www.nature.com/npjschz ARTICLE OPEN Efficacy and safety of once-monthly Risperidone ISM® in schizophrenic patients with an acute exacerbation 7✉ Christoph U. Correll1,2,3, Robert E. Litman4,5, Yuriy Filts6, Jordi Llaudó 7 , Dieter Naber8, Ferran Torres 9,10 and Javier Martínez To evaluate the efficacy and safety of Risperidone ISM® against placebo in patients with acute exacerbation of schizophrenia. A multicenter, randomized, double-blind, placebo-controlled study was conducted between June 2017 and December 2018 (NCT03160521). Eligible patients received once-monthly intramuscular injections of Risperidone ISM® (75 or 100 mg) or placebo for 12 weeks. The primary efficacy outcome was change in Positive and Negative Syndrome Scale (PANSS) total score from baseline to week 12. The key secondary efficacy outcome was change from baseline in Clinical Global Impressions-Severity of Illness scale (CGI- S) score. Altogether, 438 patients were randomized (1:1:1) and 390 included in the modified ITT efficacy set. The PANSS total score (mean difference, 95% CI) improved significantly from baseline to day 85 with Risperidone ISM® 75 and 100 mg, with placebo- adjusted differences of −13.0 (95% CI, −17.3 to −8.8); (p < 0.0001), and −13.3 (−17.6 to −8.9); (p < 0.0001), respectively. Significantly improved mean changes were also obtained for CGI-S score from baseline to day 85 for both doses of Risperidone ISM® compared with placebo −0.7 (−1.0 to −0.5); p < 0.0001, for both doses. The statistically significant improvement for both efficacy outcomes were observed as early as 8 days after first injection. The most frequently reported treatment-emergent adverse 1234567890():,; events were increased blood prolactin (7.8%), headache (7.3%), hyperprolactinemia (5%), and weight increase (4.8%). Neither new nor unexpected relevant safety information was recorded. Risperidone ISM® provided rapid and progressive reduction of symptoms in patients with acutely exacerbated schizophrenia without need of oral risperidone supplementation or loading doses. Both doses were safe and well tolerated. npj Schizophrenia (2020)6:37 ; https://doi.org/10.1038/s41537-020-00127-y INTRODUCTION RESULTS Schizophrenia remains one of the most severe mental disorders1. Patient flow and characteristics The suboptimal adherence to prescribed oral antipsychotic From 565 screened patients, 438 were randomized. One of them medication regimens is prevalent in patients with schizophrenia withdrew consent before receiving the study treatment; therefore and has been associated with relapse and worsening of long-term 437 were included in the safety population, and 390 counted for functional and mental health outcomes2–4. Long-acting formula- the modified intent-to-treat (mITT) population. The study comple- tions of antipsychotic medications (LAIs), developed to promote tion was highest in the Risperidone ISM® 75 mg group (73.8%) and treatment adherence, have helped to improve adherence and lowest in the placebo group (59.9%) (Fig. 1). thus efficacy5–8. Despite currently available atypical LAIs, some Demographic and baseline characteristics were similar among unmet clinical needs remain, e.g., the use of loading doses or oral treatment groups (Table 1 and Supplementary Table 1). The cotreatment for 2–3 weeks still is needed to achieve therapeutic patients had a mean age of 42 years; 67% were men, and 48.5% drug levels when starting treatment with a LAI9 (except for one were white. Most subjects were enrolled in the USA (61.1%), and subcutaneous version of risperidone)10. Laboratorios Farmacéuti- the mean BMI was 28.33 kg/m2. The mean (SD) PANSS score at cos ROVI, S.A., has developed a new injectable formulation of baseline was 96.4 (7.21), 96.3 (8.47), and 96.1 (8.42) for placebo, 75 risperidone using the ISM® technology. This technology is based and 100 mg of Risperidone ISM®, respectively. on an in situ forming solid polymeric matrix system that contains risperidone. The suspension obtained after reconstitution pro- duces an early, rapid, and sustained release of risperidone by 2 h Efficacy after its administration and for 1 month11. There was a statistically significant difference of both Risperidone Risperidone ISM® is a new long-acting injectable (LAI) intramus- ISM® 75 mg and 100 mg versus placebo on PANSS total score cular (IM) formulation of risperidone, for monthly administration mean change from baseline to Day 85. The placebo-adjusted without the need of using oral supplementation or loading doses, Lawrence and Hung (LH) mean change from baseline to Day 85 as it has been shown previously11,12. was −13.0 (95% confidence interval [CI]: −17.3 to −8.8; The aim of this placebo-controlled study was to evaluate the p < 0.0001) and −13.3, (95% CI: −17.6 to −8.9; p < 0.0001) for short-term efficacy and safety of Risperidone ISM® in patients with Risperidone ISM® 75 mg and 100 mg, respectively (Hommel an acute exacerbation of schizophrenia. adjusted p value < 0.0001, for both groups) (Table 2). In addition, 1 Department of Psychiatry Research, The Zucker Hillside Hospital, Glen Oaks, NY, USA. 2Department of Psychiatry and Molecular Medicine, Donald and Barbara Zucker School of Medicine at Hofstra/Northwell, Hempstead, NY, USA. 3Department of Child and Adolescent Psychiatry, Charité Universitätsmedizin Berlin, Berlin, Germany. 4CBH Health LLC, Gaithersburg, MD, USA. 5Department of Psychiatry, Georgetown University Medical School, Washington, DC, USA. 6Communal Noncommercial Enterprise of Lviv Regional Council, Lviv Regional Clinical Psychiatric Hospital, Lviv, Ukraine. 7Medical Department, Laboratorios Farmacéuticos ROVI, S.A, Madrid, Spain. 8Department of Psychiatry and Psychotherapy, Hamburg-Eppendorf University, Hamburg, Germany. 9Medical Statistics core facility, Clinical Pharmacology Department, IDIBAPS-Hospital Clinic, Barcelona, Spain. 