Asenapine, Iloperidone and Lurasidone

Critical Appraisal of the Most Recently Approved Pharmacotherapies for Schizophrenia in Adults

William V Bobo

Disclosures

Expert Rev Clin Pharmacol. 2013;6(1):61-91. 

In This Article

Clinical Efficacy

Acute-phase Treatment

Iloperidone A summary of results from four short-term, multi-site, randomized, controlled efficacy studies of iloperidone in adults (aged 18–65 years) with acute schizophrenia is presented in Table 3. The first study by Cutler et al. was a 4-week, Phase III study of 606 patients randomized to fixed doses of iloperidone (12 mg b.i.d.), ziprasidone (160 mg daily) or placebo.[35] Eligible patients had Clinical Global Impression-Severity(CGI–S)[36] scores >4 (indicating at least moderate illness severity) at baseline, a Positive and Negative Syndrome Scale (PANSS)[37] total score of >70, and ratings of >4 on at least two PANSS positive symptom items (hallucinations, delusions, suspiciousness/persecution and conceptual disorganization) at screening and at baseline. Study drugs were titrated to their target doses over 7 days. Iloperidone and ziprasidone resulted in significantly greater improvement in PANSS-total scores, the primary efficacy end point, compared with placebo (Table 3). Similar results were observed for PANSS-positive and -negative subscale, and CGI-S scores. A significantly higher proportion of iloperidone-treated patients were classified as positive treatment responders (Table 3), defined as those who achieved a >20% reduction in PANSS-positive symptom subscale scores from baseline. Neither iloperidone nor ziprasidone treatment resulted in significantly greater improvement in Calgary Depression Scale for Schizophrenia (CDSS)[38] scores compared with placebo.

Another report by Potkin et al. summarized results of three 6-week, Phase III, randomized, double-blind, placebo- and active comparator-controlled trials of identical design (Studies 1, 2 and 3).[39] Results were presented for each study separately (Table 3), and in a pooled data analysis. In all studies, eligible subjects had a diagnosis of schizophrenia and a PANSS total score of >60 at screening and baseline. Randomization occurred following a 3-day placebo run-in period. Study drugs were titrated to target doses over the first 7 days of follow-up. In Study 1, 621 patients were randomly assigned to receive one of three fixed doses of iloperidone (2, 4 or 6 mg b.i.d.), haloperidol (15 mg daily) or placebo.[39] Compared with placebo, there was significantly greater improvement in PANSS total scores with iloperidone 6 mg b.i.d. and with haloperidol (Table 3). Improvements in PANSS total scores with 4 or 8 mg of iloperidone were not significantly greater than with placebo. Iloperidone 6 mg b.i.d. and haloperidol (but not the lower iloperidone doses) were also associated with greater improvement in Brief Psychiatric Rating Scale (BPRS)[40] scores than placebo (Table 3). There was significantly greater improvement in PANSS-positive subscale scores in the haloperidol-treated group (Table 3) and a strong statistical trend-level advantage for iloperidone (p = 0.06) compared with placebo. None of the iloperidone groups resulted in significant separation from placebo on PANSS-negative subscale scores.

In Study 2,[39] 616 subjects were randomized to one of two iloperidone-treated groups (4–8 or 10–16 mg/day), risperidone (4–8 mg/day) or placebo. Significantly greater improvement in BPRS scores was observed with both iloperidone groups and with risperidone compared with placebo (Table 3). Similar results were observed for most secondary efficacy measures, including PANSS-total and -positive subscale scores, and CGI-S scores (Table 3). Only the iloperidone 10–16-mg and risperidone groups were associated with significantly greater improvement in PANSS-negative subscale scores than placebo (Table 3).

