- Email: [email protected]
Efficacy of iloperidone in schizophrenia: A PANSS five-factor analysis
Schizophrenia Research 131 (2011) 75–81
Contents lists available at ScienceDirect
Schizophrenia Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c h r e s
Efficacy of iloperidone in schizophrenia: A PANSS five-factor analysis Leslie Citrome a,⁎, Xiangyi Meng b, Marla Hochfeld b a b
New York University School of Medicine, New York, NY, USA Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA
a r t i c l e
i n f o
Article history: Received 20 January 2011 Received in revised form 19 May 2011 Accepted 20 May 2011 Available online 22 June 2011 Keywords: Factor Iloperidone PANSS Schizophrenia
a b s t r a c t Background: The Positive and Negative Syndrome Scale (PANSS) total score is widely used to assess antipsychotic efficacy, however schizophrenia is a multi-dimensional disorder. We conducted a 5-factor analysis for evaluating the efficacy of iloperidone vs. placebo across these different domains in the treatment of schizophrenia. Method: The 5-factor model was determined from pooled data from 7 clinical trials (4 placebo- and activecontrolled and 3 non-inferiority active-comparator trials of iloperidone) in schizophrenia (N = 3580).Five factors were derived (excitement/hostility [P4,P7,G8,G14], depression/anxiety [G1,G2,G3,G4,G6], cognition [P2,N5,N7,G5,G10,G11,G12,G13,G15], positive [P1,P3,P5,P6,G9], and negative [N1,N2,N3,N4,N6,G7,G16]) from a factor analysis on the covariance matrix of 30 baseline PANSS items using a varimax rotation; factors retained had eigenvalues of ≥0.5. These newly derived 5 factors differ only slightly from other 5-factor analyses published by others using different datasets. The analysis of covariance model was then applied to assess these efficacy outcomes from the 4–6 week double-blind placebo and active controlled clinical trials of iloperidone. Results: Based on the placebo-controlled trials, iloperidone improvements from baseline (least squared mean change± standard error) were as follows: excitement/hostility, 0.4 ± 0.21 for 10–16 mg, 0.6 ± 0.43 for 20–24 mg vs. −1.0 ± 0.23 for placebo; P b 0.001 for both iloperidone doses vs. placebo; depression/anxiety, 1.9 ± 0.21 for 10–16 mg, 1.9 ± 0.41 for 20–24 mg vs. 1.1 ± 0.22 for placebo; P b 0.05 for 10–16 mg dose vs. placebo; cognition, 2.8 ± 0.35 for 10–16 mg, 3.9 ± 0.69 for 20–24 mg vs. 1.6 ± 0.38 for placebo; P b 0.05 for both iloperidone doses vs. placebo; positive, 3.7 ± 0.26 for 10–16 mg, 4.1 ± 0.53 for 20–24 mg vs. 2.7 ± 0.29 for placebo; P b 0.05 for both iloperidone doses vs. placebo; and negative, 2.2 ± 0.29 for 10–16 mg, 2.5 ± 0.58 for 20–24 mg vs. 1.3 ± 0.32 for placebo; P b 0.05 for 10–16 mg vs. placebo. Active controls validated iloperidone efficacy. Conclusions: Iloperidone demonstrated positive treatment effects on these newly derived PANSS factors. The 10–16 mg and 20–24 mg dose groups had similar efficacy on the PANSS factors, with the exception of the depression/anxiety and negative factors, on which the 10–16 mg dose group showed statistical separation from placebo and the 20–24 mg dose group did not. At 6 weeks, the lack of separation from placebo for the higher dose group may have been due to the much smaller sample size in that group. © 2011 Elsevier B.V. All rights reserved.
1. Introduction The 30-item Positive and Negative Syndrome Scale (PANSS) (Kay et al., 1987)is commonly used in clinical trials to assess the psychopathology associated with schizophrenia. Although the PANSS has 3 subscales – positive symptoms, negative symptoms, and general psychopathology – some of the items assigned to one subscale may perhaps be better conceptualized as belonging to another symptom construct such as “mood” or “cognitive”. This, together with the growing understanding that schizophrenia is a multidimensional disorder with core psychotic, negative, mood, and
⁎ Corresponding author at: 11 Medical Park Drive, Suite 106, Pomona, NY 10970. Tel.: + 1 845 362 2081; fax: + 1 845 362 8745. E-mail addresses: [email protected] (L. Citrome), [email protected] (X. Meng), [email protected] (M. Hochfeld). 0920-9964/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2011.05.018
cognitive components has led to the development of multiple-factor models of the PANSS (Lindenmayer et al., 1994; White et al., 1997; Marder et al., 1997; Van den Oord et al., 2006; see review by Lehoux et al., 2009). These can be used to describe the outcomes of clinical trials for the treatment of schizophrenia and discern differences in efficacy based on symptom dimensions (Lindenmayer et al., 2004). Iloperidone is a second-generation antipsychotic that in 2009 received US Food and Drug Administration approval and has a current indication for the treatment of schizophrenia in adults (Citrome 2009; Citrome 2010; Novartis Pharmaceuticals Corporation 2011). A pooled analysis was conducted to identify 5 PANSS factors using baseline ratings of 3580 subjects with schizophrenia (excluding schizoaffective disorder) across 7 studies (3 non-inferiority active-controlled and 4 placebo- and active-controlled Phase III studies involving iloperidone). After deriving these 5 PANSS factors, the panels were then used to evaluate the efficacy of iloperidone in subjects with schizophrenia from 4 pooled, placeboand active-controlled clinical trials of 4 or 6 weeks’ duration.
76
L. Citrome et al. / Schizophrenia Research 131 (2011) 75–81
2. Method 2.1. Design and objective This was a post hoc analysis of pooled data from 7 pivotal trials of iloperidone in order to build a 5-factor PANSS model and to subsequently apply it to assess the multi-dimensional psychopathological outcomes across the 4 short-term controlled trials of iloperidone that were originally designed to test the overall efficacy of iloperidone versus placebo in subjects with schizophrenia. 2.2. Data sources The PANSS factor analysis was conducted using 7 placebo- and active-controlled studies including 3582 subjects with schizophrenia (Table 1). Because 6 of the 7 studies also included subjects with schizoaffective disorder, the analysis was repeated using baseline ratings of 4150 subjects with schizophrenia or schizoaffective disorder to determine whether or not this would have changed the factor structure in any substantial manner. The rationale for constructing a 5-factor model from this dataset was to maximize the internal validity for any subsequent analysis involving change in factor scores over time. All studies that were included enrolled acute patients; the longer term studies (Kane et al., 2008) required subjects to demonstrate improvement during a 6-week stabilization phase in order to be eligible to participate in the 46-week maintenance phase. Four randomized, double-blind, placebo- and active-controlled, multicenter studies that evaluated the efficacy and safety of iloperidone for either 6 weeks (3 studies) or 4 weeks (1 study) were evaluated using the newly derived 5-factor PANSS analysis (Table 1). Male or female subjects in these 4 studies were aged 18–65 years with a Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) diagnosis of schizophrenia (all studies) or schizoaffective disorder (only the 6-week studies); data from subjects with schizophrenia but not
schizoaffective disorder were pooled and used for the present analysis. For the efficacy analyses, all randomized subjects with schizophrenia and at least one post-baseline efficacy measurement obtained while on study medication were eligible for inclusion. Note that the data source for the efficacy analysis is a subset of the data source that was used for the derivation of the PANSS factors. 2.3. Data analysis Similar to prior analyses by other investigators (Lançon et al., 1998; Van den Oord et al., 2006; Levine and Rabinowitz, 2007), factors were derived from an analysis on the covariance matrix of 30 baseline PANSS items using a varimax rotation. Five factors were retained that had eigenvalues of ≥ 0.5 (Table 2). A total of 7 treatment groups were subsequently evaluated for efficacy using the newly derived 5 factors: placebo (n = 447), iloperidone 4–8 mg/d (n= 276), iloperidone 10–16 mg/d (n= 381), and iloperidone 20–24 mg/d (n = 394), with active controls for assay sensitivity haloperidol 15 mg/d (n = 70), risperidone 4–8 mg/d (n = 229), and ziprasidone 160 mg/d (n = 144). Dose groups for iloperidone were selected based on the doses tested in the individual studies. Numbers of subjects per group for each time point are provided in Table 3. The changes from baseline to study end point (Week 4 or Week 6, depending on the study) for each of the 5 PANSS factors were evaluated using an analysis of covariance (ANCOVA) model with treatment and study as factors and baseline as a covariate. Last observations before Week 4 or 6 were carried forward until Week 4 or 6, respectively. For any missing observations, the last recorded measurement was used. Because differential dropout rates, possibly related to titration, were observed between iloperidone and other treatments at the start of one pivotal trial (Potkin et al., 2008), additional analyses were also carried out for the subset of subjects who continued study treatment for more than 2 weeks.
Table 1 Studies used for the 5-factor Positive and Negative Syndrome Scale analysis and the subsequent efficacy analyses. Duration (weeks)
Descriptiona
N with schizophrenia available for analysis
Factor analysis only ILP3001 (Kane et al., 2008)
52
548
ILP3002 (Kane et al., 2008)
52
ILP3003 (Kane et al., 2008)
52
International, multicenter, double-blind, randomized, parallel-group study to compare the antipsychotic effect of iloperidone 4–16 mg/d with that of haloperidol 5–20 mg/d in 600 patients with schizophrenia or schizoaffective disorder (454 iloperidone, 146 haloperidol) International, multicenter, double-blind, randomized, parallel-group study to compare the antipsychotic effect of iloperidone 4–16 mg/d with that of haloperidol 5–20 mg/d in 557 patients with schizophrenia or schizoaffective disorder (420 iloperidone, 137 haloperidol) International, multicenter, double-blind, randomized, parallel-group study to compare the antipsychotic effect of iloperidone 4–16 mg/d with that of haloperidol 5–20 mg/d in 487 patients with schizophrenia or schizoaffective disorder (365 iloperidone, 122 haloperidol)
US and India, multicenter, double-blind, randomized, parallel-group, placebo- and activecontrolled study to evaluate the efficacy of iloperidone 24 mg/d in 606 patients with schizophrenia (303 iloperidone, 151 ziprasidone 160 mg/d, 152 placebo) US, multicenter, double-blind, randomized, parallel-group, placebo- and active-controlled study to evaluate the efficacy of three fixed doses of iloperidone (4, 8, and 12 mg/d) in 621 patients with schizophrenia or schizoaffective disorder (370 iloperidone, 124 haloperidol 15 mg/d, 127 placebo) International, multicenter, double-blind, randomized, parallel-group, placebo- and activecontrolled study to evaluate the efficacy of two non-overlapping dose ranges of iloperidone (4–8 mg/d and 10–16 mg/d) in 616 patients with schizophrenia or schizoaffective disorder (307 iloperidone, 153 risperidone 4–8 mg/d, 156 placebo) International, multicenter, double-blind, randomized, parallel-group, placebo- and activecontrolled study to evaluate the efficacy of two non-overlapping dose ranges of iloperidone (12–16 mg/d and 20–24 mg/d) in 706 patients with schizophrenia or schizoaffective disorder (389 iloperidone, 157 risperidone 6–8 mg/d, 160 placebo)
597
Study
Factor and efficacy analyses VP-VYY-683-3101 (Cutler et al., 2008)
4
ILP3000ST (Potkin et al., 2008)
6
ILP3004ST (Potkin et al., 2008)
6
ILP3005ST (Potkin et al., 2008)
6
a
Doses of all medications administered twice daily. All subjects were adults.
