Lurasidone for schizophrenia: a review of the efficacy and safety profile for this newly approved second-generation antipsychotic

Authors


  • Disclosures No writing assistance or external financial support was utilised in the production of this article. Leslie Citrome, is a consultant for, has received honoraria from, or has conducted clinical research supported by the following: Abbott Laboratories, AstraZeneca Pharmaceuticals, Avanir Pharmaceuticals, Azur Pharma Inc, Barr Laboratories, Bristol-Myers Squibb, Eli Lilly and Company, Forest Research Institute, GlaxoSmithKline, Janssen Pharmaceuticals, Jazz Pharmaceuticals, Merck, Novartis, Pfizer Inc, Sunovion and Vanda Pharmaceuticals. As Psychiatry Section Editor for the Journal, Leslie Citrome withdrew from the review process and deferred all editorial decisions to Graham Jackson.

Leslie Citrome, 11 Medical Park Drive, Suite 106, Pomona, NY 10970, USA
Tel.: + 1 845 362 2081
Fax: + 1 845 362 8745
Email: nntman@gmail.com

Summary

Objective:  To describe the efficacy and safety of lurasidone for the treatment of schizophrenia.

Data sources:  The pivotal registration trials were accessed by querying the literature databases PubMed, EMBASE, ISI Web of Knowledge, as well as http://www.fda.gov and http://www.clinicaltrials.gov for the search term ‘lurasidone’. Product labelling provided additional information.

Study selection:  All available clinical reports of studies were identified.

Data extraction:  Descriptions of the principal results and calculation of number needed to treat (NNT) and number needed to harm (NNH) for relevant dichotomous outcomes were extracted from the available study reports, abstracts and posters. Additional safety outcomes subject to NNH analysis were obtained from product labelling.

Data synthesis:  Lurasidone is a second-generation antipsychotic approved for the treatment of schizophrenia at a recommended starting dose of 40 mg/day administered once daily with food (≥ 350 calories). The maximum recommended dose is 80 mg/day. Regulatory approval was based primarily on a clinical trial programme that included four 6-week randomised clinical trials demonstrating efficacy vs. placebo in acute patients with schizophrenia. One additional Phase II clinical trial was considered a failed study because neither lurasidone nor the active control, haloperidol, separated from placebo on the primary outcome measure. One additional Phase III study was completed after the new drug application was submitted to the US Food and Drug Administration. Efficacy outcomes appear consistently in favour of lurasidone 80 mg/day vs. placebo on multiple measures of psychopathology, however, at least two studies also demonstrated efficacy for the doses of 40 and 120 mg/day. NNT vs. placebo was 3–6 for response as defined by ≥ 20% reduction in psychopathological rating scale total scores from baseline, depending on the study and the dose. Response as defined by a ≥ 30% improvement yielded NNTs ranging from 7 to 13. The most common adverse events in the clinical trials were somnolence (broadly defined), akathisia, nausea, parkinsonism and agitation. As estimated from product labelling, NNH vs. placebo was dose dependent for somnolence, with a NNH of 6 for lurasidone 120 mg/day, compared with NNHs of 8, 11 and 20, for 80, 40 and 20 mg/day, respectively. For akathisia NNH was 6 for lurasidone 120 mg/day, compared to NNHs of 9, 13 and 34 for 80, 40 and 20 mg/day, respectively. Lurasidone is associated with minimal weight gain and no clinically meaningful alterations in glucose, lipids, prolactin or the ECG QT interval.

Conclusions:  Lurasidone 40 and 80 mg/day appear efficacious and tolerable in the treatment of schizophrenia. Doses above 80 mg/day do not appear to confer added benefit and may be associated with a dose-related increase in certain adverse reactions. Principal advantages over some other second-generation antipsychotics are lurasidone’s highly favourable metabolic profile and once-daily dosing regimen. Additional data regarding long-term efficacy and effectiveness will help characterise this new agent when used in maintenance treatment.

Review Criteria

The pivotal registration trials were accessed by querying the literature databases PubMed, EMBASE, ISI Web of Knowledge, as well as http://www.fda.gov and http://www.clinicaltrials.gov for the search term ‘lurasidone’. Product labelling provided additional information.

Message for the Clinic

Lurasidone is an oral second-generation antipsychotic that received regulatory approval on 28 October 2010 by the US Food and Drug Administration for the treatment of schizophrenia in adults. Lurasidone 40 and 80 mg/day administered once daily with food appears efficacious and tolerable in the treatment of schizophrenia. Doses above 80 mg/day do not appear to confer added benefit and may be associated with a dose-related increase in certain adverse reactions. Principal advantages over some other second-generation antipsychotics are lurasidone’s highly favourable metabolic profile and once-daily dosing regimen.

Introduction

Lurasidone is a second-generation antipsychotic that received regulatory approval on 28 October 2010 by the US Food and Drug Administration (FDA) for the treatment of schizophrenia in adults (1).

Lurasidone joins a long list of second-generation antipsychotics, some of which have only recently been commercialised. Differentiating among the different choices for the individual patient is complex, requiring consideration of prior history of therapeutic response, prior history of tolerability with other agents and individual patient values and preferences. With regard to the latter, obstacles to adherence include lack of insight, lack of perceived benefit, as well as the presence of specific tolerability problems that an individual patient may be particularly sensitive to.

The aim of this review is to synthesise the available data about the efficacy, tolerability and safety of lurasidone for the treatment of schizophrenia, and to place this new antipsychotic into clinical perspective.

Data sources

Clinical trial information was accessed by on-line query on 4 October 2010 of the literature databases PubMed, EMBASE, ISI Web of Knowledge, as well as http://www.fda.gov and http://www.clinicaltrials.gov for the search term ‘lurasidone’. No date or language constraints were utilised. The manufacturer’s website was examined. Information was also extracted from product labelling (2).

Study selection

Query of the PubMed literature database identified six records that contained the search term ‘lurasidone’. Of these, only one was a clinical trial report (3), the remainder included a description of lurasidone’s pharmacological profile (4), a review focused on lurasidone (5), a general review of 5-HT1A receptor agonism (6) and two reports of studies conducted in rats (7,8). Query of EMBASE and the ISI Web of Knowledge identified 40 and 23 records, respectively, yielding several abstracts of presentations at professional meetings concerning lurasidone clinical trials (9–22) and one of comparative receptor binding profiles (23), as well as one additional review paper (24). Query of http://www.fda.gov did not provide any additional information. Inspection of the product label revealed additional efficacy and safety information (2). Query of http://www.clinicaltrials.gov revealed 24 studies and these are summarised in Table 2. Not all completed trials were found on http://www.clinical.trials.gov, for example there were posters describing additional clinical trials (12,16). A search on the manufacturer’s website produced copies of slides of a presentation for investors (25). Press releases have at times included categorical responder rates not otherwise available (26). Following FDA approval, additional information about a third Phase III study became publically available (27). Additional posters available to the author include those from meetings personally attended (28–34) for which abstracts are not available in indexed publications. All studies of lurasidone reported to date have been conducted by its manufacturer.

Table 2.   Lurasidone clinical trials registered at http://www.clinicaltrials.gov*
ClinicalTrials.gov identifier and manufacturer’s identifier(s)Study titleLengthPhaseConditionNLurasidone doses (once-daily) and comparator(s)Comments
  1. *Accessed 4 October 2010, with additional information on treatment arms obtained at http://www.controlled-trials.com on 13 October 2010. As much as possible, missing details regarding length of trial and dosing were filled in using information from posters and other publically available sources.