In Study 3,[39] 706 patients were randomized to receive iloperidone (12–16 or 20–24 mg/day), risperidone (6–8 mg/day) or placebo. Both iloperidone 20–24 mg and risperidone demonstrated significantly greater improvement in BPRS scores than placebo (Table 3), while a trend-level difference was noted with iloperidone 12–16 mg (p = 0.09 vs placebo). Significantly greater improvement in CGI-S scores was observed in both iloperidone dose groups and the risperidone group, compared with placebo (Table 3). However, only the iloperidone 20–24 mg and risperidone groups were associated with significantly greater improvement in PANSS total and subscale scores (Table 3).

A pooled analysis of Studies 1–3[39] examined the combined effects of each iloperidone dose group (4–8, 10–16 or 20–14 mg/day), haloperidol, risperidone, or placebo on baseline-to-end point change in BPRS scores. The pooled data set consisted of 1,553 patients who had at least 2 weeks of post-randomization follow-up. All active treatment groups, including all three iloperidone dose ranges, were associated with significantly greater improvement in BPRS scores, relative to placebo. In another pooled analysis that included data from all four prospective, randomized, double-blind, placebo- and active-controlled trials of patients with schizophrenia and schizoaffective disorder, significantly greater improvement in PANSS total, PANSS-positive and -negative subscale, and BPRS scores was also reported with iloperidone (10–16 and 20–24 mg/day) compared with placebo.[41] Similar results were obtained when the data from patients with schizoaffective disorder were excluded.

Asenapine The efficacy of asenapine for acute-phase treatment of schizophrenia has been demonstrated in two randomized, Phase III, multi-site, double-blind, placebo- and active comparator-controlled trials (Table 4). Potkin et al. conducted a 6-week randomized, fixed-dose study of asenapine 10 mg/day (5 mg b.i.d.), risperidone 6 mg (3 mg b.i.d.) or placebo.[42] Eligible patients were adults (>18 years of age) with acutely exacerbated schizophrenia (n = 182 randomized subjects), defined by baseline CGI-S scores of >4, PANSS total score of >60, and a score >4 on at least two PANSS-positive subscale items (hallucinations, delusions, suspiciousness/persecution, grandiosity or conceptual disorganization). Eligible subjects also had to have had a prior history of positive antipsychotic drug treatment response. Study drugs were titrated to target dose over 3–5 days. All asenapine-treated subjects also received two solid oral placebo tablets, while risperidone-treated subjects also received sublingual placebo. Patients in the placebo group received both solid oral and sublingual placebo b.i.d. formulations. Baseline-to-end point improvement in PANSS total scores, the primary study end point, was significantly greater with asenapine than placebo but not with risperidone (Table 4). Significantly greater decreases in PANSS total scores in the asenapine group (vs placebo) were observed beginning at week 2 and continued through the end of the study period. As compared with placebo, both asenapine and risperidone resulted in significantly greater improvement in PANSS-positive subscale and CGI-S scores (Table 4).[42] However, only asenapine resulted in significantly greater improvement in PANSS-negative and -general psychopathology subscale scores, compared with placebo (Table 4).

In a second 6-week study by Kane et al., 458 patients (aged 18 years or older) with acutely exacerbated schizophrenia were randomized to receive one of two fixed doses of asenapine (5 or 10 mg b.i.d.), haloperidol (4 mg b.i.d.) or placebo.[43] Eligible subjects had a PANSS total score of >60, a score of >4 on at least two PANSS-positive subscale items (delusions, hallucinations, suspiciousness/persecution, grandiosity, or conceptual disorganization) and CGI-S score of >4 at baseline. In the primary last-observation-carried-forward analysis, asenapine 5 mg b.i.d. and haloperidol, but not asenapine 10 mg b.i.d., resulted in significantly greater baseline-to-end point improvement in PANSS total scores (the primary efficacy end point) and PANSS-positive subscale scores (Table 4).[43] Advantages of asenapine 5 mg b.i.d. and haloperidol over placebo on PANSS-total and -positive subscale scores were apparent beginning at week 3, and continued throughout the remainder of the study period. All active treatments were superior to placebo on change in PANSS Marder positive factor scores (Table 4).[44] Only asenapine (5 mg b.i.d.) resulted in significantly greater improvement in PANSS-negative and -general psychopathology subscale scores, while no active treatment was superior to placebo for change in PANSS Marder negative factor scores (Table 4). Only haloperidol was associated with significantly greater improvement in PANSS Marder hostility/excitement factor scores than placebo, and only asenapine 5 mg b.i.d. was associated with greater improvement in PANSS Marder disorganized thought processes factor (Table 4). Both asenapine dose groups, but not the haloperidol group, were associated with a significantly higher proportion of positive treatment response, defined a priori as a >30% decrease from baseline in PANSS total scores (Table 4). A significantly higher proportion of subjects achieving a CGI-Improvement score of 1 (very much improved) or 2 (much improved) was observed in the asenapine 5 mg b.i.d. group, but not the asenapine 10 mg b.i.d. or haloperidol groups, relative to placebo (Table 4).

