Effectiveness and safety of olanzapine in the treatment of schizophrenia among Asian patients switching from conventional antipsychotics

Progress in Neuro-Psychopharmacology & Biological Psychiatry 31 (2007) 32 – 40 www.elsevier.com/locate/pnpbp

Effectiveness and safety of olanzapine in the treatment of schizophrenia among Asian patients switching from conventional antipsychotics Zheng Lu a , Jian Hu b , Chih-Ken Chen c , Tsuo-Hung Lan d , Gregorio L. Diokno Jr. e , Byoung Yong Lee f , Heather McElroy g , Gavan Harrison g , Qiuqing Ang h,⁎ a Psychiatry Department, Tongji University; 389 Xin Cun Road, Shanghai 200065, PR China Psychiatry Department, Institute of Mental Health, Harbin Medical University; No. 23, Youzheng Street, Nangang District, Harbin City, 150001, PR China Department of Psychiatry, Chang Gung Memorial Hospital; Chang Gung University School of Medicine; No.222, Mai-Chin Rd, Keelung 204, Taiwan, R.O.C. d Psychiatric Department, Yu-Li Hospital; No.448, Jhonghua Rd, Yuli Township, Hualien County 981, Taiwan, R.O.C. e St. Lukes Medical Center; Rm 814, South Tower, Cathedral Heights Building, E. Rodriguez Avenue, Quezon City, Philippines f Department of Psychiatry, Chuncheon National Hospital; 713 Wonchang5-Fi, Dongsan-Myun, Chuncheon City, Kangwon-Do, South Korea g Intercontinental Information Sciences (ICIS), Eli Lilly Australia; Level 1, 16 Giffnock Ave, Macquarie Park, NSW, 2113, Australia h Eli Lilly Asia, Inc. (Shanghai); 21F, 1 Corporate Avenue, 222 Hu Bin Rd, Shanghai, 200021, PR China

b c

Received 1 February 2006; received in revised form 24 May 2006; accepted 9 June 2006 Available online 14 July 2006

Abstract Objectives: The objectives of this study were: (1) to investigate, in a clinical practice setting, the effectiveness of olanzapine in the treatment of schizophrenia among partially-responding, symptomatic Asian patients who switch from conventional antipsychotic treatment, (2) to assess the safety of olanzapine and (3) to assess the change in quality of life in Asian patients with schizophrenia who switch to olanzapine. Methods: Effectiveness, safety and quality of life were assessed in outpatients with schizophrenia (n = 1267) who lacked symptomatic control with conventional antipsychotics and were switched to olanzapine therapy. Data for this prospective, observational study were collected for 12 months from Asian patients in China, Hong Kong, the Philippines, South Korea and Taiwan. Results: Significant clinical improvements (P < 0.05) were observed following 12 months of olanzapine treatment and 87.3% of the subjects responded to treatment at endpoint (i.e. Brief Psychiatric Rating Scale Total score reduced by ≥ 30% relative to baseline; last observation carried forward). Abnormal involuntary movements (mean change in Abnormal Involuntary Movement Scale: −3.20, P < 0.001) and quality of life were significantly improved in patients treated with olanzapine. However, some patients experienced significant weight gain (3.60 ± 4.50 kg, P < 0.001) with olanzapine treatment, relative to baseline. Conclusions: This study shows that switching to olanzapine may be effective in improving symptoms, may be well-tolerated and may improve the quality of life in Asian patients who are only partially responsive to treatment with conventional antipsychotics. The pragmatic design and naturalistic setting of this large study make the findings relevant for treating patients from some Asian countries in routine clinical practice. © 2006 Elsevier Inc. All rights reserved. Keywords: Asia; Olanzapine; Quality of life; Safety; Schizophrenia; Treatment effectiveness

1. Introduction

Abbreviations: AIMS, Abnormal Involuntary Movement Scale; BPRS, Brief Psychiatric Rating Scale; CGI-S, Clinical Global Impressions-Severity; LOCF, Last Observation Carried Forward; WHO-QoL-BREF, World Health Organization Quality of Life Scale. ⁎ Corresponding author. Tel.: +86 21 2302 1333; fax: +86 21 2302 1487. E-mail address: [email protected] (Q. Ang). 0278-5846/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.pnpbp.2006.06.006

Atypical antipsychotics are increasingly being recommended as first-line therapy for the treatment of schizophrenia (The Expert Consensus Guideline Series: Treatment of Schizophrenia, 1999; National Institute of Clinical Excellence, 2002; Mauskopf et al., 2002). However, atypical antipsychotics are underused throughout much of Asia. A recent survey of six Asian countries showed that atypical agents comprised only

