OnabotulinumtoxinA for the Treatment of Poststroke Distal Lower Limb Spasticity: A Randomized Trial

OnabotulinumtoxinA for the Treatment of Poststroke Distal Lower Limb Spasticity: A Randomized Trial

Accepted Manuscript OnabotulinumtoxinA for the Treatment of Post-Stroke Distal Lower-Limb Spasticity: A Randomized Trial Theodore Wein, MD, Alberto Es...

2MB Sizes 0 Downloads 32 Views

Accepted Manuscript OnabotulinumtoxinA for the Treatment of Post-Stroke Distal Lower-Limb Spasticity: A Randomized Trial Theodore Wein, MD, Alberto Esquenazi, MD, Wolfgang H. Jost, MD, Anthony B. Ward, MD, Grace Pan, MS, Rozalina Dimitrova, MD PII:

S1934-1482(17)30072-2

DOI:

10.1016/j.pmrj.2017.12.006

Reference:

PMRJ 2037

To appear in:

PM&R

Received Date: 13 January 2017 Revised Date:

29 September 2017

Accepted Date: 11 December 2017

Please cite this article as: Wein T, Esquenazi A, Jost WH, Ward AB, Pan G, Dimitrova R, OnabotulinumtoxinA for the Treatment of Post-Stroke Distal Lower-Limb Spasticity: A Randomized Trial, PM&R (2018), doi: 10.1016/j.pmrj.2017.12.006. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

OnabotulinumtoxinA for the Treatment of Post-Stroke Distal Lower-

RI PT

Limb Spasticity: A Randomized Trial Theodore Wein, MD,1 Alberto Esquenazi, MD,2 Wolfgang H. Jost, MD,3 Anthony B. Ward, MD,4 Grace Pan, MS,5 Rozalina Dimitrova, MD5

McGill University, Department of Neurology and Neurosurgery, Montreal General

SC

1

Hospital, Montreal, QC, Canada; 2MossRehab Gait and Motion Analysis Laboratory,

M AN U

Elkins Park, PA, USA; 3University of Freiburg, Department of Neurology, Freiburg, Germany; 4Staffordshire University, Faculty of Health and North Staffordshire Rehabilitation Centre, Haywood Hospital, Stoke on Trent, UK; 5Allergan plc, Irvine, CA, USA

EP

TE D

Address correspondence to: Theodore Wein, MD, FRCPC Mc Gill University, L7 312 Montreal General Hospital 1650 Cedar Avenue Montreal, QC H3G 1A4 CANADA [email protected] (P) 514-938-5535 (F) 514-934-8265

AC C

Cover Title: REFLEX Primary Tables: 4

Figures: 3

Keywords: Muscle spasticity, onabotulinumtoxinA, stroke Taxonomys: Spasticity, Stroke, Randomized Clinical Tria Total word count: 3239 words (5000 max)

1

ACCEPTED MANUSCRIPT

Funding Source: Allergan plc (Dublin, Ireland)

RI PT

Device Status: no medical devices were used in this study.

These results were presented in part at AAPM&R 2015 and 2016 Annual Meetings:

SC

Esquenazi A, Wein T, Jost W, Ward A, Kwan T, Pan G, Dimitrova R.

OnabotulinumtoxinA treatment in adult patients with post-stroke lower limb spasticity:

M AN U

Results from a double-blind, placebo-controlled, phase 3 clinical trial. American Academy of Physical Medicine & Rehabilitation (AAPM&R) 2015 Annual Meeting, 2015. Esquenazi A, Geis C, Wein TH, Ward AB, Liu C, Dimitrova R. Muscle Selection Patterns for Injection of OnabotulinumtoxinA in Adult Patients With Post-Stroke Lower-Limb

TE D

Spasticity Influence Outcome: Results From a Double-Blind, Placebo-Controlled Phase 3 Clinical Trial. American Academy of Physical Medicine & Rehabilitation (AAPM&R)

EP

2016 Annual Meeting, 2016.

Patel A, Ward AB, Geis C, Liu C, Jost WH, Dimitrova R. Impact of Early Intervention

AC C

With OnabotulinumtoxinA Treatment in Adult Patients With Post-Stroke Lower Limb Spasticity. American Academy of Physical Medicine & Rehabilitation (AAPM&R) 2016 Annual Meeting, 2016.

2

ACCEPTED MANUSCRIPT

Abstract

2

Background: Post-stroke distal lower limb spasticity impairs mobility, limiting activities

3

of daily living, requiring additional caregiver time.

4

Objective: To evaluate the efficacy, safety, and sustained benefit of onabotulinumtoxinA

5

in adults with post-stroke lower limb spasticity (PSLLS).

6

Design: A multicenter, randomized, double-blind, phase 3, placebo-controlled trial.

7

Setting: 60 study centers across North America, Europe, Russia the United Kingdom, and

8

South Korea.

9

Patients: Adult patients (18 to 65 years of age) with PSLLS (Modified Ashworth Scale

M AN U

SC

RI PT

1

[MAS] ≥3) of the ankle plantar flexors and the most recent stroke ≥3 months prior to study

11

enrollment. .

12

Interventions: During the open-label phase, patients received ≤3 onabotulinumtoxinA

13

treatments (≤400 U) or placebo at approximately 12-week intervals. Treatments were into

14

the ankle plantar flexors (onabotulinumtoxinA 300 U into ankle plantar flexors; ≤100 U,

15

optional lower limb muscles).

16

Main Outcome Measurements: The double-blind primary endpoint was MAS change

17

from baseline (average score at weeks 4 and 6). Secondary measures included physician-

AC C

EP

TE D

10

1

ACCEPTED MANUSCRIPT

assessed Clinical Global Impression of Change (CGI), MAS change from baseline in

19

optional muscles, Goal Attainment Scale (GAS), and pain scale.

20

Results: Of 468 patients enrolled, 450 (96%) completed the double-blind phase and 413

21

(88%) completed the study. Small improvements in MAS observed with

22

onabotulinumtoxinA during the double-blind phase (onabotulinumtoxinA, –0.8; placebo, –

23

0.6, P=0.01) were further enhanced with additional treatments through week 6 of the third

24

open-label treatment cycle (onabotulinumtoxinA/onabotulinumtoxinA, –1.2;

25

placebo/onabotulinumtoxinA, –1.4). Small improvements in CGI observed during the

26

double-blind phase (onabotulinumtoxinA, 0.9; placebo, 0.7, P=0.01) were also further

27

enhanced through week 6 of the third open-label treatment cycle

28

(onabotulinumtoxinA/onabotulinumtoxinA, 1.6; placebo/onabotulinumtoxinA, 1.6).