524
470
422
479
542
L. Citrome et al. / Schizophrenia Research 131 (2011) 75–81
77
Table 2 Positive and Negative Syndrome Scale factor analysis results. PANSS Item
Varimax Rotations (eigenvalues)
P2: Conceptual Disorganization N5: Difficulty in Abstract Thinking N7: Stereotyped Thinking G5: Mannerisms and Posturing G10: Disorientation G11: Poor Attention G12: Lack of Judgment and Insight G13: Disturbance of Volition G15: Preoccupation G1: Somatic Concern G2: Anxiety G3: Guilt Feelings G4: Tension G6: Depression P4: Excitement P7: Hostility G8: Uncooperativeness G14: Poor Impulse Control N1: Blunted Affect N2: Emotional Withdrawal N3: Poor Rapport N4: Passive/Apathetic Social Withdrawal N6: Lack of Spontaneity and Flow of Conversation G7: Motor Retardation G16: Active Social Avoidance G9: Unusual Thought Content P1: Delusions P3: Hallucinatory Behavior P5: Grandiosity P6: Suspiciousness/Persecution
Factor Category
Factor 1 (6.21)
Factor 2 (3.23)
Factor 3 (1.74)
Factor 4 (1.14)
Factor 5 (0.80)
0.06927 0.21867 0.20526 0.19993 0.12376 0.18291 0.18402 0.36260 0.25600 − 0.02155 − 0.00590 0.01528 0.00764 0.15698 − 0.20282 0.10129 0.29846 0.00134 0.65576 0.77740 0.65400 0.77805 0.67097 0.55519 0.63294 − 0.02886 − 0.08101 0.04786 − 0.24742 0.11006
0.61525 0.54674 0.54540 0.45492 0.37966 0.62405 0.42130 0.47473 0.42331 0.12663 0.03426 − 0.06280 0.22590 − 0.13051 0.27643 0.08825 0.26600 0.22037 0.27410 0.18205 0.34115 0.16360 0.34243 0.21073 0.07630 0.37076 0.10408 0.10024 0.15138 − 0.05973
0.13850 0.01915 0.15531 0.11487 0.07067 0.20043 0.16825 0.08457 0.12138 0.01390 0.21617 0.01051 0.35765 − 0.01561 0.55246 0.70928 0.56788 0.63940 − 0.07029 0.07017 0.18664 0.03613 0.02111 − 0.13534 0.18413 0.08478 0.08991 0.03505 0.16870 0.24141
0.26177 0.09480 0.17789 0.02826 0.04833 0.04788 0.22465 − 0.05454 0.27014 0.17183 0.09695 0.10293 0.04176 0.03593 0.16913 0.17792 0.03908 0.12032 0.03375 0.04081 0.07792 0.05842 − 0.13576 − 0.13651 0.14994 0.61985 0.76073 0.50704 0.38348 0.48263
− 0.03064 − 0.09024 0.03259 0.08661 0.00388 0.06184 − 0.22885 0.10842 0.12712 0.33807 0.70026 0.53079 0.55620 0.59553 0.23172 0.10456 − 0.03990 0.14949 0.00348 0.00350 − 0.07110 − 0.00345 − 0.01503 0.14453 0.12745 0.03874 0.13911 0.14380 0.01453 0.31324
Cognition Cognition Cognition Cognition Cognition Cognition Cognition Cognition Cognition Depression/Anxiety Depression/Anxiety Depression/Anxiety Depression/Anxiety Depression/Anxiety Excitement/Hostility Excitement/Hostility Excitement/Hostility Excitement/Hostility Negative Negative Negative Negative Negative Negative Negative Positive Positive Positive Positive Positive
G = general; N = negative; P = positive; PANSS = Positive and Negative Syndrome Scale.
3. Results 3.1. Demographics and descriptive data A total of 3580 subjects with schizophrenia were included when constructing the 5-factor model (2 subjects excluded due to missing baseline data). The efficacy analysis included 1941 subjects (see Table 3). Subject demographic and baseline characteristics for the population are shown in Table 4. Disposition is provided in Table 5. 3.2. PANSS factor analysis Five factors were derived from PANSS analysis for 3580 patients with schizophrenia: excitement/hostility, anxiety/depression, cognition, positive, and negative (see Table 2). The robustness of these
factors was supported by a similar pooled analysis of the same 7 studies using baseline ratings of 4150 subjects with schizophrenia or schizoaffective disorder, which yielded the same 5 factors. The variance explained by each factor was 50%, 26%, 14%, 9% and 6% for the negative, cognition, excitement/hostility, positive and anxiety/ depression factor, respectively. 3.3. Iloperidone efficacy Overall, subjects receiving iloperidone showed greater changes from baseline in all 5 PANSS factor analysis scores vs. those receiving placebo. Among all subjects, those receiving iloperidone 10–16 mg/d showed significant improvement at Week 4 and Week 6 on all factors vs. placebo. Subjects receiving iloperidone 20–24 mg/d showed significant improvement at Week 4 and Week 6 on the cognition,
Table 3 Numbers of patients analyzed for efficacy measurements by week and treatment group for the total population and patients receiving treatment for greater than 2 weeks a. Week
All patients
Patients treated for N2 weeks
1 2 3 4 5 6 1 2 3 4 5 6
Iloperidone 4–8 mg/d
10–16 mg/d
20–24 mg/d
276 276 276 276 276 276 217 217 217 217 217 217
379 380 380 380 380 380 294 295 295 295 295 295
392 394 394 394 111 111 315 316 316 316 90 90
to 381 or 381 or 381 or 381 or 381 or 381 or 295
Risperidone 4–8 mg/d
Haloperidol 15 mg/d
Ziprasidone 160 mg/d
Placebo
229 229 229 229 229 229 194 194 194 194 194 194
70 70 70 70 70 70 49 49 49 49 49 49
141 143 143 143 0 0 108 109 109 109 0 0
444 446 446 446 306 306 337 339 339 339 233 233
Note: Weeks 1–4 include data pooled from 4 clinical trials, while Weeks 5–6 include data pooled from 3 trials. a Variations within a given week due to missing efficacy data.
or or or or
142 144 144 144
or or or or
109 110 110 110
or or or or or or or or or or or or
445 447 447 447 307 307 338 340 340 340 234 234
78
L. Citrome et al. / Schizophrenia Research 131 (2011) 75–81
Table 4 Patient characteristics (all patients with schizophrenia in the 4 efficacy studies with evaluable data)a. Characteristic
Risperidone 4–8 mg/d (n = 230)
Haloperidol 15 mg/d (n = 70)
Ziprasidone 160 mg/d (n = 144)
Placebo (n = 447)
315 (80) 39 (18–65)
162 (70) 38.5 (17–67)
53 (76) 40 (19–59)
108 (75) 40 (20–61)
320 (72) 39 (18–66)
248 (65) 106 (28) 8 (2) 19 (5)
184 (47) 166 (42) 27 (7) 17 (4)
161 (70) 54 (24) 3 (1) 12 (5)
29 (41) 35 (50) 2 (3) 4 (6)
50 72 12 10
(35) (50) (8) (7)
217 (49) 182 (41) 19 (4) 29 (7)
60 (16) 151 (40) 157 (41) 10 (3) 3 (1)
67 (17) 179 (45) 138 (35) 8 (2) 2 (1)
38 (17) 88 (38) 96 (42) 6 (3) 2 (1)
11 (16) 29 (41) 28 (40) 1 (1) 1 (1)
24 (17) 56 (39) 61 (42) 1 (1) 2 (1)
75 (17) 198 (44) 153 (34) 13 (3) 8 (2)
17 (5) 302 (79) 1 (0.3) 61 (16) 94.3 (14.7)
23 (6) 323 (82) 0 48 (12) 93.5 (14.0)
16 (7) 179 (78) 0 35 (15) 95.7 (15.9) [n = 229]
2 (3) 56 (80) 0 12 (17) 98.3 (15.5)
3 (2) 122 (85) 0 19 (13) 90.9 (11.6)
23 (5) 371 (83) 0 53 (12) 93.8 (15.4)
54.4 (8.3)
54.7 (8.1)
55.0 (9.2) [n = 229]
56.9 (9.0)
53.3 (6.8)
54.3 (8.8) [n = 444]
4.8 (0.7) [n = 281]
4.8 (0.7) [n = 386]
4.8 (0.7) [n = 215]
NA
4.6 (0.7) [n = 142]
4.7 (0.8) [n = 356]
Iloperidone 4–8 mg/d (n = 275)b
Male, n (%) 205 (75) Median age, 39 (18–68) yr (range) Race, n (%) White 131 (48) Black 119 (43) Asian 6 (2) Other 19 (7) Age at diagnosis, n (%) b18 yr 59 (22) 18–24 yr 110 (40) 24–44 yr 98 (36) 45–66 yr 4 (2) Missing 4 (2) DSM-IV schizophrenia classification, n (%) 10 Disorganized 25 (9) 30 Paranoid 203 (74) 60 Residual 0 90 Undifferentiated 47 (17) 96.2 (15.7) Mean baseline PANSS total score (SD) 55.7 (8.9) Mean baseline BPRSd total score (SD) Mean baseline CGI-S 4.8 (0.7) score (SD)c [n = 105]
10–16 mg/d (n = 381)
20–24 mg/d (n = 394)
261 (69) 40 (18–68)
BPRSd = PANSS-derived Brief Psychiatric Rating Scale; CGI-S = Clinical Global Impression of Severity; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, 4th Edition; NA = Not assessed; PANSS = Positive and Negative Symptoms Scale; SD = standard deviation. a Due to missing data points, not all patients were included in the efficacy analysis. b Excludes 1 patient included in efficacy assessments. c Evaluated in 3 studies only.
excitement/hostility, and positive factors. The largest least-squared mean (Standard Error) reductions from baseline in subjects treated with iloperidone were observed in the cognition and positive factors at Week 6, as follows: cognition, all subjects: −2.8 (0.35) with iloperidone 10– 16 mg/d and −3.9 (0.69) with iloperidone 20–24 mg/d (Table 6); cognition, subjects treated N 2 weeks: -4.1 (0.38) with iloperidone 10– 16 mg/d and −4.8 (0.76) with iloperidone 20–24 mg/d (Table 7); positive, all subjects: −3.7 (0.26) with iloperidone 10–16 mg/d and −4.1 (0.53) with iloperidone 20–24 mg/d (Table 6); positive, subjects treated N 2 weeks: −4.9 (0.29) with iloperidone 10–16 mg/d and −5.0 (0.58) with iloperidone 20–24 mg/d (Table 7). Larger numeric
improvements on all 5 factors were observed in subjects receiving treatment for more than 2 weeks compared to the entire evaluated population (Table 7). Subjects receiving active controls (risperidone, haloperidol, and ziprasidone) also showed significant changes from baseline vs. placebo in the 5-factor PANSS scores. Iloperidone in the dose range of 4–8 mg/d was not efficacious, with the exception of the excitement/hostility factor. Speed of improvement on the different factors is qualitatively slower for iloperidone. Categorical results for an at least 20% improvement from baseline by 1 and 2 weeks for each intervention on each factor is outlined in Table 8.
Table 5 Patient disposition (all patients with schizophrenia in the 4 efficacy studies). Iloperidone
Patients randomized Completers Discontinued Missing Primary reason for discontinuation Protocol deviation Adverse event(s) a Lost to follow-up Death Patient withdrew consent Unsatisfactory therapeutic effect Other b a b
4–8 mg/d (n = 300)
10–16 mg/d (n = 402)
20–24 mg/d (n = 409)
300 (100.0) 124 (41.3) 176 (58.7) 0
402 207 193 2
(100.0) (51.5) (48.0) (0.5)
409 264 144 1
3 (1.0) 14 (4.7) 15 (5.0) 0 56 (18.7) 86 (28.7) 2 (0.7)
4 15 10 0 56 104 4
(1.0) (3.7) (2.5) (13.9) (25.9) (1.0)
3 22 3 0 69 43 4
(100.0) (64.5) (35.2) (0.2)
(0.7) (5.4) (0.7) (16.9) (10.5) (1.0)
Includes Adverse Events, Abnormal Laboratory Value, Abnormal Test Procedure, and Pregnancy. Includes Condition No Longer Requires Study Drug, Administrative Problem, and Other reasons.
