The effects of lurasidone on five symptom domains and individual items in schizophrenia: Post-hoc analysis from Phase 3 study

Abstract

The aim of this analysis was to examine the impact of treatment with lurasidone on a broad range of symptoms during an acute episode of schizophrenia. In patients with an acute exacerbation of schizophrenia treated with lurasidone in a 6-week, double-blind, placebo-controlled trial, a mixed-effect model for repeated measures was used to analyze weekly change in both the five Marder Factors and the 30 individual items of the Positive and Negative Syndrome Scale (PANSS). Treatment with lurasidone was associated with significant endpoint improvement on all five Marder PANSS factors with effect sizes ranging from 0.22 to 0.44. Significant early (week 2), sustained improvement was observed on four of the five Marder Factors, with significant improvement occurring by week 4 on the Marder Negative Symptom factor. In addition to significant improvement on five out of eight Marder Positive Symptom items, the following non-Positive PANSS items were notable for being especially responsive to lurasidone treatment in terms of significant early improvement and endpoint effect sizes >0.3: anxiety, tension, preoccupation, active social avoidance, and conceptual disorganization. The results of this non-USA study confirm results from previous studies indicating that lurasidone is an effective antipsychotic medication.

Introduction

The treatment of schizophrenia during the acute phase primarily relies on the use of antipsychotic medications that target positive symptoms, notably hallucinations and delusions (Kahn et al., 2015). However, negative symptoms, cognitive impairment, and mood symptoms are also evident prodromally, before the first acute episode, which is typically characterized by the onset of prominent positive symptoms. These other symptom domains persist during the acute phase (Yasui-Furukori, 2012). In addition, negative symptoms, mood symptoms, and cognitive impairment represent an important therapeutic challenge during maintenance treatment and are often associated with functional impairment and increased risk of relapse or recurrence (Heinrichs and Carpenter, 1985; Wang et al., 2018; Yamada et al., 2022; Starzer et al., 2023).

The efficacy of second-generation antipsychotic medications in the treatment of schizophrenia has been confirmed in numerous clinical trials, primarily in the context of acute exacerbation and first-episode cases, as evidenced by the changes observed in the Positive and Negative Symptom Scale (PANSS) scores (Kay et al., 1987). The PANSS has demonstrated good reliability and validity and is widely used as a measure of symptom severity in schizophrenia (Kay et al., 1987). One way to investigate the broad effects of antipsychotics in the treatment of schizophrenia is to examine changes in the full range of PANSS factors (particularly Marder 5-factors) and the items they are derived from (Marder et al., 1997; Weiden et al., 2022). Although focusing on changes in each of the 30 individual PANSS items has limitations, capturing patterns of change in individual items can potentially lead to a deeper understanding of the effectiveness of medications in each symptom domain (Fleischhacker et al., 2019; Le Moigne et al., 2021). Furthermore, since various symptoms are believed to persist for a longer duration than positive symptoms, it is desirable to observe a broad efficacy early on. Therefore, capturing the temporal changes in several scores is considered effective.

Lurasidone is a second-generation atypical antipsychotic that acts as an antagonist of D₂, 5-HT_2A, and 5-HT₇ receptors and a partial agonist of 5-HT_1A receptor, with very weak binding affinity at H₁ and M₁ receptors (Ishibashi et al., 2010; Horisawa et al., 2013). Compared with 17 other antipsychotics, lurasidone was ranked among the best in terms of metabolic outcomes in a network meta-analysis (Pillinger et al., 2020).

To date, short-term randomized, double-blind studies in schizophrenia have reported efficacy for lurasidone based on measures of change in PANSS total, subscale, and/or factor scores (Meltzer et al., 2011; Loebel et al., 2013, 2015; Nasrallah et al., 2013; Ogasa et al., 2013; Iyo et al., 2021). No report we are aware of has simultaneously compared changes in the Marder 5-factor model and each item of the PANSS between lurasidone and placebo in the same double-blind randomized controlled trial, including changes over time. The goal of the current post-hoc analysis was to investigate the temporal progression of improvement in both the 5 Marder factors and the 30 individual constituent PANSS items during acute treatment with lurasidone.