NCT01082146, D1050262A Phase I study to investigate the absorption, metabolism, and excretion of [ISOTHIAZOLYL-3-14C]-LURASIDONE following postprandial single oral dose administration in healthy male subjects7 daysIMen, healthy normal subjects614C-Labelled 40 mgCompleted; conducted August 2008–August 2008; United States
NCT01082276, D1050270Rifampin drug–drug interaction study with lurasidone HCl8 daysIMen, healthy normal subjects2040 mg with rifampin 600 mgCompleted; conducted August 2008–August 2008; United States
NCT00549666, D1050246, OC-DDIA randomised, placebo-controlled, two-period, crossover study to evaluate the effect of lurasidone HCl on oral contraceptive pharmacokinetics in healthy female subjects56 daysIHealthy women2440 mg with Ortho Tri-CyclenCompleted; conducted August 2007–December 2007; United States
NCT01082289, D1050279A Phase I, open-label, drug–drug interaction study to determine the effect of repeated-dose lurasidone 120 mg administration on the pharmacokinetics of orally administered digoxin 0.25 mg in patients with schizophrenia or schizoaffective disorder8 daysISchizophrenia, schizoaffective disorder24120 mg with digoxin 0.25 mgCompleted; conducted September 2008–October 2008; location(s) not specified
NCT01082263, D1050269Midazolam drug–drug interaction study with lurasidone HCl10 daysISchizophrenia24120 mg with midazolam 5 mgCompleted; conducted October 2008–November 2008; United States
NCT01082250, D1050263An open-label, randomised, three-period, two-sequence, crossover, repeated-dose, replicate design study to determine the bioequivalence of two different lurasidone formulations in patients with schizophrenia, schizoaffective, or schizophreniform disorder3 weeksISchizophrenia, schizoaffective disorder, or schizophreniform disorder52120 mgCompleted; conducted July 2008–September 2008; United States
NCT01074632, D1050267A Phase I, open-label study to determine the effect of calories and fat content on the pharmacokinetics of repeated-dose lurasidone 120 mg in subjects with schizophrenia, schizoaffective disorder, or schizophreniform disorder30 daysISchizophrenia, schizoaffective disorder, or schizophreniform disorder26120 mgCompleted; conducted May 2009–September 2009; United States
NCT01074073, D1050247A Phase I, open-label, 2-period, sequential, drug–drug interaction study to determine the effect of lithium 600 mg bid twice daily on the safety and pharmacokinetics of lurasidone 120 mg QD in patients with schizophrenia or schizoaffective disorder8 daysISchizophrenia24120 mg with lithium 600 mg bid twice dailyCompleted; conducted August 2008–October 2008; United States
NCT00044044, D1050049A 6-week, double-blind, randomised, fixed-dose, parallel-group study of the efficacy and safety of three dose levels of SM-13496 compared with placebo and haloperidol in patients with schizophrenia who are experiencing an acute exacerbation of symptoms6 weeksIISchizophrenia33020, 40 and 80 mg, haloperidol 10 mg/day, placeboCompleted; conducted July 2002–October 2003; United States
NCT00044005, D1050174A randomised, open-label, dose-blinded, multicenter, 6-month study of safety and tolerability of 3 dose levels of SM-13496 in patients with schizophrenia6 monthsIISchizophrenia150Open-label doses not specifiedCompleted; conducted September 2002–December 2003; United States; extension for study D1050049
NCT00088634, D1050196A double-blind fixed-dose study antipsychotic and placebo in the treatment of schizophrenia6 weeksIISchizophrenia16080 mg, placeboCompleted; conducted May 2004–December 2004; Published (3)
NCT00088621, D1050199An open-label, multicenter, 12-month study of safety and tolerability in the treatment of schizophrenia12 monthsIISchizophrenia100Open-label dose(s) not specifiedCompleted; conducted July 2004–December 2005; United States; extension for study D1050196
NCT00549718, D1050229, PEARL 1A Phase III randomised, placebo-controlled, clinical trial to study the safety and efficacy of three doses of lurasidone HCl in acutely psychotic patients with schizophrenia6 weeksIIISchizophrenia48040, 80 and 120 mg, placeboNo longer recruiting; start October 2007; United States, France, India, Malaysia, Romania, Russian Federation, Ukraine
NCT00615433, D1050231, PEARL 2A Phase III randomised, placebo-and active comparator-controlled, clinical trial to study the safety and efficacy of two doses of lurasidone HCl in acutely psychotic patients with schizophrenia6 weeksIIISchizophrenia47840 and 120 mg, olanzapine 15 mg/day, placeboCompleted; conducted January 2008–December 2009; United States, Colombia, India, Lithuania, Philippines
NCT00790192, D1050233, PEARL 3A Phase III randomised, double-blind, placebo- and active comparator-controlled clinical trial to study the efficacy and safety of two doses of lurasidone in acutely psychotic subjects with schizophrenia (PEARL 3)6 weeksIIISchizophrenia48080 and 160 mg, quetiapine XR 600 mg/day, placeboCompleted; conducted October 2008–July 2010; United States, Colombia, India, Romania, Russian Federation, Ukraine
NCT00789698, D1050234, PEARL 3 ExtA Phase III randomised, double-blind, active comparator-controlled clinical trial to study the safety and efficacy of lurasidone in subjects with schizophrenia (PEARL 3 extension study)12 monthsIIIChronic schizophrenia24040–160 mg, quetiapine XR 200–800 mg/dayRecruiting; conducted December 2008–Present; United States, Colombia, India, Romania, Russian Federation, Ukraine; extension for study D1050233
NCT00641745, D1050237Long-term safety, tolerability, and effectiveness of lurasidone in subjects with schizophrenia or schizoaffective disorder: a randomised, active comparator-controlled trial18 monthsIIISchizophrenia, schizoaffective disorder60040–120 mg, risperidone 2–6 mg/dayNo longer recruiting; conducted March 2008–Present; United States, Argentina, Brazil, Chile, Croatia, Israel, South Africa, Thailand
NCT00711269, D1001002, JapicCTI-080585Randomised, placebo-controlled, double-blind, parallel-group, confirmatory study of SM-13496 (lurasidone HCl) in patients with schizophrenia?IIISchizophrenia44040 and 80 mg, risperidone, placeboRecruiting; start July 2008; Japan, Korea, Taiwan
NCT01143077, D1050289, Switch CoreA randomised, 6-week, open-label, study evaluating the safety, tolerability, and efficacy of lurasidone for the treatment of schizophrenia or schizoaffective disorder in subjects switched from other antipsychotic agents (Protocol No. 1050289)6 weeksIIISchizophrenia, schizoaffective disorder2403 open-label-arms; dose not specifiedRecruiting; conducted June 2010–Present; United States
NCT01143090, D1050290, Switch ExtA 24-Week, flexible-dose, open-label extension study of subjects switched to lurasidone for the treatment of schizophrenia or schizoaffective disorder (Protocol No. D1050290)24 weeksIIISchizophrenia, schizoaffective disorder?Open-label dose not specifiedRecruiting; United States; extension for study D1050289
NCT00868452, D1050235, EUDRACT No. 2008-007482-23A randomised, 6-week, double-blind, placebo- controlled, flexible-dose, parallel-group study of lurasidone adjunctive to lithium or divalproex for the treatment of bipolar I depression6 weeksIIIBipolar I, depression340Lurasidone dose not specified + (lithium or divalproex)|, placebo + (lithium or divalproex)Recruiting; conducted April 2009–Present; United States
NCT00868699, D1050236, EUDRACT No. 2008-007457-13A randomised, 6-week, double-blind, placebo-controlled, fixed-flexible dose, parallel-group study of lurasidone for the treatment of bipolar I depression6 weeksIIIBipolar I, depression504Lurasidone dose not specified, placeboRecruiting; conducted April 2009–Present; United States
NCT00868959, D1050256, EUDRACT No. 2008-007483-42A 24-Week, flexible-dose, open-label extension study of lurasidone for the treatment of bipolar I depression24 weeksIIIBipolar I, depression500Open-label dose not specifiedRecruiting; conducted April 2009–Present; United States; extension study for studies D1050235 and D1050236
NCT01173874, TENETS01Clinical and biomarker assessment of efficacy of cognitive remediation in patients with schizophrenia stabilised on lurasidone6 monthsIIISchizophrenia14040–120 mgNot yet recruiting; conducted July 2010–Present; United States; this is an investigator initiated trial under a separate IND

Data extraction

Principal findings are summarised. Number needed to treat (NNT), and number needed to harm (NNH), comparing lurasidone with placebo are calculated where appropriate. The methodology of this technique is described in detail elsewhere (35–38), but essentially NNT and NNH are measures of effect size and indicate how many patients would need to be treated with one agent instead of the comparator in order to encounter one additional outcome of interest. Lower NNTs are evidenced when there are large differences between the interventions in question. For example, a NNT of 2 would be a very large effect size, as a difference is encountered after treating just two patients with one of the interventions vs. the other. A NNT of 50 would mean little difference between the two interventions, as it would take treating 50 patients to encounter a difference in outcome. NNH is used when referring to undesirable events. A useful medication is one with a low NNT and a high NNH when comparing it with another intervention; a low NNT and a high NNH would mean one is more likely to encounter a benefit than a harm. A rule of thumb is that single digit NNTs for efficacy measures suggests that the intervention has potentially useful advantages, and that double digit or higher NNHs for adverse outcomes indicate that the intervention is potentially tolerable. Where possible and applicable, for each NNT or NNH, 95% confidence intervals (CI) are provided.

