Metabolic profiles of second-generation antipsychotics in early psychosis: Findings from the CAFE study

Schizophrenia Research 111 (2009) 9–16

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 e v i e r. c o m / l o c a t e / s c h r e s

Metabolic profiles of second-generation antipsychotics in early psychosis: Findings from the CAFE study Jayendra K. Patel a,⁎, Peter F. Buckley b, Sandra Woolson c, Robert M. Hamer c,d, Joseph P. McEvoy e, Diana O. Perkins d, Jeffrey A. Lieberman f, for the CAFE investigators g a

Department of Psychiatry, University of Massachusetts Medical School, Worcester, MA, United States Department of Psychiatry and Health Behavior, Medical College of Georgia, GA, United States c Department of Biostatistics, University of North Carolina School of Medicine, Chapel Hill, NC, United States d Department of Psychiatry, University of North Carolina School of Medicine, Chapel Hill, NC, United States e Duke University Medical Center, Durham, NC, United States f Department of Psychiatry, College of Physicians and Surgeons, Columbia University, New York, NY, United States g AstraZeneca Pharmaceuticals LP, Wilmington, DE, United States b

a r t i c l e

i n f o

Article history: Received 15 October 2008 Received in revised form 11 March 2009 Accepted 16 March 2009 Available online 26 April 2009 Keywords: Antipsychotic Metabolic

a b s t r a c t Objective: To further define the metabolic profiles of second-generation antipsychotics during the treatment of young patients with early psychosis, with a view to better inform prescribing clinicians. Method: Weight, body mass index (BMI), glucose, and serum lipids were measured in the 52-week Comparison of Atypicals for First Episode (CAFE) study, in which olanzapine, quetiapine, and risperidone were evaluated, and whose primary outcomes have been reported elsewhere. These metabolic data were analyzed using a mixed random coefficients model for continuous longitudinal measures and a logistic regression model for categorical responses. Results: Of the 400 patients recruited, 31% were overweight and 18% were obese at baseline, and 17 (4.3%) patients met criteria for metabolic syndrome. After 12 and 52 weeks of treatment, weight gain ≥7% from baseline was reported in 29.2% and 50.0% of quetiapine-treated patients, 59.8% and 80.0% of olanzapine-treated patients, and 32.5% and 57.6% of risperidone-treated patients, respectively. Weight gain after 12 and 52 weeks of treatment was estimated as [Least Squares Mean (SE)] 15.6 (±1.1) and 24.2 (±1.9) lb for olanzapine, 8.6 (±1.1) and 14.0 (±1.9) lb with risperidone and 7.9 (±1.1) and 12.1 (±1.8) lb for quetiapine respectively. In women, greater weight gain occurred during risperidone treatment compared with quetiapine treatment. By week 52, increases in BMI ≥1 unit occurred with significantly higher frequency in olanzapine-treated patients compared with quetiapine- or risperidone-treated patients. By 52 weeks, treatment-emergent metabolic syndrome was reported in 51 individuals (13.4% of the total population), of whom 22 were receiving olanzapine, 18 quetiapine, and 11 risperidone. Risperidone was associated with the smallest elevations in triglyceride and total cholesterol levels. Conclusion: Weight gain and metabolic syndrome occur commonly even in young patients receiving antipsychotic treatment for early psychosis. Targeted interventions are therefore warranted from the onset of antipsychotic therapy. © 2009 Elsevier B.V. All rights reserved.

1. Introduction ⁎ Corresponding author. Department of Psychiatry, University of Massachusetts Medical School, 361 Plantation Street, Worcester, MA, 01605, United States. Tel.: +508 334 7058; fax: +508 856 4854. E-mail address: [email protected] (J.K. Patel). 0920-9964/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2009.03.025

In the National Institute of Mental Health (NIMH)sponsored Clinical Antipsychotic Trials for Intervention Effectiveness (CATIE) study, major differences were apparent

10

J.K. Patel et al. / Schizophrenia Research 111 (2009) 9–16

in the tolerability profiles of the second-generation antipsychotics (SGAs), with 30%, 16%, and 14% of patients treated with olanzapine, quetiapine, and risperidone, respectively, gaining substantial weight beyond the already elevated mean baseline body mass index (BMI) of 29.7 ± 7 kg/m2 (Lieberman et al., 2005). Additionally, the 41% prevalence of metabolic syndrome (MetS)1 in the overall study population is a sobering reminder of their comorbidities and the side effect burden associated with antipsychotics (Kumra et al., 2008; Lieberman et al., 2005; McEvoy et al., 2005, 2006; Stroup et al., 2006). The risks of weight gain and MetS have now become major considerations when individualizing antipsychotic maintenance treatment. Information available on the weight gain and metabolic changes associated with SGAs during the treatment of patients with early psychosis, though scarce, suggests rapid increases in obesity rates following SGA treatment in this population (Addington et al., 2003, 2006; Crespo-Facorro et al., 2006; Kahn et al., 2008; Perez-Iglesias et al., 2008; Sengupta et al., 2005; Zipursky et al., 2005). In two large, randomized, double-blind studies involving patients with early psychosis, mean weight gain of 10 and 16 lb was reported after 12 weeks and 17 and 34 lb after 2 years of treatment with risperidone and olanzapine, respectively (Schooler et al., 2005; Zipursky et al., 2005). Another openlabel study reported an increase in the proportion of overweight or obese patients from 36% to 60% during 1 year of treatment with various SGAs (Kahn et al., 2008). In the European First-Episode Schizophrenia Trial (EUFEST), a randomized open-label study, the weight gain with olanzapine and quetiapine after 1 year was approximately 31 lb and 23 lb respectively (Kahn et al., 2008). In an open-label study, 12 weeks of treatment with quetiapine resulted in mean weight gain of 9 lb (Tauscher-Wisniewski et al., 2002). Furthermore, concerns regarding changes in serum lipids (Graham et al., 2005; Kahn et al., 2008; Kopala et al., 2006; Sengupta et al., 2005; Wu et al., 2006) have been raised during SGA treatment of early psychosis. Thus, clinical data as well as clinical experience suggest that younger patients and those in the early stages of illness seem most prone to the weight and metabolic side effects. Moreover, data from the National Health and Nutrition Examination Survey (NHANES-III), an ongoing complex, multistage probability sample of the U.S. noninstitutionalized civilian population, suggest that the mean BMI of the adolescent and adult populations has increased every year since the 1990s such that the obesity rates in these groups have dramatically increased too (CDC, 2006; Ogden et al., 2006). In this article, we report weight and metabolic data from the Comparison of Atypicals for First Episode (CAFE) study comparing three commonly used SGAs, olanzapine, quetiapine, and risperidone, in patients with early psychosis.

