Moclobemide Response in Depressed Patients: Association Study with a Functional Polymorphism in the Monoamine Oxidase A Promoter

Monoamine oxidase A (MAO-A) is one of the key enzymes in the metabolism of monoaminergic neurotransmitters, which supposedly play an important role in the etiology of affective disorders. Moreover, MAO-A inhibitors such as moclobemide are effective in drug therapy strategies for depressive syndromes [2] [10]. Moclobemide brought about observable patient-to-patient differences in response and side effects, which may be related to the patients’ genetic make-up.

Recently, a novel functional repeat polymorphism consisting of a 30-bp repeated sequence, present in 3, 3.5, 4, or 5 copies in the promoter of the MAO-A gene on chromosome X, has been reported [8]. The longer alleles (3.5, 4, 5) proved to be more active than the shorter one (3) in luciferase reporter gene assays [3]. Two studies have suggested that the longer alleles might be associated with major depression [9] and panic disorder [3] in female patients. The authors concluded that an increased MAO-A activity might be regarded as a risk factor for major depression and panic disorder in females.

The underlying aim of this study was to analyze whether the length of alleles may be associated with clinical response to the MAO-A inhibitor moclobemide. The study included sixty-two patients (14 males, 48 females) with major depressive disorders. Lifetime consensus diagnoses according to DSM-IV criteria [1] were conducted by trained psychiatrists on the basis of a multi-dimensional phenotype characterization inventory including a personal structured interview (SADS-L) [4], family history method (FISC) [5], OPCRIT documentation [6], and a systematic review of medical records. The mean age was 52 years (SD = 13 years), the mean age of onset was 41 years (SD = 13 years). After a detailed description of the study, written informed consent was obtained from all patients prior to examination.

Response to moclobemide was measured weekly with the following instruments: Three clinician-rated scales (HAMD, HAMA, GAS) and three patient-rated scales (SDS, SAS, Bf-S/Bf-S’) to control for doctor-patient differences in the perceived outcome. Wherever possible, ratings were carried out for 42 days, thus assessing treatment response at 6 different points of time. Following the last interview, one additional follow-up assessment was performed two months later. The baseline HAMD score was assessed, and those patients who dropped to 50 % or less of their baseline score were categorized as moclobemide responders. During this assessment period, all patients received a minimum dose of 300 mg moclobemide per day, most of them receiving 450 - 600 mg moclobemide per day. Genotyping was performed according to the protocol of Deckert et al. [3].

The mean value of the baseline HAMD score in the patient sample was 21.6 (SD = 6.4). According to our response criteria, 29 patients were moclobemide responders, whereas 33 were moclobemide non-responders.

Statistical analyses were performed with the ‘last observation carried forward’ (LOCF) method; potential significance was analyzed using an ANOVA/ANCOVA design with the software package Stata 6.0. The allele groups were formed on the basis of functional characterization. The short allele group contained all alleles with 3 repeats; the long allele group contained all alleles with 3.5, 4 and 5 repeats.

The quantitative dependent variable was defined as the differences for the various scales from time 0 to the final observation, assuming possible alternative definitions in treatment response. These variables were weighted by a covariate, which was assigned to consider the interview scores of the first interview (day 0) in order to control for the varying score levels at which the various individuals entered the study. The various alleles represented the independent variables (factors). Genotypic analyses were performed under two coding models for genotypes: ‘Model 1’ considering the actual genotypes for males and females, and ‘Model 2’ transforming the actual genotypes into a binomial variable (0,1) with the same rule as for alleles collapsing into two groups. Allelic analyses were performed by using the actual allele distributions in males and females. However, no significant association was found between a given allelic or genotypic distribution and degree of response to moclobemide in any of the six rating scales. (Allelic distribution: p-values from 0.25 - 0.79, genotypic distribution (model 1 and 2): p-values from 0.29 - 0.97).

Several reasons may account for this negative finding: Firstly, considering the limited sample size of our study (48 females and 14 males), a possible effect might have been missed. Secondly, although all patients were treated for major depression, major depression is likely to be heterogeneous, and its current classification remains controversial [7]. Therefore, response to medication may be influenced by the underlying specific depressive ‘subtype’ or by other factors. Thirdly, the contribution of a single variant to the total response variance is likely to be limited. Although we measured plasma levels at least twice to check patient compliance, another limitation of this study arises from the fact that we have not measured the enzymatic activity of MAO-A or MAO-B. Thus, we lacked an objective measure for the real MAO-inhibition. Given the baseline HAMD score of 21.6 (SD = 6.4), one may argue that our patients were affected by relatively minor depressions; therefore, a low response rate would potentially prevent the detection of a response. On the other hand, our patient sample has been divided into two groups, each of them reaching nearly 50 % as regards the response to moclobemide (46.8 % responders and 53.2 % non-responders), which should represent a sufficient response rate to perform appropriate statistical analyses. In conclusion, although our results cannot definitively exclude the involvement of the MAO-A gene in clinical response to the MAO-A inhibitor moclobemide, they suggest that this influence is not likely to have a major effect.

References

• 1 American Psychiatric Association 1994: Diagnostic and statistical manual of mental disorders. Fourth edition. American Psychiatric Association Washington DC; 1994
• 2 Casacchia M, Carolei A, Barba C, Frontoni M, Rossi A, Meco G, Zylberman M R. A placebo-controlled study of the antidepressant activity of moclobemide, a new MAO-A inhibitor.  Pharmacopsychiat. 1984;  17 122-125
• 3 Deckert J, Catalano M, Syagailo Y V, Bosi M, Okladnova O, Di Bella D,. et al . Excess of high activity monoamine oxidase A gene promoter alleles in female patients with panic disorder.  Hum Mol Genet. 1999;  8 621-624
• 4 Endicott J, Spitzer R L. A diagnostic interview: The Schedule for Affective Disorders and Schizophrenia.  Arch Gen Psychiatry. 1978;  35 837-844
• 5 Mannuzza S, Fyer A J, Endicott J, Klein D F. Family Informant Schedule and Criteria (FISC). New York; Anxiety Disorder Clinic, New York State Psychiatric Institute 1985
• 6 McGuffin P, Farmer A, Harvey I. A polydiagnostic application of operational criteria in studies of psychotic illness. Development and reliability of the OPCRIT system.  Arch Gen Psychiatry. 1991;  48 764-770
• 7 Parker G. Classifying depression: should paradigms lost be regained?.  Am J Psychiatry. 2000;  157 1195-1203
• 8 Sabol S Z, Hu S, Hamer D S. A functional polymorphism in the monoamine oxidase A gene promoter.  Hum Genet. 1998;  103 273-279
• 9 Schulze T G, Müller D J, Krauss H, Scherk H, Ohlraun S, Syagailo Y V,. et al . Association between a functional polymorphism in the monoamine oxidase A gene promoter and major depressive disorder.  Am J Med Genet (Neuropsych Genet). 2000;  96 801-803
• 10 Stassen H H, Angst J, Delini-Stula A. Fluoxetine versus moclobemide: cross-comparison between the time courses of improvement.  Pharmacopsychiat. 1999;  32 56-60