Z. Lu et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 31 (2007) 32–40

28% of antipsychotic prescriptions (Chong et al., 2004). Predominant prescribing of conventional antipsychotics is likely to reflect the relatively low cost of conventional antipsychotics, their widespread availability and traditional prescribing practices (Chong et al., 2004). This prescribing pattern in Asia may change, however, as the clinical benefits and limitations associated with atypical and conventional antipsychotics become more clearly established. Increased prescribing of atypical antipsychotics, even with the relatively higher acquisition costs, may be warranted due to the superior effectiveness of atypical antipsychotics, particularly against negative and neurocognitive symptoms (Foster and Goa, 1998, 1999), and the reduced burden of extrapyramidal side effects associated with atypical antipsychotic use (Foster and Goa, 1998, 1999). Olanzapine, an atypical antipsychotic, is now widely available in Asia. Olanzapine treatment has been reported to be effective and well-tolerated, particularly in patients who have switched from conventional antipsychotics (Lee et al., 2002; Voruganti et al., 2002; Ritchie et al., 2003). Ethnic and cultural factors can influence antipsychotic treatment and treatment outcomes (Binder and Levy, 1981; Jann et al., 1992; Pi, 1998; Zhang-Wong et al., 1998; Hopper and Wanderling, 2000; Emsley et al., 2002; Bradford, 2005). Although the efficacy and safety of olanzapine treatment have been studied in clinical trials in Asia (Ishigooka et al., 2001a,b; Chiu et al., 2003), studies in routine clinical practice are lacking. In terms of clinical relevance and safety, new drugs should be studied in a routine clinical practice setting (FDA, 2002). In addition to this regulatory recommendation, there is a recognized gap between the findings of randomized controlled trials in mental health research and the effectiveness of treatments in real-world clinical practice (Wells, 1999; Essock et al., 2003). This disparity, which may limit the applicability of clinical trial results in clinical practice, reinforces the need to study the effectiveness and safety of antipsychotic treatments in clinical practice settings. The primary objective of this study was to investigate, in a clinical practice setting, the effectiveness of olanzapine in the treatment of schizophrenia among partially-responding, symptomatic Asian patients who switch from conventional antipsychotic treatment. The secondary objectives of the study were to assess the safety of olanzapine and to assess the change in the quality of life in Asian patients with schizophrenia who switch to olanzapine. 2. Material and methods 2.1. Patients Patients with schizophrenia who lacked symptomatic control with conventional antipsychotics were selected for this study. Patients were eligible to enter the study if they: (1) were diagnosed with schizophrenia (Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition [DSM-IV] [APA, 1994] or ICD-10 [WHO, 1992]), (2) were previously prescribed a conventional antipsychotic and were assessed by their psychiatrist as being likely to benefit from a switch to olanzapine treatment

33

based on their clinical judgment, (3) were Asian, (4) presented during the normal course of care, (5) were aged between 18 and 60 years of age, (6) had not used olanzapine or any other atypical antipsychotic in the 6 months before enrolment and (7) were not pregnant. The data for this report were obtained from 64 investigator sites throughout China (n = 13), Hong Kong (n = 2), the Philippines (n = 20), South Korea (n = 15) and Taiwan (n = 14). The study was conducted in accordance with the local ethics and regulatory requirements of each region. All participants provided informed consent as required by local regulations. 2.2. Study design This was a prospective, one-year, non-randomized, multicentre, observational study (Study code: F1D-AA-HGKM). The study was initiated in May 2002 and completed in September 2004. To ensure that the study reflected real-life clinical practice, patient care was at the discretion of the participating psychiatrist. Subjects were prescribed commercially-available olanzapine (at their own expense) within the recommended dose range of 5 to 20 mg/day or as stated in the package insert. The dose prescribed was at the discretion of the psychiatrist. Other concomitant medications were allowed, including other atypical or conventional antipsychotics. No treatment instructions were given other than for each participating psychiatrist to manage subjects as per their usual practice. Subjects were discontinued from the study if they stopped taking olanzapine. Follow-up of the subjects was not limited by the study's schedule of assessments. 2.3. Outcome measures Effectiveness, safety and quality of life assessments were made at baseline (Visit 1) and at approximately 4 ± 1 weeks (Visit 2), 8 ± 1 weeks (Visit 3), 26 ± 2 weeks (Visit 4) and 52 ± 2 weeks (Visit 5). The effectiveness of olanzapine treatment was assessed using the Brief Psychiatric Rating Scale (BPRS [Overall and Gorham, 1988]; Total score and Positive and Negative subscales) and the Clinical Global Impressions-Severity (CGI-S) rating scale (National Institute of Mental Health, 1976a). The Positive subscale of the BPRS comprised BPRS item numbers 4, 11, 12 and 15 (scale 0 to 24, higher score indicating worse result). The Negative subscale of the BPRS comprised BPRS item numbers 3, 13 and 16 (scale 0 to 18, higher score indicating worse result). Subjects were classified as having responded if their BPRS Total score dropped by 30% or more, relative to baseline. Safety was assessed using the 10 item Abnormal Involuntary Movements Scale (AIMS [National Institute of Mental Health, 1976b]) and quality of life was assessed using the brief version of the World Health Organization Quality of Life Scale (WHO-QoL-BREF; Physical Health, Psychological Health, Social Relations and Environment subscales [WHO, 1996]). Subject demographics were recorded at baseline. Weight, blood pressure and pulse were recorded at baseline and at the

