Rasagiline In Parkinsons Disease

Published Date: 02 Nov 2017

RASAGILINE IN PARKINSON’S DISEASE: A systematic review and meta-analysis.

Sara Mínguez-Mínguez1, Julián Solís-García del Pozo 2*, Joaquín Jordán1,3*

1 Departamento de Ciencias Médicas. Facultad de Medicina de Albacete. Universidad de Castilla-La Mancha (UCLM).

2 Servicio de Medicina Interna. Hospital General de Villarrobledo.

3 Grupo de Neurofarmacología. Instituto de Investigación en Discapacidades Neurológicas-UCLM

*Authors for correspondence:

Julian Solis Garcia del Pozo

Servicio de Medicina Interna, Hospital General de Villarrobledo. Avenida Miguel de Cervantes s/n. Villarrobledo. Albacete-02600. Spain. Telephone: +34-967133000. e-mail: [email protected]

Joaquín Jordán

Grupo Neurofarmacología. Departamento de Ciencias Médicas. Facultad de Medicina de Albacete. Universidad de Castilla-La Mancha. Calle Almansa, 14. Albacete-02008. Spain. Telephone. +34-967599200. Fax+34-967599327. e-mail: [email protected]

Abstract

Rasagiline (Azilect®) is a selective and irreversible monoamine oxidase B inhibitor. Rasagiline is well tolerated, safe, improves motor symptoms, prevents motor complications in Parkinson disease, has beneficial effects on quality of life parameters, is effective as monotherapy or in adjunctions to levodopa-therapy, is beneficial in early and late stages of PD. Herein, we aimed to compare rasagiline efficacy versus placebo treatment for decreasing the outcomes of PD patients. Major databases (Medline, the Cochrane Library) were systematically searched to identify trials of Rasagiline. Randomized controlled trials comparing rasagiline were selected. As outcome variables, the Unified Parkinson Disease Rating Scale (UPDRS) for rasagiline monotherapy and reduction in off-time for combined treatment were used. Rasagiline monotherapy, in early stages of the disease, reduces UPDRS score [-3.11 (95% CI -3.64, -2.58, P <0.00001)]. In combination with levodopa, rasagiline (1 mg/d) reduced off-time [-0.86 h (95% CI -1.15, -0.56, P <0.00001)]. Despite, although trials with delayed-start design rasagiline reduces UPDRS score [-0.89 (95% CI from -1.78 to 0, P = 0.05)] we found a disagreement between studies and doses, making it difficult to interpret this result. In conclusion, our results demonstrates efficacy of rasagiline in PD. But, it remains to establish the clinical significance of these data. Furthermore, delayed-start design studies did not establish with certainty the neuroprotective effect. It would be advisable to carry out comparative trials with other drugs used in Parkinson's disease.

Introduction

Parkinson’s disease (PD) is a neurodegenerative disorder, characterized by resting tremor, rigidity or stiffness, bradykinesia and postural instability. Although PD was described almost two centuries ago, its etiology remains uncertain, and only descriptive alterations as progressive loss of dopaminergic neurons and a depletion in the levels of dopamine in the substantia nigra are considered PD markers. Pharmacologists have pay attention in the monoamine oxidase (MAO) as a plausible target. Physiologically, MAO is involved in oxidative deamination of monoamine neurotransmitters (e.g., dopamine, norepinephrine, serotonin), as well as biogenic amines, such as tyramine. Two isoforms of MAO, types A and B, have been described; they differ with respect to localization and substrate specificity. MAO-B is the predominant isoform in the human brain, where it acts in the breakdown of dopamine, as well as in the deamination of -phenylethyl-amine, an endogenous amine that stimulates release of dopamine and inhibits its neuronal reuptake. These facts convert MAO-B an ideal pharmacological target for PD treatments, in fact, belong the several types of medications approved by the US Food and Drug Administration (FDA) for PD treatment, are the MAO-B inhibitors, as rasagiline and selegiline.

