Improvement in Quality-of-Life-Related Outcomes Following Treatment with IncobotulinumtoxinA in Adults with Limb Spasticity: A Pooled Analysis

A strong correlation has been reported between patient-reported quality of life (QoL) and the investigator-rated Disability Assessment Scale (DAS) in patients with spasticity. The current analysis evaluates the effect of incobotulinumtoxinA on QoL-related outcomes (limb position abnormality, as well as dressing- and hygiene-related disability, measured with the DAS) in adults with upper limb spasticity, using pooled data from six studies. Separate analyses for each DAS domain were performed using data from patients with disabilities for that domain (DAS score ≥1). Results showed that a significantly greater proportion of incobotulinumtoxinA-treated compared with placebo-treated patients achieved a ≥1-point reduction from baseline in each of the DAS domains (improvement) 4 weeks after the first injection. The benefits of incobotulinumtoxinA were observed regardless of the baseline severity of DAS impairment and of the time elapsed since stroke. The effects of incobotulinumtoxinA 4 weeks after injection were maintained or enhanced over multiple injection cycles for all three DAS domains, supporting the use of repeated injection cycles to provide sustained QoL benefit. IncobotulinumtoxinA represents an important treatment option to achieve better QoL-related outcomes for patients with upper limb spasticity, irrespective of the duration of their condition.


Introduction
Spasticity is a sign of many central nervous system conditions, including stroke, multiple sclerosis, cerebral palsy, spinal cord injury, and traumatic brain injury [1][2][3][4][5][6].Spasticity is estimated to occur in 25% of patients after stroke, rising to 40% in patients with paresis [7].Upper limb spasticity affects coordinated movement; limits activities such as grasp, release, and self-care activities; and predisposes to secondary complications, such as joint contractures and pain [8,9].Botulinum toxin type A (BoNT-A) is recommended as first-line treatment for spasticity [10,11].The safety and efficacy of incobotulinumtoxinA have been demonstrated in a number of studies in the treatment of limb spasticity and reduction of spasticityassociated pain [12][13][14][15][16][17][18].Additional investigations revealed that incobotulinumtoxinA is also associated with quality-of-life (QoL) improvement in adults with spasticity [16,19].
The Disability Assessment Scale (DAS) evaluates upper limb functional impairments and activity limitations commonly seen in patients with poststroke spasticity in four domains: limb position abnormality, dressing-and hygiene-related disability, and pain [20].Higher DAS scores indicate greater impairments in each of the four domains [20].Although the DAS was developed to assess poststroke spasticity-related impairments and activity limitations, its use is not only limited to stroke-related problems [21][22][23].
As previously demonstrated, upper limb spasticity has a significant negative effect on health-related QoL (HRQoL) [24].A spasticity-specific QoL scale (the Patient Reported Impact of Spasticity Measure; PRISM) has been used to demonstrate that spasticity in general can result in social avoidance or anxiety, and more severe spasticity in the arm may be associated with higher levels of perceived social embarrassment [25].A strong correlation has been reported between patient-reported HRQoL and the investigator-rated DAS dressing-and hygiene-related disability and pain domains [26] when used to classify problems relevant to patients with upper limb poststroke spasticity, which include hygiene, dressing, and limb posture.These DAS domains are, therefore, important to patients.Poststroke spasticity and the related limitations to performing basic activities can also have a societal, emotional, and economic impact on caregivers [27,28].Regarding the DAS, caregiver burden and the need for assistance are directly related to the level of disability in the dressing and hygiene domains [26].
The impact of incobotulinumtoxinA on upper limb spasticity-related limb position abnormality, dressing and hygiene has been investigated using the DAS in several studies as a primary [16,18] or secondary outcome [12,14,15,17].Most of these studies reported DAS outcomes with respect to the principal therapeutic target (PTT), which was the DAS domain identified by the patients in conjunction with the investigators as the most relevant and important for potential improvement with BoNT-A.The PTT was required to have a score ≥2 at screening/baseline.
The aim of the current study was to assess the benefit of treatment with incobo-tulinumtoxinA in adults with upper limb spasticity by considering its effect on QoL-related outcomes (limb position abnormality, dressing-and hygiene-related disability, as measured using the DAS), using pooled data from six phase 2 or 3 clinical studies.The pain domain data were excluded from this analysis since the effects of incobotulinumtoxinA on spasticity-associated pain from these six studies have recently been presented elsewhere [29].We hypothesized that QoL-related outcomes would improve after treatment with BoNT-A in a similar manner to spasticity-associated pain [29].To include the largest possible sample size and to assess whether domains that were not identified as the PTT also benefitted from incobotulinumtoxinA treatment, the analyses were based on all evaluable patients with data for the individual DAS items rather than just the PTT data.
In patients with stroke-related upper limb spasticity, time since stroke had little effect on the efficacy of incobotulinumtoxinA with respect to response rates at week 4 for each of the three DAS domains investigated (Table 2).Response rates were numerically similar across all time intervals assessed.In patients with stroke-related upper limb spasticity, time since stroke had little effect on the efficacy of incobotulinumtoxinA with respect to response rates at week 4 for each of the three DAS domains investigated (Table 2).Response rates were numerically similar across all time intervals assessed.

