Efficacy

In 1985 [23], published by then well-established researchers in the field, including W. Birkmayer, J. Knoll, P. Riederer and M. Youdim, a long-term study (9 years) showed a significant increase in life expectancy resulting from the addition of selegiline to Madopar® (co-beneldopa). Even as an open-label, nonrandomized trial, the results were remarkable. Earlier, in 1983 [24, 25] the discovery of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) toxicity had been published, which produced selective loss of dopaminergic cells in the substantia nigra. When it was later shown that selegiline could inhibit the dopaminergic neurotoxicity of MPTP in monkeys [26], several clinical trials were prepared.

In 1993, Myllylä showed significant symptomatic benefit for selegiline as monotherapy in early PD compared with placebo [27]. The study also indicated that selegiline could significantly delay the need for supplementary levodopa treatment.

Neuroprotective potential

As early as 1987, the Parkinson Study Group initiated the Deprenyl and Tocopherol Antioxidant Treatment On Parkinson’s disease (DATATOP) study. This study was the first clinical study that tried to detect a neuroprotective effect in PD. A total of 800 patients with early PD were randomized in a 2 × 2 factorial design to selegiline and/or the antioxidant vitamin E, tocopherol or placebo, and were evaluated clinically over 18+months. There were no beneficial effects of tocopherol in comparison with placebo, but selegiline significantly delayed the onset of disability requiring levodopa by about 9 months, representing a 57% risk reduction of starting levodopa during the first year of treatment [28].

An extension study was carried out with 368 patients from the original DATATOP study who were on levodopa who agreed to continuing the selegiline treatment or changing to matching placebo under double-blinded conditions. During an average of 2 years follow-up, the patients treated with selegiline showed slower motor decline in terms of Unified Parkinson’s Disease Rating Scale (UPDRS) scores, and total, activities of daily living, motor and on–off motor fluctuations or freezing of gait, but significantly more dyskinesias [29].

To further control for a possible symptomatic effect, a prospective, longitudinal, double-blinded controlled study of the effect of selegiline and levodopa on the progression of the signs and symptoms of PD over 14 months was performed [30]. A total of 101 untreated patients with mild PD were randomized to one of four treatment groups: group 1, selegiline and Sinemet® (carbidopa/levodopa); group 2, placebo-selegiline and Sinemet®; group 3, selegiline and bromocriptine; and group 4, placebo-selegiline plus bromocripitine. The primary endpoint was a change in motor score between an untreated baseline visit and an untreated final visit 2 months after withdrawal of the study drugs, i.e. selegiline or its placebo, and 7 days after withdrawal of Sinemet® or bromocriptine. The decline measured by UPDRS score with this study design was used as an index of disease progression. Placebo-treated patients deteriorated by 5.8 ± 1.4 points for the total UPDRS score and the deprenyl-treated patients by 0.4 ± 1.3 points (P < 0.001). The authors concluded that “these findings are not readily explained by the drug’s symptomatic effects and are consistent with the hypothesis that selegiline has a neuroprotective effect.” However, it is possible that the washout period was too short, so that a prolonged symptomatic effect may complicate the explanation of a neuroprotective effect.

In the MONOCOMB (Use of Selegiline as Monotherapy and in Combination with Levodopa in the Management of Parkinson’s Disease) study [31–33], previously untreated patients with idiopathic PD (n = 157) were randomized to receive selegiline 10mg/day or placebo until levodopa was required; experimental medication was then withdrawn for 8 weeks. Patients were then randomized to levodopa (50mg/day, titrated in 50mg/day increments to 150mg/day) plus either selegiline or placebo. Treatment was continued until patients required additional antiparkinsonian therapy or up to 7 years after initial randomization. The impact of selegiline monotherapy on time to the start of levodopa therapy was first investigated, followed by a comparison of the progression of PD in patients treated with individualized levodopa plus selegiline or placebo. Selegiline significantly delayed the time when levodopa therapy became necessary during the monotherapy phase by more than 4 months, even though the mean total UPDRS scores at the time of initiation of levodopa were similar in both groups. In the combination therapy phase of the study, the use of selegiline as an adjunct to levodopa demonstrated a lower total consumption of levodopa. Patients treated with selegiline plus levodopa also exhibited a distinct (P = 0.005) slowing in the anticipated increase in the UPDRS scores over time, as, for example, shown by a mean UPDRS total score after 5 years that was 10 points lower than in patients on levodopa and placebo.

