human psychopharmacology

Hum. Psychopharmacol Clin Exp 2009; 24: 574583. Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/hup.1067

Remission in schizophrenia: one-year Italian prospective study of risperidone long-acting injectable (RLAI) in patients with schizophrenia or schizoaffective disordery
`2, Vincenzo Del Curatolo3, Francesco Scapati4, Alessandro Rossi1*, Anna Bagala 5z Micaela Maria Bernareggi and Maria Grazia Giustra5x on behalf of the Risperidone Long-Acting Trial Investigators (R-LAI)
1 2

Department of Experimental Medicine, University of LAquila, Italy Mental Health Department, Palmi, Reggio Calabria, Italy 3 Mental Health Department, Barletta, Bari, Italy 4 Mental Health Department, Taranto, Italy 5 Medical Affairs, Janssen-Cilag SpA, Cologno Monzese, Milano, Italy

Objectives To evaluate the maintenance of efcacy of risperidone long-acting injectable (RLAI) in stable patients with schizophrenia or schizoaffective disorders. The prevalence of patients who met standardized remission criteria will be also evaluated as well as the predictors factors of remission according to psychopathological, psychosocial and subjective correlates. Methods 52-week, open-label prospective trial in 347 stable patients switching directly to RLAI from any previous antipsychotic treatment. Results One year of treatment was completed by 70% of patients. Positive and Negative Syndrome Scale (PANSS) total and subscale, Global Assessment of Functioning (GAF) and Clinical Global Impression (CGI) scores improved from baseline at each assessment visit ( p < 0.001, p < 0.001 and p < 0.05, respectively). Drug Attitude Inventory 30 (DAI30) scores improved signicantly from month 3 onwards. 32% of patients met sustained remission at week 52. In a logistic regression model less severe positive and negative PANSS scores at baseline predicted remission ( p < 0.001). RLAI treatment was well tolerated: one-third of patients reported mild to moderate adverse events (AEs). Eleven patients (3.2%) discontinued treatment due to an AE. No signicant weight gain ( p 0.093) was reported. Conclusions RLAI treatment up to one year improved symptoms and global functioning versus baseline, indicating that an established and accepted antipsychotic therapy can enable patients with schizophrenia to achieve and maintain remission. Copyright # 2009 John Wiley & Sons, Ltd. key words schizophrenia; antipsychotics; long-acting; remission; PANSS; DAI30

INTRODUCTION Pharmacotherapy for schizophrenia and related disorders is well established in clinical psychiatry and offers considerable benet in controlling symptoms and preventing relapse (Lehman et al., 2004). Shortterm efcacy and relapse-prevention studies have the value to establish a quantitative aspect of pharmacotherapy

* Correspondence to: Prof. A. Rossi, Department of Experimental Medicine, University of LAquila, Coppito II, 67100 LAquila (Italy). Tel: 390862-433602. Fax: 39-0862-433602. E-mail: [email protected] y The other authors state that no potential conicts do exist. z Clinical Research Manager for the Global Clinical Operation- Medical Affairs Operations of Janssen Cilag Italy x Medical Affairs Manager for the Medical Affairs Department of Janssen Cilag Italy.

related mainly to efcacy and safety (Kane et al., 2003; Honer et al., 2007). Although longitudinal studies continue to point to a large number of patients who experience improvements in their condition over time (Davidson et al., 2008), few studies have addressed the issue of remission. The goals of treatment during the stable phase are to ensure that symptoms remission or control is sustained, the patients level of functioning and quality of life are maintained or improved, and relapses are prevented and, if they occur, are effectively treated. There is growing attention to the issue of remission in psychiatry (Kupfer, 1991; Ballenger, 2001) and more recently in schizophrenia (Andreasen et al., 2005; Leucht et al., 2007) as a critical component of recovery. In a naturalistic cohort, 29% of patients with a diagnosis of schizophrenia or schizoaffective disorder in different
Received 23 March 2009 Accepted 29 July 2009

Copyright # 2009 John Wiley & Sons, Ltd.

remission in schizophrenia with risperidone lai

575

treatment settings met the criteria for remission at study endpoint. These patients showed better global functioning and insight than those who did not meet the criteria (De Hert et al., 2007). Analyses of trial data using the remission criteria indicate that remission may represent a meaningful clinical target (Lasser et al., 2005; Sethuraman et al., 2005). To further address this issue, we studied the pattern of symptoms related to remission in a population of patients with schizophrenia or schizoaffective disorder who were switched to risperidone long-acting injectable (RLAI). RLAI is the rst atypical antipsychotic to become available in long-term injectable form and has well-proven efcacy, safety and tolerability (Fleischhacker et al., 2003; Kane et al., 2003). This prospective trial followed a cohort of patients who were switched to RLAI. The primary aim was to investigate the maintenance of antipsychotic efcacy and safety of RLAI injected every 2 weeks over a period of 52 weeks. The secondary aim was to investigate prospectively the prevalence of patients who met standardized remission criteria (Andreasen et al., 2005) and the psychopathological, psychosocial and subjective predictors of achieving remission. METHODS Study design This 52-week, prospective, open-label, single-arm study was conducted at 47 sites in Italy between January 2005 and April 2007. Patients with schizophrenia or schizoaffective disorder who were receiving treatment with any antipsychotic medication and who required a long-term antipsychotic therapy were switched directly to RLAI without an oral risperidone run-in. They were not to be considered either optimally treated or symptom-free. Patients were either outpatients or living in residential structures at the time of the enrolment as well as throughout the study. The trial was performed in accordance with the guidelines of the International Conference on Harmonization for Good Clinical Practice as stipulated in the Declaration of Helsinki. Local Ethics Committees approval was obtained at each of the participating study sites.