10 Biostatistics Unit, Faculty of Medicine, Universitat Autònoma de Barcelona, Barcelona, Spain. ✉email: jmartinez@rovi.es Published in partnership with the Schizophrenia International Research Society C.U. Correll et al. 2 Assessed for eligibility (n=565) Excluded (n=127) • Adverse Event (n=2) • Withdrawal of consent (n=37) • Eligibility criteria not met (n=81) • Physician decision (n=6) • Other (n=1) Randomized (n=438) Placebo (n=147) Risperidone ISM 75 mg (n=145) Risperidone ISM 100 mg (n=146) • Received allocated intervention • Received allocated intervention • Received allocated intervention (n=147) (n=144) (n=146) • Did not receive allocated intervention (withdrawal of consent) (n=1) Completed (n=88) Completed (n=107) Completed (n=95) Did not complete (n=59): Did not complete (n=37): Did not complete (n=51): • AEs (n=3) • AEs (n=5) • AEs (n=5) • Positive pregnancy or drug test (n=3) • Withdrawal of consent (n=14) • Positive pregnancy or drug test (n=2) • Withdrawal of consent (n=20) • Lost to follow-up (n=6) • Withdrawal of consent (n=23) • Lost to follow-up (n=6) • Insufficient response (n=6) • Lost to follow-up (n=5) • Physician decision (n=1) • Other (n=6) • Non-compliance (n=1) 1234567890():,; • Non-compliance (n=1) • Hospitalization due to worsening of • Hospitalization due to worsening of schizophrenia (n=2) schizophrenia (n=3) • Insufficient response (n=9) • Insufficient response (n=15) • Other (n=4) • Other (n=7) mITT Analysis (n=132) mITT Analysis (n=129) mITT Analysis (n=129) Excluded from analysis (n=15): Excluded from analysis (n=16): Excluded from analysis (n=17): • No postbaseline PANSS (n=2) • Not dosed in DB phase (n=1) • No postbaseline PANSS (n=1) • Potential unintentional unblinding • No postbaseline PANSS (n=1) • Potential unintentional unblinding (n=13)* • Potential unintentional unblinding (n=16)* (n=14)* Fig. 1 Patient disposition. Where patients marked with “*” were included in the ITT analyses (ITT: Intention-to-Treat; mITT: modified ITT). the statistically significant improvement in PANSS total score (p < 0.0001 for both groups) (Table 2). These differences from mean change against placebo was shown as early as Day 8 for placebo were significant at each assessment point from Day 8 and Risperidone ISM® 100 mg (LS Mean difference, 95% CI: −3.9, −6,4 beyond for both Risperidone ISM® doses. to −1.5; p = 0.001), and Day 15 for Risperidone ISM® 75 mg For overall response rate at endpoint, the difference in (Fig. 2a). These significant differences remained until the end of proportions versus placebo was 39.2% (95% CI: 27.5–49.2) for the study (Day 85). Risperidone ISM® 75 mg and 33.8% (95% CI: 22.0–43.8) for For patients with baseline PANSS score ≥95, the placebo- Risperidone ISM® 100 mg (p < 0.0001 for both groups, adjusted least square mean (LSM) difference in change from Mantel–Haenzel Test) (Table 2) translating into a number-needed- baseline to Day 85 was −13.8 (95% CI: −19.5 to −8.1) and −15.6 to-treat of three for both doses. The first significant difference from (95% CI: −21.4 to −9.9) for Risperidone ISM® 75 mg and 100 mg placebo in overall response rate was at day 8 for Risperidone ISM® group, respectively (p < 0.0001 for both groups). 100 mg (difference in proportion = 11.8%; 95% CI: 3.3–20.5; p = Significant differences versus placebo in mean change from baseline were observed at each assessment time point since Day 8 0.005, Mantel–Haenzel test) and at Day 15 for Risperidone ISM® for PANSS Positive subscale, and Day 15 for PANSS Negative in 75 mg (difference in proportion = 22.0%; 95% CI: 12.1–31.5; p < both Risperidone ISM® groups, and since Day 8 for General 0.0001, Mante–Haenzel test) (Fig. 4). Psychopathology subscales in Risperidone ISM® 100 mg (Fig. 3). Similar results for the same efficacy outcomes were obtained The placebo-adjusted CGI-S score LSM difference in change when the full analysis set was evaluated (Supplementary Table 2). from baseline at endpoint was: −0.7 (95% CI: −1.0 to −0.5; p < 0.0001) for 75 mg and −0.7 (95% CI: −1.0 to −0.5; p < 0.0001), Safety and tolerability for 100 mg group (Hommel adjusted p value < 0.0001, for both groups) (Table 2). LSM change from baseline at all assessments Overall, 239 (54.7%) patients experienced at least one Treatment- since Day 8 was significantly greater with both doses of Emergent Adverse Event (TEAE) (Supplementary Table 3), most of Risperidone ISM® compared with placebo (Fig. 2b). them were mild (67.8%) or moderate (28.0%). Of the TEAEs, 169 For CGI-I score, the placebo-adjusted LSM difference at Day 85 (38.7%) were considered related with the study treatment. The was −0.8 (95% CI: −1.0 to −0.5) for Risperidone ISM® 75 mg, and most frequently reported drug-related TEAEs in ≥2% of patients −0.7 (95% CI: −1.0 to −0.4) for Risperidone ISM® 100 mg on placebo were dizziness and headache, whereas in both npj Schizophrenia (2020) 37 Published in partnership with the Schizophrenia International Research Society C.U. Correll et al. 3 Table 1. Demographic and baseline characteristics (mITT population). Baseline variable statistic/ Placebo Risperidone ISM 75 mg Risperidone ISM 100 mg All Risperidone ISM Overall category N = 132 N = 129 N = 129 N = 258 N = 390 Age (years) n 132 129 129 258 390 Mean (SD) 40.0 (11.35) 42.6 (10.63) 42.6 (11.14) 42.6 (10.87) 41.7 (11.08) Sex, n (%) Male 85 (64.4) 88 (68.2) 84 (65.1) 172 (66.7) 257 (65.9) Female 47 (35.6) 41 (31.8) 45 (34.9) 86 (33.3) 133 (34.1) Race, n (%) White 71 (53.8) 69 (53.5) 65 (50.4) 134 (51.9) 205 (52.6) Black or African American 59 (44.7) 58 (45.0) 63 (48.8) 121 (46.9) 180 (46.2) Asian 1 (0.8) 1 (0.8) 1 (0.8) 2 (0.8) 3 (0.8) Other 1 (0.8) 1 (0.8) 0 1 (0.4) 2 (0.5) Ethnicity, n (%) Hispanic or Latino 10 (7.6) 6 (4.7) 3 (2.3) 9 (3.5) 19 (4.9) Not Hispanic or Latino 122 (92.4) 123 (95.3) 126 (97.7) 249 (96.5) 371 (95.1) Country, n (%) Ukraine 56 (42.4) 57 (44.2) 57 (44.2) 114 (44.2) 170 (43.6) United States 76 (57.6) 72 (55.8) 72 (55.8) 144 (55.8) 220 (56.4) BMI (kg/m²)a n 132 129 129 258 390 Mean (SD) 28.19 (4.715) 27.77 (5.226) 28.26 (5.268) 28.02 (5.243) 28.07 (5.065) Years since Schizophrenia Diagnosis n 132 129 129 258 390 Mean (SD) 14.3 (9.74) 16.1 (10.67) 15.7 (10.43) 15.9 (10.53) 15.4 (10.29) Time since Acute Exacerbation or relapse (days) n 132 129 129 258 390 Mean (SD) 3 (1.65) 3 (1.95) 3 (3.67) 3 (2.95) 3 (2.55) a BMI (kg/m2) is calculated as BMI = 1002 × weight (kg)/[Height (cm)2]. Note: presented statistics, frequencies and the denominator used for percentages are based on patients in the modified ITT population and the treatment received. Risperidone ISM® groups (75 and 100 mg) were blood prolactin No noteworthy differences between treatment groups were increase, hyperprolactinaemia, akathisia, and headache (Table 3). documented for Columbia-Suicide Severity Rating Scale (C-SSRS). No patient died during the study. Serious TEAEs were reported Two patients (1.4%) in each treatment group were documented to in 12 patients: 5 (3.4%), 2 (1.4%), and 5 (3.4%) with placebo, 75 and have had treatment-emergent suicidal behavior or ideation. 