Data synthesis

Pharmacodynamics and pharmacokinetics

Lurasidone belongs to the chemical class of benzoisothiazol derivatives (2). Lurasidone is also described as an azapirone derivative and its pharmacological properties are summarised in (4). Relevant receptor binding affinities are also outlined in a poster (23). Studies assessing pharmacodynamics involved the use of cloned human receptors or membrane fractions prepared from animal tissue, and may not necessarily reflect biological activity in a patient but can nevertheless help predict what effects can be anticipated. As expected for a second-generation antipsychotic, lurasidone is a full antagonist at dopamine D2 and serotonin 5-HT2A receptors. In addition, lurasidone also has strong affinity for serotonin 5-HT7 receptors, which may have implications for cognition (39). Lurasidone is a partial agonist at 5-HT1A; this may be linked to potential antidepressant and pro-cognitive effects (6,40). Lurasidone has moderate affinity for alpha 2C noradrenergic receptors, and minimal affinity for alpha 1 noradrenergic receptors, the latter predicting a low propensity for orthostatic hypotension (40). Lurasidone has minimal affinity for 5-HT2C receptors and virtually no affinity for histamine H1, predicting a low liability for weight gain (41). No affinity for histamine H1 would also predict a lower risk for sedation compared with agents that have strong H1 activity, although sedation or somnolence can also be produced through other mechanisms. Lurasidone’s lack of affinity for cholinergic M1 receptors would predict a low propensity for causing anticholinergic side effects (40). These predictions would need to be tested in clinical trials (28,31). Lurasidone also exhibits a high selectivity for the D2 receptor subtype compared with the D1 (264-fold higher), D3 (16-fold higher) and D4 (30-fold higher) receptors (23), the clinical relevance of which is not known. Lurasidone’s dopamine D2 receptor occupancy was examined in 21 healthy male subjects using positron-emission tomography (5), with mean D2 receptor occupancies for the 10, 20, 40, 60 and 80 mg groups ranging from 41.3% to 43.3%, 51.0% to 54.8%, 63.1% to 67.5%, 77.4% to 84.3% and 72.9% to 78.9%, respectively. Assuming that 60–80% D2 receptor occupancy would be necessary for antipsychotic response, this could be potentially achieved with oral doses of lurasidone starting at 40 mg/day.

Oral administration results in rapid absorption with a time to maximum concentration of 1–3 h (2). Approximately 9–19% of the administered dose is absorbed (2). At single doses among patients with schizophrenia in the range of 120–160 mg/day, the mean half-life ranged from approximately 29–37 h (5), somewhat higher than the mean half-life of 18 h reported for 40 mg as noted in the product label (2). Following administration of lurasidone 40 mg the mean apparent volume of distribution is estimated to be 6173 l (2). Apparent clearance is 3902 ml/min (2). Area under the curve (AUC) and maximum concentration (Cmax) increases linearly with oral dosing within the range of 20–160 mg (2). Steady-state is reached within 7 days (2). Lurasidone is highly protein-bound (99.8%) (2,5), with affinity for albumin and α-1-glycoprotein (5).

CYP3A4 is the primary metabolic pathway for lurasidone (2,5), and this has implications regarding the use of lurasidone in the presence of inducers and inhibitors of CYP3A4 (see section on Potential drug–drug interactions) (34). Lurasidone is not a substrate of CYP1A1, CYP1A2, CYP2A6, CYP4A11, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6 or CYP2E1 enzymes (2). The major biotransformation pathways are oxidative N-dealkylation, hydroxylation of norbornane ring, and S-oxidation (2) producing two pharmacologically active metabolites, ID-14283 (the exo-hydroxy metabolite) and ID-14326 representing approximately 25% and 3% of the parent exposure, respectively (5,34), as well as a third active metabolite ID-11614 (34) present at extremely low levels (1%) (5) and not mentioned in product labelling. ID-14283 has a similar pharmacological profile as the parent compound but has a shorter half-life (approximately 7.5–10 h) (5). There are also two major non-active metabolites, ID-20219 and ID-20220 (2).

Food can affect the absorption of lurasidone, akin what can be seen with ziprasidone (42,43), but possibly with a lower caloric threshold than necessary with ziprasidone. In a study that included 26 patients with schizophrenia, schizophreniform or schizoaffective disorder, lurasidone 120 mg/day was administered under six different meal conditions: fasting, 350 calories/high fat, 500 calories/low fat, 500 calories/high fat, 800–1000 calories/low fat, and 800–1000 calories/high fat (34). Lurasidone exposure as measured by AUC is increased twofold in the presence of food. Cmax is increased up to threefold. Time to maximum concentration increased by 0.5–1.5 h when lurasidone was administered with food. The actual caloric count or fat composition did not substantially impact the magnitude of the food effect. The metabolites of lurasidone, ID-14283, ID-14326, and ID-11614, showed similar changes in exposure under fed vs. fasting conditions as the parent compound.

In elderly patients with psychosis (ages 65–85 years), lurasidone concentrations when dosed at 20 mg/day were similar to those in young subjects (2).

Renal and hepatic impairment can increase exposure to lurasidone (2). After administration of lurasidone 40 mg (single dose) to patients with mild, moderate and severe renal impairment, mean Cmax increased by 40%, 92% and 54%, respectively, and AUC increased by 53%, 91% and 100%, respectively, compared with healthy matched controls (2). In a single-dose study of lurasidone 20 mg, mean AUC was 1.5-times higher in subjects with mild hepatic impairment (Child-Pugh Class A), 1.7-times higher in subjects with moderate hepatic impairment (Child-Pugh Class B) and 3-times higher in subjects with severe hepatic impairment (Child-Pugh Class C) compared with the values for healthy matched controls (2). Mean Cmax was 1.3, 1.2 and 1.3-times higher for mild, moderate and severe hepatically impaired patients, respectively, compared with the values for healthy matched controls (2).

Mean AUC is slightly higher (18%) in women than in men and the apparent oral clearance of lurasidone was lower in women as assessed by population pharmacokinetic analysis (2). No impact of race on kinetics was observed (2).

After a single dose of radio-labelled lurasidone, 80% of radioactivity was recovered in faeces and 9% recovered in urine (2).

Efficacy in clinical trials

Three Phase II and three Phase III 6-week studies were completed that compared lurasidone at fixed doses ranging from 20 to 160 mg/day with double-blinded placebo in patients with schizophrenia experiencing an acute episode (Table 1). Also reported is an 8-week dose-finding study of inpatients and outpatients in Japan (12), as well as a 3-week study that compared lurasidone with ziprasidone in stable outpatients with schizophrenia (5,16). With the exception of one failed study (see below), efficacy outcomes were consistent in favour of lurasidone 80 mg/day vs. placebo on measures of psychopathology, however, at least two studies also demonstrated efficacy for the doses of 40 and 120 mg/day. In the Japanese dose-finding study, the poster stated that the 40 mg/day produced the most improvement in the overall study population, but that the 80 mg dose appeared more effective in inpatients (12). One study (D1050233) was completed after the manufacturer submitted the New Drug Application to the FDA (27).