Patient characteristics A total of 347 patients, aged 18 years or older, with schizophrenia or schizoaffective disorder according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) (APA, 1994) and
Copyright # 2009 John Wiley & Sons, Ltd.

who required long-term antipsychotic therapy were eligible for inclusion. At the time of recruitment, all patients were symptomatically stable and had been taking the same dose of antipsychotic agents for at least one month before the baseline visit. Patients were considered stable if there had been no appreciable change in symptoms over the previous month, regardless of the severity of their symptoms. Patients who had received clozapine during the previous 3 months, who had participated in an investigational drug trial in the previous 30 days, or who had previously been shown to be either intolerant or nonresponsive to risperidone therapy were excluded. Other exclusion criteria included the presence of a serious unstable medical condition, such as a history or current symptoms of tardive dyskinesia or a history of neuroleptic malignant syndrome. Pregnant or breastfeeding patients were also ineligible, and all other female patients of childbearing potential who did not use adequate contraception were excluded. All patients or their legal representatives provided their written informed consent prior to enrolment in the study. Three hundred and forty seven patients were included in the study and received at least one dose of RLAI (safety population), whereas 326 had at least one post-baseline efcacy evaluation intention-to-treat (ITT) population. Overall, 243 patients (70%) completed the 52-week study period in accordance with the study protocol (per-protocol (PP) population, i.e. treatment completers). All these patients were in charge to the local unit on the Mental Health Centre who followed up the patients within the local community program as established by the National Mental Health Care Service guidelines. The patients were contacted each month. The most common reason for discontinuation of RLAI treatment was withdrawal of consent (15.6%, Table 1). Of the 347 patients, the majority (61.9%) were male. Mean SD age was 44.2 11.4 years. On the basis of DSM-IV criteria, patients were diagnosed with schizophrenia (n 260; 74.9%) or schizoaffective disorder (n 87; 25.1%). The mean age at rst onset of symptoms was 26.9 9.4 years and mean age at rst antipsychotic treatment was 28.6 9.5 years. Patients had an average of 4.6 5.9 hospitalizations over a period of 6 months prior to inclusion in the study. The demographic characteristics of the PP population were comparable to those of the overall population. The demographic and clinical characteristics of schizophrenic patients were comparable to those of schizoaffective patients but the former had baseline higher negative symptoms (t-test: p < 0.02).
Hum. Psychopharmacol Clin Exp 2009; 24: 574583. DOI: 10.1002/hup

576
Table 1. Reasons for discontinuation of RLAI treatment N Total number of subjects Completed 52-week treatment (PP) Discontinued Reasons for discontinuation Withdrawal of consent Adverse event Patients lost to follow-up Insufcient response Non-compliance Death Other PP: Per Protocol Population. 347 243 104 54 11 11 10 10 1 7

a. rossi

ET AL.

% 100 70.0 30.0 15.6 3.2 3.2 2.9 2.9 0.3 2.0

More than the half of the 347 patients were on antipsychotic monotherapy at baseline (n 201; 57.9%). The most common antipsychotic medications were risperidone (n 71; 35.3%). The choice of previous antipsychotic reected clinical judgement only. Insofar as we are unable to verify the impact of different antipsychotic among politherapy patients on outcome measure, monotherapy patients (i.e. risperidone vs. conventional neuroleptics) were contrasted. No signicant difference between completers and discontinued patients were observed in relation to their previous antipsychotic therapy (x2 p ns). At the onset of RLAI treatment, the vast majority (97.1%) of patients received 25 mg as the initial dose, whereas the remainder received 37.5 mg. At the endpoint, 41.2% of patients were receiving the 25 mg dose, whereas 30.0% and 28.8% were receiving 37.5 mg and 50 mg, respectively. The modal dose at each visit was 25 mg. During the 52-week study period excluding the rst 3 weeks, 23.3% of patients received oral risperidone supplementation at a mean dose of 2.1 0.1 mg and for a mean duration of 46.0 5.1 days. This supplementation was given to manage a brief burst of symptoms. Trial medication Risperidone long-acting injectable (RISPERDAL1 prolonged-release powder and solvent for suspension for intramuscular injection, Cilag AG, Schaffhausen, Switzerland) was administered by intramuscular (gluteal) injection every 2 weeks, at a starting recommended dose of 25 mg, which could be higher (37.5 or 50 mg) in case of persistence of symptoms or if the patient was known to respond only to higher dosages of antipsychotics. According to the patients symptoms and response to treatment, the dosage could be adjusted or, if necessary,
Copyright # 2009 John Wiley & Sons, Ltd.

