Risperidone for Treatment-Refractory Major Depressive Disorder

Context Does augmentation with an atypical antipsychotic improve symptoms in patients with major depression that is suboptimally responsive to antidepressant monotherapy? Contribution This double-blind randomized trial found that 6 weeks of treatment with risperidone improved symptoms more than placebo in 274 adults with major depression that was suboptimally responsive to antidepressant monotherapy. Risperidone, compared with placebo, resulted in more remissions (25% vs. 11%), as well as more cases of somnolence (5% vs. 2%) and dry mouth (5% vs. 1%). Caution The trial duration was short, and 19% of risperidone recipients did not complete treatment. Implication Risperidone augmentation may improve symptoms in some patients with suboptimal response to antidepressant monotherapy. The Editors Major depressive disorder is the leading cause of disability worldwide (1), affecting nearly 121 million people. It is associated with increased overall mortality (2, 3); premature cardiovascular-related death (4, 5); and morbidity, including disability, lost productivity, and decreased wages, compared with individuals without depression (6). Selective serotonin reuptake inhibitors (SSRIs) and serotoninnorepinephrine reuptake inhibitors are generally considered first-line treatments for depression. Despite a sufficient dose and duration of antidepressant treatment, symptoms do not resolve in 30% to 40% of patients with major depressive disorder (710). Remission, defined as the virtual absence of depressive symptoms, is important because it is associated with improved patient prognosis and functioning relative to response (generally defined as 50% reduction in symptoms) (11, 12). In an effort to improve the therapeutic efficacy of antidepressant monotherapy, several pharmacologic strategies to modulate different or additional neurotransmitters (1315) are commonly used, despite a paucity of evidence from randomized clinical trials. These strategies include switching monotherapy within (10, 1618) or between antidepressant classes (10, 1921), combination regimens consisting of 2 antidepressants (2226), and augmentation of antidepressants with nonstandard treatmentsfor example, thyroid hormone (27); S-adenosyl-l-methionine (28); lithium (29); cyclooxygenase-2 inhibitors (30); or an atypical antipsychotic (3133), such as risperidone. Findings of several open-label studies (3135) suggest that atypical antipsychotic augmentation provides benefit in major depressive disorder that is suboptimally responsive to antidepressant monotherapy. Risperidone, whose activity includes blockade of certain serotonin and dopamine receptors, is considered an atypical antipsychotic. It is approved in the United States for treatment of bipolar mania and schizophrenia and in other countries for additional uses. The efficacy of risperidone has also been studied for various psychoses and behavioral disorders, including treatment-resistant depression and anxiety spectrum disorders (34, 3639). We hypothesized that low-dose risperidone augmentation would reduce symptoms of major depressive disorder, enhance clinical response and remission rates, and reduce disability and improve quality of life compared with continued antidepressant monotherapy. Preliminary data suggest that persons with a confirmed suboptimal response to previous treatment may respond to risperidone augmentation within 4 to 6 weeks (38). We therefore conducted a large, multicenter, double-blind, placebo-controlled trial to assess the clinical efficacy of risperidone augmentation from the clinician and patient perspectives and to evaluate tolerability in patients with major depressive disorder who continue to experience symptoms despite an adequate trial of standard antidepressants. Methods Design Consenting persons entered a 4-week prospective open-label run-in period during which they received their current antidepressant monotherapy at the dosages recommended in product labeling, so that we could confirm an insufficient response to standard monotherapy. The run-in period was followed by a 6-week double-blind treatment phase; during this phase, patients who continued to experience depressive symptoms, defined as a Clinical Global ImpressionSeverity of illness (CGI-S) score of 4 or greater and a Carroll Depression Scale (40) score of 20 or greater, were randomly assigned to receive risperidone augmentation therapy or placebo. A 4-week follow-up phase of open-label risperidone augmentation therapy was available to participants who completed at least 4 weeks of double-blind treatment (results not shown). Setting and Participants Outpatients 18 to 65 years of age who had received antidepressant monotherapy for at least 4 weeks and met the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (41), criteria for unremitting major depressive disorder (single or recurrent episodes) and a CGI-S score of 4 or more (42) at the start of both the open-label