ADVERTISEMENT

Issue Navigator

Volume 12 No. 01
Earn CME
Accepted Papers





Scientific Investigations

Post Hoc Analysis of Data from Two Clinical Trials Evaluating the Minimal Clinically Important Change in International Restless Legs Syndrome Sum Score in Patients with Restless Legs Syndrome (Willis-Ekbom Disease)

William G. Ondo, MD1; Frank Grieger, Dipl Stat2; Kimberly Moran, PhD3; Ralf Kohnen, PhD4; Thomas Roth, PhD5
1University of Texas Health Science Center-Houston, Houston, TX; 2UCB Pharma, Monheim am Rhein, Germany; 3UCB Pharma, Smyrna, GA; 4Formerly of RPS Research Germany, Nurnberg, Germany (deceased); 5Henry Ford Hospital, Detroit, MI

ABSTRACT

Study Objectives:

Determine the minimal clinically important change (MCIC), a measure determining the minimum change in scale score perceived as clinically beneficial, for the international restless legs syndrome (IRLS) and restless legs syndrome 6-item questionnaire (RLS-6) in patients with moderate to severe restless legs syndrome (RLS/Willis-Ekbom disease) treated with the rotigotine transdermal system.

Methods:

This post hoc analysis analyzed data from two 6-mo randomized, double-blind, placebo-controlled studies (SP790 [NCT00136045]; SP792 [NCT00135993]) individually and as a pooled analysis in rotigotine-treated patients, with baseline and end of maintenance IRLS and Clinical Global Impressions of change (CGI Item 2) scores available for analysis. An anchor-based approach and receiver operating characteristic (ROC) curves were used to determine the MCIC for the IRLS and RLS-6. We specifically compared “much improved vs minimally improved,” “much improved/very much improved vs minimally improved or worse,” and “minimally improved or better vs no change or worse” on the CGI-2 using the full analysis set (data as observed).

Results:

The MCIC IRLS cut-off scores for SP790 and SP792 were similar. Using the pooled SP790+SP792 analysis, the MCIC total IRLS cut-off score (sensitivity, specificity) for “much improved vs minimally improved” was −9 (0.69, 0.66), for “much improved/very much improved vs minimally improved or worse” was −11 (0.81, 0.84), and for “minimally improved or better vs no change or worse” was −9 (0.79, 0.88). MCIC ROC cut-offs were also calculated for each RLS-6 item.

Conclusions:

In patients with RLS, the MCIC values derived in the current analysis provide a basis for defining meaningful clinical improvement based on changes in the IRLS and RLS-6 following treatment with rotigotine.

Citation:

Ondo WG, Grieger F, Moran K, Kohnen R, Roth T. Post hoc analysis of data from two clinical trials evaluating the minimal clinically important change in international restless legs syndrome sum score in patients with restless legs syndrome (Willis-Ekbom Disease). J Clin Sleep Med 2016;12(1):63–70.


INTRODUCTION

Restless legs syndrome (RLS; Willis-Ekbom disease) is a chronic, circadian sensory-motor neurological disorder.1 Symptoms of RLS are typified by an urge to move the legs due to uncomfortable sensations that are worse at night or during periods of inactivity or rest.2 Although the pathophysiology of RLS remains elusive, animal models and studies with postmortem human tissue suggest that dopaminergic dysfunction coupled with an imbalance in brain and spinal cord iron metabolism may have a role in the etiology of the disease.3,4 The effectiveness of dopamine replacement with levodopa and dopamine agonists for RLS symptomology supports at least an indirect involvement of dopaminergic dysfunction, although this is still a controversial issue.

BRIEF SUMMARY

Current Knowledge/Study Rationale: The minimal clinically important change (MCIC) is defined as the minimum change in score that would be perceived as beneficial. This is largely unexplored in restless legs syndrome (RLS). This post hoc analysis of data from two double-blind, randomized clinical studies determined the MCIC for the international restless legs syndrome (IRLS) and restless legs syndrome 6-item questionnaire (RLS-6), two validated patient-reported scales used in the assessment of RLS treatment, in patients with moderate to severe RLS treated with the rotigotine transdermal system.