1 Metabolic syndrome was defined as the presence of three or more of the following American Heart Association (AHA) criteria: 1) abdominal obesity (men N40 in.; women N35 in.); 2) fasting triglycerides N 150 mg/dL; 3) HDL (men b40 mg/dL; women b 50 mg/dL); 4) blood pressure N 130/85 mm Hg or use of antihypertensive medication; and 5) fasting glucose ≥ 100 mg/dL or receiving insulin or hypoglycemic medication.

2. Methods 2.1. Study design The CAFE study was a double-blind, flexible-dose, multisite study of patients with early psychosis randomized to receive treatment with olanzapine, quetiapine, or risperidone for 52 weeks. The main efficacy and safety results have been reported previously in The American Journal of Psychiatry (McEvoy et al., 2007). 2.2. Study participants Study participants, aged between 16 and 40 years, met Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria for schizophrenia, schizophreniform disorder, or schizoaffective disorder. Patients had been continuously ill for at least 1 month but not more than 5 years. Patients were excluded if they had a prior psychotic episode that remitted for ≥3 months or a history of antipsychotic treatment encompassing more than 16 cumulative weeks. 2.3. Study treatments Patients were randomly assigned to olanzapine (2.5– 20 mg/day), quetiapine (100–800 mg/day), or risperidone (0.5–4 mg/day). Any previous antipsychotic therapy was discontinued during the first 2 weeks of double-blind treatment. Anticholinergic medications for the treatment of extrapyramidal symptoms were permitted for a total of 2 weeks over the course of the trial. Antidepressants and traditional mood stabilizers were not permitted during the first 8 weeks of blinded treatment. Otherwise, concomitant medications could be used without restriction. 2.4. Study assessments Study eligibility was determined using the Structured Clinical Interview for DSM-IV (SCID) (First et al., 1997), medical history, physical examination, vital signs, and laboratory tests. Weight was recorded at baseline, weekly for the first 6 weeks, every other week for the next 6 weeks, and monthly thereafter. All laboratory assessments were measured at baseline and weeks 12, 24, and 52, or at time of study termination if before week 52. Laboratory tests and analyses evaluating glucose, hemoglobin A1c, and lipid levels were performed at a single centralized laboratory. At each blood draw, patients reported the number of hours since they had last consumed any food or caloric drink. Fasting was defined as no caloric consumption for at least 8 h prior to the blood draw. Blood pressure was recorded after 5 to 10 min in a seated position. Weight, height, and waist circumference (measured at the narrowest part of the torso using standard technique) were recorded. Weight was quantified using four different measures: 1) mean weight; 2) BMI (used to categorize patients as normal [BMI = 18.5–24.9], overweight [BMI = 25–29.9], or obese [BMI ≥ 30]); 3) the proportion of patients experiencing a BMI increase of one unit (used as an indicator of meaningful weight gain (Marder et al., 2004); and 4) the proportion of patients experiencing ≥7% weight gain from baseline.

J.K. Patel et al. / Schizophrenia Research 111 (2009) 9–16

3. Results

Table 1 Patient demographics.

Age (years), mean (SD) Sex, n (%) Male Female Ethnicity, n (%) White Black Other DSM-IV diagnosis, n (%) Schizophrenia Schizophreniform disorder Schizoaffective disorder Duration of previous antipsychotic (weeks), mean (SD) Duration of illness (months), mean (SD) Drug naïve, n (%)

11

Olanzapine (n = 133)

Quetiapine (n = 134)

Risperidone (n = 133)

24.7 (5.8)

25.0 (6.1)

23.9 (5.5)

101 (76) 32 (24)

92 (69) 42 (31)

99 (74) 34 (26)

61 (46) 61 (46) 11 (8)

66 (49) 60 (45) 8 (6)

78 (59) 51 (38) 4 (3)

81 (61) 35 (26) 17 (13) 6.9 (8.8)

75 42 17 6.6

75 (56) 38 (29) 20 (15) 5.4 (5.0)

(56) (31) (13) (7.3)

11 (12.9)

15.1 (20.0)

12.7 (17.9)

32 (24.2)

36 (26.9)

28 (21.1)

Four hundred patients with early psychosis from 26 sites were randomly allocated to olanzapine (n = 133), quetiapine (n = 134), or risperidone (n = 133) treatment groups. Treatment groups did not differ significantly in terms of demographic characteristics (McEvoy et al., 2007) (Table 1). Discontinuation rates were very similar during this study (olanzapine 68.4%; quetiapine 59.9% and risperidone 71.4%). Kaplan–Meier survival curves showed similar discontinuation rates across the entire range of the study (see Fig. 2 in McEvoy et al., 2007, page 1054). The majority of treatment discontinuations were secondary to patient decision (41.5%) followed by unacceptable side effects (10%), inadequate therapeutic effect (10.8%) and administrative causes (8%) (for more details, see Fig. 1 in McEvoy et al., 2007, page 1053). Patients with cumulative exposure to antipsychotics ≤2 weeks were categorized as the Minimal Antipsychotic Exposed Group (MAEG; n = 152), and the remainder were classified as the Antipsychotic Exposed Group (AEG; n = 243).

