Summary

• RISPERDAL is indicated for the treatment of irritability associated with autistic disorder, including symptoms of aggression toward others, deliberate self-injuriousness, temper tantrums, and quickly changing moods.¹
• Efficacy was established in 3 short-term trials^2-4 in children and adolescents (ages 5 to 17 years). Refer to sections (14.4) Clinical Studies and (8.4) Pediatric Use of the RISPERDAL Prescribing Information for additional information.¹
• A systematic review and meta-analysis found that based on the Aberrant Behavior Checklist (ABC) scale, short-term (≤8 weeks) and long-term (>8 weeks) use of risperidone (RIS) effectively improved overall clinical condition in children and adolescents with autism spectrum disorder (ASD). Weight gain was observed with RIS treatment, regardless of the duration of treatment.⁵
• Additional double-blind (DB), placebo (PBO)-controlled trials found RIS to be safe and effective in the treatment of children with autism/autism spectrum disorders.^6,7
• During a 2-month, DB autism trial, RIS and aripiprazole appeared to be equally effective in improving irritability and agitation, lethargy and social withdrawal, stereotypic behavior, hyperactivity and noncompliance, and inappropriate speech.⁸
• The combination of RIS with other treatments provided improvement in certain subscale scores of the Aberrant Behavior Checklist-Community (ABC-C) Rating Scale.^8-22
• The dosage of RISPERDAL in the treatment of irritability associated with autistic disorder in children and adolescents should be individualized according to response and tolerability. Refer to the RISPERDAL Prescribing Information for additional information.¹

clinical data

Meta-analysis

Mano-Sousa et al (2021)⁵ conducted a systematic review and meta-analysis evaluating the efficacy of RIS in children and adolescents with ASD using the ABC scale. The effect on weight gain and waist circumference was also evaluated.

Study Design/Methods

Studies comparing the effect of RIS treatment with the baseline values, and/or to the PBO were included in this meta-analysis.

Results

Overall, 41 studies^{2-4,7,8,23-57} were included in the meta-analysis (males, 78.28%; mean age, 8.13 years; age range, 2-17 years). The studies were grouped based on short-term (≤8 weeks) and long-term (>8 weeks) RIS treatment. RIS effectively improved overall clinical condition of the ABC scale with both short-term (P=0.000; 95% confidence interval [CI], -0.70 to -0.30; heterogeneity [I²]=0) and long-term (P=0.004; 95% CI, -1.08 to 0.20; I²=12.16) treatment. Reduction in irritability was observed with both short-term (P=0.000; 95% CI, -1.06 to -0.35; I²=84.472) and long-term (P=0.000; 95% CI, -1.89 to -0.79; I²=89.396) RIS; the reduction was greater with the long-term use of RIS. Compared to PBO, short-term use of RIS reduced irritability with larger effect (P=0.000; 95% CI, -1.58 to -0.81; I²=61.021).

Short-term RIS decreased hyperactivity, inappropriate speech, lethargy, and stereotypic behavior compared to PBO. Similar results were seen with long-term treatment. Short-term treatment effectively reduced inappropriate speech and lethargy compared with long-term treatment. For stereotypic behavior, the effect size was greatest with short-term treatment.  

Weight gain was observed with RIS regardless of the duration of treatment when compared to PBO. Additionally, weight gain was associated with increased waist circumference with long-term use.

Pivotal DB Trials

Kent et al (2013)² conducted a 6-week, randomized, DB, PBO-controlled study evaluating the safety and efficacy of 2 fixed RIS doses in 96 children and adolescents with autism. The screening phase (3 weeks) was followed by a DB phase (6 weeks). The open-label extension (OLE) continued for 6 months.³⁴

Inclusion criteria were: weight ≥20 kg; Aberrant Behavior Checklist-Irritability (ABC-I) parent rating ≥18; Clinical Global Impressions-Severity (CGI-S) ≥4 at screening and baseline; mental age >18 months; no psychotic medication ≥1 week prior to baseline (4 weeks for fluoxetine; 8 weeks for depot medications). Patients with a previous or current psychotic disorder diagnosis of pervasive developmental disorder (PDD) other than autism were excluded. The primary efficacy endpoint was the mean change in ABC-I subscale score from baseline to DB endpoint.