Risperidone 4–8 mg/d (n = 241)
Haloperidol 15 mg/d (n = 78)
Ziprasidone 160 mg/d (n = 149)
Placebo (n = 472)
241 (100.0) 157 (65.1) 83 (34.4) 1 (0.4)
78 (100.0) 33 (42.3) 45 (57.7) 0
149 (100.0) 98 (65.8) 51 (34.2) 0
472 (100.0) 235 (49.8) 236 (50.0) 1 (0.2)
3 (1.2) 14 (5.8) 15 (6.2) 0 24 (10.0) 27 (11.2) 0
2 (2.6) 8 (10.3) 4 (5.1) 0 18 (23.1) 13 (16.7) 0
1 (0.7) 13 (8.7) 0 0 23 (15.4) 12 (8.1) 2 (1.3)
2 (0.4) 28 (5.9) 12 (2.5) 0 54 (11.4) 133 (28.2) 8 (1.7)
L. Citrome et al. / Schizophrenia Research 131 (2011) 75–81
79
Table 6 Least-squared mean changes from baseline in the 5-factor PANSS model at Weeks 1 through 6for all patients with schizophrenia. Factor scores for all patients, (SE) Cognition
Excitement/ hostility
Depression/ anxiety
Positive
Negative
Iloperidone
Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week
1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6
4–8 mg/d
10–16 mg/d
20–24 mg/d
− 0.6 (0.3) − 0.9 (0.36) − 1.4 (0.39) − 1.6 (0.42) − 1.7 (0.42) − 1.8 (0.43) 0.0 (0.2) − 0.1 (0.24) − 0.1 (0.25)* − 0.2 (0.26)‡ − 0.2 (0.26)† − 0.1 (0.26)† − 0.9 (0.21) − 1.1 (0.23) − 1.4 (0.25) − 1.6 (0.26) − 1.7 (0.25) − 1.7 (0.26) − 1.3 (0.23) − 1.8 (0.27) − 2.3 (0.3) − 2.6 (0.32) − 3.0 (0.32) − 3.0 (0.33) − 1.1 (0.28) − 1.3 (0.32) − 1.5 (0.35) − 1.4 (0.36) − 1.4 (0.36) − 1.5 (0.36)
−0.8 (0.24) − 1.6 (0.28) − 2.2 (0.31)* − 2.5 (0.33)† − 2.7 (0.34)* − 2.8 (0.35)* − 0.1 (0.16) − 0.2 (0.19)† − 0.4 (0.2)‡ − 0.5 (0.21)‡ − 0.4 (0.21)‡ − 0.4 (0.21)‡ − 0.9 (0.16) − 1.3 (0.18) − 1.8 (0.2)* − 2.0 (0.20)* − 1.8 (0.2)* − 1.9 (0.21)* −1.5 (0.18) − 2.3 (0.21) − 3.0 (0.24) − 3.3 (0.25)* − 3.6 (0.26)† − 3.7 (0.26)† − 1.0 (0.22) − 1.8 (0.25) − 2.2 (0.27)* − 2.4 (0.29)* − 2.2 (0.29)* − 2.2 (0.29)*
− 1.1 − 2.3 − 2.7 − 3.0 − 3.5 − 3.9 0.2 − 0.2 − 0.2 − 0.3 − 0.5 − 0.6 − 0.7 − 1.3 − 1.3 − 1.6 − 1.8 − 1.9 − 1.7 − 2.9 − 3.3 − 3.6 − 3.9 − 4.1 − 1.3 − 1.8 − 1.9 − 2.2 − 2.4 − 2.5
(0.29) (0.34)* (0.37) † (0.39)‡ (0.67) † (0.69)† (0.19) (0.22)* (0.24)† (0.25)‡ (0.42)† (0.43)‡ (0.19) (0.22) (0.24) (0.24) (0.4) (0.41) (0.21) (0.26) (0.28)† (0.30)† (0.52)* (0.53)* (0.26) (0.3) (0.33) (0.34) (0.58) (0.58)
Risperidone 4–8 mg/d
Haloperidol 15 mg/d
Ziprasidone 160 mg/d
Placebo
− 1.9 (0.31)* − 3.5 (0.37) ‡ − 3.8 (0.41)‡ − 4.4 (0.43)‡ − 4.7 (0.45)‡ − 4.9 (0.47)‡ − 0.9 (0.21)‡ − 1.3 (0.25)‡ − 1.4 (0.26)‡ − 1.5 (0.27)‡ − 1.6 (0.28)‡ − 1.5 (0.29)‡ − 1.5 (0.21)* − 2.0 (0.24)* − 2.3 (0.26)‡ − 2.5 (0.26)‡ − 2.7 (0.27)‡ − 2.8 (0.28)‡ − 2.5 (0.23) † − 4.1 (0.28) ‡ − 4.6 (0.31)‡ − 5.2 (0.33)‡ − 5.5 (0.35)‡ − 5.8 (0.35)‡ − 1.0 (0.29) − 2.2 (0.33)* − 2.6 (0.36) † − 3.0 (0.37)‡ − 3.2 (0.39)‡ − 3.1 (0.39)‡
− 1.8 (0.57) − 2.7 (0.68) − 3.5 (0.75)† − 3.9 (0.79)† − 4.0 (0.82) † − 4.0 (0.85)† − 0.7 (0.38)* − 1.1 (0.45) † − 1.1 (0.48)‡ − 1.2 (0.50)‡ − 1.1(0.51) ‡ − 1.1 (0.52)‡ − 1.5 (0.39) − 2.4 (0.44)* − 2.4 (0.47)* − 2.4 (0.48)* − 2.6 (0.5)* − 2.4 (0.51)* − 2.3 (0.43) − 3.1 (0.51) − 3.7 (0.57)* − 4.2 (0.60)* − 4.4 (0.64)* − 4.3 (0.65)* − 1.0 (0.53) − 2.2 (0.6) − 2.4 (0.66) − 2.4 (0.69) − 1.7 (0.71) − 2.3 (0.72)
− 1.4 − 2.3 − 2.3 − 3.0
(0.45) (0.53) (0.58) (0.61)†
− 0.1 − 0.3 − 0.2 − 0.1
(0.3) (0.35)* (0.37)* (0.39)*
− 1 (0.21) − 1.4 (0.25) − 1.3 (0.27) − 1.3 (0.28) − 1.5 (0.37) − 1.6 (0.38) 0.2 (0.14) 0.5 (0.16) 0.7 (0.17) 0.9 (0.18) 0.9 (0.23) 1.0 (0.23) − 0.9 (0.14) − 1.3 (0.16) − 1.1 (0.17) − 1.3 (0.18) − 1.2 (0.22) − 1.1 (0.22) − 1.7 (0.15) − 2.4 (0.19) − 2.4 (0.21) − 2.5 (0.22) − 2.6 (0.28) − 2.7 (0.29) − 0.7 (0.19) − 1.4 (0.22) − 1.3 (0.24) − 1.4 (0.25) − 1.4 (0.31) − 1.3 (0.32)
− 1.1 (0.3) − 1.4(0.34) − 1.5 (0.37) − 1.7 (0.37)
− 2.3 − 3.5 − 3.8 − 3.9
(0.33) (0.4) † (0.44)† (0.47)†
− 2.0 − 2.3 − 2.5 − 2.2
(0.41) † (0.46) (0.51)* (0.53)
*P b 0.05; †P b 0.01; ‡ P b 0.001 vs. placebo. PANSS = Positive and Negative Syndrome Scale; SE = standard error.
4. Discussion 4.1. Overview Using this 5-factor PANSS model, clinical efficacy was observed in subjects with schizophrenia who received iloperidone 10–16 mg/d for 4 or 6 weeks in terms of a consistent separation from placebo for all 5 factors. Iloperidone 20–24 mg/d demonstrated separation from placebo on 3 of the 5 factors (positive, cognition and excitement/ hostility), but not on the depression/anxiety and negative factors. There was no clear evidence of a dose response for iloperidone above 4–8 mg/day across all the factors. The greatest reductions from baseline in 5-factor PANSS analysis scores were observed in the cognition and positive factors at both Weeks 4 and 6. Apart from some small numerical changes, the overall pattern of results was unchanged
when factor scores were computed for only those patients who completed greater than 2 weeks of treatment. Similar to iloperidone 10–16 mg/d, risperidone 4–8 mg/d also separated from placebo across all 5 factors, consistent with the results of a pooled analysis of two North American trials where risperidone was compared with placebo and haloperidol 20 mg/d (Marder et al., 1997). The model developed for risperidone was similar to the model we present here, with only minor differences. These differences include the cognitive factor being referred to as disorganized thought in the Marder model, and somatic concern (G1),stereotyped thinking (N7) and lack of judgment and insight (G12) being assigned to the positive factor in the Marder model rather than to the depression/ anxiety factor (G1) or the cognitive factor (N7, G12) in our model. The Marder model has been used by others for different study datasets, such as in a study of asenapine versus placebo with haloperidol as the
Table 7 Least-squared mean changes from baseline in the 5-factor PANSS model at Week 4 and Week 6. for all patients with schizophrenia who received treatment for more than 2 weeks. Factor scores for patients with N 2 weeks of treatment, (SE) Cognition Excitement/hostility Depression/anxiety Positive Negative
Week Week Week Week Week Week Week Week Week Week
Iloperidone
4 6 4 6 4 6 4 6 4 6
4–8 mg/d
10–16 mg/d
20–24 mg/d
− 2.6 − 2.9 − 1.0 − 0.8 − 2.3 − 2.4 − 3.5 − 3.9 − 2.0 − 2.1
− 3.7 (0.36)† − 4.1 (0.38)* − 1.4 (0.21)‡ − 1.3 (0.22)‡ − 2.8 (0.22)† − 2.7 (0.22)‡ − 4.4 (0.27)† − 4.9 (0.29)‡ − 3.2 (0.32)* − 2.9 (0.33)
− 3.8 − 4.8 − 0.9 − 1.4 − 2.1 − 2.5 − 4.2 − 5.0 − 2.7 − 3.3
(0.45) (0.47) (0.26)‡ (0.27)‡ (0.27) (0.28)* (0.35) (0.36) (0.41) (0.41)
*P b 0.05; †P b 0.01; ‡ P b 0.001 vs. placebo. PANSS = Positive and Negative Syndrome Scale; SE = standard error.