100 mg of Risperidone ISM® groups, respectively. Only one serious Suicidal behavior or ideation worsened from baseline in two TEAE (agitation) was related to study drug in the Risperidone ISM® patients (1.4%) in the placebo group and one patient each (0.7%) 100 mg group. The incidence of patients who discontinued due to in the Risperidone ISM® 75 mg and 100 mg groups. No suicide TEAE was 7.5% with placebo, 4.2% with Risperidone 75 mg and attempts, aborted, or interrupted attempts were reported in any 6.2% in Risperidone 100 mg group. patient of any treatment group during the study. Among treatment groups, the frequency of injection site The Abnormal Involuntary Movement Scale (AIMS), Barnes reaction (ISR: redness, swelling, or induration) ranged from 6.1% Akathisia Rating Scale (BARS), and Simpson-Angus Scale (SAS) were with placebo, to 8.3% with Risperidone ISM® 75 mg and 9.6% with used to assess extrapyramidal symptoms. Regarding all three scales, treatment groups were comparable and no relevant changes from Risperidone ISM® 100 mg. Overall, the most frequently reported baseline to end of treatment were observed in any treatment group, ISR was redness (6.2%), followed by swelling (1.8%) (Table 4). i.e., mean scores (SD) for AIMS: 0.1 (1.36) for placebo, −0.1 (1.98) for Patients reported median Visual Analog Scale (VAS) scores Risperidone ISM 75 mg and −0.1 (1.22) for Risperidone ISM 100 mg; (0–10), indicating injection site pain were 2.0 in all study groups for BARS: 0.1 (0.98) for placebo, 0.0 (0.59) for Risperidone ISM 75 mg (Table 5). and 0.1 (0.87) for Risperidone ISM 100 mg, and for SAS: Mean (SD) For hematology or biochemistry parameters, there were no 0.145 (2.5) for placebo, 0.085 (0.7) or Risperidone ISM 75 mg and notable differences between treatment arms from baseline 0.048 (1.2) for Risperidone ISM 100 mg. through end of treatment and no notable changes in either Almost all patients (95.9%) took medication prior to the first treatment arm, except for prolactin. Prolactin values decreased in dose of randomized treatment. The most common pharmacolo- placebo group from baseline to end of treatment, whereas they gical groups of pre-baseline medications were antipsychotics increased in both Risperidone ISM® groups, with mean (SD) (88.3%) and anxiolytics (27.5%). The most common medications endpoint prolactin levels being 220.6 (257.4) mIU/L with placebo, were risperidone (59.3%), lorazepam (23.1%), quetiapine fumarate 875.4 (1080.7) mIU/L with Risperidone ISM® 75 mg and 904.8 (16.9%), tropicamide (16.0%), and proxymetacaine hydrochloride (810.6) mIU/L with Risperidone ISM® 100 mg. (13.3%). Medications taken prior to the first dose of the Published in partnership with the Schizophrenia International Research Society npj Schizophrenia (2020) 37 C.U. Correll et al. 4 Table 2. Primary and secondary efficacy assessments at endpoint (mITT population). Efficacy assessment Placebo N = 132 Risperidone ISM 75 mg N = 129 Risperidone ISM 100 mg N = 129 PANSS total score (mean change)1 Mean baseline score (SD) 96.4 (7.21) 96.3 (8.47) 96.1 (8.42) LS mean change (SE), 95% CIa −11.0 (1.56), −14.1 to −8.0 −24.6 (1.51), −27.5 to −21.6 −24.7 (1.54), −27.7 to −21.6 Treatment difference (SE), 95% CIb −13.0 (2.19), −17.3 to −8.8 −13.3 (2.21), −17.6 to −8.9 P valuec <0.0001 <0.0001 CGI-S total score (mean change)2 Mean baseline score (SD) 4.9 (0.52) 5.0 (0.65) 4.9 (0.48) LS mean change (SE), 95% CIa −0.6 (0.09), −0.8 to −0.4 −1.3 (0.09), −1.5 to −1.2 −1.3 (0.09), −1.5 to −1.2 Treatment difference (SE), 95% CIb −0.7 (0.13), −1.0 to −0.5 −0.7 (0.13), −1.0 to −0.5 P valuec <0.0001 <0.0001 CGI-I score3 LS means (SE), 95% CI 3.3 (0.10), 3.1–3.5 2.5 (0.10), 2.3–2.7 2.5 (0.10), 2.3–2.7 LS means difference (SE), 95% CI −0.8 (0.14), −1.0 to −0.5 −0.7 (0.14), −1.0 to −0.4 p value <0.0001 <0.0001 Overall response3 Responders, n (%) 27 (20.5) 77 (59.7) 70 (54.3) 95% CI (%) 13.9–28.3 50.7–68.2 45.3–63.1 Difference in proportions (%) 39.2 33.8 95% CI 27.5 to 49.2 22.0–43.8 P value <0.0001 <0.0001 PANSS positive subscale3 LS Means (SE), 95% CI −4.1 (0.50), −5.1 to −3.1 −8.0 (0.49), −9.0 to −7.1 −8.7 (0.50), −9.7 to −7.8 LS Means Difference (SE), 95% CI −3.9 (0.70), −5.3 to −2.5 −4.6 (0.71), −6.0 to −3.2 p value <0.0001 <0.0001 PANSS negative subscale3 LS means (SE), 95% CI −1.7 (0.40), −2.5 to −0.9 −3.8 (0.38), −4.5 to −3.0 −3.7 (0.39), −4.4 to −2.9 LS means difference (SE), 95% CI −2.1 (0.55), −3.1 to −1.0 −2.0 (0.55), −3.1 to −0.9 p value <0.001 <0.001 PANSS general psychopathology subscale3 LS means (SE), 95% CI −5.6 (0.82), −7.2 to −4.0 −12.8 (0.79), −14.4 to −11.3 −12.4 (0.81), −14.0 to −10.8 LS means difference (SE), 95% CI −7.3 (1.13), −9.5 to −5.0 −6.8 (1.15), −9.1 to −4.6 p value <0.0001 <0.0001 1 Primary efficacy endpoint. 2 Key secondary efficacy endpoint. 3 Secondary efficacy endpoint. CI confidence interval, CGI-I Clinical Global Impression-Improvement Scale, CGI-S Clinical Global Impression-Severity of Illness Scale, ITT Intent-to-Treat, KM Kaplan–Meier, PANSS Positive And Negative Syndrome Scale. a Data were analyzed using a mixed model repeated measures (MMRM) approach. b Difference (Risperidone ISM minus placebo) in least squares mean change from baseline adjusted by Lawrence and Hung method. C Hommel adjusted p value. randomized treatment were similar across treatment groups. No treatment of the acute schizophrenia. Superiority of active differences were noted in efficacy between patients who received treatment versus placebo was shown for the primary efficacy different prior treatments. outcome, with a statistically significant advantage of both Concomitant medications were taken by 56.5% of all patients Risperidone ISM® 75 mg and 100 mg to placebo on PANSS total during the study. The most common pharmacological groups score mean change from baseline to Day 85 in the mITT were anxiolytics (37.8%), anti-inflammatory and antirheumatic population,. The key secondary efficacy variable, CGI-S score products, non-steroidal anti-inflammatory agents (14.9%), and mean change from baseline to Day 85, was also superior to other analgesics and antipyretics (10.3%). The most common placebo for both doses of Risperidone ISM®. concomitant medications were lorazepam (34.6%), ibuprofen Although comparisons between studies should be interpreted (13.7%), paracetamol (9.2%), and