29

Physician- and patient-assessed GAS scores improved with each subsequent treatment. No

30

new safety signals emerged.

31

Conclusions: OnabotulinumtoxinA significantly improved ankle MAS, CGI, and GAS

32

scores compared with placebo; improvements were consistent and increased with repeated

33

treatments of onabotulinumtoxinA over 1 year in patients with PSLLS.

34

Clinical Trial Registration URL:

35

https://clinicaltrials.gov/ct2/show/NCT01575054?term=NCT01575054&rank=1

36

Unique Identifier: NCT01575054

AC C

EP

TE D

M AN U

SC

RI PT

18

2

ACCEPTED MANUSCRIPT

Introduction

38

Approximately 20% to 40% of stroke survivors develop spasticity, the severity of which

39

depends on several factors and may be influenced by paresis, the degree of neurological

40

deficit and disability, the degree of neuropathic or nociceptive pain and the time since

41

onset.[1-4] Spasticity impairs activities of daily living, including self-care and ambulation,

42

and requires additional caregiver time and effort.[5] Lower limb spasticity is frequently

43

complicated by limited mobility and impaired motor function.[6] Limited research into the

44

humanistic burden among patients with lower-limb spasticity has demonstrated that

45

patients with this condition experience impairments and complications that reduce quality

46

of life.[6]

47

Post-stroke lower-limb spasticity (PSLLS) management includes physical modalities (eg,

48

physical therapy, positioning, orthotics, ultrasound therapy, magnetic stimulation, and

49

transcutaneous electrical stimulation), oral medications, and surgery.[7,8] However,

50

treatment patterns vary widely,[6] and some treatments are associated with significant side

51

effects or provide inadequate responses.[7,8]

52

OnabotulinumtoxinA (BOTOX®, Allergan plc, Dublin, Ireland) has been shown to reduce

53

upper-limb spasticity (measured by the Modified Ashworth Scale [MAS])[9] and is

54

approved by the US Food and Drug Administration (FDA) for this indication in adults.[10]

55

Patient-centered outcomes such as goal attainment (measured by the Goal Attainment

56

Scale [GAS]) and global benefit (reported by patients and investigators) have been

57

reported to significantly improve with onabotulinumtoxinA (relative to placebo) in patients

AC C

EP

TE D

M AN U

SC

RI PT

37

3

ACCEPTED MANUSCRIPT

with upper-limb spasticity.[9] OnabotulinumtoxinA was recently approved by the FDA for

59

treating lower-limb spasticity in adults.[10] To date, information in the medical literature

60

on the optimal dosing and duration of effect of onabotulinumtoxinA for PSLLS is limited.

61

The purpose of this clinical trial, which helped to inform the FDA approval decision, was

62

to investigate the efficacy and safety of onabotulinumtoxinA in patients with PSLLS of the

63

ankle over up to 4 treatment cycles.

AC C

EP

TE D

M AN U

SC

RI PT

58

4

ACCEPTED MANUSCRIPT

Methods

65

Study Design

66

This was a multicenter (60 sites in North America, Europe, Russia, the United Kingdom,

67

and Asia), randomized, double-blind, placebo-controlled phase 3 study of the efficacy and

68

safety of onabotulinumtoxinA injection for treating PSLLS of the ankle, with an open-label

69

extension (Figure 1). The double-blind phase included 7 study visits: screening (days –42

70

to –4), randomization/treatment (week 0/day 1), and follow-up (weeks 2, 4, 6, 8, and 12).

71

At screening, patients underwent a physical examination, vital sign measurement,

72

electrocardiogram, and laboratory work. In the open-label extension, onabotulinumtoxinA

73

was administered in up to 3 additional 12-week treatment cycles, with 2-week

74

posttreatment telephone follow-ups and clinic visits every 6 weeks. Study participation did

75

not exceed 60 weeks.

76

The study (clinical trial registration: NCT01575054) was approved by an Institutional

77

Review Board or Independent Ethics Committee and conducted in compliance with Good

78

Clinical Practice. Written informed consent was obtained from each patient.

79

Study Treatments

80

At the double-blind treatment visit (day 1), patients were randomly assigned (1:1, block

81

size=4) to receive intramuscular injection of onabotulinumtoxinA or placebo. An

82

automated interactive voice- or web-response system (IVRS/IWRS) managed

83

randomization and treatment assignment, based on a randomization scheme prepared by

84

Allergan Biostatistics, and provided medication kit numbers for each patient. Sites

AC C

EP

TE D

M AN U

SC

RI PT

64

5

ACCEPTED MANUSCRIPT

dispensed medication according to the instructions received from the IVRS/IWRS system.

86

Study drug (100 U onabotulinumtoxinA, 0.5 mg human albumin, and 0.9 mg sodium

87

chloride) and placebo (0.9 mg sodium chloride) were provided in sterile, vacuum-dried

88

form without preservative in identical packaging. Study personnel with no patient

89

interaction reconstituted the study drug with 4 ml preservative-free sterile saline (0.9%

90

sodium chloride) and filled the syringes for injection. The injector and patient were blinded

91

to treatment allocation. Treatment groups were designated as

92

“onabotulinumtoxinA/onabotulinumtoxinA” or “placebo/onabotulinumtoxinA” per

93

double-blind treatment assignment.

94

The dose for each muscle was evenly distributed across the number of specified injection

95

sites for that muscle, including 3 sites for each of the mandatory ankle muscles (ie, medial

96

and lateral gastrocnemius, soleus, tibialis posterior). An optional total additional dose of

97

≤100 U was injected into additional muscles (ie, flexor digitorum longus, flexor digitorum

98

brevis, flexor hallucis longus, extensor hallucis longus, rectus femoris), if clinically

99

indicated (doses permitted into each of the additional muscles are provided in Table 1 ).

EP

TE D

M AN U

SC

RI PT

85

Muscles were injected using instrumented muscle localization techniques (ie,

101

electromyography, electrical-stimulation, sonography), targeting the motor endplate

102

region.

103

Eligible patients who completed the 12-week double-blind phase entered the open-label

104

phase, in which they could receive ≤400 U of onabotulinumtoxinA at approximately 12-

105

week intervals. Targeted muscles for the open-label phase included all mandatory and

106

additional muscles in the double-blind phase plus the hamstrings. To receive treatment, the

AC C

100

6

ACCEPTED MANUSCRIPT

identified muscles required a MAS score of ≥1+ (excluding the toe muscles). Each muscle

108

had a maximum dose and number of injection sites Table 1).

