Efficacy and safety of lumateperone for bipolar depression and schizophrenia: a systematic review and meta-analysis

Abstract

This study aimed to evaluate the efficacy and safety of lumateperone in treating bipolar disorder and schizophrenia. A comprehensive literature search was conducted across multiple databases and websites from inception to July 16, 2024, to identify both published and unpublished randomized controlled trials (RCTs). Meta-analyses were performed using random-effects or fixed-effects models depending on statistical heterogeneity. Relative risks (RRs) or standardized mean differences (SMDs) with 95% confidence intervals (CIs) were used to summarize the effects. Out of 931 records screened, 7 RCTs (four focusing on bipolar depression and 3 on schizophrenia) were eligible for inclusion. Lumateperone was efficacious in reducing depressive symptoms in bipolar depression (SMDs = −0.36, 95% CI: −.59 to −.13). In treating schizophrenia, lumateperone exhibited a lower combined SMD of -0.14 (95% CI: −.27 to 0, P = .051, I² = 49.6%), showing no significant difference from the placebo group, although the P-value approached significance. The lumateperone group showed significantly higher response rates compared with placebo in both bipolar depression (RRs = 1.27, 95% CI = 1.07 to 1.51) and schizophrenia (RRs = 1.44, 95% CI = 1.12 to 1.86). Common treatment-emergent adverse events included somnolence, dry mouth, dizziness, nausea, and headache (RRs = 1.30 to 3.29). Importantly, lumateperone did not significantly increase extrapyramidal symptoms (EPS, RRs = 1.46, 95% CI = .84 to 2.53). Lumateperone is effective in treating bipolar depression but does not significantly reduce symptom severity in schizophrenia. It has a favorable safety and tolerability profile. However, caution is warranted in interpreting these findings due to the limited number of studies included.

INTRODUCTION

Bipolar disorder (BD) is a recurrent and progressive psychiatric disorder characterized by episodes of mania, hypomania, and depression (McIntyre et al., 2020). With a global prevalence exceeding 1%, BD constitutes a significant public health challenge (Smedler et al., 2019; McIntyre et al., 2020). Depressive episodes in BD are more frequent and longer-lasting than manic or hypomanic phases, accounting for 72% of the illness duration. Consequently, the burden and costs associated with BD predominantly stem from its depressive phases, which are associated with poorer prognosis, increased disability, reduced quality of life, and heightened suicide rates. While some medications effectively manage manic symptoms in BD, treating the depressive phase remains challenging and limited (Yatham et al., 2018; Malhi et al., 2021). Currently, only 5 antipsychotics (APDs) have received Food and Drug Administration (FDA) approval for treating depressive episodes in bipolar I disorder (Goes, 2023). Treatment options for depression in bipolar II disorder are even more restricted, with lumateperone and quetiapine (including extended-release formulations) being the sole approved treatments (Goes, 2023). However, the clinical side effects of APDs often limit their utility and benefit for patients, and few medications offer therapeutic effects through novel mechanisms of action (Carvalho et al., 2020; Keramatian et al., 2023). Therefore, exploring new treatment approaches that offer high efficacy while minimizing adverse effects is critical.

Schizophrenia affects nearly 24 million people globally, presenting as a chronic and severe mental illness characterized by positive, negative, affective, and cognitive symptoms that significantly impair functioning (GBD 2019 Diseases and Injuries Collaborators, 2020; GBD 2021 Causes of Death Collaborators, 2024). Patients with schizophrenia have a life expectancy of approximately 15 years shorter than the general population and face heightened suicide risks throughout their lives (McCutcheon et al., 2020). The primary treatment for schizophrenia involves APDs (Kaul et al., 2024). While effective in managing positive symptoms such as hallucinations and delusions, current APDs often fail to fully alleviate negative symptoms, cognitive impairment, and social functioning. Moreover, all available APDs are associated with clinically significant side effects of varying severity, including motor impairments, prolactin abnormalities, metabolic disturbances, weight gain, and cardiovascular risks (Correll et al., 2015; Solmi et al., 2017, 2020; Huhn et al., 2019). These adverse effects exacerbate the already high rates of morbidity and mortality associated with schizophrenia, contributing to poor treatment tolerability and adherence (Correll et al., 2015, 2017, 2020; Chan et al., 2023). Therefore, there is a pressing need for safer, more effective APDs capable of addressing a broader spectrum of symptoms across multiple domains while improving tolerability.