Table 1.   Completed lurasidone double-blind randomised trials in patients with schizophrenia
StudyDuration (weeks)Disease stateN randomisedLurasidone dose and NActive comparator dose and NN placeboComments
  1. †Where possible, Ns in this column represent the number of patients available for efficacy analyses and obtained from (32). ‡These four clinical trials were the basis of regulatory approval of lurasidone by the US Food and Drug Administration (2). BPRS, Brief Psychiatric Rating Scale; CGI-I, Clinical Global Impression-improvement; CGI-S, Clinical Global Impression-severity; LOCF, last observation carried forward; MADRS, Montgomery-Asberg Depression Rating Scale; MMRM, mixed model repeated measures; NSA, Negative Symptom Assessment Scale; OC, observed case; PANSS, Positive and Negative Syndrome Scale

D1050006‡ (5,19,20,22,32)6Acute schizophrenia14940 [49] and 120 mg/day [47]None50Phase II. Lurasidone 40 and 120 mg/day were each statistically significantly superior to placebo at end-point in mean changes from baseline for the BPRS total (placebo, −4.3; 40 mg, −10.0; 120 mg, −11.3), CGI-S score (placebo, 0; 40 mg, −0.7; 120 mg, −0.8), CGI-I score (placebo, 4.0; 40 mg, 3.2; 120 mg, 3.0) and between lurasidone 120 mg and placebo for PANSS total score (placebo, −5.8; 40 mg, −14.1; 120 mg, −17.4). No extension study was available.
D1050049 (5)6Acute schizophrenia35320, 40 and 80 mg/day (total N = 209)Haloperidol 10 mg/day [72]72Phase II. Neither haloperidol 10 mg/day, nor lurasidone 20, 40, or 80 mg/day separated statistically from placebo on the major outcomes (BPRS, PANSS, CGI-S) and thus this is a failed study. Patients were eligible for a 6-month extension (98 enrolled).
D1050196‡ (3,5,15,17,20–22,32)6Acute schizophrenia18080 mg/day [90]None90Phase II. This is the only study published in a peer-reviewed journal at this time. Lurasidone 80 mg/day was statistically significantly superior to placebo on the BPRS (OC change from baseline: placebo, −9.7; 80 mg, −14.7; LOCF: placebo, −4.2; 80 mg, −8.9), PANSS total score (LOCF: placebo, −5.5; 80 mg, −14.1), PANSS positive subscale (LOCF: placebo, −1.7; 80 mg, −4.3), negative subscale (LOCF: placebo, −1.3; 80 mg, −2.9) general psychopathology subscale (LOCF: placebo, −2.7; 80 mg, −7.0), and CGI-S (LOCF: placebo, −0.2; 80 mg, 0.6). Other evidence of superiority was noted for the PANSS cognitive component, PANSS depression, and MADRS. Mean point reduction on the MADRS for patients with a baseline MADRS of at least 12 was 6.1 for lurasidone vs. 2.7 for placebo (LOCF). Effects of lurasidone vs. placebo on the BPRS, PANSS total score and CGI-S were significant starting at 3 days post randomisation. A 12-month extension study was available (61 enrolled).
D1050229‡, ‘PEARL 1’ (15,25,27,32)6Acute schizophrenia49640 [122], 80 [119], and 120 mg/day [124]None127Phase III. Limited information is publically available. Lurasidone 80 mg/day, but not 40 or 120 mg/day, was statistically significantly superior to placebo on the PANSS total score and CGI-S (both LOCF). An open-label extension study with lurasidone 40–120 mg/day was available.
*D1050231‡, ‘PEARL 2’ (10,15,26,27,32)6Acute schizophrenia47340 [119] and 120 mg/day [118]Olanzapine 15 mg/day [122]114Phase III. Lurasidone 40 and 120 mg/day were each statistically significantly superior to placebo at 6 weeks on the PANSS total score (MMRM: placebo, −16; 40 mg, −25.7; 120 mg, −23.6), PANSS positive subscale (MMRM: placebo, −5.4; 40 mg, −7.7; 120 mg, −7.5), PANSS negative subscale (MMRM: placebo, −3.6; 40 mg, −6.0; 120 mg, −5.2), CGI-S (MMRM: placebo, −1.1; 40 mg, −1.5; 120 mg, −1.4). Olanzapine 15 mg/day also produced significantly greater improvements than placebo on the PANSS total score (−28.7), PANSS positive subscale (−9.3), PANSS negative subscale (−6.2), and CGI-S (−1.5). An open-label extension study with lurasidone 40–120 mg/day for up to 6 months was available; results demonstrated maintenance of clinical effects.
D1050233, ‘PEARL 3’ (25,27)6Acute schizophrenia48880 and 160 mg/dayQuetiapine XR 600 mg/dayNot availablePhase III. Limited information is publically available pending presentation at a scientific meeting. Lurasidone 80 and 160 mg/day were each statistically significantly superior to placebo on the PANSS total score and the CGI-S. Quetiapine XR 600 mg/day also produced statistically significant improvement relative to placebo. A double-blind flexible dose extension study with lurasidone 40–160 mg/day and quetiapine XR 200–800 mg/day was available.
(12)8Inpatients and outpatients with schizophrenia19520 [65], 40 [72], and 80 mg/day [58]NoneNoneDescribed as a dose-finding study. Conducted in Japan. No washout period took place prior to starting lurasidone but if the prior antipsychotic exceeded the equivalent of haloperidol 12 mg/day, this was reduced to 12 mg/day prior to randomisation. LOCF analyses of the PANSS and BPRS demonstrated significant improvement from baseline for the 40 or 80 mg/day patient groups, with the 40 mg/day group being superior to the 20 mg/day group on the BPRS. A 44-week extension was available.
D1050254 (5,16)3Stable outpatients with schizophrenia301120 mg/day [150]Ziprasidone 160 mg/day [151]NoneDescribed as a Phase Ib safety study when described in a review (5), and as a Phase II study in a poster (16). Ns are as provided for the efficacy analyses (16). Overall efficacy as measured by the change in PANSS total score was similar for both treatment groups at week 3 (MMRM: lurasidone, −6.2; ziprasidone, −4.5), although scores at week 1 showed an advantage for lurasidone (MMRM: lurasidone, −4.1; ziprasidone, −1.6). There may be an advantage for lurasidone in terms of the PANSS negative subscale at end-point (LOCF: lurasidone, −1.3; ziprasidone, −0.6), but on MMRM at 3 weeks this was not significant.

In an integrated analysis from four positive studies (studies D1050006, D1050196, D1050229 and D1050231), change in the Positive and Negative Syndrome Scale (PANSS) total score using mixed model repeated measures (MMRM) demonstrated similar time of onset (in the first 3–7 days) and similar trajectory of improvement across the 40–120 mg/day range (32). Although there were no differences suggesting a dose–response relationship in Clinical Global Impression-severity (CGI-S) at end-point within the 40–120 mg/day dose range, change in CGI-S scores demonstrated an earlier onset (by Day 7) for the 80–120 mg/day doses. (32). Five standard PANSS factors were also evaluated using MMRM: positive, negative, disorganised thought, depression/anxiety and hostility (13). Lurasidone was significantly superior to placebo in improving all five PANSS factor scores. Week 6 change scores were significant compared with placebo among patients treated with 40, 80 and 120 mg/day on the PANSS positive factor (−7.92, −8.48, −8.25, respectively), negative factor (−5.59, −4.96, −5.21), disorganised thought (−4.86, −5.10, −5.22), hostility (−2.33, −2.58, −2.87) and depression/anxiety (−3.14, −3.23,−3.01). No clear dose–response relationship could be identified. Generalisability may be limited to the type of patient enrolled in these clinical trials, with mean age noted to range from approximately 37 to 41 years, depending on the treatment arm, with 69–79% being men, and 32–52% being White (32). Mean baseline PANSS total score ranged from approximately 93 to 97, depending on the treatment arm, and mean baseline CGI-S ranged from 4.8 to 5.0 (32).

There is one failed trial (study D1050049) where neither lurasidone (20, 40, or 80 mg/day) nor haloperidol 10 mg/day separated from placebo on the psychopathology outcome measures (5). Placebo response in clinical trials can be substantial, diminishing the possibility to observe a drug-placebo difference. In an effort to understand drug-placebo differences observed in clinical trials, the manufacturer sent out questionnaires to Principal Investigators at 40 investigative sites in the US that had participated in Phase III clinical trials of lurasidone in schizophrenia (9). The majority of study sites in the sample recruited subjects from outside their own practice. A larger lurasidone-placebo effect size and smaller placebo response were observed in study sites that had greater proportions of patients with no prior research study experience and lower proportions of patients with a history of response to high doses of antipsychotic agents.

In the extant study directly comparing lurasidone 120 mg/day with another antipsychotic, lurasidone’s efficacy among stable outpatients with schizophrenia was found to be similar to that for ziprasidone 160 mg/day (16). Additional cognitive outcomes were systematically assessed. Subjects were tested at baseline and end-point with a battery of cognitive tests and also received an interview-based assessment of cognitive functioning (18). There were no between-group differences in these ratings, but on testing at week 3 lurasidone demonstrated significant within group-improvement from baseline whereas ziprasidone did not.

The long-term efficacy profile of lurasidone has been assessed in both uncontrolled and controlled clinical trials (see Table 2), but results have not been publically disclosed to date.