Response Rates Following Multiple Injection Cycles
Response rates for all three DAS domains investigated increased after each repeated incobotulinumtoxinA injection cycle, showing a cumulative positive effect on QoL-related outcomes (Figure 3).At each injection visit, many patients had a maintained response to incobotulinumtoxinA (>26%) and response rates at these visits increased with each subsequent injection cycle.
Four weeks after the fourth injection cycle, response rates in patients with limb position abnormality at baseline reached 59.5%.Response rates in patients with dressing disability reached 49.2%, and response rates in patients with hygiene-related disability reached 55.6%.

Response Rates Following Multiple Injection Cycles
Response rates for all three DAS domains investigated increased after eac incobotulinumtoxinA injection cycle, showing a cumulative positive effect on Q outcomes (Figure 3).At each injection visit, many patients had a maintained r incobotulinumtoxinA (>26%) and response rates at these visits increased with e quent injection cycle.Four weeks after the fourth injection cycle, response rates in patients with tion abnormality at baseline reached 59.5%.Response rates in patients with dr ability reached 49.2%, and response rates in patients with hygiene-related reached 55.6%.

Response Rates for the PTT
In patients with a DAS disability domain as the PTT, a higher response ra served with incobotulinumtoxinA treatment than placebo 4 weeks after the fir (52.0% vs. 25.8%, a difference (95% CI) of 26.2 (19.2-33.2)%;p < 0.0001).When a PTT severity at baseline (moderate or severe disability was a requirement), numtoxinA-treated patients showed a significantly higher response rate at w those treated with placebo irrespective of severity: Response rates were 37.4% for those with moderate disability and 66.9% vs. 35.6%for those with severe d < 0.0001 for both).Patients with 3-5, 6-10, and >10 years since stroke had n similar response rates with incobotulinumtoxinA compared to patients with since stroke (46.7%, 51.7%, and 64.5%, respectively, vs. 51.2%);all response rate tistically higher with incobotulinumtoxinA than with placebo (p < 0.01 for all stroke).
Response rates for the PTT increased after repeated incobotulinumtoxin cycles to 65.9% 4 weeks after the fourth injection, showing a cumulative positiv each injection visit, many patients had a maintained response to incobotulin

Response Rates for the PTT
In patients with a DAS disability domain as the PTT, a higher response rate was observed with incobotulinumtoxinA treatment than placebo 4 weeks after the first injection (52.0% vs. 25.8%, a difference (95% CI) of 26.2 (19.2-33.2)%;p < 0.0001).When analyzed by PTT severity at baseline (moderate or severe disability was a requirement), incobotulinumtoxinA-treated patients showed a significantly higher response rate at week 4 than those treated with placebo irrespective of severity: Response rates were 37.4% vs. 18.7% for those with moderate disability and 66.9% vs. 35.6%for those with severe disability (p < 0.0001 for both).Patients with 3-5, 6-10, and >10 years since stroke had numerically similar response rates with incobotulinumtoxinA compared to patients with 0-2 years since stroke (46.7%, 51.7%, and 64.5%, respectively, vs. 51.2%);all response rates were statistically higher with incobotulinumtoxinA than with placebo (p < 0.01 for all times since stroke).
Response rates for the PTT increased after repeated incobotulinumtoxinA injection cycles to 65.9% 4 weeks after the fourth injection, showing a cumulative positive effect.At each injection visit, many patients had a maintained response to incobotulinumtox-inA (36.9%-51.4%)and response rates at these visits increased with each subsequent injection cycle.