Lumateperone is a novel approved APD by the FDA for treating adults with schizophrenia or bipolar I and II depression, either as monotherapy or in conjunction with lithium or valproate. Differing from other atypical APDs, lumateperone concurrently acts as a potent serotonin 5-HT_2A receptor antagonist, a dopamine D₂ receptor presynaptic partial agonist and postsynaptic antagonist, a D₁ receptor-dependent modulator of glutamate, and a serotonin reuptake inhibitor (Snyder et al., 2015; Davis and Correll, 2016; Titulaer et al., 2022). This pharmacologic profile suggests efficacy not only against positive symptoms but also negative symptoms, depressive symptoms, mixed features, and cognitive deficits (Jawad et al., 2022; Lobo et al., 2022). At a therapeutic daily dose of 42 mg, lumateperone exhibits dopamine D₂ receptor occupancy (D₂RO) of approximately 40% in schizophrenia patients, substantially lower than other atypical APDs (Farde et al., 1992; Vanover et al., 2019b). Given the positive correlation between D₂RO and risks of extrapyramidal side effects and hyperprolactinemia, lumateperone’s lower D₂RO likely contributes to its favorable tolerability profile (Kapur et al., 2000). These unique properties, alongside its lack of interaction with off-target receptors implicated in adverse effects of other APDs, position lumateperone as a promising option for managing schizophrenia and bipolar depression (Snyder et al., 2015; Davis and Correll, 2016).

Hence, this study aims to conduct a comprehensive meta-analysis and systematic review to assess the efficacy and safety of lumateperone in the treatment of bipolar depression and schizophrenia.

METHODS

This systematic review and meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (Page et al., 2021). The study protocol was registered with PROSPERO (CRD42023461824).

Search Strategy and Study Selection

We conducted a comprehensive search across 6 databases (PubMed, Embase, the Cochrane Central Register of Controlled Trials, Web of Science, Scopus, and PsycINFO), trial registries (International Clinical Trials Registry Platform, ClinicalTrials.gov, EU Clinical Trials Register), and pharmaceutical company website (Intra-Cellular Therapies, Inc.) to identify both published and unpublished RCTs from the inception to July 16, 2024. In addition, we manually searched for citations in relevant reviews. Our search strategy combined both Medical Subject Headings and free-text terms, focusing on 4 key terms: lumateperone, Caplyta, ITI-007, and ITI-722. Detailed search terms and the corresponding results for each database are provided in the Supplementary materials (Table S1).

We included all randomized placebo-controlled trials evaluating the efficacy and safety of lumateperone in individuals diagnosed with bipolar depression or schizophrenia according to standard diagnostic criteria (e.g., DSM-5 and ICD-11 or older versions). Even unpublished RCTs were considered for inclusion as long as relevant outcome data were available. Exclusion criteria were as follows: (1) open-label trials; (2) pooled analysis or post-hoc analysis; (3) animal or in vitro designs; (4) reviews, comments, or errata; (5) containing duplicated or overlapping data; and (6) no available data. We selected the report with the most comprehensive dataset for trials reported multiple times to avoid data duplication. Low fixed doses (14 or 28 mg/d) deemed subtherapeutic were excluded.

Two independent researchers (Hanrui Peng and Kewen Yan) conducted initial screening of titles and abstracts. Studies deemed potentially eligible were evaluated in full text. Consensus resolved disagreements, and a third reviewer (Zejun Li) arbitrated unresolved disagreements. Inter-rater reliability was assessed using a kappa score (0.90). Figure 1 illustrates the selection process and reasons for study exclusion.

Figure 1.

Flow diagram for identification of studies.

Data Extraction and Quality Assessment

Data were systemically extracted into a Microsoft Excel template, including basic details (title, first author, date of publication, and study IDs), demographic information (age range, mean age, proportion of males, race, and sample size per group), study characteristics (study design, phase, study period, and area of enrollment), clinical characteristics (baseline severity, diagnosis, and diagnostic manual), intervention details (dosage, duration of treatment, and whether monotherapy), outcome information (efficacy, safety, and acceptability), and any information related to bias. SDs of partial means were calculated from SEs (Cumpston et al., 2019). We combined the reported subgroups into a unified group in studies with multiple dosage arms.

The risk of bias (quality evaluation) of the included RCTs was assessed using the Cochrane Risk of Bias tool across 7 domains: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other potential bias (Z. Li et al., 2024b).

Outcomes

In this study, the primary outcomes were (1) the efficacy of lumateperone, defined by the mean overall change scores from baseline to completion of treatment on the Montgomery-Åsberg Depression Rating Scale (MADRS) or Positive and Negative Syndrome Scale (PANSS); and (2) the safety including the types of treatment-emergent adverse events (TEAEs) and the incidence of common TEAEs.