increased 4 weeks after treatment but not earlier than 2 weeks after the previous injection. Tolerability of oral risperidone should be investigated before the rst injection of RLAI in patients with no history of previous risperidone use by administering two 1 mg risperidone tablets once daily for 2 days. Patients were switched from their previous antipsychotic agent(s) directly to RLAI without an oral risperidone run in. When switching from an oral antipsychotic, the agent was administered at the same dose for 21 days after the rst injection of RLAI, and then stopped or tapered off over 3 days. If the switching was from conventional depot neuroleptics, the RLAI treatment regimen was as follows: one injection of depot neuroleptic 7 days before and 7 days after the rst RLAI dose, if the patient received medication every 2 weeks. If the interval of the depot neuroleptic was every 3 weeks, patients were injected 21 days before the rst RLAI injection, plus a nal injection of depot agent into the other buttock on the same day as the rst RLAI injection. Finally, patients received a nal depot dose 7 days before the rst RLAI injection if the interval of administration of the conventional depot was every 4 weeks. Temporary oral supplementation with risperidone (12 mg/day) was permitted when considered necessary by the investigator to treat breakthrough psychosis, and also for up to 21 days following an increase in the dose of RLAI if an immediate clinical effect was needed. Patients could receive other psychotropic medication that had been initiated before the trial for other reasons (e.g. sleep induction or sedation). Anticholinergic medication was allowed up to 8 weeks past baseline with a gradual tapering. Finally, benzodiazepines could be used as rescue medication for a short period (45.8% of patients (n 159) taking these during the 52 weeks of the study). Outcome measures Efcacy. All efcacy assessments were performed at baseline and after 4, 12, 26, 38 and 52 weeks of treatment with RLAI. Symptom severity was assessed using the Positive and Negative Syndrome Scale (PANSS) (Kay, 1991; Pancheri et al., 1995). Total, positive, negative, general psychopathology PANSS scales and PANSS cognitive cluster score were evaluated as reported by Daneluzzo et al. (2002) and by Ehmann et al. (2004). The Clinical Global Impression-Severity (CGI-S) scale (Guy, 1976) was used to assess patient condition at baseline and the CGI-Change score was used at all other time points. Changes in functioning from baseline were recorded
Hum. Psychopharmacol Clin Exp 2009; 24: 574583. DOI: 10.1002/hup

remission in schizophrenia with risperidone lai

577

using the Global Assessment of Functioning (GAF) scale (Endicott et al., 1976). Personal attitude towards RLAI treatment was rated by a subjective questionnaire, the Drug Attitude Inventory 30 (DAI30) (Rossi et al., 2001). Safety and tolerability All adverse events (AEs), including any serious AEs that occurred from the rst up to the last trial-specic procedure were recorded. Changes in weight and body mass index (BMI) were also recorded at baseline and at all trial time points. Statistical analysis All patients who received at least one dose of RLAI were included in the safety analysis (safety population). All patients who received at least one dose of RLAI and completed at least one post-baseline efcacy assessment were included in the intention-to-treat (ITT) population and analysed for the primary efcacy parameter using a last-observation-carried-forward (LOCF) approach with respect to endpoint visits. Demographic characteristics were analysed using descriptive statistics. Patients in the ITT population who completed the study according to study protocol (i.e. completers) were included in the PP population. The primary efcacy measure (i.e. PANSS total and subscales score change from baseline to endpoint) was analysed for the ITT and PP populations. Secondary efcacy measures (i.e. CGI-C, GAF and DAI30 score change from baseline to week 52, and remission data) were analysed for the PP population. Three nonparametric tests were used for comparative statistical analyses of efcacy measures: Wilcoxon RankSum test, Wilcoxon SignedRank test and KruskalWallis test. A p < 0.05 was considered statistically signicant. The PANSS scores at each visit were analysed (PP population) to determine whether patients met the criteria for remission as dened by the Remission in Schizophrenia Working Group (Andreasen et al., 2005). This denition requires the simultaneous attainment of a score of 3 (mild), 2 (minimal) or 1 (absent) for at least 6 months for all of the following symptoms (PANSS items): delusions (P1); concept disorganization (P2); hallucinatory behaviour (P3); unusual thought content (G9) or mannerisms and posturing (G5); blunted affect (N1); passive/apathetic social withdrawal (N4) and lack of spontaneity and ow of conversation (N6). For visits up to week 26, only the severity criteria were applied, but both severity
Copyright # 2009 John Wiley & Sons, Ltd.

and duration criteria were applied for subsequent visits (i.e. patients were considered to be in remission). A binary logistic regression model was used to test the power of the predictors to estimate the remission. From the predictor variables the 8 PANSS remission items were removed. RESULTS Efcacy The mean change from baseline to endpoint in the PANSS total score was reduced signicantly ( p < 0.001) at week 52 and at the endpoint. Table 2 shows values at baseline and at each time point in the two analysed populations (ITT and PP). A signicant improvement was seen as early as week 4 of treatment with RLAI, and further improvements were observed with continued treatment throughout the one-year study period (Table 2). Signicant improvements from baseline to endpoint were observed in all the subscores for the PANSS positive, negative, general psychopathology subscales and in the PANSS cognitive cluster score (Table 2). When patients who completed the 52-week study period (PP population; PANSS data for 12 patients were not available at all time points and therefore excluded from analysis) were stratied on the basis of their improvement in PANSS total score from baseline to week 52, 57.6% (n 133) had  20% improvement. Of these patients, 16% (n 37) showed  30% improvement, 6.1% (n 4) had  40% improvement and 5.6% (n 13) achieved  50% improvement. No differences in the extent of improvement in the PANSS total score from baseline to week 52 were observed among treatment completers (PP population) taking 25, 37.5 or 50 mg of RLAI as nal doses (Figure 1A), although these groups had signicantly different ( p 0.012) baseline PANSS total scores (Figure 1B). At baseline, 46.5% of patients (n 113, PP population) were considered to be moderately ill according to the CGI-S. The percentage of patients who experienced any degree of improvement according to the CGI-Change between assessment visits increased signicantly from 34.8% (V2-V1) to 65.8% (V6-V5) (x2 Test, p < 0.05). Psychosocial functioning as measured by GAF improved signicantly from baseline (49.1 11.2) to week 52 (58.9 13.3; PP population, p < 0.001). There was an overall improvement in the personal attitude towards medication during treatment with RLAI, as demonstrated by the signicant increase
Hum. Psychopharmacol Clin Exp 2009; 24: 574583. DOI: 10.1002/hup

578

a. rossi

ET AL.