(0.42)† (0.76)* (0.25)‡ (0.43)‡ (0.26) (0.44) (0.32)* (0.58)* (0.38) (0.65)
Risperidone 4–8 mg/d
Haloperidol 15 mg/d
Ziprasidone 160 mg/d
Placebo
− 4.7 (0.45)‡ − 5.4 (0.50)‡ − 1.6 (0.26)‡ − 1.7 (0.28)‡ − 2.9 (0.27)† − 3.1 (0.29)‡ − 5.7 (0.34)‡ − 6.4 (0.38)‡ − 3.4 (0.40)* − 3.5 (0.43)*
− 5.4 (0.90)† − 5.6 (0.99)* − 2.2 (0.53)‡ − 2.1 (0.57)‡ − 2.9 (0.55) − 2.9 (0.58)* − 5.2 (0.69)* − 5.3 (0.75)* − 3.6 (0.81) − 3.6 (0.85)
− 4.1 (0.67)*
− 2.5 (0.32) − 2.9 (0.42) 0.2 (0.18) 0.4 (0.24) − 1.8 (0.19) − 1.6 (0.25) − 3.3 (0.24) − 3.5 (0.32) − 2.3 (0.28) − 2.2 (0.36)
− 0.7 (0.39)* − 2.0 (0.41) − 4.8 (0.51)† − 3.0 (0.60)
80
L. Citrome et al. / Schizophrenia Research 131 (2011) 75–81
Table 8 Percent who have improved by at least 20% in the 5-factor PANSS model at Week 1 and Week 2 for all patients with schizophrenia. Estimated percentage using Kaplan-Meier (SE) Cognition Excitement/hostility Depression/anxiety Positive Negative
Iloperidone
Week1 Week2 Week1 Week2 Week1 Week2 Week1 Week2 Week1 Week2
4–8 mg/d
10–16 mg/d
20–24 mg/d
9.5 (1.80) 18.4 (2.50) 20.9 (2.50) 34.7 (3.06) 24.9 (2.65) 37.2 (3.08) 18.8 (2.40) 35.1 (3.07) 15.9 (2.25) 29.1 (2.92)
10.5 26.8 22.1 41.3 20.9 37.5 17.6 40.9 14.0 31.6
8.2 (1.41) 24.9 (2.34) 21.5 (2.11) 40.3 (2.63) 17.0 (1.94) 34.3 (2.57) 18.5 (2.00) 39.5 (2.63) 14.8 (1.83) 31.7 (2.49)
(1.61) (2.45) (2.17) (2.71) (2.13) (2.66) (2.00) (2.72) (1.82) (2.55)
Risperidone 4–8 mg/d
Haloperidol 15 mg/d
Ziprasidone 160 mg/d
Placebo
17.8 32.7 29.5 45.9 26.3 40.9 25.8 56.0 15.2 30.3
15.4 29.2 23.1 49.0 27.2 53.6 32.3 53.5 13.8 37.6
14.7 28.0 28.2 42.7 23.2 44.9 21.2 49.0 23.4 46.6
11.5 27.5 21.9 35.9 22.8 42.0 21.4 40.0 14.7 32.6
(2.55) (3.20) (3.05) (3.43) (2.94) (3.35) (2.92) (3.41) (2.40) (3.16)
(4.47) (6.08) (5.22) (6.64) (5.47) (6.65) (5.80) (6.63) (4.28) (6.62)
(3.04) (3.99) (3.83) (4.39) (3.59) (4.39) (3.49) (4.56) (3.61) (4.51)
(1.55) (2.30) (2.00) (2.43) (2.03) (2.53) (1.99) (2.51) (1.72) (2.40)
PANSS = Positive and Negative Syndrome Scale; SE = standard error.
active control (Kane et al., 2010). We undertook a sensitivity analysis and reanalyzed our data using the Marder factors — the pattern of results was essentially unchanged. Haloperidol at a dose of 15 mg/d failed to separate statistically from placebo on the negative symptom factor; in the aforementioned study of risperidone, of all the PANSS factors the one that was associated with the smallest effect size versus placebo was for haloperidol and negative symptoms. In a study of patients with suboptimal response to antipsychotics, the negative factor demonstrated significant worsening for haloperidol (Lindenmayer et al., 2004). Ziprasidone 160 mg/d failed to statistically separate from placebo on the depression/anxiety and negative factors. The latter outcome is not consistent with another study that examined negative symptoms in patients with schizophrenia treated with ziprasidone or amisulpride, albeit for a longer period of time — 12 weeks (Olie et al., 2006). An important caveat is that the iloperidone registration trials for acute schizophrenia were powered to detect differences between iloperidone and placebo. The active controls, haloperidol, risperidone and ziprasidone, were used as active controls for assay sensitivity. Any dose response or drug-drug comparisons will require the conduct of appropriately designed and powered randomized controlled trials. 4.2. Safety and tolerability The PANSS factor analysis presented here is focused exclusively on antipsychotic efficacy. However real-world effectiveness is dependent on a medication's tolerability profile and a patient's willingness to actually adhere. Further information regarding iloperidone's tolerability profile can be found elsewhere (Citrome 2009; Citrome 2010; Novartis Pharmaceuticals Corporation 2011). 4.3. Limitations Although the overall numbers of subjects available for this analysis was substantial, allowing for the construction of a robust 5 factor PANSS model, the numbers available for the efficacy analyses were reduced as there were multiple treatment arms, including 3 dose ranges for iloperidone. The pooling of 4 short-term efficacy studies also assumed that the study populations were comparable, however this may not be entirely the case as the three 6-week studies that included schizoaffective disordered patients (studies ILP3000ST, ILP3004ST, and ILP3005ST; Table 1) were done in a different epoch of drug development than the 4-week study (study VP-VYY-683-3101). Furthermore, at five and six weeks post-baseline, the 20–24 mg dose group had almost thirty percent as many evaluable patients as the other dose groups, which reduced statistical power to separate from placebo. The analysis of schizoaffective patients enrolled in studies ILP3000ST, ILP3004ST, ILP3005ST, although smaller in number than the subjects diagnosed with schizophrenia, may yield different results
on the 5 factors, particularly the depression/anxiety factor. However, diagnostic uncertainty regarding the differentiation of schizophrenia from schizoaffective disorder may make this type of analysis difficult to interpret (Kantrowitz & Citrome, 2011). 5. Conclusions In a pooled sample of patients with acute schizophrenia, iloperidone demonstrated positive treatment effects across the multiple dimensions of psychopathology commonly associated with schizophrenia: positive, negative, cognitive, excitement/hostility and depression/anxiety. The dose range of 4–8 mg/d was sub-therapeutic. The two higher doses had similar efficacy on the PANSS factors, with the exception of the depression/anxiety and negative factors, on which the 10–16 mg dose group showed statistical separation from placebo and the higher dose group did not. At 5 and 6 weeks, the lack of separation from placebo for the higher dose group may have been due to the much smaller sample size in that group. Further studies examining which symptom dimensions iloperidone is best suited for will require specifically designed studies that include patients from the specific populations in question, for example patients with schizophrenia with predominant negative, cognitive, or depressive symptoms. Role of funding source Funding for this study was provided by Novartis Pharmaceuticals Corporation. Employees of Novartis Pharmaceuticals Corporation were involved in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.
Contributors Leslie Citrome, Xiangyi Meng, and Marla Hochfeld participated in the design and execution of the study, as well as in the preparation of the poster presentation that preceded the paper. Xiangyi Meng undertook the statistical analyses. Leslie Citrome wrote the first draft of the manuscript. All authors contributed to and have approved the final manuscript.
Conflicts of interest 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, BristolMyers Squibb, Eli Lilly and Company, Forest Research Institute, GlaxoSmithKline, Janssen Pharmaceuticals, Jazz Pharmaceuticals, Merck, Novartis Pharmaceuticals Corporation, Noven Pharmaceuticals, Pfizer Inc, Sunovion, Valeant Pharmaceuticals and Vanda Pharmaceuticals. Xiangyi Meng, and Marla Hochfeld are full-time employees and stock holders of Novartis Pharmaceuticals Corporation.
Acknowledgments This work was funded by Novartis Pharmaceuticals Corporation. Presented as a poster at the 23rd United States Psychiatric and Mental Health Congress, Orlando, Florida, November 18–21, 2010.
L. Citrome et al. / Schizophrenia Research 131 (2011) 75–81
References Citrome, L., 2009. Iloperidone for schizophrenia: A review of the efficacy and safety profile for this newly-commercialised second-generation antipsychotic. Int. J. Clin. Pract. 63 (8), 1237–1248. Citrome, L., 2010. Iloperidone redux: a dissection of the Drug Approval Package for this newly commercialised second-generation antipsychotic. Int. J. Clin. Pract. 64 (6), 707–718. Cutler, A.J., Kalali, A.H., Weiden, P.J., Hamilton, J., Wolfgang, C.D., 2008. Four-week, doubleblind, placebo- and ziprasidone-controlled trial of iloperidone in patients with acute exacerbations of schizophrenia. J. Clin. Psychopharmacol. 28 (2 Suppl 1), S20–S28. Kane, J.M., Lauriello, J., Laska, E., DiMarino, M., Wolfgang, C.D., 2008. Long-term efficacy and safety of iloperidone: results from 3 clinical trials for the treatment of schizophrenia. J. Clin. Psychopharmacol. 28 (2 Suppl 1), S29–S35. Kane, J.M., Cohen, M., Zhao, J., Alphs, L., Panagides, J., 2010. Efficacy and safety of asenapine in a placebo- and haloperidol-controlled trial in patients with acute exacerbation of schizophrenia. J. Clin. Psychopharmacol. 30 (2), 106–115. Kantrowitz, J.T., Citrome, L., 2011. Schizoaffective disorder, a review of current research themes and pharmacologic management. CNS Drugs 25 (4), 317–331. Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr. Bull. 13 (2), 261–276. Lançon, C., Aghababian, V., Llorca, P.M., Auquier, P., 1998. Factorial structure of the Positive and Negative Syndrome Scale (PANSS): a forced five-dimensional factor analysis. Acta Psychiatr. Scand. 98 (5), 369–376. Lehoux, C., Gobeil, M.H., Lefebvre, A.A., Maziade, M., Roy, M.A., 2009. The Five-Factor Structure of the PANSS: A Critical Review of its Consistency Across Studies. Clin. Schizophr. Relat. Psychoses 3 (2), 103–110.
81
Levine, S.Z., Rabinowitz, J., 2007. Revisiting the 5 dimensions of the Positive and Negative Syndrome Scale. J. Clin. Psychopharmacol. 27 (5), 431–436. Lindenmayer, J.P., Bernstein-Hyman, R., Grochowski, S., 1994. A newfive factor model of schizophrenia. Psychiatr. Q. 65 (4), 299–322. Lindenmayer, J.P., Czobor, P., Volavka, J., et al., 2004. Effects of Atypical Antipsychotics on the Syndromal Profile in Treatment-Resistant Schizophrenia. J. Clin. Psychiatry 65 (4), 551–556. Marder, S.R., Davis, J.M., Chouinard, G., 1997. The effects of risperidone on the five dimensions of schizophrenia derived by factor analysis: combined results of the North American trials. J. Clin. Psychiatry 58 (12), 538–546. Novartis Pharmaceuticals Corporation, 2011. Fanapt (iloperidone) Tablets. Revised 4/2011 (accessed 03 June 2011) http://www.pharma.us.novartis.com/product/pi/pdf/ fanapt.pdf. Olie, J.P., Spina, E., Murray, S., Yang, R., 2006. Ziprasidone and amisulpride effectively treat negative symptoms: results of a 12-week, double-blind study. Int. Clin. Psychopharamacol. 21 (3), 143–151. Potkin, S.G., Litman, R.E., Torres, R., Wolfgang, C.D., 2008. Efficacy of iloperidone in the treatment of schizophrenia: initial phase 3 studies. J. Clin. Psychopharmacol. 28 (2 Suppl 1), S4–S11. Van den Oord, E.J., Rujescu, D., Robles, J.R., et al., 2006. Factor structure and external validity of the PANSS revisited. Schizophr. Res. 82 (2–3), 213–223. White, L., Harvey, P.D., Opler, L., Lindenmayer, J.P., 1997. Empirical assessment of the factorial structure of clinical in schizophrenia.A multisite, multimodel evaluation of the factorial structure of the Positive and Negative Syndrome Scale. The PANSS Study Group. Psychopathology 30 (5), 263–274.
