et al., 2019; De Berardis et al., 2020; Minhas et al., 2021; Segura-Garcia et al., 2021; Yang et al., 2021; Di Nicola et al., 2022; Mattingly et al., 2022; Papalexi et al., 2022; Wang et al., 2022), and 10 case reports or case series (involving 13 patients) were included in the qualitative analysis (Cetin and Kose, 2018; Ozkan, 2019; Okumus, 2020; Yildiz, 2020; Gunes and Inan, 2021; Songur, 2021; Yoshimura et al., 2021; Isik et al., 2022; Tunc and Tunc, 2022; Yilbas, 2022). The 2 primary reasons for study exclusion were publication type ineligibility and non–real-world studies (Figure 1). Table 2 provides a summary of the basic characteristics of each trial used for quantitative analysis. The included studies reported prospective (8/11, 72.7%) and retrospective (3/11, 27.3%) real-world studies from 11 countries. The mean age of participants was 36.0 years (SD 18.5), and approximately one-third of the sample population was male (37.3%). At least 5 studies reported that participants were treated with other antidepressants before receiving vortioxetine (Chin et al., 2018; Chokka et al., 2019; De Berardis et al., 2020; Yang et al., 2021; Mattingly et al., 2022), and participants in at least 4 studies received vortioxetine in combination with other antidepressants (De Berardis et al., 2020; Yang et al., 2021; Mattingly et al., 2022; Wang et al., 2022). The median follow-up after treatment was 24 weeks (range 8−52 weeks). Of the included case reports or case series, 9 were from Turkey and 1 was from Japan (Yoshimura et al., 2021), with most of them reporting AEs (8/10, 80.0%) (supplemental Table 1) (Cetin and Kose, 2018; Ozkan, 2019; Okumus, 2020; Yildiz, 2020; Gunes and Inan, 2021; Songur, 2021; Isik et al., 2022; Tunc and Tunc, 2022).

Table 2.

Characteristics of the Studies Included for Quantitative Analysis

Study	Study design	Follow-up period	Area recruited from	Setting	Type of depression	Sample sizea	Mean age (SD)	Male (%)b	Dosage (mean ± SD, median) (mg/d)	Depression scales	Combined with other antidepressants
(Chin et al., 2018)	Multi-country, multi-site, prospective, non-interventional study (August 2016 and April 2017)	3 mo ± 14 d	Malaysia, Philippines, Singapore, Thailand	Not stated	MDD	131	39.7 (12.5)	34.1	−, 10.0	PHQ-9	Not stated
(Chokka et al., 2019)	Interventional, open-label, single-cohort study (February 2015 to July 2017)	52 wk	Canada	Not stated	MDD	199	40.8 (12.5)	30.6	15.2 ± 5.1, −	QIDS-SR	Not stated
(De Berardis et al., 2020)	Retrospective chart review study (January 2018 to February 2019)	8 wk	Italy	Outpatients	SSRI-resistant MDD	36	31.1 (5.8)	61.1	16.4, −	PHQ-9, HRDS-17	100%
(Di Nicola et al., 2022)	Retrospective analysis (July 2017 to January 2019)	6 mo	Italy	Outpatients	MDD	56	50.4 (12.7)	35.7	9.9 ± 2.7, −	MADRS	Not stated
(Liguori et al., 2019)	Retrospective, observational, single-center, questionnaire analysis based on individual charts review (February 2016 to August 2017)	6 mo	Italy	Inpatients	MDD and insomnia	15	43.6 (11.5)	40.0	Not stated (range: 5−10)	BDI	Not stated
(Mattingly et al., 2022)	Multinational, observational, prospective cohort study (November 2017 to January 2021)	24 wk	Canada, France, Italy, USA	Outpatients	MDD	737	49.3 (15.4)	35.8	12.9 ± 6.5, −	PHQ-9	170/737, 23.1%
(Minhas et al., 2021)	Non-interventional, multi-centred, cross-sectional, prospective study (2019-2020)	3 mo ± 14 d	Pakistan	Outpatients	MDD	498	34.6 (11.3)	51.2	10.8 ± 2.7, −	PHQ-9	Not stated
(Papalexi et al., 2022)	Non-interventional, prospective, multicenter study (January to September 2019)	3 mo	Greece	Outpatients	MDD	336	47.9 (14.3)	35.7	Not stated (range: 5−20)	PHQ-9, MADRS	Not stated
(Segura-Garcia et al., 2021)	Naturalistic, uncontrolled pilot study (August 2017 to October 2019)	24 wk	Italy	Outpatients	MDD and BED	30	42.0 (12.9)	13.3	15.8 ± 5.6, 20	BDI, HRSD-17	Not stated
(Wang et al., 2022)	Observational, multisite, single-arm, prospective, cohort study (December 2018 to March 2020)	24 ± 4 wk	China	Outpatients	MDD	859	33.1 (13.5)	34.1	11.1 ± 3.5, −	PHQ-9, HRSD-17	≥385/859, 44.8%
(Yang et al., 2021)	Non-interventional, prospective, multi-site, cohort study (June 2019 to August 2020)	3 mo	China	Outpatients	MDD	242	41.7 (14.0)	34.3	−, 9.5	PHQ-9	91/228, 39.9%

