Current Psychiatry Reports (2019) 21:124 https://doi.org/10.1007/s11920-019-1114-0 SCHIZOPHRENIA AND OTHER PSYCHIATRIC DISORDERS (AK PANDURANGI, SECTION EDITOR) Long-Acting Injections in Schizophrenia: a 3-Year Update on Randomized Controlled Trials Published January 2016–March 2019 Luisa Peters 1 & Amanda Krogmann1 & Laura von Hardenberg 1 & Katja Bödeker1 & Viktor B. Nöhles1 & Christoph U. Correll1,2,3 # Springer Science+Business Media, LLC, part of Springer Nature 2019 Abstract Purpose of Review This study was conducted in order to review randomized controlled trial (RCT) data published January 2016– March 2019 on long-acting injectable antipsychotics (LAIs) for schizophrenia. Recent Findings Thirty-one RCTs (primary studies = 7; post hoc analyses = 24; n = 4738) compared LAIs vs. placebo (studies = 11, n = 1875), LAIs vs. oral antipsychotics (OAPs) (studies = 7, n = 658), and LAI vs. LAI (studies = 13, n = 2205). LAIs included two new formulations, aripiprazole lauroxil nanocrystal dispersion and subcutaneously injectable risperidone Perseris, as well as aripiprazole lauroxil, aripiprazole once-monthly, paliperidone once-monthly, paliperidone 3-monthly, and risperidone- LAI. Regarding prevention of relapse and hospitalization, LAIs consistently outperformed placebo, being partly superior to OAPs, without relevant LAI–LAI differences. LAIs were comparable to OAPs regarding all-cause discontinuation, functioning, quality of life, and tolerability, being associated with higher patient satisfaction and service engagement. Recent meta-analyses yielded mixed results, but never favoring OAPs over LAIs. Summary In RCTs, LAIs are superior to placebo, but only in some aspects, superior to OAPs. Comparative effectiveness of LAIs vs. OAPs requires further study, ideally in generalizable/real-world samples. Keywords Schizophrenia . Antipsychotics . Long-acting injectables . Relapse . Hospitalization . Symptom improvement . Acceptability . Discontinuation . Meta-analyses Introduction comealongwaywiththeinitialdevelopment of antipsychotics and atypical antipsychotics [2–5], which have remained the only Schizophrenia remains one of the most severe medical diseases approved and effective pharmacologic treatment for schizophre- [1]. The treatment of this chronic and often disabling disorder has nia [4, 6•]. Despite pharmacologic advances in recent years, an Luisa Peters and Amanda Krogmann contributed equally to this work. This article is part of the Topical Collection on Schizophrenia and Other Psychiatric Disorders * Christoph U. Correll Viktor B. Nöhles [email protected] [email protected] Luisa Peters [email protected] 1 Department of Child and Adolescent Psychiatry, Charité Universitätsmedizin, Augustenburger Platz 1, Amanda Krogmann 13353 Berlin, Germany [email protected] 2 Laura von Hardenberg Psychiatry Research, The Zucker Hillside Hospital, Northwell [email protected] Health, 75-59 263rd Street, Glen Oaks, NY 11004, USA Katja Bödeker 3 Department of Psychiatry and Molecular Medicine, Zucker School of [email protected] Medicine at Hofstra/Northwell, Hempstead, NY, USA 124 Page 2 of 31 Curr Psychiatry Rep (2019) 21:124 effective treatment of schizophrenia remains an issue, both due to acceptability of LAIs. Since a comprehensive review of the insufficient treatment of cognition and negative symptoms as literature on the use of LAIs up to 2015 already exists [13], we well as the challenge of treatment nonadherence, which is strong- focused on the time frame 2016–2019. ly linked to recurrent relapses [7, 8]. Given the need for long- term maintenance treatment in schizophrenia [9•], relapse pre- vention is a major goal [10, 11•, 12]. Methods In order to improve adherence to prescribed antipsychotics through less frequent dosing schedules and readily observable We aimed to summarize the recent evidence of RCTs compar- nonadherence, long-acting injectable antipsychotics (LAIs) ing LAIs to placebo, OAPs, or another LAI for schizophrenia- have been developed [13••]. Although in randomized con- spectrum disorders. Therefore, we conducted a systematic lit- trolled trials that systematically include more adherent patients erature search of articles published in PubMed between 01 who know that their medication adherence will be monitored January 2016 and 26 March 2019, using the following search closely [14], LAIs were not superior to oral antipsychotics terms: ((schizophrenia OR schizophreniform OR (OAPs) regarding relapse and treatment continuation [15], schizoaffective OR psychosis OR psychotic) AND (depot and meta-analyses of mirror image and cohort studies that OR long-acting injectable OR injection* OR microsphere* arguably enroll more generalizable patients [14]demonstrated OR decanoate OR palmitate