Available categorical response rates and the corresponding NNTs vs. placebo are contained in Table 3. The information available is limited to lurasidone and placebo; no data are available regarding the responder rates for the active comparators when present. The definition for response differs from study to study, making comparisons difficult. In general, the lower the threshold for response, the more robust the NNT (i.e. NNT was 3, 4 or 6 for a reduction of psychopathology of ≥ 20% from baseline). A threshold of ≥ 30% improvement from baseline on the PANSS total score resulted in statistically significant NNT estimates for lurasidone 80 mg/day in study D1050229 (NNT 8) and lurasidone 40 mg/day in study D1050231 (NNT 7). Remission as an outcome was reported in study D1050006, with higher remission rates observed for lurasidone 120 and 40 mg/day (34% and 31%, respectively) compared with placebo (6.1%) at end-point, with remission defined as a score of mild or less on 8 core PANSS items. NNT was four for both doses of lurasidone, with 95% CIs of 3–10 and 2–8 for 40 and 120 mg/day, respectively (19).

Table 3.   Responder rates and number needed to treat for lurasidone vs. placebo and 95% confidence intervals for the four short-term trials used for regulatory approval
StudyPlaceboLurasidone
40 mg/day80 mg/day120 mg/day
Responders (%)Responders (%)NNT (95% CI)Responders (%)NNT (95% CI)Responders (%)NNT (95% CI)
  1. *Responder defined as a decrease in the Brief Psychiatric Rating Scale total score of at least 20% from baseline or a Clinical Global Impressions – Improvement score of 2 or less. †Responder defined as a decrease in the Positive and Negative Syndrome Scale total score of at least 20% from baseline. ‡Responder defined as a decrease in the Positive and Negative Syndrome Scale total score of at least 30% from baseline. CI, confidence interval; NNT, number needed to treat; ns, not significant; Numerators were estimated by multiplying the stated percentage by the provided denominator, hence the NNH and 95% CI provided here can differ slightly from calculations made using actual numbers of patients experiencing the event.

D1050006 (5)*22.255.33 (2–7)50.04 (3–11)
D1050196 (3)†26.744.46 (4–26)
D1050229 (25)‡384613 (ns)528 (4–63)509 (ns)
D1050231 (26)‡38537 (4–43)4712 (ns)

Safety in clinical trials

Safety of lurasidone in individual placebo-controlled trials has been disclosed in a study report (3), and several posters (10,12,20,29). An integrated analysis is also available (14). Data were pooled from five double-blind, placebo-controlled studies of patients who met Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV criteria for schizophrenia with an acute exacerbation in order to summarise the key safety parameters of lurasidone (14). Included were 1004 subjects receiving lurasidone, 72 receiving haloperidol, 122 receiving olanzapine and 455 receiving placebo. NNH is provided in Table 4a for lurasidone vs. placebo for four adverse events with an incidence ≥ 10% and at least twice the rate of placebo. From an older poster of a pooled safety analysis that included only four of the short-term trials, data are provided by dose and the respective NNH estimates are provided in Table 4b (30). Akathisia appears dose-related, with a NNH of 6 for lurasidone 120 mg/day, compared with NNHs of 15 and 10 for 40 and 80 mg/day, respectively (30). Somnolence also appears dose-related, with a NNH of 12 for lurasidone 120 mg/day, compared with NNHs of 25 and 17 for 40 and 80 mg/day, respectively (30).

Table 4.   (a) Percentage of subjects with common adverse reactions and number needed to harm vs. placebo and 95% confidence intervals in acute schizophrenia trials.* (b) Percentage of subjects with common adverse reactions and number needed to harm vs. placebo and 95% confidence intervals in acute schizophrenia trials, by lurasidone dose†
Adverse eventPlacebo (N = 455)Lurasidone (pooled) (N = 1004)
Rate (%)Rate (%)NNH (95% CI)
(a)
Akathisia3.315.09 (7–12)
Nausea3.912.013 (10–19)
Sedation5.511.916 (11–29)
Somnolence4.610.717 (12–30)
Adverse eventPlacebo (N = 339)Lurasidone
40 mg/day (N = 241)80 mg/day (N = 282)120 mg/day (N = 173)
Rate (%)Rate (%)NNH (95% CI)Rate (%)NNH (95% CI)Rate (%)NNH (95% CI)
  1. *Source data from (14), studies D1050006, D1050049, D1050196, D1050229, D1050231. Numerators were estimated by multiplying the stated percentage by the provided denominator, hence the NNH and 95% CI provided here can differ slightly from calculations made using actual numbers of patients experiencing the event. †Source data from (30), studies D1050006, D1050049, D1050196, D1050229. CI, confidence interval; NNH, number needed to harm; ns, not significant.

  2. Numerators were estimated by multiplying the stated percentage by the provided denominator, hence the NNH and 95% CI provided here can differ slightly from calculations made using actual numbers of patients experiencing the event. A NNH of ∞ occurs when the rates for the outcome of interest are the same for the two interventions being compared.

(b)
Akathisia4.111.215 (9–39)14.910 (7–17)21.46 (5–10)
Nausea6.510.824 (ns)12.816 (10–63)6.5∞ (ns)
Sedation6.211.220 (11–381)12.816 (9–52)11.619 (ns)
Somnolence4.78.725 (ns)10.617 (10–61)13.312 (8–33)

Discontinuations caused by adverse events were 8% in the lurasidone group vs. 4% on placebo (NNH 25, 95% CI 16–65), with no dose-related differences in the 40–120 mg/day range (14). Measures of extrapyramidal symptoms were similar to placebo for lurasidone and olanzapine. On metabolic parameters, the median end-point change (mg/dl), in the lurasidone, olanzapine, haloperidol and placebo groups, respectively, was −8.0, +9.0, −8.0 and −10.0 for total cholesterol; −5.0, +25.0, −3.0 and −7.0 for triglycerides; and 0.0, +4.0, +2.0 and +1.0 for glucose. Weight gain (≥ 7%) was more frequent with olanzapine (34.4%) compared with lurasidone (5.6%), haloperidol (4.2%) or placebo (4.0%), with NNH vs. placebo of four (95% CI 3–5) for olanzapine, 63 (ns) for lurasidone, and 500 (ns) for haloperidol. Mean QTcF change was minimal for lurasidone and placebo (+1.5 vs. +1.9 ms). Mean end-point change in prolactin (ng/ml) was +1.1 on lurasidone, +3.7 on olanzapine, +8.5 on haloperidol and −0.5 on placebo.

The comparative safety and efficacy of lurasidone vs. ziprasidone over a 3-week period was tested in stable adult outpatients diagnosed with DSM-IV schizophrenia or schizoaffective disorder (16). Subjects had no hospitalisation or acute exacerbation of psychosis in the previous 3 months. After a 1–3 day single-blind, placebo run-in period, patients were randomised to 21 days of treatment with lurasidone (80 mg/day for 3 days, followed by 120 mg/day) or ziprasidone (40 mg bid twice daily for 3 days, followed by 80 mg bid twice daily). This was double-blinded with lurasidone patients receiving a matched placebo for the second dose of the day. Baseline PANSS total scores were approximately 69–70. See Table 5 for NNH of the adverse events where incidence was greater than 5%. The only statistically significant difference was an advantage for lurasidone over ziprasidone in terms of sedation, with a NNH of −16 (i.e. a NNT of +16). The median end-point change in weight and metabolic variables was essentially similar for both lurasidone and ziprasidone. No patients in either treatment group had clinically significant elevations in the ECG QT interval, although mean change in msec was +0.3 for lurasidone and +3.3 for ziprasidone.

Table 5.   Percentage of subjects with common adverse reactions and number needed to harm for lurasidone vs. ziprasidone and 95% confidence intervals*
Adverse eventZiprasidone 160 mg/day (N = 151)Lurasidone 120 mg/day (N = 150)
Rate (%)Rate (%)NNH (95% CI)
  1. *Source data from (16). CI, confidence interval; NNH, number needed to harm; ns, not significant.

  2. Positive numbers for NNH represent a disadvantage for lurasidone vs. ziprasidone; negative numbers for NNH represent an advantage for lurasidone vs. ziprasidone.