Study Impact: This post hoc analysis established the MCIC in IRLS and RLS-6 scores related to the ability to discern a clinically relevant improvement in RLS symptoms following treatment with rotigotine. Defining the MCIC for the IRLS and RLS-6 gives clinicians an anchor for determining the effectiveness of RLS therapy, and to better understand the different factors that may influence perceptions of clinical improvement.

Clinical studies in patients with RLS commonly use patient and clinician assessment scales such as the international restless legs syndrome study group rating scale (IRLSSG), clinical global impressions severity scale (CGI-1), clinical global impressions of change (CGI-2), and the restless legs syndrome 6-item questionnaire (RLS-6) as outcome measures. The international restless legs syndrome scale (IRLS) is a validated, patient-reported, disease-specific instrument for assessing RLS symptom severity and is regarded as the clinical standard for measuring RLS disease severity.5 The RLS-6 consists of six separate items that are used to assess the severity profile of RLS during the day and night, as well as daytime tiredness and sleep satisfaction. Although both scales provide useful metrics for measuring changes in RLS severity, determining whether a change in score for a given patient is clinically meaningful has never been evaluated.

A minimal clinically important change (MCIC) is defined as the minimum change in score that would be perceived as beneficial, and can be determined using an anchor-based approach, such as using the seven-point CGI-2 as a measure of perceptible change in disease status.6 The CGI-2 is a validated, generic scale that is familiar to healthcare providers, and provides an assessment of change (improvement or worsening) in a patient's basal condition and response to therapy.

Determining the MCIC in assessment scales such as the IRLS and RLS-6 is an important step in clinicians being able to evaluate benefits of therapeutic interventions for RLS. The efficacy of the rotigotine transdermal system in patients with moderate to severe idiopathic RLS versus placebo was established using the IRLS sum score in two 6-mo, randomized, double-blind studies conducted in Europe (SP790; NCT00136045) and in the USA (SP792; NCT00135993).7,8 Least square mean (SE) differences at end of maintenance (EoM) in IRLS sum score in favor of rotigotine compared to placebo, ranged from −2.3 (1.2) to −8.2 (1.3) at the approved rotigotine dosages of 1, 2, and 3 mg/24 h.7,8 Secondary outcome measures included the CGI-2 and RLS-6 scales.7,8 Data from these studies, therefore, provide an opportunity to determine and validate a MCIC for the IRLS and for the RLS-6 items. The objective of this analysis was to determine the MCIC in the IRLS and RLS-6 scales in patients with moderate to severe RLS treated with rotigotine, by analyzing the changes in scores corresponding to the CGI-2 categories “much improved versus minimally improved,” “much improved/very much improved versus minimally improved or worse,” and “minimally improved or better versus no change or worse.” This analysis focused on these comparisons because they represent the categories in which the clinician may best relate a change in IRLS score with a clinically meaningful benefit to the patient.

METHODS

SP790 and SP792 Study Design

This post hoc analysis was performed using data from two randomized, double-blind, placebo-controlled studies of rotigotine in adult patients (aged 18–75 y) with moderate to severe idiopathic RLS. Detailed methods have been reported previously.7,8 Briefly, eligibility criteria included a diagnosis of idiopathic RLS according to the IRLSSG,9 a positive response to previous dopaminergic therapy or no previous dopaminergic therapy, a baseline IRLS sum score ≥ 15 (at least “moderate severity”), and a CGI item 1 severity score (CGI-1) of ≥ 4 points (at least “moderately ill”). Patients were excluded if they had RLS secondary to iron deficiency, pregnancy, or end-stage renal disease, a history of sleep disorders other than RLS, other central nervous system disorders (Parkinson disease, dementia, progressive supranuclear palsy), or clinically relevant concomitant diseases (polyneuropathy, akathisia, claudication).7,8 Patients receiving dopamine receptor agonists were required to discontinue therapy for at least 4 w prior to the baseline visit whereas for previous levodopa therapy the washout was 1 w. It should be noted that SP790 included three rotigotine dose groups (1 mg/24 h, 2 mg/24 h, and 3 mg/24 h), and SP792 included four dose groups (0.5 mg/24 h, 1 mg/24 h, 2 mg/24 h, and 3 mg/24 h). The pooled SP790+SP792 analysis included data from only the 1 mg/24 h, 2 mg/24 h, and 3 mg/24 h dose groups. Following titration, patients received their randomized fixed dose of rotigotine (0.5 mg/24 h [SP792 only], 1, 2, or 3 mg/24 h rotigotine) or placebo for a 6-mo maintenance period.