Abbreviations: BDI, Beck Depression Inventory; BED, binge eating disorder; HRSD, Hamilton Rating Scale for Depression; MADRS, Montgomery-Åsberg Depression Rating Scale; MDD, major depressive disorder; PHQ-9, Patient Health Questionnaire-9; QIDS-SR, Quick Inventory of Depressive Symptomatology–Self-Report; SSRI, selective serotonin reuptake inhibitor.

^aFull analysis set.

^bThe proportion of males in some studies was calculated based on the safety analysis set.

Quality Assessment and Sensitivity Analysis

The quality assessment (risk of bias assessment) for the studies included in this meta-analysis is summarized in supplemental Figure 1 and supplemental Table 2. The most common methodological issue was the inability to limit the effect of confounds (i.e., high risk of bias in the “causality” domain). Of the 11 studies included in the quantitative analysis, 5 were rated high on risk of bias (Liguori et al., 2019; De Berardis et al., 2020; Segura-Garcia et al., 2021; Yang et al., 2021; Wang et al., 2022), 2 as unclear (Di Nicola et al., 2022; Mattingly et al., 2022), and 4 as low (Chin et al., 2018; Chokka et al., 2019; Minhas et al., 2021; Papalexi et al., 2022). Three studies had a high risk of bias in the “selection” domain (Liguori et al., 2019; De Berardis et al., 2020; Segura-Garcia et al., 2021), 2 in the “causality” domain (Yang et al., 2021; Wang et al., 2022), and 1 in the “reporting” domain (Liguori et al., 2019) (supplemental Table 2). The sensitivity analysis indicated that none of the included studies had an inordinate influence on the results of effectiveness (improvement in depression severity, cognitive function, and functioning) and safety (incidence of nausea) (supplemental Figure 2). Egger’s test (P = .138, .103, .203, .823, respectively) showed a low likelihood of publication bias, although the funnel plot revealed a slight asymmetry (supplemental Figure 3). The trim-and-fill analysis suggested that the effect of unpublished studies on the meta-results was minimal (supplemental Figure 3).

Effectiveness on Depression Severity

Vortioxetine treatment significantly improved depression severity from the patient’s (self-rated scales) perspectives (number of studies = 9, SMD = 2.25, 95% CI = 1.60−2.89) (Figure 2A). Other rated scales also showed a significant improvement in depression severity from baseline to posttreatment (n = 5, SMD = 3.73, 95% CI = 2.78−4.69) (Figure 2B). From the physician’s (CGI-S) perspectives, we found a similarly favorable result, with an SMD of 2.40 (n = 8, 95% CI = 1.98−2.82) (supplemental Figure 4A).

