The "Seroquel" Outcomes Study (SOS): Efficacy and tolerability oí
Transcription
The "Seroquel" Outcomes Study (SOS): Efficacy and tolerability oí
I International Journal oi Psychiatry in Clinical Practice, 2007; 11(3): 222-232 informa healthcare ORIGINAL ARTICLE The "Seroquel" Outcomes Study (SOS): Efficacy and tolerability oí quetiapine in a long-term, naturalistic study oí patients with schizophrenia JUAN GIBERT1, JosÉ GINER2, JULIO BOBES3, MÓNICA TAFALl.A\ SANTIAGO HERRANZ\ CARMEN OVEJER05 & FERNANDO RICO-VIllADEMOROS5 for the SOS GROUP* 1Facultad de Medicina, Departamento de Farmacología, Cádiz, Spain, 2Departamento de Psiquiatría, Hospital Universitario "Virgen Macarena", Universidad de Sevilla, Sevilla, Spain, 3Departamento de Psiquiatría, Facultad de Medicina, Universidad de Oviedo, Oviedo, Spain, 4Departamento Médico, AstraZeneca, Madrid, Spain, and 5Biométrica, Madrid, Spain Abstraet Objective. The "Seroquel" Outcomes Study (SOS) aimed to assess the efficacy and tolerability of quetiapine in patients with schizophrenia in the clinical practice setting. Methods. A 6-month, non-comparative, open-label study in adults with schizophrenia in a standard care setting in Spain. Outpatients received tlexibly dosed quetiapine. Efficacy was evaluated using the BriefPsychiatry Rating Scale (BPRS) and the Clinical Globallmpression (CGI) scale. BPRS response was defined as ~30% decrease from baseline. Tolerability was assessed using the Simpson-Angus Scale (SAS) and a modified Udvalg for Kliniske Undersogelser (UKU) side-effeets scale. Results. A total of 2029 patients enrolled. Significant changes from baseline to Month 6 were recorded for BPRS total and subscale scores (P <0.001). Compared with doses of~600 mg/day, doses of <400 mg/day were a strong predictor of a lower response rate (OR 0.62; 95% CI: 0.48, 0.82) and higher withdrawal rate (OR 3.3; 95% CI; 2.5, 4.4). Mean change in weight was minimal (+0.4 kg). Somnolence (26.7%), asthenia (12.5%), and constipation (9.8%) were the most common adverse events. Conclusion. Quetiapine was found to improve symptoms of schizophrenia, as indicated by a significant decrease in BPRS scores, and was well tolerated by patients in clinical practice. Key Words: Quetiapine, antipsychotics, schizophrenia, naturalistic, outpatient Introduction Schizophrenia is a chronic and debilitating illness that impacts on the cognitive, affective, behavioural, and motivational functioning of patients [1]. The World Health Organization has reported a worldwide prevalence rate of 0.1-1.7% [2], while in the US, the lifetime prevalence of the disease is estimated at 1% [3]. The development of the so-caBed "atypical" antipsychotics in the 1990s extended the treatment options for patients with schizophrenia. In addition to effectively treating the positive symptoms of schizophrenia, these agents are effective against negative, affective and cognitive symptoms. Their * See Appendix for tbe complete list of SOS Group members. broad therapeutic profile and improved tolerability compared with the conventional antipsychotics, particularly with regard to extrapyramidal symptoms (EPS), means that the atypical antipsychotics are considered first-line treatment for schizophrenia in most clinical situations [4]. The modero standard for evaluating new medicines is the randomised clínical tríal [5]. Although the use of proper controls, randomisation and blinding decreases bias and the potential for error, the conditions under which such trials are conducted differ from everyday clínical practice [6,7]. In the "reallife" setting, the typical patient often presents with comorbid disorder(s), has not responded to previous treatment efforts, and may require a number of different treatments. Further, trial patients are usually carefully selected and must meet strict study Correspondence: Mónica Tafalla, MD, Medical Department, AstraZeneca, Parque Norte, Edificio Roble, Serrano Galvache 56, 28033 Madrid, Spain. Tel.: +34 91 301 9647. Fax: +3491 301 9104. E-mail: [email protected] (Received 10 February 2006; accepted 7 December 2006) ISSN 1365-1501 print/lSSN 1471-1788 online iD 2007 Taylor & Francis DOI: 10.1080/13651500601176963 NaturaJistic study of quetiapine for schizophrenia inclusion/exclusion criteria. Naturalistic studies closely mirror daily clinical practice, but have the disadvantage of being uncontrolled and as they are observational are potentially open to bias. How a new medicine performs in clinical practice is a proper indication of its ttue benefit to risk ratio [8]. It is, therefore, important for the prescribing clinician to consider clinical trial data in conjunction with data from naturalistic studies as well as his/her own experience when selecting an atypical antipsychotic for a given patient [9]. Quetiapine is a dibenzothiazepine atypical antipsychotic with affinity for various neurotransmitter receptors including serotonin, dopamine, histamine, and adrenergic receptors and its chemical strueture is similar to that of clozapine [10]. Three pivotal, randomised, double-blind clinical trials demonstrated that quetiapine efIectively reduces psychotic symptoms in patients with schizophrenia as indicated by significant difIerences between quetiapine and placebo on the Brief Psychiatty Rating Scale (BPRS) total, BPRS positive-symptom, Scale for the Assessment of Negative Symptoms (SANS) and Clinical Global Impression (CGI) severity of illness scales [11-13]. Quetiapine has shown comparable efficacy to chIorpromazine [14] and haloperidol [15,16]. The main objective of the "Seroquel" Outcomes Study (SOS) was to assess the efficacy and tolerability of the atypical antipsychotic, quetiapine ("Seroquel"), in a large representative group of patients with schizophrenia under naturalistic conditions. It also assessed factors relevant to the use of quetiapine in the clinical setting, such as the reduction of side efIects in patients who had switched to quetiapine from previous antipsychotic medications. Methods Patients Male or female outpatients aged ~ 18 years with a DSM-IV