Methods

This analysis utilized data from the double-blind phase of the JEWEL study, which was a multicenter, randomized, placebo-controlled, phase III trial of lurasidone (JEWEL study; EudraCT number: 2016-000060-42; study initiated in May 2016 and completed in November 2018) (Iyo et al., 2021).

Study design and participants

The Japan and Europe Working to Evaluate Lurasidone (JEWEL) study evaluated the efficacy and safety of lurasidone 40 mg/day compared with placebo over a period of 6 weeks. The study included a screening/washout phase, followed by a double-blind treatment phase (6 weeks) and a follow-up visit (7 days after the last dose of study drug). Hospitalization was required during the washout phase and up to postbaseline week 2. This study was conducted in accordance with the International Conference on Harmonization Guidelines for Good Clinical Practice and the Declaration of Helsinki at 73 clinical sites in five countries (Japan, Ukraine, Russia, Romania, and Poland). An ethics committee at each participating center approved the study protocol, and written informed consent was obtained from each patient following a detailed explanation of study procedures.

Potentially eligible patients (age 18–74) experiencing an acute exacerbation of schizophrenia were diagnosed by clinical interview using the Mini-International Neuropsychiatric Interview (MINI) (Sheehan et al., 1998) 6.0.0 diagnostic interview in conjunction with the Diagnostic and Statistical Manual of Mental Disorder, 4th Edition, Text Revision (DSM-IV-TR) criteria (APA, 2000). Eligible patients also met the following key inclusion criteria: PANSS total score ≥80; a PANSS subscale score ≥4 (moderate) on two or more of the following PANSS subscale items: delusions, conceptual disorganization, hallucinations, suspiciousness, or unusual thought content at both screening and baseline; a score of four (moderately ill) or higher on the Clinical Global Impressions-Severity of Illness (CGI-S) (Guy, 1976) at screening and baseline; and an acute exacerbation of mainly positive symptoms for no longer than 2 months before the screening visit and marked deterioration of function from baseline. The main exclusion criteria for the JEWEL study included continuous hospitalization for more than 90 days in the 3 months preceding screening, continuous hospitalization due to an acute onset of psychotic symptoms for more than 14 days before screening while receiving adequate doses of one or more antipsychotic drugs, and a reduction of 20% or more in the PANSS total score between screening and baseline visits.

The enrolled patients were randomly assigned in a 1 : 1 ratio to receive either lurasidone 40 mg/day fixed dose or placebo once daily.

Outcomes

For the current report, we focused on changes from baseline to week 6 in each PANSS Marder 5-factor score and from baseline to week 6 in each of the 30 individual PANSS items. The Marder factors consist of the following (Marder et al., 1997): (1) Positive symptoms with eight items [including delusions (P1), hallucinatory behavior (P3), grandiosity (P5), suspiciousness (P6), stereotyped thinking (N7), somatic concern (G1), unusual thought content (G9), and lack of judgment/insight (G12)], (2) Negative symptoms with seven items [blunted affect (N1), emotional withdrawal (N2), poor report (N3), passive social withdrawal (N4), lack of spontaneity and flow of conversation (N6), motor retardation (G7), and active social avoidance (G16)], (3) Disorganized thought with seven items [conceptual disorganization (P2), difficulty in abstract thinking (N5), mannerism and posturing (G5), disorientation (G10), poor attention (G11), disturbance of volition (G13), preoccupation (G15)], (4) Hostility/excitement with four items [excitement (P4), hostility (P7), uncooperativeness (G8), poor impulse control (G14)], and (5) Anxiety/depression with four items [anxiety (G2), guilt feeling (G3), tension (G4), depression (G6)].

Statistical analysis

The population for the analysis was patients who received at least one dose of the investigational drug and had at least one postbaseline evaluation of the PANSS total score (intention-to-treat population).