Discussion
This is one of the largest analyses to consider the effect of BoNT-A on individual DAS domains; a previous pooled analysis of data from the same studies confirmed the beneficial effect of incobotulinumtoxinA on the pain DAS domain [29].Of the 937 patients included in these analyses, the vast majority had upper limb position abnormality (98.0%), dressing disability (96.8%) and hygiene-related disability (92.3%) at baseline, usually as a result of stroke-related spasticity.For many of the patients, these limitations were of sufficient severity to be considered as the PTT for treatment.
Results showed that a significantly greater proportion of incobotulinumtoxinA-treated compared with placebo-treated patients achieved a ≥1-point improvement in limb position abnormality, dressing-and hygiene-related disability DAS scores 4 weeks after the first injection.Compared with patients receiving a placebo, incobotulinumtoxinA-treated patients were at least three times more likely to achieve a reduction in DAS limb position abnormality score, almost three times more likely to achieve a reduction in DAS hygiene-related disability score, and at least twice as likely to achieve a reduction in DAS dressing disability score.Importantly, the benefits of incobotulinumtoxinA were observed regardless of the baseline severity of DAS impairment and, in those with stroke-related spasticity, of the time elapsed since stroke, although differences versus placebo were not always statistically significant in the subgroups.Specifically, patients with mild upper limb position abnormality or hygiene-related disability did not have statistically significantly higher response rates with incobotulinumtoxinA than with placebo.These nonsignificant findings may have been a result of the smaller numbers of patients with mild disability, the likelihood that individuals with mild disability have residual control of voluntary movements, which allows improvements in task performance in the placebo groups, or the fact that the opportunity for large improvements in those with only mild disability is reduced.Nonsignificant differences from placebo seen in time-since-stroke subgroups generally occurred with no discernible pattern, and again, patient numbers were reduced as time since stroke increased.
Time since stroke (<36 months vs. ≥36 months) was a significant predictor of active motor function in a study of onabotulinumtoxinA in upper limb spasticity, with a shorter time since stroke predicting a better functional response to treatment [31].Furthermore, in a study of lower limb spasticity, initiating treatment with BoNT earlier after stroke (≤24 months vs. >24 months) was associated with significant benefits in a greater number of efficacy outcomes than later treatment initiation [32].In contrast, in the current study, response rates for all three investigated DAS domains were either consistent across the assessed time periods since stroke or appeared to be higher with increased time since stroke, and the difference in treatment effect between incobotulinumtoxinA and placebo was actually largest in those with >10 years since stroke when upper limb position abnormality and dressing disability were considered.This demonstrates that incobotulinumtoxinA efficacy can still be achieved in patients who have gone untreated for spasticity for many years, and treatment with incobotulinumtoxinA can be started in patients with spasticity at any time, regardless of when their spasticity was first diagnosed.Similarly, improvements were significantly better in patients who received one or two cycles of abobotulinumtox-inA "late" in the disease course than those who received it "early" poststroke, although benefits in achieving an individual functional goal, and ease of care and hygiene, were seen irrespective of timing of treatment with respect to stroke, particularly after four injection cycles [33].