Insomnia9.310.772 (ns)
Vomiting4.08.025 (ns)
Nausea4.67.338 (ns)
Headache4.66.750 (ns)
Somnolence9.96.7−31 (ns)
Sedation11.34.7−16 (−8 to −191)
Akathisia6.63.3−31 (ns)
Dizziness6.62.7−26 (ns)
At least 1 adverse event6657−12 (ns)

Additional information about the safety and tolerability of lurasidone 160 mg/day compared with 80 mg/day, quetiapine XR 600 mg/day and placebo will become available from a 6-week Phase III study (D1050233) (27). Preliminary information is that lurasidone was well tolerated with a lower discontinuation rate than placebo. The most common adverse events reported for the lurasidone group (greater than 5% and at least twice the rate of placebo) were akathisia, nausea, parkinsonism, dizziness and somnolence.

Efficacy vs. tolerability

To be effective in real-world clinical practice, medications must be efficacious enough, tolerable enough and the patient has to take it consistently enough. A proxy measure for effectiveness can be whether or not patients stop their medication. In a clinical trial this decision may be impacted by a strong desire by all parties to continue participation until the end of the study, but there are often instances when discontinuation is required because of lack of efficacy or because of an adverse event. The thresholds for discontinuing a patient from a clinical trial because of these events may differ considerably from what would be expected in clinical practice and does differ from patient to patient. Nonetheless it can be useful to calculate NNT for discontinuation because of lack of efficacy and NNH for discontinuation because of an adverse event, with the assumption that lurasidone is superior to placebo for efficacy and inferior to placebo in terms of tolerability. Table 6 includes the patient disposition data from an integrated analysis from four positive studies of lurasidone (32), and the NNT and NNH for the two reasons for discontinuation that are of interest. The potential effect of the omission of the failed trial (study D1050049) on the analysis is not known.

Table 6.   Patient disposition from an integrated analysis from four positive studies (studies D1050006, D1050196, D1050229, and D1050231)* and number needed to treat or harm with 95% confidence intervals vs. placebo for discontinuation because of lack of efficacy or for an adverse event, and likelihood to be helped or harmed
 PlaceboLurasidone
40 mg/day80 mg/day120 mg/day
  1. *Source data from (32). CI, confidence interval; LHH, likelihood to be helped or harmed; NNH, number needed to harm; NNT, number needed to treat; ns, not significant.

  2. Numerators were estimated by multiplying the stated percentage by the provided denominator, hence the NNH and 95% CI provided here can differ slightly from calculations made using actual numbers of patients experiencing the event.

Randomised455360282291
Completed53%57%60%59%
Discontinued47%43%40%41%
 Discontinued because of lack of efficacy24%18%9%11%
 Discontinued because of adverse event4%8%7%9%
 Discontinued because of other reasons19%17%24%21%
NNT for discontinuation because of lack of efficacy (95% CI)17 (9–239)7 (5–11)8 (6–13)
NNH for discontinuation because of adverse event (95% CI)25 (14–150)34 (ns)20 (12–80)
LHH1.54.92.5

In each comparison of NNT and NNH the benefit with lurasidone (avoidance of discontinuation because of lack of efficacy) is encountered more frequently than the harm (discontinuation because of an adverse event). Likelihood to be helped or harmed (LHH) can be calculated, and is simply the ratio of NNH to NNT (44–46). The optimal dose appears to be lurasidone 80 mg/day where the NNH for discontinuation because of an adverse event is 34 (and actually not statistically significant), and the NNT for avoidance of discontinuation because lack of efficacy is seven The LHH of 4.9 for lurasidone 80 mg/day means that lurasidone 80 mg/day compared with placebo is 4.9 more likely to lead to an avoidance of discontinuation because of lack of efficacy than discontinuation because of an adverse event. Unaccounted for in this analysis are the potential differences in actual symptom relief among the different doses of lurasidone, but as noted earlier no dose–response relationship was noted for PANSS total scores or PANSS factor scores (13,32).

Potential drug–drug interactions

Potential pharmacokinetic interactions of lurasidone with other medications were described in a poster outlining the results of seven studies (33), in product labelling (2), and in a review (5). Concomitant administration of lurasidone 10 mg and ketoconazole 400 mg (a potent CYP3A4 inhibitor) in male patients with schizophrenia, schizophreniform, or schizoaffective disorder resulted in a ninefold increase in lurasidone AUC and an almost sevenfold increase in Cmax. Diltiazem, a moderate CYP3A4 inhibitor, when administered at a dose of 240 mg with lurasidone 20 mg in healthy subjects, resulted in approximately twofold increases in lurasidone AUC and Cmax. Rifampin, a potent CYP3A4 inducer, when administered at a dose of 600 mg with lurasidone 40 mg in healthy subjects, resulted in a decrease in lurasidone AUC and Cmax by approximately 85%.

Lurasidone itself can exhibit weak CYP3A4 inhibition, and resulted in a negligibly small increase in midazolam (a CYP3A4 substrate) exposure when midazolam 5 mg was given to patients with schizophrenia, schizophreniform or schizoaffective disorder administered lurasidone 120 mg. As per the product label, lurasidone increased Cmax and AUC of midazolam by approximately 21% and 44%, respectively (2).

P-glycoprotein transport effects were tested in patients with schizophrenia, schizophreniform or schizoaffective disorder administered lurasidone 120 mg with digoxin 0.25 mg. No clinically relevant interaction was observed, suggesting that lurasidone is not a substrate of the P-glycoprotein transporter. As per the product label, lurasidone increased Cmax and AUC of digoxin by approximately 9% and 13%, respectively (2).

No interaction was observed when lurasidone 120 mg was administered with lithium 600 mg bid twice daily in patients with schizophrenia, schizophreniform or schizoaffective disorder.

No changes in ethinyl estradiol and norelgestromin plasma exposure were observed when this combination contraceptive pill was administered with lurasidone 40 mg in healthy female subjects.

Additional safety information as noted in the product label

Class level warnings and precautions are listed, including increased mortality in elderly patients with dementia-related psychosis (boxed warning), cerebrovascular adverse reactions including stroke, neuroleptic malignant syndrome, tardive dyskinesia, metabolic changes, hyperprolactinemia, leucopenia/neutropenia/agranulocytosis, orthostatic hypotension and syncope, seizures, potential for cognitive and motor impairment, body temperature regulation, suicide, dysphagia and use in patients with concomitant illness (2).

Somnolence (including hypersomnia, hypersomnolence, sedation and somnolence) in short-term trials was reported in 224/1004 (22.3%) patients receiving lurasidone vs. 45/455 (9.9%) receiving placebo (2), yielding a NNH of 9 (95% CI 7–12). This was dose-related, with somnolence reported in 77/291 (26.5%) for patients receiving lurasidone 120 mg/day (NNH vs. placebo 6, 95% CI 5–10). See also Table 7b.

Table 7.   (a) Percentage of subjects with common adverse reactions and number needed to harm vs. placebo and 95% confidence intervals in acute schizophrenia trials.* (b) Percentage of subjects with akathisia or somnolence and number needed to harm vs. placebo and 95% confidence intervals in acute schizophrenia trials, by lurasidone dose.† (c) Percentage of subjects with extrapyramidal symptoms and number needed to harm vs. placebo and 95% confidence intervals in acute schizophrenia trials, by lurasidone dose‡
Adverse eventPlacebo (N = 455)Lurasidone (pooled) (N = 1004)
Rate (%)Rate (%)NNH (95% CI)
(a)
Somnolence10229 (7–12)
Akathisia3159 (7–11)
Nausea61217 (12–33)
Parkinsonism51117 (12–32)
Agitation3634 (20–118)
Adverse eventPlacebo (N = 455)Lurasidone
20 mg/day (N = 71)40 mg/day (N = 360)80 mg/day (N = 282)120 mg/day (N = 291)
Rate (%)Rate (%)NNH (95% CI)Rate (%)NNH (95% CI)Rate (%)NNH (95% CI)Rate (%)NNH (95% CI)
(b)
Akathisia3634 (ns)1113 (9–23)159 (7–14)226 (5–8)
Somnolence101520 (ns)1911 (8–24)238 (6–14)266 (5–10)
Adverse eventPlacebo (N = 455)Lurasidone
20 mg/day (N = 71)40 mg/day (N = 360)80 mg/day (N = 282)120 mg/day (N = 291)
Rate (%)Rate (%)NNH (95% CI)Rate (%)NNH (95% CI)Rate (%)NNH (95% CI)Rate (%)NNH (95% CI)
  1. *Source data from (2). CI, confidence interval; NNH, number needed to harm.

  2. Somnolence includes hypersomnia, hypersomnolence, sedation, and somnolence. Except for somnolence, numerators were estimated by multiplying the stated percentage by the provided denominator, hence the NNH and 95% CI provided here can differ slightly from calculations made using actual numbers of patients experiencing the event.