Figure 2.

The forest plot of meta-analysis with vortioxetine treatment for major depressive disorder (MDD). (A) Improvement in depression severity (assessed using self-rated scales). (B) Improvement in depression severity (assessed using other rated scales). (C) Improvement in functional disability (assessed using the patient-reported Sheehan Disability Scale). (D) Improvement in cognitive function (assessed by the Perceived Deficits Questionnaire-Depression). (E) Incidence of nausea after transformation by double arcsine. (F) Remission rate after vortioxetine treatment. (G) Response rate after vortioxetine treatment.

However, there was substantial heterogeneity (I² = 98.9%) in the effect sizes of the included studies. Therefore, we conducted subgroup and meta-regression analyses to explore potential heterogeneity in the effectiveness outcome (improvement in depression severity based on self-rated scales). In the subgroup analysis (Figure 3A), the pooled SMD varied across different countries (P < .001), continents (P < .001), study designs (P = .025), literature quality (P = .010), whether combined with other antidepressants (P = .002), and type of depression scale (P = .001). However, significant heterogeneity persisted in all subgroups (range 90.1%−99.5%), indicating that these variables did not fully explain the heterogeneity.

Figure 3.

Subgroup analysis of improvement in depression severity (assessed using self-rated scales) (A) and cognitive function (B).

We found that the high-quality study group had a lower combined SMD of 1.38 (95% CI = 0.90−1.85) compared with the low-quality study group, which had an SMD of 3.06 (1.98−4.15). Studies that used vortioxetine alone or did not report whether other antidepressants were used had a higher pooled SMD (2.80, 1.84−3.76) compared with studies that combined vortioxetine with other antidepressants (SMD = 1.24, 95% CI = 1.00−1.48). Regarding the self-rated scales, most studies used the Patient Health Questionnaire-9 (6/9), with a pooled result of 2.06 (95% CI = 1.33−2.80) (Chin et al., 2018; Minhas et al., 2021; Yang et al., 2021; Mattingly et al., 2022; Papalexi et al., 2022; Wang et al., 2022).

Univariate meta-regression showed that country, literature quality, and whether combined with other antidepressants were likely to affect the heterogeneity of the results (Table 3). In multivariate meta-regression analysis, we found that study design, whether combined with other antidepressants, and type of self-rated scales were significant factors influencing depression severity improvement (R² = 62.06%), contributing to the observed heterogeneity (Table 3).

Table 3.

Results of Univariate and Multivariate Meta-Regression Analyses for Improvement in Depression Severity (Self-Rated Scales) After Vortioxetine Treatment

Variables		Coefficient	95% CI	Std. Err	P value
Univariate analysis
Country	China Canada Greece Italy Pakistan Multi-site	ref 2.3746 0.9235 0.7888 2.9904 0.7281	ref (0.9836, 3.7656) (−0.4348, 2.2818) (−0.4522, 2.0298) (1.6259, 4.3548) (−0.3861, 1.8423)	ref 0.7097 0.6930 0.6332 0.6961 0.5685	ref .001 .183 .213 <.001 .200
Study design	Prospective study Retrospective study	ref 1.6076	ref (−0.7040, 3.9193)	ref 1.1794	ref .173
Quality assessment	Low High Unclear	ref −1.4843 −1.6991	ref (−2.6524, −0.3162) (−3.4840, 0.0858)	ref 0.5960 0.9107	ref .013 .062
Combined with other antidepressants	Not stated Yes	ref −1.5500	ref (−2.478, −0.6220)	ref 0.4735	ref .001
Type of depression scale	BDI PHQ-9 QIDS-SR	ref −0.0858 1.3920	ref (−1.6843, 1.5128) (−0.9313, 3.7152)	ref 0.8156 1.1853	ref .916 .240
Follow-up period	<6 mo ≥ 6 mo	ref −0.2880	ref (−0.6178, 1.0418)	ref 0.6785	ref .671
Multivariate analysis
Study design	Prospective study Retrospective study	ref 2.7678	ref (0.6869, 4.8486)	ref 1.0617	ref .009
Combined with other antidepressants	Not stated Yes	ref −1.6766	ref (−2.6408, −0.7125)	ref 0.4919	ref .001
Type of depression scale	BDI PHQ-9 QIDS-SR	ref 1.9556 2.5874	ref (0.5127, 3.3985) (0.8309, 4.3440)	ref 0.7362 0.8962	ref .008 .004