Table 2. PANSS total and subscale scoresa and PANSS cognitive cluster score at each time point Baseline Subscales Positive Negative Cognitive General Total ITT PP ITT PP ITT PP ITT PP ITT PP ITT PP (n 326) (n 243) 17.9 6.3 18.0 6.1 24.4 8.0 25.2 7.8 21.2 6.9 21.6 6.7 44.9 11.9 45.2 11.8 87.2 22.5 88.4 22.1 4 weeks (n 326) (n 242) 16.9 6.4b 17.3 6.4b 23.4 8.0b 24.2 7.6b 20.3 6.8b 20.9 6.5b 42.7 12.3b 43.6 12.2b 83.0 23.3b 85.1 22.8b 12 weeks (n 298) (n 239) 15.7 6.0b 16.0 6.1b 22.5 7.8b 23.0 7.6b 19.4 6.5b 19.8 6.4b 40.6 11.7b 40.9 11.7b 78.8 22.3b 79.9 22.1b 26 weeks (n 265) (n 236) 15.5 6.5b 15.5 6.4b 22.1 7.4b 22.1 7.5b 19.2 6.7b 19.2 6.6b 40.0 12.8b 40.0 12.9b 77.6 23.8b 77.6 23.9b 38 weeks (n 248) (n 232) 14.8 5.9b 14.6 5.7b 21.1 7.5b 20.9 7.4b 18.4 6.5b 18.2 6.3b 38.2 12.1b 37.7 12.0b 74.1 22.9b 73.2 22.5b 52 weeks (n 231) (n 231) 13.7 5.7b 13.7 5.7b 19.9 7.4b 19.9 7.4b 17.4 6.5b 17.4 6.5b 36.1 12.2b 36.1 12.2b 69.6 22.9b 69.6 22.9b Endpoint (n 326) (n 242) 14.5 6.3b 13.9 6.0b 20.3 7.7b 20.1 7.3b 17.9 6.8b 17.6 6.5b 37.6 12.7b 36.6 12.4b 72.5 24.1b 70.6 23.1b

PANSS Positive and Negative Syndrome Scale. ITT Intention To Treat population. PP Per Protocol Population. a Scores are expressed as mean SD. b Wilcoxon SignedRank Test, p < 0.001, shifts versus baseline.

versus baseline in DAI30 total scores observed at week 12 and thereafter up to week 52 (baseline, n 215, 42.6 5.1; week 12, n 199, 43.7 5.1; week 52, n 180, 43.7 5.5; PP population, p < 0.05). This improvement was also observed in both Attitude towards Medication (baseline, 18.7 2.5; week 12, n 199, 19.1 2.4; week 52, 19.2 2.6; PP population, p < 0.05) and Subjective Response DAI30 scores (baseline, 23.9 3.2; week 12, n 199, 24.6 3.2; week 52, 24.5 3.5; PP population, p < 0.05). In the PP population (n 243), the PANSS severity criteria for remission, measured at each time point, were met by 35 patients (14.4%) at baseline and by 110 patients (45.3%) at week 52 (Figure 2). Additionally, 22 of 35 patients (62.9%) who met the severity criteria

at baseline also met them at week 52. The number of patients who met sustained remission (both severity and duration criteria) increased from 22 (9.1%) at week 26 to 77 (31.7%) at week 52 (Figure 2). Of the 208 patients who did not meet the severity criteria at baseline, 88 (42%) met them at week 52, and 55 (26.4%) reached and maintained sustained remission (both severity and duration) at week 52. Treatment completers (PP population) who remitted during the study had signicantly lower baseline PANSS total scores than did non-remitted patients (Wilcoxon RankSum Test, z 6.9, p < 0.001; Table 3). In the Table 3 are reported the clinical ndings of remitted patients compared with nonremitted patients that showed a worse baseline clinical

Figure 1. Improvement of symptoms when patients were stratied according to the nal RLAI dose (25, 37.5, or 50 mg). A, Change in PANSS total scores did not differ signicantly between patients taking different nal RLAI doses (KruskalWallis test; p > 0.05). B, Baseline PANSS total scores differed signicantly between patients taking different nal RLAI doses (KruskalWallis test; p 0.012)

Copyright # 2009 John Wiley & Sons, Ltd.

Hum. Psychopharmacol Clin Exp 2009; 24: 574583. DOI: 10.1002/hup

remission in schizophrenia with risperidone lai

579

Figure 2. Percentage of patients (PP population n 243) who met remission criteria at each time point. The number of patients who met severity criteria or who met both severity and duration criteria at each time point is indicated above each bar

prole. The difference in baseline CGI-S and GAF scores between groups was statistically signicant (Wilcoxon RankSum Test, p < 0.001) with more severe scores in the non-remitting population (Table 3). Baseline DAI30 scores did not differ between groups, although the change in DAI30 scores (i.e. from baseline to week 52) was signicantly ( p < 0.05) greater among remitted patients versus non-remitted patients (Table 3). In a logistic regression model, baseline PANSS positive and negative scores were the strongest predictive factors for remission, representing 13% and 14%, respectively, of the explained variance ( p < 0.001) with less severe scores predicting remission (Table 4).
Table 3. PANSS, CGI-S, GAF and total DAI30 scores at baseline and DAI30 change between baseline and week 52 in remitted patients versus non-remitted patientsa (PP population)
b Non-remitted patients (n 166)

