Summary

• Results from a 12-week, double-blind (DB), double-dummy trial in patients with schizophrenia comparing risperidone long-acting injection (RISPERDAL CONSTA) and oral risperidone (RIS) showed significant improvements in both groups on the Positive and Negative Syndrome Scale (PANSS) total and factor scores.¹
• A 48-week, single-blind, single-center, randomized study showed similar efficacy and tolerability between schizophrenia patients continuing on RIS and those switched to RISPERDAL CONSTA.²
• Results from a 6-month, proof-of-concept study to evaluate the differences in heavy alcohol use found a statistical trend (P=0.054) toward significance in the difference between changes in heavy drinking days (≥5 drinks/day) for the RISPERDAL CONSTA group vs RIS.³
• A randomized, open-label (OL) study enrolling 83 patients with recent onset schizophrenia compared treatment with RIS and RISPERDAL CONSTA. After 12-months of treatment, the rate of psychotic exacerbation and/or relapse was significantly lower in the RISPERDAL CONSTA group (5%) than in the RIS group (33%). In addition, the rate of psychiatric hospitalizations was also significantly lower in patients treated with RISPERDAL CONSTA than in those treated with RIS (5% vs 18.6%).⁴

CLINICAL STUDIES

Double-blind Trial

Chue et al (2005)¹ described the results of a 12-week, DB, double-dummy, multicenter, parallel-group study comparing RISPERDAL CONSTA with RIS tablets in a noninferiority analysis.

Study Design/Methods

• Adult patients with a diagnosis of schizophrenia and a PANSS score ≥50 were included.
• Patients were given RIS for up to 8 weeks during a run-in phase. Other antipsychotics were discontinued during the first 2 weeks of the run-in phase while RIS was introduced. The RIS dose was adjusted during the next 2 weeks. Patients were then continued their optimal dose of RIS 2 mg, 4 mg, or 6 mg/day during the last 4 weeks of the run-in period.
• Patients were then randomized to a 12-week, DB, double-dummy study if they were symptomatically stable at the end of the 8-week run-in phase. After run-in, patients who were on a stable dose of RIS were randomized to receive either RISPERDAL CONSTA 25 mg, 50 mg, or 75 mg every 2 weeks plus placebo tablets daily or RIS 2 mg, 4 mg, or 6 mg daily plus placebo injections every 2 weeks. Patients in the RISPERDAL CONSTA group received RIS supplementation during the first 3 weeks of the DB study. Dose level was determined by the dose of RIS given during the run-in period.
• Assessments included the PANSS Structured Clinical Interview, the Clinical Global Impressions (CGI) scale, adverse events (AE), vital signs, clinical laboratory tests, electrocardiogram (ECG), and the Extrapyramidal Symptom Rating Scale (ESRS).
• A noninferiority analysis of between-group differences in PANSS total change scores from baseline to endpoint was performed to determine whether the switch from one risperidone formulation to the other altered symptom control. A noninferiority analysis is designed to show that one group is not worse than the other.

Results

• Of the 779 patients with schizophrenia who entered the run-in period at 95 sites in Europe, North America, and Africa, 640 patients received DB treatment (n=321, RIS; n=319, RISPERDAL CONSTA).
• Baseline characteristics of the 2 groups were similar although there was a marginally significant difference between treatment groups in the number of previous hospitalizations (P=0.048).
• The average patient was approximately 40 years of age with 86% of patients previously receiving antipsychotics during the 6 months before the run-in period (44% previous depot, 60% previous risperidone).
• Mean PANSS total scores at the start of the run-in period were 75.8±0.66 and decreased to 68.9±0.66 at randomization.
• Approximately 80% of patients completed the DB phase and received the 6 scheduled injections. The Table: Most Common Reasons for Study Discontinuation outlines the discontinuations for both groups.

Efficacy

• The noninferiority analysis included patients who received 4 or more injections: RIS group (n=275); RISPERDAL CONSTA group (n=266).
• Similar improvements in PANSS total scores were seen for both groups with an average improvement of approximately 5-6 points (see Table: Efficacy Measures at Endpoint). RISPERDAL CONSTA was as efficacious as RIS in improving PANSS total scores. Significant reductions were also seen in both groups for all 5 PANSS factors (P<0.05).