Abbreviations: BDI, Beck Depression Inventory; PHQ-9, Patient Health Questionnaire-9; QIDS-SR, Quick Inventory of Depressive Symptomatology–Self-Report.

Effectiveness on Cognitive Function and Functioning

Overall, vortioxetine treatment was associated with significant improvement in both functional disability and cognitive function (Figure 2C–D), with a combined SMD of 1.86 (n = 8, 95% CI = 1.11−2.62) and 1.71 (n = 6, 1.14−2.29), respectively. However, there was considerable heterogeneity among the studies regarding cognitive function (I² = 99.3%), and this heterogeneity persisted across all subgroups (range 95.2%−99.7%). In particular, studies with lower quality and those that combined vortioxetine with other antidepressants had superior outcomes. Among different continents, one-half of the studies were conducted in Asia (4/8) (Chin et al., 2018; Minhas et al., 2021; Yang et al., 2021; Wang et al., 2022), with a summary SMD of 1.43 (95% CI = 0.59−2.26) (Figure 3B).

A multivariate meta-regression analysis revealed that continent, study design, and whether combined with other antidepressants were significant factors contributing to the heterogeneity of the results for cognitive function (R² = 80.35%) (Table 4).

Table 4.

Results of Univariate and Multivariate Meta-Regression Analyses for Improvement in Cognitive Function (Perceived Deficits Questionnaire-Depression) After Vortioxetine Treatment

Variables		Coefficient	95% CI	Std. Err	P value
Univariate analysis
Country	China Canada Greece Italy Pakistan Multi-site	ref 1.5625 4.6353 −0.0962 1.7014 0.5178	ref (0.2794, 2.8456) (3.3420, 5.9287) (−1.4022, 1.5527) (0.4395, 2.9633) (−0.5171, 1.5527)	ref 0.6547 0.6599 0.6663 0.6438 0.5280	ref .017 <.001 .885 .008 .327
Continent	Asia Europe North America Multi-continent	ref 1.6104 0.8966 −0.5706	ref (−0.4569, 3.6778) (−1.7670, 3.5601) (−3.2205, 2.0794)	ref 1.0548 1.3590 1.3520	Ref .127 .509 .673
Study design	Prospective study Retrospective study	ref −1.3673	ref (−3.7110, 0.9764)	ref 1.1958	ref .253
Quality assessment	Low High Unclear	ref −2.2151 −2.2111	ref (−3.8164, −0.6138) (−3.8208, −0.6015)	ref 0.8170 0.8213	ref .007 .007
Combined with other antidepressants	Not stated Yes	ref −1.7278	ref (−2.8815, −0.5741)	ref 0.5886	ref .003
Follow-up period	<6 mo ≥6 mo	ref −1.2577	ref (−2.7865, 0.2710)	ref 0.7800	ref .107
Multivariate analysis
Continent	Asia Europe North America Multi-continent	ref 3.2856 0.2127 0.0948	ref (2.0694, 4.5017) (−.9926, 1.4180) (−1.0719, 1.2614)	ref 0.6205 0.6149 0.5952	ref <.001 .729 .874
Study design	Prospective study Retrospective study	ref −4.7315	ref (−6.1591, −3.3040)	ref 0.7283	ref <.001
Combined with other antidepressants	Not stated Yes	ref −1.3492	ref (−2.3155, −0.3829)	ref 0.4930	ref .006

Response and Remission

Data on response rates were available in 8 studies, and remission rates were reported in 9 studies. The estimated response rate was 66.4% (95% CI = 51.2%−81.5%), and the remission rate was 58.0% (95% CI = 48.9%−67.1%) (Figure 2F–G).