Both studies used the IRLS sum score and the CGI-1 as coprimary efficacy outcomes. Secondary outcomes included the RLS-6 and CGI-2. The IRLS is a patient survey that rates 10 items from 0 to 4, generating a sum score of 0 (no symptoms) to 40 (very severe symptoms).10 Thus, the IRLS sum score reflects the sum of all 10 IRLS item subscores as a single score of 0 to 40. The RLS-6 severity scale uses 11 points to measure the severity of RLS symptoms over six individual items each rated from 0 to 10 (0 = not present; 10 = very severe): (1) satisfaction with sleep over the previous 7 d, the severity of RLS symptoms; (2) when falling asleep; (3) at night; (4) during the day when at rest; (5) during the day when active; (6) the degree of daytime sleepiness during the previous 7 d.11 The CGI-2 (global rating of change of condition) was assessed by the investigators as a secondary outcome. Patients were rated by the investigator as: (1) very much improved; (2) much improved; (3) minimally improved; (4) no change; (5) minimally worse; (6) much worse; (7) very much worse, in comparison with their baseline condition.12 The CGI was completed by an investigator who was blind to any other RLS assessments.7,8

Both studies were completed in accordance with the Declaration of Helsinki and the Good Clinical Practice guidelines. The protocols of both studies were reviewed and approved by the relevant institutional review boards and ethics committees. Written informed consent was obtained from all participating patients.

MCIC Analysis

Post hoc analyses were performed using data from SP790 and SP792 separately, and for pooled data from the two studies (SP790+SP792). All analyses were performed with data from SP790 and SP792 using the full analysis set (as defined in the initial studies: all randomized patients with at least one valid postbaseline value for the coprimary efficacy variables, IRLS sum score, and CGI-1) with the data as observed. For the purposes of this post hoc analysis, patients also were required to have at least one CGI-2 score.

The MCIC was defined as the mean change in IRLS score and RLS-6 among patients who were rated by the investigator as “much improved vs minimally improved,” “much improved/very much improved vs minimally improved or worse,” and “minimally improved or better versus no change or worse” on the CGI-2. Mean changes from baseline to EoM in IRLS sum score were determined for each CGI-2 score, and Pearson and Spearman correlation coefficients were calculated for the change in IRLS versus CGI-2. An anchor-based approach13 was used to determine MCIC for the IRLS sum scores and RLS-6. The independent standard, or anchor, for these analyses was CGI-2 ratings. Receiver operating characteristic (ROC) curves were calculated in order to determine the IRLS score cut-offs that best distinguished a “much improved versus minimally improved,” “much improved/very much improved versus minimally improved or worse,” and “minimally improved or better versus no change or worse.”

RESULTS

The mean (standard deviation, [SD]) changes from baseline to EoM in IRLS sum scores for patients randomized to rotigotine or placebo in each CGI-2 category are shown in Table 1. A greater proportion of patients taking rotigotine were categorized as “minimally improved,” “much improved,” or “very much improved” compared to patients taking placebo (Table 1) in SP790, SP792, and for the pooled SP790+SP792 analysis. For the rotigotine-treated patients, the mean (SD) change from baseline in IRLS score to achieve a CGI-2 response of “minimally improved,” was −8.2 (6.7) points in study SP790, −6.6 (6.1) points in study SP792, and −7.5 (6.9) points for the SP790+SP792 pooled analysis. In the placebo group, the mean (SD) change from baseline in IRLS score to achieve “minimally improved” on the CGI-2 was −10.7 (6.1), and −5.9 (5.5) for SP790 and SP792, and −7.7 (6.1) for the pooled analysis (Table 1).

