Exp Neurobiol 2011; 20(1): 1-17
Published online March 31, 2011
© The Korean Society for Brain and Neural Sciences
Peter Riederer1* and Gerd Laux2
1Clinic and Policlinic for Psychiatry, Psychosomatic and Psychotherapy, University of Wuerzburg, 97080 Wuerzburg, 2Academic Hospital of Psychiatry, Psychosomatic Medicine, Psychotherapy and Neurology, Gabersee, 83512 Wasserburg a. Inn, Germany
Correspondence to: *To whom correspondence should be addressed.
TEL: 49-931-201-77200, FAX: 49-931-201-77220
Monoamine oxidase inhibitors (MAO-I) belong to the earliest drugs tried in Parkinson's disease (PD). They have been used with or without levodopa (L-DOPA). Non-selective MAO-I due to their side-effect/adverse reaction profile, like tranylcypromine have limited use in the treatment of depression in PD, while selective, reversible MAO-A inhibitors are recommended due to their easier clinical handling. For the treatment of akinesia and motor fluctuations selective irreversible MAO-B inhibitors selegiline and rasagiline are recommended. They are safe and well tolerated at the recommended daily doses. Their main differences are related to (1) metabolism, (2) interaction with CYP-enzymes and (3) quantitative properties at the molecular biological/genetic level. Rasagiline is more potent in clinical practise and has a hypothesis driven more favourable side effect/adverse reaction profile due to its metabolism to aminoindan. Both selegiline and rasagiline have a neuroprotective and neurorestaurative potential. A head-to head clinical trial would be of utmost interest from both the clinical outcome and a hypothesis-driven point of view. Selegiline is available as tablet and melting tablet for PD and as transdermal selegiline for depression, while rasagiline is marketed as tablet for PD. In general, the clinical use of MAO-I nowadays is underestimated. There should be more efforts to evaluate their clinical potency as antidepressants and antidementive drugs in addition to the final proof of their disease-modifying potential. In line with this are recent innovative developments of MAO-I plus inhibition of acetylcholine esterase for Alzheimer's disease as well as combined MAO-I and iron chelation for PD.
Keywords: selegiline, rasagiline, moclobemide, phenelzine, tranylcypromine
Monoamine oxidase (MAO) inhibitors (MAOIs) at present can be classified into 3 types:
The clinical indications and efficacy of the MAOIs are established for disorders as follows:
Other potential therapeutic uses and indications can be smoking cessation, attention deficit hyperactivity disorder and cognitive deficits in dementia for moclobemide (Chan-Palay, 1992; Berlin et al., 1995).
There is overwhelming evidence that selegiline has beneficial effects on motoricity and motorfluctuations as shown already in the first clinical trial (Birkmayer et al., 1975) and in follow-up clinical studies as summarized by Gerlach et al. (2007).
In all these and additional clinical studies (eg Rabey et al., 2000; Thebault et al., 2004; Rascol, 2005; Biglan et al., 2006; Siderowf and Stern, 2006) it could be shown that rasagiline is well tolerated, safe, improves motor symptoms, prevents motor complications in PD, has beneficial effects on quality of life parameters, is effective as monotherapy or in adjunctions to L-DOPA-therapy, is beneficial in early and late stages of PD, is safe when combined with all other PD-relevant therapies including COMT-inhibitors and may have disease-modifying properties.