2.5. Statistical analyses 3.1. Baseline characteristics Continuous longitudinal measures were analyzed using a mixed random coefficients model, using fixed effects for time and treatment group, and random effects for intercept and time, with baseline as a covariate. Time was calculated as log2 (number of weeks from baseline). Least squares mean changes in outcome parameters were compared at weeks 12 and 52 and were not corrected for multiple comparisons. For categorical responses, a logistic regression model was used, containing treatment and baseline if appropriate. Pairwise estimated odds ratios were compared at weeks 12 and 52 and were not corrected for multiple comparisons.

Overall, approximately 31% of patients were overweight (BMI 25–29.9 kg/m2) and 18% were obese (BMI N30 kg/m2) at baseline. Mean baseline BMI in the 3 groups ranged between 25.5 and 26.1 kg/m2. Fasting glucose ≥100 mg/dL was reported in 5.1% to 10.4% patients, while hemoglobin A1c was elevated (N6.1%) in 2.4% of patients. Systolic and diastolic blood pressure was elevated (≥130 mm Hg and ≥85 mm Hg) in 18.3% to 25.8% and 12.1% to 14.4% of patients, respectively. Elevated levels of fasting triglycerides (TG; N150 mg/dL) and total cholesterol (TC; ≥200 mg/dL) were recorded in 7.7% to

Table 2 Metabolic data for minimal antipsychotic exposed and antipsychotic exposed patient groups. Variable, n (%) [unless otherwise stated] BMI N 30 kg/m2 Male Female Waist circumference N40q (male) Waist circumference N35q (female) Systolic BP ≥130 mm Hg Diastolic BP ≥ 85 mm Hg Fasting triglycerides N 150 mg/dL Fasting glucose ≥ 100 mg/dL Fasting total cholesterol ≥200 mg/dL Fasting HDL cholesterol b 40 mg/dL (male) Fasting HDL cholesterol b 50 mg/dL (female) Hemoglobin A1c N6.1% Weight (lb), mean (SD) Male Female BMI (kg/m2), mean (SD) Male Female Patients with ≥7% weight gain Male Female

Minimal antipsychotic exposed group (n = 152)

Antipsychotic exposed group (n = 243)

Baseline

Week 12

Week 52

Baseline

Week 12

Week 52

17 (16.4) 11 (25.0) 18 (17.1) 23 (54.8) 33 (22.5) 23 (15.7) 9 (10.3) 7 (8.1) 12 (13.8) 22 (34.4) 14 (60.9) 4 (2.9)

– – – – – – – – – – – –

– – – – – – – – – – – –

22 (12.2) 18 (29.0) 28 (16.0) 28 (46.7) 49 (20.2) 28 (11.5) 28 (18.9) 11 (7.4) 46 (31.1) 29 (25.0) 10 (31.3) 5 (2.1)

– – – – – – – – – – – –

– – – – – – – – – – – –

172.4 (44.6) 167.7 (57.5)

11.2 (0.9)# 9.5 (1.3)#

17.1 (1.6)# 15.1 (2.4)#

176.8 (33.7) 159.1 (46.1)

13.1 (1.3)# 8.6 (1.6)#

20.6 (2.2)# 13.8 (3.0)#

25.16 (5.65) 27.3 (7.64)

1.63 (0.13)# 1.58 (0.22) #

2.49 (0.22)# 2.50 (0.41)#

25.43 (4.40) 26.89 (7.06)

1.98 (0.19)# 1.48 (0.28)#

3.14 (0.32)# 2.38 (0.50)#

– –

50 (37.9) 17 (37.0)

26 (59.1) ⁎ 11 (61.1)

– –

33 (55.0) 11 (35.5)

19 (73.1) 6 (60.0)

Abbreviations: BMI, body mass index; BP, blood pressure; HDL, high-density lipoprotein. # Values reported at 12 and 52 weeks represent the mean change from baseline. ⁎ p b .05.

12

J.K. Patel et al. / Schizophrenia Research 111 (2009) 9–16

Fig. 1. Weight gain after 12 and 52 weeks of treatment with olanzapine, quetiapine, and risperidone.

20.5% and 21.8% to 27.9% patients, and were higher in AEG compared with MAEG patients (Table 2).

significant relationship between changes in total positive and negative syndrome scale (PANSS) score and changes in weight after 12 weeks of treatment (p = .936).

3.2. Weight and metabolic changes after 12 weeks 3.3. Weight and metabolic changes after 52 weeks After 12 weeks, the mean weight gain and BMI increases associated with olanzapine were significantly higher than those associated with risperidone or quetiapine (p b .001; Fig. 1). Weight gain after 12 weeks of treatment was estimated as [Least Squares Mean (SE)] 15.6 (±1.1) lb for olanzapine, 8.6 (±1.1) lb with risperidone and 7.9 (±1.1) lb for quetiapine, respectively. Weight gain ≥7% occurred significantly more often among olanzapine- (59.8%) than risperidone-, (32.5%, p b .001) or quetiapine-treated patients (29.2%, p b .0001; Fig. 2). Mean fasting TG levels were significantly higher during quetiapine treatment than during risperidone treatment (43.7 vs 7.8 mg/dL, respectively; p b .05). There was no statistically

At 52 weeks, mean weight gain was highest among olanzapine-treated patients [24.2 lb (±1.9), 14.0 (±1.9) lb, and 12.1 (±1.8) lb, for olanzapine, risperidone, and quetiapine, respectively; p b .001]. Mean weight gain per month was 3.88 (standard deviation [SD]: 4.59) lb, 2.81 (SD 5.07) lb, and 2.85 (SD 7.66) lb for olanzapine, risperidone, and quetiapine, respectively (p N .05). BMI increases were significantly greater during olanzapine treatment compared with risperidone or quetiapine treatment (p b .001). BMI increases ≥1 unit occurred significantly more frequently with olanzapine (88.6%) compared with risperidone (69.7%; p b .05) and quetiapine (63.3%;

Fig. 2. Weight gain after 12 and 52 weeks of treatment with olanzapine, quetiapine, and risperidone.