The PANSS consists of 30 items (seven positive, seven negative, and 16 general psychopathology) and was administered weekly throughout the study by trained interviewers. Each item is rated on a 7-point scale (1 = absent, 2 = minimal, 3 = mild, 4 = moderate, 5 = moderate-severe, 6 = severe, and 7 = extreme) to evaluate the current severity level of each symptom (item) (Kay et al., 1987). The PANSS individual items and 5-factor measure were analyzed using a mixed effects model for repeated measures, with fixed factors of pooled study center, visit (weeks 1, 2, 3, 4, 5, and 6 as a categorical variable), treatment (lurasidone and placebo), and treatment-by-visit interaction, and baseline scores on the respective individual items and 5-factor score as a covariate. A specific contrast was included in the model to evaluate changes at week 6. An unstructured covariance matrix was used for within-subject correlation. The Kenward–Roger approximation was used to calculate the degrees of freedom of the denominator. Effect sizes were calculated as the estimated least squares mean difference at week 6 divided by the model estimate of SD. Additional mixed models examined changes from baseline to each weekly assessment and from baseline to week 6 for all PANSS items. P-values less than 5% (two-sided test) were considered statistically significant. Analyses were conducted using SAS Version 9.4.

Results

Baseline characteristics of sample

A total of 483 patients were randomized to lurasidone 40 mg/day or placebo, and the intention-to-treat population was 245 and 233, respectively (Fig. 1). The lurasidone and placebo groups were similar in terms of sex, age, duration of illness, and frequency of more than four hospitalizations (Table 1). The mean PANSS total score at baseline was approximately 100 in both groups, and the Marder 5-factors showed a mean score of 30.6 for positive symptoms, 24.0 for negative symptoms, 24.0 for disorganized thoughts, 11.4 for hostility/excitement, and 12.8 for anxiety/depression. The effect of the entry criteria that enriched the treatment sample for positive symptoms was reflected in the smaller percentage of positive symptom items (3/8; 37.5%) with a severity score ≤3.8 (Table 1). A severity score ≤3.8 indicates that the majority of patients rated the severity of that item as absent (score 1), minimal (score 2), or mild (score 3). In contrast, the percentage patients with low item severity scores was much higher at baseline for negative symptoms 6/7; 85.7%), disorganized thought (4/7; 57.1%), hostility/excitement (4/4; 100%), and anxiety/depression (3/4; 75.0%).

Fig. 1.

Patient disposition.

Table 1.

Baseline characteristics of samples

	Lurasidone (N = 245)	Placebo (N = 233)
Sex (male), n (%)	119 (48.6)	120 (51.5)
Age (years), mean (SD)	41.0 (11.0)	39.3 (11.4)
Disease duration (years), mean (SD)	10.6 (8.0)	10.4 (8.3)
Prior hospitalization (≥4), n (%)	141 (57.6)	135 (57.9)
PANSS total score, mean (SD)	102.8 (11.0)	101.7 (11.5)
PANSS subscale score, n (%)
Positive subscale ≥ negative subscale	157 (64.1)	144 (61.8)
Positive subscale < negative subscale	88 (35.9)	89 (38.2)
Positive symptomsa	30.6 (4.1)	30.3 (4.1)
P1: Delusions	4.7 (0.8)	4.6 (0.8)
P3: Hallucinatory behavior	4.3 (1.1)	4.2 (1.2)
P5: Grandiosityb	2.7 (1.3)	2.9 (1.4)
P6: Suspiciousness	4.4 (0.9)	4.4 (0.8)
N7: Stereotyped thinkingb	3.4 (0.8)	3.3 (0.9)
G1: Somatic concernb	3.1 (1.9)	3.0 (1.3)
G9: Unusual thought content	4.0 (0.8)	3.9 (0.8)
G12: Lack of judgment/insight	4.0 (0.9)	4.0 (0.8)
Negative symptomsa	24.0 (4.3)	23.6 (4.6)
N1: Blunted affectb	3.5 (0.9)	3.5 (0.9)
N2: Emotional withdrawalb	3.7 (0.8)	3.6 (0.9)
N3: Poor rapportb	3.2 (0.8)	3.2 (0.8)
N4: Passive social withdrawalb	3.8 (1.0)	3.7 (1.0)
N6: Lack of spontaneity and flow of conversationb	3.2 (1.0)	3.1 (1.0)
G7: Motor retardationb	2.7 (1.0)	2.7 (1.0)
G16: Active social avoidance	3.9 (0.9)	3.8 (0.8)
Disorganized thoughta	24.0 (3.4)	24.2 (3.8)
P2: Conceptual disorganization	3.9 (0.8)	3.9 (0.8)
N5: Difficulty in abstract thinking	3.9 (1.0)	3.9 (1.0)
G5: Mannerism and posturingb	3.0 (0.8)	3.0 (0.9)
G10: Disorientationb	2.6 (0.8)	2.7 (1.0)
G11: Poor attentionb	3.3 (0.8)	3.3 (0.8)
G13: Disturbance of volitionb	3.4 (0.7)	3.5 (0.8)
G15: Preoccupation	3.9 (1.0)	3.8 (1.0)
Hostility excitementa	11.4 (3.0)	11.0 (3.3)
P4: Excitementb	3.0 (0.9)	3.0 (1.0)
P7: Hostilityb	2.7 (1.0)	2.6 (1.1)
G8: Uncooperativenessb	2.9 (1.0)	2.7 (0.9)
G14: Poor impulse controlb	2.7 (1.0)	2.6 (1.0)
Anxiety/depressiona	12.8 (3.1)	12.6 (3.5)
G2: Anxiety	3.9 (1.1)	3.7 (1.2)
G3: Guilt feelingb	2.4 (1.2)	2.5 (1.2)
G4: Tensionb	3.4 (1.0)	3.3 (1.1)
G6: Depressionb	3.2 (1.1)	3.1 (1.1)