This means that even removing PANSS remission criteria from the prediction, severity of symptoms still have a role in the remission prediction. A total of ten patients (2.9%) discontinued RLAI treatment due to insufcient response. Half of these patients were receiving the 37.5 mg dose at the time of leaving the trial, while the remainder were receiving the 50 mg dose. Safety and tolerability A total of 347 patients were treated and included in the safety analysis. Treatment with RLAI was well tolerated: 30.2% of patients experienced one or more treatment-emergent AEs during the 52-week treatment period. The majority of these events were considered to be mild (51.9%) or moderate (37.7%) in severity, and 81.0% of these events required no change in study treatment. Table 5 lists the most frequently reported treatment-emergent AEs that occurred in  5% of patients. Treatment-emergent endocrine-related AEs were reported by 18 patients (7.7%; mainly amenorrhea, galactorrhea and impotence). Only 11 patients (3.2%) discontinued treatment due to a reported treatment-emergent adverse event, with extrapyramidal symptoms (4/11) as the most frequently reported reason for discontinuation. Four per cent of patients (n 14) were hospitalized due to a psychotic exacerbation during one-year follow up. One patient died during the study due to suicide, which was considered by the investigator to be unrelated to treatment with RLAI. This patient had received 4 injections of the 25 mg dose before death. There were no clinically signicant changes in vital signs over the 52-week study period. Overall, there was no signicant increase in mean body weight from
Hum. Psychopharmacol Clin Exp 2009; 24: 574583. DOI: 10.1002/hup

Remitted patients (n 77) 74.6 20.9d 14.8 6.2d 20.2 7.4d 18.3 6.7d 2.8 1.4d 39.6 10.5d 4.3 0.9d 54.6 11.4d 43.0 5.5 (n 63) 2.70 5.3 (n 53)e

PANSS Total PANSS Positive PANSS Negative PANSS Cognitive cluster score PANSS Insight item G12 PANSS General Psychopathology CGI-S GAF DAI30 Total DAI30 changec
a b

94.7 19.6 19.5 5.5 27.5 6.9 23.1 6.2 3.5 1.5 47.7 11.5 4.7 0.8 46.5 10.2 42.5 5.0 (n 152) 0.51 4.9 (n 115)

Data are expressed as mean SD. 110 patients out of 166 at the end of 52 weeks reached severity remission criteria only, but not duration. The examination of the variables assessed for this group showed intermediated values between remitted (i.e. better prole) and non-remitted (n 56) with worse prole (ANOVA KruskalWallis analysis). c DAI30 change from baseline to week 52. d Wilcoxon RankSum Test, p < 0.001 versus non-remitted patients. e Wilcoxon RankSum Test, p < 0.05 versus non-remitted patients.

Copyright # 2009 John Wiley & Sons, Ltd.

580
Table 4. Predictive factors for remission by a logistic regression

a. rossi

ET AL.

Regression Model 95% CI for Exp (B) Predictors PANSS Positive (baseline) PANSS Negative (baseline) a Sex PANSS General Psychopathology (baseline) DAI30 (baseline) Age (years) GAF (baseline) Age onset rst symptoms (years) CGI-C (baseline) b .193 .210 .542 .023 .025 .023 .030 .010 .064 S.E. .065 .060 .382 .029 .037 .021 .027 .022 .293 Wald 8.784 12.348 2.008 .659 .463 1.209 1.251 .218 .048 df 1 1 1 1 1 1 1 1 1 Sig. .003 .000 .156 .417 .496 .271 .263 .640 .826 Exp(B) .824 .811 .582 1.024 .975 .977 1.030 1.010 .938 lower .725 .721 .275 .968 .907 .939 .978 .968 .528 upper .937 .911 1.231 1.083 1.049 1.018 1.085 1.054 1.666

Binomial Logistic Regression used to model the value of a dependent variable Remission based on its relationship to predictors. The regression model explained cumulatively 35% of the variation. Considering the diagnosis as variable of stratication in the Binomial Logistic Regression signicative differences were not observed between the two groups (schizophrenic vs. schizoaffective patients). S.E.: standard error. a Dummy variable: male gender was scored 1 Table 5. Treatment-emergent adverse events occurring in  5% of patients N Any adverse event Exacerbation of disease Endocrine-related Extrapyramidal symptoms Insomnia Cardiovascular disorders Respiratory disorders Mood disorders 105 23 18 17 14 13 13 12 % 30.2 9.9 7.7 7.3 6.0 5.6 5.6 5.1

baseline (81.3 16.1 kg) to week 52 (82.1 16.5 kg, paired t-test; p 0.093) and in mean BMI (28.7 5.5 kg/m2 vs. 28.9 5.4 kg/m2; p 0.102). A weight increase of greater than or equal to 5% was observed in 15.3%of patients. DISCUSSION The main nding of this clinical trial is that one-year treatment with RLAI is associated with sustained antipsychotic efcacy and also with a reduction in the severity of symptoms, as demonstrated by the signicant improvements in PANSS total and subscales scores, CGI-S and -C scores and psychosocial variables such as patients overall functional capacity and treatment choice acceptance as assessed by physicians in the GAF and by patients in the DAI-30 questionnaire. These results are further supported by the high completion rate (70%) and the very low (3.2%) dropout rate due to AEs. Furthermore, these improvements were obtained without treating patients with the highest available
Copyright # 2009 John Wiley & Sons, Ltd.