The secondary outcomes included the following: (1) changes in scores from baseline to end of treatment on the Clinical Global Impression Scale-Bipolar Version-Severity Scale (CGI-BP-S) depression subscale and overall bipolar illness subscale, Quality of Life Enjoyment and Satisfaction Questionnaire-Short Form among patients with bipolar depression; (2) improvement in scores on PANSS positive, negative, and general psychopathology subscales and the PANSS-derived prosocial factor from baseline to post-treatment completion among patients with schizophrenia; (3) response to treatment (defined as a decrease ≥50% in MADRS total score and ≥30% in PANSS total score from baseline) and remission (defined as a MADRS total score ≤12); (4) the acceptability (defined as the proportion of patients who discontinued treatment for any reason at the end of treatment); (5) the proportion of patients with ≥7% increase in body weight.

Data Synthesis and Analysis

We conducted a meta-analysis using Stata 17.0 software. We calculated summary SMDs (Cohen d) with corresponding 95% CIs for continuous outcomes and RRs with 95% CI for dichotomous outcomes. We employed the Galbraith radial plot and calculated the I² statistics to assess the heterogeneity among the included studies. A random-effects meta-analysis was conducted when there was considerable statistical heterogeneity (I² > 50% or P <.1); otherwise, a fixed-effects Mantel–Haenszel model was chosen. In addition, we conducted an influence analysis, a form of sensitivity analysis, to identify any significant influence of individual studies on the pooled estimates. All analyses were 2-tailed, with α = .05.

RESULTS

Search Results and Study Characteristics

Figure 1 shows the flow diagram of the screening process. A total of 931 records were systematically retrieved. Following screening titles and abstracts, a thorough full-text assessment was conducted on 68 reports deemed potentially eligible for inclusion. Primary reasons for exclusion included duplicated data from the same trial (n = 33), pooled or post-hoc analyses (n = 14), open-label trials (n = 6), unavailability of data (n = 4), and reviews (n = 3). Ultimately, our meta-analysis included 7 RCTs, corresponding to 4 publications (4 trials) (Lieberman et al., 2016; Correll et al., 2020; Calabrese et al., 2021; Suppes et al., 2023) and 4 nonformally unpublished records (3 trials, NCT02600494, NCT04285515, NCT02469155). Of the 7 RCTs, 4 focused on bipolar depression (NCT02600494, NCT04285515) (Calabrese et al., 2021; Suppes et al., 2023), while the remaining 3 investigated schizophrenia (NCT02469155) (Lieberman et al., 2016; Correll et al., 2020).

Our analysis involved 2480 participants: 1278 in the lumateperone group and 1202 in the placebo group. Among them, 1579 patients had bipolar depression, while 901 patients were diagnosed with schizophrenia. The mean ages ranged from 38.3 to 46 years, with an age range of 18 to 75 years. All studies were multicenter, with participants predominantly from Black, White, and Asian ethnic backgrounds. Among the 3 schizophrenia trials, 2 were phase 3 studies (NCT02469155) (Correll et al., 2020) and 2 included another active-controlled group (NCT02469155) (Lieberman et al., 2016). All 3 trials were conducted in inpatient settings across the United States, with predominantly male participants ranging from 71.5% to 86.7%. Enrolled patients with schizophrenia were in the acute exacerbation phase and received monotherapy for either 4 or 6 weeks. One trial included lumateperone treatment groups with 42 mg and 84 mg (Lieberman et al., 2016). Regarding the 4 bipolar depression trials, 1 involved adjunctive treatment (Suppes et al., 2023), 3 were conducted across multiple countries, and the proportion of males ranged from 36.5% to 47.3%. All 4 trials were phase 3 studies with a 6-week treatment duration. Most included patients with bipolar depression experienced depressive episodes, with 1 study specifically focusing on patients diagnosed with bipolar depression with mixed features (NCT04285515). Table 1 provides a concise summary of the key characteristics of each trial.

Table 1.