• The CGI-severity scale scores improved in both treatment groups from DB baseline to endpoint.
• The percentage of patients rated as not ill or with mild illness increased from baseline levels of 46.9% in the RIS group and 49.2% in the RISPERDAL CONSTA group to 57.8% and 57.9%, respectively, at endpoint.

Safety

• During the DB phase, AEs were reported in 59.9% of patients in the RIS group vs 61.1% of patients in the RISPERDAL CONSTA group. AEs reported in ≥5% of patients in any group were anxiety, insomnia, psychosis, and headache.
• No significant between-group differences in ESRS scores or changes in scores during treatment were noted. During weeks 4-12, 6.4% and 6.8% of patients in the risperidone oral and RISPERDAL CONSTA groups reported movement disorder-related AEs, respectively. Treatment-emergent tardive dyskinesia was reported in 4 patients in the long-acting risperidone group. This transient event was considered related to the unmasking of dyskinesia associated with the reduction or discontinuation of previous antipsychotic drugs.
• Body weight changes were minimal in both groups (0.3±0.2 kg, RIS; 0.5±0.2 kg, RISPERDAL CONSTA).
• Injection-site pain was low (mean scores of 18-20 on a 100-point visual analog scale [VAS]) and was similar after receiving both risperidone and placebo injections.
• No clinically relevant drug-related abnormalities were detected in laboratory tests and ECG values.
• One death from heart failure occurred in a 55-year-old woman in the RIS group and was considered by the investigators to be unrelated to risperidone.
• During DB treatment, adverse effects potentially related to prolactin elevation were reported in 2.5% (n=8) of patients in the RIS group and 1.3% (n=4) of patients in the RISPERDAL CONSTA group. Prolactin levels were elevated at baseline in many patients in each group (245 patients in the oral group and 217 in the long-acting group) and decreased to within normal range in 29 patients in the oral group and 28 patients in the long-acting group at endpoint. Mean prolactin levels were 38.9±1.6 and 37.4±1.7 ng/ml at baseline in the RIS and RISPERDAL CONSTA groups, respectively. At endpoint, the mean levels remained almost unchanged, 38.0±1.8 ng/ml (P=0.025), in the RIS group, and decreased to 32.6±1.6 ng/ml (P<0.001) in the RISPERDAL CONSTA group (P=0.025 between groups).

Single-Blind Trial

Green et al (2015)³ conducted a proof-of-concept, randomized, 6-month study (n=95; mean age 42 years) to evaluate the differences in heavy alcohol use (≥5 drinks per day) in patients treated with RISPERDAL CONSTA vs RIS at 4 U.S. study centers.

Study Design/Methods

• Patients were recruited from community mental centers and Veterans Affairs clinics where patients presented for treatment and evaluation.
• Patients were diagnosed with schizophrenia or schizoaffective disorder and current alcohol use disorder.
• Investigators evaluating patients for overall response to treatment and AEs were unblinded while raters assessing patients for alcohol and other substance abuse, psychiatric symptoms and treatment utilization were blinded.
  • During the 2nd visit patients viewed an alcohol education videotape and were encouraged to seek support through local self-help groups and to continue psychosocial treatment at their clinic.
• Treatment was initiated with RISPERDAL CONSTA 25 mg every 2 weeks and titrated up to a target dose of 37.5 mg. Those randomized to receive RIS were titrated over a 2week period to a target dose of 4 mg/day.
  • The dose could be increased (to a maximum dose of RISPERDAL CONSTA 50 mg or RIS 6 mg) or decreased depending on tolerability and psychiatric symptom response.
• Both the number of days of heavy alcohol use and intensity of drinking (number of drinks per day) were measured. Analyses included data from weeks 5 (risperidone long-acting injection [RLAI] steady state achieved) through 23 (to avoid end of study effects on drinking behavior).

Results

• 68 patients remained in the study for 6 months and 36 patients discontinued their study medication; 8 patients switched to another antipsychotic but remained in the study.
• Heavy drinking in the RIS group worsened over time (t₆₇=2.31; P=0.024) with a statistical trend toward significance in the difference between changes in heavy drinking days in the oral and RISPERDAL CONSTA groups (t_63.5=-1.96; P=0.054).
• The estimated change in the RISPERDAL CONSTA group was -0.11 heavy drinking days/week (t₈₂=0.12; P>0.9) compared to an increase of 0.68 days/week in the OR group during the same period (t₆₇=2.31; P=0.024).
• The 2 groups differed in the mean number of drinking days per week (t₈₈=-2.14; P=0.035); however, between-group differences were not significant over time.
• There were no significant differences in total PANSS, GAF and CGI or neurological side effects scales (Simpson Angus, AIMS and BARS).
• There were no differences in the frequency of adverse effects between groups. The most common AEs included pain (11.6%), upper respiratory tract infections (5.3%), drooling (5.3%), nausea or stomach discomfort (6.3%), diarrhea (6.3%), vomiting (5.3%), and other digestive or hepatic problems (8.4%).