34

Z. Lu et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 31 (2007) 32–40

final visit or upon discontinuation. Olanzapine dose and concomitant medication use were recorded at every visit. 2.4. Sample size and statistical analysis The minimum sample size for this study was calculated as follows. Based on previous research (Revicki et al., 1999), the expected mean percentage decrease in the BPRS Total score after 12 months of olanzapine treatment is 40%. However, given that the subjects in this study were partially-responding patients who had been switched from their previous conventional antipsychotic to olanzapine, the minimum decrease in the BPRS Total score was expected to be 30%. According to the standard sample size formula for a single sample, a minimum of 896 subjects were required to detect a 30% decrease in the BPRS Total score, with an absolute precision level of 3% at the 5% significance level. Assuming an expected dropout rate of 20%, the minimum sample size was set at 1100 subjects. Statistical analyses were performed using SAS© Version 8.2 for Windows™ (SAS Institute, Cary, N.C.). Continuous variables are described using summary statistics such as means and standard deviations or 95% confidence intervals (C.I.). Categorical variables are described using frequencies and percentages. All endpoint analyses used a last observation carried forward (LOCF) algorithm where the last available values served as endpoints. For the analyses of change from baseline to endpoint, only subjects with a baseline and at least one postbaseline measure were included. Significance of the absolute change for a given variable was evaluated using the paired t-test. Post hoc analysis was conducted to examine differences in change between countries using ANCOVA with baseline score and country as covariates. For this report, P < 0.05 was considered significant. However, results should be interpreted with caution because of the large number of tests undertaken. To supplement the LOCF analyses of effectiveness at the end of the study, mixed effects models were fitted to analyse the continuous outcomes that were repeatedly measured throughout the study, using the MIXED procedure in SAS© Version 8.2. Estimates of mean change from baseline to each visit are reported here using mixed effects models with visit number and investigator as categorical fixed effects and an exchangeable correlation structure (i.e. compound symmetry) for withinsubject error correlations. The exchangeable correlation assumption was chosen as it was the simplest structure that adequately modeled the data. In the analysis of WHO-QoL BREF Psychological domain, an unstructured approach was used as the model using compound symmetry failed to converge. The association between response and weight gain (7% or greater of baseline body weight) was assessed using logistic regression adjusted by country and is reported as odds ratios.

eligibility criteria and were excluded from the analyses. Of the remaining 1267 subjects (China n = 475, Hong Kong n = 19, Korea n = 173, the Philippines n = 229 and Taiwan n = 371), 56.2% were male. The mean age (±S.D.) at baseline was 33.0 ± 10.5 years. The majority of the subjects indicated they were of Chinese origin (68.2%), 13.7% of subjects indicated South Korean origin and 18.2% of subjects indicated ‘Other’ origins. Based on the mean CGI-S (4.71 ± 1.01), the subjects were moderately ill at baseline. Most subjects (87.1%) had been using conventional antipsychotics as monotherapy when they entered the study. Before the study period, the three most commonly used conventional antipsychotics were: haloperidol (31.3%), chlorpromazine (29.3%) and perphenazine (10.5%). The mean body weight (± S.D.) was 61.9 ± 11.35 kg. Table 1 provides the patient characteristics at study initiation for each country/region. Of the 1267 subjects, 954 subjects (75.3%) completed the study. The reasons for discontinuation included being lost to follow-up (180 subjects), patient decision (43 subjects), psychiatrist decision (47 subjects), adverse event (7 subjects), participation in another study (2 subjects), death (not suicide; 2 subjects) or other (32 subjects). The rate of study completion varied widely among regions. China (including Hong Kong) had the highest rate of completers (90.9%). This was followed by the Philippines (79.9%), Taiwan (59.6%) and South Korea (58.4%). 3.2. Dosage and treatment patterns The mean dose of olanzapine varied during the study. The mean dose (± S.D.) prescribed at baseline was 8.40 (± 3.93) mg/ day. This increased to 10.96 (± 4.64) mg/day by Visit 2 and then tapered down to 9.00 (± 4.85) mg/day by Visit 5 (data from n = 972 subjects). The mean dose at endpoint was 9.25 (± 4.87) mg/day (data from n = 1203 subjects). For the individual countries/regions, the mean dose at endpoint varied somewhat: China 7.72 (±3.97), the Philippines 7.60 (± 2.79), Korea 11.18 (± 4.64) and Taiwan 11.69 (± 5.87) mg/day.

Table 1 Patient characteristics at study initiation for each country/region China ⁎ Korea Taiwan Philippines Overall

3.1. Subjects

Number of patients 494 173 371 Male, % 50.8 57.2 63.3 Age in years, mean 27.9 37.8 36.7 (S.D.) (9.5) (9.6) (9.6) Prior conventional 93.7 75.7 94.1 antipsychotic as monotherapy (%) Most common prior conventional antipsychotic(s) ⁎⁎ Haloperidol 19.0 74.0 21.3 Chlorpromazine 33.8 26.6 4.3 Perphenazine 24.5 4.6 0.8 No prior concomitant 61.5 22.6 6.1 psychotropic medication, %

A total of 1272 Asian subjects with schizophrenia were enrolled into the study. Five subjects did not meet all of the

* Including patients from China and Hong Kong. ** Patients may be counted more than once as they may take more than one antipsychotic.

3. Results

229 55.5 34.5 (9.6) 69.9

1267 52.6 33.0 (10.5) 87.1

41.9 62.0 0.4 69.9

31.3 29.3 29.3 41.8

Z. Lu et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 31 (2007) 32–40

Fig. 1. Concomitant psychotropic medication use during the study.

Most subjects changed their olanzapine dose during the course of the study. The dose increased above the baseline dose for almost half of the subjects (46.1%). For some subjects (18.2%), the olanzapine dose was decreased below the baseline dose, while for a small proportion of subjects (6.2%), the olanzapine dose both increased and decreased around their baseline dose. The dose was unchanged for 29.5% of subjects.

number of patients (75.9% and 75.6%, respectively) responding at endpoint. Consistent with the high rate of responders, significant improvements were observed in clinical status following olanzapine treatment. This is shown by the reduction in symptom scores at endpoint relative to baseline (Table 2). The mean change from baseline to endpoint represented a 67.5% reduction for the BPRS Total score, a 68.9% reduction for the BPRS Positive score, a 62.2% reduction for the BPRS Negative score and a 45.3% reduction for the CGI-S score. Furthermore, the change in scores from baseline to each visit based on the mixed effects models (Fig. 2), were similar to the change from baseline to the LOCF endpoint scores (Table 2). Changes for the better in clinical symptoms were observed in all countries/regions. However, results from ANCOVA with baseline score and country as a covariate, showed statistically significant differences between countries/regions (P < 0.001). Mean change from baseline to endpoint for BPRS total score was greatest in China (− 32.0; 95% C.I. − 32.9 to − 31.1), Table 2 Change in effectiveness, safety and quality of life scales from baseline to endpoint Outcome Measure Baseline Endpoint a Change endpointbaseline (mean ± (mean ± S.D.) (mean; 95% C.I.) S.D.)