diagnosis of schizophrenia or schizophreniform disorder for whom study investigators prescribed quetiapine as part of their normal clinical practice were included in the study. Patients were excluded if contraindicated by the quetiapine product label, for example, known hypersensitivity to the medication or any of its ingredients. This study was conducted in accordance with the Declaration of Helsinki [17]. Written informed consent was obtained from all patients prior to enrolment and the study was reported to the Ministty of Health in accordance with Spanish regu1ations at the time of the trial. Study design and dosing This was a non-comparative, open-Iabel, naturalistic, multicentre study conducted in Spain and 223 involved 279 investigators (see Appendix). Eligible patients received open treatment with quetiapine (flexiblydosed according to the clinical judgement of the study investigator) and were followed for 6 months (Figure 1). The use of concomitant medications was permitted at the discretion of the clinician. Patients were categorised into two groups at baseline: (1) patients not currently receiving antipsychotic treatment (first or breakthrough episode); (2) patients currently receiving antipsychotic treatment and requiring combination treatment with quetiapine or a switch to quetiapine due to an inadequate response or intolerable adverse events. Inadequate response was defined as persistent positive, negative, cognitive or severe depressive symptoms, persistent agitation, or treatment-refractory schizophrenia. Intolerable adverse events were determined by the psychiatrist and included sedation, weight gain, EPS, cognitive, anticholinergic, cardiovascular or sexual/reproductive adverse events, tardive dyskinesia or neuroleptic malignant syndrome. Efficacy assessments The primary efficacy assessments were changes from baseline on the BPRS [18] total and subscale scores and the CGI [19] severity of illness scale at Months 1,3 and 6. The BPRS comprises 16 items covering general, positive and negative symptoms and is scored on a severity scale of 1-7 (higher scores indicate more severe symptoms). The CGI severity of illness scale has a seven-point scale (1 is normal and 7 is extremely ill). Secondary efficacy variables included the BPRS response rate (response was defined as a ~ 30% decrease from baseline in BPRS score) and the CGI improvement scale, which measures the change in the state of the patient's illness relative to baseline scored from 1 (very much improved) to 7 (very much worse) and was completed at each visitoIn addition, a CGI improvement of side efIects scale (a scale identical to the CGI improvement scale used for the efficacy assessment) was fulfilled at each visit for patients with intolerable adverse efIects to previous antipsychotic treatment. A CGI global improvement rating of at least "much improved" constituted a therapeutic response for both an improvement in the illness and an improvement in side efIects. Investigators also recorded if patients exhibited any of the following: aggression/ violence, agitation/excitement, insomnia, dysphoria, suicidal behaviour, comorbid substance abuse, cognitive difficulties or other problems. Categories were primarily based on those described in the Expert Consensus Guideline for the Treatment of Schizophrenia [19]. At the end of the study, patients completed a Likert-type scale (7 points) to report their satisfaction with treatment, where 1 indicated "extremely 224 J Gibert et al. Excluded N=109 Missing baseline visit 3 (0.1 %) Missing post-baseline visit 68 (3.3%) No post-basaline efficacy evaluation 38 (1.9%) Withdrawn N=120: Lost to follow-up, N=3 Adversa event, N=40 Lack of efficacy, N=22 Patient withdrew cansen!, N=42 Other, N=13 5 pis not available at Month 1 Withdrawn N=155: Lost to follow-up, N=15 Adversa event, N=46 Lack of efficacy, N=22 Palient withdrew consen!, N=52 Other, N=20 5 PIs not available at Month 1 Withdrawn N=174: Lost to follow-up, N=45 Adversa event, N= 18 Lack of efficacy, N=43 Patient withdrew cansent, N=40 Other, N=28 Withdrawn N=175: Lost to follow-up, N=46 Adverse event, N=18 Lack of efficacy, N--43 Palient withdrew cansent, N=40 Other, N=28 Withdrawn N=l42: Lost to follow-up, N=48 Adversa event, N=5 Lack of efflcacy, N=46 Patient withdrew cansent, N=27 Other, N=16 Withdrawn N=142: Lost to follow-up, N=48 Adversa event, N=5 Lack of efflcacy, N=46 Palient withdrew consent, N=27 Other, N=16 5 pis not available at Month 3 but available al Month 6 5 PIs not available at Month 3 but avaiJable at Month 6 Figure l. Patient disposition. satisfied", 4 "neither satisfied nor dissatisfied", and 7 "extremely dissatisfied" with treatment. Tolerability assessments Medical and psychiatric histories were taken at base1ine and a physical examination was carried out. Vital signs and weight were measured by the investigators using their normal clinical practice methods (non-standardised procedures) at baseline and at Months 1,3 and 6. Body mass index (BMI) was calculated. Weight gain was defmed as any increase in weight (kg) from baseline. The Simpson-Angus Scale (SAS), a 10-point scale measuring EPS using standardised clinical assessment of parkinsonian side-effects including tremor, rigidity and salivation, was completed at baseline and at each follow-up visito Treatment-emergent adverse events, including sedation and cognitive side effects, were identified by patient report through open questioning ("Have you had any discomfort that could be related to the antipsychotic medication?") and categorised by the investigator using a modified Udvalg for K1iniskeUndersogelser [UKU] Side Effect Rating Scale [20], a 23-point scale covering cognitive, whole body abnormalities and sexual dysfunction, at baseline and each follow-up visitoThe investigators used their own judgement to determine the probability of a causal relationship of an adverse event to the medication as "absent", "possible", or "probable". In addition, adverse events classed as "intolerable", such as tardive dyskinesia, were evaluated by the clinician