N (%) or mean (SD) is expressed.

PANSS, Positive and Negative Syndrome Scale.

^aMarder 5 factors model score.

^bBaseline PANSS item score ≤3.8.

The co-administration rate of anxiolytics was 26.7% in the lurasidone group and 31.5% in the placebo group. The co-administration rates of hypnotics and sedatives were 20.2% and 22.6% in the lurasidone and placebo groups, respectively.

Outcomes: Positive and Negative Syndrome Scale Marder 5-Factors

All domains of the Marder 5-factors showed a significantly greater change from baseline at week 6 in the lurasidone group than in the placebo group based on the mixed effect model (Table 2). The effect sizes for positive symptoms, negative symptoms, disorganized thoughts, hostility/excitement, and anxiety/depression were 0.44, 0.22, 0.32, 0.28, and 0.30, respectively. For the four Marder domains, excluding the negative symptom factor, change in the lurasidone group was significantly greater than in the placebo group beginning at week 2 and continuing through to week 6 (Fig. 2). Additionally, the change in the Marder negative symptom factor was significantly greater in the lurasidone group than in the placebo group beginning at week 4 and continuing through to week 6.

Table 2.

Change from baseline at 6 weeks in PANSS Marder 5-factors: Results for the total sample and for subjects not treated with anxiolytics or sedative/hypnotics

Total sample	Lurasidone 40 mg/day (N = 245)	Placebo (N = 233)	P value (Lurasidone vs Placebo)	Effect size
Positive symptoms	−6.6 (0.4)	−4.3 (0.4)	<0.001	0.44
Negative symptoms	−3.8 (0.3)	−2.9 (0.3)	0.029	0.22
Disorganized thought	−3.7 (0.3)	−2.5 (0.3)	0.001	0.32
Hostility/excitement	−2.0 (0.2)	−1.2 (0.2)	0.006	0.28
Anxiety/depression	−3.4 (0.2)	−2.5 (0.2)	0.003	0.30
Subjects not treated with anxiolytics or sedative/hypnotics	Lurasidone 40 mg/day (N = 167)	Placebo (N = 142)	P value (Lurasidone vs Placebo)	Effect size
Positive symptoms	−7.8 (0.4)	−5.4 (0.5)	<0.001	0.47
Negative symptoms	−4.3 (0.3)	−3.4 (0.4)	0.049	0.24
Disorganized thought	−4.3 (0.3)	−3.4 (0.3)	0.015	0.30
Hostility/excitement	−2.6 (0.3)	−1.9 (0.3)	0.053	0.24
Anxiety/depression	−3.8 (0.3)	−3.0 (0.3)	0.032	0.26

Mean change estimate (SE) is expressed.

