European Neuropsychopharmacology (v.22, #11)

Contents (OBC).

Editorial by Christine Gispen-de Wied; Fernando de Andres-Trelles (769).

Placebo, a historical perspective by Efrat Czerniak; Michael Davidson (770-774).
Substances and interventions with no specific therapeutic effect have been in use since the dawn of history. The term placebo has first been mentioned in the Scriptures, but it was not until the 19th century that it appeared in a medical context. Although lay people like Voltaire, and physicians such as Sir William Osler, have raised the possibility that much of what physicians did had no specific therapeutic effect, this notion was not shared by the public at large or by the medical profession. It was only by the end of the 18th century that a placebo-controlled trial has been conducted, repudiating the therapeutic effect of mesmerism. The advent, in the late 1940s, of effective treatments, which also had serious adverse effects, made the distinction between placebo and putative, active drug effects more relevant and urgent, and cleared the way for double-blind, randomized, placebo-controlled trials. This in turn triggered an ethical debate on the use of placebo, both in research and in clinical practice. Anthropologists, sociologists, physiologists, and medical researchers are all focusing their efforts on understanding the mechanism, role and modulating factors of placebo.
Keywords: Placebo; History; RCT;

Placebo fascinates and mystifies. Even with today's medical science we still do not know how and if it works. The use of placebo both in therapy and in research evokes ethical problems that are not easily resolved either. Placebo is intrinsically linked to deception, while veracity is a basic tenet in today's thinking of a doctor–patient relationship. In research ethics placebo, though considered the golden control condition, leads to the question of the therapeutic obligation. This narrative review presents an overview of these ethical questions and offers considerations that are of relevance to daily medical and research practice both in psychiatry and elsewhere.
Keywords: Placebo; Psychiatry; Research; Ethics; Informed consent; Doctor–patient relationship;

It is widely believed that in randomized controlled trials of antidepressants the difference between drug and placebo response rates is rather small (around 20%), leading to a common perception that antidepressants have limited efficacy.The aim of the present paper was to present an alternative calculation and interpretation of antidepressant–placebo difference in the treatment response to antidepressant in drug trials which may shed a new light on the efficacy of antidepressants.We have previously highlighted several controversial points concerning the calculation of antidepressant and placebo response rates in randomised controlled trials, which may influence views concerning the efficacy of drugs, and demonstrated several factors which may lead to overestimation of the placebo effect and underestimation of antidepressant efficacy. The traditional interpretation of antidepressant–placebo difference in randomized controlled trials on major depression has been also challenged previously from at least five points of view but all leading to a conclusion that currently prevailing opinions concerning relative placebo and antidepressant response rates overestimate placebo response, and thereby underestimate efficacy of antidepressant drugs. In our present paper we propose another method for calculating placebo and antidepressant response rates which may shed new light on an overlooked aspect of the efficacy of these drugs.We contend that opinions on the effectiveness of antidepressants should be reconsidered, and comparisons with placebo should be more carefully applied. Interpretation of the placebo response is of crucial importance for establishing the efficacy of antidepressive medications, and psychiatry should not become the hostage of placebo.
Keywords: Antidepressant; Placebo; Antidepressant efficacy; Antidepressant drug response;

Placebo effect in child and adolescent psychiatric trials by Mara Parellada; Carmen Moreno; Miguel Moreno; Ana Espliego; Enrique de Portugal; Celso Arango (787-799).
Much literature has been written in the field of child psychiatry regarding the placebo as a tool to test drug efficacy in clinical trials, but quite little regarding the placebo effect itself or its clinical use in child psychiatry. In this article, we aim to critically review the literature regarding the placebo effect in children and adolescents with mental disorders, focusing especially on factors influencing the placebo effect and how they may influence the interpretation of clinical trials. The placebo effect seems to be more marked in children than adults, and particularly in children and adolescents with depression, although it is pervasive across ages and is present in non-psychiatric conditions as well. The use of a placebo in clinical trials as a comparator with drugs that have moderate efficacy at most makes it difficult to obtain positive results, and much effort is needed to design very high quality clinical trials that may overcome the limitations of using a placebo. In addition, the placebo effect across ages and clinical conditions must be tested directly (compared with no treatment whenever possible), in order to characterise which placebos work for what and to determine their use in clinical settings.
Keywords: Placebo effect; Child psychiatry; Adolescent psychiatry; Clinical trials;

