lable at ScienceDirect

Parkinsonism and Related Disorders 20 (2014) 1345e1351

Contents lists avai

Parkinsonism and Related Disorders

journal homepage: www.elsevier .com/locate/parkreldis

Effects of long-term treatment with rotigotine transdermal system ondyskinesia in patients with early-stage Parkinson's disease

Nir Giladi a, b, *, Liesbet Ghys c, Erwin Surmann d, Babak Boroojerdi e, Joseph Jankovic f

a Department of Neurology, Tel-Aviv University, Tel-Aviv, Israelb Sago School of Neuroscience, Tel-Aviv University, Tel-Aviv, Israelc UCB Pharma, Brussels, Belgiumd UCB Pharma, Monheim am Rhein, Germanye UCB Biosciences Inc., Raleigh, NC, USAf Parkinson's Disease Center and Movement Disorders Clinic, Baylor College of Medicine, Department of Neurology, Houston, TX, USA

a r t i c l e i n f o

Article history:Received 21 May 2013Received in revised form4 September 2014Accepted 8 September 2014

Keywords:RotigotineParkinson's diseaseDyskinesiaLevodopa

* Corresponding author. Movement Disorders UniTel-Aviv Sourasky Medical Center, Sackler School of MWeizmann Street, 64239, Tel-Aviv, Israel. Tel.: þ976974153.

E-mail address: nirg@tlvmc.gov.il (N. Giladi).

http://dx.doi.org/10.1016/j.parkreldis.2014.09.0161353-8020/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

Purpose: In two 6-month, double-blind, placebo-controlled studies, rotigotine transdermal system waswell-tolerated and efficacious monotherapy in early-stage PD. This post hoc analysis of the long-termopen-label extensions (NCT00594165; NCT00599196) of these studies assessed incidence and severity ofdyskinesia in participants treated with rotigotine, with or without concomitant levodopa, for up to 6years.Methods: Open-label rotigotine was titrated to optimal dose (�16 mg/24 h). Concomitant levodopa waspermitted. Dyskinesia data, recorded using the Unified Parkinson's Disease Rating Scale Part IV, werepooled from the two open-label studies.Results: Of 596 participants who received open-label rotigotine, 299 (50%) remained at trial closure; nopatient discontinued due to dyskinesia. In the two studies, median exposure to rotigotine was 1910 days(~5 years, 3 months), and 1564.5 days (~4 years, 3 months). During up to 6 years of open-label rotigotine,423/596 (71%) received levodopa. Dyskinesias were reported in 115/596 (19%) participants, 90/115 (78%)of who developed dyskinesia after levodopa was added; 25 reported dyskinesia in the absence oflevodopa (includes patients who never received open-label levodopa, and those who reported dyskinesiabefore starting concomitant levodopa). Dyskinesia severity data were available for 107 of the 115 par-ticipants. In 56/107 (52%) participants, dyskinesia was considered ‘not disabling’ for all occurrences; theworst-case severity was ‘mildly disabling’ for 33/107 (31%), and ‘moderately’ or ‘severely disabling’ for18/107 (17%; 3% of total participants).Conclusion: During treatment with rotigotine in patients with PD for up to 6 years the incidence ofdyskinesia was low, and the dyskinesia was generally ‘not disabling’ or ‘mildly disabling’.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Although levodopa remains the most efficacious symptomatictreatment for Parkinson's disease (PD), in most patients its long-term use eventually becomes complicated by the development ofmotor fluctuations and dyskinesia [1,2]. Dyskinesia (in a form of

t, Department of Neurology,edicine, Tel-Aviv University, 62 3 6974790; fax: þ972 3

chorea, stereotypy, or dystonic involuntary movements) can resultin marked disability and may negatively impact quality of life [3].Within 4e6 years of initiating levodopa therapy around 40% ofpatients develop dyskinesia, with almost all patients eventuallydeveloping dyskinesia with prolonged treatment [4].

While the pathogenesis of dyskinesia is not fully understood,preclinical and clinical findings suggest that the development oflevodopa-induced dyskinesia is related to levodopa's short plasmahalf-life, which results in pulsatile stimulation of striatal dopaminereceptors and altered basal ganglia output [5,6]. Normal striataldopamine receptor stimulation is thought to consist of both low-frequency continuous tonic stimulation and higher-frequencyphasic stimulation. Based on this tonic/phasic hypothesis of

N. Giladi et al. / Parkinsonism and Related Disorders 20 (2014) 1345e13511346

dopamine release, the concept of continuous dopaminergic drugdelivery has emerged as a clinical strategy that may more closelymimic physiologic dopaminergic stimulation [6,7]. Dopamine re-ceptor agonists, which have a longer plasma half-life than levo-dopa, are often used as first-line therapy in patients with early-stage PD as a strategy to postpone the need for or reduce thedose of levodopa, thereby minimizing the risk of dyskinesia andother levodopa-related motor complications [8,9].