Rasagiline (Azilect®, N-propargyl-1 (R)-aminoindan) is a second generation propalgylamine, which irreversible inhibits MAO-B, with a potency 10-fold greater than selegiline, is indicated for the treatment of idiopathic PD as monotherapy (without levodopa) or as adjunct therapy (with levodopa) in patients with end-of-dose fluctuations. Rasagiline is administered at a dose of 1 mg once daily, with no titration required. For clinical purposes, the severity of PD is categorized according to the outcome variables, the Unified Parkinson Disease Rating Scale (UPDRS) and reduction in off-time.

Clinical trials have studied Rasagiline both as monotherapy in the TEMPO (Rasagiline Mesylate [TVP-1012] in Early Monotherapy for PD Outpatients) and ADAGIO (Attenuation of Disease Progression with Azilect Given Once-daily) studies and as adjunct therapy in combination with levodopa in the PRESTO (Parkinson´s Rasagiline: Efficacy and Safety on the Treatment of "OFF") and Lasting effect in Adjunct therapy with Rasagiline Given Once daily (LARGO) studies . However, mixed results have been reported, possibly because of sample heterogeneity or methodological inconsistencies.

The use of meta-analysis allows us to make the best use of all the information we have gathered in our systematic review by increasing the power of the analysis. And, although present limitations, the statistical combination of the results of similar studies consent us to improve the precision of our estimates of treatment effect, and assess whether treatment effects are similar in similar situations.  Using this toll, we have recently shown that among the most frequently reported adverse effects for rasagiline as monotherapy are: headache, dizziness and insomnia. Depression, dizziness, somnolence, and other sleep disorders are reported when used in combination therapy. Our previous analysis demonstrated that the most frequently reported adverse effects in trials did not occur more often with rasagiline than placebo . Indeed, elsewhere recent meta-analysis concluded that rasagiline was effective for PD treatment, but the number of studies included was limited or .

So, our aim in this study is to review the evidence supporting the use of rasagiline in Parkinson's disease, both as monotherapy and in combination therapy. A statistical analysis to determine the efficacy of rasagiline vs. placebo by changes in UPDRS or off-time was performed. To achieve this goal we have collected data from clinical trials using techniques of meta-analysis.

Material and methods

A search in MEDLINE and in the Cochrane Database of Controlled trials (CENTRAL) with the term "rasagiline" (up to March 2013) was made. At the same time, a review of the literature provided by each of the papers found was made in search of new items. Reports that could not be retrieved online were sought through contact with study authors. Reference lists from selected studies were searched for additional reports, and all relevant findings were added to the selected pool. Two investigators (SMM and JSGdP), independently extracted those publications identified here that describe controlled studies of rasagiline given in PD. Disagreements were resolved in discussion with a third investigator (JJ). Clinical trials published employing rasagiline in PD patients were included in our review. Only trials with fully characterized patients in terms of diagnosis, duration, dosage of treatment were selected. These trials had to compare the rasagiline effectiveness and placebo or other anti-Parkinson drugs, or different doses of rasagiline. It must also include clearly the outcome variable used in terms of efficacy. Studies in which the parameters listed above are not clear or where the objectives and results are confusing or unclear, or where diagnostic criteria or selection of participants are not sufficiently specified were not included in this review. Overall score of UPDRS or the score of UPDRS subscales were selected as the main efficacy variables. In patients with motor fluctuations, the reduction of the daily off-time is selected as efficacy variable.

After selecting the clinical trials included, the following data were extracted: author and year of publication, total number of patients, number of patients in each group, age and sex, degree of evolution of the disease, if the patients were taking other anti-parkinsonian drugs, time of treatment with rasagiline, dosage, outcome variable used and its results in the study groups.

The quality of each study was determined by identifying the possible following biases: a) Selection bias: it can be divided into bias in random sequence generation, and bias in the allocation concealment. b) Performance bias: blinding of participants and personnel. c) Detection bias: concealment or blinding in the evaluation of the results. d) Bias because incomplete outcome data. And e) Bias because selective reporting. Each of these potential biases has been evaluated in one of the following three categories: low risk of bias, high risk of bias or unclear risk of bias. This grading was performed according to the instructions of the Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (The Cochrane Collaboration, 2011) .

Statistical analysis to determine the efficacy of rasagiline vs. placebo by changes in UPDRS or off-time was performed. It has also made a comparison in efficacy between different doses or between rasagiline and other drugs if the selected studies allow. Quantitative variables are expressed as mean (standard deviation) or as mean (95% confidence interval).