  3. †Source data from (2). CI, confidence interval; NNH, number needed to harm; ns, not significant.

  4. Somnolence includes hypersomnia, hypersomnolence, sedation and somnolence. Except for somnolence rates for placebo and lurasidone 120 mg/day, numerators were estimated by multiplying the stated percentage by the provided denominator, hence the NNH and 95% CI provided here can differ slightly from calculations made using actual numbers of patients experiencing the event.

  5. ‡Source data from (2). CI, confidence interval; EPS, extrapyramidal symptoms; NNH, number needed to harm; ns, not significant.

  6. For akathisia see Table 8. Numerators were estimated by multiplying the stated percentage by the provided denominator, hence the NNH and 95% CI provided here can differ slightly from calculations made using actual numbers of patients experiencing the event. A negative NNH represents an advantage of lurasidone vs. placebo.

(c)
All EPS events910100 (ns)247 (5–11)266 (5–9)394 (3–5)
All EPS events, excluding akathisia/restlesness56100 (ns)1313 (9–26)1117 (10–55)226 (5–9)
Dystonia0.70−143 (ns)4.229 (18–78)4.626 (16–75)6.518 (12–35)
Parkinsonsim56100 (ns)1020 (12–77)750 (ns)179 (6–14)
Restlessness21−100 (ns)450 (ns)1−100 (ns)3100 (ns)

Additional information is available in product labelling regarding categorical changes in fasting glucose, cholesterol and triglycerides by lurasidone dose. These rates are listed in Table 8, together with NNH for lurasidone vs. placebo. Changes in weight have already been discussed. The data are consistent with a benign metabolic profile for lurasidone.

Table 8.   Categorical metabolic outcomes and number needed to harm for lurasidone vs. placebo and 95% confidence intervals*
EventPlaceboLurasidone
20 mg/day40 mg/day80 mg/day120 mg/day
RateRateNNH (95% CI)RateNNH (95% CI)RateNNH (95% CI)RateNNH (95% CI)
  1. *Source data from (2). CI, confidence interval; NNH, number needed to harm; ns, not significant.

  2. A negative NNH represents an advantage of lurasidone vs. placebo.

Serum glucose ≥ 126 mg/dl34/397 (8.6%)7/60 (11.7%)33 (ns)47/328 (14.3%)18 (10–93)24/241 (10.0%)72 (ns)26/260 (10.0%)70 (ns)
Total cholesterol ≥ 240 mg/dl23/350 (6.6%)8/58 (13.8%)14 (ns)21/287 (7.3%)135 (ns)15/216 (6.9%)269 (ns)9/238 (3.8%)−36 (ns)
Triglycerides ≥ 200 mg/dl39/312 (12.5%)7/49 (14.3%)56 (ns)37/264 (14.0%)66 (ns)17/196 (8.7%)−27 (ns)22/209 (10.5%)−51 (ns)

The proportion of patients with prolactin elevations ≥ 5-times the upper limit of normal was 3.6% for lurasidone vs. 0.7% for placebo (NNH 35, 95% CI 24–68). For women, this occurred in 8.3% for lurasidone vs. 1% for placebo (NNH 14, 95% CI 9–29). For men, this occurred in 1.9% for lurasidone vs. 0.6% for placebo (NNH 77, ns). Product labelling also notes that increases in prolactin concentration are dose-related, with median change in prolactin (ng/ml) from baseline among all patients to be −1.1, +0.3, +1.1 and + 3.3 for lurasidone 20, 40, 80 and 120 mg/day, respectively, compared with −0.6 for placebo (2). For women, median changes were −0.7, −0.9, +2.0 and +6.7 for lurasidone 20, 40, 80 and 120 mg/day, respectively, compared with −1.5 for placebo (2). For men, median changes were −1.2, +0.5, +0.9 and +3.1 for lurasidone 20, 40, 80 and 120 mg/day, respectively, compared with −0.5 for placebo (2).

Orthostatic hypotension occurred in 4/1004 of patients receiving lurasidone vs. 1/455 receiving placebo, yielding a NNH of 560 (ns). Syncope occurred in 1/1004 patients receiving lurasidone and in none receiving placebo giving a NNH of 1004 (ns). Vital sign change of ≥ 20 mmHg decrease in systolic blood pressure and ≥ 10 bpm increase in pulse from sitting to standing or supine to standing positions in short-term clinical trials occurred with a frequency of 0.9% with placebo and 0.8% with lurasidone 40 mg, 1.4% with 80 mg and 1.7% with 120 mg (NNH −1000, 200 and 125, respectively, all ns) (2).

Seizures occurred in 1/1004 patients receiving lurasidone compared with 1/455 receiving placebo [NNH –833 (ns)] (2).

The incidence of treatment-emergent suicidal ideation was 6/1004 for patients receiving lurasidone compared with 2/455 receiving placebo [NNH 633 (ns)]. No suicide attempts or completed suicides were reported in these short-term placebo-controlled trials (2).

The product label describes tolerability and safety outcomes from uncontrolled longer-term studies (primarily open-label extension studies). In these, lurasidone was associated with a mean change in glucose (mg/dl) of +1.6 at week 24 (N = 186), +0.3 at week 36 (N = 236) and +1.2 at week 52 (N = 244) (2). Mean changes in total cholesterol (mg/dl) and triglycerides (mg/dl) were −4.2 (N = 186) and −13.6 (N = 187) at week 24, −1.9 (N = 238) and −3.5 (N = 238) at week 36 and −3.6 (N = 243) and −6.5 (N = 243) at week 52, respectively (2). Mean changes in weight (kg) were −0.38 at week 24 (N = 531), −0.47 at week 36 (N = 303) and −0.71 at week 52 (N = 244) (2). Median changes in prolactin (ng/ml) were −1.9 at week 24 (N = 188), −5.4 at week 36 (N = 189) and −3.3 at week 52 (N = 243) (2).

Product labelling notes that the most commonly observed adverse reactions (incidence ≥ 5% and at least twice the rate of placebo) in patients treated with lurasidone in short-term trials were somnolence, akathisia, nausea, parkinsonism and agitation (2). These are presented in Table 7a and these data differ slightly from that in Table 4a. Notably somnolence is more broadly defined in product labelling. Table 7b provides the rates of akathisia and somnolence by dose (2).

Additional detailed information is available regarding extrapyramidal symptoms observed in the short-term trials. The incidence of reported extrapyramidal symptom-related events, excluding akathisia and restlessness, was 14.7% for lurasidone vs. 5.1% for placebo (NNH 11, 95% CI 8–16). Akathisia is as described above and data shown in Table 7b. The greatest frequency of parkinsonism and dystonia occurred with 120 mg/day, the highest dose of lurasidone in the short-term trials assessed by the FDA (Table 7c). On rating scale scores for extrapyramidal symptoms, akathisia and dyskinesias, the mean change from baseline for lurasidone-treated patients was similar to those receiving placebo, with the exception of the Barnes Akathisia Scale (BAS) where a small increase in the global score (0.2) was observed for lurasidone and no change for placebo (2). The proportion of patients who shifted from normal to abnormal on the BAS was 16% for lurasidone compared with 7.6% for placebo (NNH 12) (2). The proportion of patients who shifted from normal to abnormal on the Simpson Angus Scale for extrapyramidal symptoms was 5.3% for lurasidone compared with 2.5% for placebo (NNH 36). Dystonia occurred in 4.7% of lurasidone patients, compared with 0.7% for placebo (NNH 25, 95% CI 19–41), with rates differing by dose (Table 7c) (2). Seven patients (0.7%, 7/1004) discontinued clinical trials because of dystonic events with four receiving lurasidone 80 mg/day and three receiving 120 mg/day (2).

Pooled data reveal that 94/1004 (9.4%) of patients receiving lurasidone discontinued treatment because of an adverse reaction(s), compared with 27/455 (5.9%) of those receiving placebo, yielding a NNH of 30 (95% CI 16–165) (2). There were no adverse reactions associated with discontinuation in subjects treated with lurasidone that were at least 2% and at least twice the placebo rate (2).