using the CGI improvement of side effects scale as described above. Statistical analysis Al1efficacy analyses were performed on the intention to treat (fIT) sample using a last observation carried forward approach (LOCF). The ITT population comprised patients who met the study entry criteria, were prescribed quetiapine and had at least one postbaseline efficacy evaluation. For inclusion of patient data in the ITT sample for multi-item scales such as the BPRS, data for at least 80% of all items was required. Logistic regression analysis was carried out to determine the relationship between patients' clinical characteristics, dose of quetiapine at endpoint (categorised as low, <400; intermediate, ~400 to <600; and high, ~600 mglday) and response rateo A Student's paired t-test was used to analyse changes from baseline to endpoint in the BPRS total and clusters scores. AlI statistical tests were two-tailed with a P value of ~0.05 considered statistically significant [21]. Descriptive statistics were used to summarise modified UKU and SAS total score data for the tolerability population (all patients who entered the study and had at least one follow-up visit). Any increase from the baseline severity score in a UKU item rated as possibly or probably related to n more than one medication. Naturalistic study of quetiapine for schizophrenia quetiapine was considered to be an adverse event; missing evaluations of causal relationship were also considered possibly related to quetiapine. For the SAS, changes from baseline scores were assigned to three categories: "improved" (change <O), "no change" (change =0) and "worsened" (change >0), and the frequency distributions of the categories calculated for each study visito A logistic regression analysis was performed to investigate the relationship between patients' clínical characteristics and quetiapine dose with regard to response rateo The following risk factors were included: age, sex, first episode, breakthrough episode, inadequate response to previous treatment, intolerable adverse reactions to previous treatment, inadequate response and intolerable adverse reactions to previous treatment, treatment -refractory schizophrenia, dose (categorised as "low", "intermediate" and "high"; see definitions above), prominent depression (BPRS item #9 score > 4), prominent hostility symptoms (BPRS item #10 score >4), prominent anxiety symptoms (BPRS item #2 score >4), diagnostic complications as defined above were introduced as dichotomous variables, and finally the total BPRS score at baseline. Results A total of 2029 patients were enrolled in the study. Demographic and clínical characteristics are pre- sented in Table 1. Overall, 109 patients were excluded from the ITT analyses (n = 1920) due to missing base1ine visit (n =3), missing post-baseline visit (n = 68) or no post-baseline efficacy evaluation (n =38). At study entry, 78% of patients (n = 1575) were already receiving antipsychotic treatment: 63.3% were switched from their existing medication to quetiapine monotherapy and 14.7% received quetiapine as adjunctive therapy. Of the patients switched from previous antipsychotics to quetiapine, 9% had previously been treated with two or more antipsychotics (n = 68) and 91 % with one antipsychotic (n = 717). Of the patients treated with one previous antipsychotic (n = 717), 35.4% had received risperidone, 27.1 % olanzapine, 2.8% clozapine, 15.5% haloperidol, 0.4% other atypicals and 19.3% other conventional antipsychotics. The most common therapies in the adjunctive group were risperidone (23.2%), zuclopenthixol (15.2%), olanzapine (14.3%), fluphenazine (14.3%) and haloperidol (10.3%). Of those previously on antipsychotic medication, 42.5% had an inadequate response to previous treatment, 17.3% had experienced intolerable adverse events and 40.2% cited both reasons. Agitation (25.3%), dysphoria (25.1 %), insomnia (19.1%), cognitive deficits (18.8%) and aggression/ violence (15.3%) were among the most common diagnostic complications (occurring in ~ 15% patients) reported by the investigators. Table I. Patient charaeteristics. 2.9 7.4 13 7.4 14.7 0.2 0.5 4.7 24.1 58.8 %7.1 0.4 0.3 0.9 28.1 33.2 2.3 11.8 1.0 1.1 1.4 11.9 8.0 6.1 10.4 6.9 61.1 8.7 16.2 Lithium Anticonvulsants Co-morbidities 0.6 0.8 31.2 21.8 38.4 2.4 12.2 59.7 66.8 Mean SD 55.7 12.6 Mean (SD) weight, kg 1239 Nsystem 10 1194 488 177 95 4 59 328 A1zheimers Parkinsons Other Cardiovascuiar Diabetes Other 1212 1354 163 571 442 283 240 247 780 633 124 21 19 9 abnormalities 241 673 150 46 47 22 48 29 16 12 6 12.4 N76.1 141 Cataraets 1130 151 264 36.8 24.1 Nerwus abnormalities Antidepressants Hypertension Anxiolyticslhypnotics Ischemic Obesity cardiopathy Optical Antiparkinsonians abnormalities 225 Endocrinelnutritionallmerabolic Cardiac insufficiency Other circulatory abnormalities abnormalities Existing medication' 226 J Gibert et al. Overall, 541 (26.7%) patients withdrew from the study: 162 (8%) were lost to follow-up, 131 (6.5%) withdrew consent; 112 (5.5%) withdrew due to lack of efficacy; 69 (3.4%) withdrew due to adverse events and 67 (3.3%) withdrew for other reasons. At study endpoint, the mean (SD) dose of quetiapine was 449 (206.7) mg/day; 24.7% of patients received <400 mg/day quetiapine; 44.3% received ~400 to <600 mg/day and 31.0% received ~ 600 mg/day. ·6.0 • Monlh1 • Monlh3 O Monlh6 ·5.0 .s j~ ~ li - CI) ~ f~ ·4.0 ·3.0 -1.0 ~ -1.2.0 .