Acceptability and Tolerability

In terms of acceptability, vortioxetine demonstrated a low estimated dropout rate of 8.4% (n = 9, 95% CI = 4.6%−13.2%). Furthermore, in regards to tolerability, vortioxetine was well tolerated, with a pooled dropout rate of 3.5% (n = 8, 1.8%−5.8%) (supplemental Figure 4B–C).

Safety

Eight studies (3181 patients) were included for quantitative analysis to evaluate the safety of vortioxetine (Chin et al., 2018; Chokka et al., 2019; De Berardis et al., 2020; Minhas et al., 2021; Yang et al., 2021; Di Nicola et al., 2022; Mattingly et al., 2022; Wang et al., 2022). Common AEs associated with vortioxetine were identified as nausea, headache, dizziness, and pruritus, which were reported in 8 studies (Chin et al., 2018; Chokka et al., 2019; De Berardis et al., 2020; Minhas et al., 2021; Yang et al., 2021; Di Nicola et al., 2022; Mattingly et al., 2022; Wang et al., 2022), 6 studies (Chin et al., 2018; Chokka et al., 2019; De Berardis et al., 2020; Minhas et al., 2021; Yang et al., 2021; Di Nicola et al., 2022; Mattingly et al., 2022; Wang et al., 2022), 5 studies (Chin et al., 2018; Chokka et al., 2019; Minhas et al., 2021; Yang et al., 2021; Wang et al., 2022), and 4 studies (Minhas et al., 2021; Yang et al., 2021; Mattingly et al., 2022; Wang et al., 2022), respectively (Table 5). Nausea was found to be the most frequently reported AE, with an estimated combined rate of 8.9% (95% CI = 3.8%−15.8%) (Figure 2E). Severe adverse drug reactions were reported in only 2 studies, affecting 6 participants (0.48%) (Yang et al., 2021; Wang et al., 2022) (Table 5). In addition, AEs were reported in 8 of the 10 included case reports or case series, with one-half of them (4/8) being skin-related AEs (Cetin and Kose, 2018; Okumus, 2020; Yildiz, 2020; Gunes and Inan, 2021) (supplemental Table 1).

Table 5.

Common Adverse Events of Vortioxetine and Their Incidence

Study	Safety population	Total AEs	Severe AEs	Total ADRs	Severe ADRs	Nausea	Headache	Dizziness	Pruritus	Other
(Chin et al., 2018)	138	N	N	13, 9.4%	N	3, 2.2%	1, 0.7%	1, 0.7%	−	Abdominal discomfort, 4, 2.9%; somnolence, 2, 1.4%; urticaria, 2, 1.4%; hypersensitivity, 2, 1.4%; rash, 1, 0.7%
(Chokka et al., 2019)	219	N	N	16, 7.3%	N	64, 29.2%	26, 11.9%	13, 5.9%	−	Insomnia, 20, 9.1%; nasopharyngitis, 15, 6.8%; anxiety, 14, 6.4%
(De Berardis et al., 2020)	36	N	0, 0%	N	N	7, 19.4%	3, 8.3%	−	−	Dry mouth, 2, 5.6%
(Di Nicola et al., 2022)	56	2	N	N	N	2, 3.6%a	−	−	−	−
(Mattingly et al., 2022)	985	209	23, 2.3%	N	N	81, 8.2%	15, 1.5%	−	15, 1.5%	Anxiety, 14, 1.4%
(Minhas et al., 2021)	498	N	N	0, 0%	0, 0%	15, 3.0%a	15, 3.0%	5, 1.0%	5, 1.0%	Severe irritability, 15, 3.0%; orthostatic hypotension (black out on change of posture), 5, 1.0%; abdominal discomfort, <1%
(Wang et al., 2022)	996	419	42, 4.2%	N	5, 0.5%	182, 18.7%	13, 1.3%	32, 3.2%	24, 2.4%	Vomiting, 31, 3.1%; decreased appetite, 11, 1.1%; somnolence, 10, 1.0%
(Yang et al., 2021)	253	17	9, 3.6%	9, 3.6%	1, 0.4%	3, 1.2%	−	2, 0.79%	2, 0.79%	Hospitalization (not related to vortioxetine), 8, 3.2%; gastrointestinal disturbances, 2, 0.8; suicide attempt, 1, 0.4%; diarrhea, 1, 0.4%
Summation	3181	647	74	38	6	340	73	53	46	−