J.K. Patel et al. / Schizophrenia Research 111 (2009) 9–16

p b .05). Eighty percent of olanzapine-compared with 57.6% risperidone- (p b .05) and 50.0% quetiapine- (p b .01) treated patients recorded weight gain ≥7% from baseline (Fig. 2). At baseline, mean fasting glucose, TC, and high-density lipoprotein (HDL) cholesterol levels were not significantly different across treatments. There was a significant correlation between weight gain and the change in TC levels during treatment with olanzapine (p b .01) and quetiapine (p b .05; data not shown). The increase in fasting TG levels was significantly lower during treatment with risperidone compared with quetiapine (8.2 vs. 44.3 mg/dL, respectively; p b .05). There was no statistically significant relationship between changes in total PANSS score and changes in weight after 52 weeks of treatment (p = .338). 3.4. The effect of gender on weight gain and metabolic indices At baseline, 14% of male and 28% of female patients were obese. Across treatment groups, mean waist circumference ranged between 34.9 and 35.4 in. in men and 34.8 and 36.2 in. in women. HDL cholesterol was below normal levels (b40 mg/dL for men and b50 mg/dL for women) in 26.6% to 29.8% of men and 40.9% to 50.0% of women. Women in the quetiapine treatment group showed the lowest mean weight gain and the smallest increase in mean BMI of all treatment groups. Women treated with quetiapine were significantly less likely to gain ≥7% of their body weight than women treated with olanzapine after 12 weeks (p b .01) and women treated with risperidone after 52 weeks (p b .05) (Fig. 2). Weight gain ≥7% was significantly more common in men during olanzapine treatment relative to risperidone, after 12 (p b .01) and 52 (p b .05) weeks of treatment. At both time points, more women than men registered a waist circumference N35 and N40 in., respectively (Fig. 2). 3.5. Metabolic syndrome At baseline, 4.3% patients met AHA criteria for MetS (Table 3). The baseline incidence of MetS was higher among women than men (11.2% vs. 2.1%, respectively; p b .001). After 52 weeks, 18.6% of the total sample (57/307 patients) met

13

AHA criteria for MetS, the incidence of which was comparable between men and women (12.7% vs. 15.5%, respectively; p N .05). Treatment-emergent MetS occurred in 51 individuals (13.4% of the total population), of whom 22 were receiving olanzapine, 18 quetiapine, and 11 risperidone. Treatmentemergent MetS was reported in a significantly lower proportion of African-American patients than in patients of other ethnicities (8.5% vs. 17.2% respectively; p b .05), and in significantly fewer MAEG than AEG patients (8.6% vs. 16.2%, respectively; p b .05; Table 3). The mean time to the development of treatment-emergent MetS was 12.4 weeks. 4. Discussion This study compared the metabolic profiles of three commonly prescribed SGAs in young patients with early psychosis. The study design, duration of treatment, and large sample size are its major strengths. All-cause treatment discontinuation rates were comparable for olanzapine, quetiapine, and risperidone (McEvoy et al., 2007). At baseline, nearly half of the patients were overweight or obese. Mean BMI, already in the overweight category, was higher than that reported in previous studies of early psychosis (Alvarez-Jimenez et al., 2006; Perez-Iglesias et al., 2008; Zipursky et al., 2005). The overweight/obesity rates of 27–34% at baseline reported previously (Addington et al., 2006; Perez-Iglesias et al., 2008) were lower than those observed in our study. According to recent data from NHANES-III for the years 2003 to 2004, 17% and 57% of the population aged 12 to 19 years and 20 to 39 years, respectively, were overweight or obese. Thus, baseline overweight/obesity rates reported in our study are broadly comparable to those of the U.S. population during the study recruitment period, implying that patients with early psychosis confront similar negative consequences of overweight/ obesity to the general population even before meaningful antipsychotic exposure. Furthermore, in our study, more women than men were obese at baseline, even with minimal or no exposure to antipsychotics. In the NHANES-III study, 18% of men and 16% of women aged 12 to 19 years were overweight, while 28% of men and 29% of women aged 20

Table 3 Baseline and treatment-emergent MetS. Metabolic syndrome Baseline N (total) All patients Sex Male Female Ethnicity African-American Other Treatment history MAEG AEG Current treatment Olanzapine Quetiapine Risperidone

%

Time to metabolic syndrome Treatment-emergent

Total

Treatment-emergent

Total

N (total)

%

N (total)

%

Weeks (SD)

Weeks (SD)

17 (385)

4.3

51 (380)

13.4

57 (307)

18.6

12.4 (2.1)

12.3 (1.4)

6 (287) 11 (98)

2.1 11.2

36 (283) 15 (97)

12.7 15.5

38 (228) 19 (79)

16.7 24.1

– –

– –

5 (172) 12 (228)

2.9 5.3

14 (165) 37 (215)

8.5 17.2

15 (170) 42 (227)

8.8 18.5

– –

– –

12 (153) 5 (247)

7.8 2.0

12 (139) 39 (241)

8.6 16.2

17 (151) 40 (426)

11.3 16.3

12.0 (1.7) 12.5 (2.3)

12.0 (1.4) 12.5 (2.2)

7 (133) 6 (134) 4 (133)

5.3 4.5 3.0

22 (125) 18 (126) 11 (129)

17.6 14.3 8.5

24 (132) 21 (132) 12 (133)

18.2 15.9 9.0

12.9 (2.8) 11.6 (1.3) 12.5 (1.2)

12.9 (2.7) 11.7 (1.2) 12.5 (1.1)

MetS defined according to AHA criteria. Abbreviations: MAEG, Minimal Antipsychotic Exposed Group; AEG, Antipsychotic Exposed Group.