3.3. Concomitant medications

Effectiveness BPRS Total

Concomitant psychotropic medication use decreased during the study (Fig. 1). The proportion of subjects taking any concomitant psychotropic medications was 55.4% at baseline, and decreased to 31.5% based on subjects who completed the study. Country/region variations in concomitant psychotropic medication prescribing were observed, with the highest level at baseline for Taiwan (86.8% decreasing to 80.9% at study completion) and Korea (72.3% to 37.1%), compared to China (40.9% to 14.1%) and the Philippines (23.1% to 7.1%). Much of this change was attributed to the decreased use of benzodiazepines, which decreased in the overall sample from 39.1% at baseline to 22.9% by the end of the study (completers). However, in Taiwan, 74.6% of patients were taking benzodiazepines at the end of the study. Overall, the use of non-neuroleptic psychotropic medications decreased from 48.8% at baseline to 27.4% at the end of the study. The use of other antipsychotics also decreased during the study (baseline: 10.1%; 52 weeks: 4.5%).

BPRS Positive

3.4. Effectiveness Most subjects responded to olanzapine treatment. For the subjects who completed the study (n = 954), 93.3% of subjects (95% C.I.: 91.5% to 94.8%) responded to olanzapine treatment (BPRS Total reduced by ≥ 30% relative to baseline). The proportion of responders decreased slightly to 87.3% (95% C.I.: 85.2% to 89.1%) based on endpoint data (n = 1186). The highest response rate at endpoint was observed in China (95.9%) and the Philippines (93.5%), with Taiwan and Korea having a lower

35

37.24 ± 15.95 11.67 ± 5.28 6.74 ± 3.35 4.71 ± 1.01

12.00 ± 13.25 3.54 ± 4.27 2.35 ± 2.65 2.48 ± 1.20

4.60 ± 7.07 61.7 ± 11.2

Quality of Life (WHO QoL) Physical health 44.76 ± 15.60 Psychological 39.32 ± 16.37 Social 38.27 ± relationships 17.98 Environment 42.23 ± 15.23

BPRS Negative CGI-Severity

Safety AIMS Weight b

P value

− 25.23 (− 26.18,− 24.29) <0.001 − 8.13 (− 8.45,− 7.81)

<0.001

− 4.39 (− 4.59,− 4.19)

<0.001

− 2.23 (− 2.31,− 2.15)

<0.001

1.40 ± 3.64 65.3 ± 11.3

− 3.20 (− 3.55, − 2.85)

<0.001

3.60 (3.32, 3.88)

<0.001

63.72 ± 14.06 61.19 ± 17.66 57.97 ± 19.11 61.18 ± 17.71

18.96 (17.82, 20.11)

<0.001

21.87 (20.60, 23.13)

<0.001

19.70 (18.36, 21.04)

<0.001

18.95 (17.77, 20.13)

<0.001

95% C.I. = 95% confidence interval; AIMS = Abnormal Involuntary Movements Scale; BPRS = Brief Psychiatric Ratings Scale; CGI = Clinical Global Impression; WHO QoL = World Health Organisation Quality of Life. a Endpoint represents the last observation carried forward. Data were available for the 52-week visit from the following proportion of subjects: BPRS Total data — 82% (967/1186), BPRS Positive data — 81% (968/1195), BPRS Negative data — 80% (970/1206), CGI-S data — 81% (975/1209), AIMS — 81% (972/1204), all WHO QoL data — 82% (961/1172). b Unlike other measures in this table, weight was only measured at baseline and final visit (52-week visit or at early discontinuation when possible). Therefore, only 989 subjects are included in this analysis, 95% of whom provided weight data at the 52-week visit.

36

Z. Lu et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 31 (2007) 32–40

Fig. 2. Change in BPRS Total score (A), Positive score (B), Negative score (C) and CGI-S (D) score over time, relative to baseline. Errors bars represent 95% confidence intervals. All changes were significantly different from baseline (P < 0.001). Results are estimates from mixed effects models using visit number and investigator as fixed effects.

followed by the Philippines (− 29.3; 95% C.I.: − 30.6 to − 27.9), Korea (− 17.3; 95% C.I. −18.8 to − 15.7) and Taiwan (− 16.3; 95% C.I. − 17.4 to − 15.2). Mean change from baseline to endpoint for CGI-S showed a similar pattern: China (−3.00; 95% C.I. −3.09 to−2.92), the Philippines (−2.28; 95% C.I. −2.41 to −2.15), Korea (−1.66; 95% C.I. −1.81 to −1.52) and Taiwan (−1.33; 95% C.I. −1.43 to −1.23). Improvements in clinical symptoms were observed shortly after the initiation of olanzapine therapy. Improvements in BPRS Total score, BPRS Positive score, BPRS Negative score and CGI-S score were most rapid in the first 8 weeks of treatment (Fig. 2). After this time, clinical symptoms continued to improve but at a slower rate. Some assessments of effectiveness were made outside of the scheduled visit window and the visit number itself, as recorded on the case report forms, was used instead of the actual time of measurement. However, analyses that excluded visits that occurred outside the scheduled visit window gave similar results (data not shown).