É 0.0 Bfficacy Quetiapine significantly improved symptoms from Month 1 through to study end. Mean BPRS total score decreased significantly by 17.3 points from base1ineto endpoint (from 51.5 to 34.2, respective1y; p <0.001; Figure 2). Significant decreases from baseline to Month 6 in BPRS positive, negative and Factors I-V (thought disturbance, anergia, anxiety, activation, hostility/suspicion) subscale scores were also observed (P <0.001; Figure 3). After 1 month of quetiapine treatment, 22.6% (95% CI: 20.7, 24.4) of patients were c1assified as responders (~30% decrease from baseline in BPRS) with this proportion increasing to 46.1 % at Month 3 (95% CI: 43.8, 48.3), and to 55.8% (95% CI; 53.5, 58.0) at Month 6, the study endpoint. Similarly, the CGI global improvement response rates increased over the 6-month study period with responses of 22.3% (95% CI: 20.4, 24.2), 44.1% (95% CI; 41.8, 46.3) and 52.8% (95% CI: 50.6, 55.0) at Months 1, 3 and 6, respectively. Both the BPRS and CGI response rates were statistically significant at Months 3 and 6 (P <0.001) versus Month 1. The factors that were the strongest predictors of a lower response to treatment (< 30% decrease from baseline in BPRS) were the use of quetiapine at a dose of <400 mg/day compared with ~600 mg/day (odds ratio [OR] 0.62; 95% CI; 0.48, 0.82), and initiating treatment because of an inadequate response to previous therapy (OR: 0.63; 95% CI: 0.51, 0.79). A substantial number of patients initiated treatment with quetiapine monotherapy because of -20.0 Month 1 "'1><0.001 vs b8se1ine Figure 2. Mean change froro baseline in BPRS total score. "·p<O.OOl F~Qr$I·V Figure 3. Mean change froro baseline in BPRS positive, negative and Factor 1-V subscale scores. intolerable adverse events with their previous antipsychotic treatment. By Month 6, a CGI improvement of side efIects score of:::;;2 ("much" or "very much" improved) was achieved for more than 80% of those who switched due to experiencing EPS with their previous antipsychotic therapy, for 78.2% of patients who switched due to sexual side efIects and for two-thirds of patients who experienced adverse cognitive efIects with their previous medication (Table 11). For patients who experienced tardive dyskinesia during previous treatment, more than half (58%) of the 24 evaluable patients were "much" or "very much" improved after 6 months of quetiapine treatment. At the end of the study, two-thirds of the lIT population were satisfied with their treatment, with 41.6% of these noting that they were either "very satisfied" or "extreme1y satisfied". Of the 1485 patients who completed the 6-month study, 86.9% were satisfied with treatment, of whom 55.3% were "very satisfied" or "extremely satisfied". 1blerability The tolerability population comprised 1961 patients (who had at least one follow-up visit). In general, quetiapine was well tolerated. Patients experienced considerable improvements in EPS on quetiapine treatment, as indicated by reductions in SAS scores. Mean SAS scores decreased significantly from a baseline score of6.8 (95% CI: 6.5, 7.1) to 2.8 (95% CI: 2.6, 3.0) at Month 6 (P <0.05). Weight changes that occurred durlng the study were mínimal, with a mean weight increase from baseline to Month 6 of 0.4 kg. Analysis according to baseline BMI showed that patients with a normal BMI (Le. 20-25 kglm2) had only a modest mean increase of 1.6 kg while those who were clinically obese or severely obese (BMI categories ~30 kglm2) had mean weight reductions at the end of the study , Naturalistic study 01quetiapine lor schizophrenia Table II. Improvement of adverse reactions according to CGI-I at Month 6 in patients with intolerable antipsycbotic treatment. No. of patients AlI" N=1575 Side effect EPS Tardive dyskinesia Sexual dysfunctions Weight gain Sedation Cognitive Anticholinergic Cardiovascular Other reproductive Other side effeets effects CGI score:S;2 (%) "Much" Switcbedb N = 1276 323 24 270 88 325 79 281 94 79 91 40 74 AlI" N=1575 7 33 5 26 27 26 23 (Figure 4). A weight change of>7% was reported for 10.1% ofpatients (n= 1718). The most common side effects (~5% patients) according to the UKU Side Effects Rating Scale were somnolence (25.7%; 95% CI: 23.8, 27.6), asthenia/ lassitude (12.0%; 95% CI: 10.6, 13.4) constipation (9.3%; 95% CI: 8.0, 10.6), orthostatic hypotension (7.8%; 95% CI: 6.6, 9.0), dry mouth (6.9%; 95% CI: 5.8, 8.0), difficulty concentrating (5.6; 95% CI: 4.6,6.6) and reduced libido (5.0%; 95% CI: 5.0,6.0) (Table lII). Spontaneously reported adverse events followed a similar pattern with the only event reported in over 5% of patients being somnolence (18.5%). There was a reduction in the severity of somnolence in 52.0 and 69.7% of the patients presenting with this adverse event at the first evaluation (n = 300) after 3 and 6 months, respectively, indicating that tolerance to somnolence developed during the study. The effect of concomitant psychotropic medications on somnolence and sexual function was analysed. Treatment with antipsychotics other than quetiapine was associated with a significant increase in orgastic dysfunction (P=0.035) (this analysis could not be conducted for individual antipsychotic agents due to small numbers). In addition, treatment with anxiolyticsJhypnotics was associated with a significant decrease in the incidence 3.0 2.5 BMI(kglm'l 2.0 .! .<20 1.5 J- 1.0 .20-25 .2S-30 ~30-40 0>40 Adverse event 3 ·1.0 ·1.5 Somnolence Asthenia/lassitude -3.5 Month3 Figure 4. Mean change in weight according baseline. Monlh 6 to BMI category at 81.6 57.9 78.2 54.0 67.1 66.2 54.5 40.0 69.2 63.6 (Le. switching or combination). Table III. Adverse events recorded for ;?:5% of patients using the UKU Side Effeets Rating Scale .o.S 0.0 ·3.0 N = 1276 In a naturalistic clínical setting, patients treated with quetiapine experienced significant reductions in symptoms of schizophrenia and improvements in overall functioning. Over the 6-month treatment period, quetiapine was well tolerated with over half le s.f. -2.5 improved Discussion 0.5 -2.0 Switcbedb to previous ofsomnolence (P=0.006) and a significant increase in reports ofincreased sexual desire (P=0.015). Five deaths occurred during the study and none of these were considered related to the study drug in the investigator's opínion. Deaths were due to suicide (n = 2) and myocardial infarction, pneumonia, and following laryngeal surgery. Using the logistic regression model with "withdrawal due to any cause" as the dependent variable, the strongest predictor of withdrawal was the use of a lower dose of quetiapine (mean final dose <400 mg/day) with an OR of 3.3 (95% CI: 2.5, 4.4). Patients who received a mean final dose of ~ 400 to <600 mg/day were also slightly more likely