Contents lists available at ScienceDirect
Schizophrenia Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c h r e s
Efficacy of iloperidone in schizophrenia: A PANSS five-factor analysis Leslie Citrome a,⁎, Xiangyi Meng b, Marla Hochfeld b a b
New York University School of Medicine, New York, NY, USA Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA
a r t i c l e
i n f o
Article history: Received 20 January 2011 Received in revised form 19 May 2011 Accepted 20 May 2011 Available online 22 June 2011 Keywords: Factor Iloperidone PANSS Schizophrenia
a b s t r a c t Background: The Positive and Negative Syndrome Scale (PANSS) total score is widely used to assess antipsychotic efficacy, however schizophrenia is a multi-dimensional disorder. We conducted a 5-factor analysis for evaluating the efficacy of iloperidone vs. placebo across these different domains in the treatment of schizophrenia. Method: The 5-factor model was determined from pooled data from 7 clinical trials (4 placebo- and activecontrolled and 3 non-inferiority active-comparator trials of iloperidone) in schizophrenia (N = 3580).Five factors were derived (excitement/hostility [P4,P7,G8,G14], depression/anxiety [G1,G2,G3,G4,G6], cognition [P2,N5,N7,G5,G10,G11,G12,G13,G15], positive [P1,P3,P5,P6,G9], and negative [N1,N2,N3,N4,N6,G7,G16]) from a factor analysis on the covariance matrix of 30 baseline PANSS items using a varimax rotation; factors retained had eigenvalues of ≥0.5. These newly derived 5 factors differ only slightly from other 5-factor analyses published by others using different datasets. The analysis of covariance model was then applied to assess these efficacy outcomes from the 4–6 week double-blind placebo and active controlled clinical trials of iloperidone. Results: Based on the placebo-controlled trials, iloperidone improvements from baseline (least squared mean change± standard error) were as follows: excitement/hostility, 0.4 ± 0.21 for 10–16 mg, 0.6 ± 0.43 for 20–24 mg vs. −1.0 ± 0.23 for placebo; P b 0.001 for both iloperidone doses vs. placebo; depression/anxiety, 1.9 ± 0.21 for 10–16 mg, 1.9 ± 0.41 for 20–24 mg vs. 1.1 ± 0.22 for placebo; P b 0.05 for 10–16 mg dose vs. placebo; cognition, 2.8 ± 0.35 for 10–16 mg, 3.9 ± 0.69 for 20–24 mg vs. 1.6 ± 0.38 for placebo; P b 0.05 for both iloperidone doses vs. placebo; positive, 3.7 ± 0.26 for 10–16 mg, 4.1 ± 0.53 for 20–24 mg vs. 2.7 ± 0.29 for placebo; P b 0.05 for both iloperidone doses vs. placebo; and negative, 2.2 ± 0.29 for 10–16 mg, 2.5 ± 0.58 for 20–24 mg vs. 1.3 ± 0.32 for placebo; P b 0.05 for 10–16 mg vs. placebo. Active controls validated iloperidone efficacy. Conclusions: Iloperidone demonstrated positive treatment effects on these newly derived PANSS factors. The 10–16 mg and 20–24 mg dose groups had similar efficacy on the PANSS factors, with the exception of the depression/anxiety and negative factors, on which the 10–16 mg dose group showed statistical separation from placebo and the 20–24 mg dose group did not. At 6 weeks, the lack of separation from placebo for the higher dose group may have been due to the much smaller sample size in that group. © 2011 Elsevier B.V. All rights reserved.
1. Introduction The 30-item Positive and Negative Syndrome Scale (PANSS) (Kay et al., 1987)is commonly used in clinical trials to assess the psychopathology associated with schizophrenia. Although the PANSS has 3 subscales – positive symptoms, negative symptoms, and general psychopathology – some of the items assigned to one subscale may perhaps be better conceptualized as belonging to another symptom construct such as “mood” or “cognitive”. This, together with the growing understanding that schizophrenia is a multidimensional disorder with core psychotic, negative, mood, and
⁎ Corresponding author at: 11 Medical Park Drive, Suite 106, Pomona, NY 10970. Tel.: + 1 845 362 2081; fax: + 1 845 362 8745. E-mail addresses: [email protected] (L. Citrome), [email protected] (X. Meng), [email protected] (M. Hochfeld). 0920-9964/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2011.05.018
cognitive components has led to the development of multiple-factor models of the PANSS (Lindenmayer et al., 1994; White et al., 1997; Marder et al., 1997; Van den Oord et al., 2006; see review by Lehoux et al., 2009). These can be used to describe the outcomes of clinical trials for the treatment of schizophrenia and discern differences in efficacy based on symptom dimensions (Lindenmayer et al., 2004). Iloperidone is a second-generation antipsychotic that in 2009 received US Food and Drug Administration approval and has a current indication for the treatment of schizophrenia in adults (Citrome 2009; Citrome 2010; Novartis Pharmaceuticals Corporation 2011). A pooled analysis was conducted to identify 5 PANSS factors using baseline ratings of 3580 subjects with schizophrenia (excluding schizoaffective disorder) across 7 studies (3 non-inferiority active-controlled and 4 placebo- and active-controlled Phase III studies involving iloperidone). After deriving these 5 PANSS factors, the panels were then used to evaluate the efficacy of iloperidone in subjects with schizophrenia from 4 pooled, placeboand active-controlled clinical trials of 4 or 6 weeks’ duration.
76
L. Citrome et al. / Schizophrenia Research 131 (2011) 75–81
2. Method 2.1. Design and objective This was a post hoc analysis of pooled data from 7 pivotal trials of iloperidone in order to build a 5-factor PANSS model and to subsequently apply it to assess the multi-dimensional psychopathological outcomes across the 4 short-term controlled trials of iloperidone that were originally designed to test the overall efficacy of iloperidone versus placebo in subjects with schizophrenia. 2.2. Data sources The PANSS factor analysis was conducted using 7 placebo- and active-controlled studies including 3582 subjects with schizophrenia (Table 1). Because 6 of the 7 studies also included subjects with schizoaffective disorder, the analysis was repeated using baseline ratings of 4150 subjects with schizophrenia or schizoaffective disorder to determine whether or not this would have changed the factor structure in any substantial manner. The rationale for constructing a 5-factor model from this dataset was to maximize the internal validity for any subsequent analysis involving change in factor scores over time. All studies that were included enrolled acute patients; the longer term studies (Kane et al., 2008) required subjects to demonstrate improvement during a 6-week stabilization phase in order to be eligible to participate in the 46-week maintenance phase. Four randomized, double-blind, placebo- and active-controlled, multicenter studies that evaluated the efficacy and safety of iloperidone for either 6 weeks (3 studies) or 4 weeks (1 study) were evaluated using the newly derived 5-factor PANSS analysis (Table 1). Male or female subjects in these 4 studies were aged 18–65 years with a Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) diagnosis of schizophrenia (all studies) or schizoaffective disorder (only the 6-week studies); data from subjects with schizophrenia but not
schizoaffective disorder were pooled and used for the present analysis. For the efficacy analyses, all randomized subjects with schizophrenia and at least one post-baseline efficacy measurement obtained while on study medication were eligible for inclusion. Note that the data source for the efficacy analysis is a subset of the data source that was used for the derivation of the PANSS factors. 2.3. Data analysis Similar to prior analyses by other investigators (Lançon et al., 1998; Van den Oord et al., 2006; Levine and Rabinowitz, 2007), factors were derived from an analysis on the covariance matrix of 30 baseline PANSS items using a varimax rotation. Five factors were retained that had eigenvalues of ≥ 0.5 (Table 2). A total of 7 treatment groups were subsequently evaluated for efficacy using the newly derived 5 factors: placebo (n = 447), iloperidone 4–8 mg/d (n= 276), iloperidone 10–16 mg/d (n= 381), and iloperidone 20–24 mg/d (n = 394), with active controls for assay sensitivity haloperidol 15 mg/d (n = 70), risperidone 4–8 mg/d (n = 229), and ziprasidone 160 mg/d (n = 144). Dose groups for iloperidone were selected based on the doses tested in the individual studies. Numbers of subjects per group for each time point are provided in Table 3. The changes from baseline to study end point (Week 4 or Week 6, depending on the study) for each of the 5 PANSS factors were evaluated using an analysis of covariance (ANCOVA) model with treatment and study as factors and baseline as a covariate. Last observations before Week 4 or 6 were carried forward until Week 4 or 6, respectively. For any missing observations, the last recorded measurement was used. Because differential dropout rates, possibly related to titration, were observed between iloperidone and other treatments at the start of one pivotal trial (Potkin et al., 2008), additional analyses were also carried out for the subset of subjects who continued study treatment for more than 2 weeks.
Table 1 Studies used for the 5-factor Positive and Negative Syndrome Scale analysis and the subsequent efficacy analyses. Duration (weeks)
Descriptiona
N with schizophrenia available for analysis
Factor analysis only ILP3001 (Kane et al., 2008)
52
548
ILP3002 (Kane et al., 2008)
52
ILP3003 (Kane et al., 2008)
52
International, multicenter, double-blind, randomized, parallel-group study to compare the antipsychotic effect of iloperidone 4–16 mg/d with that of haloperidol 5–20 mg/d in 600 patients with schizophrenia or schizoaffective disorder (454 iloperidone, 146 haloperidol) International, multicenter, double-blind, randomized, parallel-group study to compare the antipsychotic effect of iloperidone 4–16 mg/d with that of haloperidol 5–20 mg/d in 557 patients with schizophrenia or schizoaffective disorder (420 iloperidone, 137 haloperidol) International, multicenter, double-blind, randomized, parallel-group study to compare the antipsychotic effect of iloperidone 4–16 mg/d with that of haloperidol 5–20 mg/d in 487 patients with schizophrenia or schizoaffective disorder (365 iloperidone, 122 haloperidol)
US and India, multicenter, double-blind, randomized, parallel-group, placebo- and activecontrolled study to evaluate the efficacy of iloperidone 24 mg/d in 606 patients with schizophrenia (303 iloperidone, 151 ziprasidone 160 mg/d, 152 placebo) US, multicenter, double-blind, randomized, parallel-group, placebo- and active-controlled study to evaluate the efficacy of three fixed doses of iloperidone (4, 8, and 12 mg/d) in 621 patients with schizophrenia or schizoaffective disorder (370 iloperidone, 124 haloperidol 15 mg/d, 127 placebo) International, multicenter, double-blind, randomized, parallel-group, placebo- and activecontrolled study to evaluate the efficacy of two non-overlapping dose ranges of iloperidone (4–8 mg/d and 10–16 mg/d) in 616 patients with schizophrenia or schizoaffective disorder (307 iloperidone, 153 risperidone 4–8 mg/d, 156 placebo) International, multicenter, double-blind, randomized, parallel-group, placebo- and activecontrolled study to evaluate the efficacy of two non-overlapping dose ranges of iloperidone (12–16 mg/d and 20–24 mg/d) in 706 patients with schizophrenia or schizoaffective disorder (389 iloperidone, 157 risperidone 6–8 mg/d, 160 placebo)
597
Study
Factor and efficacy analyses VP-VYY-683-3101 (Cutler et al., 2008)
4
ILP3000ST (Potkin et al., 2008)
6
ILP3004ST (Potkin et al., 2008)
6
ILP3005ST (Potkin et al., 2008)
6
a
Doses of all medications administered twice daily. All subjects were adults.