Selegiline and rasagiline are "suizide-inhibitors" which inhibit the enzyme irreversibly and for a rather long time. Therefore only newly synthesized enzyme will recover MAO activity sufficient to metabolize its substrates in an adequate amount. Positron-emission tomography (PET) studies have shown a biological half-life time (HLT) of MAO-B recovery after selegiline- (Fowler et al., 1994) or rasagiline- (Freedman et al., 2005) induced blockade of about 30~40 days. As described earlier (Riederer and Lachenmayer, 2003; Gerlach et al., 2007) PET-studies are only in part suitable to detect the HLT of MAO-inhibitors as they only detect the distribution of radiolabeled inhibitor and its metabolites. In addition the exact kinetic properties underlying these PET-studies have been described only at random (Arnett et al., 1987). In contrast, measurement of PEA, the pure MAO-B substrate, in the urine of healthy individuals after selegiline dosing showed recovery (decline) of this amine concentration to normal values already 2~3 days after withdrawal of selegiline (Clarke et al., 2001). Such data agree with other pharmacological studies as described by Riederer and Lachenmayer (2003) and point to the conclusion that the HLT of MAO-B recovery after irreversible inhibition by selegiline is about 7 days. According to Green et al. (1977) an 80% inhibition of MAO-isoenzymes is necessary to increase concentration of biogenic amines significantly. This means that newly synthesized enzyme in the range of 20 to 30% is sufficient to lose any symptomatic behavioural effect after withdrawal of any MAO-inhibitor. In line with this are studies by Youdim and Tipton (2002) who could not detect selegiline-induced stereotyped behaviour due to PEA increase 4 days after withdrawal of selegiline. At this time MAO-B recovery was already 20%. Taking these findings into account a wash-out phase of two weeks or after selegiline/rasagiline withdrawal is sufficient to avoid any symptomatic effect of these MAO-I's. Therefore, the interpretations of the DATATOP-study and other follow-up trials to demonstrate a "neuroprotective" effect have to be reconsidered. If so, a neuroprotective effect of selegiline cannot be excluded (Riederer and Lachenmayer, 2003; Gerlach et al., 2007). Table 2 (taken in part from Riederer and Lachenmayer 2003 with permission) illustrates the major outcomes of the most important clinical long-term trials with selegiline. While treatment of patients in an early phase of PD with selegiline shows sufficient results regarding motoricity and disease-modification the unconclusive results of the DATATOP-follow-up clinical trials may demonstrate a bias in severity of patients grouping as given in the first follow-up study (Parkinson Study Group, 1996a) or that the patients have non-response based on disease-progression and/or accompanying diseases (Table 3). As there were no final conclusions about the neuroprotective effect of selegiline the development of more efficient clinical trial designs have been investigated including PET-controlled studies as shown with dopaminergic agonists. More recently the "delayed-start study design" has been developed as discussed by D'Agostino (2009). Rasagiline has been the first drug tested in such clinical trial. In fact the ADAGIO-study has given profound evidence for a disease-modifying effect of 1 mg/day rasagiline, while there was not such benefit at a 2 mg/day dosis (Table 4). Although - and in comparison to the negative outcome of a delayed-start-study with pramipexol (Schapira et al., 2009) - there is ample evidence for a significant disease-modification with low dosis rasagiline this clinical outcome has been questioned recently (Ahlskog and Uitti, 2010; Sampaio and Ferreira, 2010; Schwarzschild, 2010). Caslake et al. (2009) conclude in their Cochrane Database report: "MAO-B inhibitors are one option for the early treatment of PD although they have weaker symptomatic effects than levodopa and dopamine agonists. They may reduce the rate of motor fluctuations compared with initial levodopa therapy and may have fewer significant adverse effects than the older agonists but data are too few to provide reliable conclusions".
Neuroprotective, neurorestaurative and disease-modifying effects of selegiline and rasagiline are substantiated by in vitro-as well as in vivo experimental studies (Table 5). There is ample evidence that compounds attached to a propargylamino group induce molecular processes which act against cell death mechanisms like apoptosis, excitotoxicity and oxidative stress. These are non-MAO based mechanisms and they act not only on a functional level but rather on influencing gene expression and protein synthesis. These more recent discoveries are in line with the earlier assumption of neuroprotective/neurorestaurative/disease-modifying properties especially of this chemical class of compounds.
With a prevalence of approximately 40% depression is considered to be the most frequent psychiatric manifestation in PD (Burn, 2002; McDonald and DeLong, 2003; Veazey et al., 2005). Particular clinical features of the depressive symptom profile in patients with PD have been described (Wermuth and Bech, 2006; Brand et al., 2007) and recommendations towards better recognition given (Burn 2002). About 50% of the patients meet the criteria for major depressive disorder (MDD), 50% have minor depression or dysthymia (Chaudhuri et al., 2006). Depression in PD is associated with increased disability and reduced quality of life, the impact of depressive symptoms in early PD especially should not be underestimated (Ravina et al., 2007).