lisinopril (8.0%). There were no with caution because of the differences in patient characteristics clinically relevant differences among the study groups. and study methodology, the placebo-adjusted reduction of the PANSS total score with Risperidone ISM® (−13.0 and −13.3) was higher than those obtained by other LAIs in similar acute DISCUSSION schizophrenia studies, being approximately twice that of RBP- The data obtained in the present study demonstrate the efficacy, 7000 90 mg/120 mg at 8 weeks (−6.1 and −7.2)13 and Paliper- safety, and tolerability of Risperidone ISM® in the monthly idone Palmitate 25 mg eq/100 mg eq/150 mg eq at 13 weeks npj Schizophrenia (2020) 37 Published in partnership with the Schizophrenia International Research Society C.U. Correll et al. 5 a a LS Mean (SE) Change From Baseline in PANSS 0 0 LS Mean (SE) Change From Baseline in PANSS -8 -2 ** * * *** Total Score -16 *** -4 ** positive subscale *** ** *** *** *** -24 -6 *** *** *** *** *** -32 -8 *** 0 4 8 15 29 57 85 *** Study Day *** -10 0 4 8 15 29 57 85 Placebo Risperidone ISM 75 mg Risperidone ISM 100 mg Study Day b 0 LS Mean (SE) Change From Baseline in CGI-S Placebo Risperidone ISM 75 mg Risperidone ISM 100 mg * ** ** ** b 0 LS Mean (SE) Change From Baseline in PANSS -1 ** ** Score ** ** ** ** -2 * * Negative Subscale * ** -2 *** 0 4 8 15 29 57 85 ** Study Day -4 ** *** Placebo Risperidone ISM 75 mg Risperidone ISM 100 mg Fig. 2 LS mean change from baseline at each time point (mITT -6 0 4 8 15 29 57 85 population). a PANSS total score, where mean PANSS score at baseline for placebo = 96.4 (SD: 7.21), for Risperidone ISM 75 mg = Study Day 96.3 (SD: 8.47) and for Risperidone ISM 100 mg = 96.1 (SD: 8.42). The Placebo Risperidone ISM 75 mg Risperidone ISM 100 mg error bars represent SE and P values are for Risperidone ISM 75 mg and Risperidone ISM 100 mg dose group versus placebo (*p < 0.01, **p < 0.001, ***p < 0.0001), in b CGI-S Score, where mean CGI-S score at baseline for placebo = 4.9 (SD: 0.54), Risperidone ISM 75 mg = 4.9 c 0 (SD: 0.63) and Risperidone ISM 100 mg = 4.8 (SD: 0.53). The error LS Mean (SE) Change from Baseline in PANSS bars represent SE and P values are for Risperidone ISM 75 mg and -2 General Psychopathology Subscale Risperidone ISM 100 mg dose group versus placebo (*p < 0.01, -4 **p < 0.0001). -6 * ** -8 *** (−5.1, −8.7, and −9.8),14 as well as similar to Aripiprazole once- -10 *** *** monthly 400 mg at 10 weeks (−15.1),15 and Aripiprazole Lauroxil -12 *** *** 441 mg/882 mg at 12 weeks (−10.9 and −11.9),16 where patients *** received also oral aripiprazole during 2 and 3 weeks after -14 *** 0 4 8 15 29 57 85 randomization, respectively. Study Day The onset of significant improvement in PANNS total and CGI-S score mean change was shown as early as Day 8 for the 100 mg Placebo Risperidone ISM 75 mg Risperidone ISM 100 mg dose and was maintained until the end of treatment period. Thus, Risperidone ISM® could address an unmet medical need and be Fig. 3 LS mean change from baseline in PANSS subscale score used as an early antipsychotic therapy at the admission of acutely (mITT population). a PANSS positive subscale, where mean PANSS positive subscale score at baseline for placebo = 25.3 (SD: exacerbated schizophrenic patient for a rapid and effective 3.11), for Risperidone ISM 75 mg = 25.1 (SD: 3.15) and for reduction of severe or moderate psychotic symptoms. Risperidone ISM 100 mg = 25.5 (SD: 3.40). The error bars Risperidone ISM® 75 mg and 100 mg were also superior to represent SE and P values are for Risperidone ISM 75 mg and placebo for improving the patients’ Positive and Negative Risperidone ISM 100 mg dose group versus placebo (*p < 0.05, Symptom Scores (PANSS), overall response rates, and CGI-I scores. **p < 0.001, ***p < 0.0001). b PANSS Negative Subscale, where Improvement in CGI-I scores and PANSS-positive symptom scores mean PANSS score at baseline for placebo = 23.5 (SD: 3.34), for compared with placebo was shown as early as Day 8 and Risperidone ISM 75 mg = 23.3 (SD: 4.19) and for Risperidone ISM 100 mg = 23.1 (SD: 3.73). The error bars represent SE and P values continued until Day 85. are for Risperidone ISM 75 mg and Risperidone ISM 100 mg dose Similarly, a significant improvement in overall response rate was group versus placebo (*p < 0.05, **p < 0.01, ***p < 0.001). c PANSS also seen as early as Day 8 for Risperidone ISM® 100 mg and for General Psychopathology Subscale, where Mean PANSS Score at Risperidone ISM® 75 mg at Day 15, being significantly higher in baseline: for placebo = 47.7 (SD: 4.90), for Risperidone ISM 75 mg both doses of Risperidone ISM® versus placebo, at all subsequent = 47.8 (SD: 5.48) and for Risperidone ISM 100 mg = 47.4 (SD: time points, with a high and clinically very meaningful number- 5.06). The error bars represent SE and P values are for Risperidone needed-treat of three. ISM 75 mg and Risperidone ISM 100 mg dose group versus Likewise, consistent results were demonstrated in the PANSS placebo (*p < 0.05, **p < 0.001, *** p < 0.0001). positive, negative, and General Psychopathology subscales. For the PANSS positive and negative subscales, statistically differences Published in partnership with the Schizophrenia International Research Society npj Schizophrenia (2020) 37 C.U. Correll et al. 6 60% Table 5. Injection site pain (safety population). 50% Responders (%) 40% Risperidone ISM Risperidone ISM 30% Time point Placebo 75 mg 100 mg 20% Statistic N = 147 N = 144 N = 146 10% VAS pain score 0% Mean (SD) 2.4 (2.56) 2.3 (2.31) 2.5 (2.38) Day 4 Day 8 Day 15 Day 29 Day 57 Day 85 Placebo Risperidone ISM 75mg Risperidone ISM 100mg 95% CI 1.9 to 2.8 1.9 to 2.7 2.1 to 2.9 Fig. 4 Overall response rate at each time point (mITT Population). Median 2.0 2.0 2.0 Where the responders are those patients who achieved a decrease IQR 0.0 to 4.0 0.0 to 3.0 0.0 to 4.0 from baseline in PANSS Total Score >30% or CGI-I at least much Min 0 0 0 improved. Dropouts prior to a presented time point are treated as non-responders and those patients who do not achieve a response Max 10 9 10 are censored on the day of withdrawal/completion from the CI confidence interval, IQR interquartile range, SD standard deviation, VAS treatment. Presented statistics, frequencies, and denominator used visual analog scale. The overall summary presents incidence for each for percentages are based on all patients in the ITT population and reaction