In the comparison of efficacy variables (UPDRS score and reduced off-time), the weighted mean difference with 95% confidence interval was used. A random-effects model was used in the case of significant heterogeneity between studies. To measure heterogeneity Cochran Q statistic and I2 of inconsistency has been used. For all statistical test a signification level of p <0.05 was selected. Analyses were carried out using RevMan.

Results

Figure 1 depicts the results of the search strategy. The conducted search yielded 391 articles. Of them, 336 were initially excluded because they were reviews, letters, editorials, opinion pieces, laboratory or animal studies. Of the 55 remaining articles were subsequently excluded another 32: 13 for being describing isolated cases, 7 because were non-clinical trials, 4 for being economic or cost-effectiveness studies, 4 due to were trials conducted in healthy volunteers and not in patients with PD, 2 articles for being validation questionnaires, 1 for being a paper describing the design of a clinical trial subsequently conducted, and another 1 for being published in Russian language (Figure 1). The remaining 23 studies were considered in this study (Table 1). Four major trials were found: The Rasagiline (TVP-1012) mesylate in Early Monotherapy for PD Outpatients (TEMPO) study , Parkinson's Rasagiline Efficacy and Safety on the Treatment of "OFF" (PRESTO) , Lasting effect in Adjunct Therapy With Rasagiline Given Once daily (LARGO) and Attenuation of Disease Progression with Azilect Given Once daily (ADAGIO) . Noteworthy that 11 of the included studies were substudies and secondary publications derived from the above 4 mentioned studies. In addition, another 8 clinical trials were also included .

a. Description of clinical trials

The TEMPO study was a randomized double-blind trial, performed during 6-months, which enrolled 404 patients with PD who did not require treatment with levodopa. Patients were divided into 3 groups: the first one received rasagiline 1 mg/d (n = 134), the second 2 mg/d (n = 132), and the third received placebo (n = 138). The primary response variable was the change in the UPDRS between baseline and 26 weeks of treatment. Mean changes were 0.1 (SD 6.8), 0.7 (5.8), and 3.9 (7.5) at doses of 1 mg/d, 2 mg/d and placebo. After adjustment for baseline score and the center where the patient was treated, changes in dose of 1 mg/d compared to placebo at 26 weeks were -4.20 (-5.66 to -2.73), and 2 mg/d compared to placebo was -3.56 (-5.04 to - 2.08).

As a continuation of the TEMPO study, the Parkinson Study Group published a new article introducing rasagiline at 2 mg/d in the placebo group, which resulted in a "delayed-start" design . The treatment time was extended for another six months, with a total of 52 weeks of treatment. In this phase, of the initial 404 patients, 371 patients were enrolled. The mean changes in the UPDRS score were 3.01 (8.26), 1.97 (7.49) and 4.17 (8.83) for the daily dose of 1 mg or 2 mg, and placebo followed by 2 mg/d, respectively. The difference in outcome between the 1 mg/d rasagiline group and placebo-2mg/d group was favorable to the first group with -1.82 (-3.64 to 0.01, P = 0.05), and the difference for the 2 mg/d group versus placebo-2mg group was favorable to the 2 mg/d group with -2.29 (-4.11 to -0.48, P = 0.01).

Biglan et al performed a TEMPO study subanalysis focusing in the changes in the Parkinson Disease Quality of Life scale (PDQUALIF). Of the 404 patients included in the first 6 months of TEMPO study, improvement was detected on PDQUALIF compared to placebo. This result was obtained for the 1 mg/d rasagiline (difference -2.91, 95% CI - 5.19 to -0.64, P = 0.01) and for the 2 mg/d (-2.74, 95% CI -5.02 to -0.45, P = 0.02). In the second part of this study (design delayed start), continued evaluation of the quality of life up to 12 months, but here only 266 patients could be evaluated. There was no difference between the group who started rasagiline 12 months before and the group who started rasagiline only 6 months before. The quality of life improvement appeared to be due mainly to the symptomatic effect of rasagiline. Noteworthy the absence of relationship between changes in scores measuring motor symptoms (UPDRS III) and the quality of life score at 52 weeks.