Repeated injections were administered in five of the six pooled studies.Multiple injection cycles of incobotulinumtoxinA have been shown to have a cumulative effect on DAS pain in spasticity [29] and pain in cervical dystonia [34].Our analyses of the remaining three DAS domains showed that many patients had a maintained response to incobotulinumtoxinA at the time of reinjection.In common with other BoNT-A analyses [21,35], we observed that the effects of repeated injection cycles of incobotulinum-toxinA at 4 weeks postinjection appeared to plateau after about the third cycle.However, the beneficial effects on DAS scores (measured using the PTT) of incobotulinumtoxinA at 12 weeks postinjection were not found to have plateaued after four injection cycles in a post hoc analysis of data from two trials [36].Many study publications using the DAS tool focus on only the PTT, either pooling findings across the PTT domains [14,17,19,37,38] or reporting domain-specific data only from patients for whom the domain was the PTT [13,18], including a meta-analysis of five BoNT-A studies [39].There is some evidence showing that BoNT-A improves each of the four domains [12,13,35], although significant improvement may not be possible for domains that are only mildly impaired at baseline [40,41].
Spasticity develops in about 25% of stroke survivors, although the incidence varies according to the follow-up time poststroke since spasticity can develop late or resolve; for example, in a meta-analysis of data from 24 studies, the incidence was 31.6% if follow-up was within 1 month, 21.8% if in 1-3 months, 26.3% if in 3-6 months, and 24.2% if after more than 6 months [7].The findings of this analysis show that, overall, efficacy was not affected by the timing of incobotulinumtoxinA relative to stroke, which will, therefore, be of interest to many patients who have survived stroke.
The strengths of this study include that the patient population came from a wide (international) geographic spread and that the DAS, a standard measure of functional disability, was used across the six studies included in the analyses.Additional strengths include that most of the incorporated studies were placebo-controlled for the first injection cycle, data were obtained over multiple incobotulinumtoxinA injection cycles and the analyses were performed on a background of stable antispastic medications (centrally acting muscle relaxants, benzodiazepine) and physical/occupational therapy.Using a multipattern treatment approach, incobotulinumtoxinA produced effective improvements in QoL-related outcomes, a finding that should be considered along with the well-established safety profile and low immunogenicity profile of incobotulinumtoxinA .Furthermore, the mixed etiology of spasticity evident in this study may make the findings applicable to a wider range of patients, although most patients had poststroke spasticity.
Limitations of this study include potential bias due to different patient numbers from the different studies used in the pooled analyses.For example, the PURE study [14] contributed around 40% of patients analyzed, whereas study MRZ_60201_03071 contributed only 3%.Additionally, the contributing studies were not all in the same drug development phase.Their designs differed, including two that were not placebo-controlled, and the DAS was not defined as a primary endpoint in most of the studies.Finally, as with other published pooled analyses, we have not used a systematic search approach and instead only selected relevant clinical trials conducted by the study sponsor -Merz Therapeutics GmbH.The reason for this was to ensure data availability at the level required to conduct the pooled analyses.