to withdraw (OR: 1.12; 95% CI: 0.9, 1.5) than those who received a mean final dose of>600 mg/day. When crude withdrawal rates were analysed according to dosage group, the proportions were 49.7% (272/485), 18.5% (161/867) and 17% (103/607) for the <400, ~400 to <600 and ~600 mg/day quetiapine categories, respectively. j~ i- or "very much" 79.5 58.4 77.0 52.9 68.9 67.1 52.5 57.1 69.2 61.5 19 'Patients with intolerable adverse events (ITT sample) who initiated treatment with quetiapine bpatients who were switcbed to quetiapine from previous antipsychotic treatment. Monlh 1 adverse reactions 227 Constipation Orthostatic hypotension Drymouth Difficulty concentrating Reduced libido No. (%) of patients (n =1961) 503 235 182 153 136 110 (25.7) (12.0) (9.3) (7.8) (6.9) (5.6) 98 (5.0) 95% CI 23.8,27.6 10.6, 13.4 8.0, 10.6 6.6,9.0 5.8,8.0 4.6,6.6 5.0,6.0 I 228 J. Gibert et al. the population of patients who completed the study indicating a high degree of satisfaction (very or extremely satisfied) with the treatment that they received. The response rate of 55.8% obtained in the study was very similar to that reported in trials by Arvanitis et al (51% with 300 mg/day quetiapine) and Small et al (50% with <250 mg/day quetiapine and 53% with ~250 to :::;;750mg/day quetiapine) using the same criteria of a ~ 30% reduction in BPRS total score [11,13]. However, it is important to note that a substantial proportion of patients in our study ínitiated treatment with quetiapine solely because of the presence of intolerable side efIects with previous antipsychotics. In addition, our study included a greater proportion of patients with comorbidities and, possibly, more treatment-refractory patients. Over 30% of patients who were classed as nonresponders (:::;;30%decrease in BPRS score) reported that they were satisfied with treatment. This suggests that it is important to use both objective and subjective measures to assess patient response to treatment in a naturalistic clínical setting. A low withdrawal rate of26. 7% was observed. This is comparable with the 26.5% withdrawal rate encountered with olanzapine in the same setting with a 6-month study duration [22] but substantially lower than that observed elsewhere [23]. A Cochrane Group review of 42 trials of quetiapine reported dropout rates of 36-64%, despite the fact that most ofthe trials were short term [23]. More recently, the CATIE [24] (Clínical Antipsychotic Trials of Intervention EfIectiveness) study, reported that 74% of all patients discontinued antipsychotic treatment before 18 months; in the quetiapine group 82% of patients discontinued treatment (the mean modal dose of quetiapine was 543.4 mg/day). Although a naturalistic design, CATIE was a double-blind clínical trial and the design and duration of follow-up may have impacted on withdrawal rates. In addition, it was relatively easy for patients to switch treatments. One can only speculate about the reasons for this difIerence in withdrawal rates, since many factors afIect treatment compliance. Logistic regression analysis undertaken in the present study indicated that dropouts were associated with the use of quetiapine at doses of <400 mg/day when controlling the baseline difIerences between dose groups. This is an important finding because, in clínical practice, withdrawal from treatment could be considered to be a measure of treatment efIectiveness. In clinical practice, addressing adverse events that may occur with antipsychotics is problematic. Several therapeutic strategies are available for the management of adverse events, including adjusting the dose, using an antidote (e.g., antiparkinsonian medications, amantadine, nizatidine, bromocriptine) or switching to another antipsychotic [1,25]. The actual management of antipsychotic-induced side efIects in the clinical setting is therefore an interesting matter to investigate. In the present naturalistic study, we found that more than two-thirds of patients who experienced EPS, sexual side efIects, other hormonal efIects (e.g., amenorrhoea), cognitive side efIects or sedation with their previous antipsychotic medication experienced considerable improvement in these adverse events when switched to quetiapine. Significant improvements in EPS have also been reported in a previous naturalistic study in patients with schizophrenia switched from other antipsychotics to quetiapine [26]. Significant increases in weight have been reported for clozapine and olanzapine [27,28]. While there was a slight increase in mean patient weight of 0.4 kg during the 6-month treatment period in our study, this is a mínimal change compared with that observed with clozapine and olanzapine [29,30]. In addition, in patients who were switched to quetiapine from other antipsychotics, weight gain was improved or very much improved in over 50% of patients. Furthermore, when weight gain was stratified by baseline BMI, most weight gain occurred in those who were underweight, whereas there was an overall reduction in weight in patients who were clínically obese. Greater increases in weight were reported in a retrospective analysis of weight change in 661 patients treated with quetiapine for at least 26 weeks (mean duration of treatment = 17.8 months) [31]; the mean weight change in this analysis was 2.3 kg. As observed in the present study, the greatest weight gain occurred in patients who were underweight at base1ine, with no significant change in those who were clínically obese. The prevalence of obesity in the studied population would afIect the mean weight change and could be an indirect reflection of selective prescribing pattems in the observational studies. The incidences of somnolence and asthenia in our study were slightly higher than previously reported in placebo-eontrolled trials [13,15,32], although a recent pooled analysis of eight placebo-controlled studies reported a similar rate of somnolence (25.6%) [33]. No concomitant psychotropic medication appeared to be significantly associated with an increase in somnolence but, as this was an uncontrolled trial, a