524
470
422
479
542
L. Citrome et al. / Schizophrenia Research 131 (2011) 75–81
77
Table 2 Positive and Negative Syndrome Scale factor analysis results. PANSS Item
Varimax Rotations (eigenvalues)
P2: Conceptual Disorganization N5: Difficulty in Abstract Thinking N7: Stereotyped Thinking G5: Mannerisms and Posturing G10: Disorientation G11: Poor Attention G12: Lack of Judgment and Insight G13: Disturbance of Volition G15: Preoccupation G1: Somatic Concern G2: Anxiety G3: Guilt Feelings G4: Tension G6: Depression P4: Excitement P7: Hostility G8: Uncooperativeness G14: Poor Impulse Control N1: Blunted Affect N2: Emotional Withdrawal N3: Poor Rapport N4: Passive/Apathetic Social Withdrawal N6: Lack of Spontaneity and Flow of Conversation G7: Motor Retardation G16: Active Social Avoidance G9: Unusual Thought Content P1: Delusions P3: Hallucinatory Behavior P5: Grandiosity P6: Suspiciousness/Persecution
Factor Category
Factor 1 (6.21)
Factor 2 (3.23)
Factor 3 (1.74)
Factor 4 (1.14)
Factor 5 (0.80)
0.06927 0.21867 0.20526 0.19993 0.12376 0.18291 0.18402 0.36260 0.25600 − 0.02155 − 0.00590 0.01528 0.00764 0.15698 − 0.20282 0.10129 0.29846 0.00134 0.65576 0.77740 0.65400 0.77805 0.67097 0.55519 0.63294 − 0.02886 − 0.08101 0.04786 − 0.24742 0.11006
0.61525 0.54674 0.54540 0.45492 0.37966 0.62405 0.42130 0.47473 0.42331 0.12663 0.03426 − 0.06280 0.22590 − 0.13051 0.27643 0.08825 0.26600 0.22037 0.27410 0.18205 0.34115 0.16360 0.34243 0.21073 0.07630 0.37076 0.10408 0.10024 0.15138 − 0.05973
0.13850 0.01915 0.15531 0.11487 0.07067 0.20043 0.16825 0.08457 0.12138 0.01390 0.21617 0.01051 0.35765 − 0.01561 0.55246 0.70928 0.56788 0.63940 − 0.07029 0.07017 0.18664 0.03613 0.02111 − 0.13534 0.18413 0.08478 0.08991 0.03505 0.16870 0.24141
0.26177 0.09480 0.17789 0.02826 0.04833 0.04788 0.22465 − 0.05454 0.27014 0.17183 0.09695 0.10293 0.04176 0.03593 0.16913 0.17792 0.03908 0.12032 0.03375 0.04081 0.07792 0.05842 − 0.13576 − 0.13651 0.14994 0.61985 0.76073 0.50704 0.38348 0.48263
− 0.03064 − 0.09024 0.03259 0.08661 0.00388 0.06184 − 0.22885 0.10842 0.12712 0.33807 0.70026 0.53079 0.55620 0.59553 0.23172 0.10456 − 0.03990 0.14949 0.00348 0.00350 − 0.07110 − 0.00345 − 0.01503 0.14453 0.12745 0.03874 0.13911 0.14380 0.01453 0.31324
Cognition Cognition Cognition Cognition Cognition Cognition Cognition Cognition Cognition Depression/Anxiety Depression/Anxiety Depression/Anxiety Depression/Anxiety Depression/Anxiety Excitement/Hostility Excitement/Hostility Excitement/Hostility Excitement/Hostility Negative Negative Negative Negative Negative Negative Negative Positive Positive Positive Positive Positive
G = general; N = negative; P = positive; PANSS = Positive and Negative Syndrome Scale.
3. Results 3.1. Demographics and descriptive data A total of 3580 subjects with schizophrenia were included when constructing the 5-factor model (2 subjects excluded due to missing baseline data). The efficacy analysis included 1941 subjects (see Table 3). Subject demographic and baseline characteristics for the population are shown in Table 4. Disposition is provided in Table 5. 3.2. PANSS factor analysis Five factors were derived from PANSS analysis for 3580 patients with schizophrenia: excitement/hostility, anxiety/depression, cognition, positive, and negative (see Table 2). The robustness of these
factors was supported by a similar pooled analysis of the same 7 studies using baseline ratings of 4150 subjects with schizophrenia or schizoaffective disorder, which yielded the same 5 factors. The variance explained by each factor was 50%, 26%, 14%, 9% and 6% for the negative, cognition, excitement/hostility, positive and anxiety/ depression factor, respectively. 3.3. Iloperidone efficacy Overall, subjects receiving iloperidone showed greater changes from baseline in all 5 PANSS factor analysis scores vs. those receiving placebo. Among all subjects, those receiving iloperidone 10–16 mg/d showed significant improvement at Week 4 and Week 6 on all factors vs. placebo. Subjects receiving iloperidone 20–24 mg/d showed significant improvement at Week 4 and Week 6 on the cognition,
Table 3 Numbers of patients analyzed for efficacy measurements by week and treatment group for the total population and patients receiving treatment for greater than 2 weeks a. Week
All patients
Patients treated for N2 weeks
1 2 3 4 5 6 1 2 3 4 5 6
Iloperidone 4–8 mg/d
10–16 mg/d
20–24 mg/d
276 276 276 276 276 276 217 217 217 217 217 217
379 380 380 380 380 380 294 295 295 295 295 295
392 394 394 394 111 111 315 316 316 316 90 90
to 381 or 381 or 381 or 381 or 381 or 381 or 295
Risperidone 4–8 mg/d
Haloperidol 15 mg/d
Ziprasidone 160 mg/d
Placebo
229 229 229 229 229 229 194 194 194 194 194 194
70 70 70 70 70 70 49 49 49 49 49 49
141 143 143 143 0 0 108 109 109 109 0 0
444 446 446 446 306 306 337 339 339 339 233 233
Note: Weeks 1–4 include data pooled from 4 clinical trials, while Weeks 5–6 include data pooled from 3 trials. a Variations within a given week due to missing efficacy data.
or or or or
142 144 144 144
or or or or
109 110 110 110
or or or or or or or or or or or or
445 447 447 447 307 307 338 340 340 340 234 234
78
L. Citrome et al. / Schizophrenia Research 131 (2011) 75–81
Table 4 Patient characteristics (all patients with schizophrenia in the 4 efficacy studies with evaluable data)a. Characteristic
Risperidone 4–8 mg/d (n = 230)
Haloperidol 15 mg/d (n = 70)
Ziprasidone 160 mg/d (n = 144)
Placebo (n = 447)
315 (80) 39 (18–65)
162 (70) 38.5 (17–67)
53 (76) 40 (19–59)
108 (75) 40 (20–61)
320 (72) 39 (18–66)
248 (65) 106 (28) 8 (2) 19 (5)
184 (47) 166 (42) 27 (7) 17 (4)
161 (70) 54 (24) 3 (1) 12 (5)
29 (41) 35 (50) 2 (3) 4 (6)
50 72 12 10
(35) (50) (8) (7)
217 (49) 182 (41) 19 (4) 29 (7)
60 (16) 151 (40) 157 (41) 10 (3) 3 (1)
67 (17) 179 (45) 138 (35) 8 (2) 2 (1)
38 (17) 88 (38) 96 (42) 6 (3) 2 (1)
11 (16) 29 (41) 28 (40) 1 (1) 1 (1)
24 (17) 56 (39) 61 (42) 1 (1) 2 (1)
75 (17) 198 (44) 153 (34) 13 (3) 8 (2)
17 (5) 302 (79) 1 (0.3) 61 (16) 94.3 (14.7)
23 (6) 323 (82) 0 48 (12) 93.5 (14.0)
16 (7) 179 (78) 0 35 (15) 95.7 (15.9) [n = 229]
2 (3) 56 (80) 0 12 (17) 98.3 (15.5)
3 (2) 122 (85) 0 19 (13) 90.9 (11.6)
23 (5) 371 (83) 0 53 (12) 93.8 (15.4)
54.4 (8.3)
54.7 (8.1)
55.0 (9.2) [n = 229]
56.9 (9.0)
53.3 (6.8)
54.3 (8.8) [n = 444]
4.8 (0.7) [n = 281]
4.8 (0.7) [n = 386]
4.8 (0.7) [n = 215]
NA
4.6 (0.7) [n = 142]
4.7 (0.8) [n = 356]
Iloperidone 4–8 mg/d (n = 275)b
Male, n (%) 205 (75) Median age, 39 (18–68) yr (range) Race, n (%) White 131 (48) Black 119 (43) Asian 6 (2) Other 19 (7) Age at diagnosis, n (%) b18 yr 59 (22) 18–24 yr 110 (40) 24–44 yr 98 (36) 45–66 yr 4 (2) Missing 4 (2) DSM-IV schizophrenia classification, n (%) 10 Disorganized 25 (9) 30 Paranoid 203 (74) 60 Residual 0 90 Undifferentiated 47 (17) 96.2 (15.7) Mean baseline PANSS total score (SD) 55.7 (8.9) Mean baseline BPRSd total score (SD) Mean baseline CGI-S 4.8 (0.7) score (SD)c [n = 105]
10–16 mg/d (n = 381)
20–24 mg/d (n = 394)
261 (69) 40 (18–68)
BPRSd = PANSS-derived Brief Psychiatric Rating Scale; CGI-S = Clinical Global Impression of Severity; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, 4th Edition; NA = Not assessed; PANSS = Positive and Negative Symptoms Scale; SD = standard deviation. a Due to missing data points, not all patients were included in the efficacy analysis. b Excludes 1 patient included in efficacy assessments. c Evaluated in 3 studies only.
excitement/hostility, and positive factors. The largest least-squared mean (Standard Error) reductions from baseline in subjects treated with iloperidone were observed in the cognition and positive factors at Week 6, as follows: cognition, all subjects: −2.8 (0.35) with iloperidone 10– 16 mg/d and −3.9 (0.69) with iloperidone 20–24 mg/d (Table 6); cognition, subjects treated N 2 weeks: -4.1 (0.38) with iloperidone 10– 16 mg/d and −4.8 (0.76) with iloperidone 20–24 mg/d (Table 7); positive, all subjects: −3.7 (0.26) with iloperidone 10–16 mg/d and −4.1 (0.53) with iloperidone 20–24 mg/d (Table 6); positive, subjects treated N 2 weeks: −4.9 (0.29) with iloperidone 10–16 mg/d and −5.0 (0.58) with iloperidone 20–24 mg/d (Table 7). Larger numeric
improvements on all 5 factors were observed in subjects receiving treatment for more than 2 weeks compared to the entire evaluated population (Table 7). Subjects receiving active controls (risperidone, haloperidol, and ziprasidone) also showed significant changes from baseline vs. placebo in the 5-factor PANSS scores. Iloperidone in the dose range of 4–8 mg/d was not efficacious, with the exception of the excitement/hostility factor. Speed of improvement on the different factors is qualitatively slower for iloperidone. Categorical results for an at least 20% improvement from baseline by 1 and 2 weeks for each intervention on each factor is outlined in Table 8.
Table 5 Patient disposition (all patients with schizophrenia in the 4 efficacy studies). Iloperidone
Patients randomized Completers Discontinued Missing Primary reason for discontinuation Protocol deviation Adverse event(s) a Lost to follow-up Death Patient withdrew consent Unsatisfactory therapeutic effect Other b a b
4–8 mg/d (n = 300)
10–16 mg/d (n = 402)
20–24 mg/d (n = 409)
300 (100.0) 124 (41.3) 176 (58.7) 0
402 207 193 2
(100.0) (51.5) (48.0) (0.5)
409 264 144 1
3 (1.0) 14 (4.7) 15 (5.0) 0 56 (18.7) 86 (28.7) 2 (0.7)
4 15 10 0 56 104 4
(1.0) (3.7) (2.5) (13.9) (25.9) (1.0)
3 22 3 0 69 43 4
(100.0) (64.5) (35.2) (0.2)
(0.7) (5.4) (0.7) (16.9) (10.5) (1.0)
Includes Adverse Events, Abnormal Laboratory Value, Abnormal Test Procedure, and Pregnancy. Includes Condition No Longer Requires Study Drug, Administrative Problem, and Other reasons.