Abbreviations: ADRs, adverse drug reactions; AEs, adverse events; N, not stated.

^aIncluding nausea and vomiting.

DISCUSSION

To our knowledge, this meta-analysis presented the first comprehensive synthesis of current data on the effectiveness and safety of vortioxetine for MDD in clinical practice. Our findings showed that vortioxetine significantly improved depression severity, cognitive function, and functional disability with high response and remission rates. Additionally, vortioxetine was well tolerated, with no severe new safety concerns.

Vortioxetine has been approved in more than 80 countries globally, and numerous RCTs supported its efficacy (Thase et al., 2016; Cipriani et al., 2018; Inoue et al., 2018; Nishimura et al., 2018; Wagner et al., 2018; Zheng et al., 2019; Inoue et al., 2020; Thase et al., 2022; Zhang et al., 2022). This meta-analysis confirmed the superior effectiveness of vortioxetine in treating MDD, as evidenced by the significant improvements in depression severity assessed through both self- and other-rated scales. Notably, improvements in other-rated scale scores appeared more significant, potentially due to physician overestimation of study results in open-label, real-world studies. Therefore, it may be more appropriate to judge improvements in depression severity from the patient’s perspective in real-world studies. Previous meta-analyses based on RCTs have also found vortioxetine to be effective in improving depression-related scores (MADRS scale) vs placebo in adults (Baldwin et al., 2016; Thase et al., 2016; Zheng et al., 2019; Huang et al., 2022), with a possible dose-response relationship (Baldwin et al., 2016; Thase et al., 2016). A meta-analysis of RCTs conducted by Zheng et al. (2019) found that a daily dose of 10 mg vortioxetine resulted in a significant difference in MADRS total score change from baseline compared with placebo (SMD = 3.50, 95% CI = 2.17−4.83). These results were comparable with the findings in our study (SMD = 3.73, 95% CI = 2.78−4.69). Additionally, vortioxetine has shown significant efficacy in patients with MDD and a history of trauma, reducing relapse risk and improving depression and anxiety symptoms and overall functioning (Christensen et al., 2020). In patients with MDD and high anxiety levels, vortioxetine (across daily doses of 5−20 mg) was an efficacious treatment option for reducing depressive and anxiety symptoms (Baldwin et al., 2016). However, a 12-week randomized, placebo-controlled study did not find a significant difference between the combined doses of vortioxetine and placebo in the pediatric-adolescent population (Findling et al., 2022).

Several studies have examined the efficacy of vortioxetine as the first step of treatment for patients with MDD compared with other second-generation antidepressants. Although some studies showed no apparent difference in efficacy between vortioxetine and other antidepressants (Wagner et al., 2018; Zhang et al., 2022), a comprehensive network meta-analysis of RCTs comparing the efficacy of 21 antidepressants for the acute treatment of MDD in adults found that vortioxetine was more effective than other antidepressants in head-to-head studies (Cipriani et al., 2018). Although some patients included in this review were switched to vortioxetine after initial treatment with other antidepressants proved ineffective, the combined results demonstrated excellent effectiveness. These findings suggested that the selection of vortioxetine was not inferior to that of other second-generation antidepressants. When choosing initial therapy for MDD in clinical practice, it is important to consider not only drug efficacy but also patient preferences such as AEs, costs, personal involvement, and possible adherence.