baseline (AIMS > 0; n = 597), overall, 50.4% no longer suffered from these symptoms (AIMS = 0) at endpoint (China 62.1%, Korea 37.7%, Taiwan 41.6% and the Philippines 54.6%). Some patients experienced significant weight gain during olanzapine treatment. The mean (± S.D.) weight gained was 3.60 ± 4.50 kg by the final visit (n = 989), relative to baseline (P < 0.001). Over a third of the subjects (39.1%), gained 7% or more of their baseline body weight (China 42.9%, Korea 32.3%, Taiwan 32.8% and the Philippines 42.4%). Interestingly, based

3.5. Safety Significant improvements were observed in terms of safety at endpoint, relative to baseline (Table 2). The mean change in the AIMS score was − 3.20 (95% C.I.: − 3.55, − 2.85). Of the subjects that did not have any abnormal involuntary movements at baseline (AIMS = 0; n = 607), 94.9% of the subjects continued to remain free of these adverse symptoms at endpoint. Of the subjects who had some abnormal involuntary movements at

Fig. 3. Change in WHO Quality of Life domains over time. Errors bars represent 95% confidence intervals. All changes were significantly different from baseline (P < 0.001). Results are estimates from mixed effects models using visit number and investigator as fixed effects.

Z. Lu et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 31 (2007) 32–40

on endpoint data, the odds of responding (reduction in BPRS Total score ≥ 30% relative to baseline) for subjects who gained 7% or more of their baseline body weight were 2.50 times higher than those who gained less than 7% (odds ratio comparison; 95% C.I.: 1.41, 4.44; P = 0.002). 3.6. Quality of life Significant improvements in the quality of life were experienced for the four domains that were measured (Table 2). The mean change from baseline to endpoint represented a 75.8% increase in the WHO QoL Physical Health score, a 103.2% increase in the WHO QoL Psychological Health score, an 83.0% increase in the WHO QoL Social Relationships score and a 71.3% increase in the WHO QoL Environment score. Similar to the improvements in clinical symptoms, the improvements in the quality of life were most rapidly gained in the first 8 weeks of olanzapine treatment (Fig. 3). The quality of life continued to improve to the end of the study but at a slower rate. 4. Discussion Ethnic differences in the response to antipsychotics are an important consideration in the treatment of schizophrenia (Lin et al., 1995; Frackiewicz et al., 1997; Pi, 1998). Asian patients have been reported to have a lower antipsychotic dose requirement and may be more susceptible to extrapyramidal symptoms (Binder and Levy, 1981; Ruiz et al., 1996; Frackiewicz et al., 1997; Bradford, 2005). To the best of our knowledge, this paper provides new data on patients with schizophrenia, from a large and diverse Asian population who were studied in a clinical practice setting as they switched from conventional antipsychotics to olanzapine. Our study indicates that switching to olanzapine is a viable treatment alternative for Asian patients with schizophrenia, who are only partially responsive to conventional antipsychotics. For the majority of patients in our study, significant improvements were observed in clinical symptoms during olanzapine treatment. Improvements in the severity and incidence of abnormal involuntary movements and in the quality of life were statistically significant following a switch to olanzapine treatment. 4.1. Effectiveness The results of this study support and extend the findings of previous studies conducted in Asia. Our results align with shortterm, open-label, clinical trials in which the majority of subjects were switched from conventional antipsychotics to olanzapine (Ishigooka et al., 2001a; Lee et al., 2002). Ishigooka et al., 2001a showed that in Japanese patients with schizophrenia, olanzapine treatment can significantly improve the BPRS Total score after 8 weeks of treatment. Similarly, Lee et al., 2002 showed that in Asian patients, switching to olanzapine treatment can significantly improve clinical symptoms based on the CGI-S and Positive and Negative Syndromes Scales. Differences were apparent, however, between the level of clinical improvement achieved in the early phase of our study and that

37

achieved in these short-term studies (Ishigooka et al., 2001a; Lee et al., 2002). The seemingly lower level of effectiveness in the short-term studies, compared to our study, may have been due to differences in the severity of illness of the participants (Ishigooka et al., 2001a; Lee et al., 2002) or the lower dose of olanzapine that was used (Ishigooka et al., 2001a). Despite the recognized importance of observational studies (Wells, 1999; Essock et al., 2003; FDA, 2002), the lack of studies conducted in Asian patients with schizophrenia in a clinical practice setting makes it difficult to compare the findings of our study with analogous data from other work. However, a smaller, long-term, observational study conducted in Canadian patients, who lacked control with conventional antipsychotics, showed that, similar to our study, switching to olanzapine was associated with improvements in clinical symptoms based on the Positive and Negative Syndromes Scale (Voruganti et al., 2002). 4.2. Safety In terms of safety, the improvements in extrapyramidal symptoms during olanzapine treatment were generally consistent with previous studies. However, some differences in magnitude were observed. The level of improvement in extrapyramidal symptoms observed in our study (mean change in AIMS: − 3.20) was greater than those previously reported in Canadian (mean change in AIMS: − 0.75) and Latin American (mean change in AIMS: − 1.48) studies of patients who were switched from conventional antipsychotics to olanzapine (Voruganti et al., 2002; Costa e Silva et al., 2003). The greater level of improvement shown in our study may be attributed to the greater severity of abnormal involuntary movements in our study subjects at baseline compared with those studies. One explanation for the higher level of extrapyramidal symptoms at baseline and greater improvement with olanzapine treatment in our study population comes from previous research which suggests that Asian patients may be more susceptible to extrapyramidal symptoms induced by conventional antipsychotics, compared to Caucasian patients (Binder and Levy, 1981; Lam et al., 1995). The increase in weight after switching to olanzapine that was observed in our study was also consistent with findings from previous studies. Reports have indicated that Asian patients treated with olanzapine can gain on average 1.29 to 1.93 kg after 6 weeks of treatment (Lee et al., 2002) and up to 8.34 kg after 103 weeks of treatment (Lee et al., 2004). Subjects in our study gained on average 3.60 kg by their final visit. Weight gain in patients with schizophrenia has been significantly associated with better clinical outcomes, lower baseline body mass index, non-White race and olanzapine treatment (Basson et al., 2001). Our study with olanzapine supports the finding that better clinical outcomes are linked with weight gain. Direct comparator studies are needed to compare the effects of switching to olanzapine on weight gain in different ethnic groups and in patients with varying body mass indices. Strategies to mitigate weight gain should be implemented, as an increase in weight may negatively impact on treatment satisfaction and