It is worth noting that the acute-phase efficacy of asenapine in adults with schizophrenia has been evaluated in two additional 6-week placebo-controlled trials that also included olanzapine as an active control group.[45] The first study compared the efficacy of two fixed doses of asenapine with olanzapine or placebo. Only olanzapine (15 mg daily) resulted in significant baseline-to-end point decreases in PANSS total score, compared with placebo. As such, this was considered a negative trial for asenapine. The second study compared the clinical effects of flexibly dosed asenapine (5–10 mg b.i.d.), olanzapine (10–20 mg b.i.d.) and placebo. In this study, neither active treatment group showed significantly greater improvement in PANSS total scores, the primary efficacy variable, than placebo. As such, this was considered a failed trial.

Lurasidone The short-term, acute-phase efficacy of lurasidone for the treatment of schizophrenia in adults has been evaluated in several 6-week double-blind, placebo-controlled trials (of which three were Phase II studies).[33,46] One Phase II[47] and one Phase III[48] study have been published (Table 5). One 6-week, Phase II, placebo-controlled study (with an additional haloperidol-treated arm for assay sensitivity) was considered a failed trial.[46]

In the first Phase II study conducted in the USA, 149 patients with acute schizophrenia were randomized to one of two fixed doses of lurasidone (40 or 120 mg/day) or placebo.[46] No active control group was included in this study. Significantly greater mean improvement in BPRS scores (lurasidone 40 mg [-9.4 ± 1.6] vs placebo [-3.8 ± 1.6], p = 0.02; lurasidone 120 mg [-11.0 ± 1.6], p = 0.004 vs placebo), the primary study end point, as well as CGI-S and Clinical Global Impression-Improvement scores, were observed with both lurasidone doses than with placebo. The lurasidone 120-mg dose group experienced significantly greater improvement in PANSS total scores. Both lurasidone groups had higher rates of positive treatment response (defined as >20% improvement from baseline in PANSS total scores), compared with the placebo group (lurasidone 40 mg [55.3%, number needed to treat (NNT) 3, 95% CI: 2–7]; lurasidone 120 mg [50.0%, NNT 4, 95% CI: 3–11]).

In a second USA Phase II study published by Nakamura et al., 180 patients (aged 18 to 65 years) with an acute exacerbation of schizophrenia were randomized to receive lurasidone at a fixed dose of 80 mg/day or placebo.[47] Eligible subjects had a BPRS score of >42, a CGI-S score of >4 (moderately ill), and minimal EPS (Simpson-Angus Scale [SAS][49] score of <2 and an Abnormal Involuntary Movement Scale [AIMS][50] score of <3) at baseline. Significantly greater baseline-to-end point improvement in BPRS scores was observed with lurasidone than placebo (Table 5). Statistical advantages of lurasidone over placebo were observed as early as day 3, and continued throughout the remainder of the study. Lurasidone treatment also resulted in significantly greater baseline-to-end point improvement in all secondary effect measures, including PANSS total and subscale scores, as well as the CGI-S and Montgomery–Asberg Depression Rating Scale (MADRS) (Table 5). A significantly higher proportion of lurasidone-treated patients achieved a positive treatment response (Table 5), defined a priori as a >20% reduction in PANSS total score from baseline.