The MONOCOMB trial is among the largest and longest-duration placebo-controlled studies to report experience with selegiline monotherapy in the early phase of PD. The results of the MONOCOMB trial confirm that selegiline is effective in retarding the progression of early PD, that it has levodopa-sparing qualities in more advanced disease, and that it is reasonably well tolerated in long-term use.

Two meta-analyses in 2004 and 2005 [34, 6] addressed the effectiveness of MAO-B inhibitors in reducing the rate of progression and effective and safe symptomatic treatment.

Ives et al. [34] included 3525 people with PD in 17 randomized trials: 13 trials on selegiline, three on lazabemide (RO19-6327, a short-acting, reversible, highly selective inhibitor of MAO-B which, unlike selegiline, is not metabolized to active compounds) and one on rasagiline therapy. The result of the lazabemide studies were compatible with the results of the other therapies. All trials showed significantly improved scores in favor of selegiline versus controls, i.e. MAO-B inhibitors reduces disability, the need for levodopa and the incidence of motor fluctuations without significant side effects or increased mortality.

The Cochrane review on MAO-B inhibitors in 2005 [6] found similar results in 2514 PD patients from 12 trials, 11 on selegiline and one on lazabemide (several of these trials were also included in [34]). This Cochrane review also found significantly improved scores in favor of MAO-B inhibitors given from baseline to 1 year on treatment.

In the NICE guidelines of 2006 [7] for PD, recommendations based on the analyses by Ives et al. [34] and the Cochrane review [6] stated that “the benefits of MAOB inhibitors versus control in terms of rating scales were consistent with a known short-term symptomatic effect. There does not seem to be any clear increase or decrease in mortality with MAOB inhibitors” and that “further large trials with long-term follow-up are required to assess whether the MAOB inhibitors have neuroprotective properties in PD.”

Rasagiline as an adjunct therapy to levodopa for motor fluctuations in PD

The efficacy of rasagiline was demonstrated in randomized, double-blinded clinical trials. Rasagiline as an adjunct therapy to levodopa has been tested in two separate multicenter, double-blind, parallel-group clinical trials, PRESTO and LARGO.

In the PRESTO (Parkinson Rasagiline: Efficacy and Safety in the Treatment of “Off”) trial [35], 472 patients with at least 2.5h of daily “off” time in spite of optimized treatment with other anti-PD medications (levodopa or concomitant medication such as dopamine agonists or entacapone) were randomized to receive rasagiline, either 1 or 0.5mg/day, or placebo as an adjunct therapy over 26 weeks. A modest but significant reduction of “off” time was found from baseline by 1.85h (29%) for the 1mg/day dose and by 1.41h (23%) for the 0.5mg/day dose compared with a reduction of 0.49h (15%) in the placebo group.

In the LARGO (Lasting effect in Adjunct therapy with Rasagiline Given Once daily) study [36], the investigators wanted to investigate the efficacy and safety of rasagiline compared with entacapone in levodopa-treated patients with PD and motor fluctuations. A total of 687 patients with at least 1h of “off” time measured by a 24h home diary were randomly assigned to either rasagiline 1mg once daily or entacapone 200mg with every levodopa dose or placebo for 18 weeks. The results showed a significant reduction of mean daily “off” time by 1.18h for rasagiline and by 1.2h for entacapone, and by 0.4h for those patients on placebo, from baseline. This study also reported information that is useful for the physicians who treat the patients, i.e the mean daily “on” time without troublesome dyskinesias, and a modest but statistically significant increase of 0.85h was found in patients on either rasagiline or entacapone compared with the placebo group.