Rotigotine is a dopamine receptor agonist with activity acrossD1 through D5 receptors as well as select adrenergic and seroto-nergic sites [10,11]. Rotigotine transdermal system providescontinuous dopaminergic stimulation [12], and maintains stableplasma levels over 24 h with a single daily application [13]. Animalstudies have suggested that continuous administration of rotigo-tine may reduce the risk of dyskinetic involuntary movementscompared with pulsatile administration of rotigotine or levodopa[14,15]. In two 6-month, randomized, double-blind, placebo-controlled studies, rotigotine transdermal system was shown to bean efficacious and well-tolerated monotherapy in early-stage PD[16e18]. Each study offered enrollment in a prospective, long-term,open-label extension [19,20]. As the longest interventional studiesof rotigotine conducted to date, with adjunctive levodopapermitted as required during the open-label extensions, thesestudies provide an opportunity to assess the long-term incidence ofdyskinesia. In this post hoc analysis, results from these open-labelextensions were assessed to determine the long-term effect ofrotigotine on incidence and severity of dyskinesia in patients whoenrolled with early-stage PD, with or without concomitantlevodopa.

2. Methods

2.1. Design

Participants completing the 6-month maintenance period of one of two phaseIII, randomized, double-blind trials of rotigotine transdermal system in patients withearly-stage PD (SP512 [17,18]) and SP513 [16]) had the option of enrolling intoprospective, single-arm, open-label extension studies of up to 6 years' duration:SP702 (participants from SP512; ClinicalTrials.gov NCT00594165 [19]) and SP716(participants from SP513; ClinicalTrials.gov NCT00599196 [20]). Participantsenrolled in each double-blind studywere aged 30 years or older, had been diagnosedwith idiopathic PD for up to 5 years, had a Unified Parkinson's Disease Rating Scale(UPDRS) Part III score of �10, and a Hoehn and Yahr stage �3 (mild-to-moderatePD). Treatment with dopamine receptor agonists or levodopa was not permittedduring the double-blind studies or in the 1 month prior to enrollment. Participantswere also excluded from entering the double-blind studies if they had taken levo-dopa for longer than 6 months since PD diagnosis. Levodopa (in combination withbenserazide or carbidopa) was allowed, if required, after 1 month of open-labelrotigotine maintenance. In addition, the following other anti-Parkinson medica-tions were allowed after 1 month of open-label rotigotine maintenance: MAO-Binhibitors, anticholinergic agents, NMDA-antagonists (e.g., amantadine), entaca-pone, and modafinil. Complete inclusion and exclusion criteria are published[16e20].

2.2. Procedures

In the double-blind studies, participants were randomized to receive once-daily transdermal patches of placebo or rotigotine (2e6 mg/24 h) in SP512, andplacebo, rotigotine (2e8 mg/24 h), or oral ropinirole (up to 24 mg/day) in SP513. Atthe end of double-blind maintenance, treatment was de-escalated to 2 mg/24 hrotigotine. All participants entering the open-label studies, including those whowere treated with placebo or ropinirole in the double-blind studies, received open-label rotigotine titrated to their optimal dose (up to 6 mg/24 h in SP702, up to8 mg/24 h in SP716). The optimal dose was defined following discussion betweenthe participant and the investigator, taking into account the potential forimprovement of disease symptoms and the participant's adverse event (AE) profile.After the first year of open-label maintenance, up-titration to a maximum dose of16 mg/24 h was permitted. During open-label maintenance, dose adjustments ofrotigotine were permitted at any time at the discretion of the investigator. Theinvestigator was encouraged to increase rotigotine to maximum allowed dosebefore adding/increasing levodopa or other anti-Parkinson medications. Partici-pants were maintained at their optimal dose until rotigotine became commerciallyavailable or the sponsor closed the trial, whichever came first. Maintenance visits

were scheduled at the end of the first month of open-label maintenance and at 3-month intervals thereafter [19,20].

2.3. Rotigotine exposure and levodopa use

The extent of rotigotine and levodopa exposure during the open-label studieswas reported and included assessments of dose, time to treatment initiation(levodopa only), and duration of treatment.

2.4. Assessment of dyskinesia

Dyskinesia was assessed using UPDRS Part IV, Complications of therapy (in thepast week) item 32 (dyskinesia duration) and item 33 (dyskinesia disability) criteria.Item 32: “What proportion of the waking day are dyskinesia present? (historicalinformation)”, where a score of 0 ¼ none, 1 ¼1e25%, 2 ¼ 26e50%, 3 ¼ 51e75%, and4¼ 76e100%. Dyskinesia was defined as present if the response to UPDRS IV item 32was >0. Item 33: “How disabling are the dyskinesia? (historical information, whichmay be modified by office examination)”, where a score of 0 ¼ Not disabling,1 ¼ Mildly disabling, 2 ¼ Moderately disabling, 3 ¼ Severely disabling, and4 ¼ Completely disabling. The incidence and severity of dyskinesia were thereforebased on participant report, with the opportunity to modify the responses followinginvestigator examination. UPDRS Part IV assessments were performed at everyopen-label clinic visit.

The incidence of dyskinesia in relation to levodopa use was assessed, and wasdefined as the first episode of dyskinesia occurring 1) in the absence of levodopa or2) after starting levodopa. Dyskinesia in the absence of levodopa included partic-ipants who experienced dyskinesia but never received levodopa during the study,and participants whose first onset of dyskinesia occurred before their first dose oflevodopa. Therefore, participants who reported their first dyskinesia in the‘absence of levodopa’ may have subsequently received concomitant levodopa.Dyskinesia was also assessed by duration of PD at double-blind baseline (andduration of PD at the first episode of dyskinesia), and stratified by gender and age atdouble-blind baseline. The age cohorts (<50, 50e<65, 65e<75, and �75 years) arein line with age cut-offs used to identify age-related differences in AEs in studies ofrotigotine [21].