Three Phase III studies have been conducted;[46,51] however, only one study (PEARL 2) has been published.[48] The first study (PEARL 1) compared the efficacy of lurasidone at one of three fixed doses (40, 80, or 120 mg/day) with placebo in a 6-week, multination (51 sites worldwide), randomized study of 500 patients with acute schizophrenia.[46] The primary efficacy end point was baseline-to-end point change in PANSS total scores; change in CGI-S score was a secondary measure. Lurasidone 80 mg/day, but not 40 or 120 mg/day, resulted in significantly greater improvement in PANSS total and CGI-S scores, as compared with placebo. Significantly greater improvement in PANSS-positive subscale scores and a significantly higher proportion of positive treatment responders (e.g., persons who achieved a >30% decrease in PANSS total scores from baseline) were observed for both the 80-and 120-mg/day lurasidone dose groups, compared with placebo.

Meltzer et al. published results of the second Phase III study (PEARL 2), in which 478 patients were randomized to one of four fixed-dose treatment groups: lurasidone (40 or 120 mg/day), olanzapine (15 mg/day) or placebo.[48] Eligible subjects had a Mini-International Neuropsychiatric Interview-confirmed diagnosis of schizophrenia, and a CGI-S score >4, PANSS total score >80, and a score of >4 on at least two PANSS-positive subscale items (delusions, hallucinations, suspiciousness/persecution, unusual thought content, or conceptual disorganization), at baseline [52]. Baseline-to-end point improvement in PANSS total scores, the primary efficacy end point, was significantly greater for all of the active treatment groups, compared with the placebo group (Table 5). Superiority of lurasidone 40 mg and olanzapine over placebo on PANSS total score reduction was observed from week 1 onward, while significantly greater improvement in this effect measure for lurasidone 120 mg was observed from week 3 onward. Similar results for all active treatment groups were observed on secondary efficacy measures, including baseline to end point change in PANSS-positive, -negative, -general psychopathology and CGI-S scores (Table 5). Only the olanzapine group experienced significantly greater baseline-to-end point improvement in MADRS scores than the placebo group (Table 5). A significantly higher proportion of olanzapine but not lurasidone treated patients were considered positive treatment responders (Table 5).

The final Phase III trial (PEARL 3) compared the effects of two lurasidone fixed doses (80 and 160 mg/day) with placebo and quetiapine XR (600 mg/day) in adults with acute schizophrenia over 6 weeks.[46,51,53] All active treatments were associated with significant improvement from baseline to study end point in PANSS total scores, the primary efficacy end point, as well was CGI-S scores, compared with placebo. Significant improvement in PANSS total scores for both lurasidone dose groups was observed at day 4 and persisted throughout the remainder of the study period. Improvements in mean PANSS total scores from baseline were significantly greater with lurasidone 80-mg (-22.2), lurasidone 160-mg (-26.5) group, and quetiapine XR (-27.8), compared with placebo (-10.3, p < 0.001 for all comparisons).[53] All active treatments resulted in significantly greater improvement than placebo in depressive symptoms, as assessed by the Montgomery Asberg Depression Rating Scale[54] (p < 0.001 for all comparisons).

Longer-term studies

Iloperidone The long-term safety and efficacy of iloperidone was evaluated in three 52-week, randomized, double-blind, multicenter trials. Pooled data from these studies have been published.[55] Each of the studies were identical in design and compared the clinical effects of flexibly dosed iloperidone (4–16 mg/day) and haloperidol (5–20 mg/day) in patients with schizophrenia or schizoaffective disorder. A total of 1,239 patients were randomized to iloperidone and 405 to haloperidol during an initial 6-week, double-blind treatment phase. Those who completed the initial treatment phase and achieved a positive treatment response (iloperidone: n = 371; haloperidol: n = 118) were eligible for the 46-week double-blind maintenance phase.