Characteristics of the included studies

Study	Diagnosis	Study design (period)	Diagnostic criteria	Drug administrationa	Mean age (SD), range	Proportion of male (no, %)	Race	Recruitment area	ITT set, safety set (T/C)	Baseline severity of MADRS or PANSS, mean (SD)
Calabrase 2021 NCT03249376	BD Ⅰ or Ⅱ, current MDE	Phase 3, multi-center, randomized, double-blind, placebo-controlled outpatient study (Nov 2017−Mar 2019)	DSM-5	42 mg/d 6 wk monotherapy	T: 46 (14.1) C: 44 (12.9) 18-75	T: 89 (47.3%) C: 69 (36.5%)	White, Black, Asian, and other	54 clinical sites in 6 countries: United States, Bulgaria, Colombia, the Russian Federation, Serbia, and Ukraine	188/188 188/189	T: 30.8 (4.92) C: 30.2 (4.65)
Suppes 2023 NCT02600507b	BD Ⅰ or Ⅱ, current MDE	Phase 3, multi-center, randomized, double-blind, placebo-controlled study (Mar 2016−Jul 2020)	DSM-5	28 or 42 mg/d 6 wk add-on	T42: 44.7 (12.6) C: 45.1 (12.9) 18-70	T42: 68 (38.4%) C: 77 (44.0%)	White, Black, Asian, and other	89 clinical sites globally: United States, Ukraine, Bulgaria, Serbia, Russia	174/174 177/175	T42: 32.3 (4.96) C: 32.2 (5.23)
NCT02600494b	BD Ⅰ or Ⅱ, current MDE	Phase 3, multi-center, randomized, double-blind, placebo-controlled study (Dec 2015−Jul 2019)	NA	28 or 42 mg/d 6 wk monotherapy	NA 18-75	NA	NA	NA	166/177 184/185	NA
NCT04285515	Bipolar depression with mixed features	Phase 3, multi-center, randomized, double-blind, placebo-controlled study (Feb 2020−Nov 2022)	DSM-5	42 mg/d 6 wk monotherapy	T: 42.7 (14.58) C: 43.4 (13.84) 18-75	T: 92 (37.9%) C: 95 (38.8%)	NA	United States, Bulgaria, Russia, Serbia, Ukraine	211/219 240/241	> 24
Lieberman 2016 NCT01499563c	Acute exacerbation of schizophrenia	Phase 2, multi-center, randomized, double-blind, placebo- and active-controlled inpatient trial (Dec 2011−Nov 2013)	SCID-CT	42 mg/d 84 mg/d 4 wk monotherapy	T42: 38.3 (10.0) T84: 41.1 (8.9) C: 40.5 (9.8) 18-55	T42: 66 (78.6%) T84: 72 (86.7%) C: 65 (76.5%)	Black, White, Asian, and other	8 sites in United States	76/80/80 84/83/85	T42: 88.1 (11.0) T84: 84.6 (11.6) C: 86.3 (13.1)
Correll 2020 NCT02282761b	Acute exacerbation of schizophrenia	Phase 3, multi-center, randomized, double-blind, placebo-controlled inpatient trial (Nov 2014−Jul 2015)	DSM-5 DSM-IV-TR	28 or 42 mg/d 4 wk monotherapy	T42: 42.4 (10.3) C: 41.4 (10.3) 18-60	T42: 110 (73.3%) C: 123 (82.6%)	Black, White, and other	12 clinical sites in United States	148/141 150/149	T42: 90.1 (9.5) C: 90.1 (11.1)
CDER, FDA 2018 NCT02469155b,c	Acute exacerbation of schizophrenia	Phase 3, multi-center, randomized, double-blind, placebo- and active-controlled inpatient trial (Jun 2015−Aug 2016)	DSM-5 SCID-CT	14 or 42 mg/d 6 wk monotherapy	T42d: 41.9 (9.9) C: 43.5 (9.9) 18-60	T42: 123 (71.5%) C: 132 (74.2%)	Black, White, Asian, and other	13 sites in United States	162/169 172/178	T42: 90.4 (10.2) C: 90.2 (9.8)

^a42 mg/d of active moiety lumateperone is equivalent to 60 mg/d of lumateperone tosylate (also known as ITI-007). ^bLow fixed-doses (14 or 28 mg/d) that could be deemed as subtherapeutic were not included in the meta-analysis. ^cThe active controlled group was administered 4 mg of risperidone. ^dBD, bipolar disorder; C, control group; ITT, intent to treat; MADRS, Montgomery-Åsberg Depression Rating Scale; MDE, major depressive episodes; NA, not available; PANSS, Positive and Negative Syndrome Scale; SCID-CT, Structured Clinical Interview for DSM Disorders-Clinical Trials Version; T, treatment group.

The final safety assessments were usually performed approximately 2 weeks after the last dose of study medication.