2.5. Statistical analyses

Analyses of pooled data from the two open-label studies are reported; analysesinclude participants randomized to ropinirole during the SP513 double-blind study(n ¼ 160). All participants who received at least one dose of open-label studymedication were included in the safety set. UPDRS IV analyses were performed onthe safety set, as observed. End of maintenance was defined as the last availablevalue from the maintenance period. Data were analyzed descriptively and nostatistical significance testing was performed.

3. Results

3.1. Participants

Of the 632 participants who completed double-blind treat-ment, 598 (95%) chose to enter the open-label extensions; 596participants received open-label rotigotine and were included inthe safety set. 299/596 (50%) participants were still in the studywhen rotigotine became commercially available or at the time ofclosure by the sponsor (4e6 years after the onset of open-labelrotigotine treatment); 145 (24%) withdrew prematurely due toAEs, 34 (6%) due to lack of efficacy, and 120 (20%) because ofother reasons (Fig. 1) Full patient disposition are reported in theprimary manuscripts [19,20]. No participant discontinued due todyskinesia. Pooled baseline demographic data are listed inTable 1.

3.2. Rotigotine exposure and levodopa use

The mean ± SD rotigotine dose at the end of treatment was7.2± 3.4mg/24 h (median 6mg/24 h) in SP702 and 8.2± 3.8mg/24 h(median 8 mg/24 h) in SP716. The median exposure duration torotigotine during SP702 was 1910 days (~5 years, 3 months; range1e2188 days), and during SP716 was 1564.5 days (~4 years,3 months; 5e2154 days).

A total of 23 (4%) participants who entered the open-labelstudies had received levodopa (for less than 6 months) prior tothe double-blind studies. During up to 6 years of open-label

blblblbl

l l

ll

l

ll

l

ll

l

l

l

ll

l

l

l l l

ll

l l

l

l

lll

l

Fig. 1. Study flow charts. *One participant received open-label medication in the final visit of the double-blind phase, but never returned to the clinic and was therefore excludedfrom the SP702 safety set. **One participant withdrew consent before receiving any open-label medication and was therefore excluded from the SP716 safety set. yStudy partic-ipation ended either when rotigotine became commercially available or when the open-label studies were closed by the sponsor in 2008; median exposure to rotigotine duringSP702 was 1910 days (~5 years, 3 months), and during SP716 was 1564.5 days (~4 years, 3 months). SS, safety set; ‘other’ ¼ unsatisfactory compliance/major protocol deviation/withdrawal of consent/lost to follow-up/other reasons.

N. Giladi et al. / Parkinsonism and Related Disorders 20 (2014) 1345e1351 1347

rotigotine treatment, 423 participants (71%) received levodopatreatment as add-on therapy (Table 2). The median dose ofconcomitant levodopa over the entire open-label studies was317.8 mg/day (range 50e3223 mg/day). The median time fromstart of open-label treatment to initiation of levodopa was 374days (~1 year; range 1e2019 days) in SP702, and 485 days (~1year, 4 months; 14e2076 days) in SP716. In addition, 88 partic-ipants (15%) received concomitant amantadine during the open-label treatment period. The reason for concomitant amantadinewas not reported.

Table 1Pooled baseline characteristics (SP702/716) (Safety Set, double-blind baseline).

Participants (n ¼ 596)

Age, mean years ±SD (range) 62.2 ± 10.1 (31e87)Age <50 years, n (%) 65 (11)Age 50e<65 years, n (%) 269 (45)Age 66e<75 years, n (%) 205 (34)Age �75 years, n (%) 57 (10)Male, n (%) 377 (63)Female, n (%) 219 (37)Caucasian, n (%) 575 (96)Duration of PD since diagnosis, mean ± SD

(median [range]) years1.3 ± 1.3(0.9 [0e7])

UPDRS Part II score, mean ± SD (range) 8.6 ± 4.2 (1e28)UPDRS Part III score, mean ± SD (range) 22.4 ± 9.1 (5e57)UPDRS Part II þ III total score, mean ± SD (range) 31.1 ± 12.1 (7e80)Prior or concomitant PD medication, n (%)a 300 (50)

UPDRS, Unified Parkinson's Disease Rating Scale.a Medication taken for at least 1 day before the study treatment period was

considered as prior (regardless of continuing or stopping the medication).Concomitant medication was any medication taken on date of first study drugadministration (regardless of continuing or stopping the medication).