Relapse was defined on the basis of worsening symptoms (increase in PANSS total scores of >25% ****or >10 points compared with PANSS total scores taken at the end of the initial 6-week phase), global clinical state (increase in Clinical Global Impression of Change scores by >2 points), hospitalization because of psychotic symptom exacerbation or discontinuation of study medication owing to lack of efficacy.[55] Maintenance phase relapse rates were very similar between the iloperidone and haloperidol treatment groups (43.5 vs 41.2%; Table 6). Mean time to relapse was slightly longer with haloperidol (101.8 days) than iloperidone (89.8 days); however, this difference was not statistically significant. Assuming a relapse rate of 30% with haloperidol, the statistical threshold for equivalence between iloperidone and haloperidol (hazard ratio < 1.676 at α [one-sided] = 0.025) was achieved in the pooled data set. Baseline-to-end point improvement PANSS-positive subscale scores were greater with haloperidol (Table 6); however, there were no significant between-group differences in other secondary end points, including PANSS total and other subscale scores, BPRS scores, rate of discontinuation for unsatisfactory therapeutic effect, or rates of positive treatment response.

Asenapine

Two published randomized studies focused on the long-term efficacy and safety of asenapine in adults with schizophrenia,[56,57] including one relapse prevention study.[56] Two additional reports summarized the extension-phase results of trials comparing the effectiveness of asenapine and olanzapine,[58,59] one of which focused on primary negative symptoms.[59] In the relapse prevention study,[56] 700 patients were treated with open-label sublingual asenapine (10 mg b.i.d.) for up to 26 weeks. Patients who were clinically stable at the end of the open-label phase were randomized to double-blind continuation treatment with sublingual asenapine or placebo for an additional 26 weeks. Relapse was the primary end point and was broadly defined (Table 6, footnote) on the basis of symptom worsening, worsening global clinical state, or the emergence of violent or suicidal behaviors. During double-blind treatment, 12.1% of asenapine-treated patients and 47.4% of placebo-treated patients in the intent-to-treat population experienced a relapse event (p < 0.0001). Mean time to relapse (p < 0.0001) and first quartile of the time to relapse (asenapine, 156 days vs placebo, 41 days) was significantly longer with asenapine than with placebo. Rates of all-cause discontinuation were also higher with placebo (62.5%) than with asenapine (30.4%). Changes from baseline in favor of asenapine were observed for all secondary measures, including PANSS total, PANSS Marder factor,[44] CGI-S and CDSS scores (Table 6).

A second randomized, double-blind, multisite study evaluated the long-term (52 week) tolerability of asenapine (5–10 mg b.i.d., n = 913) or olanzapine (10–20 mg daily, n = 312) in adults with schizophrenia or schizoaffective disorder, and a baseline PANSS total score of >60 (including scores of >4 on two or more PANSS-positive subscale items).[57] Patients who completed the 52-week core study were given the option of continuing treatment until the study blind was broken (extension study). Clinical symptoms (as measured by the PANSS total and Marder factors), global clinical state (CGI-S) and other effectiveness measures (Subjective Well-Being Under Neuroleptic Treatment[60] and Medical Outcomes Study 12-Item Short Form[61]) were examined as secondary end points. Improvements in PANSS total scores were similar between treatment groups at week 6 (asenapine [-17.9] vs olanzapine [-19.0], p = not significant [NS]), but were significantly greater for the olanzapine group at study end point in the 52-week core study (Table 6). In the core study, no significant differences were observed between groups in a secondary observed-case analysis. Similar results were observed for baseline-to-end point change in PANSS Marder factor and CGI-S scores (Table 6). There were no significant between-group differences in SWN or SF-12 scores at any time during the study.