14

J.K. Patel et al. / Schizophrenia Research 111 (2009) 9–16

to 39 years were obese (Ogden et al., 2006); this suggests that, at baseline, the women in our study (aged 16–44 years) had a higher than average BMI for their age. These findings mimic the obesity pattern reported in women with chronic schizophrenia when compared with the general noninstitutionalized population, both before (Allison et al., 1999) and after increased use of SGAs (Homel et al., 2002). Thus, young women with early psychosis, like those suffering from chronic schizophrenia, appear more vulnerable to weight gain even before meaningful antipsychotic exposure occurs. It has been speculated that these effects may be modulated by gonadal steroid hormones (Homel et al., 2002). Future studies evaluating the metabolic effects of psychotropic drugs should take into consideration these findings. Significant weight gain occurred with high frequency across all treatment groups, but was most pronounced in the olanzapine group. This is consistent with previous studies of early psychosis (Alvarez-Jimenez et al., 2006; Crespo-Facorro et al., 2006; Graham et al., 2005; Green et al., 2006; Kahn et al., 2008; Lambert et al., 2005; Lieberman et al., 2003; Robinson et al., 2002; Sanger et al., 1999; Sengupta et al., 2005; Wu et al., 2006; Zipursky et al., 2005) in which weight gain up to 30 lb was reported after approximately 12 weeks (Alvarez-Jimenez et al., 2006; Lieberman et al., 2003; Sanger et al.,1999; Sengupta et al., 2005; Zipursky et al., 2005) and 52 weeks (Kahn et al., 2008) and 34 lb after 2 years of treatment with olanzapine (Zipursky et al., 2005). Similarly, risperidone-associated weight gain in patients with early psychosis has been reported in the range of 10 to 17 lb after 12 weeks (Alvarez-Jimenez et al., 2006; Kahn et al., 2008) and 17 lb after 2 years (Schooler et al., 2005). In our study, the mean weight gain of 8 lb at 12 weeks and 13 lb at 52 weeks associated with quetiapine was lower than that reported for olanzapine or risperidone and in line with previous reports of weight gain during quetiapine treatment of approximately 9 lb after 12 weeks (Tauscher-Wisniewski et al., 2002) and 12 to 30 lb after 1 year or more 2 years (Kahn et al., 2008; Kopala et al., 2006). Though the mean weight gain per month was not significantly different across the treatments in our study, it was numerically higher for olanzapine compared with the other two SGAs. Substantial weight gain of N7% was recorded significantly more frequently with olanzapine compared with quetiapine or risperidone, consistent with previous reports (Alvarez-Jimenez et al., 2006; Kahn et al., 2008; Zipursky et al., 2005). More patients had an increase in BMI ≥1 unit—a measure that provides an early threshold for clinical intervention to manage weight gain (Marder et al., 2004)—than ≥7% weight gain. BMI increases ≥1 unit were significantly more frequent during olanzapine treatment compared with risperidone or quetiapine. More than half of the total weight gain occurred during the first 3 months of antipsychotic treatment, a phenomenon also reported in other studies (Addington et al., 2003; Lieberman et al., 2003; Schooler et al., 2005; Zipursky et al., 2005). Thus, for weight gain measures of magnitude and frequency, quetiapine and risperidone had a better profile than olanzapine. This differential pattern of weight gain is of clinical importance, although it is equally important to recognize that ≥50% of patients treated with quetiapine or risperidone also recorded substantial weight gain. We did not find a significant correlation between changes in weight and clinical symptoms as reported previously in some studies (Cook et al., 2003; Ford et al., 2002).

Young women with early psychosis showed a differential pattern of weight gain according to the antipsychotic. Women treated with quetiapine showed significantly lower weight gain than those treated with either olanzapine or risperidone. Women treated with risperidone gained weight the most frequently. If replicated, these findings could provide some rationale for tailoring of antipsychotic selection in women with schizophrenia. The incidence of treatment-emergent MetS associated with SGAs is important. At baseline, 4.3% of patients met criteria for MetS, a prevalence rate probably lower than the general population (Ascher-Svanum et al., 2005; Czobor et al., 2002). However, 13.4% of patients developed treatment-emergent MetS over the course of 1 year, which is a worrisome increase. Significantly more AEG than MAEG patients developed treatment-emergent MetS suggesting that even short-term exposure to antipsychotics may prime patients for future metabolic changes. This is borne out by the higher number of patients with abnormal lipid levels in the AEG group at baseline. Although significantly more women met criteria for MetS at baseline, there were no significant gender differences in the incidence of MetS at 52 weeks. This may reflect recent population trends in which the prevalence of obesity is rapidly increasing in men and has plateaued in women (Ogden et al., 2006). Why significantly more non-African-Americans developed treatment-emergent MetS in our study remains unclear, but may reflect worsening lifestyle trends. The mean time to development of treatment-emergent MetS of approximately 12 weeks is rather alarming. These data, in conjunction with the high reported rates of discontinuation or switching of antipsychotic treatment (McEvoy et al., 2006; Lieberman et al., 2005), may not bode well for the long-term treatment of young patients with early psychosis unless effective interventions to improve treatment adherence as well as lifestyle changes are implemented concurrently. When individual MetS criteria were examined at 52 weeks, changes in mean fasting glucose and hemoglobin A1c levels were not significantly different across treatments, despite significant weight gain. Similar findings have been reported in patients with early psychosis (Kahn et al., 2008; Kopala et al., 2006; Lieberman et al., 2003; Sengupta et al., 2005; Zipursky et al., 2005). BMI increases also failed to correlate with changes in fasting glucose levels, a finding also reported previously (Zipursky et al., 2005). These data suggest that a considerable time lag may exist between significant weight gain and the development of type 2 diabetes mellitus in patients with early psychosis treated with SGAs (Meyer and Koro 2004), even if changes in insulin levels do occur early. Interestingly, the largest changes in TG and TC levels occurred during quetiapine treatment, during which the lowest degree of weight gain occurred. Thus, contrary to some studies (Graham et al., 2005; Kopala et al., 2006; Wu et al., 2006), changes in lipid parameters may not always correspond with changes in weight. In patients with early psychosis, some studies have reported significant increases in TG levels following olanzapine treatment associated with weight gain (Graham et al., 2005; Wu et al., 2006), although other studies have failed to find such an association (Sengupta et al., 2005). While selected studies have associated olanzapine-induced weight gain with increases in nonfasting TC levels in patients with early psychosis (Sengupta et al., 2005; Zipursky et al., 2005), others have observed