109

Patients

110

Inclusion Criteria

111

Adults (18 to 85 years) with PSLLS (MAS score ≥3) with equinus (plantar flexion of the

112

ankle) or equinovarus foot deformity, and most recent stroke occurring ≥3 months before

113

screening were enrolled. Patients were botulinum toxin treatment-naive or treated with

114

botulinum toxin ≥20 weeks before study day 1 for spasticity in the study limb or ≥12

115

weeks before study day 1 for other indications. Patients receiving muscle relaxants or oral

116

medication for spasticity were on a stable dose for ≥2 months before study day 1; those

117

receiving anti-epileptic medications were on a stable dose for ≥1 month before study day 1

118

and were not permitted to have dose adjustments during the double-blind phase of the

119

study.

120

Exclusion Criteria

121

Patients were excluded if they had lower-limb spasticity with etiology other than stroke;

122

spasticity that required treatment in the contralateral leg; fixed ankle contractures in the

123

study leg (ie, MAS=4); profound atrophy of the muscles to be injected; or previous

124

surgical intervention, phenol block, ethanol block, or muscle afferent block before

125

screening in muscles eligible for treatment or ≤6 months before screening for any other

126

upper- or lower-limb muscles. Additionally, patients were excluded if they were

127

nonambulatory; had the study limb casted ≤6 months before study day 1 or planned to cast

AC C

EP

TE D

M AN U

SC

RI PT

107

7

ACCEPTED MANUSCRIPT

the limb during the double-blind phase; had an infection of the skin, soft tissue, or joint in

129

the injection area; were pregnant, or had a known allergy or sensitivity to study

130

medication.

131

Assessments

132

In the double-blind phase, patients were assessed for treatment efficacy using MAS (each

133

follow-up visit), Clinical Global Impression of Change by Physician (CGI; each follow-up

134

visit), GAS (weeks 8 and 12), pain scale (each follow-up visit), gait speed assessment

135

(weeks 6 and 12), Modified Tardieu Scale (MTS; each follow-up visit) for the ankle

136

flexors, toe spasticity assessment (each follow-up visit), and onset of spasticity symptom

137

relief (week 2). In the open-label extension, MAS, CGI, GAS, pain, gait speed, MTS, and

138

toe spasticity were assessed at treatment day 1, week 6, and week 12 (or study exit) for

139

each treatment cycle.

140

For the analysis only, MAS scores of 0, 1, 1+, 2, 3, or 4[11] were coded as grades 0, 1, 2,

141

3, 4, or 5, respectively; the original grading system was used for screening and evaluation.

142

CGI was rated by a trained physician at each assessment time point using a 9-point scale

143

indicating change from patient’s baseline status from −4 (very marked worsening) to +4

144

(very marked improvement). GAS was an individualized, goal-oriented measurement tool

145

rated by the physician and patient to track functional improvement[12]; scores ranged from

146

–3 (worse than start) to +2 (much more than expected, improvements clearly exceeded the

147

defined therapeutic goal) based on a 6-point scale. Active and passive goals were set by

148

patient and investigator at screening. An 11-point pain scale measured the average and

AC C

EP

TE D

M AN U

SC

RI PT

128

8

ACCEPTED MANUSCRIPT

worst pain experienced in the previous 48 hours during walking, scored from 0 (no pain) to

150

10 (pain as bad as can be imagined). Gait speed was assessed as the time to walk 10

151

meters (measured in seconds). MTS quantified spasticity of the ankle using the angle of

152

catch after a fast-speed stretch (V1), the passive range of motion after a slow-speed stretch

153

(V3), and the level of dynamic contracture (R2–R1).[13] Toe spasticity was measured

154

using a 7-point, physician-assessed scale (range, –3=markedly worse to 3=markedly better)

155

among patients receiving treatment in toe muscles. Patient-reported onset of spasticity

156

symptom relief was collected at the week 2 visit when the patient was asked if, and if so

157

when, they had noticed an effect on spasticity.

158

Safety was assessed in the double-blind and open-label phases via adverse event (AE)

159

reporting, physical examinations, vital sign measurement, and laboratory tests.

160

Endpoints

161

To capture the peak effect time, the primary endpoint was the average of weeks 4 and 6

162

change from baseline in MAS for the ankle plantar flexors. Secondary endpoints included

163

average of weeks 4 and 6 CGI physician rating, change from baseline to each assessment

164

time point in MAS for optional muscles, GAS (passive and active) by physician and

165

patient at weeks 8 and 12, and change from baseline to each assessment time point in

166

average and worst pain. Tertiary measurements included gait speed, MTS score, toe

167

spasticity, and onset of spasticity symptom reduction.

AC C

EP

TE D

M AN U

SC

RI PT

149

9

ACCEPTED MANUSCRIPT

Statistical Analysis

169

The sample size of 178 patients per treatment group was calculated to provide 98% overall

170

power based on a 2-sample t test at a significance level of 0.05 and an assumed between

171

treatment group difference in the primary endpoint of change in MAS ankle from baseline

172

averaged across weeks 4 to 6 of 0.3 (SD: 0.7). The number needed to treat (NNT) was

173

calculated by the inverse of the absolute difference between the treatment and placebo

174

response rates. Efficacy was evaluated using the intent-to-treat (ITT) population (according

175

to treatment assigned). Safety was analyzed based on all treated patients (according to

176

treatment received). For the primary efficacy endpoint, MAS was analyzed using an

177

analysis of covariance (ANCOVA) model including treatment and investigational site as

178

factors, with baseline ankle MAS and muscle group injected as covariates. In the presence

179

of a large sample size, the sampling distribution can be assumed to be normally distributed,

180

and ANCOVA was therefore used to adjust for baseline covariates in the analysis. The

181

proportion of patients with ≥1-grade reduction in MAS was analyzed by Pearson’s chi

182

square or Fisher’s exact test. Change from baseline in MAS for optional muscles was

183

analyzed by ANCOVA. Effect size was calculated using Cohen's d = M1–M2/σpooled;

184

where σpooled =√[(σ12+ σ22) /2].

185

CGI was analyzed by ANCOVA as in the primary analysis of MAS. GAS was analyzed by

186

ANCOVA; the proportion of patients with GAS goal achievement and goal improvement

187

was analyzed by Pearson’s chi-square test or Fisher’s exact test. Change from baseline in

188

average and worst pain was analyzed by ANCOVA. Other efficacy outcomes, including

189

gait speed, MTS, and assessment of toe spasticity, were also analyzed using ANCOVA

AC C

EP

TE D

M AN U

SC

RI PT

168

10

ACCEPTED MANUSCRIPT

with treatment as a factor, and baseline ankle MAS, muscles injected, and baseline value

191

(if applicable) as covariates. The proportion of patients who reported onset of spasticity

192

relief by week 2 was analyzed by Pearson’s chi square or Fisher’s exact test. The

193

correlation between MAS ankle scores and CGI was also examined.