PANSS, Positive and Negative Syndrome Scale.

Fig. 2.

Mean change over 6 weeks of treatment with lurasidone and placebo for PANSS Marder 5-factors. PANSS, Positive and Negative Syndrome Scale.

Because of the relatively high rates of concomitant treatment with anxiolytics and sedative/hypnotics, we performed a sensitivity analysis that included the subgroup of patients not treated with those concomitant medications. The lurasidone versus placebo effect sizes at week 6 were very similar (Table 2), suggesting that treatment effects on PANSS factors as-needed treatment with anxiolytics and sedative/hypnotic.

Outcomes: Positive and Negative Syndrome Scale individual items

The lurasidone group demonstrated significantly greater changes than the placebo group at week 6 on the following five out of eight Marder positive symptom factor items: delusions (P1), hallucinatory behavior (P3), suspiciousness (P6), somatic concern (G1), and unusual thought content (G9); on the following one out of seven Marder negative symptom factor items: Active social avoidance (G16) item; on the following four out of seven Marder disorganized thought factor items: Conceptual disorganization (P2), difficulty in abstract thinking (N5), poor attention (G11), and preoccupation (G15); on the following two out of four Marder hostility/excitement factor items: Hostility (P7) and poor impulse control (G14) items; and on the following two out of four Marder anxiety/depression factor items: Anxiety (G2) and tension (G4) items. These results are shown as a heatmap and as a radar chart (Fig.3a,b).

Fig. 3.

(a) PANSS individual items: statistical significance of change over time at each weekly assessment. (b) PANSS individual items: statistical significance of change from baseline to week 6. PANSS, Positive and Negative Syndrome Scale.

For individual PANSS items, change from baseline was significantly greater in the lurasidone group than in the placebo group beginning at weeks 2 and continuing through to week 6 on the following items: P1, P3, G16, P2, G15, and G2; and beginning at week 3: P6 and G4; and later onset (at weeks 4 or 5) of significant efficacy were observed for the following PANSS items: G9, N5, G11, P7, G8, and G14 (Fig. 3a).

Discussion

This analysis of a double-blind, placebo-controlled, 6-week trial compared temporal changes for the lurasidone and placebo groups in the PANSS Marder 5-factors and in the individual PANSS items. Lurasidone 40 mg/day improved all Marder five factors at week 6 compared with placebo, with significance observed from week 2 onward for four of the five factors (significance was observed for negative symptoms beginning at week 4).

On individual items, treatment with lurasidone was associated with significant improvement (onset at weeks 2 or 3) on the following items: delusions, hallucinatory behavior, suspiciousness, active social avoidance, conceptual disorganization, preoccupation, anxiety, and tension. In addition, treatment with lurasidone was associated with improvement in five out of the seven items on the Marder disorganized thought factor, with onset of significant improvement observed at week 4. These results indicate that lurasidone had notable early and sustained improvement on the Marder positive and anxiety/depression symptom factors, and somewhat later onset of significant effects on the Marder disorganized thought and hostility/excitement factors.

Overall, a total of 16 PANSS items (see Fig. 3a) did not demonstrate significant improvement at week 6, and all 16 of those items had low mean severity scores at baseline (ranging from 2.4 to 3.8). For this reason, we are hesitant to draw inferences regarding the possible lack of efficacy of lurasidone for these specific items, since it is difficult to demonstrate a significant level of improvement for items whose baseline severity is minimal. This is especially true for the negative symptom factor, where only one out of the seven constituent items (active social avoidance) had a baseline score that was moderate in severity; and, notably, it was the only item that demonstrated early (week 2) and sustained significance. To our knowledge, there has been no published clinical trial that examined the efficacy of lurasidone for the treatment of schizophrenia in a population that was enriched for high levels of the negative symptom factor. Alternatively, this issue could be addressed, in an exploratory fashion, by a pooled post hoc analysis that was limited to patients who met criteria for a predominant negative symptom presentation.