J.K. Patel et al. / Schizophrenia Research 111 (2009) 9–16

only marginal relationships (Lindenmayer et al., 2003). Thus, much variability exists between studies and across treatments, an observation further compounded by a lack of consistent association between antipsychotic-induced weight gain and changes in lipid parameters. These data once again justify individualizing antipsychotic treatment for patients with early psychosis. HDL cholesterol showed nonsignificant reductions across the treatments after 52 weeks, a finding reported previously (Kahn et al., 2008; Sengupta et al., 2005). Some have suggested that obesity may itself cause decreases in HDL cholesterol levels (Meyer and Koro, 2004). Further prospective studies of the association between HDL cholesterol changes and antipsychotic treatment are warranted, as this represents a very important risk factor for atherosclerotic cardiovascular disease. 4.1. Study limitations This study was designed before aripiprazole and ziprasidone were approved and available, and hence it was not possible to include these two medications. There is evidence that both aripiprazole and ziprasidone may have less impact on weight and metabolic parameters than quetiapine, olanzapine, and risperidone and it would have been useful to include them in the same trial. The trial was designed as a noninferiority trial with all-cause treatment discontinuation as the primary outcome. Weight and metabolic parameters were secondary variables. Thus, the trial design was optimized for assessment of psychotic symptoms and assessment of allcause treatment discontinuation. The trial was designed with a 12-week acute phase and a 52-week total length, with analyses planned at each of the two major time points. The 12week phase corresponds to what was thought to be the smallest amount of time which might demonstrate differences among the treatments, while the 52-week study was designed to retain at least the majority of the subjects. Since the weight and metabolic analyses were secondary, the analyses we did should be considered exploratory, and the results suggest hypotheses that need to be confirmed on an independent set of data. Readers should not interpret these results as definitive but rather as indicative. Because this was an exploratory analysis, and because the interpretation is suggestive, we did not correct for multiple comparisons. Readers should examine the magnitudes of the effects as well as the results of the hypothesis tests in their interpretation of the results. 5. Conclusion We found that young patients with early psychosis show increased sensitivity to the side effect of substantial weight gain during SGA treatment after 1 year, with more than half of weight gain taking place during the first 12 weeks. Treatment-emergent MetS was reported in a large number of patients at the end of the study. Individualizing antipsychotic treatment as well as targeting interventions to deal with weight gain and metabolic indices is warranted from the very onset of antipsychotic therapy. Role of funding source Funding for this program was provided by AstraZeneca Pharmaceuticals LP.

15

Contributors Dr. Lieberman was the Principal Investigator of the study and developed the protocol with significant contribution from Drs. McEvoy, Perkins, and Hamer. All authors contributed to developing specific questions and a statistical plan to analyze the data for this manuscript. Dr. J. K. Patel and Dr. P. Buckley managed the literature search and analyses, Dr. Hamer and Ms. Woolson undertook the statistical analysis, and Dr. Patel wrote the first draft of the manuscript. All authors contributed to and have approved the final manuscript. The Comparison of Atypicals in First Episode of Psychosis research program was coordinated by the University of North Carolina. UNC was responsible for the integrity of all data, including data collation and analyses. Conflict of interest Dr. Patel is currently receiving research grant support from: the National Institute of Health, the Foundation for the National Institute of Health, Johnson and Johnson, Wyeth, Bristol-Myers Squibb, Forest Pharmaceuticals and the United BioSource Corporation, and has acted as a speaker for: Eli Lilly, Pfizer, AstraZeneca, Janssen Pharmaceutica, Bristol-Myers Squibb, GlaxoSmithKline, and Abbott. Dr. Buckley is currently receiving research grant support from: AstraZeneca, the National Institute of Mental Health, Pfizer, Solvay, Janssen Pharmaceutica and Wyeth, and is a consultant for National Institute of Mental Health and Janssen Pharmaceutica. Ms. Woolson reports no competing interests. Dr. Hamer has acted as a consultant for: Corcept, Eli Lilly, Epix, Johnson and Johnson, Sanofi-Aventis, and Wyeth, and is involved in a contract agreement between UNC and AstraZeneca. Dr. Hamer has served on an advisory board for: eNabledMD, Eli Lilly, and Wyeth and on a DSMB/IDMC for: Allergan, Pfizer, Schwartz, and Solvay. Dr. Hamer holds stock in BristolMyers Squibb, Amgen, Eli Lilly, Genentech, Proctor and Gamble, and Sepracor. Dr. McEvoy receives honoraria from Eli Lilly and Janssen, is on the advisory board for Eli Lilly and Organon and has research funding from Sanofi. Dr. Perkins is currently receiving grant support from Janssen, and is a consultant to Dainippon. In last 12 months, Dr. Perkins is or has been on speaker's bureau for Eli Lilly and AstraZeneca. In the past, Dr. Perkins reports having received research funding from AstraZeneca, Bristol-Myers Squibb, Otsuka, Eli Lilly, Janssen and Pfizer; and consulting and educational fees from AstraZeneca, Bristol-Myers Squibb, Eli Lilly, Janssen, Glaxo Smith Kline, Forest Labs, Pfizer and Shire. Dr. Lieberman serves as a consultant and/or advisor for Astra Zeneca, Eli Lilly, Forest Laboratories, GlaxoSmithKline, Pfizer, Bioline, Janssen, and Wyeth; and as a member of the Data Safety Management Board (DSMB) for Solvay and Wyeth. He does not receive financial compensation or salary support for his participation as a consultant or as a member of a board. He receives grant support from AstraZeneca, Allon, Bristol-Myers Squibb, Forest, GlaxoSmithKline, Janssen, Merck, Pfizer and Wyeth; and he holds a patent from Repligen. Acknowledgements The authors would like to acknowledge the editorial assistance of Eleanor Bull, Ph.D., (PAREXEL MMS). Financial support for this assistance was provided by AstraZeneca Pharmaceuticals LP.