Head to head trials have been proposed as an alternative to the ethical and methodological concerns related to placebo-controlled trials. While those studies may be particularly informative from the clinical and cost-effectiveness point-of-view, avoiding placebo poses several regulatory concerns: for superiority designs, the choice of the trial population, outcomes, dose and escalation of the comparator, as well as the comparator itself may be an issue; for non-inferiority studies, issues related to uncertain assay sensitivity and exposure of large samples to potentially ineffective or unsafe drugs make them inappropriate, in the absence of a previous positive superiority trial, for regulatory purposes. The inclusion of active comparators in regulatory trials should not be seen as an alternative, but as a useful complement to the information that can be obtained from placebo-controlled studies.
Keywords: Placebo; Non-inferiority; Methodology; Psychopharmacology;

The placebo arm in clinical studies for treatment of Psychiatric Disorders: A Regulatory Dilemma by Christine Gispen-de Wied; Violeta Stoyanova; Yang Yu; Maria Isaac; Luca Pani; Fernando de Andres-Trelles (804-811).
The use of placebo in clinical trials, and, related to this, ethical and feasibility aspects, are often debated. However, regulatory authorities must ensure that only new drugs with a positive benefit/risk would be granted a marketing authorization. It is therefore not surprising that they often put forward the need for placebo control in clinical trials in an area where many trials fail, and assay sensitivity is not self-evident. To illustrate the complexity that regulatory authorities encounter when faced with the registration dossier of products in the main psychiatric therapeutic areas, Major Depressive Disorder (MDD) and schizophrenia, the trial outcome for products receiving an opinion in the EU during the past 15 years were reviewed.European Public Assessment Reports and registration files.A total of 45 studies qualified for analysis. For the indication MDD 38% of the studies (10/26) were recorded as failed, and another 15% (4/26) as negative. For schizophrenia, these figures were 16% (3/19) and 11% (2/19). Further exploration of the trials in MDD revealed an inconsistent pattern in terms of magnitude of placebo- and drug-mediated response (i.e. similar studies with consistent placebo response provided different treatment outcomes).From a regulatory perspective the dilemma of a priori exclusion of the placebo arm in clinical trials in the domains of depression or schizophrenia cannot be solved at this time as long as factors influencing trial variability are not better identified or understood. This counts in particular for MDD where the added drug effect is not consistent across trials with almost identical inclusion criteria. Unfortunately, this trend has not changed over the past 15 years. However, all efforts should be taken to optimize the clinical development of drugs in the psychiatric domain, and improve the intrinsic quality of the clinical trials in order to allow for a different viewpoint.
Keywords: Clinical trial; Placebo; Psychiatry; Psychopharmacology; Regulatory science;

The competing hypotheses that the action onset to antipsychotic medication assumes a course of early- or a delayed-response have been tested in positive and not negative symptoms in schizophrenia. The current study aims to test the early- and delayed-onset hypotheses with regard to negative symptoms. Data were re-analyzed from three clinical trials that compared placebo or amisulpride for up to 60 day. Participants had predominantly negative symptoms of schizophrenia (n=487). Response was examined with the incremental percentage Scale for the Assessment of Negative Symptoms (SANS) reduction over time. Response to the treatment, visit and treatment–visit interaction was assessed with mixed-modeling. Effect size differences on response between the amisulpride and placebo groups were reported at each visit. Across trials, mixed modeling showed that the incremental SANS reductions by the treatment–visit interaction that tests the action–onset hypothesis were not statistically significantly different across periods. The effect size differences of medication less placebo in the incremental percent SANS reduction showed non-significant differences based on overlapping confidence intervals with a moderate improvement at 8–14 day (ES=.33; 95% CI: −.07,.31), the least improvement at 28–30 day (ES=.12; 95% CI: −.07,.31), and a moderate improvement at 42–60 day to (ES=.39, 95%, CI: .19,.60). Generally, early- and delayed-response differences to antipsychotic were limited.
Keywords: Scale for the Assessment of Negative Symptoms; Early-response; Delayed-response; Treatment; Negative symptoms;