Hauser et al. published an open label extension of the TEMPO, in which 306 patients continued with rasagiline (adding other medications if necessary because the clinical course of patients) for an average time of 3.6 ± 2.1 years, and in some cases reaching a follow up of 6.5 years. They revealed that UPDRS during this time is consistently better in the early treatment group, compared with delayed treatment group, supporting a neuroprotective effect for rasagiline. Likewise, Lew et al. , using the same patients cohort showed that rasagiline was generally well tolerated and was effective in the long term.

The principal trial was the PRESTO study , which enrolled 472 levodopa treated patients that received during 6 months, by randomized assignment, rasagiline1mg/d, rasagiline 0.5 mg/d, or a placebo. A reduction in the average in off-time was reported (1 mg/d, 1.85 h; 0.5 mg/d, 1.41 h; and placebo, 0.91 h). The rasagiline patients had a difference in off-time compared with placebo of -0.94 h (95% CI -1.36 to -0.51, P <0.001) with a dose of 1 mg/d, and a difference of -0.49 h (95% CI of - 0.91 to -0.08, P = 0.02) for the dose of 0.5 mg/d.

White et al conducted a self-monitored of blood pressure and transtelephonic control of the patients from PRESTO study, concluding that rasagiline does not induce postprandial hypertension in patients with Parkinson even having a diet without tyramine restriction.

Furthermore, we detected 3 substudies that included patients from TEMPO and PRESTO studies. Goetz CG et al. reported a similar incidence of rasagiline-induced side effect between Rasagiline and placebo groups. Indeed, when the patients were divided into 2 groups, patients younger or older than 70 years, the authors did not find difference between them. In a second study, Elmer et al (876 patients) examined the rate of cognitive and behavioral adverse effects, and changes in the UPDRS subscale for mental function comparing in the subgroup of patients who received 1 mg/d rasagiline or placebo finding no differences. deMarcaida et al conducted a study with 110 patients after completing 26 weeks of treatment with rasagiline, concluding that rasagiline, at doses from 0.5 to 2 mg/d, is not associated with clinically significant reactions to tyramine.

The third major trial was the double-blind multicenter study LARGO study . The 687 patients, with PD and motor fluctuations who were treated with levodopa, were randomized to receive: rasagiline 1 mg/d, entacapone 200 mg/d or placebo with each levodopa dose. Rasagiline (1 mg /d) reduced off-time and improved motor symptoms. Although the trial comparisons between rasagiline and entacapone were not made, rasagiline-induced reduction in off-time (-1.18 h, SD 0.15) was similar to that obtained with entacapone (-1.20 h, SD 0.15), In addition, the difference between rasagiline vs. placebo was -0.78 h (95% CI of -1.18 to -0.39, P = 0.0001).

Stocchi and Rabey, aimed a substudy of LARGO in order to measure the effect of rasagiline and entacapone in the motor symptoms of PD during the off-period. Rasagiline (32 patients), but not entacapone (36 patients), significantly improved motor symptoms compared with placebo (37 patients) in the off-time. Recently, Tolosa and Stern divided the patients from PRESTO and LARGO studies into 2 groups, patients younger or older than 70 years of age. They compared the efficacy and safety of rasagiline, founding non significant differences between older patients (-1.41 vs. off-time h with placebo -0.43, p <0.01) and younger (-0.70 vs. -1.54 h with placebo, p <0.001) patients in the reduction in Off-time.

ADAGIO study was designed to detect if rasagiline has a neuroprotective effect to modify the clinical course of PD regardless of their symptomatic effect. A total of 1176 subjects with untreated Parkinson's disease were randomly assigned to receive rasagiline (at a dose of either 1 mg or 2 mg per day) for 72 weeks (the early-start group) or placebo for 36 weeks followed by rasagiline (at a dose of either 1 mg or 2 mg per day) for 36 weeks (the delayed-start group). The two doses were associated with different outcomes. Early treatment with rasagiline (1 mg/d) provided benefits that were consistent with a possible disease-modifying effect. But, early treatment with rasagiline (2 mg/d) did not.