While early studies suggested that MAOIs were not as effective as other antidepressants, more recent studies have demonstrated that, when prescribed in adequate dosages, phenelzine and tranylcypromine are as effective as other antidepressants (Pare, 1985). Table 6 gives a summary of important controlled studies with tranylcypromine. In a recent study tranylcypromine, 60 mg daily, was found effective in the treatment of panic disorder and social anxiety disorder comorbidity (Nardi et al., 2010).
The main indications for the classical irreversible MAOIs are subgroups of depression such as atypical depression, dysthymia or for patients who do not respond to reuptake inhibitors, so-called therapy resistant depressions (Nolen et al., 1994; Paykel, 1995; Bauer et al., 2002).
The therapeutic efficacy of moclobemide has been assessed in numerous controlled studies comparing it with placebo and established antidepressants (for reviews see Laux, 1989; Fitton et al., 1992; Fulton and Benfield, 1996; Laux et al., 2002). An overview of the most relevant trials is given in Table 7.
Large trials in patients with major depression have generally confirmed that the efficacy of moclobemide is equivalent to that of TCAs beside some negative studies. Moclobemide treatment was usually fixed up to 300, 400 or 600 mg/day. It should be noted, however, that the use of dosages of up to 900 mg/d has been established due to clinical experience nowadays particularly in patients with refractory depression (Guentert et al., 1995). A meta-analysis of inpatient studies indicated that higher dosages (>450 mg/day) of moclobemide are needed for full therapeutic effect in patients with severe depression (Angst et al., 1995).
Subgroup analyses of patients revealed moclobemide to be effective in patients with dysthymia or atypical depression especially (Lonnqvist et al., 1995).
Several trials have evaluated moclobemide alone or in combination with TCAs or SSRIs in the treatment of refractory depression (Bakish et al., 1995) reporting positive results. These data are limited and great caution is necessary because of the potential to induce the serotonin syndrome when combining moclobemide with serotoninergic drugs.
Uncontrolled long term follow-up studies demonstrated continued effectiveness of moclobemide over the treatment period of 1 year with more than 60% of patients having continued response (Fulton and Benfield, 1996). The role of MAOIs in maintenance treatment of depression still has to evaluated (Kennedy, 1997).
MAO-B inhibitors like selegiline in high dosage have been used in therapy-refractory depressions probably due to non-selective MAO effects (see review Laux, 1993). Recently, transdermal selegiline has been approved and released in the USA for treatment of major depression with a target dose of 6 mg/24 hours. This dosage and application overcomes the MAO-B selectivity and leads to MAO-A inhibition, seen to be necessary for antidepressive effects. In several controlled studies selegiline transdermal system exhibited significant treatment effects on MDD including core depression symptoms, vegetative symptoms and motor retardation (Frampton and Plosker, 2007; Robinson et al., 2007). A combination of selegiline and 5-hydroxytryptophan has been tested in a pilot study and proved antidepressant efficacy (Mendlewicz and Youdim, 1978).
The scientific knowledge about the treatment with antidepressant drugs among PD patients is nearly missing. A pharmaco-epidemiological study in Danmark showed that persons treated with antiparkinson drugs have higher frequency of antidepressant drug treatment than have controls (Brandt-Christensen et al., 2007). Most authors conclude, that recommendations for the optimal drug treatment of depression in PD are difficult to give (Burn, 2002). Serotonin selective reuptake inhibitors (SSRIs) are given frequently, the benefit of SSRI treatment in PD has not been established, however (Wermuth and Bech, 2006).
Thirty-seven patients with early PD have been treated successfully with tranylcypromine (TCP), parkinsonian symptoms improved slightly, follow-up after 1.5 years on average revealed only slight worsening (Fahn and Chouinard, 1998). No other studies have been reported with TCP so far.
Ten patients with PD have been treated successfully with moclobemide vs. moclobemide with selegiline under tyramine restriction (Steuer and Ballering, 1997). Sufficient data allowing conclusions are missing.