Rasagiline in clinical trials for disease modification

The TEMPO (TVP-1012 in Early Monotherapy for PD Outpatients) study [37], was originally designed to evaluate the safety and efficacy of rasagiline given to patients not requiring dopaminergic therapy [29]. In this study, 404 patients with early PD were randomized to placebo or rasagiline 1 or 2mg/day and followed 6 months. At the end of the study period, the rasagiline groups had a robustly improved UPDRS score in comparison with placebo, of 4.2 points (P < 0.001) for the rasagiline 1mg/day group and 3.56 points (P < 0.001) for the rasagiline 2mg/day group.

These promising data resulted in the 6-month study being extended for an additional 6 months for 380 patients [38]. The patients in the rasagiline groups remained on their original dose and the patients from the placebo group who wanted to participate were given rasagiline 2mg/day. The primary endpoint was a change in total UPDRS score from baseline to week 52. The result for the whole 12-month period for the rasagiline groups on 1 or 2mg/day and the delayed 2mg/day were improvements of 3.0, 2.0 and 4.1 points, respectively, in their UPDRS scores. The 12-month rasagiline 2mg/day group had a 2.29-point smaller increase in UPDRS total score compared with subjects treated with placebo for 6 month and then rasagiline 2mg/day for 6 months. The conclusion was that subjects treated with rasagiline 2 or 1mg/day for 12 months showed less functional decline than subjects whose treatment was delayed for 6 months. The TEMPO study was not initially designed as a disease modification trial but did provide the early data on which the planning for the ADAGIO study was based.

The ADAGIO (Attenuation of Disease Progression with Azilect GIven Once daily) study [39] was larger and longer but with a similar delayed-start design. In this double-blinded study, 1176 patients with early, untreated PD were randomized to four parallel arms: rasagiline at either 1 or 2mg/day for a total of 72 weeks (i.e. the early-start groups), and placebo for 36 weeks followed by rasagiline 1 or 2mg/day for another 36 weeks (i.e. delayed-start group). This study has been referred as “a land mark study in terms of size, design and outcome.” It was the first to test prospectively the hypothesis that earlier rather than later use of a specific dopaminergic drug leads to an improvement in outcome that cannot be explained by a symptomatic effect alone [40] The primary analysis used to indicate significant disease modification required three hierarchical endpoints that the early-start group had to meet, based on the change in the total UPDRS score from baseline: (i) the superiority of early-start treatment versus placebo with respect to estimates of the rate of change from baseline in the total UPDRS score between weeks 12 and 36; (ii) superiority to delayed-start treatment in the change of the score from baseline to week 72; and (iii) the noninferiority of early-start treatment compared with delayed-start treatment in the rate of change of the score between weeks 48 and 72, i.e. that the rate of progression in the early-start group on 1 or 2mg/day after week 12 must be significantly slower than the placebo group during the first 9 months showing effect and there had to be a significant difference at 18 months (end of the study) in UPDRS score for the early- and late-start patients indicating a long-lasting effect. However, the rate of progression for UPDRS scores during months 9–18 should also at least be parallel and should not converge.

The results showed that rasagiline 1mg/day achieved all three hierarchical primary endpoints based on disease progression. However, the 2mg/day dosage met the first but failed the second, and therefore did not meet all hierarchical primary endpoints as required. The authors pointed out that the early-start regimen with rasagiline 1mg/day gave support to a possible disease-modifying effect, although 2mg/day did not meet their endpoints.

Selegiline

Selegiline is a propargyl methamphetamine and undergoes extensive hepatic first-pass metabolism to l-methamphetamine and l-amphetamine. These amphetamine metabolites might be associated with cardiac or psychiatric adverse events. Interestingly, during the long-standing DATATOP study, no differences were noted between the groups with selegiline or placebo concerning serious adverse events, cardiovascular adverse events, discontinuation of therapy or mortality. However, the amphetamine metabolites of selegiline have potential cardiovascular effects and have been associated with aggravation of orthostatic hypotension in PD patients. Long-term postmarketing has revealed that psychiatric adverse effects such as nightmares and hallucinations have also been reported frequently in patients on selegiline [42]. In the UK, the PDRG (Parkinson’s Disease Research Group) study showed, during a mean follow-up of 5.6 years, a 60% increase in mortality for patients on levodopa with selegiline compared with levodopa alone [43]. However, a population-based study of more than 12,000 patients with PD in the UK found no significant increase in the risk of mortality among patients on selegiline monotherapy or on combination levodopa/selegiline treatment [44].