IRLS sum score mean changes from baseline.

jcsm.12.1.63.t01.jpg

table icon
Table 1

IRLS sum score mean changes from baseline.

(more ...)

The MCIC IRLS cut-off scores corresponding to CGI-2 ratings for “much improved versus minimally improved,” “much improved/very much improved versus minimally improved or worse,” and “minimally improved or better versus no change or worse” for the rotigotine-treated patients and those taking placebo, are shown in Table 2. Although the ROC cut-offs were one to two points higher for the SP790 analysis than those calculated for the SP792 analysis, the MCIC cut-offs for each of the categories evaluated between SP790, SP792, and the pooled analysis were similar. Consequently, the results from the pooled SP790+SP792 analysis are described in detail. In addition, the MCIC cut-offs were higher for rotigotine than for placebo (Table 2). For the CGI-2 ratings for “much improved versus minimally improved” the corresponding MCIC in IRLS cut-off scores was −9 (0.69, 0.66) for the SP790+SP792 pooled analysis (Figure 1). For the CGI-2 ratings corresponding to “much improved/very much improved versus minimally improved or worse,” the IRLS cut-off scores for the pooled SP790+SP792 pooled analysis was −11 (0.81, 0.84) (Figure 2). The IRLS cut-off score corresponding to CGI-2 ratings of “minimally improved or better versus no change or worse” was −9 (0.79, 0.88) for the pooled SP790+SP792 analysis (Figure 3). Finally, the ROC MCIC cut-off in IRLS sum score for “minimally improved versus no change” in the SP790+SP792 pooled analysis was −5 points (0.72, 0.63) (Figure 4). For the patients randomized to rotigotine, the Pearson correlation coefficients between changes in IRLS sum scores and CGI-2 ratings were 0.67 (p < 0.0001) for SP790, 0.72 (p < 0.0001) for SP792, and 0.68 (p < 0.0001) for pooled. The Spearman correlation coefficients between changes in IRLS sum scores and CGI-2 ratings were 0.74 (p < 0.0001) for SP790, 0.72 (p < 0.0001) for SP792, and 0.71 (p < 0.0001) for pooled. In the placebo group, the Pearson coefficients between changes in IRLS sum scores and CGI-2 ratings were 0.77 (SP790), 0.69 (SP792), and 0.74 (pooled) (p < 0.0001 for all). The Spearman correlation coefficients between changes in IRLS sum scores and CGI-2 ratings were 0.83, 0.74, and 0.80 (all p < 0.0001).

Summary of minimal clinically important ROC cut-offs.

jcsm.12.1.63.t02.jpg

table icon
Table 2

Summary of minimal clinically important ROC cut-offs.

(more ...)

Minimal clinically important change receiver operating characteristic curves for the mean change in international restless legs syndrome sum scores for category “much improved versus minimally improved.”

Study SP790 (A), SP792 (B), and the SP790+SP792 pooled analysis (C).

jcsm.12.1.63a.jpg

jcsm.12.1.63a.jpg
Figure 1

Minimal clinically important change receiver operating characteristic curves for the mean change in international restless legs syndrome sum scores for category “much improved versus minimally improved.”

(more ...)

Receiver operating characteristic curves for minimal clinically important change in the mean change in international restless legs syndrome sum scores for category “much improved/very much improved versus minimally improved or worse.”

Study SP790 (A), SP792 (B), and the SP790+SP792 pooled analysis (C).

jcsm.12.1.63b.jpg

jcsm.12.1.63b.jpg
Figure 2

Receiver operating characteristic curves for minimal clinically important change in the mean change in international restless legs syndrome sum scores for category “much improved/very much improved versus minimally improved or worse.”