clear relationship to quetiapine treatment could not be ascertained. Reduced libido was the only common (~5% of patients) sexual side efIect reported, despite the use of concomitant drugs that are associated with a significant increase in sexual desire (anxiolytics) and orgastic dysfunction (other antipsychotics). A similar rate of reduced libido (3.7%) was reported in another Spanish naturalistic study in patients with schizophrenia using a difIerent evaluation method, the Psychotropic-Related Sexual Dysfunction Questionnaire [34]. Numerous naturalistic studies have been carried out with the atypical antipsychotics, and used to Naturalistic study 01quetiapine lor schizophrenia ( supplement and expand on the data from randomised controlled trials [35]. The open-Iabel QUEST study compared the efficacy and tolerability of 3 months' treatment with quetiapine or risperidone in 728 patients with psychosis [36]. The results suggest that overall, both agents have similar efficacy and tolerability, but that quetiapine has a more favourable EPS profile. Similarly, in a naturalistic study comparing risperidone and olanzapine in patients with first-episode schizophrenia, there were significant reductions in the severity of illness and patient functioning with both agents, but the incidence of EPS was significantIy higher in the risperidone group [37]. In a large, pan-European study of 8400 outpatients with schizophrenia, patients treated with risperidone were significantIy more likely to experience EPS and sexual dysfunction, while patients treated with olanzapine and elozapine had significantIy greater weight gain [38]. Naturalistic studies have a number of limitations. An open-Iabe1, uncontrolled, non-randomised design means that caution must be exercised when interpreting results and drawing conelusions. In the present study, investigators were requested to inelude consecutive patients; however, selection bias exists due to the lack of randomisation and because the study aimed to elose1yreflect elinical practice in a "real-life setting" [39]. In addition, the uncontrolled design of the study limits defmitive conclusions, as effects such as regression to the mean cannot be discounted. Conversely, an important strength of our study is that it is representative of clinical practice, or at the very least, of clinical practice in Spain (the demographics and clinical characteristics of our patients are very similar to those included in another large, naturalistic Spanish study of olanzapine [40]). In addition, by looking at patient characteristics (e.g., psychiatric comorbidity, use of concomitant medications) it appears that most patients would not have been included in clinical trials of quetiapine nor any other antipsychotic due to strict enrolment criteria. Our study not only supports the effectiveness of quetiapine (measured by scales such as the BPRS, CGI, SAS and UKU, previously reported in randomised, controlled clinical trials), but also suggests that the efficacy and tolerability results obtained with quetiapine in carefully selected trial populations remain valid in a naturalistic setting. Conclusions This study demonstrated that quetiapine was well tolerated and improved symptoms in patients during a 6-month study in a naturalistic care setting in Spain. A quetiapine dose of <400 mglday was a strong predictor of lower BPRS changes « 30% decrease) and higher withdrawals than a dose of ;::;:600mglday, suggesting that higher doses are 229 associated with increased benefits. The results from this open-Iabel, naturalistic study indicated that patients with schizophrenia who were treated with quetiapine experienced symptom improvement and reduced side effects. Key points • The main objective of the study was to assess the efficacy and tolerability of the atypical antipsychotic, quetiapine, in a large representative group of patients with schizophrenia under naturalistic conditions • Quetiapine treatment was associated with significant and progressive improvement in overall symptomatology at 1,3 and 6 months • Quetiapine < 400 mglday was a strong predictor of lower BPRS changes «30% decrease) and higher withdrawals than doses ;::;:600mglday, suggesting that higher doses are associated with increased benefits • Patients experienced considerable improvements in EPS on quetiapine treatment, as indicated by reduetions in SAS scores • Weight changes that occurred during the study were minimal, with a mean weight increase from baseline at 6 months of 0.4 kg Acknowledgements We thank Jocelyn Woodcock, MPhil, from CMC, who provided editing assistance funded by AstraZeneca. Statement of interest This study was supported by AstraZeneca, Madrid, Spain. References [1] Kane JM. Management issues in schizophrenia. London: Taylor & Francis (A Martín Dunitz Book); 2000. [2] Barbato A. Schizophrenia and public health. World Health Organization, Geneva, 1998. Accessed 27 May 2005 [available at: http://www.who.intlmentaChealth!mediaJen/55.pdf). [3] National Institute of Clínical Excellence. Schizophrenia: core interventions in the treatment and management of schizophrenia in primary and secondary careo 2002. Accessed 27 May 2005 [available at: http://www.nice.org.uk]. [4] McEvoy ]P, Scheifler PL, Frances A. Expert Consensus Guideline Series. Treatment of schizophrenia. J Clin Psychiatry 1999;60(Suppl 11):8-80. [5] Friedman LM, Furberg CD, DeMets DL. Fundamentals of c1ínicaltrials. New York: Springer-Verlag; 1998. [6] Fleischhacker WW, Hummer M. Do phase In trials have c1ínicalvalue? J Clin PsychopharmacoI1999;19:391-2. [7] Norquist G, Lebowitz B, Hyman S. Expanding the frontier of treatment research. Prevention and treatment. Vol. 2, Article OOOla, 1999, posted 21 March. Accessed 27 May 2005 [available at: http://www.journals.apa.org/prevention/ volume2/pre002000 1a.htm1]. 