3.3. Assessment of dyskinesia

3.3.1. Incidence of dyskinesiaDyskinesia was reported in 115 (19%) participants during the

open-label studies (Table 2). An assessment of the incidence ofdyskinesia during the open-label studies according to double-blindtreatment indicated that 35/156 (22%) participants who receiveddouble-blind placebo, 22/160 (14%) who received double-blindropinirole, and 58/280 (21%) participants who received double-blind rotigotine, reported dyskinesia during the open-labelstudies. As the duration of participation in the study could impactthe development of dyskinesia, the proportion of patientsremaining free of dyskinesia over the course of the study are shownin KaplaneMeier plots (Fig. 2); plots for the proportion of patientsremaining free of dyskinesia are presented for ‘all patients’ (Fig. 2A)and for ‘rotigotine only patients’ (i.e. in the absence of levodopa)(Fig. 2B). In addition, as the time from the start of levodopa use mayimpact the development of dyskinesia, a KaplaneMeier plot forproportion of patients remaining free of dyskinesia after the start oflevodopa is also presented (Fig. 3).

The majority of participants who experienced dyskinesia (78%[ie, 90 of the 115]) reported developing dyskinesia after initiatingconcomitant treatment with levodopa (Table 2). Among partici-pants with dyskinesia after starting levodopa, median levodopadose over the studies was 374.0 mg/day (range 70e1721 mg/day).The median time to reporting the onset of first dyskinesia afterinitiating levodopa therapy was 828 days (~2 years, 3 months;range 78e1861 days). Six of the 90 participants who reporteddeveloping dyskinesia after initiating levodopa had also previouslyreceived levodopa prior to the double-blind study.

Twenty-five (4%) participants reported dyskinesia in theabsence of levodopa. Two of these 25 participants had previously

Table 2Incidence of dyskinesia during up to 6 years of open-label rotigotine treatment and in relation to levodopa use.

All participantsn ¼ 596

Malen ¼ 377

Femalen ¼ 219

<50 yearsn ¼ 65

50e<65 yearsn ¼ 269

65e<75 yearsn ¼ 205

�75 yearsn ¼ 57

Received levodopa, n (% of participants) 423 (71%) 271 (72%) 152 (69%) 41 (63%) 190 (71%) 154 (75%) 38 (67%)Participants with dyskinesia, n (% of participants) 115 (19%) 61 (16%) 54 (25%) 23 (35%) 58 (22%) 29 (14%) 5 (9%)Dyskinesia in the absence of levodopa, n (% of participants)

[% of participants with dyskinesia]25 (4%)[22%]

12 (3%)[20%]

13 (6%)[24%]

6 (9%)[26%]

12 (4%)[21%]

6 (3%)[21%]

1 (2%)[20%]

Dyskinesia after starting levodopa, n[% of participants with dyskinesia]

90[78%]

49[80%]

41[76%]

17[74%]

46[79%]

23[79%]

4[80%]

All only l

l ll l

Fig. 2. KaplaneMeier plots for proportion of patients remaining free of dyskinesia over the course of the study. KaplaneMeier curve for: A) ‘All patients’ e time from start of open-label study to first dyskinesia (patients at risk are those remaining in the study), and B) ‘Rotigotine only patients’ (i.e. in the absence of levodopa) e time from start of open-labelstudy to first dyskinesia (patients at risk are those remaining in the study, and who have not started levodopa). The KaplaneMeier curves are calculated from the total number ofpatients at risk of dyskinesia at each time point.

l

l

Fig. 3. KaplaneMeier plot for proportion of patients remaining free of dyskinesia afterstart of levodopa use. KaplaneMeier curve for ‘Patients who received levodopa’ e timefrom start of levodopa use to first dyskinesia (patients at risk are those who havestarted levodopa, and who are remaining in the study). The KaplaneMeier curve iscalculated from the total number of patients at risk of dyskinesia at each time point.

N. Giladi et al. / Parkinsonism and Related Disorders 20 (2014) 1345e13511348

received levodopa prior to the double-blind study; therefore, 23participants who reported dyskinesia had never received levo-dopa. There was no apparent relationship between the incidenceof dyskinesia and dose of rotigotine (dose of longest durationduring the open-label studies): 1/19 (5%) participants whoreceived 2 mg/24 h, 0/43 (0%) who received 4 mg/24 h, 6/158(4%) who received 6 mg/24 h, 7/203 (3%) who received 8 mg/24 h, 1/47 (2%) who received 10 mg/24 h, 3/40 (8%) who received12 mg/24 h, 1/19 (5%) who received 14 mg/24 h, and 6/67 (9%)participants who received 16 mg/24 h reported dyskinesia in theabsence of levodopa. Of the 88 participants who receivedconcomitant amantadine, 25 reported dyskinesia during theopen-label studies (5 reported dyskinesia before starting treat-ment with amantadine, 13 reported the first incidence of dyski-nesia during treatment with amantadine, and 7 were no longerreceiving amantadine at the time of dyskinesia).

The median time to first onset of dyskinesia from initiation ofrotigotine (which was either in the double-blind studies or in theopen-label studies if the participant did not receive double-blindrotigotine) was 1204 days (~3 years, 3 months; range 0e2284days). For the 90 participants who reported dyskinesia after start-ing levodopa the median time to first onset of dyskinesia frominitiation of rotigotine was 1286.5 (~3 years, 6 months; range196e2284 days). For the 25 participants who reported dyskinesia inthe absence of levodopa themedian time to first onset of dyskinesiafrom initiation of rotigotine was 288 days (~9 months; range0e1927 days).

