Actu Psychiutr Scand 1997: 96: 134-141 Copyright 0 Munksgaard 1997 Printed in UK - all rights reserved ACTA PSYCHIATRICA SCANDINAVICA ISSN 0001-690X Double-blind randomized controlled study of the efficacy and tolerability of two reversible monoamine oxidase A inhibitors, pirlindole and moclobemide, in the treatment of depression Tanghe A, Geerts S, Van Dorpe J, Brichard B, Bruhwyler J, GCczy J. Double- A. Tanghe', S. Gee&, J. Van blind randomized controlled study of the efficacy and tolerability of two Dorpe3, B. Brichard4, J. Bruhwyler4, reversible monoamine oxidase A inhibitors, pirlindole and moclobemide, in J. Geczf the treatment of depression. 'Groepspraktijk voor Psychotherapie, Biologische Acta Psychiatr Scand 1997: 96: 134-141. 0 Munksgaard 1997. Psychiatrie en co - Therapie. Male hoeklaan. 13, Bruges, 'Academic Hospital St Lucas. Dept of Psychiatry, Assebroek [Bruges). 3Academic Hospital Heilig Hart, Asse and 4Therabel Research S.A.. The aim of this double-blind randomized study was to compare the efficacy Brussels. Belgium and the tolerability of moclobemide (300-600 mg daily) and pirlindole (150-300 mg daily), two reversible inhibitors of MAO-A (RIMAs), in the treatment of depression. In total 116 patients were included in the trial, 111 patients (52 patients on pirlindole and 59 patients on moclobemide) were evaluable for efficacy and safety, and 77 patients completed the whole study (42 days of administration). Both treatments produced highly significant improvements in the Hamilton Depression Rating Scale (HDRS) score, the Hamilton Anxiety Rating Scale (HARS) score and the Montgomery-Asberg Rating Scale (MADRS) score from day 7 to day 42. The pattern of development of the three scores in the two groups did not differ significantly. After 42 days of treatment, an improvement of >50%0 in the HDRS score was noted in 80% and 67% of patients in the pirlindole Key words: major depression; antidepressant; and moclobemide groups, respectively. A total of 30 (58%) patients on monoamine oxidase inhibitor; pirlindole; pirlindole and 33 (56%) patients on moclobemide experienced side-effects moclobemide that were considered to be possibly or probably related to the medication. Jacques Bruhwyler, Research, Development and The differences between the two drugs were non-significant for all types of Biostatistics, Therabel Research S.A., Rue Van side-effect, with the exception of dry mouth and tachycardia, which were Ophem 110. 1180 Brussels. Belgium significantly more frequent with moclobemide. Accepted for publication December 14, 1996 Introduction Until recently, monoamine oxidase inhibitors than tricyclic antidepressants (1).During the 1980s, (MAOIs) were regarded as 'secondary drugs' for controlled studies have provided evidence that the treatment of depression relative to the tricyclic MAOIs are very effective in the treatment of so- antidepressants for three main reasons. First, hepa- called therapy-resistant depressions, atypical depres- totoxicity was a problem with the early MAOIs, sions, phobias, panic disorders and anxiety states such as iproniazid. The second, and probably most (2, 3). In addition, low cardiotoxicity, absence of important, reason was the cheese effect, i.e. the anticholinergic side-effects, and mood-activating interaction between MAOIs and foods rich in properties may be factors in favour of MAOIs as tyramine, leading to hypertensive episodes with compared to tricyclic antidepressants (4). eventual serious consequences, including cerebral The discovery of two distinct subtypes of MA0 haemorrhage or even death. Thirdly, in true depres- (MAO-A and MAO-B), and the resulting develop- sions, MAOIs were considered to be less efficacious ment of reversible (i.e. short-acting) and selective 134 Pirlindole vs. moclobemide in major depression inhibitors of MAO-A has led to a resurgence of Depression Rating Scale (HDRS) was also interest in the use of MA0 inhibitors for the required. Patients who had been receiving lithium treatment of depression (5). By competitively and for more than 2 months and who showed a stable selectively inhibiting MAO-A, the enzyme pri- lithium plasma concentration were eligible for marily responsible for deamination of those inclusion in the study, and could continue to receive monoamines (noradrenaline and serotonin) that their treatment. The exclusion criteria were as are implicated in the aetiology of depression, while follows: psychotic disorders, other bipolar disorders leaving MAO-B unaffected, these compounds can (DSM-111-R 296.4, 296.6, 296.7), high risk of be expected to combine antidepressant activity suicide, treatment-resistant depression, defined as with a reduced risk of hypertensive crisis (6). the absence of response of the actual episode Among the new selective reversible inhibitors of to two or more antidepressant drugs prescribed MAO-A (RIMAs), moclobemide has certainly over a period of 21 days at a sufficient dose, been the most well studied (7, 8). Its efficacy and equivalent to at least 150 mg amitriptyline daily, safety have largely been