Efficacy on Bipolar Depression and Schizophrenia

The meta-analysis revealed a significant advantage of lumateperone over placebo among patients with bipolar depression, leading to greater reductions from baseline on MADRS score (SMDs = −0.36, 95% CI = −.59 to −.13, P = .002, I² = 80.0%) (Figure 2A). We found no significant difference between the lumateperone and placebo groups in terms of efficacy for treating schizophrenia, although the P value approached significance (SMDs = –0.14, 95% CI = –.27 to 0, P = .051, I² = 49.6%) (Figure 2B). In this excluded study, lumateperone 14 mg and 42 mg arms did not separate from placebo, and the placebo response was more significant than in the other 2 studies.

Figure 2.

Forest plot of meta-analysis on lumateperone for bipolar depression and schizophrenia. (A) Improvement in MADRS scores. (B) Improvement in total PANSS scores.

Secondary Outcomes

The secondary outcomes are presented in Table 2. For the treatment of bipolar depression, the lumateperone treatment group demonstrated significant improvements compared with the placebo group across 3 CGI-BP-S subscales: CGI-BP-S depression subscale (SMDs = −0.33, 95% CI = −.47 to −.18, P < .0001, I² = 20.31%), CGI-BP-S overall bipolar illness subscale (SMDs = −0.30, 95% CI = −.45 to −.16, P < .0001, I² = 0%), and Quality of Life Enjoyment and Satisfaction Questionnaire-Short Form (SMDs = 0.22, 95% CI = .07 to .36, P = .004, I² = 0%). Additionally, the lumateperone group exhibited a higher response rate compared with placebo (RRs = 1.27, 95% CI = 1.07 to 1.51, P = .006, I² = 18.68%). However, there was no significant difference in remission rates (RRs = 1.06, 95% CI = .86 to 1.30, P = .580, I² = 37.76%) and rates of discontinuation due to any reason (RRs = 1.30, 95% CI = .98 to 1.73, P = .068, I² = 38.06%) between the 2 groups (Table 2).

Table 2.

Lumateperone for bipolar disorder and schizophrenia: the secondary outcomes

Variables	No. of studies	No. of participants		SMDs/RRs (95% CI)	I2 (%)	P
		Lumateperone	Placebo
Bipolar disorder
CGI-BP-S-Da	2	362	362	-0.33 (-0.47, -0.18)	20.31	<.0001
CGI-BP-S-Oa	2	362	362	-0.30 (-0.45, -0.16)	0	<.0001
Q-LES-Q-SFa	2	362	362	0.22 (0.07, 0.36)	0	.004
Response	2	362	362	1.27 (1.07, 1.51)	18.68	.006
Remission	2	362	362	1.06 (0.86, 1.30)	37.76	.580
Acceptability	3	605	605	1.30 (0.98, 1.73)	38.06	.068
Schizophrenia
PANSSN positive symptom subscaleb	2	304	221	-0.34 (-0.52, -0.17)	0	.0001
PANSSN negative symptom subscaleb	2	304	221	-0.11 (-0.29, -0.06)	7.03	.199
PANSSNc general psychopathology subscaleb	2	304	221	-0.27 (-0.45, -0.10)	0	.003
PANSS-derived prosocial factorb	2	304	221	-0.29 (-0.46, -0.11)	0	.001
Response	2	304	221	1.44 (1.12, 1.86)	0	.005
Acceptability	3	491	408	0.82 (0.50, 1.33)	70.15	.414

^aThe raw data were based on a mixed-effects model for repeated measures (MMRM). ^bThe raw data were based on analysis of covariance (ANCOVA) with last observation carried forward (LOCF). ^cCGI-BP-S, Clinical Global Impression Scale-Bipolar Version-Severity Scale; CGI-BP-S-D, CGI-BP-S depression subscale; CGI-BP-S-O, CGI-BP-S overall bipolar illness subscale; PANSS, Positive and Negative Syndrome Scale; Q-LES-Q-SF, Quality of Life Enjoyment and Satisfaction Questionnaire-Short Form.

For the treatment of schizophrenia, the lumateperone-treated group showed greater improvements in PANSS positive symptom subscale scores (SMDs = −0.34, 95% CI = −.52 to −.17, P = .0001, I² = 0%) and PANSS general psychopathology subscale scores (SMDs = −0.27, 95% CI = −.45 to −.10, P = .003, I² = 0%). However, there was no significant difference in PANSS negative symptom subscale scores between the lumateperone and placebo groups (SMDs = −0.11, 95% CI = −.29 to −.06, P = .199, I² = 7.03%). Improvement in the PANSS-derived prosocial factor scale scores was superior in the lumateperone group compared with placebo (SMDs = −0.29, 95% CI = −.46 to −.11, P = .001, I² = 0%). The lumateperone group exhibited a higher response rate compared with placebo (RRs = 1.44, 95% CI = 1.12 to 1.86, P = .005, I² = 0%). No significant differences were observed in acceptability (RRs = 0.82, 95% CI = .50 to 1.33, P = .414, I² = 70.15%) between the 2 groups. Additionally, the proportion of participants with ≥7% increase in body weight was higher in the lumateperone group than in the placebo group (RRs = 1.86, 95% CI = 1.00 to 3.46, P = .048, I² = 0%) (Table 2).