category and maximum VAS score. the randomized treatment. *p < 0.01, **p < 0.0001. in mean change were obtained as early as Day 8 and Day 15 in Table 3. Summary of treatment-related TEAEs (safety population). positive and in negative subscales, respectively, demonstrating that Risperidone ISM® may rapidly resolve not only positive Preferred term Placebo Risperidone Risperidone N = 147 ISM 75 mg ISM 100 mg symptoms but also negative symptoms, unlike another monthly N = 144 N = 146 injectable Risperidone, in which changes in PANSS-negative scale n (%) n (%) n (%) scores was not significant different across the treatment and placebo groups13. In addition, the effect shown in the more Patients with at least 32 (21.8) 60 (41.7) 77 (52.7) severely ill subpopulation (≥95 PANSS total score at baseline) one TEAE further demonstrate the robust efficacy of Risperidone ISM®, and Hyperprolactinaemiaa 1 (0.7) 8 (5.6) 13 (8.9) particularly with the dose of 100 mg, which produced at endpoint Injection site pain 3 (2.0) 5 (3.5) 4 (2.7) a particularly large reduction over placebo of 15.6 points in PANSS total score. This finding provides further evidence for the potential Alanine 1 (0.7) 2 (1.4) 4 (2.7) aminotransferase value of Risperidone ISM® for the treatment of the acutely increased exacerbated schizophrenic patient who may need to be hospitalized owing to the severity of their symptoms. Blood prolactin 0 (0) 13 (9.0) 21 (14.4) increaseda Both doses of Risperidone ISM® were well tolerated. The adverse events (AEs) observed were those expected for oral and LAI Weight increased 2 (1.4) 5 (3.5) 6 (4.1) risperidone at therapeutic doses17 and were consistent with that Akathisia 2 (1.4) 5 (3.5) 11 (7.5) observed in previous studies with Risperidone ISM®11,12. Dizziness 4 (2.7) 4 (2.8) 5 (3.4) All TEAEs were mainly mild or moderate in most patients in Dystonia 1 (0.7) 4 (2.8) 3 (2.1) both treatment groups. Although the frequency of TEAEs was Headache 4 (2.7) 9 (6.3) 5 (3.4) lower with placebo than in the Risperidone ISM® groups, the rate was similar to those reported in a similar study in acute Somnolence 2 (1.4) 4 (2.8) 8 (5.5) schizophrenia,16 and slightly lower to those observed in previous Insomnia 1 (0.7) 3 (2.1) 4 (2.7) LAI risperidone study13. Furthermore, both Risperidone ISM® TEAEs treatment-emergent adverse event, descriptions of TEAEs are coded groups were associated with lower rate of discontinuation owing using MedDRA version 20.0; treatment-related TEAEs listed occurred in to TEAE compared with placebo, and no patient died owing to a ≥2% of risperidone ISM patients. TEAE during the study. a An increase in prolactin plasma levels were considered as an adverse Generally, the incidence of serious TEAEs and of TEAEs, leading event (either for hyperprolactinaemia or blood prolactin increased) when to study drug discontinuation was low and no clear differences any of the following criteria were present: Values above 1000 mIU/L for between treatment groups were observed. Similarly, the fre- three consecutive determinations after randomization, although no clinical quency of ISRs (redness, swelling, or induration) was low overall, symptoms are present or Values above 530 mIU/L if clinical symptoms of hyperprolactinaemia are present (e.g., headache, decreased libido, oligo- with redness being the most frequent in all treatment groups, and amenorrhea). with a slight trend for a dose-dependent increase of ISRs. No relevant differences between treatment groups were seen in the 0–10 Visual Analog Scale (VAS) score, with a median value of 2.0 in all treatment groups, which is a clinically meaningful result Table 4. Evaluation of injection site reactions (safety population). for a new LAI formulation. Similarly, the EPS, akathisia, dyskinesia, and suicidality safety scales also did not indicate significant Risperidone ISM Risperidone ISM differences between either dose of Risperidone ISM® and placebo. Time point Placebo 75 mg 100 mg There were further no significant differences in laboratory Statistic N = 147 N = 144 N = 146 measures between treatment arms from baseline through end of the study and no notable changes in either treatment arm, except Any reaction, n (%) 9 (6.1) 12 (8.3) 14 (9.6) for prolactin. The events related with the prolactin increase were Redness, n (%) 7 (4.8) 9 (6.3) 11 (7.5) between the more frequently reported TEAEs in this study, with a Swelling, n (%) 3 (2.0) 3 (2.1) 2 (1.4) comparable incidence to that described by others13,18. Induration, n (%) 0 (0) 2 (1.4) 1 (0.7) Several limitations need to be considered when interpreting the study results. As this was a short-term study in acute exacerbated npj Schizophrenia (2020) 37 Published in partnership with the Schizophrenia International Research Society C.U. Correll et al. 7 patients, it does not address long-term maintenance treatment, Treatment but the clinical efficacy of the active compound risperidone for Risperidone ISM® (Laboratorios Farmacéuticos ROVI, S.A., Madrid, Spain) maintenance treatment is very well known6,7,19,20. Nevertheless, a was available in a kit of two syringes, one containing Risperidone ISM® long-term evaluation of Risperidone ISM® is also being carried plus poly lactic-co-glycolic acid (PLGA) in the form of a solid powder, out,21 whose results will be published soon. Meanwhile, the data and the other containing dimethyl sulfoxide, the solvent required for reconstitution. Matching placebo was also available in a 2-syringes kit, shown here indicate that Risperidone ISM® could be a clinical with a similar appearance but containing only PLGA in the solid power option both for the acute and maintenance treatment of syringe. schizophrenia. Eligible patients were randomized 1:1:1 in a double-blind fashion to Patients with schizophrenia suffering a relapse need urgent Risperidone ISM® 75 mg, Risperidone ISM® 100 mg or placebo, injected attention owing to the severity of the symptomatology and future into the gluteal or deltoid muscle every 4 weeks on days 1, 29, and 57. A consequences if not treated immediately22. unique randomization number was assigned via Interactive Web In conclusion, Risperidone ISM® is a new monthly LAI Response System (IWRS) accessed immediately after eligibility confirma- antipsychotic that provides immediate and sustained drug plasma tion of a patient. The doses selected for this study were supported by the results obtained from previously