RLAI dose. Indeed, at the treatment endpoint, more than 70% of patients were receiving the 25 or 37.5 mg dose. At the end of our study, there were no differences in the degree of PANSS improvement among those taking 25, 37.5 or 50 mg doses, each group representing about one-third of all patients studied. Although we did not evaluate the response per dose per baseline PANSS subgroup, it could be suggested that at least one-third of patients did not need an increase in their drug regimen to obtain a satisfactory response, and only one-third of patients needed the 50 mg dose to reach a similar response. In so far as these groups differed in their baseline PANSS score, it is conceivable that patients with more severe symptoms at baseline (i.e. higher baseline total PANSS score) needed a higher drug dosage to achieve improvements similar to those attained by patients with better baseline PANSS scores. Beyond the maintenance of antipsychotic efcacy and safety, the achievement of remission is considered a more stringent outcome measure. Recently, an expert consensus denition of remission in schizophrenia was proposed along with specic operational criteria for the attainment of remission. (Andreasen et al., 2005). In our study, we report that of the 208 patients who did not meet the severity criteria for remission at baseline, 26% achieved the severity and duration criteria at the end of the 52-week study period. This is in line with Lasser et al. (2005) who showed that 21% of non-remitted patients at baseline were able to achieve symptom remission for at least 6 months after treatment with RLAI. Moreover, our data collected prospectively showed that one-third (32%) of patients who completed the 52-week study period reached and
Hum. Psychopharmacol Clin Exp 2009; 24: 574583. DOI: 10.1002/hup

remission in schizophrenia with risperidone lai

581

maintained remission over a one-year treatment period with RLAI and this is slightly different from studies by Kissling et al. (2005) and Helldin et al. (2007). An explanation for this difference may be that our sample is more severely ill which is indicated by the high rate of admissions in the last 6 months before inclusion and by higher PANSS total score at baseline. A much higher remission rate (64%) has been recently reported by Emsley et al. (2008a; 2008b) in a 2-year prospective trial in rst-episode schizophrenia patients. One reason proposed by Emsley et al. for this nding is that early psychotic subjects are known to respond more favourably to antipsychotic treatment. In our study, baseline PANSS positive and negative scores were the best predictors of remission (i.e. less severe scores predicted higher remission). This is not an unexpected nding as Docherty et al. (2007) using a similar model also showed that lower CGI severity scores predicted remission. Indeed, comparison of baseline characteristics between remitters and nonremitters revealed that patients with less severity in a broad range of clinical features at baseline, such as symptoms, health status and functionality, were more likely to reach remission. In so far as we removed PANSS remission criteria from the predictors, severity of symptoms still have a role in the remission prediction. Because factor analysis demonstrates that several symptoms domains are correlated, this is a not surprising nding. However this nding add some evidence to the a priori construct of remission of the Andreasen et al. (2005): if the 8 proposed items for Remission Criteria could be an innovative approach to facilitate research and treatment, severity of symptom as a whole have also to be taken into account. Furthermore, we report an overall improvement in the personal attitude towards medication, as demonstrated by the signicant increase in DAI30 total score with RLAI treatment from month 3 onwards. This pattern of improvement was also conrmed in both Attitude towards Medication and Subjective Response DAI30 subscale scores. DAI30 change correlated signicantly with GAF and PANSS positive changes, suggesting that robust clinical improvement can modify personal opinion about therapy and this in turn may motivate patients to continue treatment. Indeed, remitted patients had a signicantly greater change in DAI30 scores compared with non-remitters. These results are in agreement with those of Docherty et al. (2007) where the substantial improvement in patient-rated drug attitude can be viewed as supporting the validity of the remission criteria and as a relevant measure of moving toward a state of wellness.
Copyright # 2009 John Wiley & Sons, Ltd.

A further relevant nding is the good tolerability prole of RLAI. Only 3.2% of patients withdrew from the study due to an adverse event, which is in line with the 3% rate reported in a 12-month study by Kissling et al. (2005). Furthermore, in our study, one-third of patients experienced one or more adverse event during the 52-week treatment period. This compares favourably with adverse event rates of 6972% reported previously in similar trials (Kissling et al., 2005; Llorca et al., 2008). Additionally, no signicant weight gain (mean increase 0.8 kg, p 0.093) was seen after one year of RLAI treatment, although a possible explanation for the less-than-expected weight gain could be that a large number of patients had experienced most of their weight gain due to prior medication before switching to RLAI. However, the signicant mean weight gains of 0.9, 1.4 and 1.0 kg reported, respectively, by Moller et al. (2005), Kissling et al. (2005) and Llorca et al. (2008) insimilar patient populations and study designs underline the benet of our positive ndings. Lastly, movement disorders were reported as a treatmentemergent adverse event by 7.3% patients. Together these results underscore the low discontinuation rate at 12 months, which are consistent with the rates of 30% reported by Kissling et al. (2005) and Llorca et al. (2008). The good tolerability and safety prole of RLAI may lead to better compliance with therapy and reduced risk of relapse.

CONCLUSIONS Despite limitation of the open-label design of this study, we report a similar improvement in efcacy with a good safety prole to that reported in previously RCT programme involving over 2000 psychotic patients treated with RLAI (Kane et al., 2003; Chue et al., 2005; Keks et al., 2007). Our result supports that remission is an achievable condition for a signicant portion of the patient population with schizophrenia. Our results add a naturalistic perspective to previously reported data (De Hert et al., 2007). Other clinical factors leading to nonremission and likely to a more severe clinical picture should be further explored as potential therapeutic targets. For example, continuing treatment and followup beyond the one-year period could lead to increased remission and possibly other additional benets. In light of the high rate of medication discontinuation of 74% reported in the CATIE study at 18 months (Lieberman et al., 2005), our 70% completion rate after one year of treatment with a long-acting injectable drug
Hum. Psychopharmacol Clin Exp 2009; 24: 574583. DOI: 10.1002/hup