A secondary and post-hoc analysis of ADAGIO study was performed by Rascol et al , and concludes that rasagiline reduced the need for additional antiparkinsonian medication. Indeed they emphasize the contribution of the subscales of the activities of daily living of the UPDRS in the response to treatment with "early" rasagiline 1 mg/d group versus "delayed".

In addition to the above mentioned 4 major trials, other 5 articles have been published. Thus, in 2000 a small trial was published in which it was to assess the safety and tolerability of rasagiline administered in combined therapy in 70 patients who were already treated with levodopa for at least 6 months . . The patients were randomized in 4 groups, 3 of them received rasagiline 0.5mg/d, 1 and 2 mg/d and the fourth received placebo. This trial revealed that the incidence and types of side effects were similar in the groups of treatment whit rasagiline than in the placebo group. Neither difference in clinical effect was detected.

Later, in 2004, Stern et al assessed the rasagiline safety and tolerability in patients with early PD who had no received levodopa treatment during 10 weeks. The 56 patients were randomized into 4 groups to receive: 1, 2, 4 mg/d of rasagiline or placebo. During the study, no severe side effects were observed, and the frequency and type of side effects in patients treated with rasagiline was similar to the patients in the placebo group. Interestingly, rasagiline 2 mg/d showed significant changes in UPDRS score after treatment compared with baseline. But, this circumstance was not observed with 1mg/d or 4 mg/d doses.

Hanagasy et al published in 2011 a trial involved 48 patients, aimed in the rasagiline effect on cognitive deficits in PD patients. The patients selected suffered of cognitive impairment without dementia. After 12 weeks, they compared rasagiline 1 mg/d versus placebo and found differences in attention (p=0.005) and in executive functions such as verbal fluency (p=0.038), but no effect was seen in other areas such as memory and visuospatial functions.

Wilson et al performing an open-label (272 patients, 12 weeks) not placebo controlled, revealed that rasagiline, in a dose-independently manner (1 mg/d monotherapy, or 0.5 mg/d as combination therapy) improved the UPDRS score a week later of starting treatment.

A small pilot study involving only 6 patients was conducted by Korchounov et al with the aim to assess the antidepressant effect of rasagiline in PD patients. The 3 patients who received 2 mg/d of rasagiline have better outcome of their depression (assessed by the Hamilton Depression Score) with respect the patients received 1 mg/d of rasagiline. However, this improvement does not correlate with the improvement of the motor symptoms which was the same in both groups.

Finally, there are three other recently published works. Alvarez et al published a short paper involving 17 PD patients with hyposmia already receiving PD medication, to evaluate olfaction with rasagiline treatment. They reported that the patients have an improvement assessed with the University of Pennsylvania Smell Identification Test. Viallet el al. have performed a randomized double-blind trial to assess the safety and tolerability of rasagiline compared to pramipexole in the treatment of early PD. In this trial, rasagiline had clinically favorable difference in gastrointestinal and sleep adverse effects compared to pramipexole. Müller et al performed an open label trial in 30 PD patients to demonstrate that a switch from selegiline to rasagiline is well tolerate and safe and provide benefits in these patients in terms of UPDRS score (from 22.63 mean score to 21.4 mean score after four months with rasagiline treatment).

Meta-analysis of clinical trials

Next, we conducted a meta-analysis on the above mentioned clinical trials in an effort to obtain a better understanding of how well rasagiline works.  To do so, rasagiline efficacy was assessed using the overall UPDRS score in 3 clinical trials (Stern et al , ADAGIO and TEMPO studies). As shown in figure 2, rasagiline effectiveness, compared to placebo, is favorable for both doses of 1 mg/d or 2 mg/d. Specifically, rasagiline at 1 mg/d the overall effect in terms of UPDRS over placebo was -3.06 points (95% CI -2.31 to -3.81 at p <0.00001) with no significant heterogeneity between studies. In addition, rasagiline 2 mg/d effect is slightly higher (-3.17 points, 95% CI -3.91 to -2.42, P <0.00001) with no heterogeneity between studies (figure 2). The comparison between the dose of 1 mg/d and 2 mg/d shows the equivalence of them (mean difference 0.00, 95% CI -0.82 to 0.81).