(more ...)

Minimal clinically important change receiver operating characteristic curves for the mean change in international restless legs syndrome sum scores for categories “minimally improved or better versus no change or worse.”

Study SP790 (A), SP792 (B), and the SP790+SP792 pooled analysis (C).

jcsm.12.1.63c.jpg

jcsm.12.1.63c.jpg
Figure 3

Minimal clinically important change receiver operating characteristic curves for the mean change in international restless legs syndrome sum scores for categories “minimally improved or better versus no change or worse.”

(more ...)

Receiver operating characteristic curves for minimal clinically important change in the mean change in international restless legs syndrome sum scores for category “minimally improved versus no change.”

Study SP790 (A); SP792 (B); and the SP790+SP792 pooled analysis (C).

jcsm.12.1.63d.jpg

jcsm.12.1.63d.jpg
Figure 4

Receiver operating characteristic curves for minimal clinically important change in the mean change in international restless legs syndrome sum scores for category “minimally improved versus no change.”

(more ...)

The MCIC ROC cut-offs and Pearson correlation coefficients for each RLS-6 item are shown in Table 3. Overall, MCIC cutoffs ranged from 0 to −4 for the six items in the RLS-6 scale for each of the categories analyzed. For the RLS-6 Item 2, severity of RLS symptoms when falling asleep, the mean (SD) change from baseline to EoM in the patients achieving “minimally improved” ranged from −2.3 (2.9) to −2.5 (3.7) in the individual study and pooled analyses; for patients achieving “very much improved” the scores ranged from −5.1 (2.7) to −5.5 (3.1). The MCIC ROC cut-offs for RLS-6 Item 2 for “much improved versus minimally improved” was −4 (0.47, 0.63) for SP790, −4 (0.55, 0.63) for SP792, and −4 (0.54, 0.62), for SP790+SP792 (Table 3). Similar results were obtained for the “much improved/very much improved versus minimally improved or worse,” and “minimally improved or better versus no change or worse” calculations (Table 3).

Minimal clinically important change ROC cut-off values.

jcsm.12.1.63.t03.jpg

table icon
Table 3

Minimal clinically important change ROC cut-off values.

(more ...)

DISCUSSION

These post hoc results from two clinical studies in patients with moderate to severe RLS determined the MCIC in IRLS and RLS-6 scores related to the investigator assessed clinically relevant improvement in RLS symptoms following treatment with rotigotine. Using the pooled dataset, the analyses determined that an IRLS MCIC cut-off score of −9 points differentiates “much improved” from “minimally improved,” −11 points differentiates patients reporting “much improved/very much improved vs minimally improved or worse,” a cut-off of −9 points defines “minimally improved or better versus no change or worse,” and a cut-off of −5 points differentiates “minimally improved versus no change.” The MCIC ROC cutoffs for RLS-6 ranged from 0 to −4. For the most part, MCIC cut-offs were in the 3- to 4-point range for most of the RLS-6 items, except for item 5 “severity of RLS symptoms during the day when active.” This may be due to the circadian nature of RLS symptoms being predominantly expressed at night and because of the increase in activity during daytime hours. Many clinical studies used the change from baseline to EoM in IRLS score to evaluate the efficacy of an RLS therapy versus placebo. However, statistical superiority of the active treatment over placebo may or may not translate into clinically meaningful improvement.

The individual studies with rotigotine (SP790 and SP792) reported a least square mean treatment difference of up to 8.2 points in IRLS scores between the highest dose of rotigotine (3 mg/24 h) and placebo.7,8 The results from the current analysis suggest that a reduction of at least 5 points in IRLS score is necessary for clinicians to distinguish a patient from “minimally improved” from “no change” and a reduction of 9 points to differentiates a patient from “much improved” from “minimally improved.” Thus, although the results of the initial studies have established the effectiveness of rotigotine compared to placebo in alleviating RLS symptoms, calculation of the MCIC, as provided in this study, provides the quantitative basis for achieving meaningful and observable patient improvement. Thus, defining the MCIC for the IRLS and RLS-6 gives clinicians an anchor for determining the effectiveness of RLS therapy. Future studies may institute similar measures to determine the MCIC in order to better understand the clinical implications of new treatment approaches for RLS.