At pretreatment baseline, patients with an acute exacerbation of schizophrenia included in this study exhibited a wide range of moderate-to-severe symptoms. In clinical practice, during the acute phase of illness, positive symptoms typically are more prominent and are the focus of treatment. However, the results of the current study demonstrate the extent to which individual items in the four other PANSS Marder domains exhibit clinically relevant severity, with mean severity scores >3 on six out of seven items in the Marder negative symptom factor, on 6 out of 7 items in the disorganized thought factor, one out of four items in the hostility/excitement factor, and in three out of four items in the anxiety/depression factor.

Treatment with lurasidone 40 mg/day showed significant improvement compared with placebo from week 2 to week 6 endpoint on all five Marder factors except for the negative symptom factor, which showed significant improvement from week 4 to week 6 endpoint. The current result for the 40 mg/day dose of lurasidone is consistent with effect sizes in a previously reported pooled analysis of five clinical trials with lurasidone (40–160 mg/day; Loebel et al., 2015): Marder positive symptom factor (effect size, 0.44 vs. 0.43), hostility/excitement factor (0.28 vs. 0.31), and for the anxiety/depression factor (0.30 vs. 0.31), but not for the negative symptom factor (0.22 vs. 0.33) or for the disorganized thought factor (0.32 vs. 0.42).

Endpoint effect sizes other than the positive symptom factor were small to moderate. For the positive symptom factor, the effect size for lurasidone was 0.44, which compares favorably to the meta-analytic positive symptom factor effect sizes reported for other atypical oral antipsychotic medications approved (like lurasidone) in the past 15 years: asenapine (0.47), cariprazine (0.30), iloperidone (0.30), and brexpiprazole (0.17) (Huhn et al., 2019).

Treatment with lurasidone 40 mg/day was associated with significant reduction in symptom severity on five out of eight items on the Marder positive symptom factor, one out of seven items on the negative symptom factor, four out of seven items on the disorganized thought factor, two out of four items on the hostility/excitement factor, and two out of four items on the anxiety/depression factor. Lack of improvement on individual PANSS items may be partly attributable to the fact that 12 out of the 16 nonsignificant PANSS items had relatively low severity scores at baseline, ranging from 2.4 to 3.4, thus making it difficult to demonstrate a treatment effect.

The clinical importance of individual PANSS items is largely due to the impact each item has on key correlated outcomes. For example, delusions (P1) and suspiciousness (P6) have been associated with impairment in social functioning (Menon et al., 2008; Fan et al., 2022); conceptual disorganization (P2), tension (G4), and poor impulse control (G14) have been associated with impairment in cognitive functioning (Wang et al., 2013; Silva et al., 2021); hallucinatory behavior (P3) and difficulty in abstract thinking (N5) have associated with impairment in perceptual filtering function (indexed by acoustic startle response and P50 gating) (Storozheva et al., 2016); and hostility (P7) and poor impulse control (G14) have been associated with aggressive behavior (Lera Calatayud et al., 2012; Citrome and Volavka, 2021). More generally, PANSS items such as anxiety (G2) and active social avoidance (G16) have been associated with poor prognosis, and conceptual disorganization (P2), difficulty in abstract thinking (N5), and unusual thought content (G9) have been associated with treatment resistance (Braga et al., 2013; Montreuil et al., 2013; Robertson et al., 2014; Ortiz et al., 2020).

Moreover, several items showed score reductions from baseline as early as week 2 or 3. These items showing early change included positive symptoms, such as hallucinations and delusions, as well as factors related to social functioning, cognitive function, and other outcomes. It is possible that improvement in anxiety, tension, preoccupation, and active social avoidance may have an indirect effect associated with improvement in positive symptoms. However, the receptor profile and the results of preclinical research suggest that lurasidone may be independently effective in the treatment of, for example, anxiety and depressive symptoms and symptoms of cognitive impairment (due to 5-HT_1A agonist activity and 5-HT₇ antagonist activity) (Ishibashi et al., 2010; Celada et al., 2013; Fukuyama et al., 2023).