conducted studies,11,12 as well as levels without loading doses or oral supplementation. Risperidone pharmacokinetic modeling26. ISM®, using monthly intramuscular doses of 75 mg or 100 mg, was Patients who had never taken risperidone had a brief trial of oral well tolerated, and provided significant improvement of the risperidone 2 mg/day for 3 days during the screening period to ensure lack symptomatology and disease severity in acutely exacerbated of any hypersensitivity reactions before the first dose of study drug. patients with schizophrenia. Moreover, this efficacy was observed as early as at the 8th day after first injection and was further Study assessments improved up to 12 weeks, without requiring any loading dose or Efficacy was assessed with the PANSS, CGI-S, and Clinical Global supplementation with oral risperidone. Thus, Risperidone ISM® can Impression-Improvements (CGI-I)24 at each scheduled visit. The primary be an effective therapeutic strategy at the admission of patients efficacy endpoint was the mean change in PANSS total score from baseline with schizophrenia suffering from an acute episode with severe or to end of treatment (Day 85 or last post-baseline assessment). The key moderate psychotic symptoms. secondary efficacy endpoint was CGI-S score mean change from baseline to end of treatment. Other secondary efficacy outcomes included mean CGI-I score at endpoint and each post-baseline assessment time point, METHODS overall response rate at endpoint (defined as PANSS total score ≥30% decrease from baseline to endpoint or CGI-I score of 2 (much improved) or Study design 1 (very much improved) at endpoint), time to reach overall response, and This was a phase III multicenter, randomized, double-blind, placebo- overall response rate at each post-baseline assessment time point, among controlled clinical trial (PRISMA-3), which was conducted between June others. 2017 and December 2018 at 26 sites in the United States and Ukraine, in Safety was evaluated by assessment of AEs, vital signs, laboratory test, accordance with the Declaration of Helsinki, and Good Clinical Practice electrocardiograms, physical examinations, ISRs (redness, swelling, and principles outlined in the International Conference on Harmonization. The induration), and scales to assess injection site pain (VAS) and extrapyr- protocol, amendments, and informed consent were approved by the Ethics amidal symptoms (AIMS;27 BARS28, and SAS29) as well as suicidality (C- Committee for each site, and written informed consent was obtained from SSRS). all subjects before study participation. This study was registered at ClinicalTrials.gov (identifier: NCT03160521). Statistical analysis The study consisted of a screening period of up to 8 days, immediately A sample size of 124 patients in the mITT population in each treatment group preceding the baseline day, followed by a treatment period of 12 weeks, would have 90% power to detect a difference in means of nine (standard which ended with a 2-week follow-up period. Eligible patients were deviation = 20, effect size = 0.45) with a 2.5% two-sided significance level for a randomly assigned 1:1:1 to double-blind intramuscular treatment with Risperidone ISM® group versus the placebo group. The power to show 75 mg or 100 mg of Risperidone ISM® or placebo. After initial dosing at superiority of both Risperidone ISM® doses to placebo using the above baseline, each study drug was administered intramuscularly once every calculation would be at least 81%. Taking into account that each of the two 4 weeks during the 12-week treatment period. Risperidone ISM® groups were tested separately against the placebo group, a Bonferroni adjustment for the alfa level was performed. A common standard Patients deviation of 20 in two-group t tests was assumed. Eligible subjects were 18–65 years old, with a current diagnosis of A relatively low post-randomization dropout of 5% rate was anticipated. This schizophrenia, according to the diagnostic and Statistical Manual of Mental assumption was re-assessed at the interim analysis and used in re-estimating Disorders, Fifth Edition criteria and a body mass index between 18.5 and the total number of randomized patients required. One unblinded interim 40.0 kg/m2. Patients were currently experiencing an acute exacerbation or analysis was planned to re-estimate the sample size required for the final analysis of up to 558 patients (186 patients per arm) in the mITT population. relapse with a total score between 80 and 120 on the PANSS,23 and a score This interim analysis was to be conducted when ~50% randomized patients, ≥4 points for ≥2 of the following positive symptom items: delusions, had either reached study day 85 or withdrawn from the study. The decision of conceptual disorganization, hallucinatory behavior, and suspiciousness/ the independent Data Monitoring Committee (DMC) was to continue the study persecution. All patients had to score of ≥4 (moderately ill or worse) on the without modifying the sample size. Clinical Global Impression-Severity scale (CGI-S)24 and had previously had a The efficacy analysis was performed on the mITT population containing clinically significant beneficial response after treatment with an anti- all randomized patients who received ≥1 dose of study drug with a psychotic other than clozapine. baseline measurement and ≥1 post-baseline evaluation for the PANSS, and Patients were excluded if improvement in PANSS total score was ≥20% for whom blinding was not potentially compromised (owing to a one-off between the screening visit and baseline, or with active suicidality, error in the IWRS). indicated by having answered “yes” on item 4 or 5 of the C-SSRS25 in the A mixed effects model with repeated measurements (MMRM) approach most recent episode (within the past 2 months) or having answered “yes” was fitted for patients in the mITT population with country where enrolled, to any of the five items (suicidal behavior) with an episode occurring visit, treatment, and treatment-by-visit interaction as fixed effects, and within the last year. Patients were also excluded for the presence of baseline PANSS total score as covariate. This MMRM models were used to clinically significant comorbid neuropsychiatric disorder, lifetime history of allow for an unstructured covariance pattern between visits to be fitted, schizoaffective or bipolar disorders, or a history of any unstable medical and the visit was fitted in the MMRM as a categorical factor. To utilize the condition or laboratory abnormality that could interfere with the conduct endpoint result (primary outcome) in the MMRM, the endpoint results from of the study or compromise the well-being of the patient. Women who any early termination visits were assigned to the next planned protocol were pregnant or breastfeeding were also excluded. visit in the MMRM. Published in partnership with the Schizophrenia International Research Society npj Schizophrenia (2020) 37 C.U. Correll et al. 8 The primary efficacy analysis was supported with sensitivity analyses. All schizoaffective disorder (PRISMA-1). Int. Clin. Psychopharmacol. 