Fig. 4. Worst-case severity of dyskinesia over the course of the study*: A) Disability due to dyskinesia (UPDRS IV Item 33) and B) Duration of dyskinesia (UPDRS IV Item 32).*The worst-case severity of dyskinesia over the course of the study (Panel A. Disability and Panel B. Duration) was available for 107 of the 115 participants identified as developingdyskinesia; 8 participants reported dyskinesia during unscheduled visits, and their most severe case of dyskinesia was not captured. Participants who reported their first dyskinesiain the ‘absence of levodopa’ may have subsequently received concomitant levodopa. These participants are reported in the dyskinesia in the ‘absence of levodopa’ group, althoughthe worst-case severity of dyskinesia may have occurred while they were receiving levodopa.

N. Giladi et al. / Parkinsonism and Related Disorders 20 (2014) 1345e1351 1349

3.3.2. Severity of dyskinesiaDyskinesia severity data over the course of the study were

available for 107 of the 115 participants. The most severe case ofdyskinesia for each participant during the course of the study isshown in Fig. 4; 56 of the 107 (52%) participants reported dyski-nesia to be ‘not disabling’ for all occurrences during the study(Fig. 4A). Eighteen participants reported the worst case of dyski-nesia to be ‘moderately’ or ‘severely disabling’, which represented3% of the overall population. Most participants with dyskinesia(77/107 [72%]) reported the longest duration of dyskinesia to bebetween 1% and 25% of the day (Fig. 4B). There were no obviousdifferences in the worst-case severity of dyskinesia in relation tolevodopa use (Fig. 4A and B).

3.3.3. Dyskinesia by duration of PD, gender, and ageThe median duration of PD at double-blind baseline for the 115

participants who reported dyskinesia was 1.0 years (range 0e6years) and 0.9 years (0e7 years) for the 481 participants who didnot report dyskinesia. The median duration of PD at first onset ofdyskinesia was 1813 days (~5 years, 0 months; range 317e3940days).

Approximately the same proportion of men and womenreceived levodopa, and there were no marked differences inmedian levodopa dose (336.1 mg/day men vs. 299.6 mg/daywomen). Proportionally more women reported dyskinesia(Table 2), and sooner after starting dopaminergic treatment (roti-gotine or levodopa); time from onset of rotigotine treatment untilfirst dyskinesia: median 816.5 days (~2 years, 3 months [range0e2088 days]) in women vs. 1372 days (~3 years, 9 months[0e2284]) in men; time from onset of levodopa use until start ofdyskinesia: 714 days (~1 year, 11 months [range 78e1735 days]) inwomen vs.1019 days (~2 years, 10 months [range 93e1861 days]) inmen. Dyskinesia disability data were available for 52/54 womenwho experienced dyskinesia, and 55/61 men. Twenty of the 52(38%) women reported the disability to be ‘not disabling’ for alloccurrences compared with 36/55 (65%) of men.

There were no marked differences between the age subgroupsin the proportion of participants who received levodopa, the inci-dence of dyskinesia in relation to levodopa use (Table 2), or themedian levodopa dose (<50 years, 300.0 mg/day; 50e<65 years,317.9 mg/day; 65e<75 years, 307.4 mg/day; and �75 years,339.7 mg/day). There was a trend towards a decrease in the

incidence (Table 2) and severity of dyskinesia with increasing age;worst-case disability of dyskinesia was reported as ‘not disabling’or ‘mildly disabling’ in 15/21 (71%) of participants younger than 50years, 47/56 (84%) between 50 and 65 years, 22/25 (88%) between65 and 75 years, and 5/5 (100%) 75 years or older.

4. Discussion

In this post hoc analysis of two open-label extension studies inpatients with early-stage PD at study enrollment, incidence ofdyskinesiawas 19% during up to ~6 years' treatment with rotigotinetransdermal system. Most dyskinesias were ‘not disabling’ or‘mildly disabling’ and only 3% of participants reported dyskinesiathat was ‘moderately’ or ‘severely’ disabling.

The 19% incidence of dyskinesia observed in participants treatedwith rotigotine, with or without levodopa, for up to ~6 years issimilar to that observed in other prospective long-term studies oforal dopamine receptor agonists. In a 4-year study of pramipexoleand a 5-year study of ropinirole, each of which allowed levodopasupplementation, 24.5% [22] and 20% [23] of participants reporteddyskinesia. The majority (78%) of participants receiving rotigotinereported their first episode of dyskinesia after initiating concomi-tant levodopa, which is not surprising as levodopa is accepted asthe most important factor in inducing dyskinesia [24]. Althoughthere was no rotigotine monotherapy arm or levodopa-only arm inthe current analyses, the incidence of dyskinesia in participants inthe absence of levodopa was low (25/596; 4%) and similar to thatreported for ropinirole and pramipexole [22,23]. While 23 of the 25patients who experienced dyskinesia in the absence of levodopahad no history of prior levodopa exposure, a low incidence ofdyskinesia during dopamine receptor agonist monotherapy, andwith no prior levodopa treatment, has also been reported in pa-tients treated with pramipexole [23]. It is difficult, retrospectively,to identify the cause of dyskinesia in this small number of patients.In the current analyses, the time to onset of first dyskinesia afterinitiation of rotigotine was ~2.5 years earlier in participants whoreported dyskinesia in the absence of levodopa compared withthose reporting dyskinesia after levodopa. It may be possible thatthere is a subgroup of patients particularly sensitive to the devel-opment of dyskinesia, even on dopamine receptor agonist mono-therapy, and after a relatively short period of time. However, thesestudies were not designed as two-arm parallel trials with rotigotine

N. Giladi et al. / Parkinsonism and Related Disorders 20 (2014) 1345e13511350

monotherapy vs. levodopa; therefore temporal data are interpretedwith caution.