Risperidone 4–8 mg/d (n = 241)
Haloperidol 15 mg/d (n = 78)
Ziprasidone 160 mg/d (n = 149)
Placebo (n = 472)
241 (100.0) 157 (65.1) 83 (34.4) 1 (0.4)
78 (100.0) 33 (42.3) 45 (57.7) 0
149 (100.0) 98 (65.8) 51 (34.2) 0
472 (100.0) 235 (49.8) 236 (50.0) 1 (0.2)
3 (1.2) 14 (5.8) 15 (6.2) 0 24 (10.0) 27 (11.2) 0
2 (2.6) 8 (10.3) 4 (5.1) 0 18 (23.1) 13 (16.7) 0
1 (0.7) 13 (8.7) 0 0 23 (15.4) 12 (8.1) 2 (1.3)
2 (0.4) 28 (5.9) 12 (2.5) 0 54 (11.4) 133 (28.2) 8 (1.7)
L. Citrome et al. / Schizophrenia Research 131 (2011) 75–81
79
Table 6 Least-squared mean changes from baseline in the 5-factor PANSS model at Weeks 1 through 6for all patients with schizophrenia. Factor scores for all patients, (SE) Cognition
Excitement/ hostility
Depression/ anxiety
Positive
Negative
Iloperidone
Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week Week
1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6
4–8 mg/d
10–16 mg/d
20–24 mg/d
− 0.6 (0.3) − 0.9 (0.36) − 1.4 (0.39) − 1.6 (0.42) − 1.7 (0.42) − 1.8 (0.43) 0.0 (0.2) − 0.1 (0.24) − 0.1 (0.25)* − 0.2 (0.26)‡ − 0.2 (0.26)† − 0.1 (0.26)† − 0.9 (0.21) − 1.1 (0.23) − 1.4 (0.25) − 1.6 (0.26) − 1.7 (0.25) − 1.7 (0.26) − 1.3 (0.23) − 1.8 (0.27) − 2.3 (0.3) − 2.6 (0.32) − 3.0 (0.32) − 3.0 (0.33) − 1.1 (0.28) − 1.3 (0.32) − 1.5 (0.35) − 1.4 (0.36) − 1.4 (0.36) − 1.5 (0.36)
−0.8 (0.24) − 1.6 (0.28) − 2.2 (0.31)* − 2.5 (0.33)† − 2.7 (0.34)* − 2.8 (0.35)* − 0.1 (0.16) − 0.2 (0.19)† − 0.4 (0.2)‡ − 0.5 (0.21)‡ − 0.4 (0.21)‡ − 0.4 (0.21)‡ − 0.9 (0.16) − 1.3 (0.18) − 1.8 (0.2)* − 2.0 (0.20)* − 1.8 (0.2)* − 1.9 (0.21)* −1.5 (0.18) − 2.3 (0.21) − 3.0 (0.24) − 3.3 (0.25)* − 3.6 (0.26)† − 3.7 (0.26)† − 1.0 (0.22) − 1.8 (0.25) − 2.2 (0.27)* − 2.4 (0.29)* − 2.2 (0.29)* − 2.2 (0.29)*
− 1.1 − 2.3 − 2.7 − 3.0 − 3.5 − 3.9 0.2 − 0.2 − 0.2 − 0.3 − 0.5 − 0.6 − 0.7 − 1.3 − 1.3 − 1.6 − 1.8 − 1.9 − 1.7 − 2.9 − 3.3 − 3.6 − 3.9 − 4.1 − 1.3 − 1.8 − 1.9 − 2.2 − 2.4 − 2.5
(0.29) (0.34)* (0.37) † (0.39)‡ (0.67) † (0.69)† (0.19) (0.22)* (0.24)† (0.25)‡ (0.42)† (0.43)‡ (0.19) (0.22) (0.24) (0.24) (0.4) (0.41) (0.21) (0.26) (0.28)† (0.30)† (0.52)* (0.53)* (0.26) (0.3) (0.33) (0.34) (0.58) (0.58)
Risperidone 4–8 mg/d
Haloperidol 15 mg/d
Ziprasidone 160 mg/d
Placebo
− 1.9 (0.31)* − 3.5 (0.37) ‡ − 3.8 (0.41)‡ − 4.4 (0.43)‡ − 4.7 (0.45)‡ − 4.9 (0.47)‡ − 0.9 (0.21)‡ − 1.3 (0.25)‡ − 1.4 (0.26)‡ − 1.5 (0.27)‡ − 1.6 (0.28)‡ − 1.5 (0.29)‡ − 1.5 (0.21)* − 2.0 (0.24)* − 2.3 (0.26)‡ − 2.5 (0.26)‡ − 2.7 (0.27)‡ − 2.8 (0.28)‡ − 2.5 (0.23) † − 4.1 (0.28) ‡ − 4.6 (0.31)‡ − 5.2 (0.33)‡ − 5.5 (0.35)‡ − 5.8 (0.35)‡ − 1.0 (0.29) − 2.2 (0.33)* − 2.6 (0.36) † − 3.0 (0.37)‡ − 3.2 (0.39)‡ − 3.1 (0.39)‡
− 1.8 (0.57) − 2.7 (0.68) − 3.5 (0.75)† − 3.9 (0.79)† − 4.0 (0.82) † − 4.0 (0.85)† − 0.7 (0.38)* − 1.1 (0.45) † − 1.1 (0.48)‡ − 1.2 (0.50)‡ − 1.1(0.51) ‡ − 1.1 (0.52)‡ − 1.5 (0.39) − 2.4 (0.44)* − 2.4 (0.47)* − 2.4 (0.48)* − 2.6 (0.5)* − 2.4 (0.51)* − 2.3 (0.43) − 3.1 (0.51) − 3.7 (0.57)* − 4.2 (0.60)* − 4.4 (0.64)* − 4.3 (0.65)* − 1.0 (0.53) − 2.2 (0.6) − 2.4 (0.66) − 2.4 (0.69) − 1.7 (0.71) − 2.3 (0.72)
− 1.4 − 2.3 − 2.3 − 3.0
(0.45) (0.53) (0.58) (0.61)†
− 0.1 − 0.3 − 0.2 − 0.1
(0.3) (0.35)* (0.37)* (0.39)*
− 1 (0.21) − 1.4 (0.25) − 1.3 (0.27) − 1.3 (0.28) − 1.5 (0.37) − 1.6 (0.38) 0.2 (0.14) 0.5 (0.16) 0.7 (0.17) 0.9 (0.18) 0.9 (0.23) 1.0 (0.23) − 0.9 (0.14) − 1.3 (0.16) − 1.1 (0.17) − 1.3 (0.18) − 1.2 (0.22) − 1.1 (0.22) − 1.7 (0.15) − 2.4 (0.19) − 2.4 (0.21) − 2.5 (0.22) − 2.6 (0.28) − 2.7 (0.29) − 0.7 (0.19) − 1.4 (0.22) − 1.3 (0.24) − 1.4 (0.25) − 1.4 (0.31) − 1.3 (0.32)
− 1.1 (0.3) − 1.4(0.34) − 1.5 (0.37) − 1.7 (0.37)
− 2.3 − 3.5 − 3.8 − 3.9
(0.33) (0.4) † (0.44)† (0.47)†
− 2.0 − 2.3 − 2.5 − 2.2
(0.41) † (0.46) (0.51)* (0.53)
*P b 0.05; †P b 0.01; ‡ P b 0.001 vs. placebo. PANSS = Positive and Negative Syndrome Scale; SE = standard error.
4. Discussion 4.1. Overview Using this 5-factor PANSS model, clinical efficacy was observed in subjects with schizophrenia who received iloperidone 10–16 mg/d for 4 or 6 weeks in terms of a consistent separation from placebo for all 5 factors. Iloperidone 20–24 mg/d demonstrated separation from placebo on 3 of the 5 factors (positive, cognition and excitement/ hostility), but not on the depression/anxiety and negative factors. There was no clear evidence of a dose response for iloperidone above 4–8 mg/day across all the factors. The greatest reductions from baseline in 5-factor PANSS analysis scores were observed in the cognition and positive factors at both Weeks 4 and 6. Apart from some small numerical changes, the overall pattern of results was unchanged
when factor scores were computed for only those patients who completed greater than 2 weeks of treatment. Similar to iloperidone 10–16 mg/d, risperidone 4–8 mg/d also separated from placebo across all 5 factors, consistent with the results of a pooled analysis of two North American trials where risperidone was compared with placebo and haloperidol 20 mg/d (Marder et al., 1997). The model developed for risperidone was similar to the model we present here, with only minor differences. These differences include the cognitive factor being referred to as disorganized thought in the Marder model, and somatic concern (G1),stereotyped thinking (N7) and lack of judgment and insight (G12) being assigned to the positive factor in the Marder model rather than to the depression/ anxiety factor (G1) or the cognitive factor (N7, G12) in our model. The Marder model has been used by others for different study datasets, such as in a study of asenapine versus placebo with haloperidol as the
Table 7 Least-squared mean changes from baseline in the 5-factor PANSS model at Week 4 and Week 6. for all patients with schizophrenia who received treatment for more than 2 weeks. Factor scores for patients with N 2 weeks of treatment, (SE) Cognition Excitement/hostility Depression/anxiety Positive Negative
Week Week Week Week Week Week Week Week Week Week
Iloperidone
4 6 4 6 4 6 4 6 4 6
4–8 mg/d
10–16 mg/d
20–24 mg/d
− 2.6 − 2.9 − 1.0 − 0.8 − 2.3 − 2.4 − 3.5 − 3.9 − 2.0 − 2.1
− 3.7 (0.36)† − 4.1 (0.38)* − 1.4 (0.21)‡ − 1.3 (0.22)‡ − 2.8 (0.22)† − 2.7 (0.22)‡ − 4.4 (0.27)† − 4.9 (0.29)‡ − 3.2 (0.32)* − 2.9 (0.33)
− 3.8 − 4.8 − 0.9 − 1.4 − 2.1 − 2.5 − 4.2 − 5.0 − 2.7 − 3.3
(0.45) (0.47) (0.26)‡ (0.27)‡ (0.27) (0.28)* (0.35) (0.36) (0.41) (0.41)
*P b 0.05; †P b 0.01; ‡ P b 0.001 vs. placebo. PANSS = Positive and Negative Syndrome Scale; SE = standard error.