Although the MCIC is defined as the smallest change in clinical status that would be perceived as a meaningful improvement,14,15 we sought to define the IRLS cut-offs at which physicians may assess their patients as “much improved/very much improved,” over “minimal improvement” from their pretreatment status, as this is often achieved by placebo and this cut-off is most emphasized by regulatory agencies. Based on the IRLS validation study by Walters et al.10 which demonstrated a strong correlation between the IRLS and the CGI severity scale, Allen14 has argued that the minimal clinically significant change in IRLS is 3 points. Inasmuch as the validation study by Walters et al.10 was not designed to calculate the MCIC based on treatment differences, the study may serve to compare MCIC analyses from studies evaluating improvements in RLS symptoms following different active treatments. Thus, it would be logical to expect that treatments shown to be efficacious vs placebo in alleviating RLS symptoms would have a MCIC greater than the 3 points as defined by Allen.14 Similar analyses have used the anchor-based approach with CGI-2 to derive the MCIC for the mean changes in the Unified Parkinson's Disease Rating Scale scores and changes in “off-time” following rasagiline, ropinirole, or pramipexole treatment of Parkinson disease.6,13,16 The current level of MCIC analysis, however, has not previously been performed for RLS studies.

In this analysis, the calculation of the MCIC for the categories chosen was based on the data from the two rotigotine pivotal studies. Analyses of data from the individual studies showed slight differences in MCIC scores. The MCIC can be influenced by a number of factors including the study design (presence or absence of placebo-controlled comparison), study duration, and study population.13 Indeed, in their assessment of the MCIC in patients treated for Parkinson disease, Hauser and Auinger noted that variations between the MCIC calculation in their study with that from previous investigations were likely due to differences between the types of studies included in the analyses.13 Thus, the results from the current MCIC analyses are very specific to the included study population, namely patients with moderate to severe RLS. In the current analysis, although the patient eligibility criteria were the same for each study, patients enrolled in SP790 had more severe disease compared to the SP792 study (mean [SD] of 28.1 [6.1] versus 23.3 [5.0]). Patients with more severe disease are expected to experience a greater potential scope for improvement than patients with less severe disease. Thus, a more severely affected patient population may lead to a higher calculated MCIC.13 Other important differences between SP790 and SP792 include the range of rotigotine doses and the number of patients naïve to rotigotine therapy. Study SP792 included a 0.5 mg/24 h rotigotine dose, whereas SP790 only included 1–3 mg/24 h dose groups. In SP792, only the 2 mg/24 h and 3 mg/24 h were found to substantially improve RLS sum scores from baseline compared with placebo, whereas significant treatment differences were observed for all three rotigotine dose groups versus placebo in SP790. However, although the 0.5 and 1 mg/24 h dose groups in SP792 were not statistically significant, the results of this analysis showing the MCIC-based ROC curve cutoffs for the CGI-2 categories of improvement suggest that patients may have perceived a clinically important improvement regardless of the statistical outcome.7,8 In their analysis of the MCIC in patients with Parkinson disease, Hauser and Auinger13 noted that more efficacious therapies led to a higher MCIC.13 This included obtaining higher calculated MCIC values for the active treatment group (i.e., patients treated with rotigotine) compared with placebo. The results in the current analysis are consistent with the assumption by Hauser and Auinger given that a greater number of the treatment groups in SP790 resulted in significant improvements versus placebo than in the SP792 study. This may explain, at least in part, why the ROC curves were 1 to 2 points higher for SP790 than SP792. In addition, in all but two categories (“much improved/ very much improved vs minimally improved or worse,” and “minimally improved or better versus no change or worse” in SP792), the calculated MCIC was higher for the rotigotine treatment group than for placebo.