Dyskinesia can initially be mild, but may progress to becomequite violent and severely disabling. While the incidence of dyski-nesia appeared to be greater following levodopa treatment, dys-kinesias that were reported to be at least moderately disabling, orfor more than 25% of the day, were uncommon regardless ofwhether they were reported in the absence of levodopa or aftersupplemental levodopa. The impact of dyskinesia on participants'quality of life was not assessed in the current studies, but it is ofinterest to note that no participants withdrew from these long-term studies due to dyskinesia.

In addition to modifiable risk factors (eg, levodopa use), non-modifiable risk factors (including younger age at PD onset,increased severity of PD, female gender) may also influence theincidence of dyskinesia [24,25]. In the current analyses there wereno apparent differences in the duration of PD at double-blindbaseline between participants who reported dyskinesia and thosewho did not. However, as at least 50% of participants had beendiagnosed with PD for less than 1 year at baseline, we cannotconclude that the duration of PD has no influence on the incidenceof dyskinesia in participants treated with rotigotine.

Slightly more women than men reported dyskinesia (25%women vs. 16% men). An equal proportion of men and womenreceived levodopa, suggesting that these gender differences werenot solely explained by levodopa use. Women also appeared toexperience earlier onset of dyskinesia after starting either rotigo-tine (~1 year, 6 months earlier) or levodopa (~10 months earlier),and more disabling dyskinesia than men. Although female gendermay be associated with a milder disease course and longer lifeexpectancy [26], it has been reported as a risk factor for dyskinesia[24,25], and, in line with the current findings, female gender hasbeen associated with a significantly shorter time to onset ofdyskinesia from onset of treatment [27]. In the current patientpopulation, the median levodopa dose was slightly lower inwomen. When corrected for body weight, women have higherplasma levels of levodopa [24,28,29]. Therefore, women receivinglevodopamay be at a higher risk for dyskinesia perhaps due to theirlower body weight but not independent of other confoundingfactors [24]. In the current analyses there were no obvious differ-ences between women and men in the incidence of dyskinesia inrelation to rotigotine monotherapy or levodopa use.

Age at baseline appeared to have an influence on the incidenceof dyskinesia, with the greatest incidence occurring in youngerparticipants (35% of participants younger than 50 years), anddecreasing with increased age (9% of participants 75 years or older).The differences in dyskinesia incidence did not appear to beexplained by levodopa use per se. Although most participants re-ported their dyskinesia as ‘not disabling’ or ‘mildly disabling’, thedata also suggest that the worst-case severity of dyskinesiadecreased with increased age. Younger age has been identified asan important risk factor associated with increased occurrence ofdyskinesia [24]; 100% of young onset patients have been shown todevelop motor complications within 10 years of levodopa [30],whereas they are less common in patients with late-onset PD [31].

Because this study was not prospectively designed to assessdyskinesia, a number of study design issues limit the interpretationof these analyses. First, UPDRS IV (item 32 and 33) is commonlyused to assess dyskinesia [22,23]; using this tool the incidence/severity of dyskinesia are based on historical participant report,with the opportunity to modify responses following investigatorexamination. However, it is possible that the occurrence is under orover reported, as participants may not always correctly identifydyskinesia. Secondly, the adjunctive use of levodopa as neededduring the course of open-label follow-up prevents an estimate of

the incidence of dyskinesia with long-term rotigotine mono-therapy. Thirdly, the absence of a levodopa-only arm in either long-term study prevents comparison of the incidence and severity ofdyskinesia with levodopa in patients not initially treated withrotigotine. However, other studies of dopamine receptor agonistsincorporating levodopa-only arms suggest an expected dyskinesiarate with levodopa in excess of 45% over up to 5 years [22,23]. Inaddition, the relatively small number of participants in some of thecohorts in Table 2 limits the conclusions that can be drawn fromthese analyses. Finally, in addition to premature withdrawal fromthe study due to reasons such as AEs and lack of efficacy, approx-imately 50% of participants did not have the opportunity to com-plete the full 6 years of study participation, as participation endedeither when rotigotine became commercially available or when theopen-label studies were closed by the sponsor in 2008 due tomanufacturing problems leading to the formation of crystals in thepatch (now resolved).

Taking into account the limitations, these data from almost 600patients with PD followed for a maximum of 6 years (median of 5years, 3 months in SP702, and 4 years, 3 months in SP716), suggestthat there was a relatively low incidence of dyskinesia with long-term rotigotine treatment. It is of further note that dyskinesiawere generally not disabling or mildly disabling. Although notconclusive, these data are supportive of the concept that dopaminereceptor agonists or continuous dopaminergic stimulation maylimit the emergence of dyskinesia.