Geographical and temporal variations in clozapine prescription for schizophrenia by Jimmi Nielsen; Rasmus Røge; Ole Schjerning; Holger J. Sørensen; David Taylor (818-824).
Despite its unsurpassed efficacy in treatment-resistant schizophrenia, clozapine remains underutilized. Trends in the prescription of clozapine in patients with ICD-10 F20.x schizophrenia were assessed using data from Danish national registers. Three substudies were carried out: (i) an assessment of differences in national prescription patterns between 1996 and 2007 using a cross-sectional design; (ii) a comparison of time from first schizophrenia diagnosis to first prescription of clozapine in a five-year cohort study, using the Cox regression model, of two patient groups who were first diagnosed in 1996 and in 2003; (iii) an assessment of differences in the general psychiatric hospitals' use of clozapine in 2009. The results are as follows: (i) The percentage of schizophrenia patients receiving clozapine rose from 9.0% in 1996 to 10.1% in 2007 (p<0.001). In the same period, the percentage of patients having clozapine treatment augmented with another antipsychotic increased from 43.1% to 64.2%, p<0.001. (ii) Time from diagnosis with schizophrenia until first clozapine prescription was longer for patients diagnosed in 2003 compared to those diagnosed in 1996 (HR: 0.28 CI: 0.16–0.49). (iii) In 2009 there was significant variation in clozapine administration from one hospital to the other, with percentages of patients receiving the drug ranging from 5.7% to 16.8%, with 10.2% as the national mean. Although, the percentage of schizophrenia patients receiving clozapine increased from 1996 to 2007, the time from diagnosis of schizophrenia until first prescription of clozapine increased.
Keywords: Clozapine; Antipsychotic; Prescribing pattern; Augmentation; Polypharmacy; Schizophrenia;

A randomized, double-blind, placebo-controlled trial to assess prevention of mood episodes with risperidone long-acting injectable in patients with bipolar I disorder by Eduard Vieta; Stuart Montgomery; Ahmad Hatim Sulaiman; Rodrigo Cordoba; Benedicte Huberlant; Lupe Martinez; Andreas Schreiner (825-835).
The efficacy and safety of risperidone long-acting injectable (LAI) for preventing recurrence of mood episodes in patients with bipolar I disorder was evaluated in a randomized, placebo-controlled study. After a 12-week open-label period with risperidone LAI (n=560), patients who did not experience a recurrence entered an 18-month randomized, double-blind period with risperidone LAI (n=132) or placebo (n=135); a third treatment arm (n=131) was randomized to oral olanzapine (10 mg/day) for reference and exploratory comparisons. The primary efficacy endpoint was time to recurrence of any mood episode for risperidone LAI versus placebo in the double-blind period (Kaplan–Meier analysis). Additional efficacy endpoints included Young Mania Rating Scale, Montgomery–Asberg Depression Rating Scale and Clinical Global Impression. During the double-blind period, dosing was fixed at patients' final open-label dose (25 mg, 66% of patients; 37.5 mg, 31%; 50 mg, 4%). The primary outcome demonstrated a median time to mood episode recurrence of 198 day in the placebo arm, whereas the median was not reached in the risperidone LAI arm (p=0.057). Time to recurrence of any mood episode was significantly longer with risperidone LAI versus placebo (log-rank test stratified by region only, p=0.031); the difference was significant for time to recurrence of elevated mood episodes (p=0.005) but not depressive episodes (p=0.655). Significant improvement of manic symptoms and global condition versus placebo were observed for risperidone LAI, with no evidence of worsening of depression. In conclusion, risperidone LAI significantly delayed time to recurrence of elevated mood episodes, with a safety profile consistent with previous studies.
Keywords: Atypical antipsychotics; Bipolar disorder; Long-acting injectable; Placebo-controlled trial; Recurrence; Risperidone;

Reduced plasma levels of asymmetric Di-Methylarginine (ADMA) in patients with alcohol dependence normalize during withdrawal by Helge Frieling; Viktoria Leitmeier; Mani Haschemi-Nassab; Johannes Kornhuber; Mathias Rhein; Stefan Bleich; Thomas Hillemacher (836-840).
Asymmetric Di-Methylarginine, an endogenous inhibitor of nitric oxide synthase, is increasingly recognized as vascular risk factor. Elevated ADMA levels have been described not only in ‘typical’ vascular diseases like congestive heart failure, artherosclerosis and diabetes but also for major depression and Alzheimer's disease.As homocysteine increases ADMA levels and elevated homocysteine serum levels are present in patients with alcohol dependence, the aim of the present study was to examine plasma ADMA levels in patients with alcohol dependence during withdrawal.ADMA and homocysteine levels were measured in the plasma from 42 patients drawn at baseline, on day 1, day 3 and day 7–10 of inpatient detoxification treatment. Measurements were compared against 32 healthy controls. We found significantly lower levels of ADMA in patients at baseline and on day 1 and 3, while no differences were present at the end of treatment. Plasma ADMA levels significantly increased during withdrawal. We found no association between homocysteine and ADMA levels.Our finding of reduced ADMA levels in actively drinking alcohol dependent patients is in apparent contrast to other findings regarding cardiovascular risk factors in alcoholism. However an influence of alcohol on arginine metabolism may help explain the so called ‘French paradox’.
Keywords: ADMA; Homocysteine; Alcoholism; French paradox; Alcohol withdrawal; Nitric oxide;