230 ( J Gibert et al. [8) Awad AG. Quality of life of schizophrenic patients on medications and implications for new drug trials. Hosp Cornmunity Psychiatry 1992;43:262-5. [9) Stahl SM. Selecting an atypical antipsychotic by combining clinical experience with guidelines from clinical trials. J Clin Psychiatry 1999;60(Suppll0):31-41. [10) Nemeroff CB, Kinkead B, Goldstein J. Quetiapine: preclinical studies, pharmacokinetics, drug interactions, and dosing. J Clin Psychiatry 2002;63(SuppI13):5-1 1. [11) Arvanitis LA, Miller BG, and the Seroquel Trial 13 study group. Multiple fixed doses of "Seroquel" (quetiapine) in patients with acute exacerbation of schizophrenia: a comparison with haloperidol and placebo. Biol Psychiatry 1997;42:233-46. [12) Borison RL, Arvanitis LA, Miller BG, and the US Seroquel Study Group. ICI 204,636, an atypical antipsychotic: efficacy and safety in a multicenter, placebo-controlled trial in patients with schizophrenia. J Clin Psychopharmacol 1996;16:158-69. [l3) Small JG, Hirsch SR, Arvanitis LA, Miller BG, Link CGG, and the Seroquel Study Group. Quetiapine in patients with schizophrenia. A high- and low-dose double-blind comparison with placebo. Arch Gen Psychiatry 1997;54:549-57. [14) Peuskens J, Link CGG. A comparison of quetiapine and chlorpromazine in the treatment of schizophrenia. Acta Psychiatr Scand 1997;96:265-73. [15) Copolov DL, Link CGG, Kowalcyk B. A multicentre, double-blind, randomized comparison of quetiapine (ICI 204,636, "Seroquel") and haIoperidol in schizophrenia. Psychol Med 2000;30:95-105. [l6) Emsley RA, Raniwalla J, Bailey PJ, Jones AM, on behalf of the PRIZE Study Group. A comparison of the effeets of quetiapine ("Seroquel") and haloperidol in schizophrenic patients with a history of and a demonstrated, partial response to conventional antipsychotic treatment. Int Clin PsychopharmacoI2000;15:121-31. [17) World Medical Association Declaration of Helsinki: ethical principies for medical research involving human subjeets (revised 7 Oetober 2000). HIV Clin Trials 2001;2:92-5. [l8) Overall]E, Gorham DR The Brief Psychiatric Rating Scale. Psychol Rep 1962;10:799-912. [19) Guy W ECDEU. Assessment Manual for Psychopharmacology -Revised (DHEW Publ No ADM 76-338). Rockville, MD: US Department of Health, Education, and Welfare, Public Health Service, Alcohol, Drug Abuse, and Mental Health Administration, NIMH Psychopharmacology Research Branch, Division of Extramural Research Ptograms; 1976. p 218-22. [20) Lingjaerde O, Ahlfors UG, Bech P, Dencker SJ, EIgen K. The UKU side effeet rating scale. A new comprehensive rating scale for psychotropic drugs and a cross-sectional study of side effeets in neuroleptic-treated patients. Acta Psychiatr Scand 1987;334:1-100. [21) A1tman DG. Practical statistics for medical research. London: Chapman and Hall; 1991. [22) Gomez JC, Sacristan JA, Hemandez J, et al. The safety of olanzapine compared with other antipsychotic drugs: results of an observational prospective study in patients with schizophrenia (EFESO Study). Pharmacoepidemiologic Study of Olanzapine in Schizophrenia. J Clin Psychiatry 2000;61 :335-43. [23) Srisurapanont M, Disayavanish C, Taimkaew K. Quetiapine for schizophrenia. Cochrane Database ofSystematic Reviews 2000; CD000967. [24) Lieberman JA, Stroup TS, McEvoy]P, et al. Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. New Engl J Med 2005;353:1209-23. [25) Wirshing DA, Danovitch 1, Erhart SM, Pierre JM, Wirshing WC. Update on atypicals: practical tips to manage cornmon side effects. Curr Psychiatry (accessed 3 April 2004) [http:// www.currentpsychiatry.coml2003_03/0303_antipsych.asp). [26) De Nayer A, Windhager E, lrmansyah, et al. Efficacy and tolerability of quetiapine in patients with schizophrenia switched from other antipsychotics. Int J Psychiatry Clin Praet 2003;7:59-66. [27) AIlison DB, Mentore JL, Heo M, et al. Antipsychoticinduced weight gain: a comprehensive research synthesis. Am J Psychiatry 1999;156:1686-96. [28) Wirshing DA, Wirshing WC, Kysar L, et al. Novel antipsychotics: comparison of weight gain liabilities. J Clin Psychiatry 1999;60:358-63. [29) Eli Lilly and Company. Zyprexa® olanzapine tabletsIUS prescribing information. 2005 [http://pi lilIycomluslzyprexapipdf]. [30) Lamberti JS, Bellnier T, Schwarzkopf SB. Weight gain among schizophrenic patients treated with clozapine. Am J Psychiatry 1992;149:689-90. [31) Brecher M, Leong R, Osterling-Koskinen L, Jones M. Longterm weight change with quetiapine in schizophrenia: data review. Poster presented at the American Psychiatric Association Annual Meeting, Atlanta, GA, USA; 2005. Poster NR262. [32) Buckley PF. Efficacy of quetiapine for the treatment of schizophrenia: a combined analysis of three placebo-controlled trials. Curr Med Res Opin 2004;20:1357-63. [33) Goldstein J, Paulsson B, Sweitzer D, Zhong K. A review of the evidence for somnolence with quetiapine treatment. 158th Annual Meeting of the American Psychiatric Association, Atlanta, GA, USA, 21-26 May; 2005. NR 259. [34) Montejo Gonzalez AL, Rico-VilIademoros F, Tafalla M, Majadas S, Spanish Working Group for the Study of Psychotropic-related Sexual Dysfunction. A 6-month prospective observational study on the effeets of quetiapine on sexual functioning. J Clin Psychopharmacol 2005;25:533538. [35) Sebastian CS, Glazer W, Buckley PE Naturalistic studies of second generation antipsychotics in the treatment of schizophrenia. Curr Med Chem 2004;11:329-42. [36) MullenJ,Jibson MD, Sweitzer D, the QUEST StudyGroup. A comparison of the rclative safety, efficacy, and tolerability of quetiapine and risperidone in outpatients with schizophrenia and other psychotic disorders: the Quetiapine Experience with Safety and Tolerability (QUEST) study. Clin Ther 2001;23:1839-54. [37) Montes JM, Ciudad A, Gascon J, Gomez JC, EFESO Study Group. Safety, effcctiveness, and quaiity of life of olanzapine in first-episode schizophrenia: a naturalistic study. Prog Neuropsychopharmacol Biol Psychiatry 2003;27:667 -74. [38) Lambert M, Haro JM, Novick D, et al. Olanzapine vs. other antipsychotics in actual out-patient settings: six months tolerability results from the European Schizophrenia Outpatient Health Outcomes study. Acta Psychiatr Scand 2005; 111:232-43. [39) Tognoni G, Laporte ]R. From clinical trails to drug utiJization. In: Dukes MNG, editor. Drug utiJization studies. Methods and uses. WHO Regional Publications European: World Health Organization; 1993. [40) Gomez JC, Sacristan JA, Hemandez J, et al. The safety of olanzapine compared with other antipsychotic drugs: results of an observational prospective study in patients with schizophrenia (EFESO Study). Pharmacoepidemiologic Study of Olanzapine in Schizophrenia. J Clin Psychiatry 2000;61 :335 -43. Appendix SOS Group members Antonio Agüera Fernández, MD; Cristina del Álamo Jiménez, MD; Mercedes Alba Vallejo, MD; Ángeles Albarrán Barrado, MD; Jesús Alberdi Sudupe, MD; Beatriz Allue Torra, MD; Luis Almenar Naturalistic study o/ quetiapine /or schizophrenia ( Gay, MD; Yolanda Alonso de Armiño Fernandez, MD; Susana Álvarez Avelló, MD; Luis Miguel Alvárez Carmona, MD; Pablo Álvarez Lobato, MD; César Antón Saiz, MD; Enrique Aragués Ortiz de Zárate, MD; Juan José Arechederra Aranzadi, MD; José Ramón Arteaga Darías, MD; Javier Aztarain Diez, MD; José Vicente Baeza Alemán, MD; Maria Montserrat Bañuelos Piñol, MD; Iván Baques Rebull, MD; Concepció Bardoleti Casas, MD; Ignacio Basurte Villamor, MD; Miguel Bautista Parraga, MD; Francisco Bellver Fradas, MD; CarIes Berche Cruz, MD; José Manuel Bertolin Guillén, MD; José Javier Blanco Blanco, MD; Joan Manuel Blanqué López, MD; Moisés Bolívar Peralvarez, MD; Antonio Bordallo Aragón, MD; Juan F. Bort Ruiz, MD; David Busse i Olivé, MD; Pedro Bustos de Abajo, MD; Carmen Busuldo Vega, MD; Arturo Cabezas Sánchez, MD; Rodrigo Cabrera Gárate, MD; Joan Cadevall Dieguez, MD; Román Calabuig Crespo, MD; Manuel Camacho Muñoz, MD; José María Cámara Teruel, MD; Manuel Camarero Candela, MD; Rossend Camon Solsona, MD; Mateo Campillo Agusti, MD; Salustiano Campos Velázquez, MD; Jose Cañete Crespillo, MD; José Carmona Calvo, MD; Carlos Carmona Palau, MD; Esther Carrasco Parrado, MD; Alfonso Casas Losada, MD; Ma Rosario Cejas Méndez, MD; Ramón Collado Bayona, MD; Jesús M. Cuesta Bascones, MD; Pastora Cuevas Muñoz, MD; Miguel Ángel Cuquerella Benavent, MD; Amando de Bernardo Barrio, MD; José de BIas Soto, MD; Consuelo de Dios Perrino, MD; Jaime de La Torre Hernández, MD; Ma Teresa de Lucas Taracena, MD; Miguel Ángel de Uña Mateos, MD; Tomás de Vicente Muñoz, MD; José Francisco Delgado González, MD; Antonio Delgado Martel, MD; Juan Carlos Diaz del Valle, MD; Marina Diaz Marsá, MD; Francisco Doce Feliz, MD; Jordi Domingo Ribas, MD; María Echeveste Portugal, MD; Nabil El Jada Qawashi, MD; Vicente Elvira Cruañes, MD; Ma Luisa Embid Martin, MD; BIas Erkizia Amiliria, MD; Manuel Espiñeira Álvarez, MD; José Vicente Estalrich Canet, MD; Jose Antonio Fernández Benítez, MD; Juan Fernández Hierro, MD; Luis Fernández Menéndez, MD; José Luis Fernández Sastre, MD; Andrés Fernández-Cuevas Vicario, MD; Isabel Florez Fernández, MD; Eva Fontova Carbonell, MD; Joan Franch Barcelo, MD; Isabel Freire Santos, MD; Ma Asunción Fresno Gonzalez, MD; Javier Frometa Pérez, MD; Antonio Galbis Olivares, MD; José Mariano Galletero López, MD; Alberto Gangoiti Gurtubay, MD; Juan Carlos García Álvarez, MD; Javier García Campayo, MD; Juan Antonio García Mellado, MD; Blanca García Montañez, MD; Gorka García Serrano, MD; Ma Jesús Garrido Machiñena, MD; Jose Gascon Barrachina, MD; Francisco Gazquez Martinez, MD; Miguel Gelabert Camarasa, MD; Susana Gil Flores, MD; Patxi Gil López, MD; Mar Gil Villa, MD; Angel 231 Gilabert Senar, MD; José Manuel Giménez García, MD; Juan Carlos Giménez Morales, MD; José Gimillo Asensio, MD; Josep M. Giralt Coll, MD; María José Gómez del Castillo, MD; Isidro Gómez Pérez, MD; Felisa Gómez Robina, MD; Esther Gómez Rodriguez, MD; Gixane González García, MD; Antonio P. González González, MD; Felix Andrés González Lorenzo, MD; Emilio González Pablos, MD; Lidia González Segura (Residente), MD; Ana Isabel González Vázquez, MD; Micaela González-Quiros Menéndez de Luarca, MD; Eva Gracia Peligero, MD; Oded Graizer Rebhuhn, MD; Basilia Guerra Amador, MD; Antonio Guerrero Romero, MD; Joaquin Guinart Prados, MD; Albert Guiral Torner, MD; Miguel ÁngelHaza Duaso, MD; Miguel Ángel Hernández Lasheras, MD; Jose Luis Hernández Fleta, MD; Cristina Hernández Herrero, MD; Miguel Hernández Viadel, MD; Jose Herrera Arroyo, MD; Oscar Herreros Rodriguez, MD; Carmen Hidalgo Moratal, MD; Raúl Guillermo Ibáñez Corrales, MD; Lorenzo Iglesias, MD; Celso Iglesias García, MD; Ma Laura Iglesias Victor, MD; Jose Manuel Jaquotot Arnaiz, MD; Fernando Jiménez Mazo, MD; Luis María Larrazabal Salazar, MD; Ma Teresa Lázaro Casasus, MD; Jon Lizárraga Sobrino, MD; Patricia Uorens Rodríguez, MD; Francisco J López Ibor Aliño, MD; Ángeles López López, MD; Juan José López Plaza, MD; Luis López Sánchez, MD; Rufino Losantos Pascual, MD; Elena Lozano San Martin, MD; José Ángel Macías Fernández, MD; Iñaki Madariaga Zamalloa, MD; Pedro Malabia Lieb, MD; Juan José Mancheño Barba, MD; José Felix Marcos Frías, MD; Enrique Marcos González, MD; José Luis Marín Morales, MD; Manel Márquez Rowe, MD; Félix Martin Herguedas, MD; Rafael Martin Muñoz, MD; Sara Martinez Barrondo, MD; Antonio Martinez Fernández, MD; José Luis Martinez Fernández, MD; Ricardo Martinez Gallardo, MD; Manuel Martinez García de Castro, MD; Miguel Martinez Roig, MD; Adolf Mas-Yebra i Reverter, MD; Teresa Mayans Vázquez, MD; Sacramento Mayoral Moyano, MD; Pedro Antonio Megia López, MD; Desiderio Mejias Verdú, MD; Roger Mercade Sales, MD; María Jesús Merino García, MD; Luis Minguez Martin, MD; Ramón Mira Sempere, MD; Juan José Molina Castillo, MD; José María Mongil San Juan, MD; Ángel Luis Montejo González, MD; Jose Luis Montero Horche, MD; José Manuel Montes Rodríguez, MD; María Isabel Montes Santana, MD; Ma Virtudes Morales Contreras, MD; Carmen Rosa Morales García, MD; José Luis Morell Ocaña, MD; Crístina Moreno Corona, MD; Pablo Luis Moreno Flores, MD; Jesús Morillas Ariño, MD; Carlos Morillo-Velarde Quintero, MD; José Antonio Muñoz Martinez, MD; J. Javier Mz. de Morentin Barajoan, MD; Maria Carmen Natividad Hernández, MD; Rafael Navarro Pichardo, MD; Sergio Ocio León, MD; José Manuel Olivares Diez, MD; Jorge Alejandro Oliver Paez,