194

Results

195

Disposition and Demographics

196

This study took place from May 23, 2012, to July 1, 2015. Screening visits were conducted

197

for 564 patients; 468 patients were enrolled and randomized to receive

198

onabotulinumtoxinA (n=233) or placebo (n=235; Figure 2) in the double-blind phase. Of

199

these, 447 patients (95.5%) entered the open-label extension and received

200

onabotulinumtoxinA. No significant differences in demographics or disease characteristics

201

were observed between the 2 treatment groups (Table 2). Most patients had a baseline

202

MAS score of 3 (onabotulinumtoxinA, 92.3%; placebo, 93.2%); the remaining patients had

203

a MAS score of 4.

204

Efficacy

205

The primary endpoint, mean change from baseline in MAS (average score of weeks 4 and

206

6), was superior for onabotulinumtoxinA (−0.81 [SD=0.874]) compared with placebo

207

(−0.61 [SD=0.835], P = .01); MAS mean change from baseline at individual time points

208

was also significantly improved among those receiving onabotulinumtoxinA at weeks 2, 4,

209

and 6 (P = .006 [d= –0.24], .007 [d= –0.22], and .03 [d= –0.21], respectively; Figure 3A).

210

Raw scores for change in MAS from baseline assessed at week 4 are presented in Figure

AC C

EP

TE D

M AN U

SC

RI PT

190

11

ACCEPTED MANUSCRIPT

4A. Sustained benefits of onabotulinumtoxinA were observed in the open-label phase, and

212

the placebo/onabotulinumtoxinA group showed marked reductions in MAS beginning in

213

open-label treatment cycle 1 (Figure 3A).

214

In the double-blind phase, more onabotulinumtoxinA-treated patients than placebo-treated

215

patients achieved a ≥1-grade decrease in MAS (all time points; P ≤ .04; Figure 3B; NNT =

216

8 at weeks 2 to 6, NNT=11 at weeks 8 and 12). Raw scores for change in MAS from

217

baseline as assessed at week 4 are presented in Figure 4B. In the open-label phase, marked

218

increases in the proportion of patients (>76% of all patients at week 6 for each of the 3

219

treatment cycles) achieving ≥1-grade decrease in MAS were observed (Figure 3B).

220

In the subset of patients receiving injections in optional muscles, reductions from baseline

221

in MAS scores in these muscles were generally greater for those receiving

222

onabotulinumtoxinA than placebo. Reductions from baseline in MAS in the flexor hallucis

223

longus muscle reached statistical significance at weeks 2 (P = .02, d= –0.39) and 4 (P =

224

.008, d= –0.44) of the double-blind period. During the open-label phase, the largest

225

changes from baseline in the MAS for the optional muscles were observed at week 6 of

226

each treatment cycle, and were at least as great as the largest change from baseline in the

227

onabotulinumtoxinA group during the double-blind phase.

228

Mean CGI by physician (average score of weeks 4 and 6) was superior for

229

onabotulinumtoxinA (0.86 [SD=0.953]) compared with placebo (0.65 [SD0.902], P = .01),

230

as was CGI at weeks 2, 4, and 6 (P = .003 [d=0.34], .04 [d=0.20], and .01 [d=0.20],

231

respectively; Figure 3C). During the open-label phase, CGI scores by physician continued

AC C

EP

TE D

M AN U

SC

RI PT

211

12

ACCEPTED MANUSCRIPT

to improve and were sustained across 3 treatment cycles. Correlations were observed

233

between MAS and CGI scores at all double-blind time points (all P < .001; open-label not

234

assessed; Table 3).

235

Significant differences favoring onabotulinumtoxinA over placebo were found in mean

236

GAS for the physician-assessed passive goal at week 12 (P = .04, d=0.22) and the patient-

237

assessed passive goal at weeks 8 and 12 (P = .04 [d=0.14] and .03 [d=0.21], respectively).

238

Mean GAS for physician- and patient-assessed active goals did not reach statistical

239

significance. A greater proportion of patients treated with onabotulinumtoxinA (vs

240

placebo) progressed (GAS ≥–1) toward physician-assessed active (P=0.008, NNT=9) and

241

passive (P = .04, NNT=11) goals at week 8, and achieved (GAS ≥0) physician-assessed

242

passive goal at week 12 (P = .04, NNT=11) (Figure 3D). Both physician- and patient-

243

assessed GAS for active and passive goals generally improved over the course of the OL

244

phase.

245

Pain analyses assessing average and worst pain change from baseline and degree of pain

246

reduction did not yield any significant differences between treatment groups at any time

247

point in the double-blind phase, and generally improved over the 3 treatment cycles in the

248

open-label phase. Similarly, gait speed and MTS change from baseline were not

249

significantly different between treatment groups at any double-blind time point. In the

250

double-blind phase, toe spasticity scores showed statistically significant improvements

251

favoring onabotulinumtoxinA at all time points for the flexor hallucis longus muscle

252

(P≤0.03), all time points except week 12 for flexor digitorum longus and flexor digitorum

253

brevis muscles (P≤0.03), and at no time points for the extensor hallucis muscle. At week 2,

AC C

EP

TE D

M AN U

SC

RI PT

232

13

ACCEPTED MANUSCRIPT

a greater proportion of patients receiving onabotulinumtoxinA (vs placebo) reported

255

noticing an effect of treatment on the spasticity of the study leg since their last injection

256

(onabotulinumtoxinA, 44.3% [101/228]; placebo, 33.2% [77/232]; P = .01). In the open-

257

label phase, pain analyses, gait speed, MTS assessments, and toe spasticity scores

258

generally improved over the 3 treatment cycles.

259

Safety

260

No new safety signals were identified. Four patients (onabotulinumtoxinA, n=2; placebo,

261

n=2) discontinued the study before receiving treatment and were therefore not included in

262

the safety analysis. Across all treatment cycles, any treatment-emergent AE (TEAE) was

263

reported by 66.7% (154/231) of patients receiving onabotulinumtoxinA and 52.2%

264

(118/226) of those receiving placebo; most were mild or moderate in severity and deemed

265

by investigators to be unrelated to treatment (Table 4). TEAEs occurring in ≥2% of

266

patients in either the onabotulinumtoxinA group or placebo group during the double-blind

267

phase are shown in Table 4.