31, 323–331 analyses used the Cui, Hung, Wang adjustment,30 and the Hommel’s31 (2016). closed-testing correction procedure to present p values. Confirmatory 13. Nasser, A. et al. Efficacy, safety, and tolerability of RBP-7000 once-monthly ris- superiority of each Risperidone ISM® dose versus placebo was established peridone for the treatment of acute schizophrenia: an 8-week, randomized, when p value < 0.05. As a sensitivity analysis, the data for the primary double-blind, placebo-controlled, multicenter phase 3 study. J. Clin. Psycho- endpoint were analyzed for the mITT population using the analysis of pharmacol. 36, 130–140 (2016). covariance (ANCOVA) model. 14. Pandina, G. et al. A randomized, placebo-controlled study to assess the efficacy Point estimates and 95% CIs were obtained using methodology and safety of 3 doses of paliperidone palmitate in adults with acutely exacer- suggested by Hung and Lawrence32. This methodology constructs a point bated schizophrenia. J. Clin. Psychopharmacol. 30, 235–244 (2010). estimate and a 95% confidence interval that are motivated by an adaptive 15. Kane, J. M. et al. Aripiprazole once-monthly in the acute treatment of schizo- test statistic. phrenia: findings from a 12-week, randomized, double-blind, placebo-controlled The MMRM and ANCOVA models used both observed endpoint values study. J. Clin. Psychiatry 75, 1254–1260 (2014). and imputed33 study day 85 values performed for the mITT population. 16. Meltzer, H. et al. A randomized, double-blind, placebo-controlled trial of aripi- The ANCOVA models included country and baseline PANSS total score as prazole lauroxil in acute exacerbation of schizophrenia. J. Clin. Psychiatry 76, covariates. 1085–1090 (2015). The secondary efficacy endpoints were analyzed using the same model 17. Risperdal Consta US Package Insert. Available at: https://www.accessdata.fda.gov/ as for the primary efficacy endpoint. drugsatfda_docs/label/2010/021346_s31_s35_s38_s39lbl.pdf (2014). Safety and tolerability analyses were performed using data from the 18. Huhn, M. et al. Comparative efficacy and tolerability of 32 oral antipsychotics for safety population, which included all patients who received ≥1 dose of the acute treatment of adults with multi-episode schizophrenia: a systematic study drug. review and network meta-analysis. Lancet 394, 939–951 (2019). An independent DMC monitored patient recruitment, protocol com- 19. Kishimoto, T., Hagi, K., Nitta, M., Kane, J. M. & Correll, C. U. Long-term effectiveness pliance, reviewed safety data, and made recommendations about any of oral second-generation antipsychotics in patients with schizophrenia and existing or potential problems. related disorders: a systematic review and meta-analysis of direct head-to-head comparisons. World Psychiatry 18, 208–224 (2019). 20. Kishimoto, T. et al. Relapse prevention in schizophrenia: a systematic review and Reporting summary meta-analysis of second-generation antipsychotic versus first-generation anti- Further information on research design is available in the Nature Research psychotics. Mol. Psychiatry 18, 53–66 (2013). Reporting Summary linked to this article. 21. ClinicalTrials.gov. Study to evaluate the efficacy and safety of Risperidone ISM® in patients with acute schizophrenia: open label extension (PRISMA-3_OLE). https:// clinicaltrials.gov/ct2/show/NCT03870880?term=PRISMA-3+OLE&draw=2&rank=1 DATA AVAILABILITY (2020). The data that support the findings of this study are available from the corresponding 22. Correll, C. U., Rubio, J. M. & Kane, J. M. What is the risk-benefit ratio of long-term author upon reasonable request. antipsychotic treatment in people with schizophrenia? World Psychiatry 17, 149–160 (2018). 23. Kay, S., Fiszbein, A. & Opler, L. The positive and negative syndrome scale (PANSS) Received: 21 July 2020; Accepted: 23 October 2020; for schizophrenia. Schizophr. Bull. 13, 261–276 (1987). 24. Guy, W. The clinician global severity and impression scales. ECDEU Assessment Manual for Psycopharmacology. Superintendent of Documents, IS Government Printing Office, Publication No. 76-338. Washington, DC: US Department of Health, Education, and Welfare. 218–222 (1976). REFERENCES 25. Posner, K., Oquendo, M. A., Gould, M., Stanley, B. & Davies, M. Columbia Classi- 1. Kahn, R. S. et al. Schizophrenia. Nat. Rev. Dis. Prim. 1, 15067 (2015). fication Algorithm of Suicide Assessment (C-CASA): classification of suicidal 2. Keith, S. J. & Kane, J. M. Partial compliance and patient consequences in schi- events in the FDA’s pediatric suicidal risk analysis of antidepressants. Am. J. zophrenia: our patients can do better. J. Clin. Psychiatry 64, 1308–1315 (2003). Psychiatry 164, 1035–1043 (2007). 3. Ascher-Svanum, H. et al. Medication adherence and long-term functional out- 26. Winkler, J. et al. Population pharmacokinetic modelling and simulations of long- comes in the treatment of schizophrenia in usual care. J. Clin. Psychiatry 67, acting intramuscular Risperidone ISM. Clin. Therapeutics 37, e3–e4 (2015). 453–460 (2006). 27. Guy, W. Abnormal Involuntary Movement Scale (AIMS). ECDEU Assessment 4. Kane, J. M., Kishimoto, T. & Correll, C. U. Non-adherence to medication in patients Manual for Psychopharmacology. Rockville, MD: US Department of Health, Edu- with psychotic disorders: epidemiology, contributing factors and management cation and Welfare, Public Health Service, Alcohol, Drug Abuse, and Mental strategies. World Psychiatry 12, 216–226 (2013). Health Administration, National Institute of Mental Health, Psycopharmacology 5. Kane, J. M. Strategies for improving compliance in treatment of schizophrenia by Research Branch, Division of Extramural Research Programs. 