No clinically important differences were observed between lurasidone and placebo in changes in routine haematology, urinalysis or serum chemistry (2). In the short-term studies, the mean change from baseline in creatinine was 0.06 mg/dl for lurasidone vs. 0.03 mg/dl for placebo. A creatinine shift from normal to high (≥ 1.1 to ≥ 1.3 mg/dl depending on the definition for each study) occurred in 30/977 (3.1%) of lurasidone patients compared with 6/439 (1.4%) for placebo, yielding a NNH of 59 (95% CI 31–586). Mean changes in aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were similar for lurasidone and placebo. Proportions with elevations ≥ 3-times the upper limit of normal were 0.8% for lurasidone for each transaminase and 0.9% for AST and 1.1% for ALT in patients receiving placebo (2).

There is no warning or precaution for lurasidone regarding ECG QT interval prolongation. No postbaseline ECG QT prolongations exceeding 500 ms were reported in patients treated with lurasidone. Moreover, within a subset of patients defined as having an increased cardiac risk, no potentially important changes in ECG parameters were observed (2). No cases of torsade de pointes or other severe cardiac arrhythmias were observed in the premarketing clinical programme. In a dedicated QT study involving 87 clinically stable patients with schizophrenia or schizoaffective disorder given lurasidone 120 or 600 mg/day or ziprasidone 160 mg daily, no patients receiving lurasidone experienced a QTc increase greater than 60 ms from baseline, nor did any patients experience a QTc greater than 500 ms (2).

Administration and optimal dose

Lurasidone tablets are available in 40 and 80 mg strengths. The product label recommends a starting dose of 40 mg/day administered once daily with food (≥ 350 calories) (2). The maximum recommended dose is 80 mg/day (2). There does not appear to be a therapeutic dose–response relationship within the range of 40–120 mg/day. Doses above 80 mg/day do not appear to confer added benefit (see also Figure 1) and may be associated with a dose-related increase in certain adverse reactions (2), in particular somnolence and extrapyramidal side effects. For example NNH for any extrapyramidal symptom event for lurasidone 120 mg/day vs. placebo is 4 (95% CI 3–5), as noted in Table 7c. Dose adjustments are not required on the basis of age, gender or race (2). Dose adjustment is recommended in moderate and severe renal or hepatic impairment patients, and in those cases lurasidone dose should not exceed 40 mg/day (2). As per product label recommendations, the dose should not exceed 40 mg/day when lurasidone is co-administered with a moderate CYP3A4 inhibitor such as diltiazem (2). Administration with a strong CYP3A4 inhibitor such as ketoconazole is contraindicated (2). Likewise, co-administration with strong CYP3A4 inducers such as rifampin is also contraindicated (2).

Figure 1.

 Least Squares Mean difference from placebo in change from baseline in the four short-term trials used for regulatory approval. Source data from (2). Studies D1050006 and D1050196 used the Brief Psychiatric Rating Scale as the primary outcome measure; Studies D1050229 and D1050231 used the Positive and Negative Syndrome Scale as the primary outcome measure.

Maximum tolerated dose (MTD) for lurasidone was determined using two linked studies that enrolled patients with stable schizophrenia (11). MTD was defined as the dose where at least four patients at a given dosage level reported more than one moderate-or-severe adverse event related to lurasidone, or the dose at which greater than one patient reported a serious adverse event at least possibly related to lurasidone. The first study tested single-blind single doses of lurasidone 120, 140 and 160 mg, followed by placebo in 23 patients. The second study tested 160, 200, 240, 280, 320, 400 and 520 mg using a 7-day double-blind, placebo-controlled design. There was a separate cohort of patients who were randomised to a titration group where lurasidone was increased from 200 to 600 mg/day over an 8-day period. A total of 68 patients participated in the second study, with 52 receiving lurasidone and 16 receiving placebo. MTD was determined to be 400 mg/day, with the moderate-to severe adverse events most frequently encountered at 520 mg/day being akathisia (five patients), sedation (three patients) and restlessness (two patients). The most commonly observed adverse events at doses of 120–160 mg were anxiety, insomnia, restlessness and fatigue. Cmax and AUC generally increased with increasing dose, but were less than dose-proportional with high doses, in contrast to the linear kinetics observed in the 20–160 mg/day range (2). No clinically relevant ECG findings with lurasidone were encountered, nor were clinically significant abnormalities observed in laboratory measures, vital signs or physical examination.

Studies in bipolar depression

Studies of lurasidone for bipolar depression have been registered at http://www.clinicaltrials.gov (Table 2) and overviews of the clinical trial programme have been presented (25). Further information is not publically available at present.

Discussion

Available so far are the results of carefully conducted registration trials that enrolled subjects who fulfilled strict inclusion/exclusion criteria. Such patients can be quite different from persons with severe mental disorders encountered in routine clinical practice. For example registration trials ordinarily exclude patients with current comorbid substance use disorders, untreated medical conditions, current suicidality or who require concomitant use of other psychotropic agents (other than certain hypnotics or anxiolytics). Known non-responders to antipsychotics and those recently treated with clozapine are also usually excluded from participating. Thus, pragmatic clinical trials that are more generalisable will help place lurasidone into clinical perspective for its use in the ‘real world.’

A limitation of this review is that the majority of the data was obtained from abstracts and posters. These sources have not been subject to the peer review process of a medical journal, and the results presented in these forums may be subject to further quality review and subsequent revision prior to final publication. At present, the product label is the most authoritative source of information regarding lurasidone. It is anticipated that the publication of peer-reviewed study reports will soon occur, as well as the public availability of the FDA’s Drug Approval Package, providing additional data and explanations (47).

Still pending are the complete results from longer-term studies that will clarify the role of lurasidone for the maintenance treatment of patients with schizophrenia. Additional information about optimal dose for different patient subtypes would also be helpful, particularly addressing the question whether or not exceeding 80 mg/day would be a therapeutically useful strategy.

Lurasidone is classified as an atypical or second-generation antipsychotic as it is an antagonist at dopamine D2 and serotonin 5-HT2A receptors, and consequently has a lower propensity for extrapyramidal symptoms compared to first-generation antipsychotics. However, akin to several other second-generation antipsychotics, lurasidone is not entirely free of akathisia or parkinsonism as noted in product labelling. Although the rates of these events remain low, they appear to be dose-related.

When comparing lurasidone with the other second-generation antipsychotics that also have a low propensity for weight gain (48–51), potential advantages become apparent. For example dosing of lurasidone does not necessarily require titration (in contrast to iloperidone and perhaps ziprasidone), lurasidone is approved for once-daily administration (in contrast to ziprasidone, iloperidone and asenapine), there is no concern about possible adverse effects of partial agonism of the dopamine D2 receptor (in contrast to aripiprazole), and there is no effect on the ECG QT interval (in contrast to ziprasidone and iloperidone). A potential disadvantage is the possibility of dose-related extrapyramidal symptoms and akathisia with lurasidone (in contrast to iloperidone), as well as dose-related somnolence (as can be observed with asenapine). It must also be administered with food because of limitations in absorption (in contrast to aripiprazole, asenapine or iloperidone).

Lurasidone’s relative efficacy ranking among the different available antipsychotics is unknown. It is anticipated that meta-analytical techniques will help rank lurasidone as has been done with other antipsychotics (52–54), pending the conduct of appropriately powered head-to-head efficacy studies. This will provide additional information of the effect size of lurasidone for the treatment of psychotic symptoms, such as weighted mean difference in PANSS points when comparing lurasidone with alternatives.

There is considerable heterogeneity in antipsychotic efficacy and tolerability among the available choices for patients with schizophrenia, as well as a large individual variation in treatment response, rendering additional options welcome in treating this difficult to manage life-long chronic disorder (55). Cost may be an impediment with the availability of inexpensive generic versions of risperidone in the US, as well as other generic second-generation antipsychotics in other countries. It is hoped that new medications with entirely different mechanisms of action will one day become available that can more predictably and effectively treat schizophrenia, however, the state of the art today is to carefully match the individual patient to a medication that will optimise efficacy, tolerability and adherence.

Conclusions

Lurasidone is efficacious and well tolerated. Lurasidone’s impact on body weight is minimal, and there are no clinically relevant associated changes observed in measures of blood glucose, cholesterol, triglycerides or the ECG QT interval. Extrapyramidal side effects and akathisia (dose-related) may be encountered, particularly at doses in excess of 80 mg/day. Somnolence appears dose-related as well. It appears that efficacy and tolerability can be optimised by dosing at 80 mg/day. Additional data regarding long-term efficacy and effectiveness will help characterise this new agent when used in maintenance treatment.

Ancillary