Major depressive disorder (MDD) is a debilitating psychiatric disorder [1] that poses a significant health problem worldwide [2]. Approximately half of patients with MDD do not respond to conventional antidepressant therapy (ADT) [3,4]. They experience significant functional impairment and have increased risks for medical and psychiatric comorbidities. Recommendations for patients with inadequate response to ADT include changing the antidepressant, or adding a second antidepressant or another medication as adjunctive treatment [5,6]. The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study showed that only one-third of patients reached remission after first-line ADT and the likelihood of remission declined with subsequent lines of treatment [3]. Hence, early intervention with appropriate treatment strategies is crucial.

There is growing evidence on the use of adjunctive treatment as a strategy for addressing inadequate response to ADT [7-13]. Research supports the adjunctive use of antidepressants with non-monoaminergic activity or atypical antipsychotics for patients not responding adequately to ADT [7-13]. Recently, esketamine, a substance with glutamatergic neuromodulatory properties, has exhibited positive outcomes in reducing depressive symptoms in real-world scenarios when used as an adjunct treatment [7,8]. Additionally, the adjunctive use aripiprazole, an atypical antipsychotic, has been demonstrated to be an effective intervention in patients with MDD who had inadequate responses to ADT [9]. Brexpiprazole is an atypical antipsychotic approved by the Food and Drug Administration (FDA) for treating schizophrenia and as an adjunctive treatment for MDD with inadequate response to ADT alone [14]. It is a serotonin–dopamine activity modulator, acting as a partial agonist at the serotonin 5-HT1A and dopamine D2 receptors, and as an antagonist at the serotonin 5-HT2A and noradrenaline α1B/α2C receptors [15]. Its efficacy and safety over six weeks have been demonstrated in adult patients with MDD who had inadequate response to ADT [10-13].

Brexpiprazole was approved in Singapore in 2019 as an adjunct to ADT in patients with MDD. In a study of Asian patients with MDD, over one third discontinued their ADT due to inadequate response [16]. Yet, real-world data on adjunctive brexpiprazole treatment in these patients are lacking. This non-interventional study aimed to assess the effectiveness and safety of adjunctive brexpiprazole in Asian patients with MDD and inadequate response to ADT in a real-life clinical setting in Singapore.

This prospective, 3-month observational study was conducted at two sites between September 2020 and October 2021 in compliance with the Declaration of Helsinki, Good Clinical Practice Guidelines, and applicable local ethical and regulatory requirements. The independent ethics committee approved the study (NHG DSRB Ref: 2020/00328; PIEC Ref: PIEC/2020/009) and all patients provided written informed consent prior to study commencement.

Eligible patients were aged ≥ 21 years, clinically diagnosed with an active episode of MDD, had inadequate response to ≥ 1 antidepressant for the current episode, and were treated with an antidepressant for ≥ 6 weeks. They had initiated brexpiprazole treatment at the physician’s discretion, independently of study recruitment (Supplementary Material 1; available online).

The following assessments were performed at baseline, Weeks 2, 4, and 12 as part of routine practice: Patient Health Questionnaire-9 (PHQ-9) (for depression severity) [17], Clinical Global Impression-Severity (CGI-S) (for mental illness severity), Clinical Global Impression-Improvement (CGI-I) (for improvement or worsening in illness severity compared with baseline) [18], Sheehan Disability Scale (SDS) (for functional disability) [19], and Generalized Anxiety Disorder scale (GAD-7) (for anxiety severity) [20]. All adverse events (AEs) were recorded (Supplementary Material 2; available online).

Based on a previous study [21], 40 patients were estimated to yield 90% power in detecting difference between baseline and 3 months using one sample t test at 2.5% significance level. Assuming a 30% withdrawal rate, 60 patients were targeted for enrolment. The analysis sets are summarized in Supplementary Figure 1 (available online). The co-primary endpoints were PHQ-9 and CGI-S scores at Week 12. Other endpoints included CGI-I, SDS, and GAD-7 scores, and the incidence of AEs. PHQ-9 and CGI-S scores at Week 12 were compared with baseline using paired t test with Bonferroni adjustment for multiple testing. The linear mixed model (LMM) (prespecified for sample size < 30) was used to compare PHQ-9 and CGI-S scores at each visit with baseline, and was repeated for SDS and GAD-7 scores. Descriptive statistics were used to summarize demographic data, CGI-I scores, responders (50% reduction in PHQ-9 or CGI-S score from baseline), remitters (PHQ-9 score ≤ 4 or CGI-S score ≤ 2), and AEs. No missing data were imputed. Analyses were performed using Statistical Analysis Software (SAS)^Ⓡ version 9.4 (SAS Institute Inc.). p values < 0.05 were considered statistically significant.

Twenty patients were enrolled and 16 completed the study (Supplementary Fig. 1; available online). The median age was 27.0 (range 21.0−63.0) years (Supplementary Table 1; available online) and 70.6% (12/17) were female. Most (64.7%; 11/17) reported having their current episode for ≥ 6 months. Commonly prescribed antidepressants were fluoxetine (29.4%; 5/17), escitalopram (17.7%; 3/17), and vortioxetine (17.7%; 3/17). The majority (88.2%; 15/17) continued with brexpiprazole during the study period. Mean treatment compliance was 94.2% (standard deviation [SD] 5.1). The mean average daily dose was 1.0 (SD 0.5) mg and median treatment duration was 85.0 (range 17.0−107.0) days.