Increasingly, cognitive impairment was considered a core deficit in patients with MDD and is highly prevalent (McClintock et al., 2011). Even during remitted phases of MDD, some residual symptoms notably in executive function and attention domains often independently persist (McClintock et al., 2010, 2011; Wekking et al., 2012; Bortolato et al., 2014). Vortioxetine is the first FDA-approved treatment specifically for cognitive dysfunction in MDD (Lundbeck, 2018), and growing evidence, including the present study, supported its significant clinical benefits on cognitive function (Baune et al., 2018; Huang et al., 2022; Zhang et al., 2022). A meta-analysis of RCTs confirmed that both 10 and 20 mg of vortioxetine significantly increased Digit Symbol Substitution test scores and decreased Perceived Deficits Questionnaire-Depression scores in patients with MDD and cognitive dysfunction (Huang et al., 2022). Moreover, vortioxetine demonstrated positive influences on psychomotor speed, delayed recall (Rosenblat et al., 2015), and multiple cognitive domains, including attention, orientation, executive function, and concentration (Harrison et al., 2016; Bennabi et al., 2019; Cumbo et al., 2019). The mechanism of cognitive improvement was thought to be related to the vortioxetine-induced elevation of serotonin levels and direct modulation of serotonin receptors (Jensen et al., 2014). However, further large-scale longitudinal real-world studies are needed to validate these findings, as potential bias and confounding may affect the results of the studies included in this meta-analysis.

In the subgroup analysis, the low-quality group showed higher pooled SMD than the high-quality group, indicating a potential risk of bias. Specifically, the low-quality group demonstrated SMDs of 3.06 and 2.98 in depression severity and cognitive function, respectively, and the high-quality group had SMDs of 1.38 and 0.76 for the same outcomes. Nonetheless, both groups showed a significant ameliorative effect of vortioxetine on depressive symptoms and cognitive impairment. In comparison, the group that combined vortioxetine with other antidepressants reported worse results. This could be explained by the presence of some degree of drug resistance in patients who initially received other antidepressants but eventually were co-administered vortioxetine (De Berardis et al., 2020; Yang et al., 2021; Mattingly et al., 2022). Unfortunately, further analysis for special populations such as children and adolescents, the elderly, and pregnant women is unavailable due to the lack of relevant data. Therefore, future studies investigating the effectiveness and safety of vortioxetine for the treatment of MDD should focus on these specific populations.

This study found that vortioxetine displayed high response (66.4%) and remission (58.0%) rates in real-world clinical practice, which were comparable with, or even higher than, the results reported in other studies (Thase et al., 2016; McCue et al., 2021; Christensen et al., 2022). In addition, a double-blind, RCT that focused on the elderly population revealed no significant difference in response and remission rates between vortioxetine and sertraline (Borhannejad et al., 2020).

Our study revealed that vortioxetine had good acceptability (8.4% dropout) and tolerability (3.5% dropout), with acceptable severe AEs. Its common AEs included gastrointestinal reactions like nausea, followed by headache, dizziness, and pruritus, which was consistent with the findings of other studies (Zheng et al., 2019; De Carlo et al., 2020; Verma and Kumar, 2021). Moreover, numerous studies have demonstrated that vortioxetine exhibits excellent tolerability not only in adults (Inoue et al., 2018; Gonda et al., 2019; Inoue et al., 2020; Christensen et al., 2023; McCue et al., 2021; Bose et al., 2022) but also in children and adolescents (Findling et al., 2022), the elderly population (Borhannejad et al., 2020; Danielak, 2021), and patients with comorbid underlying diseases such as cardiovascular disease or diabetes (Baldwin et al., 2022). Importantly, vortioxetine was not found to have more or more severe AEs than other second-generation antidepressants (Shao et al., 2022). Overall, our findings suggested that vortioxetine was safe and well tolerated.