582

a. rossi

ET AL.

is encouraging. As a practical point, the clinician is well advised to consider with care whether the patient should continue or switch treatment regimens (Davis et al., 2006). As continuing treatment for up to one year produces a slight but constant improvement, careful observation of an active treatment may be a wiser approach than an urgent pressure to switch. The aim of our study was to apply the Andresean remission criteria in a prospective way. The discussion of the Andreasen criteria is beyond the scope of the present paper and in any case the scientic literature reports a favourable acceptance of the above mentioned criteria (Van Os et al., 2006) even though with some criticism (Remington and Kapur, 2005). Of course this is not the solution to all problems in this area but . . . acceptance, renement and use of these criteria should help to facilitate comparisons of effectiveness across the range of available therapeutic options . . . (from Andreasen et al., 2005).

(CZ); Dr ssa A.L. Todeschini, CSM Schio (VI); Dr R. Venanzini, DSM Fano (PS); Prof. S. Vender, Ospedale S. Macchi (VA). The study was sponsored by Janssen-Cilag Spa Medical Affairs Italy Alessandro Rossi has received grants or honoraria as a consultant from Bristol-Myers Squibb, Eli Lilly, JanssenCilag, Innova Pharma, Pzer, Sano, Shering-Plough.

REFERENCES
Andreasen NC, Carpenter WT, Jr, Kane JM, Lasser RA, Marder SR, Weinberger DR. 2005. Remission in schizophrenia: proposed criteria and rationale for consensus. Am J Psychiatr 162: 441449. APA. 1994. Diagnostic and Statistical Manual of Mental Disorders, IV. American Psychiatric Association: Washington, USA. Ballenger JC. 2001. Overview of different pharmacotherapies for attaining remission in generalized anxiety disorder. J Clin Psychiatr 62 (Suppl 19): 1119. Chue P, Eerdekens M, Augustyns I, et al. 2005. Comparative efcacy and safety of long-acting risperidone and risperidone oral tablets. European Neuropsych 15: 111117. Daneluzzo E, Arduini L, Rinaldi O, et al. 2002. PANSS factors and scores in schizophrenic and bipolar disorders during an index acute episode: a further analysis of the cognitive component. Schizophr Res 56: 129136. Davidson L, Schmutte T, Dinzeo T, Andres-Hyman R. 2008. Remission and recovery in schizophrenia: practitioner and patient perspectives. Schizophr Bull 34: 58. Davis JM, Marder SR, Tamminga CA. 2006. Switch or stay? Am J Psychiatr 163: 20322033. De Hert M, van Winkel R, Wampers M, Kane J, van Os J, Peuskens J. 2007. Remission criteria for schizophrenia: evaluation in a large naturalistic cohort. Schizophr Res 92: 6873. Docherty JP, Bossie CA, Lachaux B, et al. 2007. Patient-based and clinician-based support for the remission criteria in schizophrenia. Int Clin Psychopharmacol 22: 5155. Ehmann TS, Khanbhai I, Macewan GW, et al. 2004. Neuropsychological correlates of the PANSS Cognitive Factor. Psychopathology 37: 253258. Emsley R, Medori R, Koen L, Oosthuizen PP, Niehaus DJ, Rabinowitz J. 2008a. Long-acting injectable risperidone in the treatment of subjects with recent-onset psychosis: a preliminary study. J Clin Psychopharmacol 28: 210213. Emsley R, Oosthuizen PP, Koen L, Niehaus DJH, Medor R, Rabinowitz J. 2008b. Remission in patients with rst-episode schizophrenia receiving assured antipsychotic medication: a study with risperidone long-acting injection. Int Clin Psichoparmacol 23(6): 325331. Endicott J, Spitzer RL, Fleiss JL, Cohen J. 1976. The global assessment scale. A procedure for measuring overall severity of psychiatric disturbance. Arch Gen Psychiatr 33: 766771. Fleischhacker WW, Eerdekens M, Karcher K, et al. (2003). Treatment of schizophrenia with long acting injectable risperidone: a 12-month openlabel trial of the rst long-acting second-generation antipsychotic. J Clin Psychiatry 64: 12501257. Guy W. 1976. Clinical Global Impressions. In: ECDEU Assessment Manual for Psychopharmacology. Revised 1976. DHEW publication No. (ADM) 76338, 1976. Helldin L, Kane JM, Karilampi U, Norlander T, Archer T. 2007. Remission in prognosis of functional outcome: a new dimension in the treatment of patients with psychotic disorders. Schizophr Res 93: 160168. Honer WG, Thornton AE, Sherwood M, et al. 2007. Conceptual and methodological issues in the design of clinical trials of antipsychotics for the treatment of schizophrenia. CNS Drugs 21: 699714. Kane JM, Eerdekens M, Lindenmayer JP, Keith SJ, Lesem M, Karcher K. 2003. Long-acting injectable risperidone: efcacy and safety of the rst long-acting atypical antipsychotic. Am J Psychiatr 160: 11251132. Kay SR. 1991. Positive and Negative Syndromes in Schizophrenia:. Assessment and Research, New York: Brunner/Mazel.