and double-blind phases were eligible. The CGI-S is scored on a scale of increasing intensity from 1 to 7, with 4 representing moderate illness. We recruited participants by using various methods, including media advertisements, and enrolled persons from 75 public and private primary care (n= 41) and psychiatric (n= 34) centers in the United States between 19 October 2004 and 17 November 2005. Ethical approval was obtained from the institutional review board of each site, and written informed consent was obtained from all patients. Exclusion criteria were pregnancy; serious suicidal risk or serious medical or neurologic illness; active substance or alcohol use disorders; or current treatment with a tricyclic antidepressant, monoamine oxidase inhibitor, mood stabilizer, antiepileptic, or a centrally acting agent for attention deficit disorder/attention deficit hyperactivity disorder or narcolepsy. Randomization and Interventions Patients who were eligible for randomization continued their standard antidepressant regimen and dosage and were assigned to receive risperidone or placebo. Tablets were identical in appearance. The investigators, study staff, and patients were blinded to the treatment assignment. The randomization code was centrally generated and administered by a telephone interactive voice response system, was stratified by antidepressant class (SSRI or non-SSRI) and center, and occurred in random permuted blocks. An independent statistician provided the randomization codes. The dosing schedule was 0.25 mg for the first 3 days, 0.5 mg on days 4 to 15, and 1.0 mg on days 16 to 28. On day 29, patients with insufficient treatment response in the opinion of the investigator could continue augmentation at the current dosage, have the dosage increased to 2 mg/d, or discontinue double-blind treatment. Concomitant medications were allowed as necessary for medical conditions. Sedative agents (zolpidem, 2.5 to 10 mg/d, or zaleplon, 5 to 20 mg/d, as needed) were allowed for insomnia. During double-blind treatment, benztropine mesylate was permitted as needed for potential treatment-emergent motor effects. Measurements and Outcomes At each visit (end of the open-label phase and weeks 1, 2, 4, and 6), trained personnel administered the grid version of the 17-item Hamilton Rating Scale for Depression (HRSD-17) (43, 44) and the CGI-S (42), both clinician-rated instruments. Scores on the HRSD-17 range from 0 to 52; higher scores indicate more severe depression. Patients also independently rated their response by using the telephone interactive voice response system at baseline and weekly thereafter. Other patient-reported outcomes were scores on 3 validated instruments: the Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q) (45), the Patient Global Improvement Scale (7-point rating from 1 [very much improved] to 7 [very much worse]), and the Sheehan Disability Scale (SDS). Change in HRSD-17 total score from baseline to study end was the a priori primary efficacy measure. Response (50% reduction in HRSD-17 total score from baseline) and remission (HRSD-17 total score 7) were assessed at each time point. Secondary measures were changes in the CGI-S, Q-LES-Q, SDS, and Patient Global Improvement Scale scores. Follow-up Patients underwent physical examination, including assessment of vital signs, at regular intervals. Research staff maintained records of trial medication and assessed treatment compliance by matching doses taken with the number of treatment days. To assess adverse effects, patients were interviewed at each study visit with open-ended questions about potential adverse events; spontaneous reports were also taken into account. All adverse events were recorded, and the investigator assessed each event for severity and relationship to the study drug (probable, possible, not related). Statistical Analysis A sample of 116 patients per group was anticipated to have 90% power to detect an arbitrary difference in change in mean HRSD-17 total score of 3.0 units, assuming a common SD of 7.0 using a 2-group t test and a 2-sided significance level of 0.05. We also assumed that approximately 30% of the participants would not complete double-blind treatment and planned for 332 persons to undergo randomization. We further assumed that during the open-label run-in phase, 30% of participants would discontinue the study and 20% would respond to standard antidepressant treatment and therefore be ineligible for randomization; thus, we sought 592 persons for the open-label run-in phase. All analyses were performed by using SAS, version 8.2 (SAS Institute, Cary, North Carolina), and statistical analysis personnel were blinded to patient group assignment. Baseline patient characteristics were summarized by using descriptive statistics. Efficacy and safety analyses were based on the intention-to-treat population (that is, all persons who underwent randomization and received at least 1 dose of medication in the double-blind phase). Obser