Conclusions
This pooled analysis showed that incobotulinumtoxinA significantly improves QoLrelated outcomes, such as upper limb positioning abnormality and dressing-and hygienerelated disability, in adults with upper limb spasticity.This is the largest patient cohort analyzed to date in this setting, providing additional evidence to support the use of incobotulinumtoxinA in QoL-related outcomes.The three DAS domains analyzed are frequently selected by patients with spasticity as the PTT, indicating the impact they have on patients' lives.
The effects of incobotulinumtoxinA 4 weeks after injection were maintained or enhanced over multiple injection cycles for all three DAS domains, supporting the use of repeated injection cycles to provide sustained QoL benefit.IncobotulinumtoxinA represents an important treatment option to achieve better QoL-related outcomes for patients with upper limb spasticity, irrespective of the duration of their condition.Improvements in QoL induced by incobotulinumtoxinA can be achieved regardless of the underlying neurological condition that has resulted in spasticity, sex, ethnicity, and pretreatment status.Although it is desirable to start treating spasticity early, the results of this analysis demonstrate that initiating incobotulinumtoxinA treatment later will still lead to meaningful QoL improvements for the patients.Future studies on the benefit of BoNT therapy for QoL are warranted to further understand these improvements in patients with spasticity.

Studies Included in the Analysis
This post hoc analysis used pooled upper limb position abnormality, dressing and hygiene data from six prospective, multicenter, phase 2 or 3 studies of incobotulinum-toxinA (Xeomin ® ; Merz Therapeutics GmbH, Frankfurt, Germany) in the treatment of upper [12,14,17,18,30] or upper and lower [15] limb muscle spasticity in adults conducted by the study sponsor -Merz Therapeutics GmbH (Table 3).Studies were limited to Merz-sponsored studies in order to ensure sufficient data availability to conduct the pooled analyses.MRZ_60201_03071 was terminated prematurely due to low recruitment.It was not registered and no data have been published previously outside of pooled analyses.Four of the studies were randomized, double-blind, and placebo-controlled [12,14,17,30], whereas the remaining two studies evaluated different incobotulinumtoxinA dosing schedules [15] and dilutions [18].
The studies were conducted across the world and included a range of patient ethnicities.In the six studies, adult patients aged ≥18 years who had not received BoNT-A injections within at least 4 months of screening received incobotulinumtoxinA injections as appropriate for their condition, most commonly at a total body dose of 400 U.Each injection was followed by at least 12 weeks of observation and assessment.In studies where patients received more than one incobotulinumtoxinA injection, the time between injections was 12-14 weeks for the majority of patients [29].
All studies were conducted in accordance with the Declaration of Helsinki and Good Clinical Practice and were approved by the ethics committee for each participating site.All patients provided written informed consent prior to study participation.

Disability Assessment Scale
The current analyses used the limb position, dressing, and hygiene domains of the DAS [20] to measure functional disability and activity limitations resulting from spasticity in each of the three domains and, therefore, provide further information regarding the impact of upper limb spasticity on QoL.In all studies, these domains were assessed at baseline, every injection visit, and 4 weeks after each injection using a 4-point rating scale: score 0 (no disability), 1 (mild disability), 2 (moderate disability), and 3 (severe disability).
Response was defined as ≥1-point reduction from baseline in limb position abnormality, dressing, and hygiene score, as per previous studies [35,37,79].Apart from study MRZ_60201_03071, where no PTT was selected, the PTT was selected by the patient and physician from one of the four DAS items (limb position, dressing, hygiene, pain) with a required baseline score ≥2.

Analyses
Separate analyses for each DAS domain were performed using pooled data from patients with each limb position abnormality, dressing-, and hygiene-related disability at baseline (DAS score ≥1 for that domain).Patients with more than one domain score ≥ 1 could be included in multiple analyses.
DAS response rates over time for all injection cycles relative to baseline were recorded at injection visit 1 (baseline), at each subsequent injection visit if applicable, and at control visits 4 weeks (±3 days) after each injection.Response rates for incobotulinumtoxinA versus placebo were compared 4 weeks after the first injection using Wald tests (95% CI) and evaluated 4 weeks after each injection visit of the subsequent three cycles for in-cobotulinumtoxinA without placebo control.For the placebo-controlled studies, after the placebo injection cycle (first cycle only), data from patients who subsequently received incobotulinumtoxinA in an open-label extension phase were included with data from the original incobotulinumtoxinA groups (data assigned to first, second, and third incobo-tulinumtoxinA doses).
Data for limb position abnormality, dressing-, and hygiene-related disability at week 4 following incobotulinumtoxinA injection were analyzed in terms of the proportion of incobotulinumtoxinA-and placebo-treated patients who responded to treatment and the likelihood of response to incobotulinumtoxinA versus placebo (logistic regression analysis, presented as an OR analyzed using the chi-square test and 95% Wald CI).Furthermore, several subgroup analyses were performed using data from the first injection cycle.The difference in overall response rates to incobotulinumtoxinA versus placebo according to baseline severity of the respective DAS domain (mild, moderate, severe) was assessed, as well as the difference in overall response rates to incobotulinumtoxinA versus placebo according to time since stroke (0-2 years, 3-5 years, 6-10 years, >10 years).The proportion of incobotulinumtoxinA-and placebo-treated patients who responded to treatment according to PTT (any of the four DAS domains) was also evaluated after all injection cycles.
Analyses were based on observed cases; there was no strategy for missing postbaseline data in participants with a domain DAS score ≥1 at baseline, as few data were missing (limb position abnormality, 2.5% (n = 23), dressing disability, 2.4% (n = 22), hygiene-related disability, 2.4% (n = 21)).Analyses were performed using Statistical Analysis Software (SAS) 9.4 (SAS Institute Inc., Cary, NC, USA).Logistic regression was used to calculate ORs and associated p-values.For response differences, Wald tests and 95% CIs were computed.