ACKNOWLEDGEMENTS R-LAI investigators: Dr G. Bertolazzi, SPDC Isola della Scala (VR); Dr P. Bianco, CSM (AV); Dr G. Bif, CPS Bonate Sotto (BG); Prof. F. Bogetto, CSM ASL 01 (TO); Dr ssa D. Caporale, DSM ASL 5 Collegno Grugliasco (TO); Dr N. Capozza, CSM (KR); Dr R. Cappuccio, CSM S. Antimo (NA); Dr M. Carbone, CSM (SA); Dr A. Contu, CSM ASL 6 Sanluri (CA); Dr G. Corlito, DSM (GR); Dr A. Cucciniello, CSM ASLNA 2 Mugnano di Napoli (NA); Dr ssa A. Cumin, CSM (RI); Dr P. De Iorio, SPDC (PE); Dr M. De Vito, CSM4243 Mondragone (CE); Dr V. Del Curatolo, CSM Barletta (BA); Dr W. Di Munzio, CSM Inferiore (SA); Dr A. Erlicher, Psichiatria 1 Ospedale Niguarda (MI); Dr ssa A. Ferraro, DSM (TR); Dr B. Forti, CSM (BL); Dr ssa F.F. Foti, CSM Palmi (RC); Dr P. Galimberti, CSM ASL Roma G (RM); Dr ssa O. Gambini, SPDC S. Paolo (MI); Dr G. Giardina, CSM (TA); Dr M. Grignani, SPDC A.O. di (PG); Dr ssa O. Grillo, CSM Novi Ligure (AL); Dr G. Interlandi, DSM Caltagirone (CT); Dr ssa C. Ionio, CSM (RA); Dr L. Lalicata, CSM ` (AG); Dr V. Maffei, CSM (BA); Dr G. Malara, Canicatt SPDC Ospedali Riuniti (RC); Dr E. Marchi, DSM (LU); Dr C. Munizza, SPDC Ospedale Giovanni Bosco (TO); Dr G. Nuvoli, U.O. Salute Mentale 5 ASL 3 (GE); Dr R. Parravani, CSM ASL Roma B (RM); Dr A. Pezzoli, DSM ASL BO Sud Casalecchio di Reno (BO); Dr E. Pirfo, DSM Ospedale Amedeo di Savoia (TO); Dr ssa V. Quattrocchi, DSM Empoli (FI); Dr F. Ramaciotti, DSM Canareggio (VE); Dr E. Re Psichiatria 2 Ospedale Niguarda (MI); Dr G. Reggiani, CSM Gatteo (FC); Dr O. Rinaldi e Prof. A. Rossi, ` Operativa Psicologia, Villa SerCentro Coordinatore, Unita ` S. Angelo (PE); Dr A. Sbardella, SPDC Ospedale ena, Citta S. Filippo Neri (RM);Dr F. Selvaggio, CSM Gravina di Catania (CT); Dr S. Tagliamonte, CSM di Catanzaro Lido
Copyright # 2009 John Wiley & Sons, Ltd.

Hum. Psychopharmacol Clin Exp 2009; 24: 574583. DOI: 10.1002/hup

remission in schizophrenia with risperidone lai

Keks N, Ingham M, Khan A, Karcher K. 2007. Long-acting injectable risperidone vs. olanzapine tablets for schizophrenia or schizoaffective disorder: British. J Psychiatr 191: 131139. Kissling W, Heres S, Lloyd K, et al. 2005. Direct transition to long-acting risperidone - analysis of long-term efcacy. J Psychopharmacol 19: 1521. Kupfer DJ. 1991. Long-term treatment of depression. J Clin Psychiatr 52 (Suppl:52) 2834. Lasser RA, Bossie CA, Gharabawi GM, Kane JM. 2005. Remission in schizophrenia: results from a 1-year study of long-acting risperidone injection. Schizophr Res 77: 215227. Lehman AF, Lieberman JA, Dixon LB, et al. 2004. Practice guideline for the treatment of patients with schizophrenia, second edition. Am J Psychiatr 161: 156. Leucht S, Beitinger R, Kissling W. 2007. On the concept of remission in schizophrenia. Psychopharmacology (Berl) 194: 453461. Lieberman JA, Stroup TS, McEvoy JP, et al. 2005. Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. N Engl J Med 353: 12091223. Llorca PM, Sacchetti E, Lloyd K, Kissling W, Medori R. 2008. Long-term remission in schizophrenia and related psychoses with long-acting

583

risperidone: Results obtained in an open-label study with an observation period of 18 months. Int J Clin Pharmacol Ther 46: 1422. Moller HJ, Llorca PM, Sacchetti E, Martin SD, Medori R, Parellada E. 2005. Efcacy and safety of direct transition to risperidone long-acting injectable in patients treated with various antipsychotic therapies. Int Clin Psychopharmacol 20: 121130. Pancheri P, Brugnoli R, Carilli L, Delle Chiaie R, Marconi PL, Petrucci RM. 1995. Valutazione dimensionale della sintomatologia schizofrenica. Giorn Ital Psicopat 13: 6076. Remington G, Kapur S. 2005. Remission: whats in a name. Am J Psychiatr 162: 23932394. Rossi A, Arduini L, De Cataldo S, Stratta P. 2001. Subjective response to neuroleptic medication: a validation study of the Italian version of the drug attitude inventory (DAI). Epidemiol Psichiatr Soc 10: 107114. Sethuraman G, Taylor CC, Enerson M, Dunayevich E. 2005. A retrospective comparison of cumulative time spent in remission during treatment with olanzapine or risperidone among patients with schizophrenia. Schizophr Res 79: 337340. Van Os J, Drukker M, Campo J, Meijer J, Bak M, Delespaul P. 2006. Validation of remission criteria for schizophrenia. Am J Psychiatr 163: 20002002.

Copyright # 2009 John Wiley & Sons, Ltd.

Hum. Psychopharmacol Clin Exp 2009; 24: 574583. DOI: 10.1002/hup