38

Z. Lu et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 31 (2007) 32–40

compliance, and can increase the risk of developing a range of obesity-related conditions (Kurzthaler and Fleischhacker, 2001). Interestingly, however, while the recent Clinical Antipsychotic Trials of Intervention Effectiveness reported that the olanzapine group experienced the highest rate for discontinuation due to weight gain or metabolic effects (9%), the overall discontinuation rate before the end of the 18-month study period was lowest in this group (Lieberman et al., 2005).

patients, who were switched predominantly from conventional antipsychotics to olanzapine (Lee et al., 2002), a lower proportion of our study subjects used benzodiazepine or other nonneuroleptic medications. The increased use of these concomitant medications in the study by Lee and colleagues may reflect an increased severity of illness in their study population and the shorter treatment period used (6 weeks). 4.6. Method of switching medications

4.3. Quality of life Unfortunately, we are limited in our ability to compare our quality of life findings with the published literature. The level of information currently available on the potential effects of olanzapine on quality of life among Asian patients with schizophrenia is inadequate. However, consistent with our study, a Canadian study showed that subjects' quality of life was improved when they switched from conventional antipsychotics to olanzapine (Voruganti et al., 2002).

In our study, the method of switching was at the discretion of the treating psychiatrist. In the study by Lee et al., 2002, Asian patients with schizophrenia were equally responsive to a switch to olanzapine treatment regardless of whether they were switched directly or initially treated with both olanzapine and their usual antipsychotic. Both groups in the study experienced significant improvements in clinical and extrapyramidal symptoms after switching to olanzapine treatment, with no significant differences between the two groups (Lee et al., 2002).

4.4. Dosage

4.7. Study strengths and limitations

The dose of olanzapine used in our study was lower than that used in a number of studies conducted in Western countries. The mean dose at endpoint in our study was 9.25 mg/day whereas mean (or mean modal) doses of 12.1 mg/day to 20.9 mg/day have been reportedly used in studies conducted in Europe and other Western countries (Tran et al., 1997; Revicki et al., 1999; Voruganti et al., 2002; Gureje et al., 2003; Haro et al., 2005; Lieberman et al., 2005). In comparison, the majority of studies conducted in Asia have reported mean (or mean modal) doses of 7.4 mg/day to 12.2 mg/day (Ishigooka et al., 2001a,b; Lee et al., 2002). One study on treatment-resistant Chinese subjects did, however, report mean modal doses as high as 20.4 mg/day (Chiu et al., 2003). The high level of clinical response and improvements in the quality of life observed in our study, even with a relatively lower dose of olanzapine, supports previous reports that have suggested that Asian patients have a lower antipsychotic dose requirement (Ruiz et al., 1996; Frackiewicz et al., 1997). Additional, indirect support for this suggestion comes from a study that showed that Singaporean psychiatrists prescribed clozapine at a dose that was more than two-fold lower than that prescribed by Canadian psychiatrists (Chong et al., 2000).

In terms of the study design, our study has both strengths and limitations. The use of a pragmatic design to closely reflect the real-life clinical situation makes the findings of this study highly relevant to clinical practice in many parts of Asia. Furthermore, the study was prospective in design, included a large number of subjects and was of long-term duration. The main limitations of this study are the lack of a control group and the lack of blinding, which may have led to observer bias. We also recognize the lack of identification of precise ethnic groups, as Asian patients do not represent a homogeneous group but may be from countries that are ethnically and culturally diverse. Indeed, between-country/region differences in a number of patient characteristics at study initiation were apparent and while the individual country-level results followed similar trends over time to the overall results, differences in magnitude of some outcomes were observed between countries/regions. Therefore, adjustments for country/region or investigator were made when examining differences between subgroups. The LOCF approach used to impute data poses its own limitations. The first assumption of the LOCF approach is that a person's last observed value before early discontinuation would remain at that level had the person remained in the study. This assumption appears reasonable in this study where the effectiveness of olanzapine was seen early in the study (first 4 to 8 weeks) and then stabilised somewhat. The second assumption of the LOCF approach is that missing data occurs completely at random, which may not be very reasonable in this study. Although the majority of early discontinuation was due to subjects being lost to follow-up, some early discontinuation was related to poor outcomes (adverse event, physician decision). However, the results from the LOCF approach were consistent with the results from the mixed effects model used to analyse the repeated outcome measurements, thus providing confidence in the estimates obtained. Another limitation is that due to the non-interventional design of the study, subjects were able to supplement olanzapine