268

Across all treatment cycles, the incidence of AEs adjudicated by the investigator to be

269

related to treatment (ie, treatment-related AEs; TRAEs) was 8.5% (39/457) and decreased

270

with each treatment cycle. Three TRAEs were considered severe intensity

271

(onabotulinumtoxinA/onabotulinumtoxinA, n=3: 1 each of injection site pain, injection site

272

mass, and muscular weakness; placebo/onabotulinumtoxinA, n=1 injection site pain). No

273

deaths occurred in the double-blind phase; 5 deaths were observed in the open-label phase,

AC C

EP

TE D

M AN U

SC

RI PT

254

14

ACCEPTED MANUSCRIPT

none of which was considered by the investigator to be treatment-related. No serious

275

TRAEs were reported.

276

Discussion

277

The primary efficacy endpoint, average of weeks 4 and 6 change from baseline in ankle

278

MAS, was met for the ITT population, demonstrating the efficacy of onabotulinumtoxinA

279

for in reducing ankle spasticity in adults with PSLLS. These effects were sustained and

280

further improved during the open-label extension. These findings support the outcomes

281

reported by others after single doses onabotulinumtoxinA [14], even in patients with less

282

severe spasticity (MAS≥1) than the patients included in this study.[15,16] A corresponding

283

significant improvement in a key secondary endpoint, CGI as assessed by physician

284

(average of weeks 4 and 6), was reported for the onabotulinumtoxinA group (vs placebo);

285

also supporting the outcomes reported by others.[14] Ankle MAS and CGI assessed by

286

physician were strongly correlated at all time points, indicating that, although the

287

improvements in ankle tone were relatively small, they were associated with observed

288

clinical improvements, thereby suggesting that observed muscle tone reductions were

289

clinically meaningful. Injections into the toe flexors also resulted in significant treatment

290

benefits (data not shown); thus, including these muscles in the standard injection pattern

291

for individuals with ankle spasticity should be strongly considered.

292

Several other secondary endpoints achieved significance at 1 or more time points;

293

however, no between-group differences were observed for pain outcomes, gait speed, and

294

MTS measures. Similarly, despite observing an increase in gait speed after

AC C

EP

TE D

M AN U

SC

RI PT

274

15

ACCEPTED MANUSCRIPT

onabotulinumtoxinA 300 U, Kaji and colleagues[14] found no significant difference

296

between onabotulinumtoxinA and placebo in gait speed 4 weeks after treatment. In a

297

subgroup analysis of a study that enrolled patients with MAS≥1, those patients with the

298

most severe ambulatory incapacity (ie, house-bound patients) had the greatest

299

improvements in gait speed following treatment with onabotulinumtoxinA in association

300

with extensive inpatient rehabilitation.[15] Interestingly, when mobility was assessed using

301

a 6-minute walking test, a significant improvement in walking distance was observed for

302

onabotulinumtoxinA 200 U versus placebo, even in patients with less severe spasticity

303

(MAS≥1).[16] The Fugl-Meyer score has been used to assess the effect of botulinum toxin

304

on lower limb function, with a meta-analysis indicating botulinum toxin therapy improves

305

the Fugl-Meyer score compared with placebo.[17] As the Fugl-Meyer score correlates

306

closely with measures of spasticity such as the MAS,[18] it was not assessed in this study.

307

Data from the open-label extension indicated that repeated treatment with

308

onabotulinumtoxinA had positive and sustained effects across multiple efficacy

309

parameters. In both the onabotulinumtoxinA/onabotulinumtoxinA and the

310

placebo/onabotulinumtoxinA groups, improvements in muscle tone and global clinical

311

response were observed in all repeat cycles, and the magnitude of the improvement

312

generally increased with repeated treatments of onabotulinumtoxinA. This incremental

313

benefit with each treatment cycle is similar to that observed with other open-label studies

314

of botulinum toxins.[19,20]

315

Interestingly, during the double-blind phase of our study approximately 50% of patients on

316

onabotulinumtoxinA improved by ≥1 MAS grade after 4 to 6 weeks compared with

AC C

EP

TE D

M AN U

SC

RI PT

295

16

ACCEPTED MANUSCRIPT

approximately 80% of patients starting on onabotulinumtoxinA in the open label phase.

318

The difference in outcome for these 2 patient groups may be due to the differences in

319

treatment approach in the open-label phase of the study. Importantly, in the open-label

320

phase of the study, treatment was individualized and optimized by the investigators. For

321

example, in addition to the mandatory and optional muscles used for treatment in the

322

double-blind phase, the hamstrings could also be injected in the open-label phase.

323

Investigators were not required to inject into the mandatory muscles in the open-label

324

phase and the maximum dose of 400 U could be injected across all permitted muscles at

325

the discretion of the investigator, as long as the dose remained under the maximum

326

allowable dose per muscle site. As the percentage of patients in the

327

onabotulinumtoxinA/onabotulinumtoxinA group experiencing an improvement of ≥1 MAS

328

grade increased further at week 4 to 6 of the open-label phase, it would suggest that the

329

outcome is better when investigators are given more discretion to target treatment to the

330

affected muscles, as would be more typical in clinical practice. In addition, other factors

331

related to the open-label study design (eg, patients and physicians knowledge that the

332

patient was receiving active drug) may have contributed to the differences in outcome

333

observed between the double-blind and open label phases.

334

Reported AEs were consistent with the existing spasticity AE profile for

335

onabotulinumtoxinA.[17,21-23] While two-thirds of patients receiving

336

onabotulinumtoxinA/onabotulinumtoxinA and more than half of patients receiving

337

placebo/onabotulinumtoxinA had a TEAE, the percentage of patients with an AE

AC C

EP

TE D

M AN U

SC

RI PT

317

17

ACCEPTED MANUSCRIPT

considered by the investigator to be related to treatment (TRAE) was much lower

339

(onabotulinumtoxinA/onabotulinumtoxinA: 10.0%; placebo/onabotulinumtoxinA: 7.1%).

340

Several study strengths are thought to allow these results to be generalizable to the

341

poststroke population with ankle spasticity. This large, well-controlled study followed by

342

an open-label extention period was the first to validate onabotulinumtoxinA doses ranging

343

from 300 U to 400 U in several muscle groups. This can be valuable to physicians when

344

individualizing treatment and provides evidence for the longer-term tolerability.

345

Furthermore, the treatment paradigm (ie, dose administered and muscles injected) in this

346

study was based on doses and injection sites used previously;[10,14,24-28] a consideration

347

of previous treatments suggested to us that the dose per muscle and range of dose per

348

treatment session used in this study would be safe and well tolerated.

349

However, there are a few limitations to be considered when interpreting these results.

350

Allowing for a range of total dose per session (ie, 300 U−400 U) by the addition of

351

optional muscles precluded the randomization of patients based on the total dose received.