Subsequent, we compared the change in the overall UPDRS score between groups of the trials with "delayed start" design (ADAGIO and TEMPO), we could only find an effect of -0.89 points (95% -1.78-0) and a borderline statistical significance (p = 0.05) (Figure 3).

If we analyze distinguishing between doses (1 or 2 mg/d), we can denote that for the 1 mg/d dose the effect observed in ADAGIO was -1.68 points (p = 0.03). In TEMPO there was not delayed treatment group with 1 mg/d rasagiline, so its outcome in this group was not included. For the 2 mg/d dose of rasagiline, although TEMPO had good results, in ADAGIO these good results were not obtained, so that the overall effect in this dose group compared to delayed treatment is not significant (p = 0.53) (Figure 3).

Finally we assayed the rasagiline effect on off-time. In this case, this parameter has only been reported by LARGO and PRESTO studies. Rasagiline 1 mg/d reduced off-time over placebo in -0.86 h (95% CI -1.15 to -0.56, P <0.00001). This reduction is lower with a dose of 0.5 mg which was used in PRESTO study (Figure 4).

Discussion

Herein we reviewed the evidence supporting the rasagiline usage in PD patients. We provide an exhaustive description of each clinical trial conducted, tests performed and the estimated effect of rasagiline. Our data offer evidence on rasagiline effects in terms of efficacy, either on UPDRS or on off-time. Indeed, we distinguish between used doses, and provide an estimation of the effect on delayed-start studies.

The major strength of the study is the validated meta-analysis procedure adopted for comparing the results of different studies. Our meta-analysis includes all the published trials up to march 2013 and doing so, we provide data of interest with respect to those previously published. Indeed, we differ in the addressed objectives. Although in 2004, Ives et al. published a meta-analysis of 17 randomized trials using 3 different MAO-B inhibitors (rasagiline, selegiline and lazabemide), only one trial was conducted with rasagiline (TEMPO ). Our results are in the line of their observations, patients treated with MAO-B inhibitors obtained better scores in clinical trials (UPDRS II and III). But, they detected similar or slightly worse differences than ours to the global UPDRS with rasagiline 1 mg (difference of -3.06 points in favor of rasagiline over placebo), and 2 mg/d of rasagiline (-3.17 point difference favor of rasagiline over placebo). Even more, our results support the data from Clarke at al., meta-analysis , where they analyzed data from studies designed "delayed-start" (ADAGIO and TEMPO), founding an improvement of 0.91 points on the UPDRS (95% CI 0.01-1.8) compared with the group of "delayed" treatment, but with a borderline statistical significance (p = 0.05). furthermore the present meta-analysis differentiate between dose and the effect on off-time, detail that was not addressed by Jost WH et al, who performed an indirect comparison of the effectiveness of selegiline and rasagiline.

Our data reveal that rasagiline, 1 mg/d or 2 mg/d, decreased the score of UPDRS in early PD over placebo around 3 points. But, we would highlight a plausible problem that might arise in interpreting these results. The clinical impact of UPDRS scores differences. In this line, we denoted that there are not many jobs that value minimal clinically significant change in the UPDRS. For instance, Schrag et al. found that a change of 8 points in the overall score of the UPDRS was clinically significant. Shulman et al. found that a difference of 4.1 to 4.5 points for the overall UPDRS. In addition, more recently, Hauser et al concluded that a difference between 3 and 3.5 points of improvement in the UPDRS score in patients with early Parkinson was clinically significant. Our found results can be included between this latter range. But, we found that rasagiline reduces the time off 0.74 h. This difference does not reach the significant clinical relevance if we used the value offered by Hauser (1h) .

One explanation could be found in the intrinsic mechanism of action of MAO-B inhibitors, which require the residual availability of adequate amounts of endogenous dopamine. Furthermore, treatment efficacy is expected to diminish with the severity of PD. Age at onset of PD is variable, and the disease can present with a range of motor and nonmotor symptoms, many of which are poorly defined and nonspecific. Moreover, diagnosis of PD is currently based on presence of the classic motor symptoms, which are caused by a >50% loss of dopaminergic neurons in the substantia nigra pars compacta, leading to a >80% reduction in dopamine levels in the striatum.  So, it will be interesting knowing the effects of rasagiline in the premotor phase. It is now recognized that this probably last more than ten years. Arguably, any putative treatment to modify disease progression and prevent development of PD will have to be applied to early stages of the disease. So, we unknown a possible effect of rasagiline on "silent" reductions in nerve cell number and dopamine concentrations in the substantia nigra demonstrate early signs of nigrostriatal cell degeneration.