534–537 (1976). using a long-acting formulation of an antipsychotic: clinical studies. J. Clin. Psy- 28. Barnes, T. R. A rating scale for drug-induced akathisia. Br. J. Psychiatry 154, chiatry 64, 34–40 (2003). 672–676 (1989). 6. Kishimoto, T. et al. Effectiveness of long-acting injectable vs oral antipsychotics in 29. Simpson, G. M. & Angus, J. W. A rating scale for extrapyramidal side effects. Acta patients with schizophrenia: a meta-analysis of prospective and retrospective Psychiatr. Scand. Suppl. 212, 11–19 (1970). cohort studies. Schizophr. Bull. 44, 603–619 (2018). 30. Cui, L., Hung, H. M. J. & Wang, S. J. Modification of sample-size in group 7. Kishimoto, T., Nitta, M., Borenstein, M., Kane, J. M. & Correll, C. U. Long-acting sequential trials. Biometrics 55, 853–857 (1999). injectable versus oral antipsychotics in schizophrenia: a systematic review and 31. Hommel, G. A stagewise rejective multiple test procedure based on a modified meta-analysis of mirror-image studies. J. Clin. Psychiatry 74, 957–965 (2013). Bonferroni test. Biometrika 75, 383–386 (1988). 8. Kane, J. M., Kishimoto, T. & Correll, C. U. Assessing the comparative effectiveness 32. Lawrence, J. & Hung, H. M. Estimation and Confidence after Adjusting the Max- of long-acting injectable vs. oral antipsychotic medications in the prevention of imum Information. Biometrical J. 45, 143–152 (2003). relapse provides a case study in comparative effectiveness research in psychiatry. 33. Ratitch, B. & O’Kelly, M. Implementation of pattern-mixture models using stan- J. Clin. Epidemiol. 66, S37–S41 (2013). dard SAS/STAT procedures. PharmaSUG, SG04 (2011). 9. Correll, C. U. et al. The use of long-acting injectable antipsychotics in schizo- phrenia: evaluating the evidence. J. Clin. Psychiatry 77, 1–24 (2016). 10. Citrome, L. sustained-release risperidone via subcutaneous injection: a systematic ACKNOWLEDGEMENTS review of RBP-7000 (PERSERIS™) for the treatment of schizophrenia. Clin. Schi- We would like to acknowledge to Lourdes Anta, for drafting and editing the zophr. Relat. Psychoses 12, 130–141 (2018). manuscript; Begoña Gorostidi, Laura Vallejo, Ignacio Ayani, Ana B. López, Mateo 11. Anta, L. et al. A phase II study to evaluate the pharmacokinetics, safety, and Fernández, Lourdes Ochoa, and Ibón Gutierro, for their contributions to the study. All tolerability of Risperidone ISM multiple intramuscular injections once every of them are employees of Laboratorios Farmacéuticos Rovi, S.A., Madrid, Spain. We 4 weeks in patients with schizophrenia. Int. J. Clin. Psychopharmacol. 33, 9–87 thank the study participants, their families and cares, and the investigators for their (2018). participation. This study was funded by Laboratorios Farmacéuticos Rovi, S.A. Madrid, 12. Llaudó, J. et al. Phase I, open-label, randomized, parallel study to evaluate the Spain. It was supported also in part by a grant from Centro para el Desarrollo pharmacokinetics, safety, and tolerability of one intramuscular injection of Ris- Tecnológico Industrial (Expedient No. IDI-20160109). Previous presentations: these peridone ISM at different dose strengths in patients with schizophrenia or results have not been published previously. npj Schizophrenia (2020) 37 Published in partnership with the Schizophrenia International Research Society C.U. Correll et al. 9 AUTHOR CONTRIBUTIONS ADDITIONAL INFORMATION J.LL., J.M., CU.C., D.N., and F.T., were involved in the data analysis. J.LL., J.M., RE.L., Y.F., Supplementary information is available for this paper at https://doi.org/10.1038/ were involved in the design and interpretation. Manuscript development was s41537-020-00127-y. performed by CU.C., J.LL., and J.M. All the authors have made substantial contributions and have revised, edited, and approved the manuscript. Correspondence and requests for materials should be addressed to J.M. Reprints and permission information is available at http://www.nature.com/ reprints COMPETING INTERESTS J.LL. and J.M. are employees of Laboratorios Farmacéuticos Rovi, S.A., the sole Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims developer of Risperidone ISM®. CU.C. has been a consultant and/or advisor to or have in published maps and institutional affiliations. received honoraria from: Alkermes, Allergan, Angelini, Boehringer-Ingelheim, Gedeon Richter, Gerson Lehrman Group, Indivior, IntraCellular Therapies, Janssen/J&J, Laboratorios Farmacéuticos Rovi, S.A, LB Pharma, Lundbeck, MedAvante-ProPhase, Medscape, Merck, Neurocrine, Noven, Otsuka, Pfizer, Recordati, Servier, Sumitomo Dainippon, Sunovion, Supernus, Takeda, and Teva. He provided expert testimony for Open Access This article is licensed under a Creative Commons Janssen and Otsuka. He served on a Data Safety Monitoring Board for Lundbeck, Rovi, Attribution 4.0 International License, which permits use, sharing, Supernus, and Teva. He has received grant support from Janssen and Takeda. He is adaptation, distribution and reproduction in any medium or format, as long as you give also a stock option holder of LB Pharma. RE.L., is a Member, Speakers Bureau, for JNJ appropriate credit to the original author(s) and the source, provide a link to the Creative and for Allergan, Member, Scientific Advisory Board, Terran BioSciences. He Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless participated as Investigator Coordinator in The United States of America and as indicated otherwise in a credit line to the material. If material is not included in the Principal Investigator in the study PRISMA-3. Y.F. has participated as Investigator article’s Creative Commons license and your intended use is not permitted by statutory Coordinator in Ukraine and Principal Investigator in the study PRISMA-3. He received regulation or exceeds the permitted use, you will need to obtain permission directly fees from the sponsor for his advice in some previous task related to this study. D.N., from the copyright holder. To view a copy of this license, visit http://creativecommons. has received consulting fees from Laboratorios Farmacéuticos Rovi, S.A. He served on org/licenses/by/4.0/. the Data Monitoring Committee for Rovi at the present study. F.T. has received consulting fees from Laboratorios Farmacéuticos Rovi, S.A. He served on the Data Monitoring Committee for Rovi at the present study. © The Author(s) 2020 Published in partnership with the Schizophrenia International Research Society npj Schizophrenia (2020) 37