Acknowledgments

This study was supported by UCB Pharma, Brussels, Belgium. NirGiladi serves as Associate Editor for Journal of Neural Transmission,as a member of the Editorial Board for Current Treatment Options inNeurology, and Journal of Parkinson's Disease; serves as consultantfor Teva-Lundbeck, Intec Pharma, NeuroDerm, Armon Neuro-medical Ltd., and Pharma Two B; has received payment for lecturesfrom Teva-Lundbeck, Novartis and UCB Pharma; and receivesresearch support from Michael J. Fox Foundation for Parkinson'sResearch (Grant: AJ -6201, BDPD- 9555, PPMI- 108/2310), theNational Parkinson Foundation (PD in non-AJ- 1395), the EuropeanUnion 7th Framework Programme (V-TIME- 278169 and CuPiD-288586), and the Israel Science Foundation. Liesbet Ghys was anemployee of UCB Pharma, Brussels, Belgium, during the develop-ment of the manuscript. Erwin Surmann is an employee of UCBPharma, Monheim am Rhein, Germany. Babak Boroojerdi is anemployee of UCB Pharma, Raleigh, NC, USA, and receives stockoptions from this employment. Joseph Jankovic has receivedresearch funding and Center of Excellence Grants from AllerganInc., Allon Therapeutics, Biotie Therapies Inc., Ceregene Inc., CHDIFoundation, GE Healthcare, Huntington's Disease Society of Amer-ica, Huntington Study Group, Impax Pharmaceuticals, IpsenLimited, Lundbeck Inc., Michael J. Fox Foundation for Parkinson'sResearch, Medtronic, Merz Pharmaceuticals, National Institutes ofHealth; National Parkinson Foundation, St. Jude Medical, TevaPharmaceutical Industries Ltd, UCB Pharma, University of Roches-ter, and Parkinson Study Group; has received honoraria for servicesas a consultant or an advisory committee member from AllerganInc., Auspex Pharmaceuticals Inc., Ipsen Biopharmaceuticals Inc.,Lundbeck Inc., Merz Pharmaceuticals, Teva Pharmaceutical In-dustries Ltd, UCB Pharma, and US World Meds; has received roy-alties from Cambridge, Elsevier, Hodder Arnold, LippincottWilliams and Wilkins, and Wiley-Blackwell; and has served on thefollowing editorial boards and foundation advisory boards withoutfinancial compensation: (editorial boards) Medlink: Neurology,Expert Review of Neurotherapeutics, Neurology in Clinical Practice,The Botulinum Journal, PeerJ, Therapeutic Advances in Neurological

N. Giladi et al. / Parkinsonism and Related Disorders 20 (2014) 1345e1351 1351

Disorders, Neurotherapeutics, Tremor and Other Hyperkinetic Move-ments, Journal of Parkinson's Disease, UpToDate, (foundations)Michael J. Fox Foundation for Parkinson Research, Dystonia MedicalResearch Foundation, Benign Essential Blepharospasm ResearchFoundation, International Essential Tremor Foundation. The au-thors acknowledge Ging-Ging Li CMPP (Global Publications Man-ager, Movement& Sleep Disorders, UCB Pharma, Brussels, Belgium)for publication coordination. The authors also acknowledge EmilyThompson, PhD, Evidence Scientific Solutions, London, UK, forwriting and editorial assistance, which was contracted by UCBPharma, Brussels, Belgium.

References

[1] Jankovic J. Motor fluctuations and dyskinesias in Parkinson's disease: clinicalmanifestations. Mov Disord 2005;20(Suppl. 11):S11e6.

[2] Brotchie J, Jenner P. New approaches to therapy. Int Rev Neurobiol 2011;98:123e50.

[3] Pechevis M, Clarke CE, Vieregge P, Khoshnood B, Deschaseaux-Voinet C,Berdeaux G, et al. Effects of dyskinesias in Parkinson's disease on quality of lifeand health-related costs: a prospective European study. Eur J Neurol 2005;12:956e63.

[4] Ahlskog JE, Muenter MD. Frequency of levodopa-related dyskinesias andmotor fluctuations as estimated from the cumulative literature. Mov Disord2001;16:448e58.

[5] Calabresi P, Di Filippo M, Ghiglieri V, Tambasco N, Picconi B. Levodopa-induced dyskinesias in patients with Parkinson's disease: filling the bench-to-bedside gap. Lancet Neurol 2010;9:1106e17.

[6] Jenner P. Molecular mechanisms of L-DOPA-induced dyskinesia. Nat RevNeurosci 2008;9:665e77.

[7] Goto Y, Otani S, Grace AA. The Yin and Yang of dopamine release: a newperspective. Neuropharmacology 2007;53:583e7.

[8] Hauser RA. New considerations in the medical management of early Parkin-son's disease: impact of recent clinical trials on treatment strategy. Parkin-sonism Relat Disord 2009;15(Suppl. 3):S17e21.

[9] Jankovic J, Poewe W. Therapies in Parkinson's disease. Curr Opin Neurol2012;25(4):433e47.

[10] Jenner P. A novel dopamine agonist for the transdermal treatment of Par-kinson's disease. Neurology 2005;65:S3e5.