Figure 1 .
Figure 1.DAS domain response rates (≥1-point improvement in DAS score) at week 4 after the first injection cycle by treatment and domain.Chi-square test was used for OR and Wald test to determine between-treatment p-values.*, p < 0.01; ***, p < 0.0001.CI, confidence interval; DAS, Disability Assessment Scale; OR, odds ratio.

Figure 1 .
Figure 1.DAS domain response rates (≥1-point improvement in DAS score) at week 4 after the first injection cycle by treatment and domain.Chi-square test was used for OR and Wald test to determine between-treatment p-values.*, p < 0.01; ***, p < 0.0001.CI, confidence interval; DAS, Disability Assessment Scale; OR, odds ratio.
test was used to determine between-treatment p-values.CI, Wald confidence interval; DAS, Disability Assessment Scale.Toxins 2023, 15, x FOR PEER REVIEW 5 of 18severity numerically favored incobotulinumtoxinA over placebo for all QoL-related outcomes investigated, reaching statistical significance for dressing disability (Figure2).

Figure 2 .
Figure 2. Differences in overall DAS domain response rates (≥1-point improvement in DAS score) at week 4 after the first injection cycle, by treatment and baseline severity of the respective DAS domain.Wald test was used to determine between-treatment p-values.*, p < 0.05; **, p < 0.001; ***, p < 0.0001.DAS, Disability Assessment Scale.

Figure 2 .
Figure 2. Differences in overall DAS domain response rates (≥1-point improvement in DAS score) at week 4 after the first injection cycle, by treatment and baseline severity of the respective DAS domain.Wald test was used to determine between-treatment p-values.*, p < 0.05; **, p < 0.001; ***, p < 0.0001.DAS, Disability Assessment Scale.

Figure 3 .
Figure 3. DAS domain response rates (≥1-point improvement in DAS score from basel 4 after repeated injection cycles.CV took place 4 weeks after the previous IV; time betwe was 12-14 weeks for the majority of patients.Only the first injection was placebo-co placebo-treated subjects received incobotulinumtoxinA in subsequent cycles, their r their first cycle of incobotulinumtoxinA (at study IV2 and CV2) are included in IV1 and the subsequent two incobotulinumtoxinA injection cycles.The N-value given below eac number of patients with data at that visit.CV, control visit; DAS, Disability Assessme injection visit.

Figure 3 .
Figure 3. DAS domain response rates (≥1-point improvement in DAS score from baseline) at week 4 after repeated injection cycles.CV took place 4 weeks after the previous IV; time between injections was 12-14 weeks for the majority of patients.Only the first injection was placebo-controlled.As placebo-treated subjects received incobotulinumtoxinA in subsequent cycles, their responses to their first cycle of incobotulinumtoxinA (at study IV2 and CV2) are included in IV1 and CV1 and in the subsequent two incobotulinumtoxinA injection cycles.The N-value given below each visit is the number of patients with data at that visit.CV, control visit; DAS, Disability Assessment Scale; IV, injection visit.

Table 1 .
Characteristics of patients (n = 937) with limb position abnormality and dressing-and hygiene-related disability (score ≥1) at baseline.

Table 2 .
DAS domain response rates (≥1-point improvement in DAS score) at week 4 after the first injection cycle by treatment and time since stroke.

Table 2 .
DAS domain response rates (≥1-point improvement in DAS score) at week 4 after the first injection cycle by treatment and time since stroke.

Table 3 .
The six studies included in the pooled analyses (main period and open-label extension period).

Table 3 .
Cont.In this study, patients chose one of four domains of the DAS (dressing, limb position, pain, hygiene) as the primary therapeutic target; this was limb position in 63%, dressing in 24%, pain in 6%, and hygiene in 8% of patients.b In PURE, the primary target clinical pattern treated included flexed elbow, flexed wrist, or clenched fist at predefined fixed doses.Other clinical patterns could be treated with the remainder of the total dose as medically indicated.