Safety of Lumateperone

Safety outcomes were reported based on pooled data from studies involving both bipolar depression and schizophrenia. Table 3 illustrates the incidence of common adverse events comparing lumateperone with placebo. The results revealed that lumateperone administration was associated with an increased risk of drug-related TEAEs (RRs = 1.32, 95% CI = 1.08 to 1.62, P = .007, I² = 0%). Specifically, somnolence (RRs = 3.29, 95% CI = 2.42 to 4.47, P < .0001, I² = 42.53%), dry mouth (RRs = 2.76, 95% CI = 1.35 to 5.65, P = .006, I² = 39.65%), dizziness (RRs = 2.68, 95% CI = 1.43 to 5.38, P = .005, I² = 57.11%), nausea (RRs = 2.32, 95% CI = 1.59 to 3.38, P < .0001, I² = 0%), and headache (RRs = 1.30, 95% CI = 1.04 to 1.62, P = .019, I² = 0%) were significantly more prevalent in the lumateperone group compared with the placebo group. There was no significant difference in the rates of extrapyramidal symptoms (EPS; RRs = 1.46, 95% CI = .84 to 2.53, P = .176, I² = 0%) between the 2 groups.

Table 3.

Incidence of common adverse events with lumateperone vs placebo

Adverse events	No. of studies	Reactions/total		RRs (95% CI)a	I2 (%)	P
		Lumateperone	Placebo
Affected subjects by TEAEs	4	289/755	190/754	1.53 (1.11, 2.12)	77.44	.01
TEAEs	2	191/365	173/364	1.10 (0.95, 1.27)	0	.193
Drug-related TEAEs	2	142/365	107/364	1.32 (1.08, 1.62)	0	.007
Somnolence	6	175/1094	46/1017	3.29 (2.42, 4.47)	42.53	<.0001
Dry mouth	3	31/489	9/412	2.76 (1.35, 5.65)	39.65	.006
Dizziness	6	86/1094	29/1017	2.68 (1.43, 5.38)	57.11	.005
Nausea	6	89/1094	37/1017	2.32 (1.59, 3.38)	0	<.0001
Headache	5	156/927	121/932	1.30 (1.04, 1.62)	0	.019
Proportion with ≥7% increase in body weight	4	37/646	14/561	1.86 (1.00, 3.46)	0	.048
Extrapyramidal symptoms	4	29/687	20/691	1.46 (0.84, 2.53)	0	.176

^aTEAEs, treatment-emergent adverse events.

Quality Assessment and Sensitivity Analysis

The quality assessment (risk of bias evaluation) for the included RCTs is detailed in Table S2. Six trials were rated as having a low risk of bias, while 1 was classified as high risk due to “other potential bias.” Sensitivity analysis indicated robustness in the efficacy result for bipolar depression (reduction in MADRS scores). However, the efficacy result for schizophrenia (reduction in PANSS scores) was influenced by the study NCT02469155 (Figure S1). Additionally, Galbraith plots were used to visually represent the heterogeneity of pooled results (Figure S2).

DISCUSSIONS

This systematic review and meta-analysis include 7 RCTs, providing a comprehensive synthesis of current data on the efficacy and safety of lumateperone in patients with bipolar depression and schizophrenia. The results revealed that lumateperone effectively improved MADRS scores in bipolar depression but did not significantly reduce total PANSS scores in schizophrenia. The lumateperone group showed significantly higher response rates compared with placebo in both bipolar depression and schizophrenia. Lumateperone showed good acceptability, with common TEAEs including somnolence, dry mouth, dizziness, nausea, and headache. Notably, no significant serious safety concerns were reported, and administration of lumateperone did not lead to a significant increase in EPS. Moreover, lumateperone showed improvements in quality of life in patients with bipolar depression and enhanced social function in those with schizophrenia. These findings offer an important reference for the treatment of adult patients with bipolar depression and schizophrenia.