(0.42)† (0.76)* (0.25)‡ (0.43)‡ (0.26) (0.44) (0.32)* (0.58)* (0.38) (0.65)
Risperidone 4–8 mg/d
Haloperidol 15 mg/d
Ziprasidone 160 mg/d
Placebo
− 4.7 (0.45)‡ − 5.4 (0.50)‡ − 1.6 (0.26)‡ − 1.7 (0.28)‡ − 2.9 (0.27)† − 3.1 (0.29)‡ − 5.7 (0.34)‡ − 6.4 (0.38)‡ − 3.4 (0.40)* − 3.5 (0.43)*
− 5.4 (0.90)† − 5.6 (0.99)* − 2.2 (0.53)‡ − 2.1 (0.57)‡ − 2.9 (0.55) − 2.9 (0.58)* − 5.2 (0.69)* − 5.3 (0.75)* − 3.6 (0.81) − 3.6 (0.85)
− 4.1 (0.67)*
− 2.5 (0.32) − 2.9 (0.42) 0.2 (0.18) 0.4 (0.24) − 1.8 (0.19) − 1.6 (0.25) − 3.3 (0.24) − 3.5 (0.32) − 2.3 (0.28) − 2.2 (0.36)
− 0.7 (0.39)* − 2.0 (0.41) − 4.8 (0.51)† − 3.0 (0.60)
80
L. Citrome et al. / Schizophrenia Research 131 (2011) 75–81
Table 8 Percent who have improved by at least 20% in the 5-factor PANSS model at Week 1 and Week 2 for all patients with schizophrenia. Estimated percentage using Kaplan-Meier (SE) Cognition Excitement/hostility Depression/anxiety Positive Negative
Iloperidone
Week1 Week2 Week1 Week2 Week1 Week2 Week1 Week2 Week1 Week2
4–8 mg/d
10–16 mg/d
20–24 mg/d
9.5 (1.80) 18.4 (2.50) 20.9 (2.50) 34.7 (3.06) 24.9 (2.65) 37.2 (3.08) 18.8 (2.40) 35.1 (3.07) 15.9 (2.25) 29.1 (2.92)
10.5 26.8 22.1 41.3 20.9 37.5 17.6 40.9 14.0 31.6
8.2 (1.41) 24.9 (2.34) 21.5 (2.11) 40.3 (2.63) 17.0 (1.94) 34.3 (2.57) 18.5 (2.00) 39.5 (2.63) 14.8 (1.83) 31.7 (2.49)
(1.61) (2.45) (2.17) (2.71) (2.13) (2.66) (2.00) (2.72) (1.82) (2.55)
Risperidone 4–8 mg/d
Haloperidol 15 mg/d
Ziprasidone 160 mg/d
Placebo
17.8 32.7 29.5 45.9 26.3 40.9 25.8 56.0 15.2 30.3
15.4 29.2 23.1 49.0 27.2 53.6 32.3 53.5 13.8 37.6
14.7 28.0 28.2 42.7 23.2 44.9 21.2 49.0 23.4 46.6
11.5 27.5 21.9 35.9 22.8 42.0 21.4 40.0 14.7 32.6
(2.55) (3.20) (3.05) (3.43) (2.94) (3.35) (2.92) (3.41) (2.40) (3.16)
(4.47) (6.08) (5.22) (6.64) (5.47) (6.65) (5.80) (6.63) (4.28) (6.62)
(3.04) (3.99) (3.83) (4.39) (3.59) (4.39) (3.49) (4.56) (3.61) (4.51)
(1.55) (2.30) (2.00) (2.43) (2.03) (2.53) (1.99) (2.51) (1.72) (2.40)
PANSS = Positive and Negative Syndrome Scale; SE = standard error.
active control (Kane et al., 2010). We undertook a sensitivity analysis and reanalyzed our data using the Marder factors — the pattern of results was essentially unchanged. Haloperidol at a dose of 15 mg/d failed to separate statistically from placebo on the negative symptom factor; in the aforementioned study of risperidone, of all the PANSS factors the one that was associated with the smallest effect size versus placebo was for haloperidol and negative symptoms. In a study of patients with suboptimal response to antipsychotics, the negative factor demonstrated significant worsening for haloperidol (Lindenmayer et al., 2004). Ziprasidone 160 mg/d failed to statistically separate from placebo on the depression/anxiety and negative factors. The latter outcome is not consistent with another study that examined negative symptoms in patients with schizophrenia treated with ziprasidone or amisulpride, albeit for a longer period of time — 12 weeks (Olie et al., 2006). An important caveat is that the iloperidone registration trials for acute schizophrenia were powered to detect differences between iloperidone and placebo. The active controls, haloperidol, risperidone and ziprasidone, were used as active controls for assay sensitivity. Any dose response or drug-drug comparisons will require the conduct of appropriately designed and powered randomized controlled trials. 4.2. Safety and tolerability The PANSS factor analysis presented here is focused exclusively on antipsychotic efficacy. However real-world effectiveness is dependent on a medication's tolerability profile and a patient's willingness to actually adhere. Further information regarding iloperidone's tolerability profile can be found elsewhere (Citrome 2009; Citrome 2010; Novartis Pharmaceuticals Corporation 2011). 4.3. Limitations Although the overall numbers of subjects available for this analysis was substantial, allowing for the construction of a robust 5 factor PANSS model, the numbers available for the efficacy analyses were reduced as there were multiple treatment arms, including 3 dose ranges for iloperidone. The pooling of 4 short-term efficacy studies also assumed that the study populations were comparable, however this may not be entirely the case as the three 6-week studies that included schizoaffective disordered patients (studies ILP3000ST, ILP3004ST, and ILP3005ST; Table 1) were done in a different epoch of drug development than the 4-week study (study VP-VYY-683-3101). Furthermore, at five and six weeks post-baseline, the 20–24 mg dose group had almost thirty percent as many evaluable patients as the other dose groups, which reduced statistical power to separate from placebo. The analysis of schizoaffective patients enrolled in studies ILP3000ST, ILP3004ST, ILP3005ST, although smaller in number than the subjects diagnosed with schizophrenia, may yield different results
on the 5 factors, particularly the depression/anxiety factor. However, diagnostic uncertainty regarding the differentiation of schizophrenia from schizoaffective disorder may make this type of analysis difficult to interpret (Kantrowitz & Citrome, 2011). 5. Conclusions In a pooled sample of patients with acute schizophrenia, iloperidone demonstrated positive treatment effects across the multiple dimensions of psychopathology commonly associated with schizophrenia: positive, negative, cognitive, excitement/hostility and depression/anxiety. The dose range of 4–8 mg/d was sub-therapeutic. The two higher doses had similar efficacy on the PANSS factors, with the exception of the depression/anxiety and negative factors, on which the 10–16 mg dose group showed statistical separation from placebo and the higher dose group did not. At 5 and 6 weeks, the lack of separation from placebo for the higher dose group may have been due to the much smaller sample size in that group. Further studies examining which symptom dimensions iloperidone is best suited for will require specifically designed studies that include patients from the specific populations in question, for example patients with schizophrenia with predominant negative, cognitive, or depressive symptoms. Role of funding source Funding for this study was provided by Novartis Pharmaceuticals Corporation. Employees of Novartis Pharmaceuticals Corporation were involved in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.
Contributors Leslie Citrome, Xiangyi Meng, and Marla Hochfeld participated in the design and execution of the study, as well as in the preparation of the poster presentation that preceded the paper. Xiangyi Meng undertook the statistical analyses. Leslie Citrome wrote the first draft of the manuscript. All authors contributed to and have approved the final manuscript.
Conflicts of interest 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, BristolMyers Squibb, Eli Lilly and Company, Forest Research Institute, GlaxoSmithKline, Janssen Pharmaceuticals, Jazz Pharmaceuticals, Merck, Novartis Pharmaceuticals Corporation, Noven Pharmaceuticals, Pfizer Inc, Sunovion, Valeant Pharmaceuticals and Vanda Pharmaceuticals. Xiangyi Meng, and Marla Hochfeld are full-time employees and stock holders of Novartis Pharmaceuticals Corporation.
Acknowledgments This work was funded by Novartis Pharmaceuticals Corporation. Presented as a poster at the 23rd United States Psychiatric and Mental Health Congress, Orlando, Florida, November 18–21, 2010.
L. Citrome et al. / Schizophrenia Research 131 (2011) 75–81
References Citrome, L., 2009. Iloperidone for schizophrenia: A review of the efficacy and safety profile for this newly-commercialised second-generation antipsychotic. Int. J. Clin. Pract. 63 (8), 1237–1248. Citrome, L., 2010. Iloperidone redux: a dissection of the Drug Approval Package for this newly commercialised second-generation antipsychotic. Int. J. Clin. Pract. 64 (6), 707–718. Cutler, A.J., Kalali, A.H., Weiden, P.J., Hamilton, J., Wolfgang, C.D., 2008. Four-week, doubleblind, placebo- and ziprasidone-controlled trial of iloperidone in patients with acute exacerbations of schizophrenia. J. Clin. Psychopharmacol. 28 (2 Suppl 1), S20–S28. Kane, J.M., Lauriello, J., Laska, E., DiMarino, M., Wolfgang, C.D., 2008. Long-term efficacy and safety of iloperidone: results from 3 clinical trials for the treatment of schizophrenia. J. Clin. Psychopharmacol. 28 (2 Suppl 1), S29–S35. Kane, J.M., Cohen, M., Zhao, J., Alphs, L., Panagides, J., 2010. Efficacy and safety of asenapine in a placebo- and haloperidol-controlled trial in patients with acute exacerbation of schizophrenia. J. Clin. Psychopharmacol. 30 (2), 106–115. Kantrowitz, J.T., Citrome, L., 2011. Schizoaffective disorder, a review of current research themes and pharmacologic management. CNS Drugs 25 (4), 317–331. Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr. Bull. 13 (2), 261–276. Lançon, C., Aghababian, V., Llorca, P.M., Auquier, P., 1998. Factorial structure of the Positive and Negative Syndrome Scale (PANSS): a forced five-dimensional factor analysis. Acta Psychiatr. Scand. 98 (5), 369–376. Lehoux, C., Gobeil, M.H., Lefebvre, A.A., Maziade, M., Roy, M.A., 2009. The Five-Factor Structure of the PANSS: A Critical Review of its Consistency Across Studies. Clin. Schizophr. Relat. Psychoses 3 (2), 103–110.
81
Levine, S.Z., Rabinowitz, J., 2007. Revisiting the 5 dimensions of the Positive and Negative Syndrome Scale. J. Clin. Psychopharmacol. 27 (5), 431–436. Lindenmayer, J.P., Bernstein-Hyman, R., Grochowski, S., 1994. A newfive factor model of schizophrenia. Psychiatr. Q. 65 (4), 299–322. Lindenmayer, J.P., Czobor, P., Volavka, J., et al., 2004. Effects of Atypical Antipsychotics on the Syndromal Profile in Treatment-Resistant Schizophrenia. J. Clin. Psychiatry 65 (4), 551–556. Marder, S.R., Davis, J.M., Chouinard, G., 1997. The effects of risperidone on the five dimensions of schizophrenia derived by factor analysis: combined results of the North American trials. J. Clin. Psychiatry 58 (12), 538–546. Novartis Pharmaceuticals Corporation, 2011. Fanapt (iloperidone) Tablets. Revised 4/2011 (accessed 03 June 2011) http://www.pharma.us.novartis.com/product/pi/pdf/ fanapt.pdf. Olie, J.P., Spina, E., Murray, S., Yang, R., 2006. Ziprasidone and amisulpride effectively treat negative symptoms: results of a 12-week, double-blind study. Int. Clin. Psychopharamacol. 21 (3), 143–151. Potkin, S.G., Litman, R.E., Torres, R., Wolfgang, C.D., 2008. Efficacy of iloperidone in the treatment of schizophrenia: initial phase 3 studies. J. Clin. Psychopharmacol. 28 (2 Suppl 1), S4–S11. Van den Oord, E.J., Rujescu, D., Robles, J.R., et al., 2006. Factor structure and external validity of the PANSS revisited. Schizophr. Res. 82 (2–3), 213–223. White, L., Harvey, P.D., Opler, L., Lindenmayer, J.P., 1997. Empirical assessment of the factorial structure of clinical in schizophrenia.A multisite, multimodel evaluation of the factorial structure of the Positive and Negative Syndrome Scale. The PANSS Study Group. Psychopathology 30 (5), 263–274.