Although our combined results were meaningful, the heterogeneity of almost all pooled outcomes was substantial. This heterogeneity persisted in all subgroups (all > 90.0%), and thus none of the variables fully explained the sources of heterogeneity. Meta-regression analyses collectively showed that variables such as geographic location and medication regimen (whether combined with other antidepressants) influenced the results, potentially contributing to heterogeneity. Additionally, the type of study design (retrospective or prospective) and the type of depression scale may be sources of bias leading to heterogeneity. In real-world studies, patients were only eligible to participate after a physician prescribes vortioxetine (Yang et al., 2021; Mattingly et al., 2022; Papalexi et al., 2022; Wang et al., 2022), and there were no objective criteria for when vortioxetine was prescribed. Various factors, such as the number of previous treatment episodes, previous treatments used, treatment compliance, disease severity, the combination of other antidepressants, and vortioxetine dose among the recruited patients, can impact prescription, ultimately leading to high heterogeneity of outcomes. However, these characteristics are necessary for a realistic simulation of the actual situation in clinical practice. Moreover, the self-rating scale used to assess depression severity included multiple scales, which contributed to the apparent heterogeneity because the content of each scale was not identical. Therefore, future research should focus on improving the standardization of self-evaluated scales of depression severity. Finally, differences in the basic characteristics of the patients included in the original studies, including age range, gender, and underlying disease status, may also contribute to heterogeneity, but this information was not available for further analysis.

This work’s strength lies in the fact that it is, to our knowledge, the first comprehensive assessment of the efficacy and safety of vortioxetine in clinical practice. We achieved this through an extensive literature search, utilizing rigorous inclusion and exclusion procedures, to obtain reliable results. However, certain limitations were observed in our findings. Firstly, there was substantial heterogeneity in outcomes across studies. Secondly, the pooled results did not provide evidence for comparison with other antidepressants in terms of effectiveness. Lastly, real-world studies lack controls and strict inclusion criteria, which may lead to a low quality of evidence grade for the results of this meta-analysis. Nevertheless, our findings supported the advantages of vortioxetine in improving depressive symptoms and cognitive function.

CONCLUSION

In conclusion, our findings suggested that vortioxetine had significant advantages in improving depressive symptoms, cognitive function, and functional disability in adults with MDD in clinical practice. Moreover, it was well tolerated and safe, making it a promising and effective treatment option for MDD patients with cognitive dysfunction. However, considering the limited quantity and quality of included articles, clinicians should interpret these findings with caution. More rigorous, large-scale, multicenter longitudinal studies of vortioxetine, as well as direct comparisons with other antidepressants in real-world settings, are still necessary for a more comprehensive understanding of its efficacy and safety.

Interest Statement

The authors report no conflicts of interest.

CRediT Authorship Contribution Statement

Zejun Li, Shouhuan Liu, and Tieqiao Liu contributed to study design, article search, review, and quality assessment. Yuzhu Hao, Manyun Li, Li He, and Xin Wang reviewed the study protocol and contributed to the data collection. Zejun Li, Qiuxia Wu, Pu Peng, Qian Yang, JingLi, and Yueheng Liu contributed to the drafting of the manuscript. Shubao Chen, Yunfei Wang, Qianjin Wang, and Huixue Xu designed the statistical analysis strategy and performed statistical analyses. All authors contributed to the critical revision of the paper and have agreed to be accountable for all aspects of the work.

Ethics Approval and Consent to Participate

Because we reanalyzed existing data and no identifiable information was shared, this study was not required to undergo independent ethical review.

Data Availability

All the data extracted from included studies are publicly available in PubMed, Embase, the Cochrane Central Register of Controlled Trials, Web of Science, PsycINFO, ProQuest Dissertations, CINAHL, and LiLACS.