Limitations in this study include those related to the initial studies and those related to post hoc analyses (retrospective vs prospective analysis). For example, the IRLS is limited by floor and ceiling effects which may limit the generalizability in the associations between the IRLS and CGI-2 in patients with differing disease severity.17 There also may have been variability between individuals reporting subjective ratings of severity with the IRLS. The data used in this analysis were from clinical studies in which the enrolled patients had to meet stringent eligibility requirements for entry into the study. Thus, the population of patients may not be fully representative of all patients with RLS. Similarly, the studies involved a fixed-dose regimen of rotigotine therapy. However, Hauser and Auinger have argued that mildly efficacious treatments will likely result in responses near the MCIC, whereas highly efficacious treatments may produce results above the MCIC, making it more difficult to identify the MCIC for that treatment. Consequently, if the patients were on a flexible dose, titrated to maximum efficacy, the MCIC may have been more difficult to identify.

In conclusion, this is the first analysis of clinical study data to determine the MCIC for the IRLS and RLS-6 in patients with moderate to severe RLS. The results of this analysis quantify the change in IRLS and RLS-6 score in which clinicians and patients may perceive improvement in disease status following rotigotine therapy for RLS symptoms. Therefore, the results of this analysis help to establish a range of values for the IRLS and RLS-6 in which clinicians may predict a clinically relevant improvement in patient status to occur when treating their patients with RLS with rotigotine. Additional analyses are needed to further define the MCIC in the IRLS and RLS-6 for patients with varying disease severities of RLS, and to better understand the different factors that may influence patients' perceptions of clinical improvement.

DISCLOSURE STATEMENT

This study was supported financially by UCB Pharma. Dr. Ondo was a speaker and consultant for TEVA, Allergan, Ipsen, UCB Pharma, Merz, Lundbeck, and Novartis. Frank Grieger is an employee of UCB Pharma and received stock and/or stock options from this employment. Dr. Moran is an employee of UCB Pharma and received stock and/or stock options from this employment. Dr. Kohnen (deceased) had received personal compensation from UCB Pharma for consulting services in an advisory board. Dr Kohnen qualifies as having made an authorship-worthy contribution to the paper; he knew the paper was being submitted and agreed to that prior to death. Dr. Roth served as a consultant for Abbott, Acadia, Acoglix, Actelion, Achemers, Alza, Ancil, Arena, AstraZeneca, BMS, Cephalon, Cypress, Dove, Elan, Eli Lilly, Evotec, Forest, GlaxoSmithKline, Hypnion, Johnson and Johnson, King, Ludbeck, McNeil, MediciNova, Merck, Neurim, Neurocrine, Neurogen, Novartis, Orexo, Organon, Orginer, Prestwick, Proctor and Gamble, Pfizer, Purdue, Sanofi-Aventis, Shire, Schering Plough, Sepracor, Sevier, Shire, Somaxon, Syrex, Takeda, TransOral, Vanda, Vivometrics, Wyeth, Yamanuchi, and Xenoport. He has served on a speakers bureau for Sanofi-Aventis and Takeda. He has received research support from Sunovion. This was a postanalysis of data from two 6-mo randomized, double-blind, placebo-controlled multicenter, multinational studies. Post hoc data analyses were completed at UCB Pharma, Monheim am Rhein, Germany; each author reviewed and revised the manuscript at their institution. The contents of this manuscript do not involve off-label or investigational use of the rotigotine transdermal system.

ABBREVIATIONS

GCI-1

Clinical Global Impressions severity scale

CGI-2

Clinical Global Impressions of change

EoM

end of maintenance

IRLSSG

International Restless Legs Syndrome Study Group Rating Scale

MCIC

minimal clinically important change

ROC

receiver operating characteristic

RLS

Restless Legs Syndrome

ACKNOWLEDGMENTS

The authors acknowledge medical writing assistance provided by Richard Fay, PhD, CMPP, Evidence Scientific Solutions, Philadelphia, PA. Financial support was provided by UCB Pharma. The authors acknowledge Jesse Fishman, PharmD (UCB Pharma, Smyrna, GA) for publication coordination.