Finally, a topic of high interest in the clinical practice is the possible neuroprotective effect of rasagiline, by means of which is attributed a disease-modifying effect. This idea was addressed in two randomized delayed-start trials. But, the differences observed between doses (1 and 2 mg/d) were not clearly explained. In this line, our study points out that the scarce differences (0.89 points), observed between treatment groups early and delayed, fail to present statistical and clinical significance. We have to point that neuroprotector effects might be exerted through different target besides the inhibition of MAO-B, e.g. stabilizing mitochondrial membrane, regulation of antioxidant enzymes .

One possible general limitation of this study is that, like all meta-analyses it is based on published trials and is therefore susceptible to the so called "publication bias", that is, positive studies are more likely to be published. Most of the trials considered in this analysis were company-sponsored for registration purposes and hence registered in national and international directories before being initiated. This should prevent the selective publication of only studies with positive results. Moreover, although 23 trials were identified, most of them are grouped within the four large trials (TEMPO, PRESTO, LARGO and ADAGIO), since other trials are very heterogeneous.

In conclusion, the results of our meta-analysis and the trials mentioned above support the efficacy of rasagiline in PD patients for the outcomes compared with placebo. But, from the point of view of the clinical effects and its effectiveness as a neuroprotective drug, the issue does not seem entirely clear. We contemplate that would be very interesting, from the clinical perspective, to perform comparative studies with other drugs used in PD. So, better therapeutic regimens for PD will relay upon the development of methods for early detection of PD and the ability to monitor disease progression.

Table 1: Studies included in the review.

Author (year)

N

Other drugs

Time taking rasagiline

Dose (mg)

Primary efficacy variable

TEMPO

PSG (2002)

404 (3 groups)

No

26 weeks

1, 2

UPDRS

PSG (2004)

371 (3 groups)

If patient needs

26 weeks + 26 weeks "delayed-start" (52 total weeks)

1, 2 and placebo-delayed 2

UPDRS

Biglan KM (2006)

404 (1st phase 26 weeks).

266 (2nd phase to week 52)

No

52 weeks

1,2 and placebo-delayed 2

PDQUALIF

Hauser RA (2009)

306

If patient needs

3.6±2.1 years

1,2 and placebo-delayed 2

UPDRS

Lew MF (2010)

398

If patient needs

3.5±2.1 years

1,2 and placebo-delayed 2

UPDRS

Progression of Hoehn & Yahr stage. Need for new drug

PRESTO

PSG (2005)

472 (3 groups)

Yes (levodopa)

26 weeks

0.5, 1

Change in total daily OFF time

White WB (2008)

472 (444) 3 groups

Yes (levodopa)

26 weeks

0.5, 1

Increase of postprandial systolic blood pressure> 30 mmHg.

TEMPO / PRESTO

Goetz CG (2006)

876 (404 + 472)

Not for TEMPO patients. Yes for PRESTO patients

26 weeks

0.5, 1 in combined therapy, and 1,2 in monotherapy

Adverse effects (≥70 years old vs <70 years old).

Elmer L

(2006)

876 (404 + 472)

Not for TEMPO patients. Yes for PRESTO patients

26 weeks

0.5, 1 in combined therapy, and 1,2 in monotherapy

Cognitive and behavioral adverse effects.

UPDRS I (mental)

deMarcaida JA (2006)

110 (55 + 55)

Not for TEMPO patients. Yes for PRESTO patients

26 weeks

0.5, 1 in combined therapy, and 1,2 in monotherapy

Increase of blood pressure ≥ 30 mmHg, o reduction in heart rate < 40 bpm.