Efficacy on Bipolar Depression

A network meta-analysis (Yildiz et al., 2023) involving 68 medications for bipolar depression provided moderate evidence (category I) supporting lumateperone’s efficacy in reducing depressive symptoms. Calabrese et al. (Calabrese et al., 2021) reported a rapid and substantial improvement in MADRS scores as early as week 1 following lumateperone administration, with continued improvement throughout the 6-week study. The effect size for improvements in MADRS scores in our study (SMDs = −0.36) was comparable with other APDs approved for bipolar depression, such as olanzapine/fluoxetine combination (SMDs = −0.41) and quetiapine (SMDs = −0.35) (Yildiz et al., 2023). Another network meta-analysis (S. Li et al., 2024a) evaluating 5 FDA-approved atypical APDs for bipolar depression ranked lumateperone third in MADRS response rates, demonstrating non-inferiority to olanzapine and cariprazine. In this study, a greater proportion of patients in the lumateperone group were MADRS responders compared with placebo, although MADRS remission rates were similar across groups. This could be attributed to the study participants’ moderate to severe depression at baseline; therefore, widespread MADRS remission within the 6-week treatment period was not expected (Leelahanaj et al., 2013; Solomon et al., 2013).

The study by Calabrese et al. (Calabrese et al., 2021) demonstrated notable efficacy of lumateperone in both bipolar I and bipolar II disorders through subgroup analysis despite the relatively small sample size of participants with bipolar II disorder (38 per treatment group). APDs have been investigated as adjunctive treatment to lithium or valproate for depressive episodes associated with bipolar I disorder; currently, lumateperone (for both bipolar I and II disorders) and lurasidone (for only bipolar I disorder) are the only FDA-approved APDs for this adjunctive use (Loebel et al., 2014; Carvalho et al., 2020). Therefore, lumateperone became the first and only drug with such broad indications. In addition, lumateperone was effective in alleviating depressive symptoms in patients with mixed features (NCT04285515).

Efficacy on Schizophrenia

This study revealed that lumateperone did not significantly reduce total PANSS scores in schizophrenia. However, the sensitivity analysis indicated that lumateperone was effective in treating schizophrenia symptoms after excluding the study NCT02469155. Study NCT02469155 did not demonstrate efficacy for lumateperone 42 mg in treating schizophrenia, whereas the risperidone 4 mg arm showed separation from placebo. Both the applicant and statistical reviewers conducted multiple post-hoc sensitivity analyses indicating that all active treatments, including risperidone, very likely had no statistically significant difference from placebo. Consequently, the study would lack assay sensitivity and should be considered a “failed” study. Regardless of whether risperidone’s results are considered negative or nominally positive, the absence of an effect for lumateperone is concerning. However, these results should be interpreted cautiously due to the limited number of studies. Larger, more rigorous, long-term RCTs are needed to further explore lumateperone’s efficacy in schizophrenia and to compare it directly with other APDs.

Lumateperone improves not only positive psychotic symptoms but also general psychopathology and psychosocial functioning. Significant improvements in psychosocial functioning were indicated by enhancements in the PANSS-derived prosocial factor, which includes PANSS items on active social avoidance, passive social withdrawal, and emotional withdrawal (Harvey et al., 2017). Receptor occupancy studies have shown that lumateperone acts as an antagonist at the serotonin 5-HT_2A receptor, with a K_i of 0.54 nM, which is 60-fold greater than its affinity for D₂ receptors (K_i = 32 nM) (Snyder et al., 2015). At low doses, lumateperone selectively antagonizes the 5-HT_2A receptor, promoting sleep and reducing hostility and aggression. In contrast, at higher doses, it may improve depressive and psychotic symptoms, particularly negative symptoms (Vyas et al., 2020). However, our study was not adequately powered to detect the efficacy of lumateperone in improving negative symptoms, resulting in negative findings in this regard. This limitation may be attributed to the predominance of positive symptoms rather than negative symptoms at baseline in the included studies (Lieberman et al., 2016; Correll et al., 2020). However, 1 study featuring a specific subgroup analysis of patients with prominent negative symptoms at baseline found that lumateperone reduced severity on the PANSS negative symptom subscale (Lieberman et al., 2016), suggesting the potential of lumateperone in improving negative symptoms. To expand upon these findings, further research is warranted to investigate the efficacy of lumateperone on negative symptoms in stable patients with predominant or persistent negative symptoms.