Two fixed dose groups were evaluated:

• Low-Dose: 0.125 mg/day (<45 kg); 0.175 mg/day (≥45 kg)
• High-Dose: 1.25 mg/day (<45 kg); 1.75 mg/day (≥45 kg)

The 26-week OLE primarily evaluated Safety; efficacy was assessed as a secondary endpoint. Patients must have completed the 6-week DB phase OR discontinued from DB phase after at least 3 weeks for reasons other than tolerability (and ≤4 weeks elapsed since discontinuation); in the investigator's judgment, the patient required RIS treatment. Patients <45 kg were initiated on 0.125 mg/day and those ≥45 kg on 0.175 mg/day X 3 days; on day 4, the dose was increased to 0.25 mg for all patients. After day 14, increase in increments of 0.25 or 0.5 mg every 2 weeks was allowed.³⁴

Results: DB Phase-Efficacy:

Of the 96 patients who entered the study, 77 completed the 6-week DB phase (80%). Compared to PBO (-3.5), mean ABC-I scores significantly improved in the high-dose group (-12.4; P<0.001) but not the low-dose group (-7.4; P=0.164). The high-dose (83%; P=0.004) but not low-dose (52%; P=0.817) group had significantly higher response rates (≥25% improvement in ABC-I scores), compared to PBO (41%). In addition, the high-dose (63%; P<0.001) but not low-dose (17%; P=0.985) group had a significantly higher proportion of patients showing “much” or “very much” improved on the Clinical Global Impressions - Improvement (CGI-I) compared to PBO (15%). Similar improvements were observed on the CGI-S and Children's Yale-Brown Obsessive Compulsive Scale (CY-BOCS) compulsion subscale scores.

Results: DB Phase-Safety:

The high-dose group (87%) experienced a greater incidence of treatment-emergent adverse events (TEAEs) compared to the low-dose (60%) or PBO (80%) groups. Sedation, somnolence and increased appetite occurred more frequently in the high- vs low-dose group. Two TEAEs resulted in study discontinuation: aggression (PBO); sedation (high-dose). While no clinically relevant changes in mean Abnormal Involuntary Movement Scale (AIMS), Barnes Akathisia Rating Scale (BARS) and Simpson-Angus Scale (SAS) scores occurred during the study, extrapyramidal symptom (EPS)-related adverse events (AEs) were more frequent in the high-dose group (n=5) and were mostly akathisia. No tardive dyskinesia (TD) was reported. Significantly greater increases in mean prolactin levels from baseline to DB endpoint were observed for high-dose (+20.23 ng/mL) vs low-dose (+2.58 ng/mL) or PBO (+1.27 ng/mL) groups. One high-dose patient reported oligomenorrhea.

Open-Label Extension:

Seventy-nine patients entered the OLE phase and 56 completed the study. Insufficient response (n=7), AEs (n=5), and lost to follow-up (n=4) were the most common reasons for discontinuation. The median mode dose was 0.875 mg/day (<45 kg group) and 1 mg/day (≥45 kg group). Fifty patients experienced AEs in the OLE phase. The most common AEs (>5%) reported were increased appetite, increased weight, vomiting, sedation, pyrexia, upper respiratory tract infection, nasopharyngitis, somnolence, and fatigue. One serious AE (hydrocele) was reported in the high-dose RIS/RIS group which was not related to RIS per the investigator. However, the patient later developed idiopathic thrombocytopenia purpura and a low platelet count of 14,000/mm. ³ The ABC-I response rate improved from DB endpoint to OLE endpoints: PBO/RIS 41% to 76%; low dose RIS/RIS 52% to 79%; high dose RIS/RIS 83% to 83%. The PBO/RIS group had a numerically higher percent (69%, 20/29) of patients in the “much” or “very much” improved category in the Clinical Global Impressions - Change (CGI-C) score compared to the high dose RIS/RIS (60%, 15/25) or the low dose RIS/RIS (58%, 14/24) groups.