352

Furthermore, the second phase of the trial was open-label and non-blinded. Whereas this is

353

a well accepted design for assessing efficacy of repeat treatment and long-term safety in

354

clinical trials,[19,29] this approach may be subject to potential bias due to the lack of

355

blinding of the open-label phase. The lack of significance in pain outcomes in this study is

356

noteworthy because of its contrast with previous spasticity studies (eg, Botox Economic

357

Spasticity Trial [BEST, NCT00549783]) showing a decrease in pain in individuals injected

358

with onabotulinumtoxinA.[30] The lack of positive effect on pain may reflect nonspecific

359

wording of the pain assessment question (ie, “Please circle the one number [0 to 10] that

AC C

EP

TE D

M AN U

SC

RI PT

338

18

ACCEPTED MANUSCRIPT

best describes your pain in the last 48 hours when you walk”), which may have been

361

influenced by pain not directly associated with lower-limb spasticity. In addition, the

362

findings observed on the GAS have some inherent limitations including lack of goal

363

standardization and also challenges clinicians in predicting the likely outcome of treatment.

364

Conclusions

365

This study supports the use of onabotulinumtoxinA in the treatment of PSLLS. Patient-

366

reported improvement in PSLLS symptoms and clinically meaningful improvement in

367

muscle tone were observed as early as week 2 after a single onabotulinumtoxinA

368

treatment, along with significant improvements in function, as reflected by gains in the

369

goal achievement scale. No new safety signals were identified. The improvements

370

produced by onabotulinumtoxinA observed in this study may improve patients’ quality of

371

life and reduce the burden of post-stroke spasticity.

AC C

EP

TE D

M AN U

SC

RI PT

360

19

ACCEPTED MANUSCRIPT

Acknowledgments

373

Writing and editorial assistance were provided to the authors by Amanda M. Kelly, MPhil,

374

MSHN, and Mike Zbreski, PharmD, of Complete Healthcare Communications, LLC (West

375

Chester, PA) and Dana Franznick, PharmD and was funded by Allergan plc (Dublin,

376

Ireland).

SC

RI PT

372

377

Principal investigators for the REFLEX study included: Ziyad Ayyoub, MD; Anthony

379

Hong, MD; Alberto Esquenazi, MD; Carolyn Geis, MD; Mitchell Isaac, MD; Richard

380

Harvey, MD; Allison Brashear, MD; Atul Patel, MD; James Renfroe, MD; Vu Nguyen,

381

MD; Bruce Rubin, MD; Alvaro Padilla, MD; John Kelemen, MD; Fatma Gul, MD; John

382

Wald, MD; Ralph Gonzalez, MD; Peter Moore, MD FRCP; Anthony Ward, MD FRCP;

383

Dr. Faraz Jeddi; Theodore Wein, MD; Caroline Quartly, MD; Anra Saric, MD; Satyendra

384

Sharma, MD; Dr. Stanislav Vohanka; Dr. Ondrej Skoda; Dr. David Skoloudik; Dr. Eduard

385

Minks; Dr. Iva Stetkarova; Dr. Kirsten Zeuner; Dr. Wolfgang Jost; Dr. Andreas Ceballos-

386

Baumann; Dr. Tobias Wachter; Dr. Bernhard Haslinger; Prof. Martin Hecht; Dr. Mark

387

Obermann; Dr. Chi W. Ip; Dr. Sebastian Paus; Dr. Axel Schramm; Dr. Matthias Maschke;

388

Dr. Markus Ebke; Dr. Tibor Kovacs; Dr. Zoltan Denes; Dr. Geza Szabo; Dr. Attila Csanyi;

389

Dr. Monika Szots; Dr. Katalin Bihari; Dr. Attila Valikovics; Dr. Anna Szczepanska-Szerej;

390

Prof. Andrzej Bogucki; Dr. Monika Rudzinska; Dr. Dariusz Koziorowski; Dr. Stanislaw

391

Ochudlo; Dr. Slawomir Zaborek; Han Gil Seo, MD; Min Ho Chun, MD; Deog Young

392

Kim, MD; Nam-Jong Paik, MD; Alexander Skoromets; Dr. Sofia Timerbaeva; and Prof.

AC C

EP

TE D

M AN U

378

20

ACCEPTED MANUSCRIPT

Dmitry Pokhabov. The authors would also like to thank the patients for participation in the

394

study.

AC C

EP

TE D

M AN U

SC

RI PT

393

21

ACCEPTED MANUSCRIPT

References 1.

Wissel J, Schelosky LD, Scott J, Christe W, Faiss JH, Mueller J. Early

Neurol 2010; 257(7):1067-1072. 2.

RI PT

development of spasticity following stroke: a prospective, observational trial. J

Kong KH, Chua KS, Lee J. Symptomatic upper limb spasticity in patients with

predictors. J Rehabil Med 2010; 42(5):453-457.

Thilmann AF, Fellows SJ, Garms E. The mechanism of spastic muscle hypertonus.

M AN U

3.

SC

chronic stroke attending a rehabilitation clinic: frequency, clinical correlates and

Variation in reflex gain over the time course of spasticity. Brain 1991; 114 ( Pt 1A):233-244. 4.

Paolucci S, Martinuzzi A, Scivoletto G, et al. Assessing and treating pain

TE D

associated with stroke, multiple sclerosis, cerebral palsy, spinal cord injury and spasticity. Evidence and recommendations from the Italian Consensus Conference on Pain in Neurorehabilitation. Eur J Phys Rehabil Med 2016; 52(6):827-840. Doan QV, Brashear A, Gillard PJ, et al. Relationship between disability and health-

EP

5.

related quality of life and caregiver burden in patients with upper limb poststroke

6.

AC C

spasticity. PM R 2012; 4(1):4-10.

Khalil M, Zafar HW, Quarshie V, Ahmed F. Prospective analysis of the use of OnabotulinumtoxinA (BOTOX) in the treatment of chronic migraine; real-life data

in 254 patients from Hull, U.K. J Headache Pain 2014; 15:54.

22

ACCEPTED MANUSCRIPT

7.

Jasmin L, Burkey AR, Card JP, Basbaum AI. Transneuronal labeling of a nociceptive pathway, the spino-(trigemino-)parabrachio-amygdaloid, in the rat. J

RI PT

Neurosci 1997; 17(10):3751-3765. 8.

OECD. OECD Econmic Surveys: Sweden 2015: OECD Publishing. 2015.

9.

Turner-Stokes L, Baguley IJ, De Graaff S, et al. Goal attainment scaling in the

SC

evaluation of treatment of upper limb spasticity with botulinum toxin: a secondary analysis from a double-blind placebo-controlled randomized clinical trial. J Rehabil

10.

M AN U

Med 2010; 42(1):81-89.

BOTOX® (onabotulinumtoxinA) for injection, for intramuscular, intradetrusor, or intradermal use (package insert). Irvine, CA: Allergan plc; 2016.