Safety

The acceptability of lumateperone in patients with bipolar depression and schizophrenia was favorable. Lumateperone exhibited a favorable side effect profile in contrast to currently approved APDs, which are known to have safety concerns, including EPS, weight gain, alterations in glucose and lipid metabolism, hyperprolactinemia, and cardiovascular and cerebrovascular risks (Solmi et al., 2017; Rognoni et al., 2021; Kong et al., 2024). Data from short-term studies indicated that lumateperone did not significantly increase EPS and showed no evidence of inducing hyperprolactinemia (Calabrese et al., 2021; Suppes et al., 2023). The drug’s ability to effectively treat schizophrenia with low EPS and minimal hyperprolactinemia may be attributed to its pharmacological characteristics, including low postsynaptic D₂RO (Correll et al., 2020). Although our study noted a higher proportion of patients experiencing ≥7% weight gain in the lumateperone group compared with the placebo group, there was a trend toward narrowing this difference. Previous meta-analyses have confirmed that lumateperone leads to less weight gain than other APDs (Yildiz et al., 2023; S. Li et al., 2024a). Additionally, several original studies (Lieberman et al., 2016; Correll et al., 2020; Calabrese et al., 2021; Suppes et al., 2023) indicated that the lumateperone treatment group did not experience significant increases in triglyceride levels, cholesterol, insulin, or glucose, presenting a cardiometabolic profile similar to placebo and suggesting no increased risk of metabolic syndrome. The favorable safety profile of lumateperone may stem from its unique mechanism of action, characterized by minimal binding to histaminergic or muscarinic receptors, which have been associated with the cardiometabolic effects and other tolerability issues observed with existing APDs (Correll, 2010; Snyder et al., 2015; Solmi et al., 2017).

Limitations

Several limitations should be considered in this study. First, the relatively small number and sample size of included studies may limit the generalizability of findings. Second, data limitations prevented subgroup and meta-regression analyses, which could have offered deeper insights into between-study variability and its impact on effect size estimates. Third, variations in dosages and treatment durations among the included RCTs may introduce potential biases. Fourth, since positive findings are more likely to be published, there is a risk of overestimating lumateperone’s efficacy and safety. However, this meta-analysis included unpublished RCTs to mitigate publication bias. Fifth, the controlled conditions of clinical trials differ from everyday clinical practice, particularly in terms of patient diversity and treatment adherence, which limits the applicability of our findings. Further studies are needed to evaluate lumateperone’s efficacy and safety in real-world settings. Lastly, 4 to 6 weeks of treatment may not elucidate the long-term treatment’s safety profile and maintenance therapy’s efficacy. It is worth noting that lumateperone demonstrated a favorable safety and tolerability profile in a 1-year study involving patients with stable schizophrenia(Kimberly Vanover et al., 2019a).

CONCLUSION

Our meta-analysis indicated that lumateperone is effective in treating bipolar depression but does not significantly reduce symptom severity in schizophrenia. It has a favorable safety and tolerability profile. Lumateperone showed a low risk of EPS, hyperprolactinemia, and metabolic changes. However, caution is advised when interpreting these findings due to the limited number of studies included. The long-term efficacy (especially for schizophrenia), safety, and tolerability of lumateperone should be further comprehensively evaluated through large-scale and longer-term RCTs. Additionally, direct comparisons with other APDs are necessary to understand lumateperone’s clinical applications better.

Author Contributions

Hanrui Peng (Methodology [Equal], Writing—original draft [Equal]), Kewen Yan (Methodology [Equal], Writing—original draft [Equal]), Shouhuan Liu (Data curation [Equal], Formal analysis [Equal]), Xin Li (Data curation [Equal], Formal analysis [Equal]), Xin Wang (Formal analysis [Equal]), Pu Peng (Formal analysis [Equal]), Xueyi Li (Formal analysis [Equal]), Min Wu (Formal analysis [Equal]), Huixue Xu (Formal analysis [Equal]), Qiuxia Wu (Formal analysis [Equal]), Tieqiao Liu (Conceptualization [Equal], Project administration [Equal], Writing—review and editing [Equal]), and Zejun Li (Conceptualization [Equal], Project administration [Equal], Writing—review and editing [Equal]).

CRediT Authorship Contribution Statement

Zejun Li conceived the study and corrected the manuscript. Hanrui Peng and Kewen Yan assessed the quality of the studies and drafted the manuscript. Shouhuan Liu, Xin Li, Xin Wang, Pu Peng, and Xueyi Li analyzed the data. Hanrui Peng, Kewen Yan, Min Wu, Huixue Xu, Qiuxia Wu, and Tieqiao Liu contributed to the literature screening and data collection. All authors have agreed to be accountable for all aspects of the work.

Funding

None.

Conflict of Interest

The authors report no conflicts of interest.

Data Availability

All data produced and analyzed are available in this published article.