Bai et al (2007)⁵ investigated the efficacy and tolerability of RISPERDAL CONSTA compared to RIS in adult patients hospitalized with schizophrenia. The study also conducted a pharmacokinetic analysis to determine equivalent switching doses between the 2 groups. This was a 48-week, single-blind, randomized clinical trial conducted in a single-center in Taiwan.

Study Design/Methods

• Patients had a DSM-IV diagnosis of schizophrenia disorder, and their PANSS total scores and CGI scores were stable from the screening visit to baseline. They were required to have received RIS for ≥3 months before screening and to be in good general health.
• Patients with a history of neuroleptic malignant syndrome, disease of the central nervous system, current seizure disorder, violent behavior, or suicidal ideation (including any time within the last 6 months) were excluded from the study.
• Patients (n=50) were randomized to continue their same RIS regimen (n=25) or to receive RISPERDAL CONSTA (n=25). Patients in the RISPERDAL CONSTA group continued RIS for the first 3 weeks of therapy after which their oral dose was tapered within 3 days. Initial dosing in the RISPERDAL CONSTA group was established according to their prior stable oral dosage; those stable on ≤4 mg/day of RIS received RISPERDAL CONSTA 25 mg every 2 weeks, those receiving >4 to ≤6 mg/day of RIS received RISPERDAL CONSTA 37.5 mg every 2 weeks, and those receiving >6 mg/day of RIS received RISPERDAL CONSTA 50 mg every 2 weeks. Increases in the RISPERDAL CONSTA dose was allowed in 12.5-mg increments to a maximum of 50 mg every 2 weeks. In all groups, supplemental RIS was allowed if symptoms worsened.
• Efficacy assessments were performed at baseline and at weeks 4, 8, 12, 24, 36, and 48. Assessments included the PANSS, CGI-severity, and the Global Assessment of Functioning (GAF).
• Effects on movement were measured using Abnormal Involuntary Movement Scale (AIMS), Simpson Angus Scale (SAS), and Barnes Akathisia Scale (BAS). Additional assessments were Udvalg for Kliniske Undersogelser (UKU) Side Effect Rating Scale, the SF-36 (Medical Outcome Study Short-form Health Survey) quality-of-life measure, patient satisfaction, clinical laboratory tests, body weight, pain at injection site, and serum concentrations for plasma risperidone and 9-hydroxyrisperidone metabolites.

Results

• Of the 50 patients enrolled, 5 withdrew, all from the RISPERDAL CONSTA group.
• Five patients in the RISPERDAL CONSTA group and 10 in the risperidone oral group required dose adjustment.
• At the end of the study (week 48), no significant differences between the 2 groups were seen in changes from baseline in PANSS total score, PANSS negative and positive subscale scores, PANSS general psychopathology, CGI-severity score, GAF score, quality-of-life scores, AIMS, BAS, body weight, or any clinical laboratory tests except prolactin levels.
• Stratification of the RISPERDAL CONSTA groups by initial oral RIS dosage revealed mean improvement from baseline of 11.67±11.55 in PANSS total scores in patients receiving initial dosing with 50 mg every 2 weeks and deteriorations in PANSS total scores of 5.75±14.32 in the 37.5-mg group and 3.36±7.09 in the 25-mg group (P=0.058).
• UKU Side Effect Rating Scale (P=0.048) and SAS extrapyramidal ratings (P=0.028) were lower with RISPERDAL CONSTA than with the RIS formulation at week 48.
• Mean serum prolactin levels decreased by 17.3±36.3 ng/mL in the RISPERDAL CONSTA group and increased by 8.8±21.2 ng/mL in the risperidone oral group (P=0.046).
• The mean VAS injection site pain rating was 2.9 (range 1-10).
• Pharmacokinetic analysis showed that the RISPERDAL CONSTA group had significantly lower serum concentrations of 9-hydroxyrisperidone (P=0.003) and lower concentrations of total risperidone metabolites (P=0.081) compared to the RIS group.
• Based on the results, the authors’ dose-equivalence used in switching may require some adjustment. Results suggested that patients originally receiving daily doses of RIS <3 mg receive 25 mg of RISPERDAL CONSTA every 2 weeks, while those on >3 mg but ≤5 mg receive 37.5 mg of RISPERDAL CONSTA every 2 weeks. The higher dose of RISPERDAL CONSTA 50 mg every 2 weeks was recommended for patients previously receiving >5 mg/day of RIS.