demonstrated in placebo- known drug or alcohol addiction, pheochromo- controlled studies (1, 9) and in comparative trials cytoma, organic brain disorder, epilepsy, hepatic vs. reference therapies (1, 10-14), and have been tests severely disturbed (one enzyme > 3 times confirmed in clinical use (15). normal range), renal insufficiency (clearance of Pirlindole is a tetracyclic compound that has creatinine d 80 ml mi&), pregnancy, lactation or been characterized as a potential antidepressant lack of efficient contraception. The study was car- drug in preclinical studies (16-18), and in which ried out in accordance with the Declaration of interest has been shown due to its marked selec- Helsinki amended in Tokyo, Venice and Hong tivity as a RIMA (19). In clinical trials, the efficacy Kong. Patients were included after they had given and safety of pirlindole have been demonstra- informed consent. The study protocol was approved ted compared to placebo (20) and to reference by the Ethical Committee of the Vrije Universiteit standard drugs such as maprotiline (21), imipra- of Brussels. mine (22, 23), amitriptyline (23-25), desipramine (26) and mianserin (27). Drug treatment The aim of the present study was to compare the efficacy and the tolerability of pirlindole Patients received capsules, identical in appearance, and moclobemide according to a double-blind containing either a 150-mg moclobemide tablet or randomized controlled design in patients suffering a 75-mg pirlindole tablet and filled up with lactose. from monopolar (single (DSM-111-R 296.2) or Two capsules (one in the morning and one in the recurrent (DSM-111-R 296.3) episodes) or bipolar evening) had to be taken on days 1 and 2, three (DSM-111-R 296.5) depression. capsules (one in the morning, one at midday and one in the evening) from day 3 to day 6, three to four capsules (one or two in the morning, one at Material and methods midday and one in the evening) from day 7 to day 20, and two to four capsules (one in the morning Study design and one in the evening, or one or two in the This was a multicentre, double-blind, prospective morning, one at midday and one in the evening) clinical trial, conducted by 6 investigators in 5 from day 21 to day 42. The relationship between Belgian centres (Bruges, Oudenaarde, Assebroek, medication intake and meals was not specified, and Kortrijk and Asse). Two randomized parallel patients were not required to avoid tyramine-rich groups of subjects (in-patients and out-patients, food. The use of concomitant psychotropic medi- respectively) were treated with either pirlindole or cation was prohibited, with the exception of lithium moclobemide. The study drug was administered for for patients on a previously established regimen 6 weeks, immediately following a washout period and lormetazepam (at a maximum dose of 2 mg of 7 days. daily) or dipotassic clorazepate (at a maximum dose of 50 mg daily for in-patients only), if this was judged to be necessary by the investigator. Subjects The subjects eligible for the study were men or Assessments women aged between 18 and 65 years, either in-patients or out-patients, who fulfilled the DSM- Assessments were made at baseline (day 0) and on 111-R criteria for unipolar (single (296.2) or recurrent days 7,14,21,28 and 42 of treatment. Efficacy was (296.3) episodes) or bipolar (296.5) depression. evaluated on the basis of scores on the 17-item A total score of 318 on the 17-item Hamilton Hamilton Depression Rating Scale (HDRS), the 135 Tanghe et al. Hamilton Anxiety Rating Scale (HARS) and the in the moclobemide group) discontinued the treat- Montgomery-Asberg Depression Rating Scale ment prematurely. In all, 77 patients completed the (MADRS). A therapeutic index was also deter- 6-week treatment period (35 patients on pirlindole mined. It was a 16-point scale that combined an and 42 patients on moclobemide) (Table 1). The evaluation by the doctor, both on the therapeutic two groups did not differ significantly from each effects (4-point scale: excellent, good, poor, or no other with regard to the majority of baseline charac- effect) and on the adverse effects (4-point scale: teristics. However, the HDRS and the MADRS none, light, important, or side-effect nullifying the scores were significantly higher (P<0.05 and P<O.Ol, therapeutic benefit). Compliance was judged to be respectively) in the pirlindole group (Table 2). satisfactory when at least 80% of the tablets had been taken. Tolerability was evaluated on the basis Concomitant medication of the number and severity (mild, moderate or severe) of adverse events, the likelihood of a causal There were no significant differences between relationship (possible, probable, definite or un- the two groups with regard to prescription of known) to study drugs, vital signs (supine and lormetazepam or dipotassic clorazepate (P>0.05, standing blood pressure, and heart rate), body Chi-square test). On day 7, 75% and 71% of weight, laboratory values and ECG obtained at patients were taking lormetazepam in the pirlin- baseline and after 42 days of treatment.