The recommended initial dosage of tranylcypromine is 10~20 mg/day, of moclobemide 300 to 450 mg/day, given in 2 to 3 divided doses. Subsequent dosage increase to a maximum of 60 mg/day, 900 mg/day respectively, are made as clinically indicated (Beckmann and Laux, 1991; Fitton et al., 1992). Clinical experience and studies in the last years clearly have shown the necessity of higher dosage of moclobemide pointing out the lack of early sufficient dose-finding studies. Recommended daily doses of phenelzine are 30 to 90 mg, of isocarboxazid 30 to 60 mg, respectively.
Dietary restrictions are essential for tranylcypromine (tyramine-rich food), unnecessary for moclobemide taken at the end of a meal. A 2 week wash-out period is required for switching between tranylcypromine and other classes of antidepressants, not between moclobemide and other antidepressants.
The target dose of selegiline transdermal (not released in Europe so far) is 6 mg/24 hours.
The most frequent adverse effects of irreversible MAOIs are orthostatic hypotension, sleep disturbances and nervousness/agitation (Remick et al., 1989). The mostly limiting factor in the use of these MAOIs is the potential for dangerous interactions with tyramine-rich foods and sympathomimetic and serotoninergic substances. Therefore, prescription is only possible to patients being strongly compliant with dietary restrictions. In the case of RIMAs like moclobemide there is no need for dietary restrictions. For "selegiline transdermal system" tyramine dietary restrictions are not needed. The incidence of the most frequently reported adverse effects from patients receiving MAOIs are summarised in Table 7.
Insomnia and impotency in men have been the most frequent side effects of TCP in patients with early PD (Fahn and Chouinard, 1998).
Unlike nonselective, irreversible MAOIs and tricyclics, moclobemide has little effect on the cardiovascular system and lacks anticholinergic properties associated with TCAs (Moll et al., 1994; Fulton and Benfield, 1996). In almost all controlled clinical studies comparing moclobemide with TCAs moclobemide showed clearly superior tolerability (Versiani et al., 1990; Fitton et al., 1992; Fulton and Benfield, 1996). Headache, insomnia and agitation were the only one side effects being reported more frequently with moclobemide compared to re-uptake inhibiting antidepressants. Dizziness and nausea have been noticed additionally. No systematic changes in blood pressure were observed with moclobemide, whereas increases in both blood pressure and pulse were recorded for tranylcypromine (Laux et al., 1996). Body weight gain has been observed with phenelzine and isocarboxazid, no clearcut cases with tranylcypromine or moclobemide have been reported (Cantu and Korek, 1988). Compared to SSRIs moclobemide showed fewer gastrointestinal adverse effects and sexual dysfunction have not been reported with moclobemide (Philipp et al., 1999).
Regarding cognitive functions data in favour of moclobemide compared to other antidepressants are available: Behavioural toxicity assessed by choice reaction time is very low or missing, psychomotor functions seem not be influenced negatively perhaps even improved (Hindmarch et al., 1992).
The principal side effects of selegiline transdermal were local dermal reactions and (dose-related) insomnia (Robinson and Amsterdam, 2008).
After ingestion of up to 20 g of moclobemide no fatalities were observed, so moclobemide can be considered as a 'safe' antidepressant (DeJonghe and Swinkels, 1992; Chen and Ruch, 1993). Several fatal overdoses have been reported when moclobemide was combined with serotoninergic antidepressants like citalopram, clomipramine or fluoxetine due to occurrence of a serotonin syndrome (Neuvonen et al., 1993). In contrast, irreversible MAOIs like tranylcypromine must be regarded as less safe regarding to the fatal toxicity index (Henry et al., 1995).
Adverse reactions as seen with other dopaminergic drugs, like nausea, vomiting, orthostatic hypotension, somnolence, hallucinations and dyskinesias are tolerable in most cases. Vomiting was noteable at 1 mg/day in the PRESTO-study, as was the occurrence of dyskinesias but such adverse reactions have not been described in other clinical trials like the LARGO-trial. Cognitive and behavioural symptoms of PD are not changed/worsened at 1 mg/day rasagiline (Elmer et al., 2006).
Although interactions may be suggested when MAO-B inhibitors are combined with COMT-inhibitors, such adverse reactions have not been reported to be of relevance. In fact rasagiline is effective and well tolerated in PD patients with L-DOPA induced motor fluctuations receiving entacapone (Elmer et al., 2006).