[11] Scheller D, Ullmer C, Berkels R, Gwarek M, Lübbert H. The in vitro receptorprofile of rotigotine: a new agent for the treatment of Parkinson's disease.Naunyn Schmiedebergs Arch Pharmacol 2009;379:73e86.

[12] Pfeiffer RF. A promising new technology for Parkinson's disease. Neurology2005;65:S6e10.

[13] Elshoff JP, Braun M, Andreas JO, Middle M, Cawello W. Steady-state plasmaconcentration profile of transdermal rotigotine: an integrated analysis ofthree, open-label, randomized, phase I multiple dose studies. Clin Ther2012;34:966e78.

[14] Schmidt WJ, Lebsanft H, Heindl M, Gerlach M, Gruenblatt E, Riederer P, et al.Continuous versus pulsatile administration of rotigotine in 6-OHDA-lesionedrats: contralateral rotations and abnormal involuntary movements. J NeuralTransm 2008;115:1385e92.

[15] Stockwell KA, Scheller D, Rose S, Jackson MJ, Tayarani-Binazir K, Iravani MM,et al. Continuous administration of rotigotine to MPTP-treated commonmarmosets enhances anti-parkinsonian activity and reduces dyskinesia in-duction. Exp Neurol 2009;219:533e42.

[16] Giladi N, Boroojerdi B, Korczyn AD, Burn DJ, Clarke CE, Schapira AH. Rotigotinetransdermal patch in early Parkinson's disease: a randomized, double-blind,controlled study versus placebo and ropinirole. Mov Disord 2007;22:2398e404.

[17] Jankovic J, Watts RL, Martin W, Boroojerdi B. Transdermal rotigotine: double-blind, placebo-controlled trial in Parkinson disease. Arch Neurol 2007;64:676e82.

[18] Watts RL, Jankovic J, Waters C, Rajput A, Boroojerdi B, Rao J. Randomized,blind, controlled trial of transdermal rotigotine in early Parkinson disease.Neurology 2007;68:272e6.

[19] Elmer LW, Surmann E, Boroojerdi B, Jankovic J. Long-term safety and tolera-bility of rotigotine transdermal system in patients with early-stage idiopathicParkinson's disease: a prospective, open-label extension study. ParkinsonismRelat Disord 2012;18:488e93.

[20] Giladi N, Boroojerdi B, Surmann E. The safety and tolerability of rotigotinetransdermal system over a 6-year period in patients with early-stage Par-kinson's disease. J Neural Transm 2013;120:1321e9.

[21] Oertel W, Lewitt P, Giladi N, Ghys L, Grieger F, Boroojerdi B. Treatment ofpatients with early and advanced Parkinson's disease with rotigotine trans-dermal system: age-relationship to safety and tolerability. Parkinsonism RelatDisord 2013;19:37e42.

[22] Holloway RG, Shoulson I, Fahn S, Kieburtz K, Lang A, Marek K, et al. Prami-pexole vs levodopa as initial treatment for Parkinson disease: a 4-year ran-domized controlled trial. Arch Neurol 2004;61:1044e53.

[23] Rascol O, Brooks DJ, Korczyn AD, De Deyn PP, Clarke CE, Lang AE. A five-yearstudy of the incidence of dyskinesia in patients with early Parkinson's diseasewho were treated with ropinirole or levodopa. 056 Study Group. N Engl J Med2000;342:1484e91.

[24] Sharma JC, Bachmann CG, Linazasoro G. Classifying risk factors for dyskinesiain Parkinson's disease. Parkinsonism Relat Disord 2010;16:490e7.

[25] Garcia-Ruiz PJ, Del Val J, Fern�andez IM, Herranz A. What factors influencemotor complications in Parkinson disease?: a 10-year prospective study. ClinNeuropharmacol 2012;35:1e5.

[26] Chillag-Talmor O, Giladi N, Linn S, Gurevich T, El-Ad B, Silverman B, et al.Estimation of Parkinson's disease survival in Israeli men and women, usinghealth maintenance organization pharmacy data in a unique approach.J Neurol 2013;260:62e70.

[27] Hassin-Baer S, Molchadski I, Cohen OS, Nitzan Z, Efrati L, Tunkel O, et al.Gender effect on time to levodopa-induced dyskinesias. J Neurol 2011;258:2048e53.

[28] Arabia G, Zappia M, Bosco D, Crescibene L, Bagala A, Bastone L, et al. Bodyweight, levodopa pharmacokinetics and dyskinesia in Parkinson's disease.Neurol Sci 2002;23(Suppl. 2):S53e4.

[29] Zappia M, Annesi G, Nicoletti G, Arabia G, Annesi F, Messina D, et al. Sexdifferences in clinical and genetic determinants of levodopa peak-dose dys-kinesias in Parkinson disease: an exploratory study. Arch Neurol 2005;62:601e5.

[30] Schrag A, Ben-Shlomo Y, Brown R, Marsden CD, Quinn N. Young-onset Par-kinson's disease revisitedeclinical features, natural history, and mortality.Mov Disord 1998;13:885e94.

[31] Kumar N, Van Gerpen JA, Bower JH, Ahlskog JE. Levodopa-dyskinesia inci-dence by age of Parkinson's disease onset. Mov Disord 2005;20:342e4.