References Addington, J., Mansley, C., Addington, D., 2003. Weight gain in first-episode psychosis. Can. J. Psychiatry 48, 272–276. Addington, J., Saeedi, H., Addington, D., 2006. Weight gain in first-episode psychosis over three years. Schizophr. Res. 86, 335–336. Allison, D.B., Fontaine, K.R., Heo, M., Mentore, J.L., Cappelleri, J.C., Chandler, L.P., Weiden, P.J., Cheskin, L.J., 1999. The distribution of body mass index among individuals with and without schizophrenia. J. Clin. Psychiatry 60, 215–220. Alvarez-Jimenez, M., Gonzalez-Blanch, C., Vazquez-Barquero, J.L., PerezIglesias, R., Martinez-Garcia, O., Perez-Pardal, T., Ramirez-Bonilla, M.L., Crespo-Facorro, B., 2006. Attenuation of antipsychotic-induced weight gain with early behavioral intervention in drug-naive first-episode psychosis patients: a randomized controlled trial. J. Clin. Psychiatry 67, 1253–1260. Ascher-Svanum, H., Stensland, M.D., Kinon, B.J., Tollefson, G.D., 2005. Weight gain as a prognostic indicator of therapeutic improvement during acute treatment of schizophrenia with placebo or active antipsychotic. J. Psychopharmacol. 19, 110–117.

16

J.K. Patel et al. / Schizophrenia Research 111 (2009) 9–16

CDC, 2006. State-specific prevalence of obesity among adults—United States. MMWR 55, 985–988. Cook, S., Weitzman, M., Auinger, P., Nguyen, M., Dietz, W.H., 2003. Prevalence of a metabolic syndrome phenotype in adolescents: findings from the third National Health and Nutrition Examination Survey, 1988–1994. Arch. Pediatr. Adolesc. Med. 157, 821–827. Crespo-Facorro, B., Perez-Iglesias, R., Ramirez-Bonilla, M., Martínez-García, O., Llorca, J., Luis Vázquez-Barquero, J., 2006. A practical clinical trial comparing haloperidol, risperidone, and olanzapine for the acute treatment of first-episode nonaffective psychosis. J. Clin. Psychiatry 67, 1511–1521. Czobor, P., Volavka, J., Sheitman, B., Lindenmayer, J.P., Citrome, L., McEvoy, J., Cooper, T.B., Chakos, M., Lieberman, J.A., 2002. Antipsychotic-induced weight gain and therapeutic response: a differential association. J. Clin. Psychopharmacol. 22, 244–251. First, M., Spitzer, R., Gibbons, M., 1997. Structured Clinical Interview for DSMIV Axis I Disorders (SCID), Clinician Version. American Psychiatric Press, Washington, DC. Ford, E.S., Giles, W.H., Dietz, W.H., 2002. Prevalence of the metabolic syndrome among US adults: findings from the third National Health and Nutrition Examination Survey. JAMA 287, 356–359. Graham, K.A., Perkins, D.O., Edwards, L.J., Barrier Jr., R.C., Lieberman, J.A., Harp, J.B., 2005. Effect of olanzapine on body composition and energy expenditure in adults with first-episode psychosis. Am. J. Psychiatry 162, 118–123. Green, A.I., Lieberman, J.A., Hamer, R.M., Glick, I.D., Gur, R.E., Kahn, R.S., McEvoy, J.P., Perkins, D.O., Rothschild, A.J., Sharma, T., Tohen, M.F., Woolson, S., Zipursky, R.B., 2006. Olanzapine and haloperidol in first episode psychosis: two-year data. Schizophr. Res. 86, 234–243. Homel, P., Casey, D., Allison, D.B., 2002. Changes in body mass index for individuals with and without schizophrenia, 1987–1996. Schizophr. Res. 55, 277–284. Kahn, R.S., Fleischhacker, W.W., Boter, H., Davidson, M., Vergouwe, Y., Keet, I.P., Gheorghe, M.D., Rybakowski, J.K., Galderisi, S., Libiger, J., Hummer, M., Dollfus, S., López-Ibor, J.J., Hranov, L.G., Gaebel, W., Peuskens, J., Lindefors, N., Riecher-Rössler, A., Grobbee, D.E., 2008. Effectiveness of antipsychotic drugs in first-episode schizophrenia and schizophreniform disorder: an open randomised clinical trial. Lancet 371, 1085–1097. Kopala, L.C., Good, K.P., Milliken, H., Buiteman, C., Woodley, H., Rui, Q., Whitehorn, D., Love, L., Balshaw, R., Kiss, I., Honer, W.G., 2006. Treatment of a first episode of psychotic illness with quetiapine: an analysis of 2 year outcomes. Schizophr. Res. 81, 29–39. Kumra, S., Kranzler, H., Gerbino-Rosen, G., Kester, H.M., De Thomas, C., Kafantaris, V., Correll, C.U., Kane, J.M., 2008. Clozapine and “high-dose” olanzapine in refractory early-onset schizophrenia: a 12-week randomized and double-blind comparison. Biol. Psychiatry 63, 524–529. Lambert, M., Conus, P., Schimmelmann, B.G., Eide, P., Ward, J., Yuen, H., Schacht, M., Edwards, J., Naber, D., McGorry, P.D., 2005. Comparison of olanzapine and risperidone in 367 first-episode patients with nonaffective or affective psychosis: results of an open retrospective medical record study. Pharmacopsychiatry 38, 206–213. Lieberman, J.A., Tollefson, G., Tohen, M., Green, A.I., Gur, R.E., Kahn, R., McEvoy, J., Perkins, D., Sharma, T., Zipursky, R., Wei, H., Hamer, R.M., 2003. Comparative efficacy and safety of atypical and conventional antipsychotic drugs in first-episode psychosis: a randomized, doubleblind trial of olanzapine versus haloperidol. Am. J. Psychiatry 160, 1396–1404. Lieberman, J.A., Stroup, T.S., McEvoy, J.P., Swartz, M.S., Rosenheck, R.A., Perkins, D.O., Keefe, R.S., Davis, S.M., Davis, C.E., Lebowitz, B.D., Severe, J., Hsiao, J.K., 2005. Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. N. Engl. J. Med. 353, 1209–1223. Lindenmayer, J.P., Czobor, P., Volavka, J., Citrome, L., Sheitman, B., McEvoy, J.P., Cooper, T.B., Chakos, M., Lieberman, J.A., 2003. Changes in glucose and cholesterol levels in patients with schizophrenia treated with typical or atypical antipsychotics. Am. J. Psychiatry 160, 290–296. Marder, S.R., Essock, S.M., Miller, A.L., Buchanan, R.W., Casey, D.E., Davis, J.M., Kane, J.M., Lieberman, J.A., Schooler, N.R., Covell, N., Stroup, S., Weissman,