Nevertheless, to avoid further such discussion a sublingual galenic form of selegiline treatment has been developed named Zydis-selegiline (Clarke et al., 2003a; 2003b). This "melting-tablet" avoids a first-pass effect and therefore a significant break-down to L-amphetamine and L-metamphetamine in the range of 90%. A reduction of the daily dosis of 10 mg selegiline to 1, 25 mg selegiline is advised. The bioavailability of selegiline when given as "melting-tablet" is more homogene and better reproduceable compared to the peroral type of application (Clarke et al., 2003a; 2003b). In addition a transdermal galenic form of selegiline has been developed for the treatment of major depression again reducing selegilines first-pass biotransformation (Frampton and Plosker, 2007; Robinson et al., 2007).
Melting tablet and transdermal selegiline avoid first-pass metabolism, cause higher drug availability in MAO-B preferring organs and reduce the concentration of metabolites. Increased drug concentration may cause significant inhibition of both MAO-B and -A in brain but not in the periphery. This explains selegilines antidepressant activity when combined with 5-hydroxytryptophan without the cheese-effect and without the serotonin syndrome (Mendlewicz and Youdim, 1978).
Under clinical conditions selegiline is not more "toxic" than other dopaminergic treatment strategies. In fact the side effect and adverse reaction profile of selegiline has been evaluated as being well tolerated and "mild" compared to other PD treatment strategies, eg. orthostatic reactions have been noteable in only 3, 7% of patients (Reichmann et al., 2002).
It should be mentioned that L-amphetamine is in clinical use for the treatment of attention-deficit-hyperactivity syndrome (ADHD) without major side effects. In line with this are several studies as summarized by Reidenberg (1994) showing no abuse liability of selegiline.
The contribution of desmethylselegiline (DMS) to the effects of selegiline has been underestimated so far. DMS is a weak inhibitor of MAO-B and seems to have glutathione related properties (Heinonen, 1997; Heinonen et al., 1997).
In contrast to selegiline,
In the recommended therapeutic dosis of 1 mg/day rasagiline does not potentiate the tyramine induced cheese-effect (de Marcaida et al., 2006). This is in agreement with earlier findings in rats and cats (Finberg et al., 1981; Chen and Swope, 2005; Finberg et al., 2006).
Coadministration of SSRIs and other serotoninergic substances to tranylcypromine is contraindicated due to the risk of serotonin syndrome. With moclobemide great caution should be exercised with this combination. Severe, sometimes fatal, interactions between nonselective, irreversible MAOIs, moclobemide as well as selegiline and pethidine and dextromethorphan have been reported. Indirectly acting sympathomimetics (tyramine, ephedrine, pseudoephedrine) should be administered with caution in patients treated with MAOIs (Dingemanse, 1993).
The elimination of moclobemide is significantly reduced by cimetidine making dose adjustment necessary.
Although selegiline has been reported to have antidepressant actions when combined with 5-hydroxytryptophan (Mendlewicz and Youdim, 1978) MAO-B-inhibitors may not be safe enough to avoid the "serotonin-syndrome" when given in adjunct to serotonin-reuptake inhibitors (SSRIs) to treat depression and anxiety in PD patients. However, there is a low risk (0, 24%) to develop the "serotonin-syndrome" in selegiline treated patients (Richard et al., 1997). Also the results of a small number of PD patients treated with tricyclics and SSRIs plus rasagiline in the TEMPO-, PRESTO- and LARGO-studies do not give evidence for the "serotonin-syndrome", the population incidence of serotonin toxicity in those patients has a 9, 5% probability of being less than 1, 2% (Pannisset et al., 2007; Montgomery and Panisset, 2009). However, a clinical trial with a biostatistic power relevant to give a definite answer to the problem is missing. Therefore MAO-B inhibitors have not to be combined with drugs stimulating the serotoninergic system.
Meperidin plus selegiline has been reported to be dangerous as it causes severe hypertension (Zornberg et al., 1991).
Selegiline transdermal without tyramine restriction revealed no acute hypertensive reactions in trials, until more data are available, foods that are rich sources of tyramine should be avoided, however.