4.5. Concomitant medication In our study, the effects of olanzapine were complicated, to an extent, by the use of concomitant medications, including other atypical and conventional antipsychotics. However, concomitant medication use was limited and by the end of the study, 68.5% of subjects were not taking any concomitant medications and 95.5% were not taking other antipsychotics. Furthermore, as the concomitant use of psychotropic medications is widespread in clinical practice, their permitted use in our study enhances the clinical relevance of the findings from this study. In comparison to another international study on Asian

Z. Lu et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 31 (2007) 32–40

treatment with other atypical and conventional antipsychotics and/or other concomitant medications without detailed information of this medication use being collected. Therefore, the results, both positive and negative, may not be directly attributable to olanzapine. Other limitations include the inability to comment on patient-reported adverse events, treatment compliance and on the use of anticholinergics. Although treatment compliance was not monitored, subjects who switched or stopped taking olanzapine were discontinued from the study. 5. Conclusions In conclusion, statistically significant improvement in symptoms was observed in Asian patients with schizophrenia who were switched to olanzapine from conventional antipsychotics, with the majority of patients experiencing a clinical response. Switching to olanzapine also reduced the severity and incidence of abnormal involuntary movements and improved the subjects' quality of life. This study set in clinical practice shows that switching to olanzapine can bring clinical benefits to patients who are only partially responsive to treatment with conventional antipsychotics. For this patient group, switching to olanzapine is a viable treatment option. The pragmatic design and naturalistic setting of this large study make the findings relevant for treating patients from some Asian countries in routine clinical practice. Acknowledgements This study was funded by Eli Lilly (Eli Lilly and Company, Indianapolis, USA). The authors acknowledge the independent medical writing assistance provided by ProScribe Medical Communications (www.proscribe.com.au), funded from an unrestricted financial grant from Eli Lilly Australia Pty Ltd (Level 1, 16 Giffnock Avenue, Macquarie Park, NSW, Australia 2113). References American Psychiatric Association. DSM-IV. Diagnostic and statistical manual of mental disorders. fourth ed. Washington: American Psychiatric Association; 1994. p. 273–315. Basson BR, Kinon BJ, Taylor CC, Szymanski KA, Gilmore JA, Tollefson GD. Factors influencing acute weight change in patients with schizophrenia treated with olanzapine, haloperidol or risperidone. J Clin Psychiatry 2001;62:231–8. Binder RL, Levy R. Extrapyramidal reactions in Asians. Am J Psychiatry 1981;138:1243–4. Bradford LD. The ethnopharmacology of atypical antipsychotics. CNS Spectr 2005;10(3 Suppl. 2):S6–S12. Chiu NY, Yang YK, Chen PS, Chang CC, Lee IH, Lee JR. Olanzapine in Chinese treatment-resistant patients with schizophrenia: an open-label, prospective trial. Psychiatry Clin Neurosci 2003;57:478–84. Chong SA, Remington GJ, Lee N, Mahendran R. Contrasting clozapine prescribing patterns in the east and west? Ann Acad Med Singap 2000;29:75–8. Chong MY, Tan CH, Fujii S, Tang SY, Ungvari GS, Si T, et al. Antipsychotic drug prescription for schizophrenia in East Asia: rationale for change. Psychiatry Clin Neurosci 2004;58:61–7. Costa e Silva JA, Alvarez N, Mazzotti G, Gattaz WF, Ospina J, Larach V, et al. Olanzapine as an alternative therapy for patients with haloperidol-induced extrapyramidal symptoms: results of a multicenter, collaborative trial in Latin America. J Clin Psychopharmacol 2003;21:375–81.