In addition to the above summary, Bai et al (2006)⁶ also reported the 12-week results for the same patient population. The article is cited for your reference.

Open-Label Trials

Takacs et al (2019)⁷ conducted a comparative analysis of patients with schizophrenia who were newly initiated on a second generation antipsychotic (SGA) using data from the Hungarian National Health Insurance Fund (HNHIF) to evaluate the all-cause 1- and 2-year discontinuation of the newly initiated SGA.

Study Design/Methods

Data was collected from the HNHIF on patients who were initiated on a new SGA monotherapy (oral: amisulpride, aripiprazole, clozapine, olanzapine, quetiapine, risperidone, paliperidone, ziprasidone; or long acting injection: risperidone, olanzapine, paliperidone once monthly) between Jan. 1st, 2012 and Dec. 31st, 2013 and followed for 2 years up until Dec. 31st, 2015; total n=12,232 (RIS n=1614, RISPERDAL CONSTA n=869).

Relevant Results

• 1-year continuation rate: RIS 17% vs RISPERDAL CONSTA 32%.
• 2-year continuation rate: RIS 10% vs RISPERDAL CONSTA 17%.
• Median time to discontinuation: RIS-58 days vs RISPERDAL CONSTA-176 days.
• The unadjusted hazard ratio comparing discontinuation of RIS and RISPERDAL CONSTA: RISPERDAL CONSTA was significantly better than RIS (P<0.001).
• The hazard ratio adjusted for propensity score comparing discontinuation of RIS and RISPERDAL CONSTA: RISPERDAL CONSTA was significantly better than RIS (P<0.001).

Hori et al (2018)⁸ conducted a short-term, pilot study conducted in Japan comparing the effect of switching from RIS to RISPERDAL CONSTA vs the effect of continuing RIS in patients with stable schizophrenia.

Study Design/Methods

• Sixteen patients with schizophrenia who had received tablet or solution forms of risperidone monotherapy for at least 3 months were switched to RISPERDAL CONSTA.
• Patients were initiated on 25mg dose with supplemental RIS for a 3-week period with a target maintenance dose of RISPERDAL CONSTA 25mg every 2 weeks. Based on clinical judgement the dose could be increased to 50 mg. After the 3-week crossover period, oral supplementation was permitted for acute exacerbation but long-term use (>4 weeks) was not permitted.
• The patient’s symptoms were assessed with the Positive and Negative Syndrome Scale (PANSS), cognitive function was assessed using the BACS-J, and total risperidone dose was assessed by converting to approximate chlorpromazine equivalents (CPZeq). All assessments were done before and 24 weeks after the initial injection by independent raters. Patients with stable schizophrenia (n=14) on RIS were also assessed to exclude the possibility of learning effects on cognitive function results.

Results

• The change in PANSS score from baseline to study endpoint was not significantly different between the 2 groups.
• The change in BACS z-scores showed a significantly greater improvement in verbal memory for the RISPERDAL CONSTA group than the RIS group (p=0.047) but no other significant differences between groups for any other changes in scores including the total score.

Subotnik et al (2015)⁴ conducted a randomized, OL, 12-month study in 86 patients (mean age 21.5 years; 78% male; 49% Caucasian) with recent onset schizophrenia to evaluate exacerbations and relapse after treatment with RIS or RISPERDAL CONSTA.