REFERENCES

1 

Trenkwalder C, Hening WA, Walters AS, Campbell SS, Rahman K, Chokroverty S, authors. Circadian rhythm of periodic limb movements and sensory symptoms of restless legs syndrome. Mov Disord. 1999;14:102–10. [PubMed]

2 

International Restless Legs Syndrome Study Group. “2012 Revised IRLSSG Diagnostic Criteria for RLS.” Accessed August 8, 2014. http://irlssg.org/diagnostic-criteria/.

3 

Earley CJ, Connor J, Garcia-Borreguero D, et al., authors. Altered brain iron homeostasis and dopaminergic function in restless legs syndrome (Willis-Ekbom Disease). Sleep Med. 2014;15:1288–301. [PubMed]

4 

Ondo WG, Zhao HR, Le WD, authors. Animal models of restless legs syndrome. Sleep Med. 2007;8:344–8. [PubMed]

5 

Kohnen R, Allen RP, Benes H, et al., authors. Assessment of restless legs syndrome--methodological approaches for use in practice and clinical trials. Mov Disord. 2007;22:S485–94. [PubMed]

6 

Schrag A, Sampaio C, Counsell N, Poewe W, authors. Minimal clinically important change on the unified Parkinson's disease rating scale. Mov Disord. 2006;21:1200–7. [PubMed]

7 

Hening WA, Allen RP, Ondo WG, et al., authors. Rotigotine improves restless legs syndrome: a 6-month randomized, double-blind, placebo-controlled trial in the United States. Mov Disord. 2010;25:1675–83. [PubMed]

8 

Trenkwalder C, Benes H, Poewe W, et al., authors. Efficacy of rotigotine for treatment of moderate-to-severe restless legs syndrome: a randomised, double-blind, placebo-controlled trial. Lancet Neurol. 2008;7:595–604. [PubMed]

9 

Allen RP, Picchietti D, Hening WA, Trenkwalder C, Walters AS, Montplaisi J, authors. Restless legs syndrome: diagnostic criteria, special considerations, and epidemiology. A report from the restless legs syndrome diagnosis and epidemiology workshop at the National Institutes of Health. Sleep Med. 2003;4:101–19. [PubMed]

10 

Walters AS, LeBrocq C, Dhar A, et al., authors. Validation of the international restless legs syndrome study group rating scale for restless legs syndrome. Sleep Med. 2003;4:121–32. [PubMed]

11 

Kohnen R, Oertel W, Stiasny-Kolster K, Benes H, Trenkwalder C, authors. Severity rating of RLS: review of ten years experience with the RLS-6 scales in clinical trials. Sleep. 2003;26(Abstract Suppl):A342.

12 

National Institute of Mental Health. Guy W, editor. Clinical global impressions. ECDEU assessment manual for psychopharmacology. Rockville MD: National Institute of Mental Health, 1976:217–22.

13 

Hauser RA, Auinger P, authors. Determination of minimal clinically important change in early and advanced Parkinson's disease. Mov Disord. 2011;26:813–8. [PubMed]

14 

Allen RP, author. Minimal clinically significant change for the international restless legs syndrome study group rating scale in clinical trials is a score of 3. Sleep Med. 2013;14:1229. [PubMed]

15 

Rascol O, author. Defining a minimal clinically relevant difference for the unified Parkinson's rating scale: an important but still unmet need. Mov Disord. 2006;21:1059–61. [PubMed]

16 

Hauser RA, Gordon MF, Mizuno Y, et al., authors. Minimal clinically important difference in Parkinson's disease as assessed in pivotal trials of pramipexole extended release. Parkinsons Dis. 2014;2014:467131. [PubMed Central][PubMed]

17 

Wunderlich GR, Evans KR, Sills T, et al., authors. An item response analysis of the international restless legs syndrome study group rating scale for restless legs syndrome. Sleep Med. 2005;6:131–9. [PubMed]