OR enanthate OR monohydrate that LAIs are superior to OAPs in treatment discontinuation OR lauroxil OR nanocrystal)). [16, 17•], relapse [18] and hospitalization [16, 17], time to We included RCTs that investigated LAIs in adults with relapse [18], and hospital days [18]. As schizophrenia is often schizophrenia-spectrum disorders focusing on new LAI for- associated with functional impairments, recent research also mulations (ALNCD and RBP-7000), efficacy of LAIs for focused on the maintenance and improvement of functionality symptom improvement and prevention of relapse and hospi- and quality of life with both oral antipsychotics [19•] and talization, functionality and quality of life, tolerability and LAIs [13]. In terms of tolerability and safety, meta-analyses safety of LAIs, and attitudes toward as well as acceptability have demonstrated that LAIs are comparable to OAPs [20•, of LAIs. Additionally, we considered published expert recom- 21•], with one meta-analysis favoring LAIs over OAPs [18]. mendations for the use of LAIs. We excluded studies in chil- Furthermore, novel long-acting injectable formulations have dren and adolescents, reporting on mixed-age populations emerged, aiming to move from deep intramuscular to subcutane- without reporting results separately for adults, and LAI dose ous injections and/or abolish the need for initial oral cotreatment comparisons. Finally, we also reviewed meta-analytic results or loading injection strategies, which may increase acceptability of RCTs, as long as the meta-analysis was published in [6]. These formulations include RBP-7000 (Perseris™), a long- January 2016 or later. acting injectable formulation of risperidone [22], requiring no Results are summarized first for the individual RCTs focus- oral supplementation and being the first antipsychotic available ing on the following five characteristics/outcomes: [1]FDA as a subcutaneously (intraabdominally) administered LAI [23•], approval trial, [2] efficacy for symptom improvement and and aripiprazole lauroxil nanocrystal dispersion (ALNCD; prevention of relapse/hospitalization, [3] efficacy for Aristada Initio™), a 1-day initiation regimen for the LAI functionality/quality of life, [4] tolerability/safety, and [5]at- aripiprazole lauroxil (AL), reducing oral supplementation to the titudes/acceptability, dividing the results by the comparator first day of administration and enabling faster release of (placebo, OAP, LAI). At the end, results of the meta- aripiprazole into plasma [24]. analyses are summarized for efficacy and safety/tolerability Despite demonstrated advantages of LAIs, they remain large- of LAIs vs. the meta-analyzed comparator. ly underutilized in clinical practice, possibly due to lack of famil- iarity of many physicians, incomplete or inaccurate perceptions about safety and efficacy, medication cost considered in isolation Results without taking into account overall reductions in cost of care due to reduced relapse and hospitalizations, issues with access to Search Results and Study Characteristics treatment, and negative perceptions about injectable therapy among patients, families, and prescribers [13]. The PubMed search resulted in 821 hits. After removal of 776 In this context, this systematic review aimed to summarize articles based on title and abstract review, 45 full text articles evidence from articles published 2016–2019 that reported data were inspected for eligibility. Of these, 14 were excluded for on findings of randomized controlled trials (RCTs) regarding the following reasons: LAI dose comparison (n = 8), short- (i) new long-acting injectable formulations; (ii) the efficacy of acting injectable data (n = 4), no data on LAIs (n = 1), and LAIs for symptom improvement and prevention of relapse duplicate publication (n = 1). Ultimately, we included 31 and hospitalization, functionality, and quality of life; (iii) the RCTs (primary data studies: k = 7; post hoc analyses: k = tolerability and safety of LAIs; and (iv) attitudes toward and 24; individual patients: n = 4738), comparing LAIs vs. Curr Psychiatry Rep (2019) 21:124 Page 3 of 31 124 placebo (studies = 11, n = 1875) and LAIs vs. OAPs (studies = Global Impression Scale-Severity (CGI-S) scores (90 mg: p 7, n = 658) and comparing one LAI vs. another LAI (studies = = 0.0002; 120 mg: p < 0.0001) vs. placebo [28•](Table1). In a 13, n = 2205). Altogether, 20 studies (n = 3812) had a double- post hoc analysis of the same study [28•], the maximum blind (DB) and 11 had a randomized, open-label (ROL) (n = placebo-corrected relative decrease in PANSS score in the 926) design. RBP-7000 group was 5.4%, half of which was achieved at Additionally, five meta-analyses of RCTs of LAIs in plasma concentrations of 4.6 ng/mL