11.

Bohannon RW, Smith MB. Interrater reliability of a modified Ashworth scale of muscle spasticity. Phys Ther 1987; 67(2):206-207.

Borg J, Ward AB, Wissel J, et al. Rationale and design of a multicentre, double-

TE D

12.

blind, prospective, randomized, European and Canadian study: evaluating patient

EP

outcomes and costs of managing adults with post-stroke focal spasticity. J Rehabil Med 2011; 43(1):15-22.

Ansari NN, Naghdi S, Hasson S, Rastgoo M, Amini M, Forogh B. Clinical

AC C

13.

assessment of ankle plantarflexor spasticity in adult patients after stroke: inter-and intra-rater reliability of the Modified Tardieu Scale. Brain Inj 2013; 27(5):605-612.

14.

Kaji R, Osako Y, Suyama K, et al. Botulinum toxin type A in post-stroke lower limb spasticity: a multicenter, double-blind, placebo-controlled trial. J Neurol 2010; 257(8):1330-1337.

23

ACCEPTED MANUSCRIPT

15.

Hara T, Abo M, Hara H, et al. Effects of botulinum toxin A therapy and multidisciplinary rehabilitation on upper and lower limb spasticity in post-stroke

16.

RI PT

patients. Int J Neurosci 2017; 127(6):469-478. Tao W, Yan D, Li JH, Shi ZH. Gait improvement by low-dose botulinum toxin A injection treatment of the lower limbs in subacute stroke patients. J Phys Ther Sci

17.

SC

2015; 27(3):759-762.

Wu T, Li JH, Song HX, Dong Y. Effectiveness of Botulinum Toxin for Lower

M AN U

Limbs Spasticity after Stroke: A Systematic Review and Meta-Analysis. Top Stroke Rehabil 2016; 23(3):217-223. 18.

Katz RT, Rovai GP, Brait C, Rymer WZ. Objective quantification of spastic hypertonia: correlation with clinical findings. Arch Phys Med Rehabil 1992; 73(4):339-347.

Santamato A, Panza F, Intiso D, et al. Long-term safety of repeated high doses of

TE D

19.

incobotulinumtoxinA injections for the treatment of upper and lower limb

20.

EP

spasticity after stroke. J Neurol Sci 2017; 378:182-186. Gracies J-M, Esquenazi A, Brashear A, et al. Duration of effect of

AC C

abobotulinumtoxinA (Dysport®) in adult patients with lower limb spasticity post stroke or traumatic brain injury. Presented at: TOXINS, January 18–21, 2017; Madrid, Spain.

21.

Simpson DM, Alexander DN, O'Brien CF, et al. Botulinum toxin type A in the

treatment of upper extremity spasticity: a randomized, double-blind, placebocontrolled trial. Neurology 1996; 46(5):1306-1310.

24

ACCEPTED MANUSCRIPT

22.

Brashear A, Gordon MF, Elovic E, et al. Intramuscular injection of botulinum toxin for the treatment of wrist and finger spasticity after a stroke. N Engl J Med 2002;

23.

RI PT

347(6):395-400. Rosales RL, Efendy F, Teleg ES, et al. Botulinum toxin as early intervention for

Sci 2016; 371:6-14. 24.

Grazko MA, Polo KB, Jabbari B. Botulinum toxin A for spasticity, muscle spasms,

M AN U

and rigidity. Neurology 1995; 45(4):712-717. 25.

SC

spasticity after stroke or non-progressive brain lesion: A meta-analysis. J Neurol

Leon SF, Arimura K, Chavez AM. A re-evaluation of the mechanism of action of botulinum toxin on facial movement disorders in man. Med Hypotheses 1998; 51(4):305-307.

26.

Headache Classification Committee of the International Headache Society. The

TE D

International Classification of Headache Disorders, 3rd edition (beta version). Cephalalgia 2013; 33(9):629-808.

Olver J, Esquenazi A, Fung VS, Singer BJ, Ward AB, Cerebral Palsy I. Botulinum

EP

27.

toxin assessment, intervention and aftercare for lower limb disorders of movement

AC C

and muscle tone in adults: international consensus statement. Eur J Neurol 2010; 17 suppl 2:57-73.

28.

Dunne JW, Gracies JM, Hayes M, Zeman B, Singer BJ, Multicentre Study G. A prospective, multicentre, randomized, double-blind, placebo-controlled trial of onabotulinumtoxinA to treat plantarflexor/invertor overactivity after stroke. Clin Rehabil 2012; 26(9):787-797.

25

ACCEPTED MANUSCRIPT

29.

Turner-Stokes L, Ashford S, Jacinto J, Maisonobe P, Balcaitiene J, Fheodoroff K. Impact of integrated upper limb spasticity management including botulinum toxin

RI PT

A on patient-centred goal attainment: rationale and protocol for an international prospective, longitudinal cohort study (ULIS-III). BMJ Open 2016; 6(6):e011157. Wissel J, Ganapathy V, Ward AB, et al. OnabotulinumtoxinA improves pain in

SC

patients with post-stroke spasticity: Findings from a randomized, double-blind,

EP

TE D

M AN U

placebo-controlled trial. J Pain Symptom Manage 2016; 52(1):17-26.

AC C

30.

26

ACCEPTED MANUSCRIPT

Table 1. OnabotulinumtoxinA Injection Sites and Doses in the Double-Blind and Open-Label Phases

Required Muscles

Injection Site

Dose

75 U (25 U × 3 sites)

Gastrocnemius lateral head

75 U (25 U × 3 sites)

75 U (25 U × 3 sites)

TE D

Soleus

M AN U

SC

Gastrocnemius medial head

RI PT

Double-Blind Phase

75 U (25 U × 3 sites)

AC C

EP

Tibialis posterior

Optional Muscles

Injection Site

Dose*

Flexor digitorum longus

50 U (25 U × 2 sites)

Flexor digitorum brevis

25 U (1 site)

Flexor hallucis longus Extensor hallucis

50 U (25 U × 2 sites) 25 U (1 site)

27

ACCEPTED MANUSCRIPT

Rectus femoris

100 U (25 U × 4 sites)

Eligible Muscles

Injection Site

Dose†

75 U (25 U × 3 sites)

Gastrocnemius lateral head

Soleus

M AN U

SC

Gastrocnemius medial head

RI PT

Open-Label Phase

75 U (25 U × 3 sites)

TE D

75 U (25 U × 3 sites)

AC C

EP

Tibialis posterior

100 U (25 U × 3 sites)

Flexor digitorum longus

50 U (25 U × 2 sites)

Flexor digitorum brevis

25 U (1 site)

Flexor hallucis longus Extensor hallucis Rectus femoris

50 U (25 U × 2 sites) 25 U (1 site) 100 U (25 U × 4 sites)

28

ACCEPTED MANUSCRIPT

Hamstrings

100 U (25 U × 4 sites)

AC C

EP

TE D

M AN U

SC

RI PT

*Maximum permitted dose in the optional muscles, to a total additional dose of ≤100 U. † Maximum permitted dose in each muscle, to a total dose of ≤400 U.