Early improvement in these symptom domains is considered a favorable profile when considering treatment and personal recovery focused on the acute-phase treatment of schizophrenia. However, further research is needed to determine whether improving these items has an impact on long-term functioning and relapse. Preliminary evidence suggests that lurasidone improves cognitive function in patients with schizophrenia, alleviates depressive symptoms, and reduces relapse compared with placebo (Harvey et al., 2013; Loebel et al., 2013a; Suppes et al., 2016; Tandon et al., 2016). The clinical utility of lurasidone demonstrated in these studies may be partly attributed to the improvement in specific domains reflected in the diverse PANSS items examined here.

This study has several limitations, most notably, it was based on a post hoc analysis of a randomized clinical trial in patients meeting enrollment criteria for an acute exacerbation of schizophrenia that was enriched for PANSS-Positive symptomatology, Therefore, the generalizability of the current results to the broader population of individuals diagnosed with schizophrenia in real world clinical settings is uncertain; most notably individuals with prominent negative or cognitive symptoms. Furthermore, the clinical trial excluded patients with ≥20% screen-to-baseline reduction in PANSS total score, which may have selected for a patient population that was either less responsive to treatment or had a different symptom response trajectory. Due to the use of only a 40 mg dose of lurasidone, it is unclear whether lurasidone might exhibit dose-response effects on specific Marder PANSS factors or individual PANSS items at higher doses. An additional limitation is that the trial only lasted for 6 weeks, and a longer-term follow-up is needed to evaluate the impact of the reduction in PANSS symptom severity on improvement in social functioning, occupational functioning, remission, and recovery. Finally, the current analysis did not evaluate the extent to which the effect of lurasidone on the four Marder PANSS factors was independent of the effect on the fifth Marder PANSS Positive factor.

Conclusions

In patients with an acute exacerbation of schizophrenia, treatment with lurasidone was associated with significant endpoint improvement on all five Marder PANSS factors. Significant early (week 2), sustained improvement was observed on four of the five Marder Factors, with significant improvement occurring by week 4 on the Marder negative symptom factor. The ability to determine the efficacy of lurasidone for all 30 PANSS items (and notably for items that clustered in the negative symptom factor) was not possible because study entry criteria—designed to identify a treatment sample enriched for positive symptoms—resulted in 16 items with baseline severity that was minimal-to-mild, and thus insufficiently symptomatic to test for efficacy. These 16 PANSS items comprised the total number of items that did not demonstrate significant improvement at week 6.

Acknowledgements

We thank all patients who participated in this study. We also thank Kei Watabe and Fumiya Sano for statistical analysis.

This study was sponsored by Sumitomo Pharma Co., Ltd.

Medical writing support was funded by Sumitomo Pharma Co., Ltd. Japan and was provided by Edward Schweizer, M.D. of the Paladin Consulting Group, Plainsboro, NJ.

The datasets generated during and/or analyzed during the current study are not publicly available owing to participant privacy.

Conflicts of interest

I.M. reports receiving personal fees from Sumitomo Pharma Co., Ltd. K.T. is an employee of Sumitomo Pharma Co., Ltd. S.T. was an employee of Sumitomo Pharma Co., Ltd. M.I. reports receiving consultant fees from Eli Lilly Japan K.K., Sumitomo Pharma Co., Ltd., Pfizer Japan Inc., Abbott Japan Co., Ltd., and Janssen Pharmaceutical K.K., and report honoraria from Janssen Pharmaceutical K.K., Eli Lilly Japan K.K., Otsuka Pharmaceutical Co., Ltd., Meiji Seika Pharma Co., Ltd., Astellas Pharma Inc., Sumitomo Pharma Co., Ltd., Ono Pharmaceutical Co., Ltd., GlaxoSmithKline K.K., Takeda Pharmaceutical Co., Ltd., Mochida Pharmaceutical Co., Ltd., Kyowa Kirin Co., Ltd., MSD K.K., Eisai Co. Ltd., Daiichi-Sankyo Co. Ltd., Novartis Pharma K.K., Teijin Ltd., Shionogi & Co., Ltd., Hisamitsu Pharmaceutical Co., Inc., and Asahi Kasei Corporation.