Pandina et al (2007)³ conducted a subgroup analysis of an 8-week, randomized, DB, PBO-controlled trial of children with Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) diagnosis of PDD, including: autistic disorder, Asperger’s disorder, and PDD not otherwise specified. This trial was previously published by Shea et al (2004)²⁴. The analysis focused on children with autistic disorder (n=55) with a total score of ≥30 on the Childhood Autism Rating Scale (CARS). Patients were administered RIS or PBO oral solution, initiated at a dose of 0.01 mg/kg/day and increased to 0.02 mg/kg/day on day 3. Depending on response and tolerability, the dose could be increased on day 8 by 0.02 mg/kg/day in weekly increments to a maximum dose of 0.06 mg/kg/day. The mean modal RIS dose was 1.37±0.7 mg/day (range: 0.5 to 4.2 mg/day). Compared with PBO, RIS-treated patients experienced significantly greater reductions on the ABC-I subscale from week 2 (P<0.05) through week 8 (P=0.001) and endpoint (P=0.002). Somnolence was the most frequently reported AE, reported in 74% of RIS vs 7% of PBO patients.

RUPP-AN Trials, Extensions, and Subgroup Analyses

The RUPP-AN (Research Units on Pediatric Psychopharmacology Autism Network; McCracken et al [2002])⁴ conducted a DB, 8-week, randomized, PBO-controlled clinical trial of RIS in children with autistic disorder who had serious behavioral disturbances. Patients between the ages of 5-17 years with a DSM-IV diagnosis of autistic disorder with tantrums, aggression, or self-injurious behavior were enrolled. The primary outcome measures were the change from baseline to 8 weeks on the ABC-I based on the parent’s or primary caregiver’s ratings, and the rating on the CGI-I scale, as determined by the clinician evaluator. A positive treatment response was defined as at least 25% improvement in the ABC-I, and a rating of much improved or very much improved on the CGI-I. RIS was dosed by weight-based categories. Children over 20 kg were given 0.5 mg of RIS at bedtime. The dose was increased in 0.5 mg increments but was not to exceed 2.5 mg/day in children weighing 20 to 45 kg, and 3.5 mg for children weighing greater than 45 kg. In children weighing less than 20 kg, the initial dose was 0.25 mg daily.

Eighty-two boys and 19 girls were enrolled and randomly assigned to RIS (n=49) or PBO (n=52). The mean daily dose of RIS during the final week of the study was 1.8±0.7 mg. After 8 weeks, the RIS group had a significant reduction (56.9%) on the ABC-I subscale compared to the PBO group (14.1%; P<0.001). Significantly more RIS-treated children had a rating of much improved or very much improved on the CGI-I at week 8 compared to PBO (75.5% vs 11.5%; P<0.001). A positive treatment response was seen in 69% of RIS-treated patients compared to 12% of PBO-treated patients (P<0.001) and was maintained for 6 months in 68% (23 out of 34) of the RIS-treated children. Significant changes in the ABC subscales for the RIS vs PBO group included improvement in irritability, stereotypy and hyperactivity scores.

Safety

Significantly more patients in the PBO group withdrew from the trial than patients in the RIS group (35% vs 6%; P=0.001). There were no serious AEs in the RIS group, and no children were withdrawn from the study because of an AE. AEs reported significantly more in patients receiving RIS (n=49) than PBO (n=51) included fatigue (59% vs 27%), drowsiness (49% vs 12%), increased appetite-mild (49% vs 25%), constipation (29% vs 12%), drooling (27% vs 6%), increased appetite-moderate (24% vs 4%), dizziness (16% vs 4%), tremor (14% vs 2%) and tachycardia (12% vs 2%). Most AEs were mild and self-limited. The SAS and the AIMS showed no EPS in either group. Five neurologic side effects were reported: tremor, dyskinesia, akathisia, and difficulty swallowing.

Levine et al (2016)⁵⁸ conducted a subanalysis of this 8-week study (McCracken et al [2002])⁴ to determine the relationship between baseline symptom severity and the response to treatment with RIS or PBO. A significant interaction (P<0.01) was observed between baseline severity scores and treatment on the parent-rated ABC for irritability and lethargy. The greatest differences between RIS and PBO were observed in children with moderate to severe symptoms at baseline. There was no relationship between baseline severity and treatment for the clinician-rated CGI or ABC irritability scores, nor in the parent-rated ABC for hyperactivity, stereotypic behaviors, and inappropriate speech.

Additional follow-up trials (Phase II and III) were conducted by the RUPP-AN (2005)⁵⁹ in autistic patients with serious behavior disorders who showed response to RIS during a previous short-term, 8-week, DB, PBO-controlled trial (Phase I) by McCracken et al (2002)⁴.