29

ACCEPTED MANUSCRIPT

Placebo /

OnabotulinumtoxinA

OnabotulinumtoxinA

n=233

n=235

Mean (SD) age, y

56.0 (12.6)

57.0 (11.9)

Men, n (%)

148 (63.5)

Caucasian, n (%)

184 (79.0)

194 (82.6)

Mean (SD) weight, kg

80.2 (17.8)

79.5 (15.5)

Mean (SD) height, cm

169.6 (9.1)

169.8 (9.2)

215 (92.3)

219 (93.2)

67.1 (74.4)

61.6 (73.9)

23 (9.9)

25 (10.6)

160 (68.7)

150 (63.8)

SC

RI PT

OnabotulinumtoxinA /

TE D

Table 2. Baseline Demographics and Characteristics

M AN U

155 (66.0)

Baseline MAS score = 3,

EP

n (%)* Mean (SD) time since

AC C

stroke, mo

Stroke severity, n (%) Mild

Moderate

30

ACCEPTED MANUSCRIPT

Severe

50 (21.5)

60 (25.5)

Right leg only

13 (5.6)

15 (6.4)

Left leg only

23 (9.9)

24 (10.2)

Right arm and right leg

96 (41.2)

Left arm and left leg

101 (43.3)

102 (43.4)

0

2 (0.9)

SC 92 (39.1)

M AN U

spasticity, n (%)

RI PT

Limbs affected by

Other

TE D

MAS=Modified Ashworth Scale; NA=not applicable.

AC C

EP

*Remainder of patients had baseline MAS score = 4.

31

ACCEPTED MANUSCRIPT

Table 3. Correlation of MAS and CGI Scores in the Double-Blind Phase Spearman’s Correlation Coefficient

P Value

RI PT

Visit

Number of Patients OnabotulinumtoxinA Placebo n=233 n=235 228

233

–0.625

< .001

Week 4

225

229

–0.596

< .001

Week 6

224

224

–0.631

< .001

Week 8

227

228

–0.531

< .001

Week 12

226

226

–0.382

< .001

M AN U

SC

Week 2

AC C

EP

TE D

CGI=Clinical Global Impression; MAS=Modified Ashworth Scale

32

ACCEPTED MANUSCRIPT

Table 4. Adverse Event Profile

RI PT

Placebo / OnabotulinumtoxinA

n=231*

All TEAEs

154 (66.7)

Treatment-related AEs

23 (10.0)

Serious AEs

30 (13.0)

Treatment-related serious AEs Discontinued because of AEs

118 (52.2) 16 (7.1)

M AN U

0

n=226*

SC

AEs Across All Treatment Cycles, n (%)

OnabotulinumtoxinA / OnabotulinumtoxinA

19 (8.4) 0

8 (3.5)

6 (2.7)

2 (0.9)

3 (1.3)

OnabotulinumtoxinA

Placebo

n=231*

n=233†

95 (41.1)

80 (34.3)

Pain in extremity

11 (4.8)

11 (4.7)

Nasopharyngitis

8 (3.5)

6 (2.6)

Arthralgia

8 (3.5)

2 (0.9)

Fall

6 (2.6)

8 (3.4)

Back pain

6 (2.6)

4 (1.7)

Injection site pain

5 (2.2)

3 (1.3)

Deaths

TE D

TEAEs occurring in ≥2% of patients during the first 12 weeks of the double-blind phase, n (%)

AC C

EP

Overall

AE=adverse event; TEAE=treatment-emergent adverse event.

33

ACCEPTED MANUSCRIPT

*Patients are presented by treatment group as treated on day 1; only patients receiving ≥1 treatment of onabotulinumtoxinA, either in the double-blind or in the open-label phase, are

RI PT

counted. †

2 patients from the onabotulinumtoxinA group and 2 patients from the placebo group

AC C

EP

TE D

M AN U

SC

were not treated and therefore were not included in the safety population.

34

ACCEPTED MANUSCRIPT

Legends to Figures Study design.

RI PT

Figure 1.

CGI=Clinical Global Impression; GAS= Goal Attainment Scale;

Figure 2.

Patient disposition and exit status (ITT).

SC

MAS=Modified Ashworth Scale.

M AN U

ITT=intent-to-treat; OL=open-label; TX=treatment cycle. Four patients discontinued before receiving any dose. Figure 3.

Efficacy outcomes. (A) Mean change from baseline for ankle MAS (ITT). (B) Percentage of patients who achieved ≥1 grade reduction in MAS at all

TE D

time points (ITT). (C) Mean CGI by physician. (D) Proportion of responders for passive and active goals assessed by the physician.

EP

CGI=Clinical Global Impression (as assessed by physician); GAS=Goal Attainment Scale; ITT=intent to treat; MAS=Modified Ashworth Scale;

AC C

OL=open-label; TX=treatment cycle. (A) *P ≤ .03 vs placebo. (B) *P ≤ .04 vs placebo. (C) *P ≤ .04 vs placebo. CGI scores range from –4 = very marked worsening to +4 = very marked improvement.

35

ACCEPTED MANUSCRIPT

(D) Goal achievement = GAS ≥0; Goal improvement = GAS ≥–1. GAS scores ranged from −3 to +2 (−3 = worse than start; −2 = equal to start,

RI PT

the patient’s initial condition, no change; −1 = less than expected, slight improvement but below the defined therapeutic goal; 0 = expected goal, attained the defined therapeutic goal; +1 = somewhat more than

SC

expected, improvement slightly exceeds the defined therapeutic goal; +2 = much more than expected, improvements clearly exceeded the defined

Figure 4.

M AN U

therapeutic goal).

Distribution of change in MAS scores from baseline at week 4.

AC C

EP

TE D

A) OnabotulinumtoxinA, B) Placebo.

36

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

60

48.0 40

SC

31.1 16.9

20

1.8

2.2

-4

-3

RI PT

80

M AN U

OnabotulinumtoxinA Patients, %

A)

0

-2

-1

0

0.0 +1

Change in MAS from Baseline

TE D

B)

EP

60

60.3

AC C

Placebo Patients, %

80

40 20

0

24.5 10.9

0.9

2.6

-4

-3

0.9 -2

-1

0

+1

Change in MAS from Baseline