LARGO

Rascol O (2005)

687 (3 groups, 1 with rasagiline)

Yes (levodopa)

18 weeks

1

Change in total daily OFF time

Stocchi F (2011)

105 (3 groups, 1 with rasagiline)

Yes (levodopa)

18 weeks

1

UPDRS II y III during OFF

PRESTO/LARGO

Tolosa E (2012)

768

Yes (levodopa)

26-18 weeks

1

Change in total daily OFF time (comparison between ≥ 70 years old and <70 years old).

ADAGIO

Olanow CW (2009)

1176 (4 groups)

No

36 weeks + 36 weeks "delayed-start" (72 total weeks)

1, 2, placebo-delayed 1, placebo-delayed 2

UPDRS

Rascol O (2011)

1176 (4 groups)

No

36 weeks + 36 weeks "delayed-start" (72 total weeks)

1, 2, placebo-delayed 1, placebo-delayed 2

Need for additional antiparkinsonian treatment.

UPDRS subscales.

PFS

nM-EDL

OTHER TRIALS

Rabey JM (2000)

70 (4 groups, 3 with rasagiline)

Yes (levodopa)

12 weeks

0.5,1,2

Trial focused on safety/tolerability.

UPDRS

Stern MB (2004)

56 (4 groups)

No

10 weeks

1,2,4

UPDRS

Hanagasi HA (2011)

48

Yes

12 weeks

1

Various tests

Wilson RE (2011)

272 (2 groups)

Yes, those receiving 0.5mg daily.

12 weeks

0.5, 1

Bradykinesia subscale of the UPDRS

Korchounov A (2012)

6 (2 groups)

No

8 weeks

1,2

HAMD

Alvarez MV (2012)

17

Yes

3 months

1

UPSIT

UPDRS-III

Viallet F (2013)

109 (2 groups, 1 with rasagiline and the other with pramipexole)

No

15 weeks

1

Safety

Müller T (2012)

30

Yes (some patients)

4 moths

1

PDSS, HAMD, UPDRS, PDQ39, d2-test, safety.

UPDRS=Unified Parkinson’s Disease Rating Scale; PDQUALIF= Parkinson ’s disease Quality of Life scale; PFS= Parkinson fatigue scale; nM-EDL=non Motor experiences of daily living; HAMD=Hamilton Depression Score; UPSIT= University of Pennsylvania Smell Identification Test; PDSS=Parkinson’s Disease Sleep Scale; PDQ39= Parkinson’s Disease Questionnaire 39.

Table 2: Risk of bias in the four major studies and the other studies included in the review.

Study

Random sequence generation

Allocation concealment

Blinding of participants and personnel

Binding of outcome assessment

Incomplete outcome data

Selective reporting

TEMPO

Unclear

Unclear

Low

Low

Low

Low

PRESTO

Low

Low

Low

Low

Low

Low

LARGO

Low

Low

Low

Low

Low

Low

ADAGIO

Low

Unclear

Low

Low

Low

Low

TEMPO delayed-start

Unclear

Unclear

Low

Low

Low

Unclear

Rabey et al.

Unclear

Unclear

Low

Unclear

Low

Unclear

Stern et al.

Low

Low

Low

Unclear

Low

Unclear

Hanagasi et al.

Unclear

Unclear

Low

Low

Low

Unclear

Wilson et al.

Nonrandomized

Nonrandomized

High (open)

High (open)

Low

Unclear

Korchounov et al.

Unclear

Unclear

Low

Low

Low

Unclear

Alvarez et al. .

Nonrandomized

Nonrandomized

High (open)

High (open)

Low

Unclear

Viallet et al.

Unclear

Unclear

Low

Unclear

Low

Low

Müller et al.

Nonrandomized

Nonrandomized

High (open)

High (open)

Low

Low

Figure 1:  Flow chart outlining the search strategy and results along various steps.

Figure 2: Individual and pooled relative risk for the outcome in terms of UPDRS scores in the studies considering rasagiline (1 mg/d and 2mg/d) compared to placebo therapy in PD patients.

Figure 3: Comparison between group early and delayed treatment with rasagiline in the delayed-start trials: overall (A) and differentiating between doses (B).

Figure 4: Comparison of 0.5 mg/d and 1 mg/d of rasagiline in combination therapy versus placebo in terms of reducing daily off-time.