E.M., Wirshing, D.A., Hall, C.S., Pogach, L., Pi-Sunyer, X., Bigger Jr., J.T., Friedman, A., Kleinberg, D., Yevich, S.J., Davis, B., Shon, S., 2004. Physical health monitoring of patients with schizophrenia. Am. J. Psychiatry 161, 1334–1349. McEvoy, J.P., Meyer, J.M., Goff, D.C., Nasrallah, H.A., Davis, S.M., Sullivan, L., Meltzer, H.Y., Hsiao, J., Stroup, T.S., Lieberman, J.A., 2005. Prevalence of the metabolic syndrome in patients with schizophrenia: baseline results from the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) schizophrenia trial and comparison with national estimates from NHANES III. Schizophr. Res. 80, 19–32. McEvoy, J.P., Lieberman, J.A., Stroup, T.S., Davis, S.M., Meltzer, H.Y., Rosenheck, R.A., Swartz, M.S., Perkins, D.O., Keefe, R.S., Davis, C.E., Severe, J., Hsiao, J.K., 2006. Effectiveness of clozapine versus olanzapine, quetiapine, and risperidone in patients with chronic schizophrenia who did not respond to prior atypical antipsychotic treatment. Am. J. Psychiatry 163, 600–610. McEvoy, J.P., Lieberman, J.A., Perkins, D.O., Hamer, R.M., Gu, H., Lazarus, A., Sweitzer, D., Olexy, C., Weiden, P., Strakowski, S.D., 2007. Efficacy and yolerability of olanzapine, quetiapine, and risperidone in the treatment of early psychosis: a randomized, double-blind 52-week comparison. Am. J. Psychiatry 164, 1050–1060. Meyer, J.M., Koro, C.E., 2004. The effects of antipsychotic therapy on serum lipids: a comprehensive review. Schizophr. Res. 70, 1–17. Ogden, C.L., Carroll, M.D., Curtin, L.R., McDowell, M.A., Tabak, C.J., Flegal, K.M., 2006. Prevalence of overweight and obesity in the United States, 1999– 2004. JAMA 295, 1549–1555. Perez-Iglesias, R., Crespo-Facorro, B., Martinez-Garcia, O., Ramirez-Bonilla, M.L., Alvarez-Jimenez, M., Pelayo-Teran, J.M., Garcia-Unzueta, M.T., Amado, J.A., Vazquez-Barquero, J.L., 2008. Weight gain induced by haloperidol, risperidone and olanzapine after 1 year: findings of a randomized clinical trial in a drug-naïve population. Schizophr. Res. 99, 13–22. Robinson, D.G., Woerner, M.G., Alvir, J.M., Bilder, R.M., Hinrichsen, G.A., Lieberman, J.A., 2002. Predictors of medication discontinuation by patients with first-episode schizophrenia and schizoaffective disorder. Schizophr. Res. 57, 209–219. Sanger, T.M., Lieberman, J.A., Tohen, M., Grundy, S., Beasley Jr., C., Tollefson, G.D., 1999. Olanzapine versus haloperidol treatment in first-episode psychosis. Am. J. Psychiatry 156, 79–87. Schooler, N., Rabinowitz, J., Davidson, M., Emsley, R., Harvey, P.D., Kopala, L., McGorry, P.D., Van Hove, I., Eerdekens, M., Swyzen, W., De Smedt, G., 2005. Risperidone and haloperidol in first-episode psychosis: a longterm randomized trial. Am. J. Psychiatry 162, 947–953. Sengupta, S.M., Klink, R., Stip, E., Baptista, T., Malla, A., Joober, R., 2005. Weight gain and lipid metabolic abnormalities induced by olanzapine in first-episode, drug-naive patients with psychotic disorders. Schizophr. Res. 80, 131–133. Stroup, T.S., Lieberman, J.A., McEvoy, J.P., Swartz, M.S., Davis, S.M., Rosenheck, R.A., Perkins, D.O., Keefe, R.S., Davis, C.E., Severe, J., Hsiao, J.K., 2006. Effectiveness of olanzapine, quetiapine, risperidone, and ziprasidone in patients with chronic schizophrenia following discontinuation of a previous atypical antipsychotic. Am. J. Psychiatry 163, 611–622. Tauscher-Wisniewski, S., Kapur, S., Tauscher, J., Jones, C., Daskalakis, Z.J., Papatheodorou, G., Epstein, I., Christensen, B.K., Zipursky, R.B., 2002. Quetiapine: an effective antipsychotic in first-episode schizophrenia despite only transiently high dopamine-2 receptor blockade. J. Clin. Psychiatry 63, 992–997. Wu, R.R., Zhao, J.P., Liu, Z.N., Zhai, J.G., Guo, X.F., Guo, W.B., Tang, J.S., 2006. Effects of typical and atypical antipsychotics on glucose-insulin homeostasis and lipid metabolism in first-episode schizophrenia. Psychopharmacology (Berl) 186, 572–578. Zipursky, R.B., Gu, H., Green, A.I., Perkins, D.O., Tohen, M.F., McEvoy, J.P., Strakowski, S.M., Sharma, T., Kahn, R.S., Gur, R.E., Tollefson, G.D., Lieberman, J.A., 2005. Course and predictors of weight gain in people with first-episode psychosis treated with olanzapine or haloperidol. Br. J. Psychiatry 187, 537–543.