39

Emsley RA, Roberts MC, Rataemane S, Pretorius J, Oosthuizen PP, Turner J, et al. Ethnicity and treatment response in schizophrenia: a comparison of three ethnic groups. J Clin Psychiatry 2002;63:9–14. Essock SM, Drake RE, Frank RG, McGuire TG. Randomized clinical trials in evidenced-based mental health care: getting the right answer to the right question. Schizophr Bull 2003;29:115–23. FDA [homepage on the Internet]. Rockville: FDA; c 2002 [updated 2002 April 4; cited 2003 September 20]. Report to Congress: Reports on Postmarketing Studies [FDAMA 130] [1 screen]. Available from:http://www.fda.gov/cber/ fdama/pstmrktfdama130.htm. Foster RH, Goa KL. Risperidone: a pharmacoeconomic review of its use in schizophrenia. PharmacoEconomics 1998;14:97–133. Foster RH, Goa KL. Olanzapine: a pharmacoeconomic review of its use in schizophrenia. PharmacoEconomics 1999;15:611–40. Frackiewicz EI, Sramek JJ, Herrera JM, Kurtz NM, Cutler NR. Ethnicity and antipsychotic response. Ann Pharmacother 1997;31:1360–9. Gureje O, Miles W, Keks N, Grainger D, Lambert T, McGrath J, et al. Olanzapine vs risperidone in the management of schizophrenia: a randomized doubleblind trial in Australia and New Zealand. Schizophr Res 2003;61:303–14. Haro JM, Edgell ET, Novick D, Alonso J, Kennedy L, Jones PB, et al. Effectiveness of antipsychotic treatment for schizophrenia: 6-month results of the Pan–European Schizophrenia Outpatient Health Outcomes (SOHO) study. Acta Psychiatr Scand 2005;111:220–31. Hopper K, Wanderling J. Revisiting the developed versus developing country distinction in course and outcome in schizophrenia: results from ISoS, the WHO collaborative followup project. International Study of Schizophrenia. Schizophr Bull 2000;26:835–46. Ishigooka J, Murasaki M, Miura S, Olanzapine Early-Phase II Study Group. Efficacy and safety of olanzapine, an atypical antipsychotic, in patients with schizophrenia: results of an open-label multicenter study in Japan. Psychiatry Clin Neurosci 2001a;55:353–63. Ishigooka J, Inada T, Miura S. Olanzapine versus haloperidol in the treatment of patients with chronic schizophrenia: results of the Japan multicenter, doubleblind olanzapine trial. Psychiatry Clin Neurosci 2001b;55:403–14. Jann MW, Chang WH, Lam YW, Hwu HG, Lin HN, Chen H, et al. Comparison of haloperidol and reduced haloperidol plasma levels in four different ethnic populations. Prog Neuro-psychopharmacol Biol Psychiatry 1992;16:193–202. Kurzthaler I, Fleischhacker WW. The clinical implications of weight gain in schizophrenia. J Clin Psychiatry 2001;62(Suppl. 7):32–7. Lam YW, Jann MW, Chang WH, Yu HS, Lin SK, Chen H, et al. Intra-and interethnic variability in reduced haloperidol to haloperidol ratios. J Clin Pharmacol 1995;35:128–36. Lee CT, Conde BJ, Mazlan M, Visanuyothin T, Wang A, Wong MM, et al. Switching to olanzapine from previous antipsychotics: a regional collaborative multicenter trial assessing 2 switching techniques in Asia Pacific. J Clin Psychiatry 2002;63:569–76. Lee E, Leung CM, Wong E. Atypical antipsychotics and weight gain in Chinese patients: a comparison of olanzapine and risperidone. J Clin Psychiatry 2004;65:864–6. Lieberman JA, Stroup TS, McEvoy JP, Swartz MS, Rosenheck RA, Perkins DO, et al. Effectiveness of antipsychotic drugs in patients with chronic schiozphrenia. N Engl J Med 2005;353:1209–23. Lin KM, Anderson D, Poland RE. Ethnicity and psychopharmacology: bridging the gap. Psychiatr Clin North Am 1995;18:635–47. Mauskopf J, Muroff M, Gibson PH, Grainger DI. Estimating the costs and benefits of new drug therapies: atypical antipsychotic drugs for schizophrenia. Schizophr Bull 2002;28:619–35. National Institute of Clinical Excellence. Guidance on the use of newer (atypical) antipsychotic drugs for the treatment of schizophrenia. Technology Appraisal Guidance; Report no.43. London: National Institute for Clinical Excellence; 2002. National Institute of Mental Health. Clinical Global Impressions (CGI). In: Guy W, editor. ECDEU assessment manual for psychopharmacology. Bethesda: US Department of Health, Education and Welfare; 1976a. p. 217–22. National Institute of Mental Health. Abnormal Involuntary Movement Scale (AIMS). In: Guy W, editor. ECDEU assessment manual for psychopharmacology. Bethesda: US Department of Health, Education and Welfare; 1976b. p. 534–7.

40

Z. Lu et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 31 (2007) 32–40

Overall JE, Gorham DR. The Brief Psychiatric Rating Scale (BPRS): recent developments in ascertainment and scaling. Psychopharmacol Bull 1988;24:97–9. Pi EH. Transcultural psychopharmacology: present and future. Psychiatry Clin Neurosci 1998;52:S185–7 [Suppl]. Revicki DA, Genduso LA, Hamilton SH, Ganoczy D, Beasley Jr CM. Olanzapine versus haloperidol in the treatment of schizophrenia and other psychotic disorders: quality of life and clinical outcomes of a randomized clinical trial. Qual Life Res 1999;8:417–26. Ritchie CW, Chiu E, Harrigan S, Hall K, Hassett A, Macfarlane S, et al. The impact upon extra-pyramidal side effects, clinical symptoms and quality of life of a switch from conventional to atypical antipsychotics (risperidone or olanzapine) in elderly patients with schizophrenia. Int J Geriatr Psychiatry 2003;18:432–40. Ruiz P, Chu P, Sramek J, Rotavu E, Herrera J. Neuroleptic dosing in Asian and Hispanic outpatients with schizophrenia. Mt Sinai J Med 1996;63:306–9. Tran PV, Hamilton SH, Kuntz AJ, Potvin JH, Andersen SW, Beasley Jr C, et al. Double-blind comparison of olanzapine versus risperidone in the treatment

of schizophrenia and other psychotic disorders. J Clin Psychopharmacol 1997;17:407–18. Treatment of Schizophrenia. The expert consensus guidelines series. J Clin Psychiatry 1999;60(Suppl.11):3–80. Voruganti L, Cortese L, Owyeumi L, Kotteda V, Cernovsky Z, Zirul S, et al. Switching from conventional to novel antipsychotic drugs: results of a prospective naturalistic study. Schizophr Res 2002;57:201–8. Wells KB. Treatment research at the crossroads: the scientific interface of clinical trials and effectiveness research. Am J Psychiatry 1999;156:5–10. World Health Organization. The ICD-10 classification of mental and behavioural disorders: clinical descriptions and diagnostic guidelines. Geneva: World Health Organization; 1992. World Health Organization. WHOQOL-BREF: introduction, administration, scoring and generic version of the assessment, field trial version. Programme on Mental Health. Geneva: World Health Organization; 1996. Zhang-Wong J, Beiser M, Zipursky RB, Bean G. An investigation of ethnic and gender differences in the pharmacodynamics of haloperidol. Psychiatry Res 1998;81:333–9.