Study Design/Methods

• Previous medication nonadherence was not used to exclude patients from this trial.
• All patients were treated with RIS monotherapy for at least 3 weeks, then randomly assigned to continue treatment with RIS (n=43) or switch to RISPERDAL CONSTA (n=40).
• The modal dosage of RISPERDAL CONSTA was 25 mg (mean 26.3 mg; range, 12.5 - 37.5 mg) every 2 weeks. The modal dosage of RIS was 2 mg/day, although several patients required higher dosages (mean 3.6 mg/day; range, 1.0 - 7.5 mg/day).
• In addition to study medication, patients were also randomized to receive either cognitive remediation (to improve cognitive functioning) or healthy-behaviors training (to improve lifestyle habits and well-being).
• Patients were assessed every 2 weeks using the Brief Psychiatric Rating Scale (BPRS) to determine if they were experiencing an exacerbation and/or relapse (primary endpoint) based upon increases in unusual thought content, hallucinations, or conceptual disorganization.
• The secondary outcome was the number of psychiatric hospitalizations.

Results

• The mean length of follow-up was not significantly different for the 2 treatment groups (308±108.3 days for RISPERDAL CONSTA and 277±123.5 days for RIS).
• Discontinuations due to inadequate response were reported in 2.5% of patients treated with RISPERDAL CONSTA and 17% of patients treated with RIS 17% (P=0.01).

Efficacy

• The rate of psychotic exacerbation and/or relapse was 5% (2 of 40 patients) in the RISPERDAL CONSTA group and 33% (14 of 43 patients) in the RIS group (P<0.001).
• The risk of exacerbation and/or relapse over time was significantly lower in the RISPERDAL CONSTA group than in the RIS group (mean time to relapse 298.5 vs 218.6 days, respectively P<0.004).
• The majority of relapses (12 of 14) in patients treated with RIS occurred within 6 months of randomization whereas the 2 relapses in patients treated with RISPERDAL CONSTA occurred 4 and 8 months after randomization.
• The percent of patients with psychiatric hospitalizations was 5% in the RISPERDAL CONSTA group and 18.6% in the RIS group (P=0.05).
• Excellent levels of adherence were observed in 95% of patients treated with RISPERDAL CONSTA and 33% of patients treated with RIS (adherence graded on a scale of 1 to 5).
• The rate of return of psychosis was similar with the 2 psychosocial treatments (cognitive remediation, 22% vs healthy-behaviors training, 17%; P=0.6). However, fewer patients treated with cognitive remediation discontinued the study due to inadequate response (0% vs 19%, P<0.001).

Safety

• There were no significant differences in score changes between the 2 groups in the following AEs (see Table: AEs in Patients Treated with RIS vs RISPERDAL CONSTA) nor in the incidence of discontinuation due to intolerable AEs.

Kim et al (2008)⁹ conducted a prospective, naturalistic, controlled, OL trial that compared medication adherence and relapse rates between patients with first-episode schizophrenia who received RIS (n=28) and those who received RISPERDAL CONSTA (n=22).

Study Design/Methods

• Select inclusion criteria included age between 17 and 60 years, diagnosis of first-episode schizophrenia or schizoaffective disorder according to the DSM-IV Structured Clinical Interview, IQ above 80, and stable (≥4 weeks) outpatient treatment with RISPERDAL CONSTA or RIS.
• Patients with evidence of organic mental disorder or mental retardation and severe drug or alcohol dependence that required inpatient treatment and/or detoxification were excluded from the study.
• Primary outcome measures included medication adherence, medication nonadherence, and relapse.

Results

• Medication adherence scores were significantly higher in the RISPERDAL CONSTA group than in the RIS group (P<0.01).
• The RISPERDAL CONSTA group had a significantly lower 1-year relapse rate than the RIS group (18% vs 50%, respectively; P<0.05); the 2-year relapse rate was also lower in the RISPERDAL CONSTA group than in the RIS group (23% vs 75%, respectively; P<0.01).
• Time to nonadherence was significantly longer in the RISPERDAL CONSTA group than in the RIS group (P=0.00).
• Medication adherence and time to nonadherence were predictors of relapse.
• There were no significant differences in ESRS scores or frequency of prolactin related AEs between groups. Tardive dyskinesia-like movement occurred in 1 patient who received RISPERDAL CONSTA.

OTHER RELEVANT LITERATURE

An additional reference that has not been summarized in the response is available below.^10,11

LITERATURE SEARCH

A literature search of MEDLINE^®, Embase^®, BIOSIS Previews^®, Derwent Drug File (and/or other resources, including internal/external databases) pertaining to this topic was conducted on 31 May 2023.