Results of the extension phase of this study were reported separately.[58] Patients who continued asenapine (n = 290, mean daily dosage 13.4 ± 4.6 mg) were followed for an additional 311.0 ± 146.1 days (range: 10–653 days), while those who continued olanzapine (n = 150, mean daily dosage 13.4 ± 4.1 mg) were followed for an additional 327 ± 139.6 days (range: 15–631 days). During the extension phase, only minor additional changes in mean PANSS total scores were observed in both groups (asenapine [+1.6], olanzapine [−0.8]). No significant further change in PANSS subscales, PANSS Marder factors, and CGI or CDSS scores were observed in either treatment group.

One additional report by Buchanan et al. presented the combined results of two identically-designed 26-week extension studies that evaluated the long-term efficacy of asenapine and olanzapine on persisting primary negative symptoms.[59] Eligible participants had a diagnosis of schizophrenia, baseline PANSS-negative subscale score of >20 and a score of >4 on at least three of seven PANSS negative symptom factor items (affective blunting, emotional withdrawal, passive social withdrawal, active social avoidance, poor rapport, motor retardation, or lack of spontaneity). Each study was conducted in two phases (a 26-week, double-blind core phase, followed by a 26-week double-blind extension phase). In the extension phase, patients were continued on their core study regimens. There were no significant between-group differences in negative symptom change, as measured by 16-item Negative Symptom Assessment Scale (NSA-16),[62] PANSS-negative symptom subscale, or PANSS Marder negative factor items[44] at 26 weeks (core studies). Improvement in PANSS -positive subscale and PANSS positive symptom factor scores was noted in both groups, with greater improvement occurring in the olanzapine group. In one of the two extension studies, baseline-to-end point change in NSA-16 scores (-15.8 vs -11.0; p = 0.03) and PANSS Marder negative factor scores (-9.2 vs -7.4; p < 0.05) were significantly greater with asenapine than olanzapine, while improvement in PANSS-positive subscale scores was greater with olanzapine than asenapine (-0.4 vs -1.5; p = 0.04). There were no significant between-group differences on the PANSS-negative subscale scores in either extension study.

Lurasidone

The long-term effectiveness and safety of lurasidone in adults with schizophrenia has been evaluated in one unpublished 6-month, open-label extension study[32] and one published 52-week randomized, double-blind study.[63] In the 52-week study by Citrome et al., 639 patients with stable schizophrenia or schizoaffective disorder were randomized to flexibly dosed lurasidone (40–120 mg daily, mean 84.7 mg daily) or risperidone (2–6 mg daily; mean: 4.3 mg daily).[63] The primary objective was to evaluate the long-term safety and tolerability of lurasidone; however, clinical efficacy was also assessed. All-cause discontinuation rates were higher for lurasidone-treated patients (269/419 [64%]) than those who received risperidone (105/202 [52%], NNT 9 [95% CI: 5–26]). Median survival time to all-cause study medication discontinuation was longer for the risperidone group (293 days) than the lurasidone group (181 days), and the probability of all-cause discontinuation at 12 months was 0.52 for risperidone- and 0.64 for lurasidone-treated subjects (log-rank p = 0.018). There were no significant between-group differences in the occurrence of relapse (lurasidone [20%] vs risperidone [16%], HR 1.31 [95% CI: 0.87–1.97]), defined a priori as an increase from baseline in PANSS total score by >30% (with CGI-S >3), rehospitalization for management of exacerbated psychotic symptoms, occurrence of suicidal (or homicidal) thinking, or increased risk of harm to self or others as judged by the study clinician. A preplanned test of noninferiority between treatment groups could not be conducted owing to the small number of relapses. PANSS total scores improved between baseline and 12 months in both treatment groups (lurasidone -4.7 [95% CI -6.4 to -3.0] vs risperidone -6.5 [95% CI: -8.8 to -4.3]), with no significant differences observed. There were no significant differences between treatment groups in PANSS subscale, CGI-S or MADRS scores at study end point or during follow-up, with the exception of greater improvement in MADRS scores with lurasidone at month 12 (difference = 1.6, p = 0.007).

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....