As described by Chen and Swope (2005) CYP1A2 inhibitors (cimetidine, fluvoxamine, ciprofloxacin) increase the area under curve of rasagiline, while CYP1A2 inducers like omeprazol decrease it, as it may decrease in heavy smokers.
Minor time-dependent mechanism-based inhibition of CYP2D6 has been described for selegiline and moclobemid in experimental designs; the significance in human beings remains to be investigated (Polasek et al., 2006). For selegiline interactions with CYP 3A4 and CYP 2E1 may be relevant too.
While selective MAO-B-inhibitors demonstrate a significant benefit in PD and improve motoricity and motor fluctuations eventually causing disease-modification in general MAOIs to date play a subordinate therapeutic role in the treatment of depression in PD compared to SSRIs or reboxetine, a selective noradrenergic reuptake inhibitor. This attitude is primarily due to the risk of adverse effects and compliance problems (e.g. dietary restrictions, interaction problems) with MAOIs. Moclobemide according to its favourable adverse effect, interaction and toxicity profile can be used in depressed Parkinson patients when activating properties are necessary, especially.
Overall, the clinical use of MAOIs may be limited by the possible adverse effects of restlessness and insomnia - but see rasagiline. As far as long term and prophylactic treatments are concerned, the place of MAOIs still has to be verified as antidepressant or antidementive drugs in PD (Tariot et al., 1987; Sano et al., 1997; Filip and Kolibas, 1999; Sterling et al., 2002; Elmer et al., 2006; Carageorgiou et al., 2008). Controlled studies are urgent needed for final evaluation and recommendations.
It would be worthwile to perform clinical studies to demonstrate the capacity of MAO-B inhibitors as antidepressants and antidementive drugs. New substances like ladostigil (MAOA/B inhibitor; AchE-inhibitor) or M30 and HLA-20 (MAO-A/B inhibitor, iron-chelator) and M30P (MAO-A/B inhibitor, carbonate cholinesterase inhibitor) give future aspects of using MAO-I's in a variety of clinical indications (Weinreb et al., 2010). The pharmacological properties of aminoindan have to be elucidated as they may be neuroprotective (Bar-Am et al., 2010).
Head-to-head clinical trials are necessary to demonstrate disease-modification using improved delayed-start-study designs with selegiline and rasagiline. Otherwise no objective comparison can be made between the neuroprotective/neurorestaurative properties of selegiline and rasagiline.
Aminoindan has to be tested in vivo in order to get insight into its anti-parkinsonian efficacy.
The many molecular biological and -genetic data evolved from in-vitro studies have to be confirmed in in-vivo experiments to prove relevance in human beings. That other MAO-Is, even reversible ones might well be important in neuroprotection/disease modification is a just ongoing important issue. In addition, the respective role of MAO-A has to be elucidated in more detail.
Table 1. Monoamine oxidase inhibitors
Table 2. Nueroprotection in Parkinson's disease by selegline (prospective, randomized, double-blind, placebo-controlled studies)
From Riederer and Lachenmayer 2003.
Table 3. Datatop follow-up-clinical studies
Table 4. Summary of rasagiline clinical studies
*All studies were randomised, multi-centre, parallel-group and double-blind.
ADL: activities of daily living, CGI: clinical global impressions scale, LARGO: lasting effect in adjunct therapy with rasagiline given once daily, OFF: poor symptom response, ON: good symptom reponse, PD-QUALIF: parkinson's disease quality of life, PRESTO: Parkinson's rasagiline efficacy and safety in the treatment of OFF, TEMPO: trial of etanercept and methotrexate with radiographic patient outcomes, UPDRS-ADL: unified parkinson's disease rating scale-activities of daily living. From Rascol 2005.
Table 5A. Actions of propargylamine derived MAO-B-I
Table 5B. Non-MAO related actions of selegiline and rasagiline on molecular paramters
Table 5C. Non-MAO related actions of selegiline and rasagiline on molecular parameters
Table 6. Randomised, double-blind controlled studies with tranylcypromine (TCP) in the treatment of depressive disorders (from Laux et al., 2002)
Legend: =indicates equivalent to, >indicates more effective than references in this table are given in detail in Laux et al., 2002.
C: comparator, =indicates equivalent to, >indicates more effective/better tolerated than. *melancholic subtype.