Rasagiline

Rasagiline is not a propargyl amphetamine derivate and its principal metabolite, aminoindan, does not have an amphetamine-like chemical structure and therefore has no amphetamine-like adverse effects. The adverse events that more frequent with rasagiline than with placebo are more often gastrointestinal and dose related. In the TEMPO, LARGO and PRESTO studies, no significant differences were shown between rasagiline and placebo with respect to serious adverse events or discontinuations of active therapy.

In a postmarketing observational study [45] in patients with PD, overall 754 patients received rasagiline at a recommended dose of 1mg/day under an observation period of 4 months. Both monotherapy and combination therapy improve parkinsonian symptoms such as reduced “off” time and improved quality of life, and tolerability was rated as “good” or “very good” in 97% in monotherapy patients and in 90% in combination therapy patients. The authors emphasized that “These results under every day clinical practice conditions provide additional support for the earlier findings of the pivotal rasagiline studies in clinical settings.”

Comparison of selegiline and rasagiline in the symptomatic treatment of PD: indirect meta-analysis of randomized placebo-controlled clinical trials

No direct comparative randomized controlled study on comparison between selegiline and rasagiline in the symptomatic treatment of PD exists, but a indirect meta-analysis has been performed [46]. In a systematic literature review, a total of 83 publications on placebo-controlled studies (randomized controlled trials [RCTs]) were identified. For selegiline, there were 71 studies published between 1988 and 2009 with 38 different RCTs of which 15 were considered for analysis of efficacy and/or tolerability, and nine of the 15 met the inclusion criteria for the meta-analysis of efficacy and 12 of the 15 for analysis regarding safety and tolerability. All 12 studies for rasagiline met the criteria. Overall, rasagiline showed a “significant advantage in the primary endpoint UPDRS total scores.” Risks for adverse events such as dizziness, hallucinations, diarrhea and syncope were lower with rasagiline than with selegiline (P < 0.15 for each).

Serotonin syndrome

Psychiatric symptoms are common in PD. It is believed that up to 60% of Parkinson’s patients have depressive symptoms, which require periodic or continuous treatment with antidepressants such as tricyclic antidepressants, selective serotonin reuptake inhibitors (SSRIs) or serotonin norepinephrine reuptake inhibitors – all of these drugs have the possibility of developing serotonin syndrome. Symptoms of serotonin syndrome include hyperthermia, rigidity, myoclonus, autonomic instability with possible rapid fluctuations of vital signs, and mental status changes such as confusion, irritability and extreme agitation progressing to delirium and coma. There is a theoretical risk in the use of antidepressants (in combination with MAO-B inhibitor), but, in practice, the incidence of serotonin syndrome is extremely rare [47]. In a recent review [48], pharmacological management of mood disorders in PD were reviewed. The authors reported the results of the RCT studies of SSRI treatment of depression in PD where only citalopram and paroxetine had a positive outcome, although with varying results. Regarding RCTs and tricyclic antidepressants used for depression in PD, nortryptiline and desipramine were shown to work. Regarding serotonin and norepinephrine reuptake inhibitors for the treatment of depression in PD, venlafaxine was effective.

One safety aspect that should be highlighted in the use of contraindicated antidepressants co-prescribed with MOA-B inhibitors is that velaflaxine and fluoxetine are contraindicated. Citalopram (up to 20mg), escitalopram (up to 10mg) or sertraline (up to 100mg) can be used. Tricyclic antidepressants may be used with caution (amitriptyline up to 50mg/day or trazodone up to 100mg/day), and even paroxetine (up to 30mg/day). It must be emphasized that venlafaxine/mirtazapine should not be used concomitantly with MAO-B inhibitors or within 2 weeks of discontinuation of MAO-B inhibitors. Similarly, approximately 2 weeks should be allowed to pass before a patient treated with venlafaxine/mirtazapine should be treated with MAO-B inhibitors.