In the intention-to-treat set, the mean PHQ-9 scores improved from baseline at all visits (Fig. 1A), with mean differences of −4.8 (95% confidence interval [CI], −10.0 to 0.5; Bonferroni-adjusted p = 0.0800 from paired t test) and −4.7 (95% CI, −8.1 to −1.3; p = 0.0075 from the LMM) (Fig. 1A), corresponding to a one-grade improvement (from moderately severe to moderate) at Week 12. The mean CGI-S score improved from baseline at all visits (Fig. 1B), with mean differences of −1.3 (95% CI, −2.2 to −0.4; Bonferroni adjusted p = 0.0050 from paired t test) and −1.3 (95% CI, −1.9 to −0.6; p = 0.0002 from the LMM) (Fig. 1B), corresponding to a one-grade improvement (from moderately ill to mildly ill) at Week 12. The mean CGI-I scores improved from baseline at all visits (2.3 [SD 0.9] at Week 2, 1.9 [SD 1.0] at Week 4, and 2.3 [SD 1.3] at Week 12). One third (31.3%; 5/16) achieved response and 25.0% (4/16) achieved remission based on PHQ-9 scores at Week 12. Similar results were obtained using CGI-S scores (38% for both). The mean total SDS scores and GAD-7 scores decreased from baseline at all visits, with a mean difference of −8.5 (95% CI, −13.0 to −4.0; p < 0.001) (Supplementary Fig. 2; available online) and −6.2 (95% CI, −9.9 to −2.6; p < 0.01), respectively at Week 12 (Supplementary Fig. 3; available online). Similar results were noted in the per protocol analysis set (data not shown).

Around half reported ≥ 1 AE (55.0%; 11/20) and treatment-related AEs (50.0%; 10/20) (Table 1). Two patients (10.0%; 2/20) withdrew brexpiprazole due to AEs and one patient (5.0%; 1/20) reported a serious AE. The most common AE was weight gain (25.0%; 5/20). The mean weight increased from baseline to Week 12, with a mean difference of 2.35 kg (95% CI, 1.1 to 3.6; p = 0.0003). No death or severe AE occurred.

To our knowledge, this is the first study assessing the effects of adjunctive brexpiprazole in Asian patients with MDD who had inadequate response to ADT in a real-life setting in Singapore. Treatment with adjunctive brexpiprazole for up to 3 months resulted in clinically meaningful improvements in depression severity, mental illness severity, overall functioning, and anxiety severity, with a favorable safety and tolerability profile.

A one-grade improvement in both depression and mental illness severity (as assessed by PHQ-9 and CGI-S, respectively) was noted at the end of study. PHQ-9 was chosen for its validity, responsiveness, and reliability in assessing depression severity in an Asian setting [22], and its strong correlation to Montgomery–Åsberg Depression Rating Scale (MADRS) [23]. The observed improvements are consistent with results from Phase 3 trials that demonstrated improvements in MADRS and CGI-S scores in non-Asian patients with MDD treated with adjunctive brexpiprazole [10-13]. In the present study, patients achieved 31−38% response and 25−38% remission rates, as reported in previous trials [10-13]. The improvements in SDS and GAD-7 scores revealed the benefits of adjunctive brexpiprazole in improving overall functioning and reducing anxiety symptoms, as observed in brexpiprazole studies [10-13,24]. Collectively, these findings are encouraging considering patients had inadequate response to ≥ 1 antidepressant before initiating brexpiprazole, and the study was conducted during the COVID-19 pandemic where patients’ mental health could have been negatively impacted.

Treatment with brexpiprazole was well-tolerated for up to three months with a low rate of discontinuation due to AEs (2/20 patients). The AE profile is largely consistent with earlier studies [10-13,25,26]. No new or unexpected study concerns were noted.

The pandemic posed considerable challenges to patient recruitment [27]. Clinic visits were reduced and doctors were redeployed from their usual practices to provide care for patients with COVID-19, resulting in a smaller sample size than was planned. The LMM (prespecified for sample size < 30) was therefore used to increase the power of detecting difference between baseline and post-treatment measurements. The relatively small sample size can affect the power of the study to detect true differences and may not fully represent the broader population, therefore it is important to interpret data cautiously and ensure effect sizes are taken into account when determining treat-ment effect [28]. Our single-arm study lacked a control group, which makes it challenging to differentiate the effects of brexpiprazole from potential placebo effects or other external factors, and may introduce bias. It was conducted in one country, which limits generalizability of the results to the wider Asian population. Although long-term effects of adjunctive brexpiprazole have been previously evaluated [26], long-term data in an Asian population with MDD are still lacking. Despite these limitations, our results are consistent with earlier brexpiprazole studies [10-13,25], and can provide valuable insights into the real-world use of adjunctive brexpiprazole in Asian patients with MDD.

Our findings support the results from earlier brexpiprazole trials and provide a real-world snapshot of the use of adjunctive brexpiprazole in Asian patients with MDD in Singapore.

Dr Rohini Bose is an employee of Lundbeck Singapore Pte Ltd. Dr Keira Joann Herr was an employee of Lundbeck Singapore Pte Ltd during the conduct of this study. Dr Ken Eng Khean Ung has received honoraria from Lundbeck for clinical time spent in assessment of cases. The other authors declare no conflict of interest.

Conception and design of the study: Yee Ming Mok, Pei Lin Lynnette Tan, Keira Joann Herr, Ken Eng Khean Ung. Acquisition, analysis, or interpretation of data: all authors. Drafting or revising the manuscript: all authors. Approval of the final manuscript: all authors.