The Phase II study was a 4-month, open label trial that assessed the long-term safety and efficacy of RIS followed by an 8-week, DB, PBO-substitution phase (Phase III) to assess the feasibility of RIS withdrawal. During the 4-month open-label phase, previously established RIS doses were adjusted per clinical response and AEs. For patients weighing 15-45 kg, the maximum allowed dose was 3.5 mg/day while patients weighing >45 kg received doses up to 4.5 mg/day. During the 8-week discontinuation phase patients were randomly assigned to continued treatment with RIS or gradual PBO substitution. For patients randomized to PBO, RIS was decreased by 25% weekly so that the patient was solely receiving PBO by week 4 through week 8.

Of the 101 patients enrolled in the short-term, 8-week trial, 63 responders (mean age 8.6 years) entered the OLE protocol. Of these, 51 patients (81%) completed the full 16-week open-label phase. Using an intent-to-treat analysis for all 63 patients, the ABC-I subscale score increased slightly but lacked clinical significance. Overall a 59% reduction in the mean ABC-I subscale score was still observed at the 16-week endpoint when compared to the baseline score obtained 6 months prior to the initiation of RIS treatment. CGI-I scores continued to be rated “much” or “very much” improved by 82.5% (n=52) of the patients. For the 32 patients who completed the discontinuation phase of the study, the relapse rate was significantly higher (P=0.01) for the PBO group (62.5%, n=10) vs those receiving continued treatment with RIS (12.5%, n=2). Relapse during the discontinuation phase compared to the pre-discontinuation baseline was defined as 2 consecutive weeks in which the parent-rated ABC-I score increased by 25% and a rating of “much worse” or “very much worse” on the CGI-I scale was reported. The median time to relapse for the PBO and RIS groups was 34 and 57 days, respectively. Due to this interim analysis, Phase III of the study was terminated and the remaining PBO patients were clinically treated. Overall side effects during Phase II included mild to moderate increased appetite, tiredness, and drowsiness.

Aman et al (2015)³⁶ reported long-term safety and tolerability data from 84 children and adolescents (out of the original 101 subjects enrolled) from the RUPP-AN trial. The mean duration of follow-up was 21.4 months. Antipsychotics were used by 56 children (RIS, n=55 and quetiapine, n=1), and RIS accounted for 96.2% of all antipsychotics used over the follow-up period. The mean dose of RIS was 2.47 mg. Excessive appetite was reported in 42.1% of children who received RIS and 20.0% of children who did not receive RIS (P=0.08). Other common AEs in children treated with RIS included being tired during the day (23.2%), urinary problems/enuresis (19.6%), excessive saliva/drooling (17.5%), and diarrhea/loose stools (15.8%). Two children treated with RIS experienced new-onset seizures while none of the children not treated with RIS experienced this AE. Compared with baseline values, significant increases in weight and body mass index were reported in children treated with RIS (P<0.0001).

Additional subanalyses of the RUPP-AN trial, including core symptom domain outcomes²⁷, changes in parent or caregiver-rated target symptoms²³, changes in adaptive behavior⁶⁰, nutritional balance³⁰, and AEs²⁸, including prolactin³¹ and weight/leptin changes²⁵, have been referenced for your convenience. Comments⁶¹ on the RUPP-AN trial have also been referenced.

Comparison With Other Antipsychotics

OTHER RELEVANT LITERATURE

DB trials (N<50)^{6,7,9-15,19-21,26,29,32,43,62-65}, a single-blind trial⁶⁶, and open-label studies (N>25)^{33,37,38,67,68} and additional studies ^{16,17,35,69-77}  pertaining to the use of RIS in children and adolescents with autistic disorder have been identified and citations have been included for your reference.

literature search

A literature search of MEDLINE^®, Embase^®, BIOSIS Previews^®, and Derwent Drug File (and/or other resources, including internal/external databases) pertaining to this topic was conducted on 21 May 2025.

DB trials (N≥50) discussing the use of RIS for autistic disorders are summarized in this response. Retrospective, pharmacoepidemiologic, and pharmacogenomic studies; case reports/case series; and review articles have not been included in this reply. Please contact the Johnson & Johnson Medical Information department at 1-800-526-7736 9AM to 5PM EST for a comprehensive literature search.